ABSTRACT 
 
 
 
 
Title of Document: POLY (AMIDO AMINE) DENDRIMERS: 
TRANSEPITHELIAL TRANSPORT 
MECHANISMS AND APLICATIONS IN 
ORAL DRUG DELIVERY 
  
 Deborah Swet Goldberg, Doctor of Philosophy, 
2010 
  
Directed By: Profesor Hamidreza Ghandehari, Fischel 
Department of Bioengineering 
 
 
Smal molecule chemotherapy drugs used in clinical practice are plagued by 
dose-limiting side efects due to off-target toxicities.  In addition, because of their 
low water solubility and poor bioavailability, they must be administered 
intravenously, leading to high treatment costs and recurring hospital visits. There is a 
significant need for therapies that improve the bioavailability of chemotherapy agents 
and enhance specific drug release in the tumor environment. 
Dendrimers, a clas of highly-branched, nanoscale polymers, share many 
characteristics with traditional polymeric cariers, including water solubility, high 
capacity of drug loading and improved biodistribution. Poly (amido amine) 
(PAMAM) dendrimers have shown promise as oral drug cariers due to their compact 
size, high surface charge density and permeation across the intestinal epithelial 
  
barier. Atachment of chemotherapy drugs to PAMAM dendrimers has the potential 
to make them oraly administrable and reduce of-target toxicities. 
In this disertation we investigate the transport mechanisms of PAMAM 
dendrimers and their potential in oral drug delivery. We demonstrate that anionic 
G3.5 dendrimers are endocytosed by dynamin-dependent mechanisms and their 
transport is governed by clathrin-mediated pathways. We show that dendrimer 
celular internalization may be a requisite step for tight junction opening. We also 
demonstrate that conjugation of smal poly (ethylene glycol) chains to anionic 
dendrimers decreases their transport and tight junction opening due to reduction in 
surface charge, ilustrating that smal changes in surface chemistry can significantly 
impact transepithelial transport. Knowledge of transport mechanisms and the impact 
of surface chemistry wil aid in rational design of dendrimer oral drug delivery 
systems. 
The potential of dendrimers as oral drug delivery cariers is demonstrated by 
the evaluation of G3.5 PAMAM dendrimer-SN38 conjugates for oral therapy of 
hepatic colorectal cancer metastases, a pathology present in over 50% of colorectal 
cancer cases that is responsible for two-thirds of deaths. Conjugation of SN38, a 
potent chemotherapy drug with poor solubility and low bioavailability, to PAMAM 
dendrimers via a glycine linker increased intestinal permeability, decreased intestinal 
toxicity and showed selective release in the presence of liver carboxylesterase, 
ilustrating that PAMAM dendrimers have the potential to improve the oral 
bioavailability of potent anti-cancer therapeutics. 
 
  
 
 
 
 
 
 
POLY (AMIDO AMINE) DENDRIMERS: TRANSEPITHELIAL TRANSPORT 
MECHANISS AND APLICATIONS IN ORAL DRUG DELIVERY 
 
 
 
By 
 
 
Deborah Swet Goldberg 
 
 
 
 
 
Disertation submited to the Faculty of the Graduate School of the 
University of Maryland, College Park, in partial fulfilment 
of the requirements for the degre of 
Doctor of Philosophy 
2010 
 
 
 
 
 
 
 
 
 
 
Advisory Commite: 
Profesor Hamidreza Ghandehari, Co-Chair 
Profesor Wiliam Bentley, Co-Chair 
Profesor Robert Briber 
Profesor Silvia Muro 
Profesor Peter Swan 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
? Copyright by 
Deborah Swet Goldberg 
2010 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i 
 
Dedication 
To my wonderful husband Hirsh, for his constant, unwavering love and support. 
 
 ii 
 
Acknowledgements 
I have had the honor of working with many talented individuals during my 
PhD studies who have helped me reach this milestone in my life. I would first like to 
thank my advisor Dr. Hamid Ghandehari for his enduring commitment to my 
training as a scientist. Although Dr. Ghandehari moved to the University of Utah 
shortly after I began my research, he has always strived to maintain constant 
communication and provide advice on my project. I appreciate his atention to detail 
in his careful review of my abstracts, posters, presentations, publications and 
disertation. Dr. Ghandehari has always inspired me to do my best work, and I know 
that the impact of his guidance wil follow me into my future scientific endeavors.  
I would also like to thank Dr. Peter Swaan for his contributions to my PhD 
studies as both my co-advisor and local contact. Dr. Swan has helped me fel at 
home at the University of Maryland, Baltimore. I sincerely appreciate the many 
research discussions that we had during group metings and individual metings. I 
am also thankful for the fresh perspective he offered on my publications and 
presentations. Dr. Swan has taught me to think criticaly and continualy question 
everything, a mindset that wil serve me wel in the future. 
I also wish to thank my commite members, Dr. Silvia Muro, Dr. Robert 
Briber and Dr. Wiliam Bentley for their helpful suggestions during my research 
proposal and commite meting. Their insights have helped shape the trajectory of 
my project, and I am grateful for their guidance.  
In addition to my advisors and my commite members, two other profesors 
have had a significant impact on my PhD studies. I wil be forever indebted to Dr. 
 
 iv 
 
Rohit Kolhatkar for his guidance during the first few years of my PhD studies. 
Rohit taught me experimental techniques and advised me on planning experiments, 
interpreting results and troubleshooting chemical synthesis strategies. After Rohit left 
to become an asistant profesor at the University of Ilinois, Chicago I realized that 
his guidance had taught me everything I needed to think like a scientist and 
succesfully complete the remainder of my studies. I would also like to thank Dr. 
Anjan Nan for graciously alowing me to conduct my research in his laboratory 
space and use his equipment. Dr. Nan?s support of my long distance advisement by 
Dr. Ghandehari made it possible for me to continue my research on dendrimers. 
I wish to acknowledge al of my felow lab members past and present in the 
Ghandehari, Swan and Nan labs who have had an influence on my day-to-day life in 
the lab. Specificaly, I would like to acknowledge Dr. Mark Borgman for his 
endles technical advice on experiments and his wonderful friendship. Even after he 
graduated, Mark was only a phone cal or an email away, and has continued to be a 
great friend and knowledgeable research consultant. I would also like to thank 
Britany Avarit, Tatiana Claro da Silva and Paul Dowel for their friendship and 
support over the years.  Finaly, I would like to thank Carl the janitor who always 
managed to make me smile on his early morning rounds.  
During my PhD experience I was fortunate to complete a summer internship 
in the Department of Formulation Sciences at MedImune LC. Although this 
internship experience was distinct from my graduate research, it gave me an 
appreciation of the pharmaceutical industry and taught me new ays of approaching 
problems, which positively impacted my graduate work after the internship. I would 
 
 v 
 
like to thank my supervisor Dr. Hasige Sathish and my formulation sub-group leader 
Dr. Ambarish Shah for their positive contributions to my PhD experience and for 
providing me with this unique opportunity. 
Several felowships and grants during my graduate studies have helped to 
fund my research. I would like to thank the National Science Foundation for the 
Graduate Research Felowship, which sponsored my first thre years of graduate 
school. It was an honor to receive this felowship and it had a positive impact on my 
graduate school carer, providing me with fredom in project selection. I would also 
like to thank Dr. Robert Fischel for generously sponsoring the Fischel Felowship 
in Bioenginering, which funded my final years in graduate school. Dr. Fischel?s 
commitment to fostering new Bioengineering research is admirable and his 
enthusiasm is contagious. I also acknowledge NIH R01 EB007470 for funding 
supplies for this research.  
I would like to recognize my friends and family, whose support and 
encouragement have been critical for the succesful completion of my PhD studies. 
My parents, Shari and Rick Swet, have always ben my biggest cheerleaders, 
supporting al of my academic and personal pursuits. I truly appreciate their 
excitement about al of my acomplishments from elementary school until today and 
their constant encouragement to strive for my dreams. I would also like to thank my 
mother and father in-law, Karen and Larry Goldberg for their interest and support 
throughout my PhD studies. I would like to acknowledge the wonderful support of 
my friends Marie Jeng, Chloe Marin, Rifat Jafren and Kevin Nelson among 
many others who have shown interest in my PhD progres over the years and offered 
 
 vi 
 
countles words of encouragement.  I wish to expres my gratitude to the instructors 
at Columbia Jazzercise for providing a fun and exhilarating stres release as wel as 
al of my friends at Jazercise for their constant interest and encouragement. Finaly, 
I would like to thank my husband Hirsh who has ben my greatest supporter through 
al of the ups and the downs of my PhD research. Hirsh has always offered emotional 
support and encouragement when I was frustrated with unsuccesful experiments and 
has unselfishly taken over household responsibilities when I had to work late at night 
or on the wekend to met a deadline.  I know that his love and unwavering support 
helped me push through the most dificult chalenges of my PhD and I could not have 
done it without him. He truly deserves this degre as much as I do.  
 
 vii 
 
Table of Contents 
 
Dedication........................................................i 
 
Acknowledgements.................................................ii 
 
List of Tables.....................................................xi 
 
List of Figures....................................................xii 
 
Abreviations....................................................xv 
 
Chapter 1 : Introduction.............................................1 
 
1.1  Introduction..................................................1 
1.1.1 Polymer Therapeutics........................................1 
1.1.2 Oral Drug Delivery..........................................2 
1.1.3 Poly (amido amine) Dendrimers................................3 
1.1.4 SN38.....................................................4 
1.2 Specific Aims.................................................5 
1.3 Scope and Organization..........................................6 
 
Chapter 2 : Background.............................................8 
 
2.1 Introduction...................................................8 
2.2 Polymeric Drug Delivery.........................................8 
2.2.1 Therapeutic Advantages of Polymer-Drug Conjugates in Chemotherapy..9 
2.2.2 Polymer-Drug Conjugates Currently in Clinical Trials..............12 
2.3 Administration of Drugs via the Oral Route..........................13 
2.3.1 Physiology of the Gastrointestinal Tract.........................14 
2.3.1.1 Compartments and Functions..............................14 
2.3.1.2 Intestinal Epithelial Barier................................17 
2.3.1.3 Tight Junction Biology: Structure and Function................18 
2.3.2 Mechanisms of Transport Across the Intestinal Barier..............21 
2.3.2.1 Paracelular Transport...................................21 
2.3.2.2 Pasive Difusion.......................................23 
2.3.2.3 Carier-Mediated Transport...............................24 
2.3.2.4 Endocytosis...........................................25 
2.3.3 Physiochemical Properties that Govern Intestinal Absorption.........28 
2.3.3.1 The Lipinski Rule of 5...................................28 
2.3.3.2 The Biopharmaceutics Clasification System (BCS).............28 
2.4 Models to Predict Oral Absorption and Oral Bioavailability..............30 
2.4.1 In Silico Models...........................................30 
2.4.2 Paralel Artificial Membrane Permeability Asay..................31 
2.4.3 Caco-2 Monolayers.........................................33 
2.4.4 Fast-Caco-2 Asay.........................................36 
 
 vii 
 
2.4.5 Other Types of Cel-Monolayer Systems.........................37 
2.4.6 Everted Rat Intestinal Sac....................................38 
2.4.7 Isolated Intestinal Tisue.....................................39 
2.4.8 Rat Intestinal Perfusion......................................40 
2.4.9 In Vivo Models............................................40 
2.5 Current Strategies for Oral Drug Delivery...........................41 
2.5.1 Prodrugs.................................................42 
2.5.2 Eflux and Metabolic Inhibitors................................43 
2.5.3 Tight Junction Modulators...................................44 
2.5.4 Macromolecules...........................................45 
2.6 Dendrimers..................................................47 
2.6.1 History of Dendrimer Development............................47 
2.6.2 Dendrimer Synthesis........................................49 
2.6.3 Types of Dendrimers........................................49 
2.7 Poly (amido amine) Dendrimers..................................51 
2.7.1 Structure of PAMAM Dendrimers..............................51 
2.7.2 Biocompatibility and Biodistribution of PAMAM Dendrimers........53 
2.7.3 Applications of PAMAM Dendrimers as Drug Cariers..............55 
2.8 PAMAM Dendrimers as Oral Drug Delivery Systems..................57 
2.8.1 Transepithelial Transport of PAMAM Dendrimers.................57 
2.8.2 Cytotoxicity of PAMAM Dendrimers...........................60 
2.8.3 Surface Modification of PAMAM Dendrimers....................61 
2.8.4 PAMA Dendrimer Internalization............................62 
2.8.5 PAAM Dendrimers as Oral Drug Delivery Systems...............63 
2.8.6 Designing PAMAM Dendrimer-Drug Conjugates for Oral Delivery....64 
2.9 Colorectal Cancer.............................................65 
2.9.1 Prevalence................................................65 
2.9.2 Colorectal Cancer Screning..................................66 
2.9.3 Colorectal Cancer Diagnosis and Staging........................67 
2.9.4 Colorectal Cancer Metastasis and Treatment......................67 
2.10 SN38......................................................69 
2.10.1 Mechanism of Action of Irinotecan and SN38....................70 
2.10.2 Current SN38 Drug Delivery Systems..........................72 
2.11 Unresolved Isues in Oral Delivery by Dendrimers...................73 
 
Chapter 3 : Celular Entry of G3.5 Poly (amido amine) Dendrimers by Clathrin- 
and Dynamin-Dependent Endocytosis Promotes Tight Junctional Opening in 
Intestinal Epithelia................................................75 
 
3.1 Introduction..................................................75 
3.2 Materials and Methods..........................................77 
3.2.1 Materials.................................................77 
3.2.2 Synthesis of G3.5-OG.......................................77 
3.2.3 Caco-2 Cel Culture........................................78 
3.2.4 Cytotoxicity of Endocytosis Inhibitors..........................79 
3.2.5 Celular Uptake............................................79 
 
 ix 
 
3.2.6 Colocalization and Intracelular Traficking......................80 
3.2.7 Transepithelial Transport....................................82 
3.2.8 Caco-2 Monolayer Visualization and Ocludin Staining.............83 
3.3 Results......................................................85 
3.3.1 Cytotoxicity of Endocytosis Inhibitors..........................85 
3.3.2 Celular Uptake of G3.5-OG Dendrimers in the Presence of Endocytosis 
Inhibitors.....................................................87 
3.3.3 Intracelular Traficking.....................................90 
3.3.4 Transepithelial Transport of G3.5-OG Dendrimers in the Presence of 
Endocytosis Inhibitors...........................................93 
3.3.5 Visualization of G3.5-OG Dendrimer Interaction with Caco-2 Cel 
Monolayers...................................................93 
3.3.6 Ocludin Staining in Presence of Dendrimers with and without Dynasore 
Treatment....................................................96 
3.4 Discussion...................................................99 
3.5 Conclusion..................................................104 
 
Chapter 4 : G3.5 PAMAM Dendrimers Enhance Transepithelial Transport of 
SN38 While Minimizing Gastrointestinal Toxicity......................105 
 
4.1 Introduction.................................................105 
4.2 Materials and Methods.........................................106 
4.2.1 Materials................................................106 
4.2.2 Synthesis and Characterization of G3.5-Gly-SN38 and G3.5-?Ala-SN38 
Conjugates...................................................107 
4.2.3 Stability Studies..........................................109 
4.2.4 Cel Culture..............................................110 
4.2.5 Potential Short-Term Cytotoxicity of G3.5-SN38 Conjugates........111 
4.2.6 Potential Delayed Cytotoxicity of G3.5-SN38 Conjugates...........111 
4.2.7 Transepithelial Transport...................................112 
4.2.8 Celular Uptake...........................................112 
4.2.9 IC
50
 in HT-29 Cels........................................113 
4.3 Results.....................................................114 
4.3.1 Stability of G3.5-SN38 Conjugates............................114 
4.3.2 Short-Term Cytotoxicity....................................119 
4.3.3 Delayed Cytotoxicity......................................121 
4.3.4 Transepithelial Transport...................................121 
4.3.5 Celular Uptake...........................................125 
4.3.6 IC
50
 in HT-29 Cels........................................127 
4.4 Discussion..................................................129 
4.5 Conclusion..................................................133 
 
Chapter 5 : Transepithelial Transport of PEGylated Anionic Poly (amido amine) 
Dendrimers: Implications for Oral Drug Delivery.......................135 
 
5.1 Introduction.................................................135 
 
 x 
 
5.2 Materials and Methods.........................................136 
5.2.1 Materials................................................136 
5.2.2 Conjugation of mPEG750 to PAMAM Dendrimers................137 
5.2.3 Characterization of PEGylated G3.5 and G4.5 Dendrimers..........139 
5.2.4 Synthesis of Radiolabeled Dendrimers.........................140 
5.2.5 Caco-2 Cel Culture.......................................140 
5.2.6 Cytotoxicity Asay........................................140 
5.2.7 Celular Uptake Studies.....................................140 
5.2.8 Transepithelial Permeability Asesment........................141 
5.2.9 Ocludin Staining.........................................142 
5.3 Results.....................................................144 
5.3.1 Synthesis and Characterization of PEGylated Anionic PAMAM 
Dendrimers..................................................144 
5.3.2 Short-Term Cytotoxicity....................................146 
5.3.3 Celular Uptake...........................................150 
5.3.4 Transepithelial Transport...................................150 
5.3.5 Tight Junction Opening Monitored by Ocludin Staining...........155 
5.4 Discussion..................................................155 
5.5 Conclusion..................................................160 
 
Chapter 6 : Conclusions and Future Directions.........................162 
 
6.1 Conclusions.................................................162 
6.2 Future Directions.............................................165 
 
Apendix 1: Visualization of Intracelular Trafficking of G3.5 Dendrimers and 
Transferin in Caco-2 Cels.........................................168 
 
Apendix 2: Quantification of SN38 by High Presure Liquid Chromatography
...............................................................171 
 
Apendix 3: Quantification of PEG Content of PAMAM G3.5 and G4.5-PEG 
Conjugates by Proton Nuclear Magnetic Resonance.....................178 
 
Bibliography....................................................183 
 
 
 
 
 
 
 xi 
 
List of Tables 
Table 2.1. Physiology of the Gastrointestinal Tract.........................16 
 
Table 2.2. Physical Properties of PAMAM Dendrimers......................54 
 
Table 3.1. Endocytosis Inhibitor Concentration and % Cel Viability in Caco-2 Cels
............................................................86 
 
Table 3.2. Percent Uptake of G3.5-OG Dendrimers and Control Ligands in Caco-2 
Cels in the Presence of Endocytosis Inhibitors.........................88 
 
Table 5.1. Characteristics of PAMAM Dendrimer-PEG Conjugates...........145 
 
Table A2.1. HPLC Gradient Method for SN38 Detection...................172 
 
Table A3.1. Quantification of PEG Conjugation to G3.5 and G4.5 Dendrimers by 
1
H 
NMR.......................................................180 
 
 xii 
 
List of Figures 
 
Figure 2.1. Design and Optimization of Polymer-Drug Conjugates.............10 
 
Figure 2.2. Major Proteins Involved in Tight Junctions......................19 
 
Figure 2.3. Mechanisms of Transport through the Intestinal Barier............22 
 
Figure 2.4. Clathrin- and Caveolin-Mediated Endocytosis and Transcytosis.......27 
 
Figure 2.5. The Biopharmaceutics Clasification System.....................29 
 
Figure 2.6. Transwel
? 
System used in the Caco-2 Monolayer Permeability Asay..34 
 
Figure 2.7. Dendrimers as Multifunctional Nanocariers.....................48 
 
Figure 2.8. Divergent and Convergent Syntheses of Dendrimers...............50 
 
Figure 2.9. Chemical Structures of Common Dendrimers....................52 
 
Figure 2.10. Conversion of Irinotecan to SN38 by Carboxylesterase and Equilibrium 
of Carboxylate and Lactone forms of SN38 and Irinotecan...............71 
 
Figure 3.1. Intracelular Traficking of G3.5-OG Dendrimers over Time in Caco-2 
Cels.........................................................91 
 
Figure 3.2. Intracelular Traficking of Transferin-AF488 over Time in Caco-2 Cels
............................................................92 
 
Figure 3.3. Percent Transport of G3.5-OG Dendrimers across Caco-2 Monolayers in 
the Presence of Endocytosis Inhibitors or at 4?C.......................94 
 
Figure 3.4. Visualization of G3.5-OG Dendrimer Interaction with Caco-2 
Monolayers...................................................95 
 
Figure 3.5. Ocludin Staining in the Presence and Absence of G3.5-OG Dendrimers 
in Caco-2 Cels Treated with HBS or Dynasore.......................97 
 
Figure 3.6. Quantification of Ocludin Staining............................98 
 
Figure 3.7. Mechanisms of Dendrimer Transport Across Caco-2 Cel Monolayers.103 
 
Figure 4.1. Conjugation of SN38 to G3.5 Dendrimers via Glycine and ??Alanine 
Linkers.....................................................108 
 
 
 xii 
 
Figure 4.2. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in 
Simulated Stomach Conditions for 6 hours..........................115 
 
Figure 4.3. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in 
Simulated Intestinal Conditions for 24 hours.........................116 
 
Figure 4.4. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in 
Simulated Liver Conditions for 48 hours............................117 
 
Figure 4.5. Caco-2 Cel Viability after Treatment for 2 hours with G3.5 Dendrimers, 
G3.5-SN38 Conjugates and SN38.................................120 
 
Figure 4.6. Caco-2 Cel Viability 24 hours after 2-hour Treatment with G3.5, G3.5-
SN38 Conjugates and SN38......................................122 
 
Figure 4.7. Equivalent SN38 Flux across Diferentiated Caco-2 Monolayers Treated 
with G3.5-SN38 Conjugates and SN38.............................123 
 
Figure 4.8. Celular Uptake of G3.5-SN38 Conjugates and Fre SN38 in 
Diferentiated Caco-2 Monolayers after 2-hour Treatment on the Apical Side.
...........................................................126 
 
Figure 4.9. IC
50
 Curves of SN38, G3.5-Gly-SN38 and G3.5-?Ala-SN38 in HT-29 
Cels.......................................................128 
 
Figure 5.1. PEGylation of G3.5 Dendrimer with mPEG750..................138 
 
Figure 5.2. Zeta Potential of PEGylated G3.5 and G4.5 PAMAM Dendrimers....147 
 
Figure 5.3. Caco-2 Cel Viability in the Presence of G3.5 Native and PEGylated 
Dendrimers after a 2-hour Incubation Time..........................148 
 
Figure 5.4. Caco-2 Cel Viability in the Presence of G4.5 Native and PEGylated 
Dendrimers after a 2-hour Incubation Time..........................149 
 
Figure 5.5. Uptake of G3.5 Native and Diferentialy PEGylated Dendrimers....151 
 
Figure 5.6. Uptake of G4.5 Native and Diferentialy PEGylated Dendrimers....152 
 
Figure 5.7. Apparent Permeability of G3.5 Native and Diferentialy PEGylated 
Dendrimers...................................................153 
 
Figure 5.8. Apparent Permeability of G4.5 Native and Diferentialy PEGylated 
Dendrimers...................................................154 
 
 
 xiv 
 
Figure 5.9. Staining of the Tight Junction Protein Ocludin in the Presence and 
Absence of Dendrimers Visualized by Confocal Microscopy.............156 
 
Figure 5.10. Quantification of Ocludin Staining in the Presence and Absence of 
Dendrimers...................................................157 
 
Figure A1.1. Visualization of G3.5 Dendrimer Traficking over Time in Caco-2 Cels 
by Confocal Microscopy........................................169 
 
Figure A1.2. Visualization of Transferin Traficking over Time in Caco-2 Cels by 
Confocal Microscopy...........................................170 
 
Figure A2.1. Typical HPLC Elution Profile of SN38.......................173 
 
Figure A2.2. Standard Curve Elution Profiles.............................175 
 
Figure A2.3. HPLC Standard Curve Comparing SN38 Concentration and Peak Area
...........................................................176 
 
Figure A2.4. Precision of HPLC Detection..............................177 
 
Figure A3.1. NMR Spectra of G3.5-PEG Conjugates......................181 
 
Figure A3.2. NMR spectra of G4.5-PEG Conjugates.......................182 
 
 xv 
 
Abbreviations 
 
5FU 5 fluorouracil 
ABC ATP binding casete 
? Angstrom 
AF Alexa Fluor 
ATP Adenosine triphosphate 
AUC Area under the curve 
?Ala Beta-alanine 
?CA Bicinchoninic acid  
BCS Biopharmaceutics clasification system 
BOP Benzotriazole-1-yl-oxy-tris-(dimethylamino)-
phosphoniumhexafluorophosphate 
BSA Bovine serum albumin 
CH Carbohydrazide 
CME Caveolin-mediated endocytosis 
CPT Camptothecin 
CPT-11 Irinotecan 
CT Computed tomography 
CYP Cytochrome P450 
D
2
O Deuterated water 
Da Dalton 
DACH Diaminocyclohexane 
 
 xvi 
 
DAPI 4?,6-diamidino-2-phenylindole 
DS Drug delivery system 
DI Deionized 
DLS Dynamic light scatering 
DMEM 
Dulbeco?s modified eagle?s medium 
DMSO Dimethyl sulfoxide 
DNA Deoxyribonucleic acid 
DPBS Dulbeco?s phosphate buffered saline 
DYN Dynasore 
EDC 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide 
EDTA Ethyelendiamine tetracetic acid 
EA-1 Early endosome antigen-1 
EGFR Epidermal growth factor receptor 
EPR Enhanced permeability and retention 
FBS Fetal bovine serum 
FDA Food and Drug Administration 
FIL Filipin 
FITC Fluorescein isothiocyanate 
FOBT Fecal occult blood tests 
FOLFIRI 5FU/Leucovorin + Irinotecan chemotherapy regiment 
FOLFOX 5FU/Leucovorin + Oxaliplatin chemotherapy regiment 
FPLC Fast protein liquid chromatography 
 
 xvii 
 
G Generation 
GEN Genistein 
GFLG Glyine-phenylalanine-leucine-glycine 
Gly Glycine 
GTPase Guanosine triphosphate hydrolytic enzyme 
HBS Hank?s balanced salt solution 
HCl Hydrochloric Acid 
HEPES 
N-(2-hydroxyelthyl)piperazine-N?-179 (2 ethanesulfonic acid) 
hemisodium salt 
HPLC 
High presure liquid chromatography 
HPMA N-(2-hydroxypropyl)methacrylamide 
IC
50
 Half maximal inhibitory concentration 
IgG Imunoglobulin G 
IU International unit 
IV Intravenous 
JAM Junction adhesion molecule 
LAMP-1 Lysosome-asociated membrane protein-1 
LDH Lactate dehydrogenase 
LY Lucifer yelow carbohydrazide 
MALS Multi-angle laser light scatering 
MAPK Mitogen activated protein kinase 
M-cels Microfold cels 
 
 xvii 
 
MDC Monodansyl cadaverine 
MDR Multidrug resistant protein 
MLCK Myosin light chain kinase 
mPEG Methoxy poly (ethylene glycol) 
MS Mas spectometry 
MW Molecular weight 
MWCO Molecular weight cut off 
NHS N-Hydroxysuccinimide 
NMR Nuclear magnetic resonance 
OG Oregon gren 
PAMAM Poly (amido amine) 
PAMPA Paralel artificial membrane asay 
PAO Phenylarsine oxide 
P
ap
 Apparent permeability 
PBS Phosphate buffered saline 
PD10 Protein desalting column 
PEG Poly (ethylene glycol) 
PG 
Poly (L-glutamic acid) 
P-gp P-glycoprotein 
PG-TXL 
Poly (L-glutamic acid)- Paclitaxel conjugate 
PI Poly (propylene imine) 
PVDF Poly vinylidene fluoride 
 
 xix 
 
QELS Quasi-elastic light scatering  
QSAR Quantitative structure activity relationship 
QSPR Quantitative structure property relationship 
RI Refractive index 
RME Receptor-mediated endocytosis 
RNA Ribonucleic acid 
RPM Rotations per minute 
SEC Size exclusion chromatography 
SGF Simulated gastric fluid 
SIF Simulated intestinal fluid 
SLC Solute carier 
SN38 7-Ethyl-10-hydroxy-camptothecin 
SNAP S-nitroso-N-acetyl-DL-penicilamine 
TER Transepithelial electrical resistance 
TEM Transmision electron microscopy 
TJ Tight junction 
TFA Trifluoroacetic acid 
TNM Tumor-lymph node-metastasis clasification system 
TSP-d
4
 3-(trimethylsilyl) propionic-2,2,3,3-d
4
 acid  
UDPGA Uridine diphosphoglucuronic acid 
UDPGT Uridine diphosphate glucuronyltransferase 
UV Ultraviolet 
VEGF Vascular endothelial growth factor 
 
 xx 
 
WST-1 Water soluble tetrazolium salt 
Z-Arg-AMC Benzyloxycarbonyl-L-arginine-4-methylcoumaryl-7-amide 
ZO Zonulla occludens 
Zot Zona occludins toxin 
 
1 
 
 
 
Chapter 1 : Introduction 
 
1.1  Introduction 
1.1.1 Polymer Therapeutics 
Carier-based drug delivery strategies have been used to improve the therapeutic 
profile of smal molecule drugs by enhancing uptake at the site of action and minimizing 
off-target efects. In particular, water-soluble polymer-based drug delivery systems can 
improve the eficacy of therapeuticaly active compounds with intrinsicaly poor water 
solubility and high toxicity [1]. In the field of nanomedicine, cariers such as polymeric 
nanoparticles and liposomes have also shown promise in altering the biodistribution of 
drugs, improving eficacy and enhancing intracelular acumulation [2]. Importantly, 
targeting moieties and imaging agents can be conjugated to delivery vehicles, making 
such cariers ideal for multifunctional drug delivery [3].  
Chemotherapy drugs are promising candidates for drug delivery strategies 
because they are often plagued by poor water solubility and dose-limiting toxicities. 
Conjugation of chemotherapeutics to linear water-soluble polymers such as poly 
(ethylene glycol) (PEG) and N-(2-hydroxypropyl)methacrylamide (HPMA) copolymers 
can improve their therapeutic profile by increasing the drug concentration at the tumor 
site due to enhanced permeability and retention in the leaky tumor vasculature [4]. In 
addition, targeting moieties can be conjugated to the polymer backbone to further 
enhance tumor-specific uptake. Finaly, drug linkers can be designed to release the drug 
from the polymer in the tumor environment in response to pH changes or specific 
 
2 
 
 
enzymes [5]. Increased tumor acumulation and selective release improve the overal 
therapeutic profile compared to the fre drug by increasing the concentration of drug at 
the site of action and reducing side efects caused by drug acumulation in of-target 
tisues [6]. 
 
1.1.2 Oral Drug Delivery 
Due to their low water solubility and poor bioavailability, chemotherapy agents 
are traditionaly administered intravenously, requiring recurring hospital visits and 
significant direct and indirect costs to the patient [7]. While conjugation to water-soluble 
polymers can improve their solubility, the large size of these macromolecular constructs 
necesitates intravenous administration.  Because of the strong patient preference for oral 
therapy and the numerous advantages of polymeric drug delivery systems, there is a 
significant need for development of oraly administrable polymer-drug conjugates that 
can improve the oral bioavailabilty of the drug as wel as target the drug to the tumor site.  
Advantages of oral chemotherapy include the convenience of at-home administration, 
reduction of costs asociated with hospital procedures and with transportation to and from 
the treatment center, a more flexible dosing regimen and a decreased burden on hospitals 
and the healthcare system [8, 9]. 
Oral delivery is chalenging because systems must have appropriate solubility, 
stability in the gastrointestinal pH and enzymatic environment, and the ability to 
permeate the epithelial barier of the gut. The mucosa of the intestinal tract is composed 
of polarized enterocytes that form a tight barier to transport, preventing the difusion of 
many molecules based on size, shape and charge [10]. In addition, compounds must have 
 
3 
 
 
some degre of lipophilicity to be transported via pasive difusion, otherwise, they are 
relegated to the paracelular pathway which is generaly limited to smal hydrophilic 
molecules of 100-200 Da [11]. Therefore, oral delivery of polymer-drug conjugates 
while convenient and beneficial to the patient is also chalenging.  
 
1.1.3 Poly (amido amine) Dendrimers 
Dendrimers, a clas of highly branched polymers, have been shown to be efective 
drug delivery vehicles due to their unique physical properties including near 
monodispersity and nanoscopic size [12, 13]. With each increase in dendrimer 
generation, the diameter increases linearly while the number of surface groups increases 
exponentialy. This creates a high density of surface groups that can be conjugated to 
drug molecules, targeting moieties, and imaging agents, making dendrimers a versatile 
drug delivery platform [14]. Dendrimers are more compact than traditional linear 
polymers and thus show potential in oral drug delivery.  Specificaly, poly (amido 
amine) or PAMAM dendrimers have shown promise as oral drug delivery cariers due to 
their nanoscale size and high surface charge density. 
Originaly developed by Tomalia in 1979, PAMAM dendrimers have an ethylene 
diamine core, an amido amine branching structure and are commercialy available as 
cationic ?full? generations (G1, G2, etc.) with amine-terminated branches, and anionic 
?half? generations (G1.5, G2.5, etc.) with carboxylic acid terminated branches. Previous 
studies indicate that PAMAM dendrimers in a specified size and charge window can 
efectively cross the intestinal epithelial barier, making them suitable as drug delivery 
cariers for poorly water-soluble chemotherapeutics [15-20]. While it has been firmly 
 
4 
 
 
established that generation, charge and surface chemistry influence PAMAM dendrimer 
transport across the epithelial barier, more needs to be done to investigate how surface 
modification and drug conjugation impact the degre and mechanism of dendrimer 
transport. In addition, chemical linkers that promote stability of the dendrimer-drug 
conjugate in the gastrointestinal milieu while favoring release of drug in the tumor 
environment must be developed to achieve a functional delivery system.  
 
1.1.4 SN38 
Because of its low water solubility and poor bioavailability, SN38 (7-ethyl-10-
hydroxy-camptothecin) is an ideal candidate for polymeric delivery strategies. SN38 is 
the active metabolite of Irinotecan, a water-soluble camptothecin analogue that is 
commonly used for treatment of metastatic colorectal cancer [21]. With more than 
140,000 new cases of colorectal cancer each year and over 51,000 deaths projected in 
2010, there is a significant need for novel, targeted therapies for colorectal cancer [22]. 
The most common site of colorectal cancer metastasis is the liver, with over 20% of 
patients presenting liver metastases at the time of diagnosis and 50% in the duration of 
the disease [23]. These liver metastases are responsible for two-thirds of al colorectal 
cancer deaths, indicating the grave need for novel treatments of this condition. SN38 
works by inhibition of topoisomerase-1, and shows 100-1000 fold greater in vitro activity 
than Irinotecan. However, it has had limited clinical succes due to poor water solubility 
and adverse gastrointestinal efects. Therefore a delivery system that improves SN38 
water solubility, decreases gastrointestinal toxicity and targets SN38 to liver metastases 
 
5 
 
 
has the potential to significantly improve its therapeutic profile and increase the eficacy 
of colorectal cancer treatment. 
 
1.2 Specific Aims 
This disertation seks to investigate the mechanisms of transepithelial transport 
of anionic PAMAM dendrimers and their conjugates with SN38 and PEG, with the long-
term goal of delivering SN38 oraly to colorectal cancer metastases in the liver. The 
global hypothesis of the disertation is that anionic PAMAM dendrimers can be used to 
improve the oral bioavailability of SN38 for the treatment of colorectal hepatic 
metastases and that the degre and mechanism of transepithelial transport of PAMAM 
dendrimers wil be impacted by surface modification and drug loading. This global 
hypothesis was investigated by the following thre Specific Aims: 
1. To determine the transepithelial transport mechanisms of anionic G3.5 
PAMAM dendrimers across Caco-2 monolayers as a model of the intestinal epithelial 
barier. 
2. To evaluate the toxicity, transport and uptake of G3.5-SN38 conjugates in 
Caco-2 cel monolayers and the stability of the conjugates in simulated gastrointestinal 
and liver environments. 
3. To investigate the impact of surface modification of anionic PAMAM 
dendrimers with PEG on cytotoxicity, uptake, and transepithelial transport in Caco-2 
cels. 
 
 
 
6 
 
 
These Specific Aims are designed to test the following thre hypotheses: 
1. Transepithelial transport of anionic PAMAM dendrimers occurs via several 
pathways including transcytosis, governed by specific endocytic mechanisms, and 
paracelular transport, driven by dendrimer-mediated tight junction opening. 
2. Conjugation of SN38 to G3.5 dendrimers via a glycine or ?-alanine peptide 
linker wil increase SN38 permeability while reducing intestinal toxicity. The conjugates 
wil be stable in the gastrointestinal environment and wil release SN38 in the presence of 
liver carboxylesterase. 
3. Conjugation of PEG to anionic dendrimers wil influence uptake and transport 
across Caco-2 cel monolayers. 
 
1.3 Scope and Organization 
 Chapter 2 provides background on the benefits and chalenges of oral drug 
delivery as wel as the current strategies used to improve oral bioavailabilty and the 
laboratory methods used to predict oral absorption. Chapter 2 also offers background on 
dendrimers, a detailed description of PAMAM dendrimers and a literature review of 
current knowledge on these polymers in the context of oral drug delivery [24]. In 
addition, colorectal cancer and current limitations and delivery strategies for SN38 are 
reviewed in this Chapter. Chapter 3 describes the mechanisms of transport, uptake, 
intracelular traficking and tight junction opening of native, anionic G3.5 PAMAM 
dendrimers in Caco-2 cels [25]. In particular, Chapter 3 explores the role of endocytosis 
in dendrimer tight junction opening.  In Chapter 4, G3.5-SN38 conjugates are evaluated 
for their potential in oral drug delivery [26]. Specificaly, their intestinal toxicity, 
 
7 
 
 
stability in simulated gastrointestinal environment and in the presence of liver 
carboxylesterase as wel as their celular uptake and transport across Caco-2 monolayers 
and eficacy in HT-29 cels are investigated. Chapter 5 describes the evaluation of PEG 
surface modification of G3.5 and G4.5 dendrimers in the context of oral drug delivery, 
specificaly examining the impact of PEGylation on dendrimer cytotoxicity, uptake, 
transport and tight junction modulation in Caco-2 cel monolayers [27]. Finaly, Chapter 
6 summarizes the major findings of this disertation and suggests future directions for 
this research.  
 
8 
 
 
 
Chapter 2 : Background 
 
2.1 Introduction 
The objective of drug delivery science is to control the spatial and temporal 
distribution of a drug molecule in the body in order to improve therapeutic efect and/or 
reduce toxicity [28]. Drug delivery technologies can alter a drug?s solubility, absorption, 
metabolism, elimination and biodistribution in vivo. One of the most common drug 
delivery strategies is to atach therapeutic compounds to ?cariers? by physical and 
chemical mechanisms, with the overal goal of modifying the drug?s biodistribution and 
controlling release from the carier. Examples of drug delivery cariers include 
nanoparticles, miceles, liposomes and water-soluble polymers [29-31]. These advanced 
?formulation strategies,? can be modified to yield an optimized delivery system for the 
cargo, which can improve the therapeutic index compared to the fre drug. In this 
disertation we wil explore poly (amido amine) (PAMAM) dendrimers as oral drug 
delivery vehicles for anti-cancer therapeutics. This chapter wil provide the pertinent 
background on the concepts of polymeric drug delivery, oral drug delivery, dendrimers, 
colorectal cancer and SN38 as wel as a literature review on the current state of 
knowledge on poly (amido amine) (PAMAM) dendrimers as oral drug delivery cariers. 
 
2.2 Polymeric Drug Delivery 
 Polymeric systems are among the most widely used drug cariers. Although 
synthetic polymers were used as early as the 1830?s in the field of materials science [32], 
 
9 
 
 
water-soluble polymers were not conceptualized for use as targeted drug delivery systems 
until 1975 when Ringdsorf first wrote of ?pharmacologicaly active polymers? [33]. In 
this concept, a drug molecule is covalently conjugated to a polymeric carier through a 
selectively degradable chemical linker, which releases the drug at the site of action. 
Targeting moieties are conjugated to the polymer in order to enhance the acumulation of 
the polymer-drug conjugate at the site of action, maximizing eficacy and minimizing 
non-specific toxicities.  Several properties of the polymer backbone, including chemical 
composition, molecular weight and architecture, can be tailored for a given drug delivery 
application. In addition, the number of drug molecules, choice of chemical linker and 
type of targeting moiety can be modified to create an optimized polymeric drug delivery 
system [6]. A schematic of polymer-drug conjugate design is shown in Figure 2.1. 
 
2.2.1 Therapeutic Advantages of Polymer-Drug Conjugates in Chemotherapy 
 One of the most common applications of polymer-drug conjugates is in the delivery 
of anti-cancer therapeutics. Because chemotherapy drugs have high potency but low 
selectivity, their use is severely limited by unintended toxic efects to healthy tisues 
[34]. Conjugation of chemotherapeutics to polymeric cariers has the capacity to 
improve their therapeutic index by enhanced drug acumulation at the tumor site with 
fewer off-target efects [35]. One of the most widely studied properties of polymeric 
drug delivery systems is the enhanced permeability and retention (EPR) efect. The EPR 
efect, coined by Hiroshi Maeda, describes the propensity of macromolecules to 
acumulate in solid tumors [36]. Because of the rapid angiogenesis asociated with solid  
 
10 
 
 
 
Figure 2.1. Design and Optimization of Polymer-Drug Conjugates. Several parameters 
can be optimized when designing polymer-drug conjugates including polymer backbone 
composition, molecular weight and architecture, drug cargo identity and amount, type of 
chemical linker and nature of targeting moiety. 
 
 
11 
 
 
tumor formation, tumors have dense and leaky vasculature. This leaky tumor 
vasculature, coupled with the long half-life of macromolecules, leads to higher 
acumulation of polymer-drug conjugates in the tumor compared to other tisues that 
have leser fenestrations in endothelial cel layers. Thus, pasive targeting by the EPR 
efect alows for acumulation of the drug at the tumor site, improving the therapeutic 
index compared to the fre drug. Moreover, conjugation of antibodies and ligands to the 
polymer backbone that target specific receptors over-expresed on tumor cels can 
enhance tumor acumulation [37]. The ability of polymers to serve as backbones for the 
conjugation of drugs, targeting moieties and imaging agents makes them powerful 
multifunctional delivery systems. 
 In addition to promoting drug localization at the tumor site, polymer-drug 
conjugates can also be designed to optimize drug release at the site of action [6]. There 
are several chemical linkers that are cleaved by select enzymes. For example, the amino 
acid sequence Gly-Phe-Leu-Gly (GFLG) has been shown to be selectively cleaved by 
cathepsin B, which is over-expresed in the lysosomes of tumor cels [38]. This linker 
can be used to optimize drug release in the tumor environment, thus minimizing side 
efects. A commonly used linker in the context of colonic delivery is the azo spacer [39]. 
Azo spacers are cleaved by azo-reductase enzymes produced by bacteria in the colon, 
confining drug release to the colonic environment. Therefore, appropriate drug linker 
selection alows for specific release in the target site and further enhances the therapeutic 
efect of polymer-drug conjugates. 
 
 
12 
 
 
2.2.2 Polymer-Drug Conjugates Currently in Clinical Trials 
 There are several promising polymer-drug conjugates that are currently being 
evaluated in clinical trials. Two HPMA (N-(2-hydroxypropyl) methacrylamide)-based 
systems are currently in clinical trials for the delivery of Doxorubicin and Platinate [40]. 
FCE 28068 contains Doxorubicin conjugated to 30 kDa HPMA copolymer via a GFLG 
spacer. FCE28068 showed 2-5 fold les toxicity than fre Doxorubicin with tumor 
responses documented in breast, non-smal cel lung cancers and hepatocelular 
carcinoma [41, 42]. AP5346 is also in Phase I clinical trials and contains DACH 
(diaminocyclohexane) platinum, the active form of the marketed drug Oxaliplatin, 
conjugated to 25 kDa HPMA copolymer via a pH sensitive linker. AP5346 showed 
promising pre-clinical activity in 11 diferent tumor models and has shown both safety 
and eficacy in ovarian cancer in clinical trials [43].  In addition, poly (ethylene glycol) 
(PEG) is currently being evaluated in clinical trials for the administration of 
Camptothecin, SN38, Irinotecan, Docetaxel and Paclitaxel [44]. 
The most advanced polymer-drug conjugate currently in late stage Phase II 
clinical trials is PG-TXL, a conjugate of poly (L-glutamic acid) and Paclitaxel [45]. While 
this polymeric conjugate was originaly synthesized to improve the solubility of 
Paclitaxel, preclinical studies showed enhanced tumor acumulation of PG-TXL compared 
to fre Paclitaxel, contributing to its improved eficacy [46]. In addition, cathepsin B, 
which is over expresed in tumor cels, was found to proteolyticaly degrade the PG 
backbone, alowing for targeted release of Paclitaxel at the tumor site. Clinical trials in 
several diferent tumor types demonstrated that PG-TXL was as efective as Paclitaxel, 
but it did not demonstrate superior eficacy as has been shown in the preclinical studies. 
 
13 
 
 
However, PG-TXL had several advantages over Paclitaxel, namely a reduction in the 
incidence of hair loss, nausea and hypersensitivity reactions, which are often limiting 
factors for chemotherapy. These trials demonstrate the promise for the clinical 
aplication of polymer-drug conjugates, as wel as chalenges they face in translation to 
the clinic. 
 
2.3 Administration of Drugs via the Oral Route 
Administration of drugs via the oral route is prefered by patients owing to its 
convenience and comfort, generaly leading to higher patient compliance compared to 
parenteral treatments and self-administration routes such as transdermal and inhalation. 
In addition, oral drug administration provides a more flexible dosing regimen and a lower 
burden on hospitals and the healthcare system [8, 9]. Finaly, in diseases such as cancer, 
oral therapy minimizes treatment costs due to hospital procedures as wel as indirect costs 
asociated with lost time at work, transportation to and from the treatment center and 
additional childcare [7].  
Oral delivery of therapeutics is chalenging because drugs must have appropriate 
solubility, stability in the gastrointestinal pH and enzyme environment, and the ability to 
permeate the epithelial barier of the gut [47]. Unfortunately, many current 
chemotherapy drugs are poorly water soluble with low oral bioavailability, necesitating 
administration by the intravenous route, which requires recurring hospital visits and 
significant direct and indirect costs to the patient. Because of the strong patient 
preference for oral therapy, much research is being done to find alternative oraly 
administrable chemotherapy drugs that are as efective as traditional intravenous 
 
14 
 
 
therapies and preferably with lower off-target toxicities [48]. This section wil provide 
the necesary background on oral drug administration. 
 
2.3.1 Physiology of the Gastrointestinal Tract 
2.3.1.1 Compartments and Functions 
 The gastrointestinal tract is composed of several unique sections which each have 
a distinct pH, residence time, absorptive area and group of enzymes which must be 
considered for oral drug delivery [49]. The gastrointestinal tract begins with the oral 
cavity where the drug comes in contact with saliva. With the exception of specialy 
designed sublingual and buccal formulations, negligible drug absorption occurs in the 
oral cavity. After pasage down the esophagus, the drug encounters the stomach. The 
stomach is lined with approximately 3.5 milion gastric pits, which contain cels that 
secret mucus and gastric juice. This gastric juice contains hydrochloric acid, which 
maintains the acidic pH of the stomach, as wel as pepsin. Thus, any drug that enters the 
stomach must be resistant to acidic conditions and proteolysis by pepsin to prevent 
premature degradation. The contents of the stomach are emptied at regular intervals, 
which are dependent on the fasted or fed state of the individual, amongst other factors 
including disease state or drug treatment. After these pulsatile waves, the entire content 
of the stomach is propulsed into the smal intestine [49]. 
 The smal intestine serves as the major absorptive site in digestion and contains 
thre distinct sections: the duodenum, jejunum and ileum. Absorption is favored by a 
large surface area due to mucosal folds and vilous formation thereby maximizing the 
number of intestinal absorptive cels that come into contact with the luminal contents. In 
 
15 
 
 
addition, the large blood flow to the smal intestine serves to maintain the concentration 
gradient across the intestinal barier, again, enhancing absorption of nutrients and drug 
molecules. Biliary secretion from the galbladder into the duodenum enhances absorption 
of lipids by formation of mixed miceles, and lipid digestion is promoted by lipase 
enzymes secreted by the pancreas. In addition, Brunners glands, located in the 
duodenum, secrete bicarbonate and neutralize the acidic stomach contents. This neutral 
pH promotes enzymatic activity in the smal intestine and influences the charged state of 
drug molecules, which can enhance absorption. Food can also influence the absorption 
of drugs by binding to the drugs or competing for intestinal transporters. Another 
potential barier to absorption is the thick mucus layer that lines the intestinal epithelium, 
which creates a formidable barier to drug difusion. In order to succesfully penetrate 
through mucus, drugs must be smal enough to avoid encumbrance by the dense mucin 
fiber mesh and avoid adhesion to the fibers [50]. Post-epithelial structures include the 
intraepithelial lymphocytes, basement membrane and lamina propria mononuclear cels. 
These specialized elements of the gut-asociated lymphoid tisue contribute to the 
mucosal imune response and present antigens to the lymphatic system, but do not 
present a significant barier to transport [51].  Once drugs and nutrients are absorbed in 
the smal intestine, they are transported through the hepatic portal vein and undergo first-
pas metabolism in the liver before entering systemic circulation [49]. 
 Materials that are not absorbed in the smal intestine are transfered to the large 
intestine for further procesing. The large intestine is populated by a wide range of 
bacteria that produce enzymes to further break down foods and drugs. Finaly, the 
material is cleared from the large intestine and is transfered to the colon. Table 2.1 
 
16 
 
 
Table 2.1. Physiology of the Gastrointestinal Tract 
Compartment Absorptive 
Area (m
2
) 
Residence time Fasted pH Enzymes 
Oral cavity - - 7.0 Amylase 
Esophagus 0.02 3.5 s 6.0-7.0 - 
Stomach 0.1 2-8 h 1.3-2.1 Pepsin, lipase 
Smal Intestine: 
-Duodenum 
-Jejunum 
-Ileum 
200  
5 min 
1-2 h 
2-3 h 
 
5.5-6.5 
6.1-7.1 
7.0-8.0 
Peptidases, 
lipases, 
amylase 
Large Intestine 0.3 15-48 h - Diverse 
bacterial 
enzymes 
Colon 0.25 - 8.0 - 
Rectum - - 7.0 - 
(Adapted from [49]) 
 
17 
 
 
summarizes the major compartments of the gastrointestinal tract as wel as the absorptive 
area, residence time, pH and enzymatic activity in each compartment [49].  
 
2.3.1.2 Intestinal Epithelial Barrier 
 The intestinal epithelial barier is the main site of intestinal absorption of drugs 
and nutrients. This barier forms a boundary betwen the ?external environment? and the 
?internal environment? and is designed to alow for the absorption of nutrients while 
preventing transport of ?undesirable? components such as bacteria or toxins. While 
epithelial bariers like the skin are designed to be virtualy impermeable, the intestinal 
epithelial barier must be selectively permeable to permit fluid exchange and nutrient 
absorption, and is therefore relatively leaky in comparison [52]. The intestinal epithelium 
contains a monolayer of columnar absorptive cels, or enterocytes, that are oriented with 
the ?apical membrane? facing the intestinal lumen and the ?basolateral membrane? facing 
the serosal side. The apical membrane contains a brush border of microvili that, together 
with mucosal and vilous folds, are responsible for the large absorptive surface area of the 
smal intestine [53]. In addition to the absorptive cels, the intestinal barier also contains 
specialized M-cels (microfold cels) located in Peyer?s patches of the gut-asociated 
lymphoid tisue. These cels continuously sample antigens in the intestine and present 
them to the mucosal lymphoid tisue for initiation of the imune response [54]. M-cels 
have been popular targets for viral drug delivery vectors as wel as nanoparticles. 
Finaly, the intestinal epithelium is covered by a mucus layer, which is secreted by the 
intestinal goblet cels, and prevents large particles from approaching the epithelial barier 
and especialy, protects the sensitive intestinal crypt region. In addition, the mucus 
 
18 
 
 
serves as an unstired layer, slowing the absorption of nutrients by limiting difusion to 
the intestinal cels. The mucus layer can also enhance absorption of compounds produced 
by brush border enzymes by preventing difusion of these molecules back into the 
intestinal lumen [53]. 
 The integrity of the intestinal epithelial barier is maintained by thre types of 
cel-to-cel contacts: tight junctions, adherens junctions and desmosomes [53]. 
Desmosomes and adherens junctions form the cel-to-cel contacts that maintain the 
integrity of the barier and alow for communication betwen the intestinal cels. Tight 
junctions are formed at the apical side of the epithelial barier and serve to regulate the 
pasage of hydrophilic molecules betwen the intestinal cels. The structure of the tight 
junctions as wel as the mechanism by which they regulate paracelular permeability is 
discussed in detail in Section 2.3.1.3. A schematic showing the major proteins involved 
in cel-to-cel junctions is shown in Figure 2.2. Genetic mutations that compromise 
intestinal barier function are responsible for the pathology of several intestinal diseases 
including Crohn?s disease, ulcerative colitis and Celiac disease [52]. These conditions 
ilustrate the importance of preserving the integrity of the gastrointestinal barier to 
maintain healthy function.  
 
2.3.1.3 Tight Junction Biology: Structure and Function 
 Tight junctions regulate the openings betwen the epithelial cels, alowing for 
paracelular transport. Although the function of tight junctions has been known for 
decades, the precise protein composition and function of tight junctions was not 
discovered until the 1990?s [55]. Ocludin, a 65 kDa transmembrane protein, was the 
 
19 
 
 
 
Figure 2.2. Major Proteins Involved in Tight Junctions, Adherens Junctions and 
Desmosomes betwen the Intestinal Epithelial Cels. In the tight junctions MLCK is 
myosin light chain kinase and ZO-1 is zonulla occludens 1. (Adapted from [53].) 
 
20 
 
 
first tight junction protein to be isolated, and was reported by Furuse and colleagues in 
1993 [56]. The cytosolic face of occludin is thought to interact with several zonulla 
occludens (ZO) as wel as actin [57]. Subsequently, the same group reported the 
involvement of claudins in tight junctions in 1998 [58, 59]. There have been 18 diferent 
claudins reported with specific claudins being expresed in certain cel and tisue types 
[57]. Like occludin, claudins also interact with ZO proteins. Both occludin and claudin 
were examined in knockout studies and were found to be critical for tight junction 
function. Junction adhesion molecule (JAM) has also been implicated in tight junctions 
[60]. Finaly, tricelulin, a 64 kDa polypeptide discovered in 2005, is a necesary 
component of tight junctions in areas where thre cels come into contact [61]. In 
addition to these tight junctional components, actin and myosin form a perijunctional 
actomyosin ring that plays a role in the contraction of tight junctional proteins upon 
stimulation [53]. 
 Tight junction modulation is regulated by several complex signaling pathways 
including those controlled by protein kinases, rho kinases, myosin light chain kinase 
(MLCK) and mitogen activated protein kinase (MAPK) [62]. These pathways can be 
activated in normal physiological conditions and exploited by tight junction modulating 
drugs. One of the most common instances of tight junction opening occurs during 
nutrient absorption. Apical co-transport of Na
+
 and glucose activates an MLCK-
dependent signaling pathway that causes condensation of the perijunctional actomyosin 
ring, leading to opening of the tight junctions [63]. This alows for enhanced water and 
nutrient transport through the paracelular route. This example ilustrates the 
interconnectednes of transcelular and paracelular fluxes.  Extracelular calcium ions 
 
21 
 
 
(Ca
2+
) are also critical for tight junction function. Ma and co-workers [64] demonstrated 
that the removal of extracelular calcium caused an increase in the paracelular 
permeability and a drop in transepithelial electrical resistance (TER) of Caco-2 
monolayers. In addition, they observed a rapid contraction of actin and myosin filaments 
and movement of ZO-1 and occludin away from the membrane, which was initiated by 
MLCK activation. Addition of Ca
2+
 was found to quickly reverse these efects, 
ilustrating the critical role of extracelular calcium in the maintenance of tight junction 
integrity. 
 
2.3.2 Mechanisms of Transport Across the Intestinal Barrier 
 There are two major pathways of transport across the intestinal barier: 
paracelular transport, which occurs betwen cels, and transcelular transport, which 
occurs across cels. The following four sections discuss paracelular transport and 
transcelular transport, which can occur by pasive difusion, carier-mediated transport 
and endocytosis. Importantly, molecules may follow one or more transport pathways and 
these pathways may be interconnected. Figure 2.3 ilustrates potential mechanisms of 
transport through the intestinal barier. 
 
2.3.2.1 Paracelular Transport 
 Paracelular transport occurs when molecules pas betwen epithelial cels in the 
intestinal barier. This mechanism of transport is driven by the concentration gradient 
and is thus both energy-independent and unsaturable. Because molecules permeate 
through the restricted tight junctions, they must be water-soluble and smal (i.e. MW<200 
 
22 
 
 
 
Figure 2.3. Mechanisms of Transport through the Intestinal Barier. Transport may be 
either paracelular or transcelular. Transcelular transport includes pasive difusion, 
carier-mediated transport, which may be energy-independent (facilitated difusion) or 
energy-dependent (active transport) or endocytosis, which can be fluid phase, adsorptive 
or receptor-mediated. 
 
23 
 
 
Da). Studies of permeability of hydrophilic compounds across Caco-2 monolayers 
suggested a tight junction pore size of approximately 4.5-4.8 ? [65, 66]. Molecules such 
as mannitol, a smal sugar alcohol, and lucifer yelow carbohydrazide, a fluorescent dye, 
are known to permeate across the intestinal barier strictly by the paracelular route and 
thus can be used to monitor changes in paracelular flux [67, 68]. In general, the 
paracelular route is impermeable to macromolecules because they are too large to pas 
through the tight junction pores [69]. 
 
2.3.2.2 Passive Difusion 
 Pasive difusion is the simplest mode of transcelular transport and occurs when 
membrane-permeable molecules are transported across the intestinal barier from a region 
of high concentration on the apical side to a region of lower concentration on the 
basolateral side of the membrane. This mechanism of transport is energy-independent 
and unsaturable. Pasive difusion across the intestinal barier is governed by Fick?s Law 
of Difusion, which states that flux (J) is directly proportional to the concentration 
gradient (dC/dx), the lipid-water partition coeficient (K), the difusion coeficient (D) 
and the area of transport (A): 
dx
dC
DAKJ!= (Eq. 2.1) 
The partition coeficient, a ratio of the compound?s lipid solubility to water solubility, is a 
critical component in determining the overal flux since compounds must be able to 
difuse into the lipid-rich cel membrane in order to be transported via the transcelular 
route. Thus, more lipophilic molecules have higher rates of pasive difusion across the 
intestinal barier. However, there is an upper limit: highly lipophilic drugs tend to be 
 
24 
 
 
insoluble thus preventing their absorption. Since uncharged molecules are always more 
lipid-soluble than charged molecules, the ionization state of a compound at pH of the 
intestine is a critical determinant of its pasive permeability. Many oraly administered 
drug compounds are lipophilic, alowing for pasive difusion across the intestinal 
barier.  
 
2.3.2.3 Carrier-Mediated Transport 
 Carier-mediated transport plays an important role in intestinal transport of 
compounds such as peptides, nucleic acids, monosacharides, organic cations and 
monocarboxylic acids that cannot be transported via the paracelular pathway or pasive 
difusion [70]. In carier-mediated transport, transmembrane proteins move cargo across 
the cel membrane. After the cargo has been released, the transporters resume their initial 
configuration and are available to transport additional cargo. Importantly, because there 
are a limited number of transporters, carier-mediated transport is a saturable proces. 
Carier-mediated transport can occur by facilitated difusion, which is energy-
independent and occurs down the concentration gradient or by active transport, which 
occurs against the concentration gradient and requires and energy source.  Active 
transport can either be driven directly by ATP (ATP binding casete [ABC] 
transporters), or coupled to transport of ions such as H
+
 or Na
+
 (solute carier [SLC] 
transporters). Examples of types of membrane transport proteins include monocarboxylic 
acid transporters, human peptide transports, nucleoside transporters and organic cation 
transporters [70].  
 
25 
 
 
 In addition to influx transporters, eflux transporters can also influence intestinal 
transport. P-glycoprotein (P-gp), a member of the ABC superfamily and the multidrug 
resistance subfamily, is one of the most wel characterized eflux transporters [70]. 
Although it is intended to protect the body from harmful toxins, it also causes eflux of 
many drugs. Oraly administered P-gp substrates often have poor intestinal permeability 
because of this eflux and must be co-administered with P-gp inhibitors to overcome this 
chalenge. Therefore it is important to consider if a drug may be a P-gp substrate when 
examining its intestinal permeability [70]. 
 
2.3.2.4 Endocytosis 
 Endocytosis is a generalized term that refers to the active internalization of 
specific substrates into the cel by membrane engulfment. Endocytosis is often used to 
transport macromolecules that are too large for paracelular transport or carier-mediated 
mechanisms [71]. In the context of the intestinal epithelial barier, endocytosis 
contributes to overal transport when the cargo endocytosed at the apical membrane is 
traficked to the basolateral membrane and exits the cel. Endocytosis can be divided into 
phagocytosis, which occurs in particle uptake by macrophages, and pinocytosis, in which 
the cel membrane invaginates to enclose extracelular fluid and any molecules bound to 
the cel surface. Pinocytosis can be further clasified to receptor-mediated endocytosis 
(RME), which is either clathrin or non-clathrin-mediated, as wel as adsorptive and fluid-
phase endocytosis [71]. Adsorptive and fluid-phase endocytosis are non-specific 
mechanisms in which molecules are either physicaly adsorbed to the cel surface or 
engulfed with the extracelular fluid and brought into the cytosol. Because these 
 
26 
 
 
mechanisms are non-specific, they are typicaly unsaturable and show low substrate 
afinity. In contrast, receptor-mediated endocytosis is highly specific, saturable and 
shows high afinity for the substrate, making it a much more eficient transport method.  
Receptor-mediated endocytosis occurs when ligands bind to specific cel surface 
receptors and are internalized into the cel for further traficking [72]. RME can be 
divided into clasical clathrin-dependent RME and non-clathrin RME, such as caveolin-
mediated endocytosis (CME). Clathrin-dependent endocytosis occurs when a ligand 
binds its receptor and concentrates in a clathrin coated-pit on the plasma membrane 
presenting specific adaptor proteins. After ligand binding, the membrane closes around 
the ligand-receptor complex and dynamin, a smal GTPase, pinches off the vesicle from 
the membrane. Upon internalization, hydrogen pumps are recruited to the vesicle, 
thereby lowering its pH. Consequently, the vesicle sheds its clathrin coat and matures 
into an early endosome. From that point, the vesicle can be recycled to the apical 
surface, continue for degradation in the lysosomes or be transcytosed to the basolateral 
side of the cel. Transferin is a clasical ligand for clathrin-mediated endocytosis and is 
often used as a control to monitor endocytosis and traficking [71]. Caveolin-mediated 
endocytosis occurs by a similar proces except instead of traficking to early endosomes, 
the cargo is traficked to caveosomes and then to the endoplasmic reticulum, thus 
eliminating the pH lowering step central to clathrin RME [73]. Avoidance of low pH and 
lysosomal enzymes makes the caveolar pathway more suitable for uptake of degradation-
sensitive bioactive agents than clathrin-mediated endocytosis. Cholera Toxin B is a 
clasical caveolin endocytosis ligand and can be used as a control to monitor caveolin 
RME. These two mechanisms of RME are ilustrated in Figure 2.4. 
 
27 
 
 
 
Figure 2.4. Clathrin- and Caveolin-Mediated Endocytosis and Transcytosis. A) In 
clathrin-mediated endocytosis, ligands are transported to the early endosomes where they 
can either be recycled to the apical side, transcytosed to the basolateral side or sent to the 
lysosomes for degradation. B) In caveolin-mediated endocytosis, ligands can either be 
transcytosed or sent to the endoplasmic reticulum from the caveosomes.  
 
28 
 
 
2.3.3 Physiochemical Properties that Govern Intestinal Absorption 
2.3.3.1 The Lipinski Rule of 5 
 Due to the restricted pathways of intestinal permeation, there are several key 
physicochemical properties that govern oral absorption. The ?Lipinski Rule of 5?, 
derived from a large dataset of compounds with favorable or unfavorable permeation 
properties, states that compounds likely to have poor intestinal absorption have at least 
one of the following: 1) a molecular weight greater than 500 Da, 2) log P greater than 5, 
3) more than 5 hydrogen-bond donors, and 4) more than 10 hydrogen-bond aceptors 
[74].  These parameters are derived from the requirement for smal size and a minimum 
amount of lipophilicity for pasive difusion across the intestinal barier. However, 
compounds must also have sufficient aqueous solubility to disolve in the aqueous 
intestinal environment, creating a delicate balance betwen solubility and lipophilicity 
[75]. Importantly, the Lipinski Rule of 5 does not apply to substrates of intestinal 
transporters such as antibiotics or vitamins. 
 
2.3.3.2 The Biopharmaceutics Classification System (BCS) 
 The Biopharmaceutics Clasification System was developed in 1995 to 
characterize factors that determine the rate and extent of drug absorption in the 
gastrointestinal tract. Solubility and permeability were found to be the key parameters 
that govern intestinal absorption [76] and compounds are designated as Clas I- Clas IV 
based on these factors (Figure 2.5). The BCS is used by the Food and Drug 
Administration and the World Health Organization to set bioavailabilty and 
bioequivalence standards for oral drug approval [77]. These clases are used to predict in 
 
29 
 
 
Clas I 
High Solubility 
High Permeability 
Clas II 
Low Solubility 
High Permeability 
Clas III 
High Solubility 
Low Permeability 
Clas IV 
Low Solubility 
Low Permeability 
P
er
m
ea
b
i
l
i
t
y: 
 
H
u
man
 
i
n
t
es
t
i
n
al
 
ab
s
o
rp
t
i
o
n
 
Solubility: 
Volume of water required to disolve the highest dose 
strength acros the physiological pH range 
 
Figure 2.5. The Biopharmaceutics Clasification System. Compounds are clasified as 
BCS Clas I-Clas IV based on their solubility and permeability. (Adapted from [77]) 
 
30 
 
 
vitro/ in vivo correlations of oral drug absorption as wel as the rate-limiting step in 
absorption. For Clas I compounds, since the solubility and permeability are both high, 
the rate-limiting step in absorption is defined by gastric emptying. In contrast, for Clas I 
compounds, disolution is a rate-limiting step while for Clas II compounds, membrane 
absorption is rate-limiting. Finaly, the gastrointestinal absorption of Clas IV drugs is 
clasified as highly unpredictable, thus these drugs present significant problems for oral 
delivery and are often administered intravenously [76]. 
 
2.4 Models to Predict Oral Absorption and Oral Bioavailability 
There are several experimental models used to predict oral absorption in the human 
gastrointestinal tract. This section wil describe a series of in silico, in vitro, in vivo, ex 
vivo and in situ models as wel as discuss the merits and drawbacks of each. 
 
2.4.1 In Silico Models 
 In silico models are often used to predict the gastrointestinal absorption and oral 
bioavailability of compounds in early drug development. Because drug molecules often 
fail in later stages of development due to poor oral absorption, creation of eficient and 
predictive in silico models is critical for pharmaceutical development. One of the most 
commonly used techniques is QSAR/QSPR, or Quantitative Structure Activity/ Property 
Relationship, in which datasets of compounds with known absorption properties are 
combined with a series of molecular descriptors to predict properties of novel drug 
molecules. Molecular descriptors such as molecular weight, number of hydrogen bond 
donors, number of hydrogen bond aceptors, octanol-water partitioning coeficient 
 
31 
 
 
(logP), apparent partition coeficient (logD), and intrinsic solubility have al been used to 
predict pasive absorption [78]. These types of models are highly dependent on the size 
and quality of the dataset. While many pharmaceutical companies have large in-house 
datasets, this information is often not available to the general public. Recently, Hou and 
colleagues published one of the largest intestinal absorption datasets with 647 drugs and 
drug like compounds [79], and a corresponding data set for oral bioavailabilty with 768 
compounds [80]. These robust datasets have the capability to dramaticaly improve the 
predictive quality of in situ models. Many analysis algorithms have been developed 
including multiple linear regresion, partial least squares, recursive partitioning and 
finaly support vector machines, which have shown the highest predictive ability when 
used on training and test sets [78].  
 These in silico methodologies can aid in the prediction of absorptive behavior of 
novel compounds. They are eficient as wel as economical in early drug screning since 
they can be completed prior to compound synthesis. However, as mentioned earlier, they 
are severely limited by the training dataset. In addition, while they can reliably predict 
gastrointestinal absorption via pasive difusion, they do not acount for active transport, 
which can be a significant pathway of intestinal transport for certain molecules [78]. 
Finaly, they are very poor at predicting overal bioavailability since first pas 
metabolism can have a large impact on bioavailabilty and is dificult to model [80]. 
 
2.4.2 Parallel Artificial Membrane Permeability Asay 
 PAMPA, or the paralel artificial membrane permeability asay, was first 
introduced in 1998 by Kansy and co-workers as a rapid permeability asesment tool 
 
32 
 
 
[81]. This cel-fre system uses artificial membranes to measure the permeability of 
compounds in a 96 wel microplate format. There have been several diferent variations 
of PAMPA that difer by the type of membrane, pore size, nature and percentage of lipids 
as wel as the incubation time of the asay, which can range from 2-15 hours [82].  The 
most commonly used membrane material is a hydrophobic poly (vinylidene fluoride) 
(PVDF) membrane with 125 ?m pore size. Traditionaly, detection was achieved by a 
UV plate reader, but more recently enhanced detection is acomplished by HPLC-UV 
and HPLC-MS/MS techniques. 
 There are several advantages of PAMPA, which have lead to its widespread use. 
It is a direct measure of pasive permeability and miics the membrane nature of the 
gastrointestinal barier. In addition, it is amenable to a 96 wel plate format, making it a 
high-throughput method, which is critical for early pharmaceutical development. It is 
simple to manipulate the membrane characteristics to design a system that is suited for a 
given application. Finaly, the efect of excipients on compound permeability can be 
easily studied [82]. Unfortunately, because PAMPA is a cel-fre system, it can only 
measure pasive permeability via the transcelular route and does not acount for active 
transport or paracelular permeability, which can have significant contributions to the 
transport of certain types of compounds [82]. Therefore, while PAMPA is an excelent 
screning tool, it does not provide mechanistic insight and only measures one type of 
transport. 
 
 
33 
 
 
2.4.3 Caco-2 Monolayers 
 The Caco-2 monolayer model, first introduced by Hidalgo and colleagues in 1989 
[83], has become one of the most widely used absorption screning technologies in both 
academic and industrial laboratories [84, 85]. In this technique, Caco-2 cels grown on 
semi-permeable supports are used to miic the intestinal barier. Caco-2 cels are 
derived from human colonic adenocarcinoma, and spontaneously diferentiate into 
intestinal enterocytes when grown to confluence on semi-permeable membrane supports. 
After diferentiation, the Caco-2 cels are polarized, display apical microvili, form tight 
junctions and have apical enzymes and transport systems similar to absorptive 
enterocytes [86].  
Caco-2 cels are grown on semi-permeable membranes in a Transwel
?
 system, 
containing an apical chamber, corresponding to the intestinal lumen, and a basolateral 
chamber, corresponding to the blood stream (Figure 2.6). Caco-2 cels are seded on the 
membrane and grown for 21-28 days to produce confluent, diferentiated monolayers. 
Compounds can be added in either the basolateral or apical chamber to study directional 
transport. Apical to basolateral transport coresponds to intestinal absorption while 
basolateral to apical transport corresponds to intestinal eflux [87]. Transwel
?
 systems 
are available in microplate format from 6 to 96 wels, and can be adapted to work in a 
robotic system, alowing for simultaneous screning of many diferent compounds. The 
apparent permeability (P
ap
) of compounds in the apical to basolateral direction has been 
shown to correlate with human intestinal absorption, making the Caco-2 monolayer 
system an excelent high throughput technique to predict absorption properties of novel 
compounds [88].  
 
34 
 
 
 
Figure 2.6. Transwel
? 
System used in the Caco-2 Monolayer Permeability Asay. 
Caco-2 cels are seded on the membrane insert and alowed to grow for 21-28 days to 
produce confluent, diferentiated monolayers. Compound flux from the apical to 
basolateral chamber is used to measure permeability. 
 
35 
 
 
 Because diferentiated Caco-2 cels have wel-defined tight junctions as wel as 
developed active transport systems, the Caco-2 cel monolayer system can be used to 
study not only pasive difusion, but also active transport, paracelular transport and 
potential drug eflux. Therefore, the Caco-2 monolayer system can be used as a tool to 
study both the amount of transport, which is predictive of human intestinal absorption, as 
wel as the mechanism of transport of novel compounds [87].  
In order to form diferentiated monolayers, Caco-2 cels must be grown on 
permeable supports for 21-28 days. During this time, the cels achieve confluence and 
diferentiate, forming tighter junctions and expresing proteins similar to absorptive 
enterocytes rather than colonic cels. This transformation can be monitored by 
transepithelial electrical resistance (TER), which quantifies the tightnes of the 
monolayer. TER increases over time, and ultimately plateaus betwen 21 and 28 days 
when the cels can be used for experiments [87]. Typical TER values of competent 
diferentiated monolayers range from 600-800 ohm-cm
2
.
 
Measurement of TER serves 
as a control for monolayer integrity before and during a transport asay. Transport of 
paracelular markers such as mannitol and lucifer yelow, which should have P
ap
 les 
than 1 x 10
-6
 cm/s, can also serve as controls for monolayer integrity [87]. 
 There are several advantages to using the Caco-2 monolayer system to predict 
intestinal absorption of novel compounds in humans. The system can be performed in 
multiwel format, alowing for paralel testing of diferent compounds using minimal 
compound, reagents and asay time. In addition, because Caco-2 cels miic intestinal 
enterocytes, many aspects of intestinal absorption including pasive transport, active 
transport, paracelular transport and drug eflux can be studied in one asay.  
 
36 
 
 
Unfortunately, there are also several disadvantages to the Caco-2 monolayer 
system. Because the cels must be cultured for 21 days, this asay is labor intensive and 
requires a significant amount of serum-enriched media, increasing the cost of the asay. 
In addition, studies using the monolayer system to measure permeability of compounds in 
diferent laboratories have shown that there is a high degre of variability in the asay, 
mostly due to the intrinsic variability in Caco-2 cels [86]. Caco-2 cels contain several 
diferent sub-populations, which can be inadvertently selected for depending on the 
culture conditions, causing shifts in the population over time. These shifts can 
significantly afect the monolayer formation including the tightnes of the tight junctions 
as wel as protein expresion [89]. Many experimental conditions can also afect the 
ultimate performance of the monolayers including cel pasage number, seding density, 
filter type and pore size, cel culture media, transport buffer and monolayer age. 
Therefore it is critical to control these parameters within a given laboratory so that 
experiments are comparable. In addition, it is important to run standards of known 
permeability as wel as measure the TER to ensure monolayer consistency. Despite 
these chalenges, the Caco-2 monolayer system continues to be a powerful tool for 
studying the extent and mechanism of intestinal absorption. 
 
2.4.4 Fast-Caco-2 Asay 
 Because the 21-day growth period is both labor-intensive and expensive, 
alternative diferentiation systems have been developed for faster Caco-2 asays. 
Beckton Dickenson (BD) markets the BD Biocoat Caco-2 HTS Asay system, which can 
create diferentiated Caco-2 monolayers in 3 days instead of 21 days. In this system, 
 
37 
 
 
Caco-2 cels are seded at high density onto Transwel
?
 filters coated with fibrilar 
collagen. After the cels atach to the Transwels on the first day, diferentiation medium 
containing butyric acid is added for the next two days, causing rapid diferentiation of the 
cels into intestinal enterocytes [90]. The resulting monolayers are slightly leakier than 
traditional Caco-2 monolayers and show poor expresion of transporters, but they can 
stil provide rank order measurements for permeability. Adding fetal bovine serum to the 
seding medium reduces these problems as it facilitates cel atachment, making this 
asay system ore closely miic 21-day standard [90]. Therefore, although the 
monolayers are not exactly the same as the traditional 21-day Caco-2 monolayer system, 
the fast-Caco-2 asay can efectively determine rank order of permeability and provides 
significant time and labor savings. 
 
2.4.5 Other Types of Cel-Monolayer Systems 
 In addition to Caco-2 cels, several other cel types have been used to miic the 
intestinal barier in the Transwel
?
 system. MDCK cels, derived from the canine kidney, 
are useful alternatives to Caco-2 cels as they form diferentiated monolayers in 3 days 
instead of 21 days. They have looser tight junctions than Caco-2 cels and do not miic 
the protein expresion of human enterocytes as wel as Caco-2 cels, but they are stil 
useful as a rapid screning tool for pasive permeability and paracelular transport [87, 
91]. In addition, MDCK cels can be stably or transiently transfected, which alows 
researchers to study the impact of selected transporters, which is not possible in Caco-2 
cels. Recently, TC-7 cels, a subclone of Caco-2 cels, have been used to study intestinal 
transport [92]. TC-7 cels have higher expresion of CYP enzymes, which have very low 
 
38 
 
 
expresion in Caco-2 cels. This alows for the simultaneous study of both transport and 
metabolism, which is important for compounds that are metabolized in the intestinal 
barier. Finaly, co-cultures of mucus secreting HT-29 and MTX cels have been used 
[93]. While these show lower active transport than Caco-2 cels, they show higher 
paracelular transport and may more acurately predict the paracelular pasage of 
hydrophilic molecules. Importantly, al of these cel culture monolayer systems are able 
to isolate diferent elements of the intestinal transport proces and are useful for studying 
a variety of compounds in a high throughput system that miics the in vivo environment. 
 
2.4.6 Everted Rat Intestinal Sac 
 Often, it is desirable to use a tisue-based system rather than a cel-based system 
to more closely miic gastrointestinal absorption. The everted rat intestinal sac model, 
introduced by Wilson and co-workers in 1954, has been used to study intestinal 
absorption in an ex vivo environment [94]. In this method, the intestine is removed from 
a sacrificed rat, everted over a glas rod so that the mucosal side is facing outward and 
then sectioned into 2-4 cm sacs and filed with oxygenated cel culture medium. The sacs 
are then placed in media containing the drug of interest and the amount of drug 
transported into the sac (mucosal to serosal direction) is quantified [87]. The asay time 
is limited by the viability of the intestinal tisues, which is usualy 2 hours. There are 
several advantages to the everted rat intestinal sac model compared to cel-based models, 
including the presence of a mucosal layer and the ability to study diferent sections of the 
intestine. In addition, transport can be studied in conjunction with intestinal metabolism. 
This method is fast and relatively inexpensive compared to other animal studies [87]. 
 
39 
 
 
However, the drug must cross the muscle layer in addition to the intestinal barier, which 
can potentialy underestimate permeability, and the method is low-throughput compared 
to cel-based asays. Despite these drawbacks, the everted rat intestinal sac method is an 
efective way to measure intestinal permeability in an ex vivo environment. 
 
2.4.7 Isolated Intestinal Tisue 
 First conceptualized by Hans Using in 1949 [95] and later miniaturized by Gras 
and Swetana in 1988 [96], the Using Chamber method uses isolated intestinal tisue to 
study transport of compounds across the gastrointestinal barier. Sections of intestinal 
tisue are placed betwen two chambers and exposed to Krebs Ringer Bicarbonate bufer 
gased with 95:5 CO
2
:O
2
. The gasing is used to maintain tisue viability and promote 
flow in the chambers. Importantly, similar to the everted sac model, intestinal 
metabolism can be studied at the same time as absorption and transport. This system 
alows for studying transport across diferent sections of the intestine as wel as in 
diferent species, which can asist in choosing a relevant in vivo model [87]. In addition, 
the muscle layer can be removed from the tisue, alowing the study of the epithelial 
barier in isolation. While this method is very useful, there are several disadvantages, 
most notably the dificulty of preparation of the intestinal sheets, especialy removing the 
muscle layer. The viability of the tisue sheet during the experiment is also a significant 
concern and can vary dramaticaly. Finaly, this system is significantly lower throughput 
than in vitro cel-based models, but can provide more information and is closer to the in 
vivo environment [87].  
 
40 
 
 
2.4.8 Rat Intestinal Perfusion 
 In situ perfusion of the rat intestine most closely miics in vivo environment 
without using the whole animal. In this technique, compounds are added to the lumen of 
sections of the intestine in anesthetized rats and the disappearance of the compound is 
monitored over time. This system alows the study of the gastrointestinal transport in 
isolation without confounding factors such as first pas metabolism, biliary excretion or 
enterohepatic circulation [87].  Importantly, most in situ perfusion experiments asume 
that disappearance of the compound from the lumen indicates transport across the 
intestinal barier. If there is significant metabolism of the compound or absorption in the 
intestinal barier, this method can significantly overestimate transport. Therefore, the 
asay can be amended to sample blood from the mesenteric vein to measure transport, 
although this becomes much more technicaly chalenging. In situ perfusion serves as a 
bridge betwen tisue-based models and in vivo models and combines the advantages of 
each into a powerful experimental technique. 
 
2.4.9 In Vivo Models 
 In vivo administration of compounds by the oral route in laboratory animals is a 
commonly used method to predict oral bioavailability in humans. Because the epithelial 
barier composition betwen mouse, rat and human are similar, gastrointestinal transport 
should be similar betwen species. However, diferences in gastrointestinal transit time, 
expresion of transporters and even pH can difer betwen species, so direct comparison 
is often not possible [87]. In oral bioavailability studies, dosing by the oral or 
intraduodenal route is compared to intravenous dosing. Plasma samples are taken over 
 
41 
 
 
time and the area under the curve (AUC) of the oral administration methods are 
compared to the intravenous route to get the fraction of the dose absorbed which is 
represented as the overal percent bioavailability. Administration of the compound 
intraduodenaly instead of oraly eliminates confounding factors from the stomach such 
as residence time and gastric absorption.  Importantly, measuring bioavailability by 
plasma concentration includes the efects of both gastrointestinal absorption as wel as 
first pas metabolism. Sampling from the portal vein before the liver instead of from the 
circulation alows for isolation of intestinal transport and can be useful in determining the 
impact of absorption and metabolism on overal bioavailability [87]. Because in vivo 
studies have al of the complexity of the human absorption system, they can often provide 
the closest estimate to oral absorption in humans. However, there are several 
disadvantages of in vivo studies including the use of whole animals, dificult and tedious 
plasma sampling and analysis as wel as the aforementioned interspecies diferences. 
Despite these chalenges, in vivo studies in laboratory animals are stil the closest 
approximation to human intestinal absorption and are a good follow up experiment to 
evaluate promising candidates from high-throughput techniques. 
 
2.5 Curent Strategies for Oral Drug Delivery 
 Unfortunately, many newly discovered drugs have low solubility and poor 
intestinal permeability, which would categorize them as problematic BCS Clas IV 
compounds. In fact, the bioavailability of novel chemotherapy drugs is often so low that 
these drugs must be administered intravenously. There are several current strategies to 
improve the bioavailability of poorly water-soluble and poorly permeable therapeutics 
 
42 
 
 
including 1) prodrug approaches, 2) eflux and metabolic inhibitors, 3) tight junction 
modulators and 4) novel macromolecules. Each of these approaches wil be discussed in 
further detail below. 
 
2.5.1 Prodrugs 
 A prodrug is defined as a drug molecule chemicaly conjugated to a promoiety 
that is used to improve the drug?s physicochemical properties and is cleaved to release 
the pharmacologicaly active, fre drug [97]. Common drug delivery obstacles overcome 
by prodrug strategies include poor aqueous solubility, low permeability, fast elimination, 
off-target efects and premature metabolism [97]. In the context of oral drug delivery, 
prodrugs can be used to improve the aqueous solubility of a drug and permeability across 
the intestinal barier. Promoieties are subsequently cleaved by enzymatic mechanisms or 
by hydrolysis to yield the fre, active drug.  
Poor aqueous solubility is becoming an increasingly common problem with 40% 
of new chemical entities discovered by pharmaceutical companies in combinatorial 
screning now identified as poorly water soluble [98]. In addition, traditional 
formulation techniques to improve water solubility such as use of diferent salt and 
crystaline forms or reducing particle size are not always succesful, necesitating the use 
of new strategies. Examples of prodrug modifications to improve aqueous solubility 
include addition of phosphate groups via an ester bond, addition of amino acids via ester 
bonds or conjugation to PEG [98]. For example, phosphate prodrugs of Buparvaquone, 
used to treat leishmaniasis, increase the aqueous solubility from 0.03 ?g/ml to greater 
than 4 mg/ml [99]. The prodrug releases fre buparvaquone upon metabolism by 
 
43 
 
 
cytochrome P450. Unfortunately, these methods can also potentialy decrease membrane 
permeability due to the addition of charged groups in the case of phosphates or amino 
acids or increased size, in the case of PEG. Therefore, solubility-enhancing prodrug 
modifications must be used in delicate balance with maintaining sufficient membrane 
permeability [100].  
 Prodrugs have also been used to overcome permeability limitations in oral drug 
administration [101]. Conjugation of lipophilic constituents to polar functional groups on 
the parent drug serves to increase the overal lipophilicity of the drug and improve 
pasive permeability through the intestinal barier [101]. This is especialy important 
when the polar groups are ionizable in the pH of the smal intestine, as charge can 
severely limit intestinal permeability. These groups can be designed to be cleaved during 
first-pas metabolism in the liver so that the fre drug is liberated prior to reaching the 
general circulation. Finaly, if pasive difusion across the intestinal barier is not 
possible, prodrugs can be designed to be substrates of gastrointestinal transporters, taking 
advantage of carier-mediated transport pathways [101].  
 
2.5.2 Eflux and Metabolic Inhibitors 
 In addition to poor solubility and poor pasive permeability, drugs can have poor 
intestinal absorption due to eflux or metabolism in the intestinal barier. These problems 
are especialy prevalent with anti-cancer drugs which are often efluxed by P-
glycoprotein (P-gp) or multidrug resistance proteins (MDR), or metabolized by 
Cytochrome P-450 (CYP) enzymes [102]. To addres this isue, these drugs are often co-
administered with inhibitors of drug eflux and drug metabolism to decrease these efects. 
 
44 
 
 
For example, Paclitaxel, a taxane anti-cancer drug, shows in vivo bioavailability of only 
10%, mostly due to eflux by P-gp. Co-administration of P-gp inhibitors SDZ PSC833, a 
cyclosporine D analogue or GF120918, a non-imunosuppresive P-gp blocker, showed 
significant improvements in oral Paclitaxel bioavailability [103]. In the case of 
Docetaxel, which is a P-gp substrate and metabolized by CYP3A4, co-administration of 
Ritonavir, a CYP3A4 inhibitor, led to significant gains in oral bioavailability [103]. 
Similar improvements in oral bioavailability have been observed by co-administration of 
CYP3A4 inhibitors with Irinotecan [104] and HIV-protease inhibitors [105]. These 
examples ilustrate the potential of co-administration of metabolic and eflux inhibitors to 
improve oral bioavailability.  
 
2.5.3 Tight Junction Modulators 
 For hydrophilic molecules that cannot permeate the intestinal barier by pasive 
difusion, but are too large for the paracelular route, tight junction opening serves as a 
novel mechanism to enhance intestinal permeability. Several molecules have been 
developed to transiently open the tight junctions in the intestinal barier. Medium chain 
faty acids such as sodium caprate have been shown to open the tight junctions, as 
indicated by increased paracelular flux of markers such as mannitol. These faty acids 
are thought to activate phospholipase C on the plasma membrane, which then causes an 
increase in intracelular calcium concentration [106]. This elevated calcium concentration 
causes contraction of calmodulin-dependent actin filaments, which subsequently opens 
the tight junctions. In vivo studies of sodium caprate have been completed in several 
animal models and in humans. Modest increases in oral bioavailabilty of a variety of 
 
45 
 
 
drugs were sen with minimal damage to the intestinal barier, suggesting the potential of 
sodium caprate as a safe absorption enhancer [107].  
 Similar to sodium caprate, medium chain mono- and di-glycerides and medium- 
and long-chain faty acid esters of carnitine and choline can also open tight junctions, as 
evidenced by reduced TER and increased permeation of low molecular weight 
paracelular markers [108]. Nitric oxide donors such as S-nitroso-N-acetyl-DL-
penicilamine (SNAP), have been shown to increase permeability of 4 kDa FITC-dextran 
and influence the expresion and localization of tight junction proteins, suggesting their 
potential as absorption enhancers [62]. While SNAP showed promising results in rabbits 
and rats, the mechanism of tight junction modulation is stil unknown. Finaly, zonnula 
occludens toxin (Zot), a protein derived from Vibrio cholerae, has demonstrated tight 
junction modulation.  Fasano and colleagues report that Zot selectively opens tight 
junctions in the smal intestine in a dose-dependent and reversible manner [109]. Studies 
in rats showed a significant increase in insulin and imunoglobulin bioavailability in rats 
treated with Zot. Tight junction modulators can improve delivery of large, water-soluble 
drugs via the paracelular route. 
 
2.5.4 Macromolecules 
 In addition to the above-mentioned permeation enhancers there are several 
macromolecular permeation enhancers that are currently being investigated. 
Polyacrylates, or synthetic high molecular weight polymers of acrylic acid, have been 
shown to be mucoadhesive and to modulate tight junction integrity [110]. It is 
hypothesized that polyacrylates achieve permeation enhancement by binding calcium 
 
46 
 
 
ions that are required for tight junction maintenance. Importantly, studies in rats showed 
that long term oral administration of polyacrylates did not cause any significant toxicities 
in vivo, suggesting the safety of this strategy [111]. 
 Chitosan and chitosan derivatives have also been shown to modulate tight junction 
integrity [107]. Studies on Caco-2 cels showed that treatment with chitosan and chitosan 
derivatives caused enhanced paracelular flux of smal molecules such as mannitol and 
large markers such as FITC-dextran. The precise structure of chitosan can vary both in 
molecular weight and in the amount of deacetylation, as wel as the addition of methyl 
groups for enhanced solubility in basic pH conditions. Al of these changes alter the 
degre to which chitosan can enhance paracelular transport. It has been shown that 
chitosan binds to the intestinal cels, causing redistribution of cytoskeletal F-actin and 
ZO-1 tight junctional proteins [112]. The toxicity of chitosan is directly related to its 
degre of deacetylation and there exists an optimum amount of deacetylation that causes 
tight junction opening with minimal toxicity.  
 Recently, poly (amdio amine) (PAMAM) dendrimers have been identified as potent 
tight junction modulators. Treatment of Caco-2 cels with PAMAM dendrimers of 
diferent sizes and surface functionalities has been shown to increase the paracelular flux 
of mannitol, reduce TER and increase acesibility of actin and occludin proteins [15, 
16]. PAMAM dendrimers are transported across the intestinal barier, making them both 
permeation enhancers and potential oral drug delivery cariers [113]. The following thre 
sections wil describe a detailed background on dendrimers, applications of PAMAM 
dendrimers in drug delivery and the current knowledge of these constructs as oral drug 
delivery cariers. 
 
47 
 
 
2.6 Dendrimers 
2.6.1 History of Dendrimer Development 
Dendrimers are unique polymeric structures that are highly-branched, nanoscale 
in size and have a wide range of applications in the field of nanobiotechnology. The term 
dendrimer is derived from the grek ?dendra? meaning tre and ?meros? meaning part 
[114]. Dendrimers were first conceived of in the late 1970?s by the groups of Vogtle, 
Denkwalter, Tomalia and Newkome and were first presented to the public by Tomalia in 
1983 [14]. Since then, the number of dendrimer-related publications and technologies 
has increased rapidly, ilustrating the utility of this nanoscale structure [115]. Several 
commercial applications of dendrimers have shown promise in the past decade. 
Vivagel
TM
, a vaginal topical microbiocide for the prevention of HIV transmision, was 
granted fast track status for clinical trials in 2006 [116]. SuperFect
?
, developed by 
Qiagen, uses dendrimer technology to improve transfection in a wide range of cel lines 
[117]. Finaly, Stratus CS
?
 Acute Care Diagnostic System, uses dendrimers to rapidly 
detect myoglobin, a sensitive marker for acute coronary diseases [118]. 
One of the most promising areas of application of dendrimers is in the field of 
nanomedicine. Dendrimers have many unique physical properties including near 
monodispersity and nanoscopic size [12, 13]. With each increase in dendrimer 
generation, the diameter increases linearly while the number of surface groups increases 
exponentialy. This creates high density surface groups that can be conjugated to drug 
molecules [20, 119-124], targeting moieties [125-128], and imaging agents [129-131], or 
complexed with DNA [132, 133], making dendrimers versatile drug delivery platforms 
[14]. Figure 2.7 ilustrates the multifunctional nature of dendrimers [24].  
 
48 
 
 
 
Figure 2.7. Dendrimers as Multifunctional Nanocariers. Targeting moieties can be 
atached to the dendrimer surface groups while drugs and imaging agents can be 
conjugated to the surface, encapsulated in the dendrimer core or complexed (such as 
therapeutic DNA) with the dendrimer structure. (From [24]). 
 
49 
 
 
2.6.2 Dendrimer Synthesis 
In the early stages of development, dendrimers were synthesized by the divergent 
method, wherein repeating units were progresively added to an initiator core, increasing 
the generation number with each reaction step (Figure 2.8) [115]. While straightforward, 
this synthetic strategy is somewhat tedious, requires purification after each step and is 
dificult to scale-up. Subsequently, dendrimers were produced by convergent synthesis 
methods where reactive dendrons are synthesized and then atached to a multifunctional 
initiator core to generate the final product (Figure 2.8) [134]. Convergent synthetic 
strategies are useful for larger dendrimers as they contain fewer steps than divergent 
approaches. 
Recent innovations in dendrimer synthesis have facilitated commercial production 
of dendrimers, focusing on producing high yields of intermediates and minimizing toxic 
byproducts.  ?Lego chemistry? uses branched monomers, which facilitates production of 
dendrimers by minimizing the number of reaction and purification steps. In addition 
?click chemistry? has been developed to produce dendrimers with specific surface 
chemistries using a copper catalyst. These innovations have alowed several families of 
dendrimers to be produced commercialy, which has had a significant impact on the use 
of dendrimers in biomedical research [13].  
 
2.6.3 Types of Dendrimers 
More than 100 diferent types of dendrimers with more than 1,000 diferent 
surface modifications have been developed to date [13]. These dendrimers difer in the 
identity of the initiator core, branching units and surface groups, al of which can have a 
 
50 
 
 
 
Figure 2.8. Divergent and Convergent Syntheses of Dendrimers. In divergent synthesis, 
monomeric branches are repeatedly added to the initiator core, increasing the dendrimer 
generation with each addition. In convergent synthesis, reactive dendrons are added to a 
multifunctional core. (From [24]). 
 
51 
 
 
profound impact on their biological properties. In addition, the synthetic strategy used to 
produce each type of dendrimer is often diferent. Some of the most common types of 
dendrimers, including poly (propylene amine) (PI), polyether and poly (amido amine) 
(PAMAM) dendrimers, are ilustrated in Figure 2.9. PI dendrimers were first produced 
by Vogtle and are synthesized by the divergent method (Figure 2.9 A) [135]. They are 
commercialy available and can be amine or nitrile terminated, which significantly 
impacts their biological properties [115].  Polyether dendrimers or ?Frechet-type? 
dendrimers are synthesized by the convergent method (Figure 2.9 B) and have been used 
in applications such as light harvesting and catalysis [115]. PAMAM dendrimers, 
commercialy available as Starburst
?
 PAMAM dendrimers, have an ethylene diamine 
core and amido amine branching units with carboxyl or amine terminal groups (Figure 
2.9 C,D). They are described in detail in the following section. 
 
2.7 Poly (amido amine) Dendrimers 
2.7.1 Structure of PAMAM Dendrimers 
Poly (amido amine) (PAMAM) dendrimers were originaly developed by Tomalia 
at Dow Laboratories in 1979. PAMAM dendrimers have numerous applications in 
nanobiotechnology and are some of the most commonly used since they are both fully 
characterized and commercialy available [136-138]. Starburst
?
 PAMAM dendrimers 
have an ethylene diamine core, an amido amine repeat branching structure and are 
available in generations 0.5 through 10. As generation number is increased, the number 
of active terminal groups doubles, while the diameter increases by approximately 1 nm, 
giving PAMAM dendrimers a diameter range of 1.5 to 14.5 nm [136]. In PAMAM 
 
52 
 
 
(A) B) 
 
 
C) D) 
 
 
 
Figure 2.9. Chemical Structures of Common Dendrimers. A) G4 PI dendrimer with a 
1,4-diamnobutate core, B) G4 polyether dendrimer, C) G2 PAMAM dendrimer with 
amine terminal groups and an ethylenediamine core and D) G1.5 PAMAM dendrimer 
with carboxylic acid terminal groups and an ethylenediamine core. (From [24]). 
 
53 
 
 
dendrimers ?full? generations (G1, G2, G3, etc.) have amine-terminated branches 
whereas ?half? generations (G1.5, G2.5, G3.5, etc.) have carboxylic acid terminated 
branches (Figure 2.9 C,D). They can also be modified with terminal hydroxyl groups to 
neutralize the surface charge. The surface charge and chemistry of PAMAM dendrimers 
have a significant impact on their biological properties including toxicity and 
biodistribution. Table 2.2 summarizes the physical properties of G0-G5 dendrimers 
commercialy available from Dendritech.  
 
2.7.2 Biocompatibility and Biodistribution of PAMAM Dendrimers 
 In order to be suitable for clinical use, polymeric cariers must be non-toxic and 
non-imunogenic [139]. In addition, polymeric cariers must display biodistribution 
properties that alow for uptake in the target tisue with minimal off-site acumulation 
[139]. One of the first tests for biocompatibility is performance in in vitro toxicity 
screns against diferent cel lines. In vitro toxicology studies have shown that dendrimer 
cytotoxicity is primarily governed by surface chemistry, although the core identity can 
play a role. In general, cationic PAMAM dendrimers show increasing toxicity with 
increases in concentration and generation, while anionic dendrimers have been found to 
be non-toxic against a large variety of cel lines [139]. This charge-dependent efect is 
consistent with PAMAM dendrimer impact on red blood cels. Cationic dendrimers have 
been shown to cause changes in red blood cel morphology at lower concentrations and 
significant hemolysis at higher concentrations. In contrast, anionic PAMAM dendrimers 
of generations 3.5 through 9.5 have been found to show no toxicity against red blood 
cels up to 2 mg/ml concentration [139]. These in vitro studies ilustrate the significant 
 
54 
 
 
Table 2.2. Physical Properties of PAMAM Dendrimers 
Generation Surface 
Functionality 
Number of 
Surface Groups 
Molecular 
Weight 
Measured 
Diameter (?) 
-0.5 -COH 4 436 - 
0 -NH
2
 4 517 15 
0.5 -COH 8 1,269 - 
1 -NH
2
 8 1,430 22 
1.5 -COH 16 2,935 - 
2 -NH
2
 16 3,256 29 
2.5 -COH 32 6,267 - 
3 -NH
2
 32 6,909 36 
3.5 -COH 64 12,931 - 
4 -NH
2
 64 14,215 45 
4.5 -COH 128 26,258 - 
5 -NH
2
 128 28,826 54 
Reported by Dendritech, Inc. Midland, MI 
 
55 
 
 
impact of dendrimer surface chemistry on the toxicological properties of these novel 
cariers. 
 Malik and co-workers investigated the biodistribution of 
125
I-labeled anionic 
(generations 2.5, 3.5, 5.5) and cationic (generations 3,4) PAMAM dendrimers in Wistar 
rats after intravenous administration [140]. Cationic dendrimers were cleared rapidly 
from the circulation, while anionic dendrimers showed longer retention times in the blood 
stream. In addition, both cationic and anionic dendrimers were found to have significant 
acumulation in the liver.  Pasive acumulation of polymeric cariers at the tumor site 
due to the enhanced permeability and retention efect is critical for the applicability of 
dendrimers as anti-cancer drug delivery cariers. In a separate work, Malik et al. showed 
that G3.5 PAMAM dendrimer-Cisplatin conjugates showed significant tumor 
acumulation at a 50-fold increase relative to IV administered fre drug at the maximum 
tolerated dose, suggesting the potential of G3.5 PAMAM dendrimers for tumor delivery 
of anti-cancer agents [141]. While there are stil many in vivo studies necesary to 
characterize the toxicity and biological fate of PAMAM dendrimers, it is clear that 
surface charge significantly influences both toxicity and biodistribution. 
 
2.7.3 Applications of PAMAM Dendrimers as Drug Carriers 
 Because of their high water solubility and large number of ionizable surface 
groups, dendrimers can be used to encapsulate drugs with poor aqueous solubility. It has 
been wel established that dendrimers have a core-shel architecture, alowing for 
encapsulation of drug molecules in the voids of the dendrimer core [13]. Encapsulation 
of drug cargo in dendrimers alows for improved water solubility and thus preferential 
 
56 
 
 
presentation of the drug to the biological membrane and subsequent internalization. In 
addition, the dendrimer can also protect the encapsulated drug cargo from degradation. 
Kannan et al. improved solubility of ibuprofen by encapsulation in G3 dendrimers [142]. 
Ibuprofen encapsulated in dendrimers showed a much higher rate of celular 
internalization in A549 lung cels compared to fre ibuprofen. Drug loading was 
approximately 50% for the G3-ibuprofen complexes, indicating the high capacity of 
PAMAM dendrimers for incorporation of smal molecular weight bioactive agents. 
 Despite their ease of synthesis, PAMAM dendrimer-drug complexes are plagued 
by non-specific drug release, which can lead to unwanted toxicity in vivo, especialy in 
the case of anti-cancer drugs. Therefore, there has been much research done on PAMAM 
dendrimer-drug conjugates where drugs are covalently atached to dendrimer periphery 
groups by chemical linkers. These linkers can be designed to maximize drug release at 
the site of action. Wiwatanapatapee and colleagues studied PAMAM dendrimer-5-
aminosalisyclic acid conjugates for targeted colonic delivery [122]. The conjugates had 
two diferent azo spacers, which are susceptible to degradation by azoreductase, an 
enzyme that is abundant in the colonic environment. The conjugates showed fre drug 
release in the presence of rat cecal contents, but not in the gastric fluid, indicating the 
specificity of the release mechanism. This study ilustrates that the choice of the spacer 
can have a profound impact on the ability of the dendrimer-drug conjugate to release the 
drug in the appropriate environment and time after administration. 
 Finaly, targeting and imaging agents can be atached to dendrimer-drug 
conjugates for multifunctional delivery strategies. Majoros and co-workers conjugated 
folic acid, a targeting moiety that binds to over-expresed folate receptors on cancer cels, 
 
57 
 
 
Paclitaxel, a chemotherapeutic agent, and FITC, a fluorescent ligand, to G5 PAMAM 
dendrimers, creating a tri-functional therapeutic agent [143]. The cytotoxicity and 
celular uptake of this dendrimer system was evaluated in KB cels. The folate-labeled 
dendrimers were found to selectively internalize in folate-receptor-expresing KB cels 
and those containing Paclitaxel caused cytotoxicity similar to the fre drug. In contrast, 
folate receptor- negative cels did not internalize the dendrimers, indicating the strong 
targeting ability. This work ilustrates the potential of using dendrimers as 
multifunctional nanodevices for drug delivery.  
 
2.8 PAMAM Dendrimers as Oral Drug Delivery Systems 
2.8.1 Transepithelial Transport of PAMAM Dendrimers 
 The potential of PAMAM dendrimers as oral drug delivery cariers was first 
reported in 2000 by Wiwatanapatapee et al [144]. In this study, the tisue uptake and 
serosal transfer rates of anionic (G2.5, G3.5 and G5.5) and cationic (G3, G4) PAMAM 
dendrimers were measured in in vitro everted rat intestinal sacs. Anionic dendrimers 
were found to have high serosal transport rates and minimal tisue uptake, while cationic 
dendrimers had lower serosal transport rates and more tisue uptake, most likely due to 
nonspecific binding. Importantly, the dendrimer transport rates were higher than what is 
typicaly sen for large macromolecules, suggesting the unique potential of PAMAM 
dendrimers as polymeric drug cariers.  
 This report of dendrimer transepithelial transport was followed by a 
comprehensive investigation of the influence of dendrimer size, charge and incubation 
time on dendrimer transport across Caco-2 cel monolayers as wel as the mechanism of 
 
58 
 
 
dendrimer transport. As described in Section 2.4.3, Caco-2 cel monolayers serve as a 
model of the intestinal barier and can be used to elucidate the degre and mechanisms of 
intestinal permeability without the confounding factors asociated with in vitro intestinal 
sac models. El-Sayed et al. first explored transepithelial transport of cationic dendrimers 
across Caco-2 cel monolayers in 2002 [145]. In this study, PAMAM dendrimer 
generations G0-G4 were investigated for their potential in oral delivery. Permeability of 
the dendrimers was found to increase with concentration and incubation time. G0-G2 
dendrimers were found to be non-toxic with appreciable permeability, suggesting the 
potential of these constructs as oral drug cariers. Kitchens et al. studied the 
transepithelial transport of cationic (amine-terminated), anionic (carboxylic acid-
terminated) and surface neutral (hydroxyl-terminated) dendrimers across Caco-2 cel 
monolayers [15].  When comparing dendrimers of the same size, cationic dendrimers 
showed the highest permeability, followed by anionic dendrimers and neutral dendrimers. 
Increasing the generation of anionic dendrimers caused an increase in permeability while 
increasing the generation of cationic dendrimers lead to decreased permeability and 
increased toxicity. These studies indicate that within a specified size, charge and 
concentration window, dendrimers can enhance transepithelial transport. 
In addition to determining the transport rates of PAMAM dendrimers across 
Caco-2 cels, several studies have been performed to elucidate the mechanism of 
transport. Studies have shown that dendrimers are transported by a combination of 
transcelular and paracelular mechanisms. Tight junction opening can be monitored by 
several diferent methods including a reduction in transepithelial electrical resistance 
(TER), an increase in flux of paracelular markers such as [
14
C]-mannitol or Lucifer 
 
59 
 
 
yelow and an increase in exposure of tight junction proteins such as occludin [15]. El-
Sayed et al. reported that G2.5 and G3.5 dendrimers reduced TER and increased [
14
C]-
mannitol flux up to 6 fold compared to the control [146]. In contrast, OH-terminated 
dendrimers did not cause a significant change in TER or mannitol permeability. 
Kitchens et al. reported that cationic dendrimers increased [
14
C]-mannitol flux and 
reduced TER to a greater extent than anionic dendrimers, indicating a strong charge 
dependence of dendrimer tight junction opening ability [16]. In addition, higher 
generation dendrimers of both cationic and anionic dendrimers opened the tight junctions 
to a greater degre than lower generations. Finaly, increasing incubation time from 90 to 
210 minutes increased tight junction opening for al dendrimer generations. In a paralel 
study, both cationic and anionic dendrimers were found to interact with tight junction 
proteins, increasing occludin and actin staining, thus further establishing the ability of 
dendrimers to open tight junctions [15]. These studies ilustrate that the size and charge 
of PAMAM dendrimers have a significant impact on their interaction with diferentiated 
enterocytes. In addition, there may be an optimal window of size, charge and incubation 
time that wil lead to the best dendrimer system for a specific oral drug delivery 
application. 
In addition to opening the tight junctions, dendrimers are also transported across 
the intestinal barier by the transcelular route. In 2003 El-Sayed et al. reported energy-
dependent transport of G2 dendrimers, suggesting the involvement of transcytosis in 
addition to transport via the paracelular route [147]. Jevprasephant et al. further 
confirmed this phenomenon by investigating internalization of G3 dendrimers into Caco-
2 cels by visualizing the interaction betwen Caco-2 cels and FITC labeled G3 
 
60 
 
 
dendrimers using flow cytometry and confocal microscopy and gold-labeled G3 
dendrimer by transmision electron microscopy (TEM) [19]. These studies showed a 
significant amount of dendrimer celular internalization with minimal non-specific 
binding on the cel surface, suggesting transport of G3 dendrimers by the transcelular 
route. In addition, Kitchens and colleagues examined the impact of endocytosis 
inhibitors on celular internalization and transepithelial transport of G4 dendrimers in 
Caco-2 cels [148]. Inhibitors including brefeldin A, colchicine, filipin, and sucrose al 
decreased the uptake and transport of the dendrimers, indicating the involvement of 
endocytosis mechanisms in the transport of G4 dendrimers. 
 
2.8.2 Cytotoxicity of PAMAM Dendrimers 
 In order to be useful as oral drug delivery cariers, it is crucial that dendrimers do 
not cause toxicity to intestinal cels during transport. Toxicity of PAMAM dendrimers to 
Caco-2 cels has been investigated using cel viability asays. El-Sayed et al. established 
cationic dendrimers are significantly more toxic than anionic dendrimers and this toxicity 
increases with dendrimer generation and incubation time [146]. In contrast, anionic 
dendrimers are much les toxic and only displayed slight increases in LDH release when 
cels were treated with 10 mM concentrations of G3.5 and G4.5 dendrimers for a 90 
minute incubation time [146]. Since anionic dendrimers are much les toxic they can be 
used at higher concentrations and longer incubation times without causing damage to the 
intestinal cels. 
In addition to cel viability asays, Kitchens et al. investigated the impact of 
dendrimer treatment on the morphology of diferentiated intestinal cels, specificaly 
 
61 
 
 
examining the efect on the microvili by TEM [18]. G2 cationic dendrimers, with 16 
amine terminal groups, did not cause appreciable damage to the cel monolayer, even up 
to 1 mM concentration.  In contrast, G4 dendrimers, which have 64 amine terminal 
groups, showed significant toxicity at 1 mM. When the concentration of G4 dendrimers 
was reduced to 0.01 mM, they were non-toxic, ilustrating that G4 dendrimers can be 
used safely at lower concentrations. Finaly, even high generation G3.5 anionic 
dendrimers, which have 64 terminal carboxylic acid groups, did not cause a change in 
cel morphology up to 1 mM, indicating the low intrinsic toxicity of anionic dendrimers. 
Thus, in choosing a dendrimer carier for oral drug delivery, one must choose a 
generation and concentration that maximizes transport, while minimizing toxicity. In 
addition, it is critical to first ases the toxicity of the dendrimers before asesing 
permeability since a toxic treatment could compromise the integrity of the monolayer, 
causing artificialy high permeability measurements. 
 
2.8.3 Surface Modification of PAMAM Dendrimers 
 Because of their large number of terminal groups, PAMAM dendrimers have been 
modified with diferent ligands in order to alter their biological properties including 
cytotoxicity and permeability. Jevprasesphant and co-workers conjugated lauroyl 
chloride chains to cationic dendrimers and measured their toxicity towards Caco-2 cels 
and their permeability across Caco-2 cel monolayers [17]. Modification with lauroyl 
chains decreased the cytotoxicity of cationic dendrimers and increased their permeability 
as wel as their tight junction opening ability. It was hypothesized that the significant 
reduction in cytotoxicity was observed because the lauroyl chains were able to shield 
 
62 
 
 
some of the positive charges. In a similar study, Kolhatkar et al. investigated the impact 
of modifying G2 and G4 dendrimers with acetyl groups, efectively neutralizing half or 
al of the surface charges [149]. Acetylation of G2 and G4 dendrimers significantly 
reduced their cytotoxicity towards Caco-2 cels, with full acetylation causing the greatest 
decrease in cytotoxicity. In addition, the partialy acetylated dendrimers showed 
comparable permeability to unmodified dendrimers, indicating that acetylation can 
reduce cytotoxicity without compromising permeability. These studies ilustrate that 
surface modification of PAMAM dendrimers can be a powerful tool to modulate both 
cytotoxicity and transepithelial transport to create an optimized oral drug delivery system. 
 
2.8.4 PAMAM Dendrimer Internalization 
 Because dendrimers are thought to cross the epithelial barier by transcelular and 
paracelular routes, the mechanisms of dendrimer internalization in Caco-2 cels and 
subsequent traficking have been studied. After El-Sayed et al. established the energy 
dependent celular internalization mechanism of G2 dendrimers [147], Kitchens and 
colleagues further evaluated the celular internalization of G1.5 and G2 dendrimers in 
Caco-2 cels and monitored their colocalization with markers for clathrin, early 
endosomes and lysosomes [18]. Both G1.5 and G2 dendrimers were internalized after 20 
minutes and showed colocalization with al thre markers, indicating that they are likely 
internalized by clathrin-mediated endocytosis. Interestingly, G1.5 and G2 dendrimers 
showed slightly diferent traficking paterns, with G1.5 dendrimers showing greater 
colocalization with the lysosomes at earlier time points, ilustrating the importance of 
surface charge on dendrimer traficking.  Importantly, since dendrimers colocalize with 
 
63 
 
 
early endosomes and lysosomes, these studies suggest that these organeles could 
potentialy be targeted for drug release. 
 
2.8.5 PAMAM Dendrimers as Oral Drug Delivery Systems 
 While there has been significant progres on studying the transport characteristics 
of PAMAM dendrimers, there have been comparatively few studies using dendrimers to 
improve the oral bioavailability of therapeutics. D?Emanuele et al. investigated G3 
dendrimers to improve the oral bioavailabilty of Propranolol, which has low water-
solubility and is a P-gp eflux transporter substrate [113]. By conjugating Propranolol to 
the G3 dendrimers, the apical to basolateral flux of Propranolol increased while the 
basolateral to apical transport decreased, indicating that G3 dendrimers were able to 
prevent Propranolol eflux by P-gp. This study ilustrates that PAMAM dendrimers can 
be used to both improve drug solubility and avoid eflux, thus improving overal 
bioavailabilty. 
 One of the most promising areas for drug delivery is directing chemotherapy to 
the site of action. Chemotherapeutics often have poor bioavailabilty due to low water 
solubility and can also cause intestinal toxicity. In 2008, Kolhatkar et al. reported the 
complexation of PAMAM G4 with SN38, which enhanced the solubility, transepithelial 
transport and celular uptake of SN38 relative to fre drug, suggesting its potential as an 
oral drug delivery system [150]. Ke and co-workers examined the ability of G3 
dendrimers to improve the oral bioavailability of Doxorubicin, an anthracycline antibiotic 
used to treat a wide range of cancers [151]. The dendrimer-drug complexes were found to 
enhance Doxorubicin celular uptake and transport across rat intestinal tisue compared to 
 
64 
 
 
fre drug. In addition, oral bioavailability of Doxorubicin in rats was increased 200-fold 
relative to fre drug. These studies ilustrate the utility of dendrimers to improve the oral 
bioavailabilty of chemotherapeutics. 
 
2.8.6 Designing PAMAM Dendrimer-Drug Conjugates for Oral Delivery 
 Combining our knowledge of PAMAM dendrimer transepithelial transport with 
the chalenges asociated with the oral delivery of chemotherapeutics, we can begin to 
envision the requirements for an optimized dendrimer-drug oral delivery system. Several 
elements of the system wil afect the ultimate performance of the DS including the 
choice of dendrimer generation, drug and drug linker. As described in Section 2.8.1 there 
have been many studies examining the impact of size and charge on dendrimer transport. 
While cationic dendrimers show enhanced transport, their cytotoxicity restricts their use 
to lower concentrations. Therefore, anionic dendrimers, which have very low intrinsic 
toxicity and have shown promising in vitro results in everted rat intestinal sacs as wel as 
Caco-2 cel monolayers, may be a beter choice for the carier.  
 The choice of drug cargo is also critical for having a succesful dendrimer drug 
delivery system. In addition to having low solubility and poor bioavailabilty, the drug 
must have functional groups such as amines, acids or alcohols available for conjugation 
to the dendrimer. The drug must be highly potent, ensuring that even if only a fraction of 
the administered dose is transported across the intestinal barier that the ultimate 
concentration in the bloodstream is stil within the therapeutic window of the drug. 
 The final element of a succesful dendrimer drug delivery system is the drug 
linker. This linker must be designed such that it is stable in the gastrointestinal tract, but 
 
65 
 
 
is released at the site of action in the tumor environment. In order to stay intact through 
the gastrointestinal environment, the linker must be stable under low pH in the stomach 
(pH 1-2) as wel as the elevated pH in the smal intestine (7.5-8). In addition, the linker 
must be resistant to the peptidases in the gastric environment and in the brush border of 
the intestinal mucosa. However, in order for the drug to be active, the fre drug must be 
released at the site of action. Finaly, the linker chemistry can be adjusted by the use of a 
spacer. A spacer can alow for increased aces of enzymes to the bond of interest, 
alowing for release in the tumor environment. Selection of the proper dendrimer, drug, 
linker and spacer has the potential to create a smart dendrimer drug delivery system, 
which can enhance the oral bioavailability of chemotherapeutics, thus improving patient 
lives and treatment.  
 
2.9 Colorectal Cancer 
2.9.1 Prevalence 
Acording to the American Cancer Society Facts and Figures 2010, there are 
more than 140,000 new cases of colorectal cancer each year and over 51,000 deaths 
projected in 2010 [152]. Colorectal cancer is the third most diagnosed cancer in men and 
women and the third leading cause of cancer death in the United States [153]. Although 
development of colorectal cancer is usualy sporadic, risk factors include increasing age, 
male gender, diseases such as diabetes and inflamatory bowl syndrome and 
environmental factors including high fat / low fiber diets, excesive alcohol consumption, 
smoking, obesity and a sedentary lifestyle. In addition, approximately 6% of colorectal 
cancer cases are thought to be geneticaly inherited [154]. Colorectal cancer places a 
 
66 
 
 
significant burden on the healthcare system, comprising approximately 12% of cancer-
related healthcare costs in the United States with total costs estimated betwen $4.5 and 
$9.6 bilion dollars per year [153]. Reduction in treatment costs by novel therapies could 
improve the quality of life of patients and dramaticaly reduce the burden on the 
healthcare system.   
 
2.9.2 Colorectal Cancer Screning  
Curent colorectal cancer screning guidelines recommend testing for men and 
women over the age of 50. Colorectal cancer screning tests include fecal occult blood 
tests (FOBT), sigmoidoscopies and colonoscopies, which are recommended for every 
one, five and ten years, respectively [154]. An FOBT detects blood in a stool sample, 
which may be an indication of colorectal cancer. It is the simplest and lowest cost test, 
but positive results can indicate other pathologies such as hemorrhoids, anal fisures, 
polyps, inflamatory bowl disease or Crohn?s disease. For this reason, a positive FOBT 
result is followed up with a more extensive test such as a colonoscopy. A sigmoidoscopy 
is an endoscopic examination of the rectum and the lower portion of the colon where 
60% of colorectal cancers occur. A sigmoidoscopy involves relatively litle patient 
preparation, only requiring self-administration of a single enema. A colonoscopy is the 
most extensive screning test and involves an endoscopic examination of both the upper 
and lower colon, requiring extensive patient preparation 48 hours before the procedure 
and a specialized doctor to perform the exam.  While it is both invasive and expensive, a 
colonoscopy is also the most sensitive diagnostic tool and is becoming the standard of 
care in the United States. Al of these screning techniques have been found to be cost 
 
67 
 
 
efective ways of reducing the incidence of colorectal cancer by alowing physicians to 
remove polyps before they become cancerous and reducing mortality by diagnosing the 
cancer at an earlier, more treatable stage [153].  
 
2.9.3 Colorectal Cancer Diagnosis and Staging 
 After colorectal cancer is detected by a colonoscopy, it is further characterized by 
a biopsy of the tumor as wel as a CT scan to detect any metastatic sites [154]. The 
cancer is then staged acording to thre separate factors, known as the TNM 
clasification system. The ?T? or tumor factor, ranks the invasivenes of the primary 
tumor. The ?N? or lymph node factor, ranks the number of regional lymph nodes that 
have been invaded. Finaly, the ?M? or metastasis factor, ranks the presence and number 
of distant metastases [154]. These thre scores are combined to yield an overal stage, 
which can be used to provide the patient with a five year prognosis. For example, in 
Stage 1 where the primary tumor is adherent to the submucosa but there are no lymph 
node invasions or distant metastases, the five-year survival rate is over 97%. In contrast 
in late Stage II where many regional lymph nodes show metastatic disease, the five year 
survival rate is les than 30% [154]. These dramatic survival diferences based on the 
stage of disease highlight both the importance of early detection and the great need for 
therapies to treat metastatic disease. 
 
2.9.4 Colorectal Cancer Metastasis and Treatment 
 The most common site of colorectal cancer metastasis is the liver, with over 20% 
of patients presenting liver metastases at the time of diagnosis and 50% in the duration of 
 
68 
 
 
the disease [23]. In addition, liver metastases are responsible for two-thirds of al 
colorectal cancer deaths, indicating the grave need for novel treatments for this condition. 
Unlike early stage colorectal cancer where the tumor is confined to a smal area and can 
be removed by surgery, hepatic metastases are much more dificult to remove and often 
require a combination of surgery and chemotherapy. After a patient is diagnosed with 
hepatic metastases, the size and location of the tumors are measured to determine 
resectability. Patients must be left with 20-40% of the initial liver volume in order to 
have a functioning liver after surgery [23]. If the metastatic tumors are smal, the tumors 
can be resected imediately and folowed by adjuvant chemotherapy to prevent 
recurrence. If the tumors are initialy unresectable, patients are often given neoadjuvant 
chemotherapy to shrink the tumors, making them removable.  
 Several diferent chemotherapy drugs have been approved for the treatment of 
colorectal cancer hepatic metastases. 5-fluorouracil was first approved by the FDA in 
1962 and has played an important role in the treatment of metastatic colorectal cancer. 
Originaly used as a single agent and now combined with leucovorin to enhance activity, 
5FU works by disrupting RNA synthesis and inhibiting the enzyme thymidylate synthase 
[155]. In 2001, Capecitabine, an oral prodrug of 5FU, was approved and has shown 
higher response rates than IV administered 5FU in patient groups with advanced disease 
[156]. Recently, combination therapies using 5FU/Leucovorin coupled either with 
Irinotecan, a topoisomerase inhibitor (FOLFIRI), or Oxaliplatin, a platinum-based 
cytotoxic agent (FOLFOX), have become the standard of care for advanced metastatic 
disease. While there is some debate on which treatment regimen is beter, they both 
show significant increases in survival time compared to single agent therapy and are 
 
69 
 
 
often used together as first and second line treatments [155]. In addition, both of these 
treatment regimens have shown great potential for reducing tumor size in a neoadjuvant 
seting as wel as preventing tumor recurrence when used after surgery. Depending on 
the patient, biological agents such as Cetuximab, an anti-VEGF therapy or Bevacizumab, 
an anti-EGFR antibody, can be added to the treatment regimens to improve response. 
Coupled with surgical resection, chemotherapy plays a key role in the treatment of 
colorectal cancer hepatic metastases and serves to extend patient life and reduce potential 
relapse.  
 
2.10 SN38 
 7-Ethyl-10-hydroxy-camptothecin, or SN38, is the active metabolite of Irinotecan, 
a water-soluble analogue of camptothecin that is commonly used for treatment of 
metastatic colorectal cancer. Camptothecin, a naturaly occurring alkaloid extract from 
the plant Camptotheca acuminata, was first discovered in the 1960?s [21]. Camptothecin 
showed promising anti-tumor activity in vitro and appeared to strongly inhibit DNA and 
RNA synthesis. By the 1980?s, topoisomerase I, an enzyme required for DNA coiling 
and uncoiling, was identified as the target of camptothecin. Since then, many diferent 
camptothecin analogues have been synthesized, including Irinotecan, and Topotecan 
which have been approved to treat colorectal and ovarian cancers, respectively, as wel as 
many other analogues in clinical trials and preclinical development [21].  
 
 
70 
 
 
2.10.1 Mechanism of Action of Irinotecan and SN38 
 In order to cause anti-tumor activity, Irinotecan must be cleaved by 
carboxylesterase to release SN38 (Figure 2.10) [157].  Carboxylesterases are present in 
many sites of the body with the highest expresion in the liver, where majority of fre 
SN38 is released. After release, SN38 acts as a topoisomerase poison, efectively 
inhibiting the action of topoisomerase I, which is heavily involved in DNA and RNA 
synthesis. Importantly, the lactone ring on SN38 and CPT-11 is in dynamic equilibrium 
betwen the closed ring lactone form (active) and the open ring carboxylate form 
(inactive) (Figure 2.10). Interconversion of the lactone and carboxylate forms is pH 
dependent with the lactone form favored at low pH and in the presence of human serum 
albumin. Finaly, SN38 metabolism ocurs by glucoronidation by UDPGA and UDPGT, 
converting SN38 to SN38-glucoronide [157]. 
 SN38 has significantly higher activity than the parent Irinotecan prodrug. In 
particular, SN38 has been shown to have greater topoisomerase I inhibition, more 
significant reduction in DNA and RNA synthesis and an increase in DNA strand breaks 
relative to Irinotecan at equivalent concentrations [158]. This highlights the importance 
of fre SN38 release from the prodrug for anticancer activity. Designing a prodrug of 
SN38 with more eficient release properties would have the potential to create a drug that 
is more efective than Irinotecan. 
 
71 
 
 
 
Figure 2.10. Conversion of Irinotecan to SN38 by Carboxylesterase and Equilibrium of 
Carboxylate and Lactone forms of SN38 and Irinotecan. The closed ring/ lactone form of 
SN38 is required for anticancer activity. (Adapted from [157].) 
 
72 
 
 
2.10.2 Current SN38 Drug Delivery Systems 
 While SN38 shows 100-1000 fold greater in vitro activity than Irinotecan, it has 
had limited succes in the clinic due to poor water solubility and significant 
gastrointestinal side efects. Several drug delivery systems have been synthesized to 
overcome these limitations. Zhao and colleagues reported using a 4-arm-PEG system to 
improve the water solubility of SN38 [159]. Using several diferent amino acid spacers 
to alter stability, these SN38-PEG conjugates showed favorable release profiles under 
neutral pH conditions, high eficacy in vitro and in vivo in a mouse xenograft model. In 
addition, PEG conjugation at the 20-OH position on SN38 helped to stabilize the lactone 
ring until the point of release. The PEG-SN38 conjugate using a glycine spacer showed 
the most promising therapeutic profile and is currently in Phase I clinical trials (ENZ-
2208) [44]. Meyer-Losic et al. reported a cationic peptide-linked SN38 prodrug designed 
to reduce the gastrointestinal toxicity and hepatic metabolism of Irinotecan. The peptide 
was linked to SN38 in the 10-hydroxyl position by an ester bond [160]. The conjugates 
were tested in vitro and in vivo in a human xenograft model in mice and dogs. In 
addition to showing promising in vitro toxicity against a variety of colon, lung and breast 
cel lines, the SN38-peptide conjugate showed higher activity than Irinotecan in vivo with 
fewer gastrointestinal side efects.  
To date there have been two dendrimer-based drug delivery systems developed 
for SN38. As described in Section 2.8.5, Kolhatkar and co-workers reported the use of 
G4 dendrimers complexed with SN38 through non-covalent interactions as a potential 
oral drug delivery system [150]. The complexes enhanced the solubility, transepithelial 
transport and celular uptake of SN38 relative to fre drug. PAMAM-SN38 complexes 
 
73 
 
 
released 40% of the drug within 24 hours in PBS and 90% of the drug within 30 minutes 
at pH 5. Because of the instability in acidic conditions, these dendrimer-SN38 complexes 
would require enteric coating for use in oral administration.  In addition, Vijayalakshmi 
et al. recently reported conjugation of SN38 via glycine and ?-alanine spacers to 
carboxylic-acid terminated G3.5 PAMAM dendrimers at the 20-OH position [161]. 
These G3.5-SN38 conjugates showed promising in vitro activity against colorectal 
adenocarcinoma HCT-116 cels and produced nuclear fragmentation and cel cycle arest 
in the G2/M phase similar to fre drug. The conjugates were stable in PBS, and showed 
up to 20% release in cel culture media and rat plasma during a 72-hour incubation 
period. These initial studies established the potential of PAMAM dendrimers as drug 
delivery systems of SN38 for the treatment of colorectal cancer. Importantly, PAMAM 
dendrimers have the potential to not only improve the oral bioavailability of SN38, but 
they can also reduce its non-specific toxicity and enhance acumulation at the tumor site. 
Taken together, these studies ilustrate the potential for improvement of the therapeutic 
eficacy of SN38 by using PAMAM dendrimers. 
 
2.11 Unresolved Isues in Oral Delivery by Dendrimers 
Despite the wealth of research on dendrimers as oral drug cariers, there are stil 
many remaining questions that must be resolved for their succesful use in oral drug 
delivery. While PAMAM dendrimers are thought to be transported by a combination of 
transcelular and paracelular pathways, the mechanism of tight junction opening and 
precise endocytosis and traficking pathways are stil unknown. In depth knowledge of 
transepithelial transport pathways of PAMAM dendrimers wil help to beter design these 
 
74 
 
 
constructs as oral drug cariers. In addition, while there is a significant body of work on 
the transepithelial transport of native dendrimers, there have been comparably few studies 
on dendrimer-drug conjugates.  Detailed studies of the transport, celular uptake and 
toxicity of PAMAM dendrimer-drug conjugates are necesary to further establish their 
utility in oral drug delivery. In addition, chemical linkers that promote stability of the 
dendrimer-drug conjugates in the gastrointestinal milieu while favoring release of drug in 
the presence of carboxylesterase must be developed to achieve a functional delivery 
system for the treatment of colorectal cancer hepatic metastases. Finaly, the impact of 
surface chemistry and conjugation chemistry on the degre and mechanism of dendrimer 
transport need to be investigated.  
 
 
75 
 
 
 
Chapter 3 : Celular Entry of G3.5 Poly (amido amine) Dendrimers by 
Clathrin- and Dynamin-Dependent Endocytosis Promotes Tight Junctional 
Opening in Intestinal Epithelia 
 
3.1 Introduction 
 As discussed in Chapter 2, poly (amido amine) (PAMAM) dendrimers have shown 
promise as drug delivery cariers due to their unique physical properties including 
nanoscale size and near monodispersity [114]. With each increase in dendrimer 
generation, the diameter increases linearly while the number of surface groups increases 
exponentialy, creating high density surface groups that can be conjugated to drugs, 
imaging agents and targeting moieties, making dendrimers versatile multifunctional 
nanocariers [12, 24]. Reports from our laboratory [15, 16, 145, 146] and others [17, 19, 
20, 162] have demonstrated that PAMAM dendrimers in a specified size and charge 
window can efectively cross the epithelial layer of the gut, showing potential as oral 
drug delivery cariers. Conjugation or complexation of therapeutics with poor solubility 
and low permeability to water-soluble dendrimers that can permeate the epithelial layer 
of the gut has the potential to render these drugs oraly bioavailable [113, 150, 151]. Oral 
drug administration has many advantages including the convenience of at-home 
administration, reduction of direct and indirect costs, and a more flexible dosing regimen, 
resulting in higher patient compliance and a lower burden on hospitals and the healthcare 
system [8].  
 
76 
 
 
 PAMAM dendrimers are known to cross the epithelial barier by a combination of 
transcelular and paracelular mechanisms. As described in Chapter 2, transport of 
cationic dendrimers has been found to be energy-dependent and is reduced in the 
presence of endocytosis inhibitors, signifying the importance of endocytosis in dendrimer 
transport [148]. In addition, dendrimers were found to colocalize with clathrin and early 
endosome markers, suggesting the involvement of clathrin-mediated endocytosis in 
dendrimer internalization [18].  Dendrimers have also been reported to interact with tight 
junctions, transiently opening them to alow for paracelular transport [15]. While there 
has been significant progres in understanding the mechanisms by which dendrimers 
enter cels and are transported across the epithelial barier, several important questions 
remain to be addresed. In particular, the contributions of diferent endocytosis pathways 
to dendrimer celular uptake and transcytosis and the specific mechanisms by which 
dendrimers open tight junctions have yet to be elucidated.  Additionaly, the majority of 
mechanistic studies to date have focused on cationic dendrimers, which are promising 
due to their high transport rates, but are limited by their significant cytotoxicity. Anionic 
dendrimers show comparably low cytotoxicity but stil appreciable transport rates, 
making them wel suited for oral delivery [27, 144].  
 In this Chapter, the mechanisms of celular uptake, transepithelial transport and 
tight junctional modulation of anionic G3.5 PAMAM dendrimers are investigated by 
examining the impact of endocytosis inhibitors on dendrimer interaction with Caco-2 
cels and diferentiated Caco-2 monolayers. In addition, we present the intracelular 
traficking of dendrimers from endosomes to lysosomes over time. Knowledge of the 
 
77 
 
 
specific pathways of endocytosis, intracelular traficking and transepithelial transport 
wil aid in rational design of dendrimers for oral delivery. 
 
3.2 Materials and Methods 
3.2.1 Materials 
G3.5 PAMAM dendrimers (reported molecular weight=12,931), lucifer yelow 
CH dipotasium salt (LY), oregon gren carboxylic acid succinimidyl ester (OG), 
monodansyl cadaverine (MDC), phenylarsine oxide (PAO), filipin (FIL), genistein 
(GEN) and dynasore (DYN) were purchased from Sigma Aldrich (St. Louis, MO). 
Superose 12 HR 10/300 GL column was obtained from Amersham Pharmacia Biotech 
(Piscataway, NJ) and WST-1 cel proliferation reagent from Roche Applied Sciences 
(Indianapolis, IN). Caco-2 cels were obtained from American Type Cel Culture 
(Rockvile, MD). 
 
3.2.2 Synthesis of G3.5-OG 
Purified G3.5 PAMAM dendrimers were first modified with pendant primary 
amine groups to facilitate OG labeling [144]. 50 mg dendrimer was disolved in 
deionized (DI) water to a final concentration of 10 mg/ml and the pH was adjusted to 6.5. 
1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC) was added at a 5:1 molar ratio 
to the dendrimer and stired for 30 minutes at room temperature, after which ethylene 
diamine was added at a 5:1 molar ratio and stired for 4 hours at room temperature. The 
sample volume was reduced to 1 ml by rotoevaporation of water and then run through a 
PD10 column folowed by purification with an amicon centrifugal filter (MWCO 4000) 
 
78 
 
 
to remove the EDC and ethylene diamine. Size exclusion chromatography was used to 
confirm the absence of low molecular weight impurities. The number of amines per 
dendrimer was determined to be 2.5 by the ninhydrin asay. 10 mg of amine-modified 
dendrimer was disolved in 25 ml of DI water and 1 equivalent of OG per dendrimer was 
added and stired for 30 minutes. The water was removed by rotoevaporation and the 
product was redisolved in methanol and added dropwise to diethyl ether to precipitate 
the G3.5-OG conjugate. The solution was centrifuged, the ether decanted and then the 
precipitate was dried overnight under vacuum. The precipitate was then redisolved in 1 
ml of water, and pased through a PD10 column to remove unreacted OG. Size exclusion 
chromatography was used to confirm the absence of fre dye. OG content was 
determined by a fluorescence standard curve (?
excitation
= 485 nm, ?
emision
= 535 nm) to be 
0.75 molecules of OG per dendrimer on average.  
 
3.2.3 Caco-2 Cel Culture 
Caco-2 cels (pasages 30?45) were grown at 37?C in an atmosphere of 95% 
relative humidity and 5% CO
2
. Cels were maintained in T-75 flasks using Dulbeco's 
Modified Eagle's Medium (DMEM) suplemented with 10% fetal bovine serum (FBS), 
1% non-esential amino acids, 10,000 units/mL penicilin, 10,000 ?g/mL streptomycin 
and 25 ?g/mL amphotericin B. Media was changed every other day and cels were 
pasaged at 80?90% confluence using a 0.25% trypsin/ethylenediamine tetracetic acid 
(EDTA) solution. Incubation bufer used in asays consisted of Hank's balanced salt 
solution  (HBS), suplemented with 1.0 mM N-(2-hydroxyelthyl)piperazine-N?-179 (2 
ethanesulfonic acid) hemisodium salt (HEPES) bufer (pH 7.4). 
 
79 
 
 
3.2.4 Cytotoxicity of Endocytosis Inhibitors 
Potential short-term cytotoxicity of endocytosis inhibitors was asesed in Caco-2 
cels to ensure cel viability during uptake and transport asays. Chemical inhibitors were 
prepared at a range of concentrations known to reduce clathrin-mediated endocytosis 
(phenylarsine oxide (1-20 ?M), monodansyl cadaverine (100-300 ?M), caveolin-
mediated endocytosis (filipin (1-4 ?M), genistein (100-300 ?M) or dynamin-dependent 
endocytosis (dynasore (40-80 ?M) [163-165]. Cytotoxicity of the inhibitors was 
asesed by the water soluble tetrazolium salt (WST-1) asay. Caco-2 cels were seded 
at 50,000 cels per wel in 96 wel cel culture plates (Corning, Corning, NY) and 
maintained at 37?C, 95% relative humidity and 5% CO
2
 for 48 hours. Cels were washed 
with warm HBS buffer and incubated for 2 hours with 100 ?L of varying concentrations 
of endocytosis inhibitors. After 2 hours, the inhibitor solutions were removed and the 
cels were washed with warm HBS buffer. 10 ?L WST-1 cel proliferation reagent in 
100 ?L of HBS buffer was added to each wel and incubated for 4 hours at 37?C. 
Absorbance at 460 nm and background at 600 nm were measured using a SpectraMax 
384 plate reader (Molecular Devices, Sunnyvale, CA). HBS was used as a negative 
control for 100% cel viability. Cytotoxicity of inhibitor concentration was asesed in 
four replicates. Cel viability of greater than 85% was clasified as aceptable for uptake 
and transport asays.  
 
3.2.5 Celular Uptake 
Celular uptake of G3.5-OG dendrimers was determined in the presence and 
absence of endocytosis inhibitors. Inhibitors were used at concentrations that showed a 
 
80 
 
 
minimum of 85% cel viability during the 2-hour asay period. Caco-2 cels were seded 
at 300,000 cels per wel in 12 wel plates (Corning, Corning, NY) and maintained for 48 
hours at 37?C, 95% relative humidity and 5% CO
2
. Cels were washed with HBS and 
pretreated with endocytosis inhibitors or HBS for 1 hour at 37?C. Endocytosis 
inhibitors were removed and 25 ?M G3.5-OG, 5 ?M Transferin-AF488 (Molecular 
Probes, Carlsbad, CA) or 5 ?M Cholera Toxin B-AF488 (Molecular Probes, Carlsbad, 
CA) was added in the presence of endocytosis inhibitor solutions or HBS for one hour 
at 37?C. Cels were washed once with cold HBS, trypsinized for 5 minutes and then 
complete cel culture media was added to halt the trypsinization proces. The cels were 
removed from plates, collected in microcentrifuge tubes and were centrifuged for 5 
minutes at 1,000 rpm after which the supernatant was removed. The cels were washed 
with PBS and finaly fixed in 1% paraformaldehyde in PBS at a final concentration of 
500,000 cels/ml. Flow cytometry was used to ases celular fluorescence using a BD 
LSR I flow cytometer (Becton Dickenson, Franklin Lakes, NJ) with a 530/30 bandpas 
filter. Twenty-five thousand to forty-five thousand events were collected per sample. 
Percent uptake was determined for two diferent cel populations by the shift in mean 
fluorescence in the presence of endocytosis inhibitors compared to HBS control. 
 
3.2.6 Colocalization and Intracelular Trafficking 
Caco-2 cels were seded at 40,000 cels/cm
2
 on colagen-coated 8-chamber 
slides. Slides were used when cels were 90% confluent, typicaly 4-5 days after seding. 
Cels were washed with DPBS and then incubated in DPBS for 30 minutes at 37?C. Cels 
were treated with G3.5-OG (1 ?M) or Transferin-AF488 (250 ?g/ml) for 30 minutes at 
 
81 
 
 
4?C to alow for atachment but not internalization (pulse). Subsequently, the cels were 
washed thre times with ice cold DPBS to remove unbound ligand and incubated with 
warm DPBS at 37?C for 5, 15, or 30 minutes (chase) after which they were fixed with 4% 
paraformaldeyde, 4% sucrose in DPBS for 20 minutes. Al subsequent steps were caried 
out at room temperature. The cels were washed twice with 25 mM glycine and then 
once with DPBS, permeabilized with 0.2% Triton X-100 in blocking solution (3% 
Bovine Serum Albumin (BSA)/DPBS) and then incubated with blocking solution to 
prevent non-specific binding. Primary antibodies for early endosomes (rabbit polyclonal 
early endosome antigen-1 (EA-1)) and lysosomes (rabbit polyclonal lysosome-
asociated membrane protein 1 (LAMP-1)) (Molecular Probes, Carlsbad, CA) were 
added to separate chambers and incubated for one hour. The antibodies were removed, 
cels were washed thre times with blocking solution and Alexa Fluor-568 goat anti-
rabbit IgG (Molecular Probes, Carlsbad, CA) was added at 1:400 in blocking solution for 
one hour. The cels were then washed thre times with DPBS, and incubated with 300 
nM 4',6-diamidino-2-phenylindole (DAPI) 10 minutes to stain the nuclei. The cels were 
then washed once with DPBS, once with DI water and the chambers were removed. The 
slides were mounted, covered with glas coverslips, alowed to dry for a minimum of 2 
hours before sealing and stored at 4?C prior to visualization. 
Images were acquired using a Nikon Eclipse TE2000 inverted confocal laser 
scanning microscope (Nikon Instruments, Melvile, NY). Excitation and emision 
wavelengths for DAPI, OG/ AF488 and AF568 were 405/450, 488/515 and 543/605, 
respectively. Four Z-stacks were obtained for each treatment using the following 
microscope setings: 60x oil objective, 2x optical zoom, 60 ?m pinhole and 512 x 512 
 
82 
 
 
image size. Z-stacks contained 41, 0.5 ?m slices to encompas the entire cel layer.  
Detector gains were set to be constant betwen samples to facilitate sample comparison.  
 Colocalization betwen G3.5-OG or Transferin-AF488 with early endosomes 
and lysosomes was quantified using Volocity 3D Imaging software (Improvision, 
Lexington, MA). The extent of colocalization betwen the gren and red channels (M
x
) 
was calculated by the software using the following equation: 
!
=
i
i
col
x
,
 (Eq. 3.1)
 
where x
i,coloc
 is the value of voxel i of the overlapped red and gren components and x
i
 is 
the value of the gren component. M
x 
is reported for each treatment as an average of the 
four regions. Transferin-AF488 was used as an endocytosis control ligand to establish 
the validity of the asay methods for monitoring intracelular traficking over time. 
 
3.2.7 Transepithelial Transport 
Caco-2 cels were seded at 80,000 cels/cm
2
 onto polycarbonate 12-wel 
Transwel
?
 filters of 0.4 ?m ean pore size with 1.0 cm
2
 surface area (Corning, Corning, 
NY). Caco-2 cels were maintained under standard incubation conditions where media 
was changed every other day and cels were used for transport experiments 21-25 days 
post-seding. Prior to experiments, the transepithelial electrical resistance (TER) of 
each monolayer was measured with an epithelial voltohmmeter (World Precision 
Instruments, Sarasota, FL). Monolayers with TER > 600 ??cm
2
 were used for asays. 
Cel monolayers were washed with HBS and then incubated in the presence of HBS or 
endocytosis inhibitors for 1 hour at 37?C. Inhibitors were used at concentrations that 
 
83 
 
 
showed a minimum of 85% cel viability during the 2-hour asay period. The solutions 
were removed and G3.5-OG (10 ?M) was added to the apical compartment in the 
presence of HBS or endocytosis inhibitor and the corresponding solution was added to 
the basolateral compartment. After incubating for one hour, samples were taken from the 
basolateral compartment. Transport was quantified by measuring fluorescence in the 
basolateral compartment using a SpectraMax Gemini XS spectrofluorometer (Molecular 
Devices, Sunnyvale, CA) with excitation and emision wavelengths of 485 and 535 nm, 
respectively. Percent transport is reported comparing dendrimer transport in the presence 
of inhibitors to dendrimer transport in HBS alone. The standard deviation for each 
reported value is calculated using propagation of eror for the quotient of two 
experimentaly determined values. 
LY permeability was also monitored in the presence of HBS and endocytosis 
inhibitors to ensure the integrity of the monolayers. LY apparent permeability was les 
than 1 x 10
-6
 for al conditions tested, which is within the acepted range of LY 
permeability for diferentiated monolayers [166]. There was not a significant diference 
betwen LY permeability in the presence of HBS or inhibitors, confirming that 
endocytosis inhibitors do not afect tight junctional integrity (data not shown). 
 
3.2.8 Caco-2 Monolayer Visualization and Ocludin Staining 
After acquisition of samples for transport asays, Caco-2 cel monolayers were 
prepared for visualization by confocal microscopy. In particular, Caco-2 cel monolayers 
treated with G3.5-OG or HBS in buffer or in the presence of dynasore were analyzed for 
occludin acesibility. The monolayers were washed twice with ice cold PBS, fixed, 
 
84 
 
 
permeabilized and blocked by the same procedures used in the imunofluorescence 
studies (Section 3.2.6). Subsequently, monolayers were treated with mouse anti-occludin 
(2 ?g/ml) overnight at 4?C. The next day, the monolayers were washed twice with 
blocking solution and incubated with the same solution for 30 minutes. Alexa Fluor 568 
goat anti-mouse IgG (1:400) was added in blocking solution for one hour. The cels were 
then washed and stained with DAPI. The membranes were carefully excised from the 
Transwel
?
 supports using a scalpel, mounted on glas slides, covered with a glas 
coverslip and then sealed with clear nail polish. Slides were stored at 4?C prior to 
visualization. 
Images were acquired using a Nikon Eclipse TE2000 inverted microscope using 
the same setings as described in Section 3.2.6. To visualize the G3.5 dendrimer 
interaction with the cel monolayer, Z-stacks were obtained with 51 0.5 ?m slices using 
red, gren and blue channels. To quantify occludin staining, four z-stacks were obtained 
per region, and only the red channel was used. Images were procesed using Volocity 
software.  Red voxels, corresponding to occludin staining, were quantified by 
thresholding the intensity betwen 20% and 100%. The number of red voxels was 
quantified for four z-stacks for each treatment. Results are reported as mean ? standard 
deviation and statistical significance was determined by a one-way analysis of variance 
with Bonferoni post-hoc correction. 
 
 
85 
 
 
3.3 Results 
3.3.1 Cytotoxicity of Endocytosis Inhibitors 
Five endocytosis inhibitors were chosen to examine the pathways of celular 
uptake and transepithelial transport of G3.5 PAMAM dendrimers. Before initiating 
uptake and transport studies, Caco-2 cel viability during the 2-hour asay time was 
confirmed in the presence of these inhibitors, whose concentrations were chosen based on 
literature reported values [163-165]. Inhibitors included chemicals known to prevent 
clathrin-mediated endocytosis (phenylarsine oxide (1-20 ?M); monodansyl cadaverine 
(100-300 ?M), caveolin-mediated endocytosis (filipin (1-4 ?M); genistein (100-300 
?M) and dynamin-dependent endocytosis (dynasore (40-80 ?M). We have shown 
previously that G3.5 dendrimers do not cause a reduction in Caco-2 cel viability up to 
100 ?M. Since the maximal dendrimer concentration used in these experiments was 25 
?M, toxicity due to the dendrimers was not a concern. 
Short-term cytotoxicity of endocytosis inhibitors was asesed by the WST-1 cel 
viability asay at 5 diferent concentrations in the reported range of each inhibitor (Table 
3.1). 85% cel viability was chosen as the minimum alowable for use in uptake and 
transport asays. Monodansyl cadaverine and filipin did not show appreciable toxicity at 
any concentration tested and were used at their maximum reported concentrations of 300 
?M and 4 ?M, respectively.  Phenylarsine oxide and genistein showed significant 
toxicity and were used at their lowest efective concentrations. Dynasore showed 
increasing toxicity over the range of concentrations tested, with the most aceptable (i.e. 
>85% viability) toxicity profile at 50 ?M (Table 3.1).  
 
86 
 
 
Table 3.1. Endocytosis Inhibitor Concentration and % Cel Viability in Caco-2 Cels. 
Endocytosis Inhibitors 
Concentration 
(?M) 
% Cell Viability 
Phenylarsine Oxide (PAO) 1 90.0 ? 5.7% 
Clathrin 
Inhibiting 
Monodansyl Cadaverine (MDC) 30 92.7 ? 3.4%
 
Filipin (FIL) 4 95.9 ? 2.7% 
Caveolin 
Inhibiting 
Genistein (GEN) 10 86.7 ? 2.9% 
Dynamin 
Inhibiting 
Dynasore (DYN) 50 106.1 ? 5.0% 
Results are reported as mean +/- standard deviation (n=4). 
 
 
87 
 
 
3.3.2 Celular Uptake of G3.5-OG Dendrimers in the Presence of Endocytosis Inhibitors 
As chemical inhibitors have varied efectivenes in diferent cel lines and can be 
somewhat non-specific, we took a multi-pronged approach, selecting one inhibitor for the 
general proces of dynamin-dependent endocytosis and two inhibitors for distinct 
clathrin- and caveolin-mediated proceses. In addition, we monitored the celular uptake 
of transferin and cholera toxin B, control ligands for clathrin- and caveolin-mediated 
endocytosis, respectively, to determine the eficacy and specificity of the selected 
inhibitors. Dynasore was used to block vesicular endocytosis by selectively inhibiting 
dynamin 1 and dynamin 2 GTPases, which are responsible for vesicle scision during 
both clathrin- and caveolin-mediated endocytosis [164]. Monodansyl cadaverine and 
phenylarsine oxide were used to block clathrin-mediated endocytosis. Monodansyl 
cadaverine is known to stabilize clathrin coated pits on the cel membrane, thereby 
preventing internalization [167]. Phenylarsine oxide has also been shown to inhibit 
clathrin endocytosis at low micromolar concentrations, but its mechanism is unknown 
[168, 169]. Filipin and genistein were selected to inhibit caveolin-mediated endocytosis. 
Filipin binds cholesterol and has been shown to disrupt caveolae-mediated endocytic 
pathways [170]. Genistein inhibits protein tyrosine kinases and, amongst other efects, 
has been shown to block internalization by caveolae [171]. The percent uptake relative to 
buffer control was calculated for G3.5 and control ligands in the presence of five 
endocytosis inhibitors (Table 3.2). 
 
 
 
88 
 
 
Table 3.2. Percent Uptake of G3.5-OG Dendrimers and Control Ligands in Caco-2 Cels 
in the Presence of Endocytosis Inhibitors. 
 Percent Uptake 
Endocytosis Inhibitors G3.5 Transferrin Cholera Toxin B 
Phenylarsine Oxide (PAO) 84.4 ? 0.6% 12.3 ? 1.5% 10.5 ? 4.0% 
Clathrin 
Inhibiting 
Monodansyl Cadaverine (MDC) 57.8 ? 2.1% 65.6 ? 2.5% 61.8 ? 5.1% 
Filipin (FIL) 81.0 ? 0.2% 10.2 ? 0.1% 14.4 ? 3.4% 
Caveolin 
Inhibiting 
Genistein (GEN) 2.4 ? 0.7% 69.8 ? 2.4% 20.4 ? 0.7% 
Dynamin 
Inhibiting 
Dynasore (DYN) 20.4 ? 3.5% 15.0 ? 2.7% 57.1 ? 19.6% 
Results are reported as mean +/- standard deviation (n=2). 
 
89 
 
 
 G3.5 PAMAM dendrimers show reduction in celular uptake in the presence of al 
endocytosis inhibitors tested, suggesting the involvement of both clathrin- and caveolin-
mediated endocytosis pathways in dendrimer celular uptake. As expected, dendrimers 
showed the greatest reduction in uptake in the presence of dynasore, a selective chemical 
inhibitor of dynamin, a protein integral to the halmark event of vesicle pinching from the 
plasma membrane during receptor-mediated endocytosis. The significant decrease in 
uptake of dendrimers and control ligands in the presence of dynasore confirms the 
endocytosis of G3.5 PAMAM dendrimers by dynamin-dependent pathways. Dendrimers 
showed decreased uptake in the presence of both clathrin inhibitors, with a greater 
decrease sen in the presence of monodansyl cadaverine. Transferin uptake was not 
reduced in the presence of phenylarsine oxide, indicating that this is not an efective 
clathrin-endocytosis inhibitor in Caco-2 cels at the concentration used and the decrease 
in dendrimer uptake may be due to a non-specific efect. Alternatively, dendrimer 
traficking may be more sensitive to phenylarsine oxide compared to transferin, which is 
afected only at higher phenylarsine oxide concentrations. In contrast, transferin shows 
reduced uptake in the presence of monodansyl cadaverine, ilustrating the eficacy of 
monodansyl cadaverine in Caco-2 cels. Cholera toxin B also shows reduced uptake in 
the presence of monodansyl cadaverine, but this decrease in uptake was expected since 
cholera toxin B can be endocytosed by clathrin- and caveolin-mediated pathways in 
Caco-2 cels [172]. Dendrimers also showed reduced uptake in the presence of both 
caveolin inhibitors, markedly sen with genistein; however, filipin did not reduce uptake 
of cholera toxin B, and this may represent a cel-specific efect. While genistein reduces 
the uptake of transferin, it reduces the uptake of cholera toxin B to a much larger extent, 
 
90 
 
 
making it an efective and relatively specific inhibitor of caveolin-mediated endocytosis. 
The most efective and specific inhibitors (monodansyl cadaverine, genistein and 
dynasore) were chosen for further investigation in transport asays. 
  
3.3.3 Intracelular Trafficking 
 In addition to determining the mechanism of dendrimer uptake, we investigated 
the intracelular traficking of G3.5 PAMAM dendrimers in Caco-2 cels. Dendrimers 
colocalized with early endosomes (EA-1) and lysosomes (LAMP-1) over time (Figure 
3.1). Transferin traficking (Figure 3.2), which has been wel-characterized in this cel 
line, was used as a control for the study [18]. Sample confocal images from each 
treatment and time point are included in Appendix 1. Five minutes after incubation, 
dendrimers showed initial localization in the early endosomes and lysosomes. By 15 
minutes, however, they showed increasing localization in the lysosomes, indicating quick 
traficking to these celular compartments. Interestingly, at 30 minutes, the level of 
acumulation in the lysosomes remained unchanged, while their presence in endosomes 
increased. This suggests that the traficking pathway was saturated, causing dendrimer 
retention in early endosomes once lysosomes were occupied. In contrast, transferin 
showed almost constant presence in the early endosomes with increasing presence in the 
lysosomes over time. This is typical of transferin, a clasical ligand for clathrin-
mediated endocytosis known to acumulate in the early endosomes, confirming the 
validity of the asay methods.  
 
91 
 
 
 
 
Figure 3.1. Intracelular Traficking of G3.5-OG Dendrimers over Time in Caco-2 Cels. 
Colocalization (M
x
) with early endosomes (EA-1) and lysosomes (LAMP-1) is shown. 
Results are reported as mean +/- standard deviation (n=4). 
 
92 
 
 
 
Figure 3.2. Intracelular Traficking of Transferin-AF488 over Time in Caco-2 Cels. 
Colocalization (Mx) with early endosomes (EA-1) and lysosomes (LAMP-1) is shown. 
Results are reported as mean +/- standard deviation (n=4). 
 
93 
 
 
3.3.4 Transepithelial Transport of G3.5-OG Dendrimers in the Presence of Endocytosis 
Inhibitors 
 Transepithelial transport of G3.5-OG dendrimers was monitored in the presence 
of endocytosis inhibitors and at 4?C and compared to transport in buffer at 37?C (Figure 
3.3). Transport of PAMAM G3.5 was significantly reduced at 4?C, ilustrating strong 
energy dependence. Similar to Caco-2 uptake studies, transport was also reduced in the 
presence of dynasore and monodansyl cadaverine, indicating the importance of dynamin-
dependent and clathrin-mediated endocytosis mechanisms in transepithelial transport. 
However, contrary to its efect on celular uptake of dendrimers (Table 3.2), genistein 
does not significantly impact dendrimer transport across Caco-2 monolayers, suggesting 
that caveolin-mediated endocytosis may not play a significant role in this proces. It has 
been suggested that fully diferentiated Caco-2 cels lack caveolae [172], opening the 
possibility that, while caveolae play an important role in dendrimer endocytosis in 
undiferentiated Caco-2 cels, they are les important in dendrimer transepithelial 
transport because of their lower expresion in diferentiated enterocytes. 
 
3.3.5 Visualization of G3.5-OG Dendrimer Interaction with Caco-2 Cel Monolayers 
 After cel monolayers were used for transport asays, they were fixed and stained 
for occludin and nuclear DNA. By excising the stained membranes from the Transwel
?
 
supports, we were able to visualize the dendrimer interacting with diferentiated Caco-2 
cel monolayers. Although there have been many studies documenting dendrimer 
interaction with Caco-2 cels grown on microscope slides [18], this is the first to show 
 
 
94 
 
 
 
Figure 3.3. Percent Transport of G3.5-OG Dendrimers across Caco-2 Monolayers in the 
Presence of Endocytosis Inhibitors or at 4?C.Results are reported as mean +/- standard 
deviation (n=4). ** indicates a significant diference (p<0.01) from 100% transport 
(buffer alone). 
 
 
95 
 
 
 
Figure 3.4. Visualization of G3.5-OG Dendrimer Interaction with Caco-2 Monolayers. 
Dendrimers are localized inside cel monolayers but avoid the nucleus (A) and smal 
vesicles of G3.5 dendrimers (circled) can be sen interacting with cels (B). Scale bar = 
21 ?m. 
 
96 
 
 
interaction with fully diferentiated and confluent monolayers. Figure 3.4 shows a 
representative image of the cel monolayer. The nuclear staining was omited from 
Figure 3.4 B to alow for easier visualization of the dendrimer and tight junctions. In 
both figures, dendrimer staining is observed inside the confluent cels, confirming 
internalization. In addition, there are punctate regions strongly resembling vesicles 
(circled in Figure 3.4 B), which confirm the involvement of vesicular endocytosis in 
dendrimer transepithelial transport. Finaly, dendrimers cannot be detected in the nuclear 
region of the cel, confirming their localization in the cel interior and without further 
traficking into the cel nucleus. These images serve as complementary evidence of the 
importance of the transcelular pathway in dendrimer transport. 
 
3.3.6 Ocludin Staining in Presence of Dendrimers with and without Dynasore Treatment 
 Increased occludin staining is a wel-established indicator for tight junctional 
opening in Caco-2 cel monolayers and has shown strong correlation with reduction in 
TER and increase in paracelular marker permeability [173]. Previous studies have 
shown that monolayers treated with dendrimers showed increased occludin staining 
relative to cels treated with HBS alone, indicating that dendrimers open tight junctions 
[15]. We examined occludin acesibility in Caco-2 cel monolayers treated with 
dendrimers or HBS in the presence of dynasore or buffer alone (Figures 3.5 and 3.6). In 
cels treated with buffer, dendrimers significantly increased occludin staining relative to 
untreated cels, indicating tight junction opening (Figure 3.5 A,B). In contrast, in cels 
treated with dynasore, no diference could be detected in occludin staining betwen cels 
 
97 
 
 
 
Figure 3.5. Ocludin Staining in the Presence and Absence of G3.5-OG Dendrimers in 
Caco-2 Cels Treated with HBS or Dynasore. A) G3.5/ HBS, B) HBS only C) 
G3.5/Dynasore and D) Dynasore only. Main panels ilustrate the xy plane; horizontal bars 
ilustrate the xz plane; vertical bars ilustrate the yz plane. Scale bars equal 21 ?m. 
 
98 
 
 
 
Figure 3.6. Quantification of Ocludin Staining. Cels treated with G3.5-OG dendrimers 
in the presence of HBS show a significant increase in occludin staining from untreated 
cels. (***) indicates p<0.001. Cels treated with dendrimer in the presence of dynasore 
do not show a significant change in occludin staining relative to the control. Results are 
reported as mean +/- standard deviation (n=4). 
 
 
99 
 
 
treated with dendrimers or dynasore alone (Figure 3.5 C,D). This ilustrates that 
dendrimers are unable to open tight junctions in cels where dynamin-dependent 
endocytosis is inhibited, suggesting that dendrimers must first be internalized to modulate 
celular tight junctions.  
Two major mechanisms have been established for opening tight junctions: 
depletion of divalent cations and disruption of intracelular tight junctional components. 
Disodium ethylenediaminetetracetate, a calcium chelator, and polyoxyethelyne, a 
surfactant, both lower extracelular calcium levels, causing disociation of tight junctions 
[55].  Other compounds such as sodium caprate increase tight junctional opening by 
initiating a biochemical signaling cascade which results in the contraction of actin 
microfilaments, efectively dilating the intracelular tight junctions [55]. While it is 
possible that dendrimers interact with the tight junctions in multiple ways, it is clear that 
prevention of dendrimer endocytosis reduces both transcelular and paracelular transport, 
suggesting that dendrimer endocytosis is at least in part responsible for tight junction 
modulation. Dendrimers are most likely acting on intracelular cytoskeleton components 
to induce tight junction opening. 
 
3.4 Discusion 
 PAMAM dendrimers have shown promise as oral drug delivery cariers due to 
their ability to translocate across the epithelial layer of the gut, taking poorly-bioavailable 
drug cargo in tow [150]. While many studies have suggested that dendrimers transiently 
open tight junctions and are transported through the intestinal barier by transcelular and 
paracelular pathways [16, 19, 148], the details of these proceses are poorly understood. 
 
100 
 
 
In order to design dendrimers as oral drug delivery cariers, it is critical to know the 
detailed mechanisms of their celular entry, traficking, transport and interaction with 
celular tight junctions. In this Chapter, we have uncovered some of the details of anionic 
dendrimer transport, which can have significant implications for oral drug delivery. 
Caco-2 celular uptake of G3.5 dendrimers was found to occur primarily through 
dynamin-dependent endocytosis pathways, specificaly clathrin- and caveolin-mediated 
endocytosis. Previous reports have suggested the involvement of clathrin in dendrimer 
endocytosis [18]. The present study confirms, for the first time, the involvement of 
caveolin-mediated endocytosis in dendrimer internalization. This suggests that 
dendrimers are not relegated to a single means of celular entry, but instead take 
advantage of several specific endocytosis pathways. This has significant implications for 
drug delivery, as intracelular traficking is largely dependent on initial pathway of cel 
entry. Therefore, it is to be expected that a portion of dendrimer dose applied to 
enterocytes wil be traficked to the lysosomes by the clathrin-mediated endocytosis 
pathway, while the dendrimers that enter via the caveolae may end up in the cel cytosol, 
alowing them to be targeted to either compartment depending on the desired efect.  
Dendrimer transport across diferentiated Caco-2 cel monolayers was found to be 
dependent on dynamin- and clathrin-mediated endocytosis, but independent of caveolin-
mediated endocytosis. This result was expected since fully diferentiated Caco-2 cels 
lack caveolae. The diferences betwen dendrimer transport across diferentiated 
epithelial cels and uptake in undiferentiated cels can be potentialy exploited by drug 
delivery strategies that aim to specificaly target cancer cels while simultaneously 
 
101 
 
 
avoiding intestinal cels by designing linkers that would be cleaved in caveolae-mediated 
transport pathways.  
Intracelular traficking studies shed further light on the environments that 
dendrimers encounter after celular internalization. Kitchens and co-workers [18] 
examined G1.5 and G2 dendrimer colocalization with endosomal and lysosomal markers 
in Caco-2 cels and reported that dendrimers show constant presence in the early 
endosomes at 20 and 60 minutes, with time-dependent traficking to the lysosomes. In 
this study G3.5 dendrimers were found to localize in the early endosomes and lysosomes 
after 5 minutes, displayed fast traficking to the lysosomes after 15 minutes and increased 
endosomal and lysosomal acumulation at 30 minutes, likely due to pathway saturation. 
Often drugs are conjugated to dendrimers via pH-sensitive linkers that are cleaved in the 
acidic environment of mature endosomes or peptide linkers that are cleaved by lysosomal 
enzymes such as cathepsin B [174]. These intracelular traficking studies corroborate 
the validity of such strategies, as dendrimers can be found in both environments 
following celular internalization. Cleavage of pH-sensitive linkers in the endosomes 
may be promising as dendrimers are shown to acumulate in these compartments over 
time. Knowledge of the celular uptake and intracelular traficking pathways of 
dendrimers can have significant impact on designing them for use as drug delivery 
vehicles. In the context of oral drug delivery, knowledge of subcelular localization in 
intestinal cels can enable design of conjugates that are robust in these compartments and 
absorbed intact to the blood stream. 
One of the most intriguing atributes of PAMAM dendrimers is that within a 
specified size and charge window, they have been shown to catalyze their own transport 
 
102 
 
 
via the paracelular pathway [16]. Previous reports have shown that dendrimers decrease 
TER, increase paracelular marker flux (e.g. mannitol) and increase occludin 
acesibility in Caco-2 cels, confirming that they open tight junctions [16, 145, 146]. 
However the mechanism behind this phenomenon remained largely elusive. In this 
Chapter we examined the role of dendrimer celular internalization on tight junction 
opening by monitoring the efects of dendrimers on confluent monolayers in buffer or in 
the presence of dynasore. While dendrimers were able to open tight junctions in the 
presence of buffer alone, they could not do so in the presence of dynasore, suggesting 
that dendrimer internalization is requisite for tight junction opening.  This has significant 
implications for oral delivery using dendrimers because it shows that tight junction 
opening is at least in part modulated by dendrimers within the cels. Whether afecting 
tight junctional structures from the apical environment outside the cels by dendrimers 
also plays a role remains to be examined. Therefore, this mechanism (opening tight 
junction by internalization) results in temporary tight junction opening, which is 
consistent with our previous reports that TER returns to pre-treatment values after 24 
hours [149]. Taken together, these data establish that dendrimers can be used safely as 
oral drug cariers or penetration enhancers since depending on generation, concentration 
and incubation time their efects on tight junctions can be transient, not permanent. 
Figure 3.7 summarizes some of the possible G3.5 transport pathways across Caco-2 cel 
monolayers. 
 
 
103 
 
 
 
Figure 3.7. Mechanisms of Dendrimer Transport Across Caco-2 Cel Monolayers. 
Once endocytosed, dendrimers can interact with tight junctions (A), alowing for 
paracelular transport (B), or they can be degraded by the lysosomes (C) or transcytosed 
(D). 
 
 
104 
 
 
3.5 Conclusion 
In this Chapter we described the detailed mechanisms of celular uptake, 
intracelular traficking, transport and tight junction modulation of G3.5 PAMAM 
dendrimers in Caco-2 cels. We found that G3.5 PAMAM dendrimers enter 
undiferentiated Caco-2 cels by clathrin-, caveolin-, and dynamin-dependent pathways 
but that their transepithelial transport across confluent monolayers is governed by 
clathrin- and dynamin-dependent pathways only. Dendrimers were quickly traficked to 
the lysosomes, but show increased endosomal acumulation once the lysosomal 
compartments become saturated. Finaly, it was demonstrated that dendrimer 
endocytosis promotes tight junction opening, ilustrating the interconnected nature of the 
transcelular and paracelular pathways in dendrimer transepithelial transport. 
Knowledge of detailed mechanisms of dendrimer celular uptake, intracelular and 
transepithelial transport wil asist in the design of PAMAM dendrimer-based oral drug 
delivery strategies by providing appropriate linker chemistry consistent with 
transepithelial transport and celular traficking pathways. 
 
 
 
105 
 
 
 
Chapter 4 : G3.5 PAMAM Dendrimers Enhance Transepithelial Transport of 
SN38 While Minimizing Gastrointestinal Toxicity 
 
4.1 Introduction 
 Polymer-based drug delivery systems have shown promise due to their ability to 
improve the eficacy of traditional drugs [6]. Conjugation of smal molecule therapeutics 
to a polymeric carier can enhance the drug?s solubility, increase acumulation at the 
target site and minimize non-specific toxicity. Because chemotherapy drugs are often 
plagued by poor water solubility and dose-limiting toxicities, they are promising 
candidates for polymeric drug delivery strategies. Atachment of chemotherapy drugs to 
water soluble polymers enhances solubility, alows for acumulation of the polymer-drug 
conjugate at the tumor site due to the enhanced permeability and retention efect, 
improves eficacy and reduces side efects [1]. Several polymer-drug conjugates using 
N-(2-hydroxypropyl)methacrylamide (HPMA) and poly (ethylene glycol) (PEG) as 
cariers are currently being evaluated in clinical trials [38, 44].  
While conjugation of chemotherapy drugs to water-soluble polymers can improve 
their solubility and tumor uptake, the large size of these macromolecular constructs 
necesitates intravenous administration. Oral administration is typicaly limited to smal, 
lipophilic drugs that can permeate the cel membrane, smal, hydrophilic drugs that pas 
through the tight junctions or drugs that are substrates for intestinal transporters [175]. 
Studies have shown that compared to intravenous administration, oral chemotherapy is 
the prefered method of administration by cancer patients given similar eficacy for both 
 
106 
 
 
treatments [7]. Therefore, the combination of the distinct therapeutic advantages of 
polymer-drug conjugates with strong patient preference and lower costs of oral 
chemotherapy supports a significant need for oraly bioavailable polymer therapeutics. 
 Because of its low water solubility and poor bioavailability, SN38 (7-ethyl-10-
hydroxy-camptothecin), a potent topoisomerase-1 poison used to treat colorectal cancer 
and hepatic metastases, is an ideal candidate for polymeric delivery strategies [158]. 
While SN38 shows 100-1000-fold higher activity than CPT-11 in vitro, its use is limited 
by low water solubility and significant intestinal toxicity including diarhea [176, 177]. 
Conjugation of SN38 to PAMAM dendrimers has the potential to alow for oral 
administration while also improving water solubility and minimizing gastrointestinal 
toxicity. Synthesis, characterization, and bioactivity of G3.5 PAMAM-SN38 conjugates 
against colorectal carcinoma cel lines has been previously established [161]. To 
succesfully advance these systems for oral administration, their stability in the 
gastrointestinal tract and transport across the epithelial barier of the gut must be 
determined. In this Chapter we examine G3.5 PAMAM-SN38 conjugates for their in 
vitro release profiles, cytotoxicity against Caco-2 and HT-29 colorectal cancer cels, 
celular uptake, and transepithelial transport across Caco-2 monolayers as models of the 
intestinal epithelial barier. 
 
4.2 Materials and Methods 
4.2.1 Materials 
PAMAM G3.5 dendrimers (reported molecular weight=12,931), lucifer yelow 
CH dipotasium salt (LY), pepsin from porcine gastric mucosa, pancreatin from porcine 
 
107 
 
 
pancreas and carboxylesterase from rabbit liver were purchased from Sigma Aldrich (St. 
Louis, MO). Simulated Intestinal Fluid (SIF) and Simulated Gastric Fluid (SGF) were 
obtained from Rica Chemical Company (Arlington, Texas). 7-ethyl-10-hydroxy 
camptothecin (SN38) was obtained from AK Scientific Company (Mountain View, CA). 
WST-1 cel proliferation reagent was purchased from Roche Applied Sciences 
(Indianapolis, IN). Caco-2 cels and HT-29 cels were obtained from American Type Cel 
Culture (Rockvile, MD). 
 
4.2.2 Synthesis and Characterization of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates 
 G3.5-Gly-SN38 and G3.5-?Ala-SN38 were synthesized as previously described 
[161]. Briefly, SN38 was modified at the 20-OH position via an ester linker with glycine 
or ??alanine [178]. The modified SN38 molecules were then conjugated to carboxylic 
acid-terminated G3.5 dendrimers using EDC / NHS as a coupling agent. The products 
were dialyzed against distiled water using 3,500 MWCO membranes to remove low 
molecular weight impurities and further purified using preparative fast protein liquid 
chromatography (FPLC). 
1
H NMR was used to quantify the number of SN38 molecules 
per dendrimer by comparing the area of the dendrimer protons betwen 2.1 and 3.8 ppm 
with the methyl protons of SN38 betwen 0.9-1.1 ppm. The drug loading per dendrimer 
was 2.9 and 4 for G3.5-Gly-SN38 and G3.5-?Ala-SN38, respectively. Figure 4.1 
ilustrates the conjugation strategy used to synthesize G3.5-Gly-SN38 and G3.5-?Ala-
SN38.  
 
108 
 
 
 
Figure 4.1. Conjugation of SN38 to G3.5 Dendrimers via Glycine and ??Alanine 
Linkers. (Adapted from [161]). 
 
 
109 
 
 
4.2.3 Stability Studies 
 G3.5-Gly-SN38 and G3.5-?Ala-SN38 conjugates were prepared at 0.15 and 0.10 
mg/ml respectively in the release buffer of interest and incubated at 37?C with rotation in 
glas vials. Release buffers included SGF with and without 0.32% w/v pepsin [179], SIF 
with and without 1% w/v pancreatin [179] and PBS with and without carboxylesterase 
(17.3 IU/ml) [180]. Release in gastric conditions was monitored for up to 6 hours and in 
the intestinal conditions for up to 24 hours, miicking the expected residence time in 
each of these compartments [49]. Enzyme activity was confirmed at each time point 
using bovine serum albumin [181], Z-Arg-AMC [182] and p-nitrophenol acetate [183] as 
model substrates for pepsin, pancreatin and carboxylesterase, respectively.  
During the release experiment, 100 ?L of sample was taken at each time point and 
added to an additional 400 ?L of release buffer. Released SN38 was separated from 
G3.5-SN38 conjugate by pasing the sample through a 3,000 MWCO Amicon Ultra 0.5 
centrifugal concentrator (Milipore, Bilerica, MA) by centrifuging at 14,000 x g for 30 
minutes. Amicon concentrators were pre-treated with 5% v/v Triton X-100 to minimize 
non-specific binding as per the manufacturer?s instructions. SN38 is known to exist in a 
pH-dependent equilibrium betwen a closed ring lactone form and an open ring 
carboxylate form, with the lactone form favored at low pH [157]. For SIF and PBS, 350 
?L of the filtrate was taken and acidified to pH 2 with the addition of an appropriate 
volume of 1 N HCl and subsequently incubated at 37?C for 1 hour to convert SN38 to the 
lactone form. This acidification step was omited for SGF since it is at a pH of 1-2. 
Next, the acidified SN38 was extracted by adding 200 ?L of acetonitrile folowed by 200 
?L of chloroform. The solution was vortexed and centrifuged at 250 x g for 2 minutes to 
 
110 
 
 
separate the layers. The organic layer was isolated, the chloroform addition was repeated 
two more times, and the organic extracts were pooled. The extracts were then dried 
under nitrogen gas, redisolved in 70 ?L of 50/50 DMSO/0.1N HCl and then measured 
by high presure liquid chromatography (HPLC).  
SN38 was quantified by HPLC using a system containing a Waters 1525 Binary 
Pump, Waters 717plus Autosampler and Waters 2487 dual wavelength UV detector 
(Waters Corporation, Milford, MA) set at 375 nm with a Phenomenex C18 column (250 
x 4.6 m, 5 ?m) (Phenomenex, Torrance, CA). A gradient method with methanol and 
water with 0.1% TFA was used with a total flow rate of 1 ml per minute and an injection 
volume of 20 ?L. A calibration curve using the peak area versus concentration was 
generated for each release buffer by extracting known concentrations of SN38. Detailed 
HPLC methods and standard curves are included in Appendix 2. Extraction eficiencies 
comparing extracted standards to direct injection of SN38 standards were found to be 
86%, 91%, 75%, 80%, 70% and 83% in SGF, SGF/ pepsin, SIF, SIF/ pancreatin, PBS, 
and PBS/ carboxylesterase, respectively, and were time and concentration-independent. 
 
4.2.4 Cel Culture 
Caco-2 cels (pasages 20-40) were cultured as described in Section 3.2.3. 
HT-29 cels (pasages 130-140) were cultured under the same incubation conditions as 
Caco-2 cels using McCoy?s 5A media suplemented with 10% fetal bovine serum (FBS), 
1% non-esential amino acids, 10,000 units/mL penicilin, 10,000 ?g/mL streptomycin 
and 25 ?g/mL amphotericin B. 
 
111 
 
 
4.2.5 Potential Short-Term Cytotoxicity of G3.5-SN38 Conjugates 
Potential short-term cytotoxicity of G3.5-SN38 conjugates was asesed in Caco-
2 cels to ensure cel viability during uptake and transport asays. Unmodified G3.5 
dendrimers, G3.5-Gly-SN38 and G3.5-?Ala-SN38 were prepared at 10 and 100 ?M in 
HBS transport bufer. SN38 was prepared as a concentrated stock in DMSO at 40,000 
?M and used to make solutions in HBS at 4, 40, and 400 ?M. Cytotoxicity was 
asesed by the water soluble tetrazolium salt (WST-1) asay as described in Section 
3.2.4. HBS was used as a negative control for 100% cel viability and 0.01% Triton X-
100 was used as a positive control. UV absorbance of G3.5-SN38 conjugates alone was 
also asesed to confirm that SN38 absorbance did not interfere with the cel viability dye 
(data not shown). 
 
4.2.6 Potential Delayed Cytotoxicity of G3.5-SN38 Conjugates 
 Potential delayed cytotoxicity of G3.5-SN38 conjugates was asesed in Caco-2 
cels by measuring cel viability 24 hours post-exposure. Cel viability 24 hours after 
short-term exposure miics the potential long-term efects of the conjugates on the 
intestinal cels that may not be apparent when cel viability is measured imediately 
post-treatment. Caco-2 cels were prepared and treated the same as in Section 4.2.5. 
After the 2-hour exposure, the cels were washed twice with HBS and incubated with 
100 ?L of cel culture media for an additional 24 hours. The WST-1 asay was used to 
ases the cel viability 24 hours later, with HBS used as a negative control for 100% 
cel viability and 0.01% Triton X-100 as a positive control.  
 
112 
 
 
4.2.7 Transepithelial Transport 
Caco-2 cel monolayers were grown as described in Section 3.2.7. Monolayers 
were washed with HBS and then 0.5 ml of 10 or 100 ?M G3.5-Gly-SN38, G3.5-?Ala-
SN38 or 4 ?M SN38 was added to the apical compartment and 1.5 ml HBS was added 
to the basolateral compartment. After a 2-hour incubation, samples were taken from the 
basolateral compartment. Transport was quantified by measuring fluorescence in the 
basolateral compartment using a SpectraMax Gemini XS spectrofluorometer (Molecular 
Devices, Sunnyvale, CA) with excitation and emision wavelengths of 375 and 550 nm, 
respectively, and compared to fluorescence standard curves for each conjugate and fre 
SN38. Presence of fre SN38 in the basolateral compartment of monolayers treated with 
G3.5-SN38 conjugates was determined by the extraction methods described in Section 
4.2.3. Equivalent SN38 flux was calculated by multiplying the measured molar flux of 
the conjugates with the number of SN38 molecules per dendrimer. Apical to basolateral 
flux is reported as the average of four replicates. Statistical significance was determined 
by analysis of variance followed by Tukey?s multiple comparison test. LY permeability 
was also monitored in the presence of HBS to ensure the integrity of the monolayers. 
LY apparent permeability was les than 1 x 10
-6 
cm/s, which is within the acepted range 
of LY permeability for diferentiated monolayers (data not shown). 
 
4.2.8 Celular Uptake 
 Celular uptake of G3.5-SN38 conjugates and fre SN38 was asesed in 
diferentiated Caco-2 monolayers. After the transport asay, the cels were washed twice 
with HBS. 300 ?l of 0.1% Triton X-100 was added to the apical side of each wel and 
 
113 
 
 
incubated for 2 hours at 37?C to solubilize the cels. The cels were then removed from 
the Transwel
?
 by pipete and transfered to a microcentrifuge tube. The cel debris was 
removed by centrifugation at 1000 RPM for 5 minutes. 100 ?l of the clear supernatant 
was taken and the uptake of G3.5-SN38 conjugates and SN38 was quantified by 
fluorescence with excitation at 375 nm and emision at 550 nm as described in Section 
4.2.7. Presence of fre SN38 in the celular compartment was determined by the 
extraction methods described in Section 4.2.3. Uptake is reported as an average of four 
replicates and normalized to total protein as determined by the Bradford Protein Asay 
(Bio-Rad, Hercules, CA). Statistical significance was determined by analysis of variance 
followed by Tukey?s multiple comparison test. 
 
4.2.9 IC
50
 in HT-29 Cels 
 The IC
50 
values of G3.5-SN38 conjugates and fre SN38 were determined in HT-
29 cels to ases the eficacy of the conjugates in colorectal cancer cels compared to the 
fre drug. HT-29 cels were seded at 2,500 cels/ per wel in 96 wel cel culture plates 
and incubated at 37?C for 24 hours. After 24 hours, the cels were treated with diferent 
concentrations of SN38, G3.5-Gly-SN38 and G3.5-?Ala-SN38 in media and incubated 
for an additional 48 hours. Cels were also treated with comparable concentrations of 
G3.5 dendrimer alone to ensure that the carier did not cause any long-term cytotoxicity. 
After 48 hours, the media was aspirated by pipete, the cels were washed with HBS 
buffer and the WST-1 asay was used to ases cel viability. Cel viability was 
determined by the % absorbance relative to the control cels, which were treated with 
 
114 
 
 
media alone. GraphPad Prism software (La Jolla, CA) was used to generate the IC
50
 
curves and values using the sigmoidal dose-response non-linear curve fiting routine. 
 
4.3 Results 
4.3.1 Stability of G3.5-SN38 Conjugates 
 Stability of G3.5-SN38 conjugates in simulated gastric, intestinal and liver 
carboxylesterase conditions was asesed to determine the suitability of these conjugates 
for oral delivery of SN38 for the treatment of colorectal cancer hepatic metastases. In 
particular, stability of the conjugates was monitored in the presence of SGF with and 
without pepsin (Figure 4.2), SIF with and without pancreatin (Figure 4.3) and PBS with 
and without carboxylesterase (Figure 4.4).  
Both G3.5-Gly-SN38 and G3.5-?Ala-SN38 showed litle to no release of fre 
SN38 in the gastric environment (Figure 4.2). G3.5-?Ala-SN38 did not release more 
than 0.5% SN38 during 6 hours in SGF with or without pepsin. In contrast, G3.5-Gly-
SN38 showed a burst release of SN38 after 1 hour in SGF with and without pepsin and 
ultimately released approximately 9% in SGF with pepsin and 6% in SGF without pepsin. 
This smal amount of gastric release could be prevented by use of enteric coating. These 
studies ilustrate that although G3.5-Gly-SN38 released more SN38 than G3.5-?Ala-
SN38 in the gastric environment, they were both relatively stable in acidic conditions and 
were not substrates for pepsin.   
 
115 
 
 
 
Figure 4.2. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in Simulated 
Stomach Conditions for 6 hours. Results are reported as mean +/- standard deviation 
(n=2). G3.5-Gly-SN38 is represented by squares and G3.5-?Ala-SN38 is represented by 
circles. Release in SGF with 0.32% w/v pepsin is depicted by solid lines with filed 
symbols, and release in SGF alone is depicted by dashed lines with open symbols. 
 
116 
 
 
 
Figure 4.3. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in Simulated 
Intestinal Conditions for 24 hours. Results are reported as mean +/- standard deviation 
(n=2). G3.5-Gly-SN38 is represented by squares and G3.5-?Ala-SN38 is represented by 
circles. Release in SIF with 1% w/v pancreatin is depicted by solid lines with filed 
symbols, and release in SIF alone is depicted by dashed lines with open symbols. 
 
117 
 
 
 
 
Figure 4.4. Stability of G3.5-Gly-SN38 and G3.5-?Ala-SN38 Conjugates in Simulated 
Liver Conditions for 48 hours. Results are reported as mean +/- standard deviation (n=2). 
G3.5-Gly-SN38 is represented by squares and G3.5-?Ala-SN38 is represented by circles. 
Release in PBS with carboxylesterase (17.3 IU/ml) is depicted by solid lines with filed 
symbols, and release in PBS alone is depicted by dashed lines with open symbols. 
 
 
118 
 
 
In comparison to SGF, G3.5-SN38 conjugates showed increased susceptibility for 
fre drug release in the presence of SIF (Figure 4.3). G3.5-?Ala-SN38 showed up to 4% 
SN38 release in 24 hours in the presence of SIF, but the addition of pancreatin did not 
increase this release. In contrast, G3.5-Gly-SN38 released up to 10% SN38 in the 
presence of SIF and up to 20% with the addition of pancreatin. These results show that 
both conjugates are inherently more susceptible to hydrolysis in the basic pH of SIF 
compared to the acidic pH of SGF, which is common for ester linkages. In addition, 
while G3.5-?Ala-SN38 did not appear to be a substrate for pancreatin, G3.5-Gly-SN38 
showed increased release in the presence of pancreatin compared to SIF alone. However, 
with a maximum of 20% release after 24 hours, both of these conjugates showed a low 
extent and slow rate of release in the intestinal environment. 
Finaly, we examined the stability profile of the conjugates in the presence of liver 
carboxylesterase and in PBS alone (Figure 4.4). Carboxylesterase is highly expresed in 
the liver environment and can be used to estimate the release in this milieu. Both 
conjugates showed similar release profiles in PBS and SIF up to 24 hours, and the linear 
release kinetics continued until 48 hours, suggesting similar rates of hydrolysis in SIF and 
PBS without enzymes. G3.5-?Ala-SN38 did not show any additional release in the 
presence of carboxylesterase, suggesting that it is not a substrate for this enzyme. In 
contrast, G3.5-Gly-SN38 showed a significant increase in SN38 release in the presence of 
carboxylesterase, achieving 56% release after 48 hours, ilustrating that the ester bond in 
G3.5-Gly-SN38 can be cleaved by carboxylesterase to release fre SN38 in the liver 
environment. Taken together, these stability studies show that G3.5-?Ala-SN38 is stable 
in al thre environments and is not a substrate for pepsin, pancreatin or carboxylesterase, 
 
119 
 
 
while G3.5-Gly-SN38 shows les stability in the gastric and intestinal environments and 
the greatest release in the presence of carboxylesterase, making it a potential candidate 
for oral delivery of SN38 to colorectal hepatic metastases. 
 
4.3.2 Short-Term Cytotoxicity 
Short-term cytotoxicity of G3.5 dendrimers, G3.5-SN38 conjugates and SN38 
was asesed in Caco-2 cels by the WST-1 cel viability asay. This asay is predictive 
of the short-term efects of the conjugates on the intestinal barier and also serves to 
ensure that cel viability is not compromised during the 2-hour time needed for transport 
and uptake asays. Figure 4.5 shows the cel viability of Caco-2 cels treated for 2 hours 
with unmodified G3.5 dendrimers and G3.5-SN38 conjugates. SN38 is also tested at 4, 
40 and 400 ?M for comparison, corresponding to 1, 10 or 100% of the drug loading on 
the G3.5-?Ala-SN38 conjugate.  
G3.5 dendrimers and G3.5-SN38 conjugates did not cause a significant reduction 
in cel viability up to 100 ?M. In contrast, despite the short treatment time, SN38 shows 
a significant reduction in cel viability at 40 and 400 ?M, corresponding to approximately 
10% and 100% of the drug loading on the conjugates. This suggests that conjugation of 
SN38 to G3.5 dendrimers is able to significantly reduce intestinal toxicity of SN38. In 
addition, it is observed that G3.5-Gly-SN38 and G3.5-?Ala-SN38 can be used in 
transport and uptake asays up to 100 ?M without compromising cel viability. 
 
 
120 
 
 
 
Figure 4.5. Caco-2 Cel Viability after Treatment for 2 hours with G3.5 Dendrimers, 
G3.5-SN38 Conjugates and SN38. Results are reported as mean +/- standard deviation 
(n=6). (**) and (***) indicate a statisticaly significant decrease in cel viability 
compared to HBS control with p<0.01 and p<0.001, respectively. G3.5 dendrimers and 
G3.5-SN38 conjugates do not show a reduction in cel viability up to 100 ?M while 
SN38 shows a significant cytotoxic efect at 40 and 400 ?M. 
 
 
 
121 
 
 
4.3.3 Delayed Cytotoxicity 
In order to ases the potential long-term efects of G3.5 dendrimers and G3.5-
SN38 conjugates on the intestinal barier, a delayed cytotoxicity asay was performed. In 
this asay, Caco-2 cels were treated for 2 hours, the treatment was removed and the cels 
were incubated for an additional 24 hours in cel culture media. This alows for the 
asesment of any potential delayed-onset responses of the cels to dendrimer treatment 
(e.g. apoptosis), which can be detected after 24 hours (Figure 4.6). Even after 24 hours, 
G3.5 dendrimers and G3.5-SN38 conjugates did not cause a statisticaly significant 
decrease in cel viability, with the exception of G3.5-Gly-SN38 at a 10 ?? concentration, 
which displayed 85.4 % +/- 8.1% viability. While this is a statisticaly significant 
decrease from the HBS control, 85% viability is stil considered to be aceptable in such 
viability asays and 10 ?M treatment with G3.5-Gly-SN38 does not show a significant 
diference from 100 ?M treatment with the same conjugate. In contrast, SN38, which is 
known to cause apoptosis by inhibition of topoisomerase-1 [158], had a significant 
impact on cel viability 24 hours post treatment at 4, 40 and 400 ?M concentrations. This 
ilustrates that by conjugating SN38 to dendrimers, intestinal toxicity is minimized 
compared to the fre drug and that G3.5-SN38 conjugates should be safe for oral 
administration. 
 
4.3.4 Transepithelial Transport 
Transepithelial transport of G3.5-SN38 conjugates and fre SN38 was measured 
across diferentiated Caco-2 monolayers in the apical to basolateral direction and 
expresed as the equivalent SN38 flux calculated at 2 hours (Figure 4.7). In this study 
 
122 
 
 
 
Figure 4.6. Caco-2 Cel Viability 24 hours after 2-hour Treatment with G3.5, G3.5-SN38 
Conjugates and SN38. Results are reported as mean +/- standard deviation (n=6). (*) and 
(***) indicate a statisticaly significant decrease in cel viability relative to HBS control 
with p<0.05 and p<0.001, respectively.  
 
123 
 
 
 
Figure 4.7. Equivalent SN38 Flux across Diferentiated Caco-2 Monolayers Treated with 
G3.5-SN38 Conjugates and SN38. Treatment concentrations of conjugates were 10 and 
100 ?M, corresponding to 29, 290 and 40, 400 ?M equivalents of SN38 for G3.5-Gly-
SN38 and G3.5-?Ala-SN38 conjugates, respectively. Equivalent SN38 flux was 
calculated by multiplying the measured molar flux of the conjugates with the number of 
SN38 molecules per dendrimer. Results are reported as mean +/- standard deviation 
(n=4). (***) indicates a significant diference with p<0.001. 
 
124 
 
 
SN38 was tested at 4 ?M concentration since significant cytotoxicity was observed in 
Caco-2 cels treated with SN38 at 40 and 400 ?M.  It has been shown that SN38 is 
transported across Caco-2 cel monolayers by an active transport pathway, and that 
increasing the concentration from 2.5 ?M to 25 ?M does not increase transepithelial flux 
[184]. This suggests that 4 ?M SN38 should be beyond the saturation limit for transport 
and uptake, thus permiting direct comparison of SN38 flux to that of G3.5-SN38 
conjugates despite the diference in total drug concentration. 
Presence of fre SN38 in the basolateral compartment was found to be les than 
5% of the amount transported for G3.5-Gly-SN38 and G3.5-?Ala-SN38 conjugates (data 
not shown). Hence the measured flux is due almost entirely to transport of intact 
conjugate. G3.5-Gly-SN38 at 100 ?M and G3.5-?Ala-SN38 at 10 ?M and 100 ?M 
showed a statisticaly significant increase in apical to basolateral SN38 flux relative to 
fre drug (p<0.001). Taking into acount the drug loading on each conjugate (2.9 SN38 
per dendrimer for G3.5-Gly-SN38 and 4.0 SN38 per dendrimer for G3.5-?Ala-SN38), the 
overal SN38 flux increase compared to fre drug is 13, 159, 69 and 89-fold for G3.5-
Gly-SN38 at 10 ?M and 100 ?M and G3.5-?Ala-SN38 at 10 ?M and 100 ?M, 
respectively.  These significant increases in the amount of SN38 transported across the 
monolayer indicate that G3.5 dendrimers are efective oral drug delivery cariers and can 
increase the permeability of SN38.  
 
125 
 
 
Interestingly, the impact of concentration on transport is diferent for G3.5-Gly-
SN38 and G3.5-?Ala-SN38 conjugates. G3.5-Gly-SN38 shows approximately a 10-fold 
increase in transport with a 10-fold increase in concentration, indicating that difusion-
driven proceses (i.e. paracelular transport) are predominantly involved. In contrast, 
G3.5-?Ala-SN38, shows minimal increase in transport with a 10-fold increase in 
concentration, suggesting that its epithelial flux may be controlled by a saturable, energy-
dependent mechanism with minimal paracelular transport. Dendrimers have been shown 
to cross epithelia both by transcelular and paracelular pathways and the relative 
importance of each has been shown to be charge and surface chemistry dependent [149]. 
Factors such as the number of conjugated surface groups as wel as the ability of these 
groups to shield the dendrimer surface charge have been found to impact the ability of 
dendrimers to open tight junction and overal mechanism of transport. Therefore, both the 
number of SN38 molecules conjugated and the identity of the linker could potentialy 
impact the ultimate transport pathway. These studies further confirm the importance of 
dendrimer surface chemistry and drug linker chemistry in the degre and mechanism of 
transport. More detailed mechanistic studies, however, are required to explain the 
diferences in transepithelial transport profiles of the two conjugates. 
 
4.3.5 Celular Uptake 
Celular uptake studies shed light on the contribution of the transcelular pathway 
to overal transepithelial transport as wel as the ability of the carier to promote drug 
uptake. Celular uptake of G3.5-SN38 conjugates and SN38 was measured in 
diferentiated Caco-2 monolayers after a 2-hour incubation time (Figure 4.8). 
 
126 
 
 
 
Figure 4.8. Celular Uptake of G3.5-SN38 Conjugates and Fre SN38 in Diferentiated 
Caco-2 Monolayers after 2-hour Treatment on the Apical Side. Results are reported as 
mean +/- standard deviation (n=4). (*), (**), and (***) show statistical diferences in 
uptake betwen groups with p < 0.05, 0.01 and 0.001, respectively. Al conjugates show 
a statisticaly significant increase in uptake relative to SN38. 
 
 
127 
 
 
Similar to the transport studies, fre SN38 in the cels after a 2-hour treatment was 
found to be les than 5% of conjugate uptake (data not shown). Al conjugates tested 
showed a significant increase in uptake relative to fre SN38. This ilustrates that G3.5 
dendrimers can enter cels more eficiently than SN38, and thus are suitable for celular 
delivery. Both G3.5-Gly-SN38 and G3.5-?Ala-SN38 conjugates showed a significant 
increase in uptake (p<0.001) with increase in concentration. However, for both 
conjugates the increase in uptake was les than the corresponding increase in 
concentration, suggesting the involvement of a saturable uptake mechanism, such as 
receptor-mediated endocytosis. Interestingly, both conjugates showed similar uptake for 
the 100 ?M treatment, suggesting that diferences in overal transport may be due to 
diferences in paracelular transport rather than transcelular transport at this 
concentration. These results confirm that celular uptake of G3.5-SN38 conjugates by 
Caco-2 cels plays a significant role in G3.5-SN38 conjugate transport.  
 
4.3.6 IC
50
 in HT-29 Cels 
Toxicity of G3.5-SN38 conjugates and fre SN38 was determined in HT-29 cels 
for 48 hours in order to compare the activity of the fre drug and dendrimer-drug 
conjugates (Figure 4.9). HT-29 cels are derived from human colon adenocarcinoma, are 
wel suited for IC
50 
studies of anti-cancer activity due to their uniform cel growth and 
morphology. 
 
128 
 
 
 
Figure 4.9. IC
50
 Curves of SN38, G3.5-Gly-SN38 and G3.5-?Ala-SN38 in HT-29 Cels. 
SN38 is represented by inverted triangles, G3.5-Gly-SN38 by squares and G3.5-?Ala-
SN38 by circles. Results are reported as mean +/- standard deviation (n=6). IC
50
 values 
were determined from nonlinear sigmoidal dose response curve fiting by GraphPad 
Prism software and are 66.3 nM, 0.60 ?M and 3.59 ?M for SN38, G3.5-Gly-SN38 and 
G3.5-?Ala-SN38, respectively. 
 
 
129 
 
 
SN38 showed the highest activity against HT-29 cels with an IC
50
 value of 66.3 
nM. In contrast, SN38 conjugated to G3.5 dendrimers showed much lower activity with 
IC
50 
values of 0.60 ?M and 3.59 ?M for G3.5-Gly-SN38 and G3.5-?Ala-SN38 
conjugates, respectively, corresponding to approximately 10 and 60 times les potency 
than the fre drug. Treatment with comparable G3.5 dendrimer concentrations did not 
cause any reduction in cel viability (data not shown), confirming that the carier does not 
cause toxicity. Since cytotoxicity is dependent on fre drug release from the dendrimer 
backbone, the lower IC
50
 value for G3.5-Gly-SN38 compared to G3.5-?Ala-SN38 is 
consistent with the higher release rate of SN38 from the G3.5-Gly-SN38 conjugate. 
Because the conjugates showed some release under acidic conditions and greater release 
at neutral pH, it is likely that the activity in HT-29 cels is due to a combination of 
extracelular release of SN38 in cel culture media and intracelular release. Further 
detailed studies are required to examine the relative contribution of extra- and 
intracelular drug release to cytotoxicity. Importantly, despite the loss in eficacy 
compared to the fre drug, the IC
50
 values of the G3.5-SN38 conjugates are stil in the 
nanomolar to micromolar range, which is aceptable for therapy. 
 
4.4 Discusion 
Dendrimers have shown promise as oral drug delivery cariers due to their ability 
to translocate across the epithelial layer of the gut. The impact of dendrimer properties 
such as generation, charge and concentration on transepithelial transport have been 
described in detail, but comparatively few reports have been published using dendrimers 
to translocate drugs across the intestinal barier. Previously Kolhatkar and co-workers 
 
130 
 
 
reported the use of G4 dendrimers complexed with SN38 through non-covalent 
interactions as a potential oral drug delivery system [150]. While the conjugates 
increased the transepithelial transport and celular uptake of SN38, they released 40% of 
the drug within 24 hours in PBS and 90% of the drug within 30 minutes at pH 5. 
Because of the instability in acidic conditions, these dendrimer-SN38 complexes would 
have significant limitations for oral administration. In addition, because of the high 
intrinsic toxicity of the G4 dendrimer carier [16], these complexes could only be used up 
to a 10 ?M concentration, limiting the amount of drug transported across the intestinal 
barier and into the bloodstream. 
Vijayalakshmi and colleagues reported the preliminary synthesis and 
characterization of G3.5-SN38 conjugates with glycine and ??alanine linkers [161]. In 
this Chapter, we investigated the potential of these G3.5-SN38 conjugates for oral 
therapy of colorectal hepatic metastases. The G3.5-SN38 conjugates have several 
advantages compared to the previously reported G4-SN38 complexes. Because the drug 
is covalently conjugated to the dendrimer rather than complexed, the conjugates are 
relatively stable in the gastric and intestinal environments, minimizing premature drug 
release upon oral administration. In addition, the G3.5-Gly-SN38 conjugates release the 
drug in the presence of carboxylesterase, alowing for the targeted treatment of colorectal 
hepatic metastasis. These systems can be used at higher concentrations (100 ?M vs. 10 
?M) due to the low intrinsic toxicity of G3.5 dendrimers, alowing for greater drug 
transport across the gastrointestinal barier and a higher dose at the site of action. 
Therefore, these conjugates show distinct advantages over previously published 
 
131 
 
 
dendrimer-SN38 complexes and are a significant step towards a functional dendrimer-
based oral drug delivery system. 
In order to determine their suitability for oral delivery, the stability of the 
conjugates in the gastrointestinal milieu as wel as in the presence of liver 
carboxylesterase was investigated. G3.5-?Ala-SN38 conjugates were significantly more 
stable than G3.5-Gly-SN38 under the conditions studied, releasing a maximum of 1% 
drug in SGF after 6 hours, 4% in SIF after 24 hours and 8% in carboxylesterase after 48 
hours. In addition, in each of the release buffers, the addition of an enzyme did not 
increase the release of SN38, ilustrating that G3.5-?Ala-SN38 is not susceptible to 
enzymatic cleavage. In contrast, G3.5-Gly-SN38 showed much higher release in al of 
the conditions and showed increased rate and extent of release in the presence of 
pancreatin and carboxylesterase compared to buffer alone. Previous studies conjugating 
CPT to poly(l-lysine) dendrimers using glycine and ?-alanine linkages at the 20-OH 
position showed similar results with the glycine linker indicating higher rates of 
hydrolysis than the stable ??alanine linker [185]. Although the ??alanine linker provides 
an extra methyl group as a spacer, this group appears to stabilize the bond against 
hydrolysis and enzymatic degradation.  
The impact of linker chemistry on SN38 release shows a direct correlation with 
the IC
50 
values in which the G3.5-Gly-SN38 conjugates had six-fold greater eficacy than 
G3.5-?Ala-SN38 conjugates, ilustrating the importance of SN38 release for anti-cancer 
activity. Since the conjugates were relatively stable at lower pH it is likely that only a 
smal percentage of the drug was released in the intracelular acidic environment with the 
majority released in the extracelular media. Stability studies in PBS for 48 hours 
 
132 
 
 
showed a two-fold increase in SN38 release from G3.5-Gly-SN38 compared to G3.5-
?Ala-SN38, suggesting additional enzymatic mechanisms may play a role in enhancing 
G3.5-Gly-SN38 eficacy by 6-fold. Comparing the release profiles with and without 
enzymes, an increased release of SN38 from G3.5-Gly-SN38 in the presence of 
pancreatin and carboxylesterase was observed, but no such increase for G3.5-?Ala-SN38 
was sen, indicating the increased susceptibility of the glycine linker to enzymatic 
degradation. 
In addition to difering release and toxicity profiles, G3.5-Gly-SN38 and G3.5-
?Ala-SN38 conjugates showed significant diferences in transepithelial transport. Both 
conjugates showed increased flux of SN38 relative to fre drug, which is critical for 
improving the oral bioavailability of SN38. In addition, neither conjugate showed short- 
or long-term efects on Caco-2 cels after a 2-hour treatment, ilustrating that conjugation 
of SN38 to G3.5 dendrimers can minimize intestinal toxicity while maximizing transport.  
Interestingly, transport of G3.5-Gly-SN38 was highly concentration-dependent while 
G3.5-?Ala-SN38 flux was unchanged betwen treatment with 10 and 100 ?M 
concentrations. This suggests that G3.5-Gly-SN38 may be transported primarily by a 
concentration gradient-driven proces, such as paracelular difusion, whereas a saturable 
proces, such as transcelular transport, governs G3.5-?Ala-SN38 transport. This 
phenomenon, however, needs further examination. While the uptake data suggests 
comparable uptake of the conjugates at 100 ?M this measurement could include surface 
bound dendrimer or dendrimer that would be degraded in the cel and not transcytosed. 
Therefore, reduction in G3.5-?Ala-SN38 transport at 100 ?m compared to G3.5-Gly-
SN38 transport could be due to decreases in both transcelular and paracelular pathways. 
 
133 
 
 
In Chapter 5 we wil show the impact of surface chemistry on the mechanism of transport 
and uptake of PAMAM dendrimers. Specificaly, reduction of surface charge on G3.5 
dendrimers by the addition of low molecular weight poly (ethylene glycol) is found to 
reduce tight junction opening, transepithelial transport and uptake in Caco-2 cels. Thus, 
it is possible that addition of a methyl group in the ?-alanine linker as wel as the increase 
of drug loading from 2.9 to 4.0 molecules of SN38 increased the hydrophobicity of the 
conjugates, hence reducing the degre to which G3.5-?Ala-SN38 conjugates opened the 
tight junctions, resulting in an overal transport mechanism dominated by the transcelular 
route. Importantly, these studies ilustrate the choice of drug linker and the degre of 
drug loading not only impact drug release and ultimate eficacy but can also impact 
transport. Therefore, dendrimer-drug conjugates for oral delivery must be carefully 
designed to met transport, release and eficacy demands.  
 
4.5 Conclusion 
In this Chapter we investigated G3.5-SN38 conjugates for oral delivery of SN38 
by determining their in vitro release profiles in simulated gastric and intestinal conditions 
and in the presence of carboxylesterase, their toxicity against intestinal cels and target 
colorectal cancer cels, and their transepithelial transport and celular uptake in Caco-2 
monolayers. We demonstrated that conjugation of SN38 to G3.5 dendrimers increased 
the transepithelial transport while simultaneously reducing intestinal toxicity compared to 
fre SN38, ilustrating the potential for these conjugates in oral drug delivery. A 
significant impact of linker chemistry on drug release and eficacy in HT-29 cels was 
shown with G3.5-Gly-SN38 showing lower in vitro stability and higher eficacy than 
 
134 
 
 
G3.5-?Ala-SN38 conjugates. The drug linker chemistry and drug loading also impacted 
the transport pathway with G3.5-Gly-SN38 having a concentration-dependent transport 
profile and G3.5-?Ala-SN38 conjugates having a saturable transport profile. G3.5-Gly-
SN38 shows promise for oral delivery of SN38 for the treatment of colorectal hepatic 
metastases. Treatment with 100 ?M G3.5-Gly-SN38 caused a 159-fold increase in SN38 
transepithelial transport compared to fre SN38 ilustrating its potential to enhance SN38 
bioavailability. In addition, G3.5-Gly-SN38 was relatively stable in gastric and intestinal 
milieu with increased release in the presence of liver carboxylesterase. Finaly, G3.5-
Gly-SN38 shows an IC
50
 of 0.60 ?M in HT-29 cels, which is aceptable for cancer 
therapy. Together these results show that PAMAM dendrimers have the potential to 
improve the oral bioavailability of potent anti-cancer therapeutics and that appropriate 
selection of drug linker is a critical step in designing dendrimers for oral drug delivery 
applications. 
 
 
 
135 
 
 
 
Chapter 5 : Transepithelial Transport of PEGylated Anionic Poly (amido 
amine) Dendrimers: Implications for Oral Drug Delivery 
 
5.1 Introduction 
As ilustrated in Chapter 4, PAMAM dendrimers can efectively enhance 
transepithelial transport of SN38. Previous studies in our laboratory [15, 18, 24, 148, 
149] and others [19, 144, 186, 187] indicate that dendrimers in a specified size and 
charge window can efectively translocate across the gastrointestinal epithelia. In 
addition, several studies have demonstrated that conjugation or complexation of drugs 
with PAMAM dendrimers can enhance the oral bioavailabilty of drugs normaly limited 
to intravenous administration, supporting dendrimers as viable oral drug delivery cariers 
[20, 113, 150, 151]. Due to their intrinsicaly low cytotoxicity and appreciable 
transepithelial permeation characteristics across Caco-2 monolayers and everted rat 
intestinal sac models [15, 144], anionic dendrimers show distinct advantages as vehicles 
for oral drug delivery, with higher generation dendrimers showing the greatest potential 
because of their large number of modifiable surface groups. As demonstrated in Chapter 
4, in addition to the surface charge, the drug linker can also play a significant role in 
transport and drug release. This Chapter aims to evaluate the potential of G3.5 and G4.5 
dendrimers in facilitating the delivery of drugs across the gastrointestinal tract, 
specificaly examining the efect of poly (ethylene glycol) (PEG) surface modification. 
PEGylation of drug delivery systems and bioactive agents is known to reduce 
toxicity and imunogenicity, influence pharmacokinetics and biodistribution and 
 
136 
 
 
enhance water solubility [188]. Okuda and colleagues [189, 190] showed that 
PEGylation of dendritic systems produces desirable biodistribution efects upon 
intravenous administration; however, the impact of PEGylation on transepithelial 
transport of PAMAM dendrimers across intestinal cels is presently unknown. In the 
context of oral drug delivery, PEG can be used as a surface modifier to modulate the 
degre and mechanism of transport or it can act as drug linker, altering release properties 
from the polymer backbone. In this Chapter we describe the synthesis and 
characterization of diferentialy PEGylated G3.5 and G4.5 anionic PAMAM dendrimers 
and evaluate their cytotoxicity, celular uptake and transport across Caco-2 cel 
monolayers. In addition, the efect of PEGylation on tight junction modulation was 
investigated. These studies provide the first evidence of the impact of PEG conjugation 
on dendrimer transepithelial transport. 
 
5.2 Materials and Methods 
5.2.1 Materials 
PAMAM G3.5 (reported molecular weight=12,931) and PAMAM G4.5 (reported 
molecular weight 26,258), [
14
C]-mannitol (specific activity 50mCi/mol), D
2
O, and 
Hank?s balanced salt solution (HBS bufer) salts were purchased from Sigma Aldrich 
(St. Louis, MO). [
3
H]-Acetic anhydride was purchased from American Radiolabeled 
Chemicals (St. Louis, MO). Superose 12 HR 10/300 GL column was purchased from 
Amersham Pharmacia Biotech (Piscataway, NJ). Caco-2 cels were purchased from 
American Type Cel Culture (Rockvile, MD). WST-1 cel proliferation reagent was 
 
137 
 
 
purchased from Roche Applied Science (Indianapolis, IN). The BD Biocoat Caco-2 
Asay was purchased from BD Biosciences (San Hose, CA). 
 
5.2.2 Conjugation of mPEG750 to PAMAM Dendrimers 
PEGylation of PAMAM G3.5 and G4.5 was achieved by formation of ester bonds 
betwen surface carboxyl groups of the dendrimers and hydroxyl terminated PEG 
(Molecular Weight (M
w
) 750) using benzotriazole-1-yl-oxy-tris-(dimethylamino)-
phosphoniumhexafluorophosphate (BOP) as a coupling agent and methanol as a solvent 
(Figure 5.1). 1, 2 and 4 equivalents of methoxy polyethylene glycol (mPEG) were reacted 
with dendrimers to yield samples G3.5-P1, G3.5-P1.4 and G3.5-P2.3, G4.5-P.85, G4.5-
P1.9 and G4.5-P2.8 respectively, where G represents the generation number and P 
represents the number of PEG chains conjugated. For each reaction, 75 mg of dendrimer 
and thre equivalents of BOP per equivalent of mPEG were disolved in anhydrous 
methanol. mPEG750 was disolved in methanol to a concentration of 100 mg/mL and the 
appropriate molar equivalent was added to the mixture. The solution was stired at room 
temperature for 72 hours, after which methanol was evaporated to leave the crude 
product. This product was disolved in distiled water and purified by dialysis against 
distiled water using 3,500 molecular weight cut off (MWCO) membranes (Spectrum 
Laboratories, Rancho Dominguez, CA). Subsequently, the product was freze dried and 
stored at 4?C. 
 
 
138 
 
 
 
 
Figure 5.1. PEGylation of G3.5 Dendrimer with mPEG750. 
 
 
139 
 
 
5.2.3 Characterization of PEGylated G3.5 and G4.5 Dendrimers 
Dendrimer-PEG conjugates were characterized by size exclusion chromatography 
(SEC) using an Acta FPLC system (Acta UPC 900, P-920, INV-907 from GE 
Healthcare) and phosphate buffered saline (PBS, pH 7.4) with 0.05% sodium azide as the 
eluent. Samples were injected onto the FPLC system at 5 mg/mL and simultaneously 
monitored for Ultra-Violet (UV), Refractive Index (RI), Multi-Angle Laser Light 
Scatering (MALS) and Dynamic Light Scatering (DLS) detection. SEC was used to 
confirm the absence of low molecular weight impurities and to compare the elution 
volumes of modified and native dendrimers. In addition, the dendrimer peak was 
analyzed to determine the hydrodynamic radius using a Wyat Quasi-Elastic Light 
Scatering Detector (QELS) and the calculations were performed using the Astra 5.3.4 
software. Al samples were run in duplicate. 
1
H NMR was used to determine the 
number of PEG chains conjugated per dendrimer. 
1
H NMR samples were prepared at 
approximately 8 mg/mL in D
2
O with 0.05 wt % 3-(trimethylsilyl) propionic-2,2,3,3-d
4
 
acid (TSP-d
4
). 
1
H Nuclear Magnetic Resonance (NMR) spectra were obtained using 
Varian 500 MHz FT NMR and were procesed using Spinworks software (Kirk Marat, 
University of Mannitoba, Winnipeg, Canada, ? 2008). The number of PEG chains per 
dendrimer was determined from ratios of integral values for peaks asigned to PEG (3.6-
3.7 ppm) and dendrimer (2-3.6). Finaly, the zeta potential of PEGylated dendrimers was 
obtained using the Malvern Nano-ZS system. Dendrimer solutions were prepared at 5 
mg/mL in DI water and analyzed in triplicate. 
 
 
 
140 
 
 
5.2.4 Synthesis of Radiolabeled Dendrimers 
Dendrimers and dendrimer-PEG conjugates were radiolabeled using [
3
H]-acetic 
anhydride (American Radiolabeled Chemicals, St. Louis, MO) which reacts with internal 
amines known to be present in carboxyl terminated dendrimers due to defects formed 
during their synthesis [191].
 
Dendrimers and dendrimer-PEG conjugates were disolved 
in methanol and reacted with thre equivalents of [
3
H]-acetic anhydride in the presence of 
exces triethylamine overnight. The methanol was dried under a stream of nitrogen and 
the product was redisolved in distiled water. Triethylamine and unreacted acetic 
anhydride were removed by Sephadex G-25 protein desalting (PD10) columns. Specific 
activities of the radiolabeled compounds were also calculated. 
 
5.2.5 Caco-2 Cel Culture 
Caco-2 cels (pasages 20-40) were cultured as described in Section 3.2.3. 
 
5.2.6 Cytotoxicity Asay 
Cytotoxicity of unmodified and diferentialy PEGylated PAMAM dendrimers 
was asesed by the water soluble tetrazolium salt (WST-1) asay as described in Section 
3.2.4. 
 
5.2.7 Celular Uptake Studies 
The efect of PEGylation on celular uptake of radiolabeled dendrimers was 
investigated. Caco-2 cels were seded at 40,000 cels/ wel in 24-wel cel culture plates 
(Corning, Corning NY) and maintained at 37?C, 95% relative humidity and 5% CO
2
 for 
 
141 
 
 
48 hours. Cels were washed with warm HBS buffer and 300 ?L of 0.02 mM dendrimer 
solution in HBS was added for 30 or 60 minutes. After the given incubation period, the 
cels were washed twice with ice cold HBS to halt the uptake proces. They were then 
lysed with NaOH and neutralized with HCl. Uptake was measured by quantifying the 
cel-asociated radioactivity using a liquid scintilation counter (Beckman Coulter, 
Fullerton, CA) with Econosafe scintilation cocktail (Research Products International, 
Mount Prospect, IL). Uptake was normalized to total protein content using the 
Bicinchoninic acid (BCA) protein asay kit (Pierce, Evanston, IL). Statistical significance 
was determined by a two-way analysis of variance and Bonferoni post-hoc correction. 
 
5.2.8 Transepithelial Permeability Asesment 
The efect of diferential PEGylation on transport of radiolabeled dendrimers 
across Caco-2 cel monolayers was asesed in the BD Biocoat HTS Caco-2 Asay 
System (BD Biosciences, San Hose, CA). The thre-day asay protocol defined by the 
manufacturer was used to prepare the monolayers. Briefly, Caco-2 cels were grown past 
confluency to a density of >250,000 cels/cm
2
 in T-25 flasks. Cels were seded at 
200,000 cels per wel, in 24-wel Transwel
?
 plates with fibrilar collagen-coated cel 
culture inserts in basal cel seding medium supplemented with MITO+ (mitogenic 
stimulating) serum extender and 10% FBS. After 24 hours, the medium was switched to 
entero-STIM, a fully defined media containing butyric acid, which induces diferentiation 
of intestinal epithelial cels and form a competent monolayer [192]. After an additional 
48 hours, the monolayers were used for experiments. Cels were washed twice with warm 
HBS buffer. To test dendrimer permeability, 100 ?L of dendrimer solution (0.1 mM) 
 
142 
 
 
was added to the apical compartment and 600 ?L of HBS added to the basal 
compartment. After 2 hours, 400 ?L from the basal compartment was taken for 
scintilation counting. Apparent permeability (P
ap
) was calculated by: 
! 
P
app
=
dQ
A"C
0
"dt
 (Eq. 5.1) 
where dQ/dt is the change in the amount of solute over time (permeability rate), A is the 
surface area of the insert, and C
0
 is the donor concentration. [
14
C]-Mannitol was used to 
monitor monolayer integrity. The transepithelial flux of the paracelular marker [
14
C]-
mannitol was 2.6 x 10
-6
 cm/s, which demonstrates monolayer integrity as shown 
previously [193, 194]. Statistical significance was determined by Tukey?s Multiple 
Comparison test. 
 
5.2.9 Ocludin Staining 
Caco-2 cels were seded at 20,000 cels/cm
2
 on colagen-coated four chamber 
culture slides (BD Biosciences, Bedford MA), maintained under normal cel culture 
conditions for 5 days and used for experiments when the cels reached confluence. The 
cels were equilibrated in HBS for 2 hours prior to the experiment. The cels were 
incubated with 300 ?l of 0.1 mM dendrimer solutions for 2 hours at 37?C and then 
washed thre times with ice cold PBS to remove the dendrimers. G3.5, G3.5-P1, G4.5 
and G4.5-P.85 were used with HBS as a control. The cels were then fixed with 300 ?l 
of 4% paraformaldehyde solution for 20 minutes at room temperature, washed twice with 
25 mM glycine and once with PBS and then permeabilized with 300 ?l of 0.2% Triton X-
100 in blocking solution made of 1% bovine serum albumin (BSA) in PBS for 20 
minutes at room temperature. Cels were washed thre times with PBS and then 
 
143 
 
 
incubated with blocking solution for 30 minutes. The blocking solution was removed and 
the cels were incubated with 300 ?l of 2 ?M mouse anti-occludin (Invitrogen (Zymed), 
Carlsbad, CA) and kept at 4?C overnight. Cels were washed thre times with blocking 
solution and then incubated with the same solution for 30 minutes. After the blocking 
solution was removed, the cels were incubated with 300 ?l of 10 ?g/mL Alexa Fluor 568 
goat anti-mouse IgG (H+L) (Invitrogen, Molecular Probes, Carlsbad, CA) for 1 hour. 
They were then washed thre times with PBS and the chambers were removed. Gel 
mount was added to each region and alowed to dry for 2 hours at room temperature. The 
slides were then sealed with clear nail polish and kept at 4?C prior to visualization. 
Slides were visualized using a Zeis LSM510 META confocal laser scanning 
microscope (Carl Zeis, Jena, Germany). Alexa Fluor 568 was visualized with excitation 
and emision wavelengths of 543 and 603 nm respectively. Z-stacks were obtained with 
the following microscope setings: 63x oil objective, 1 airy unit pinhole, 543 nm laser set 
to 20% power with 1000 gain, 0.1 amplifier ofset and 1 amplifier gain. The scan speed 
was set to 8 in line mode with mean averaging set to 4. Z-stack sections were taken 0.5 
?m apart with 15 sections per region. Thre z-stacks were acquired per treatment and 
analyzed using Volocity 3D imaging software. (Version 4.3.2; Improvision, Inc., 
Lexington, MA). Red voxels, corresponding to occludin staining, were quantified by 
thresholding the intensity betwen 25% and 100%. The number of red voxels was 
quantified for each region and normalized to the number of cels. Results are reported as 
mean ? standard deviation and statistical significance was determined using a one-tailed 
Student?s t-test. 
 
 
144 
 
 
5.3 Results 
5.3.1 Synthesis and Characterization of PEGylated Anionic PAMAM Dendrimers 
1
H NMR studies confirmed the formation of the conjugates with new peaks 
corresponding to the protons from PEG. 
1
H NMR spectra are shown in Appendix 3. 
1
H 
NMR quantification showed a concomitant increase in the number of PEG moieties 
conjugated with the number of equivalents of PEG, but this increase was les than the 
stoichiometric amount used (Table 5.1). This may be atributed to steric hindrance, which 
was more evident in lower generation dendrimers likely due to their lower number of 
available acid groups and smaler surface area. 
Characterization by SEC showed the absence of fre PEG and other smal 
molecular weight impurities (data not shown). Elution volumes of native and PEGylated 
dendrimers are shown in Table 5.1. While there is a decrease in elution volume from 
G3.5 dendrimers to G4.5 dendrimers due to the larger size, native and modified 
dendrimers of the same generation do not show any significant diferences in elution 
volume. This ilustrates that PEGylation does not increase the hydrodynamic volume of 
the dendrimers, suggesting that PEG is preferentialy wrapped around the dendrimer 
rather than protruding from the surface. Hydrodynamic radii of the conjugates measured 
by dynamic light scatering support this hypothesis (Table 5.1). Increasing the degre of 
PEGylation does not increase the hydrodynamic radii. In fact, the hydrodynamic radii are 
slightly decreased, indicating that PEG may shield the surface charge and decrease 
charge-charge repulsion of functional groups and therefore the size of the polymers in 
solution.  
 
 
145 
 
 
Table 5.1. Characteristics of PAMAM Dendrimer-PEG Conjugates 
Dendrimer 
Sample 
# mPEG 
Equivalents 
# mPEG 
Conjugated* 
Calculated 
Molecular 
Weight* 
Specific 
Activity 
(mCi/mol) 
SEC Elution 
Volume 
(ml) 
Hydrodynamic 
Radius 
(nm) 
G3.5 - - 12931 1.06 13.2 1.90 ? 0.0 
G3.5-P1 1 1.01 1368 3.25 13.2 1.65 ? 0.07 
G3.5-P1.4 2 1.36 13951 3.96 13.2 1.65 ? 0.07 
G3.5-P4.5 4 2.32 14671 3.91 13.2 1.70 ? 0.0 
G4.5 - - 26258 2.64 12.8 2.35? 0.07 
G4.5-P.85 1 0.85 26896 4.47 12.5 2.10 ? 0.0 
G4.5-P1.9 2 1.86 27653 5.1 12.6 2.15 ? 0.07 
G4.5-P2.8 4 2.80 28358 4.81 12.6 2.10 ? 0.0 
*Calculated from NMR data 
**Calculated based on NMR data and reported (Sigma) dendrimer molecular weights
 
146 
 
 
Zeta potential measurements shed light on the impact of PEGylation on dendrimer 
surface charge. Figure 5.2 ilustrates that zeta potential becomes les negative upon 
further PEGylation, with G3.5 and G4.5 dendrimers showing similar behavior, 
confirming that PEG shields surface charge. Shcharbin et al. [195] reported a zeta 
potential for G4.5 PAMAM dendrimers in deionized water as -56 ? 0.5 mV. Comparing 
this with our data shows that even one PEG is able to shield a significant amount of 
charge on the dendrimer surface. 
 
5.3.2 Short-Term Cytotoxicity 
While amine-terminated PAMAM dendrimers are known to be highly cytotoxic at 
higher generations, anionic dendrimers have been found to be much les toxic [16]. In 
this study we examined whether PEGylation of anionic dendrimers would influence their 
toxicity profile. G3.5 and G4.5 dendrimers and dendrimer-PEG conjugates were tested at 
concentrations up to 0.1 mM and did not show appreciable cytotoxicity (cel viability > 
90%) over the entire range of concentrations tested (Figures 5.3, 5.4). This toxicity 
profile confirmed that treatment of Caco-2 cels with up to 0.1 mM of the dendrimer-PEG 
conjugates would not afect cel viability during uptake or transport experiments. In 
addition, dendrimer-PEG conjugates would be safe to use for oral drug delivery 
applications. 
 
 
147 
 
 
 
Figure 5.2. Zeta Potential of PEGylated G3.5 and G4.5 PAMAM Dendrimers. Zeta 
potential becomes les negative with addition of PEG to G3.5 and G4.5 dendrimers. 
Results are reported as mean +/- standard deviation (n=3). 
 
 
148 
 
 
 
Figure 5.3. Caco-2 Cel Viability in the Presence of G3.5 Native and PEGylated 
Dendrimers after a 2-hour Incubation Time. Results are reported as mean +/- standard 
deviation (n=6).  Treatment with up to 0.1 mM G3.5 of G3.5-PEG conjugates maintains 
cel viability. 
 
149 
 
 
 
Figure 5.4. Caco-2 Cel Viability in the Presence of G4.5 Native and PEGylated 
Dendrimers after a 2-hour Incubation Time. Results are reported as mean +/- standard 
deviation (n=6). Treatment with up to 0.1 mM G4.5 and G4.5-PEG conjugates maintains 
cel viability. 
 
 
 
150 
 
 
5.3.3 Celular Uptake 
The impact of PEGylation on time-dependent celular uptake of dendrimers was 
asesed in Caco-2 cels (Figures 5.5, 5.6). Uptake studies shed light on the degre to 
which the dendrimers are transported through transcelular pathways (pasive difusion 
and vesicular mechanisms) in the gastrointestinal epithelium and their eventual uptake in 
tumor cels at the site of action. In general, PEGylation decreased uptake of G3.5 
dendrimers, but this efect is statisticaly significant only after 60 minutes. Notably, there 
is no significant impact of the degre of PEGylation on uptake, i.e. a single PEG chain 
can reduce dendrimer uptake by almost 50% after a 60-minute incubation period. In 
contrast, for G4.5 dendrimers, conjugation of 1 PEG chain appears to increase uptake, 
with further addition of PEG decreasing uptake from this point. 
 
5.3.4 Transepithelial Transport 
In order to ases the suitability of PEGylated dendrimers for oral drug delivery, 
their transport across Caco-2 cel monolayers was studied and compared to unmodified 
dendrimers. Transport experiments complement the uptake studies, giving an overal 
picture of transepithelial transport including both transcelular and paracelular pathways. 
PEGylation significantly decreased transport of both G3.5 and G4.5 dendrimers (Figures 
5.7, 5.8), yielding apparent permeabilities approximately 60-70% lower than unmodified 
dendrimers. Despite this decrease in overal transport flux, PEGylated dendrimers stil 
show appreciable transport compared to traditional linear polymers [196], indicating that 
the PEGylated dendrimers continue to enhance their own transport to some degre. 
Addition of more than one PEG per dendrimer did not further decrease transport. 
 
151 
 
 
 
Figure 5.5. Uptake of G3.5 Native and Diferentialy PEGylated Dendrimers. 
Caco-2 cels were treated at 0.02 mM for 30 and 60 minutes. Results are reported as 
mean +/- standard deviation (n=3). (**) and (***) denote significant diferences from 
unmodified dendrimers with p<0.01 and p<0.001 respectively. 
 
152 
 
 
 
Figure 5.6. Uptake of G4.5 Native and Diferentialy PEGylated Dendrimers. Caco-2 
cels were treated at 0.02 mM for 30 and 60 minutes. Results are reported as mean +/- 
standard deviation (n=3). (*),(**) and (***) denote significant diferences from 
unmodified dendrimers with p<0.05, p<0.01 and p<0.001 respectively. 
 
153 
 
 
 
Figure 5.7. Apparent Permeability of G3.5 Native and Diferentialy PEGylated 
Dendrimers. Caco-2 cel monolayers were treated at 0.1 mM for a 2-hour incubation 
time. Results are reported as mean +/- standard deviation (n=3). (*) denotes a significant 
diference from unmodified dendrimers with p<0.001. 
 
154 
 
 
 
Figure 5.8. Apparent Permeability of G4.5 Native and Diferentialy PEGylated 
Dendrimers. Caco-2 cel monolayers were treated at 0.1 mM for a 2-hour incubation 
time. Results are reported as mean +/- standard deviation (n=3). (*) denotes a significant 
diference from unmodified dendrimers with p<0.001. 
 
 
 
155 
 
 
5.3.5 Tight Junction Opening Monitored by Ocludin Staining 
To further ases the influence of dendrimer PEGylation on tight junctional 
modulation, Caco-2 cels were pretreated with native and PEGylated dendrimers and 
stained for occludin, a marker protein for tight junction integrity [15]. Previously, it was 
shown that Caco-2 cels treated with 1 mM G1.5, G2.5 and G3.5 dendrimers showed 
disrupted occludin staining paterns and increased occludin acesibility as compared to 
cels with no polymer treatment [15]. In this study, we used a ten-fold lower 
concentration (0.1 mM) in order to miic the conditions used in the transport asay. 
Control cels treated with HBS alone (Figure 5.9 E) show thin lines of red staining 
corresponding to occludin proteins linking cel membranes. In contrast, dendrimer-treated 
cels (Figure 5.9A-D) show brighter, wider bands of staining with increased intracelular 
staining, indicating acesibility of occludin protein to its antibody due to the opening of 
the tight junctions. In the case of G3.5 (Figure 5.9 A) confocal images show some 
amount of cel detachment, indicating tight junction opening. Al dendrimers studied 
showed a statisticaly significant increase in occludin staining compared to the HBS 
control (p<0.05) (Figure 5.10), ilustrating that native and PEGylated dendrimers open 
tight junctions to some degre, even at relatively low concentrations (0.1 mM). 
 
5.4 Discusion 
In this chapter, we investigated the impact of diferential PEGylation of G3.5 and 
G4.5 dendrimers on uptake, transport and tight junction modulation in the context of oral 
delivery. Diferential PEGylation of G3.5 and G4.5 dendrimers led to significant 
diferences in celular uptake, which can be atributed to charge-shielding properties of 
 
156 
 
 
 
Figure 5.9. Staining of the Tight Junction Protein Ocludin in the Presence and Absence 
of Dendrimers Visualized by Confocal Microscopy. Caco-2 cels incubated for 2 hours 
with 0.1 mM (A) G3.5, (B) G3.5-P1, (C) G4.5, (D) G4.5-P.85 and (E) with no polymer 
treatment. Main panels ilustrate the xy plane; horizontal bars ilustrate the xz plane; 
vertical bars ilustrate the yz plane. Scale bars equal 28 ?m. 
 
157 
 
 
 
Figure 5.10. Quantification of Ocludin Staining in the Presence and Absence of 
Dendrimers. Results reported are number of ?red? voxels per region normalized to the 
number of cels in each region, mean +/- standard deviation (n=3). One-tailed Student?s t-
tests were used to determine statistical significance. (*) and (**) denote significant 
diferences from unmodified dendrimers with p<0.05 and p<0.01 respectively. 
 
158 
 
 
PEG. The flexible chain of PEG can wrap around the rigid, spherical dendrimer and 
shield some of the negative charge on the dendrimer surface. Methoxy-terminated 750 Da 
PEG contains 16 subunits, giving it an overal extended chain length of 5.7 nm based on 
calculated bond lengths and angles. In comparison, G3.5 dendrimers have a reported 
diameter of approximately 4 nm and G4.5 dendrimers approximately 5 nm [136]. This 
suggests that the PEG chain, due to its flexible random coil conformation, is of 
comparable size to the dendrimers. For the smaler G3.5 dendrimers, one PEG chain is 
enough to reduce the interactions with the cels, leading to decreased celular uptake for 
al PEGylation ratios. This may be due to a stealth-like efect imposed by PEG, creating 
fewer interactions betwen the dendrimers and the cels. In contrast, for G4.5 dendrimers, 
addition of 1 PEG chain actualy increases uptake as compared to unmodified G4.5. This 
may be explained by the charge density of the dendrimers, which can be calculated by 
dividing the number of charges (64, 128) by the surface area of the dendrimer, asuming 
a spherical conformation [197]. The higher charge density of G4.5 dendrimers (1.6 
charges/nm
2
) compared to G3.5 dendrimers (1.3 charges/nm
2
) can presumably cause 
repulsion with the negatively charged cel membrane. Addition of 1 PEG chain reduces 
the charge on the surface, increasing the uptake to a point comparable to G3.5 
dendrimers. However, addition of more PEG shields more of the negative charge, again, 
creating a stealth system and decreasing celular uptake.  
Correlating the uptake of the dendrimers with the zeta potential data shows that 
there may be an ideal zeta potential around -30 mV that promotes celular uptake. In 
general, dendrimers with zeta potentials more negative than -30 mV had reduced celular 
uptake (ie. G4.5) and dendrimers with zeta potentials greater than -30 mV (ie PEGylated 
 
159 
 
 
G3.5) also show lower uptake. Interestingly, incubation time-dependent behavior was 
observed betwen dendrimer generations. While there is no statistical diference betwen 
uptake of unmodified or PEGylated G3.5 at 30 and 60 minutes, there is a significant 
increase (p<0.01) in uptake for the G4.5 dendrimers, suggesting that steady-state uptake 
of these larger dendrimers has not yet been achieved. 
Dendrimers are known to be transported across Caco-2 cel monolayers by both 
transcelular and paracelular pathways [16, 18, 147, 148]. Uptake studies and tight 
junction modulation experiments in Caco-2 cels help to elucidate the relative 
contributions of these proceses. Because PEGylation of G3.5 dendrimers decreases 
uptake, this surface modification may be of use in cases where it is desirable to transport 
the cargo across the cels while avoiding entrapment within the cels. Combining the 
uptake results with the occludin staining results alows us to understand the relative 
contributions of paracelular and transcelular transport of PEGylated dendrimers 
compared to unmodified dendrimers. In the case of G3.5 dendrimers, PEGylation causes 
a significant decrease in uptake and a moderate decrease in tight junction modulation, 
indicating that these dendrimers have les paracelular and les transcelular transport 
compared to unmodified dendrimers. By contrast, PEGylation of G4.5 dendrimers causes 
a modest increase in uptake and has only slightly more occludin acesibility than the 
control, suggesting that PEGylated G4.5 dendrimers have more transcelular transport 
and much les paracelular transport than unmodified G4.5 dendrimers. This sheds light 
on how PEGylation can impact transport mechanisms as wel as overal transport rates. 
The strong correlation betwen the transport studies and the occludin staining studies 
 
160 
 
 
suggest that overal, anionic dendrimers are transported primarily through the paracelular 
route. 
PAMAM dendrimers can be used for oral drug delivery to release drugs in the 
gastrointestinal tract and transport drugs into or across the intestinal barier. PEGylation 
of PAMAM dendrimers can modulate toxicity, functionalize the surface for drug 
conjugation and facilitate drug release. While PEGylation decreases the transepithelial 
transport of anionic dendrimers, the conjugates stil show appreciable transport, 
compared to traditional linear polymers [196], and have the potential for increased 
functionality. This work demonstrates that degre of PEGylation and dendrimer 
generation can modulate the mechanisms of transport, and can be custom tailored for 
diferent oral drug delivery applications. 
 
5.5 Conclusion 
In this Chapter we described the efect of PEGylation on cytotoxicity, uptake and 
transport of G3.5 and G4.5 anionic PAMAM dendrimers across Caco-2 cels. In the 
concentration range studied, PEGylation of these dendrimers maintained cel viability. 
PEGylation decreased uptake and transport for G3.5 dendrimers, whereas PEGylated 
G4.5 dendrimers demonstrated increased uptake with a concomitant decrease in overal 
dendrimer transport. Ocludin staining of Caco-2 cel monolayers in the presence of 
conjugates showed that PEGylated dendrimers opened the tight junctions to a leser 
extent than native dendrimers, indicating a reduction in paracelular transport. While 
PEGylated dendrimers showed decreased transport rates, they were stil transported to an 
appreciable extent compared to traditional linear macromolecules [196] and ofer 
 
161 
 
 
advantages of facilitated drug conjugation and release as wel as potential for improved 
biodistribution. Together these studies demonstrate that in the design of PAMAM-PEG 
conjugates for oral drug delivery, the extent of transepithelial transport, uptake as wel as 
the mechanism of transport can be controlled by a judicious choice of dendrimer 
generation and degre of PEGylation. These parameters can be custom- tailored to the 
specific needs of a desired drug delivery application. 
 
 
 
162 
 
 
 
Chapter 6 : Conclusions and Future Directions 
 
6.1 Conclusions 
Oral administration of polymer-drug conjugates for cancer treatment has the 
potential to improve the lives of cancer patients by reducing hospital visits and treatment 
costs while improving therapeutic eficacy of the drug by minimizing side efects and 
enhancing drug concentration at the tumor site. Due to the large number of emerging 
smal molecule chemotherapy drugs with poor water solubility, low bioavailability and 
significant off-target toxicities, there is a significant need for oraly administrable 
polymer therapeutics. As described in Chapter 2, it has been previously established that 
within a size and charge window, poly (amido amine) (PAMAM) dendrimers can 
permeate the epithelial layer of the gut. In this disertation we evaluated anionic 
PAMAM dendrimers for their potential as oral drug delivery cariers and investigated the 
mechanisms by which they are transported across the epithelial barier. 
In Chapter 3 we demonstrated that G3.5 dendrimers are endocytosed by dynamin-
dependent mechanisms, and that celular internalization occurs by clathrin- and caveolin-
mediated endocytosis while transcytosis is governed by clathrin-mediated pathways. We 
further confirmed the clathrin-endocytosis mechanism by monitoring G3.5 traficking to 
the endosomes and lysosomes up to 30 minutes after celular internalization. These 
studies ilustrate that dendrimers take advantage of receptor-mediated mechanisms for 
celular entry and can be found in the lyososomal and endosomal compartments after 
celular uptake. This knowledge wil aid in the design of PAMAM dendrimers for oral 
drug delivery in diferent potential applications. In the case of an oral therapy that is 
 
163 
 
 
localy targeted to the intestinal tisue for the treatment of diseases such as Crohn?s, a 
pH-sensitive or lysosomal enzyme-cleavable linker could be used to promote drug release 
in the intestinal cels. Alternatively, in the case of a chemotherapy drug that must be 
targeted to a distant tumor site, the linker could be designed to avoid release in these 
compartments to minimize intestinal toxicity. Thus, knowledge of the celular entry route 
and sub-celular traficking of the dendrimer carier wil aid in rational design of 
dendrimers for drug delivery applications. In Chapter 3, we monitored tight junction 
opening by dendrimers in the presence and absence of dynasore, a compound that inhibits 
dynamin-dependent endocytosis. Tight junction opening was found to be significantly 
decreased in the presence of dynasore suggesting that dendrimer celular internalization 
may be a requisite step prior to opening of tight junctions by dendrimers. This implies 
that dendrimer modulation of tight junctions is not simply due to chelation of 
extracelular calcium, but may be a more complicated mechanism triggered by an 
intracelular signaling cascade that acts on the tight junction proteins. This ilustrates the 
interconnectednes of the transcelular and paracelular routes. Importantly, this type of 
intracelular regulation is reversible, preventing permanent damage to the tight junctions 
and suggesting the safety of dendrimers as permeation enhancers. 
In Chapter 4 we investigated G3.5-SN38 conjugates for oral delivery of the 
poorly-bioavailable chemotherapy drug SN38 in targeting colorectal hepatic metastases. 
In particular, we examined G3.5-Gly-SN38 and G3.5-?Ala-SN38 conjugates. These 
conjugates reduced the toxicity of SN38 towards intestinal cels and maintained Caco-2 
cel viability with treatment concentrations of up to 100 ?M for 2 hours. This ilustrates 
that the conjugates would be safe for oral administration and can minimize the known 
 
164 
 
 
gastrointestinal toxicities asociated with SN38. In addition, both conjugates increased 
the transport of SN38 across Caco-2 cel monolayers, improving transport up to 160-fold 
compared to fre drug, showing their potential in improving the oral bioavailability of 
SN38. Release profiles of the conjugates showed that G3.5-?Ala-SN38 was very stable, 
with only 1%, 4% and 8% release in SGF with pepsin, SIF with pancreatin and PBS with 
carboxylesterase at 6, 24 and 48 hours respectively. In contrast, G3.5-Gly-SN38 showed 
lower gastrointestinal stability with 10% and 20% released in the simulated gastric and 
intestinal environments with increased release in the presence of pancreatin compared to 
SIF alone. In addition, G3.5-Gly-SN38 released up to 56% of SN38 in the presence of 
liver carboxylesterase, suggesting the potential for controlled release in the liver 
environment. Finaly, G3.5-Gly-SN38 and G3.5-?Ala-SN38 conjugates showed IC
50
 
values of 0.60 and 3.59 ?M in HT-29 cels treated for 48 hours, suggesting their eficacy 
in the micromolar range in target colorectal cancer cels. These studies demonstrate that 
conjugation of SN38 to G3.5 dendrimers can reduce intestinal toxicity and improve oral 
bioavailability.  Importantly, the linker chemistry (glycine vs. ?-alanine) impacted the 
release profiles as wel as the concentration-dependence in transport, ilustrating the 
importance of choosing the appropriate linker for a given drug delivery application. 
G3.5-Gly-SN38 conjugates showed reasonable gastrointestinal stability with increased 
release in the presence of carboxylesterase, and an IC
50 
of 0.6 ?M in HT-29 cels 
suggesting their potential for oral therapy of colorectal hepatic metastases.  
Finaly, in Chapter 5 we investigated the impact of PEGylation on G3.5 and G4.5 
dendrimers in the context of oral delivery. Conjugation of smal PEG chains to 
dendrimers can improve biocompatibility, influence transport properties and serve as a 
 
165 
 
 
potential drug spacer. PEGylation decreased transport of G3.5 and G4.5 dendrimers and 
reduced tight junction opening, ilustrating the importance of charge in tight junction 
modulation and transport. In addition, incremental PEGylation had diferential efects on 
celular uptake, with PEGylation decreasing G3.5 uptake while smal amounts of 
PEGylation increased G4.5 uptake with additional PEGylation decreasing uptake. This 
suggests a balance betwen reducing charge-charge repulsion betwen anionic 
dendrimers and the negative cel membrane, enhancing celular uptake, and creating a 
stealth system, reducing celular uptake. These studies ilustrate that even smal changes 
in surface chemistry, particularly changes that cause reduction in dendrimer charge, can 
have significant impacts on celular uptake, tight junction opening and overal 
transepithelial transport. Therefore, dendrimer-drug conjugates must be carefully 
designed in oral delivery applications to balance drug loading with permeability to create 
an efective drug delivery carier. Taken together, these studies ilustrate that anionic 
PAMAM dendrimers show potential in oral drug delivery and that with careful design 
they can be custom-tailored to a given application, enhancing intestinal permeability of 
the cargo while promoting specific release at the site of action.  
 
6.2 Future Directions 
While we have elucidated some details of the mechanism of transport and tight 
junction opening of G3.5 dendrimers, further studies should focus on identifying the 
signaling pathways involved in tight junction opening and the tight junction proteins that 
are afected. In particular, the impact of dendrimer treatment on claudin, ZO-1 and JAM 
should be investigated and the involvement of myosin light chain kinase in tight junction 
 
166 
 
 
opening should be determined. In addition, it is important to distinguish if tight junction 
modulation is due to reorganization of the tight junction proteins or if protein expresion 
is being impacted. Comparison to permeation enhancers with known mechanisms of 
action such as EGTA and sodium caprate wil also be useful to delineate the precise 
mechanism of dendrimer-mediated tight junction opening. 
G3.5-Gly-SN38 shows promising in vitro results suggesting potential for oral 
therapy of colorectal hepatic metastases. Future studies should evaluate G3.5-Gly-SN38 
in vivo activity including biodistribution after oral administration and eficacy studies in a 
colorectal hepatic metastasis mouse model. In addition, alternative linker chemistries 
should be investigated in order to optimize SN38 release in the presence of liver 
carboxylesterase. Since Gly-SN38 appears to be a substrate for carboxylesterase, 
peptide-Gly linkers or PEG-Gly linkers may enhance SN38 release by increasing 
acesibility of the ester bond to the active site of carboxylesterase. Computational 
modeling of the dendrimer-SN38 conjugate interaction with carboxylesterase may also be 
useful for linker selection. Strategies to improve drug loading on anionic dendrimers 
should also be investigated. Sub-optimal drug loading of SN38 was presumably due to 
low reactivity of the carboxylic acid terminal groups, low reactivity of the SN38 OH and 
steric hindrance. As an alternative strategy, the drug could be conjugated to amine-
terminated dendrimers with subsequent modification of the remaining amine groups to 
carboxylic acid groups to complete the half-generation. Detailed studies should addres 
the balance of drug loading with intestinal permeability. 
Finaly, while PEGylation decreased transport compared to unmodified 
dendrimers it stil may be a useful surface modification or drug spacer for oral delivery 
 
167 
 
 
applications. Dendrimer-drug conjugates using PEG spacers should be synthesized to 
investigate if PEG can improve drug release in the presence of target enzymes without 
significantly decreasing permeability. In addition, higher generation dendrimers should 
be investigated for their greater potential of drug loading and enhanced EPR efect. In 
particular, it would be interesting to investigate the influence of polymer architecture as 
dendrimer generation increases on drug release and intestinal permeability. These 
proposed studies would serve to build upon the conclusions of this work and advance this 
research towards a fully realized oral dendrimer-drug delivery system. 
 
168 
 
 
   
Appendix 1: Visualization of Intracelular Traficking of G3.5 Dendrimers 
and Transferin in Caco-2 Cels 
 
In Chapter 3, intracelular traficking of Oregon Gren-labeled G3.5 dendrimers 
and Alexa-Fluor-488-labeled transferin (control) over time in Caco-2 cels was 
described. The presence of G3.5 and transferin in the early endosomes and lysosomes at 
diferent time points was quantified by colocalization of gren (dendrimer/ transferin) 
and red (endosome/ lysosome) fluorescence by confocal microscopy. M
x
, the 
colocalization coeficient, was reported as the average of four z-stacks for each treatment 
and time point. Colocalization of dendrimers with early endosomes and lysosomes 
provides further confirmation of clathrin-mediated endocytosis [18]. In addition, 
comparison of dendrimer traficking paterns to transferin, a clasical clathrin RME 
ligand, alows for beter understanding of the traficking pathway and time course [198].  
Figures A1.1 and A2.2 show sample images from each treatment and time point 
in the traficking study. The early endosomes are present in the cels as punctate vesicles, 
often clustered toward the outside of the cel, away from the nucleus. In contrast, 
lysosomal proteins are found in close proximity to the nucleus and often overlap with the 
blue nuclear signal. These observations provide visual confirmation of the proper 
labeling of endosomal and lysosomal compartments. These images provide visual 
evidence of G3.5 dendrimer traficking over time and are complementary to the 
colocalization quantification provided in Chapter 3. 
 
 
169 
 
 
 
Figure A1.1. Visualization of G3.5 Dendrimer Traficking over Time in Caco-2 Cels by 
Confocal Microscopy. G3.5 dendrimer is labeled with Oregon Gren (gren), early 
endosomes (EA-1) and lysosomes are labeled with secondary antibody Alexa Fluor-568 
goat anti-rabbit IgG (red) and the cel nuclei are labeled with DAPI (blue). Yelow 
regions indicate overlapping gren and red signals (colocalization). Z-stacks are depicted 
as a main panel (xy plane), vertical panel (xz plane) and horizontal panel (yz plane). 
Scale bar = 21 ?m. 
 
 
170 
 
 
 
Figure A1.2. Visualization of Transferin Traficking over Time in Caco-2 Cels by 
Confocal Microscopy. Transferin is labeled with Alexa Fluor-488 (gren), early 
endosomes (EA-1) and lysosomes are labeled with secondary antibody Alexa Fluor-568 
goat anti-rabbit IgG (red) and the cel nuclei are labeled with DAPI (blue). Yelow 
regions indicate overlapping gren and red signals (colocalization). Z-stacks are depicted 
as a main panel (xy plane), vertical panel (xz plane) and horizontal panel (yz plane). 
Scale bar = 21 ?m. 
 
171 
 
 
 
 
Appendix 2: Quantification of SN38 by High Presure Liquid 
Chromatography 
 
In Chapter 4, stability studies of G3.5-SN38 conjugates were described where 
release of fre SN38 over time was monitored by high presure liquid chromatography 
(HPLC). Several HPLC methods to detect SN38 have been described in the literature 
[199, 200]. In this Appendix we wil describe the detailed HPLC methods used to 
quantify SN38 release which have been adapted from Vijayalakshmi et al. [161]. 
As described in Chapter 4, release samples were disolved in 50/50 DMSO/0.1 N 
HCl before HPLC analysis. This injection bufer was used to ensure that al SN38 is in 
the lactone, or closed ring form, which is favored at low pH. DMSO was used to 
solubilize the SN38 since SN38 has very low water solubility. A gradient method, with 
methanol and HPLC-grade water with 0.1% TFA as solvents, was used with a total flow 
rate of 1 ml per minute and an injection volume of 20 ?L. Table A2.1 shows the details 
of solvent flow during the 20-minute gradient method. Transitions betwen solvent 
compositions were acomplished by a linear gradient.  The average retention time of 
SN38 was found to be 16.200 ? 0.014 min. A characteristic elution profile of SN38 is 
shown in Figure A2.1. Note that negative peaks were observed around 4-5 minutes due 
to DMSO present in the injection bufer. 
 
172 
 
 
Table A2.1. HPLC Gradient Method for SN38 Detection 
Time 
(min) 
% A 
(Water, 0.1% TFA) 
% B 
(Methanol) 
Total Flow Rate 
(ml/min) 
0 (start) 90 10 1 
14 20 80 1 
16 0 100 1 
16.1 90 10 1 
20 (stop) 90 10 1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
173 
 
 
 
Figure A2.1. Typical HPLC Elution Profile of SN38. SN38 retention time was found to 
be 16.195 for this sample. Smal negative peaks are typicaly sen around 4-5 minutes 
when DMSO is present in the sample injection buffer and are not due to SN38. 
 
174 
 
 
In order to quantify SN38 for release studies, a standard curve was developed for 
concentrations of SN38 from 0.05 ?g/ml to 12 ?g/ml, corresponding to 0.5%-100% of 
drug content in G3.5-SN38 conjugates. Standard samples were prepared at 0.05, 0.1, 0.5, 
1, 2, 4 and 12 ?g/ml with more focus on the lower concentrations as most of the release 
samples were found to occur in this concentration range. SN38 was prepared in a 
concentrated DMSO stock at 100 ?g/ml and used to make the direct injection standards 
in 50/50 DMSO/ 0.1 N HCl injection buffer. Standards were measured in triplicate and 
the concentration versus peak area was used to develop a linear corelation. Figure A2.2 
shows the elution profiles of the 7 standard samples and Figure A2.3 shows the standard 
curve based on these direct injection standards. Finaly, Figure A2.4 displays a single 
standard run in triplicate to ilustrate the tight precision of the HPLC quantification. 
 
175 
 
 
 
Figure A2.2. Standard Curve Elution Profiles. Samples are prepared at SN38 
concentrations of 0.05, 0.1, 0.5, 1, 2, 4 and 12 ?g/ml. 
 
176 
 
 
 
Figure A2.3. HPLC Standard Curve Comparing SN38 Concentration and Peak Area. 
Samples are prepared at SN38 concentrations of 0.05, 0.1, 0.5, 1, 2, 4 and 12 ?g/ml and 
analyzed in triplicate. R
2
 of 0.9995 indicates a strong linear correlation.
 
177 
 
 
 
Figure A2.4. Precision of HPLC Detection. The HPLC chromatogram displays thre 
separate injections of 12 ?g/ml SN38. Peak areas were 444,763, 436,916 and 435,529 
AU-min, yielding a standard deviation of 4,979 or 1% of the average value.  Retention 
times for the thre peaks were 16.199, 16.201, 16.212, corresponding to an average value 
of 16.204 and a standard deviation of 0.0007 or 0.04% of the average.  
 
178 
 
 
 
Appendix 3: Quantification of PEG Content of PAMAM G3.5 and G4.5-
PEG Conjugates by Proton Nuclear Magnetic Resonance 
 
 
Nuclear Magnetic Resonance (NMR) has been applied to the characterization of 
dendrimers [201]. In particular, it is useful for monitoring conjugation reactions when 
the proton signals of the conjugated ligand are distinct from the proton signals of the 
dendrimer [149, 187]. As described in Chapter 5, 
1
H NMR was used to quantify the 
number of PEG chains conjugated to G3.5 and G4.5 dendrimers by comparison of 
integration areas of dendrimer peaks (2.0-3.6 ppm) and PEG peaks (3.6-3.7 ppm). The 
general relationship comparing measured dendrimer and PEG areas with known numbers 
of protons is described by equation A3.1. 
Dendrim
PEG
Dendrim
PEG
HArea
!
=
#
 (Eqn. A3.1) 
Solving for #PEG we obtain equation A3.2. 
PEG
Dendrim
Dendrim
PEG
Area
!=#
 (Eqn A3.2) 
Substituting known proton content for PEG750 (66), G3.5 (740) and G4.5 (1508) we 
derive equations A3.3 and A3.4 which can be used to calculate the number or PEG chains 
conjugated to G3.5 dendrimer and G4.5 dendrimers, respectively.   
 
66
740
.3/#
5.3
!=
PE
Area
PE
 (Eqn. A3.3)
 
 
179 
 
 
66
1508
5.4/#
5.4
!=
G
PE
Area
PE
 (Eqn. A3.4) 
 Table A3.1 lists the dendrimer areas and PEG areas for each dendrimer-PEG 
conjugate along with the calculated number of PEGs per dendrimer. Figures A3.1 and 
A3.2 show the NMR scans of G3.5-1P, G3.5-1.4P, G3.5-2.3P, G4.5-.85P, G4.5-1.9P and 
G4.5-2.8P with the integration areas analyzed by Spinworks Software. These spectra 
ilustrate that an incremental increase in the stoichiometric ratio of PEG used leads to an 
increase in the number of PEG chains conjugated to G3.5 and G4.5 dendrimers. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
180 
 
 
Table A3.1. Quantification of PEG Conjugation to G3.5 and G4.5 Dendrimers by 
1
H 
NMR 
Dendrimer 
Sample 
Stoichiometric 
PEG Equivalents 
PEG 
Area 
Dendrimer 
Area 
# PEG 
Conjugated 
G3.5-P1 1 1.97 21.83 1.01 
G3.5-P1.4 2 3.01 24.71 1.36 
G3.5-P4.5 4 5.48 26.48 2.32 
G4.5-P.85 1 0.49 12.99 0.85 
G4.5-P1.9 2 0.85 10.46 1.86 
G4.5-P2.8 4 3.11 25.39 2.80 
 
 
 
181 
 
 
 
A) 
 
B) 
 
C) 
 
Figure A3.1. NMR Spectra of G3.5-PEG Conjugates. Fed molar ratios are 1 (A), 2 (B) 
and 4 (C) PEGs/ dendrimer. Dendrimer signals occur betwen 2.0 and 3.6 ppm and PEG 
signals occur betwen 3.6 and 3.7 ppm. TSP-d
4 
standard is at 0 ppm. 
 
182 
 
 
(A) 
 
 
(B) 
 
 
(C) 
 
Figure A3.2. NMR spectra of G4.5-PEG Conjugates. Fed molar ratios are 1 (A), 2 (B) 
and 4 (C) PEGs/ dendrimer. Dendrimer signals occur betwen 2.0 and 3.6 ppm and PEG 
signals occur betwen 3.6 and 3.7 ppm. TSP-d
4 
standard is at 0 ppm. 
 
183 
 
 
Bibliography 
 
[1] M.J. Vicent, R. Duncan, Polymer conjugates: nanosized medicines for treating 
cancer. Trends Biotechnol. 24(1) (2006) 39-47. 
[2] V. Torchilin, Multifunctional and stimuli-sensitive pharmaceutical nanocariers. 
Eur. J. Pharm. Biopharm. 71(3) (2009) 431-444. 
[3] P. Debbage, Targeted drugs and nanomedicine: present and future. Curr. Pharm. 
Des. 15(2) (2009) 153-172. 
[4] H. Maeda, J. Wu, T. Sawa, Y. Matsumura, K. Hori, Tumor vascular permeability 
and the EPR efect in macromolecular therapeutics: a review. J. Control. Release 65(1-2) 
(2000) 271-284. 
[5] R. Duncan, Designing polymer conjugates as lysosomotropic nanomedicines. 
Biochem. Soc. Trans. 35(Pt 1) (2007) 56-60. 
[6] R. Duncan, The dawning era of polymer therapeutics. Nat. Rev. Drug. Discov. 
2(5) (2003) 347-360. 
[7] M. Findlay, G. von Minckwitz, A. Wardley, Efective oral chemotherapy for 
breast cancer: pilars of strength. Ann. Oncol. 19(2) (2008) 212-222. 
[8] K. Le Lay, E. Myon, S. Hil, L. Riou-Franca, D. Scott, M. Sidhu, D. Dunlop, R. 
Launois, Comparative cost-minimisation of oral and intravenous chemotherapy for first-
line treatment of non-smal cel lung cancer in the UK NHS system. Eur. J. Health Econ. 
8(2) (2007) 145-151. 
[9] S. Pashko, D.H. Johnson, Potential cost savings of oral versus intravenous 
etoposide in the treatment of smal cel lung cancer. Pharmacoeconomics 1(4) (1992) 
293-297. 
 
184 
 
 
[10] M.D. Donovan, G.L. Flynn, G.L. Amidon, Absorption of polyethylene glycols 
600 through 2000: the molecular weight dependence of gastrointestinal and nasal 
absorption. Pharm. Res. 7(8) (1990) 863-868. 
[11] G. Camenisch, J. Alsenz, H. van de Waterbeemd, G. Folkers, Estimation of 
permeability by pasive difusion through Caco-2 cel monolayers using the drugs' 
lipophilicity and molecular weight. Eur. J. Pharm. Sci. 6(4) (1998) 317-324. 
[12] D.A. Tomalia, L.A. Reyna, S. Svenson, Dendrimers as multi-purpose nanodevices 
for oncology drug delivery and diagnostic imaging. Biochem. Soc. Trans. 35(Pt 1) (2007) 
61-67. 
[13] S. Svenson, D.A. Tomalia, Dendrimers in biomedical applications-reflections on 
the field. Adv. Drug Delivery Rev. 57(15) (2005) 2106-2129. 
[14] C.. Le, J.A. MacKay, J.M. Fr?chet, F.C. Szoka, Designing dendrimers for 
biological applications. Nat. Biotechnol. 23(12) (2005) 1517-1526. 
[15] K.M. Kitchens, R.B. Kolhatkar, P.W. Swan, N.D. Eddington, H. Ghandehari, 
Transport of poly (amido amine) dendrimers across Caco-2 cel monolayers: Influence of 
size, charge and fluorescent labeling. Pharm. Res. 23(12) (2006) 2818-2826. 
[16] K.M. Kitchens, M.E. El-Sayed, H. Ghandehari, Transepithelial and endothelial 
transport of poly (amido amine) dendrimers. Adv. Drug Delivery Rev. 57(15) (2005) 
2163-2176. 
[17] R. Jevprasesphant, J. Penny, D. Atwood, N.B. McKeown, A. D'Emanuele, 
Engineering of dendrimer surfaces to enhance transepithelial transport and reduce 
cytotoxicity. Pharm. Res. 20(10) (2003) 1543-1550. 
 
185 
 
 
[18] K. Kitchens, A. Foraker, R. Kolhatkar, P. Swan, H. Ghandehari, Endocytosis and 
interaction of poly (amido amine) dendrimers with Caco-2 cels. Pharm. Res. 24(11) 
(2007) 2138-2145. 
[19] R. Jevprasesphant, J. Penny, D. Atwood, A. D'Emanuele, Transport of dendrimer 
nanocariers through epithelial cels via the transcelular route. J. Control. Release 97 
(2004) 259 ? 267. 
[20] M. Najlah, S. Freman, D. Atwood, A. D'Emanuele, In vitro evaluation of 
dendrimer prodrugs for oral drug delivery. Int. J. Pharm. 336(1) (2007) 183-190. 
[21] H. Ulukan, P.W. Swan, Camptothecins: a review of their chemotherapeutic 
potential. Drugs 62(14) (2002) 2039-2057. 
[22] American Cancer Society. Cancer Facts & Figures 2008. Atlanta: American 
Cancer Society; 2008. 
[23] H. Shimada, K. Tanaka, I. Endou, Y. Ichikawa, Treatment for colorectal liver 
metastases: a review. Langenbecks Arch. Surg. 394(6) (2009) 973-983. 
[24] R. Kolhatkar, D. Swet, H. Ghandehari, in: V. Torchilin (Ed.), Multifunctional 
Pharmaceutical Nanocariers, Springer, New York, 2008, pp. 201-232. 
[25] D.S. Goldberg, H. Ghandehari, P.W. Swan, Celular entry of G3.5 poly (amido 
amine) dendrimers by clathrin- and dynamin-dependent endocytosis promotes tight 
junctional opening in intestinal epithelia. Pharm. Res. 27(8) (2010) 1547-1557. 
[26] D.S. Goldberg, N. Vijayalakshmi, P.W. Swan, H. Ghandehari, G3.5 PAMAM 
dendrimers enhance transepithelial transport of SN38 while minimizing gastrointestinal 
toxicity. J. Control. Release (Submited). 
 
186 
 
 
[27] D. Swet, R. Kolhatkar, A. Ray, P. Swan, H. Ghandehari, Transepithelial 
transport of PEGylated anionic poly (amido amine) dendrimers: implications for oral 
drug delivery. J. Control. Release 138(1) (2009) 78-85. 
[28] A.M. Hilery, A.W. Lloyd, J. Swarbrick, Drug Delivery and Targeting for 
Pharmacists and Pharmaceutical Scientists, Taylor Francis, London, 2001. 
[29] S. Svenson, Carier-based Drug Delivery, Oxford University Pres, 2004. 
[30] V. Torchilin, Nanoparticulates as Drug Cariers, Imperial College Pres, London, 
2006. 
[31] G.S. Kwon, Polymeric Drug Delivery Systems, Taylor & Francis Group, Boca 
Raton, 2005. 
[32] T.A. Oswald, G. Menges, Materials Science of Polymers for Engineers, Carl 
Hanser, Munich, 2003. 
[33] H. Ringsdorf, Structure and properties of pharmacologicaly active polymers. J. 
Polym. Sci. Symp. 51 (1975) 135-153. 
[34] J.L. Richardson, G. Marks, A. Levine, The influence of symptoms of disease and 
side efects of treatment on compliance with cancer therapy. J. Clin. Oncol. 6(11) (1988) 
1746-1752. 
[35] R. Duncan, Polymer conjugates as anticancer nanomedicines. Nat. Rev. Cancer 
6(9) (2006) 688-701. 
[36] H. Maeda, G.Y. Bharate, J. Daruwala, Polymeric drugs for eficient tumor-
targeted drug delivery based on EPR-efect. Eur. J. Pharm. Biopharm. 71(3) (2009) 409-
419. 
 
187 
 
 
[37] T. Lamers, W.E. Hennink, G. Storm, Tumour-targeted nanomedicines: 
principles and practice. Br. J. Cancer 99(3) (2008) 392-397. 
[38] J. Kopecek, P. Kopeckova, HPMA copolymers: origins, early developments, 
present, and future. Adv. Drug Delivery Rev. 62(2) (2010) 122-149. 
[39] S.Q. Gao, Z.R. Lu, B. Petri, P. Kopeckova, J. Kopecek, Colon-specific 9-
aminocamptothecin-HPMA copolymer conjugates containing a 1,6-elimination spacer. J. 
Control. Release 110(2) (2006) 323-331. 
[40] R. Duncan, M.J. Vicent, Do HPMA copolymer conjugates have a future as 
clinicaly useful nanomedicines? A critical overview of current status and future 
opportunities. Adv. Drug Delivery Rev. 62(2) (2010) 272-282. 
[41] L.W. Seymour, D.R. Fery, D.J. Ker, D. Rea, M. Whitlock, R. Poyner, C. Boivin, 
S. Heslewood, C. Twelves, R. Blackie, A. Schatzlein, D. Jodrel, D. Biset, H. Calvert, 
M. Lind, A. Robbins, S. Burtles, R. Duncan, J. Casidy, Phase I studies of polymer-
doxorubicin (PK1, FCE28068) in the treatment of breast, lung and colorectal cancer. Int. 
J. Oncol. 34(6) (2009) 1629-1636. 
[42] L.W. Seymour, D.R. Fery, D. Anderson, S. Heslewood, P.J. Julyan, R. Poyner, 
J. Doran, A.M. Young, S. Burtles, D.J. Ker, Hepatic drug targeting: phase I evaluation of 
polymer-bound doxorubicin. J. Clin. Oncol. 20(6) (2002) 1668-1676. 
[43] D.P. Nowotnik, E. Cvitkovic, ProLindac (AP5346): a review of the development 
of an HPMA DACH platinum polymer therapeutic. Adv. Drug Delivery Rev. 61(13) 
(2009) 1214-1219. 
[44] G. Pasut, F.M. Veronese, PEG conjugates in clinical development or use as 
anticancer agents: an overview. Adv. Drug Delivery Rev. 61(13) (2009) 1177-1188. 
 
188 
 
 
[45] C. Li, S. Walace, Polymer-drug conjugates: recent development in clinical 
oncology. Adv. Drug Delivery Rev. 60(8) (2008) 886-898. 
[46] C. Li, D.F. Yu, R.A. Newman, F. Cabral, L.C. Stephens, N. Hunter, L. Milas, S. 
Walace, Complete regresion of wel-established tumors using a novel water-soluble 
poly(L-glutamic acid)-paclitaxel conjugate. Cancer Res. 58(11) (1998) 2404-2409. 
[47] G. Mustata, S.M. Dinh, Approaches to oral drug delivery for chalenging 
molecules. Crit. Rev. Ther. Drug Carier Syst. 23(2) (2006) 111-135. 
[48] M.D. DeMario, M.J. Ratain, Oral chemotherapy: rationale and future directions. J. 
Clin. Oncol. 16(7) (1998) 2557-2567. 
[49] F. Gabor, C. Filafer, L. Neutsch, G. Ratzinger, M. Wirth, Improving oral 
delivery. Handb. Exp. Pharmacol.(197) (2010) 345-398. 
[50] S.K. Lai, Y.. Wang, J. Hanes, Mucus-penetrating nanoparticles for drug and 
gene delivery to mucosal tisues. Adv. Drug Delivery Rev. 61(2) (2009) 158-171. 
[51] L.R. Johson, Physiology of the Gastrointestinal Tract, Raven Pres, New York, 
1994. 
[52] A.M. Marchiando, W.V. Graham, J.R. Turner, Epithelial bariers in homeostasis 
and disease. Annu. Rev. Pathol. 5 (2010) 119-144. 
[53] J.R. Turner, Intestinal mucosal barier function in health and disease. Nat. Rev. 
Imunol. 9(11) (2009) 799-809. 
[54] S.C. Corr, C.. Gahan, C. Hil, M-cels: origin, morphology and role in mucosal 
imunity and microbial pathogenesis. FEMS Imunol. Med. Microbiol. 52(1) (2008) 2-
12. 
 
189 
 
 
[55] M. Kondoh, K. Yagi, Tight junction modulators: promising candidates for drug 
delivery. Curr. Med. Chem. 14(23) (2007) 2482-2488. 
[56] M. Furuse, T. Hirase, M. Itoh, A. Nagafuchi, S. Yonemura, S. Tsukita, Ocludin: 
a novel integral membrane protein localizing at tight junctions. J. Cel Biol. 123(6 Pt 2) 
(1993) 1777-1788. 
[57] M.S. Balda, K. Mater, Transmembrane proteins of tight junctions. Semin. Cel 
Dev. Biol. 11(4) (2000) 281-289. 
[58] M. Furuse, H. Sasaki, K. Fujimoto, S. Tsukita, A single gene product, claudin-1 
or -2, reconstitutes tight junction strands and recruits occludin in fibroblasts. J. Cel Biol. 
143(2) (1998) 391-401. 
[59] M. Furuse, K. Fujita, T. Hiragi, K. Fujimoto, S. Tsukita, Claudin-1 and -2: novel 
integral membrane proteins localizing at tight junctions with no sequence similarity to 
occludin. J. Cel Biol. 141(7) (1998) 1539-1550. 
[60] I. Martin-Padura, S. Lostaglio, M. Schneemann, L. Wiliams, M. Romano, P. 
Fruscela, C. Panzeri, A. Stoppaciaro, L. Ruco, A. Vila, D. Simons, E. Dejana, 
Junctional adhesion molecule, a novel member of the imunoglobulin superfamily that 
distributes at intercelular junctions and modulates monocyte transmigration. J. Cel Biol. 
142(1) (1998) 117-127. 
[61] J. Ikenouchi, M. Furuse, K. Furuse, H. Sasaki, S. Tsukita, Tricelulin constitutes a 
novel barier at tricelular contacts of epithelial cels. J. Cel Biol. 171(6) (2005) 939-945. 
[62] M.A. Deli, Potential use of tight junction modulators to reversibly open 
membranous bariers and improve drug delivery. Biochim. Biophys. Acta. 1788(4) 
(2009) 892-910. 
 
190 
 
 
[63] J.L. Madara, J.R. Pappenheimer, Structural basis for physiological regulation of 
paracelular pathways in intestinal epithelia. J. Membr. Biol. 100(2) (1987) 149-164. 
[64] T.Y. Ma, D. Tran, N. Hoa, D. Nguyen, M. Meryfield, A. Tarnawski, Mechanism 
of extracelular calcium regulation of intestinal epithelial tight junction permeability: role 
of cytoskeletal involvement. Microsc. Res. Tech. 51(2) (2000) 156-168. 
[65] C.J. Watson, M. Rowland, G. Warhurst, Functional modeling of tight junctions in 
intestinal cel monolayers using polyethylene glycol oligomers. Am. J. Physiol. Cel 
Physiol. 281(2) (2001) C388-397. 
[66] G.T. Knipp, N.F. Ho, C.L. Barsuhn, R.T. Borchardt, Paracelular difusion in 
Caco-2 cel monolayers: efect of perturbation on the transport of hydrophilic compounds 
that vary in charge and size. J. Pharm. Sci. 86(10) (1997) 1105-1110. 
[67] S.J. Fisher, P.W. Swan, N.D. Eddington, The ethanol metabolite acetaldehyde 
increases paracelular drug permeability in vitro and oral bioavailability in vivo. J. 
Pharmacol. Exp. Ther. 332(1) (2010) 326-333. 
[68] H. Inokuchi, T. Takei, K. Aikawa, M. Shimizu, The efect of hyperosmosis on 
paracelular permeability in Caco-2 cel monolayers. Biosci. Biotechnol. Biochem. 73(2) 
(2009) 328-334. 
[69] S.M. Krug, S. Amasheh, J.F. Richter, S. Milatz, D. G?nzel, J.K. Westphal, O. 
Huber, J.D. Schulzke, M. Fromm, Tricelulin forms a barier to macromolecules in 
tricelular tight junctions without afecting ion permeability. Mol. Biol. Cel 20(16) 
(2009) 3713-3724. 
[70] V.H.L. Le, Membrane Transporters. Eur. J. Pharm. Sci. 11(Suppl 2) (2000) S41-
S50. 
 
191 
 
 
[71] P.W. Swan, Recent advances in intestinal macromolecular drug delivery via 
receptor-mediated transport pathways. Pharm. Res. 15(6) (1998) 826-834. 
[72] L. Bareford, P. Swan, Endocytic mechanisms for targeted drug delivery. Adv. 
Drug Delivery Rev. 59 (2007) 748-758. 
[73] P. Lajoie, I.R. Nabi, Lipid rafts, caveolae, and their endocytosis. Int. Rev. Cel 
Mol. Biol. 282 (2010) 135-163. 
[74] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feney, Experimental and 
computational approaches to estimate solubility and permeability in drug discovery and 
development setings. Adv. Drug Delivery Rev. 46(1-3) (2001) 3-26. 
[75] K. Beaumont, R. Webster, I. Gardner, K. Dack, Design of ester prodrugs to 
enhance oral absorption of poorly permeable compounds: chalenges to the discovery 
scientist. Curr. Drug Metab. 4(6) (2003) 461-485. 
[76] G.L. Amidon, H. Lenernas, V.P. Shah, J.R. Crison, A theoretical basis for a 
biopharmaceutic drug clasification: the correlation of in vitro drug product disolution 
and in vivo bioavailability. Pharm. Res. 12(3) (1995) 413-420. 
[77] A. Dahan, J.M. Miler, G.L. Amidon, Prediction of solubility and permeability 
clas membership: provisional BCS clasification of the world's top oral drugs. The 
APS Journal 11(4) (2009) 740-746. 
[78] T. Hou, Y. Li, W. Zhang, J. Wang, Recent developments of in silico predictions 
of intestinal absorption and oral bioavailability. Comb. Chem. High Throughput Scren. 
12(5) (2009) 497-506. 
 
192 
 
 
[79] T. Hou, J. Wang, W. Zhang, X. Xu, ADME evaluation in drug discovery. 7. 
Prediction of oral absorption by correlation and clasification. J. Chem. Inf. Model. 47(1) 
(2007) 208-218. 
[80] T. Hou, J. Wang, W. Zhang, X. Xu, ADME evaluation in drug discovery. 6. Can 
oral bioavailability in humans be efectively predicted by simple molecular property-
based rules? J. Chem. Inf. Model. 47(2) (2007) 460-463. 
[81] M. Kansy, F. Senner, K. Gubernator, Physicochemical high throughput screning: 
paralel artificial membrane permeation asay in the description of pasive absorption 
proceses. J. Med. Chem. 41(7) (1998) 1007-1010. 
[82] B. Faler, Artificial membrane asays to ases permeability. Curr. Drug Metab. 
9(9) (2008) 886-892. 
[83] I.J. Hidalgo, T.J. Raub, R.T. Borchardt, Characterization of the human colon 
carcinoma cel line (Caco-2) as a model system for intestinal epithelial permeability. 
Gastroenterology 96(3) (1989) 736-749. 
[84] S. Skolnik, X. Lin, J. Wang, X.H. Chen, T. He, B. Zhang, Towards prediction of 
in vivo intestinal absorption using a 96-wel Caco-2 asay. J. Pharm. Sci. 99(7) (2010) 
3246-3265. 
[85] V. Meunier, M. Bourrie, Y. Berger, G. Fabre, The human intestinal epithelial cel 
line Caco-2; pharmacological and pharmacokinetic applications. Cel Biol. Toxicol. 11(3-
4) (1995) 187-194. 
[86] D.A. Volpe, Variability in Caco-2 and MDCK cel-based intestinal permeability 
asays. J. Pharm. Sci. 97(2) (2008) 712-725. 
 
193 
 
 
[87] H. Bohets, P. Annaert, G. Mannens, L. Van Beijsterveldt, K. Anciaux, P. 
Verboven, W. Meuldermans, K. Lavrijsen, Strategies for absorption screning in drug 
discovery and development. Curr. Top. Med. Chem. 1(5) (2001) 367-383. 
[88] S. Ye, In vitro permeability across Caco-2 cels (colonic) can predict in vivo 
(smal intestinal) absorption in man-fact or myth. Pharm. Res. 14(6) (1997) 763-766. 
[89] H. Yu, T.J. Cook, P.J. Sinko, Evidence for diminished functional expresion of 
intestinal transporters in Caco-2 cel monolayers at high pasages. Pharm. Res. 14(6) 
(1997) 757-762. 
[90] S. Yamashita, K. Konishi, Y. Yamazaki, Y. Taki, New and beter protocols for a 
short-term Caco-2 cel culture system. J. Pharm. Sci. 91(3) (2002) 669-679. 
[91] J.D. Irvine, L. Takahashi, K. Lockhart, J. Cheong, J.W. Tolan, H.E. Selick, J.R. 
Grove, MDCK (Madin-Darby Canine Kidney) cels: A tool for membrane permeability 
screning. J. Pharm. Sci. 88(1) (1999) 28-33. 
[92] M.C. Gres, B. Julian, M. Bourrie, V. Meunier, C. Roques, M. Berger, X. Boulenc, 
Y. Berger, G. Fabre, Correlation betwen oral drug absorption in humans, and apparent 
drug permeability in TC-7 cels, a human epithelial intestinal cel line: comparison with 
the parental Caco-2 cel line. Pharm. Res. 15(5) (1998) 726-733. 
[93] C. Hilgendorf, H. Spahn-Langguth, C.G. Regardh, E. Lipka, G.L. Amidon, P. 
Langguth, Caco-2 versus Caco-2/HT29-MTX co-cultured cel lines: permeabilities via 
difusion, inside- and outside-directed carier-mediated transport. J. Pharm. Sci. 89(1) 
(2000) 63-75. 
 
194 
 
 
[94] T.H. Wilson, G. Wiseman, The use of sacs of everted smal intestine for the study 
of the transference of substances from the mucosal to the serosal surface. J. Physiol. 
123(1) (1954) 116-125. 
[95] H. Levi, H.. Using, Resting potential and ion movements in the frog skin. 
Nature 164(4178) (1949) 928-929. 
[96] G.M. Gras, S.A. Swetana, In vitro measurement of gastrointestinal tisue 
permeability using a new difusion cel. Pharm. Res. 5(6) (1988) 372-376. 
[97] V.J. Stela, in: V. J. Stela (Ed.), Prodrugs: Chalenges and Rewards, Springer, 
2007, pp. 3-36. 
[98] T. Heimbach, D. Fleisher, A. Kaddoumi, in: V. J. Stela (Ed.), Prodrugs: 
Chalenges and Rewards, Springer, 2007, pp. 157-215. 
[99] A. Mantyla, J. Rautio, T. Nevalainen, P. Keski-Rahkonen, J. Vepsalainen, T. 
Jarvinen, Design, synthesis and in vitro evaluation of novel water-soluble prodrugs of 
buparvaquone. Eur. J. Pharm. Sci. 23(2) (2004) 151-158. 
[100] V.J. Stela, K.W. Nti-Addae, Prodrug strategies to overcome poor water 
solubility. Adv. Drug Delivery Rev. 59(7) (2007) 677-694. 
[101] V.J. Stela, in: V. J. Stela (Ed.), Prodrugs: Chalenges and Rewards, Springer, 
2007, pp. 37-82. 
[102] I.E. Kuppens, P. Bredveld, J.H. Beijnen, J.H. Schelens, Modulation of oral drug 
bioavailability: from preclinical mechanism to therapeutic application. Cancer Invest. 
23(5) (2005) 443-464. 
 
195 
 
 
[103] C.M. Kruijtzer, J.H. Beijnen, J.H. Schelens, Improvement of oral drug treatment 
by temporary inhibition of drug transporters and/or cytochrome P450 in the 
gastrointestinal tract and liver: an overview. Oncologist 7(6) (2002) 516-530. 
[104] D.F. Kehrer, R.H. Mathijsen, J. Verweij, P. de Bruijn, A. Spareboom, 
Modulation of irinotecan metabolism by ketoconazole. J. Clin. Oncol. 20(14) (2002) 
3122-3129. 
[105] D.J. Kempf, K.C. Marsh, G. Kumar, A.D. Rodrigues, J.F. Denisen, E. 
McDonald, M.J. Kukulka, A. Hsu, G.R. Granneman, P.A. Baroldi, E. Sun, D. Pizuti, J.J. 
Platner, D.W. Norbeck, J.M. Leonard, Pharmacokinetic enhancement of inhibitors of the 
human imunodeficiency virus protease by coadministration with ritonavir. Antimicrob. 
Agents Chemother. 41(3) (1997) 654-660. 
[106] M. Tomita, M. Hayashi, S. Awazu, Absorption-enhancing mechanism of EDTA, 
caprate, and decanoylcarnitine in Caco-2 cels. J. Pharm. Sci. 85(6) (1996) 608-611. 
[107] M.J. Cano-Cebrian, T. Zornoza, L. Granero, A. Polache, Intestinal absorption 
enhancement via the paracelular route by faty acids, chitosans and others: a target for 
drug delivery. Curr. Drug Deliv. 2(1) (2005) 9-22. 
[108] A. Bernkop-Schnurch, in: A. Bernkop-Schnurch (Ed.), Oral Delivery of 
Macromolecular Drugs: Bariers, Strategies and Future Trends, Springer, New York, 
2009, pp. 85-101. 
[109] A. Fasano, S. Uzau, Modulation of intestinal tight junctions by Zonula occludens 
toxin permits enteral administration of insulin and other macromolecules in an animal 
model. J. Clin. Invest. 99(6) (1997) 1158-1164. 
 
196 
 
 
[110] H.E. Junginger, in: A. Bernkop-Schunurch (Ed.), Oral Delivery of 
Macromolecular Drugs: Bariers, Strategies and Future Trends, Springer, New York, 
2009, pp. 103-122. 
[111] R.C. Lindenschmidt, L.C. Stone, J.L. Seymour, R.L. Anderson, P.A. Forshey, 
M.J. Winrow, Efects of oral administration of a high-molecular-weight crosslinked 
polyacrylate in rats. Fundam. Appl. Toxicol. 17(1) (1991) 128-135. 
[112] N.G. Schipper, K.M. Varum, P. Arturson, Chitosans as absorption enhancers for 
poorly absorbable drugs. 1: Influence of molecular weight and degre of acetylation on 
drug transport across human intestinal epithelial (Caco-2) cels. Pharm. Res. 13(11) 
(1996) 1686-1692. 
[113] A. D'Emanuele, R. Jevprasesphant, J. Penny, D. Atwood, The use of a 
dendrimer-propranolol prodrug to bypas eflux transporters and enhance oral 
bioavailability. J. Control. Release 95(3) (2004) 447-453. 
[114] D.A. Tomalia, Birth of a new macromolecular architecture: dendrimers as 
quantized building blocks for nanoscale synthetic polymer chemistry. Prog. Polym. Sci. 
30 (2005) 294-324. 
[115] D.A. Tomalia, J.M. Frechet, Dendrimers and Other Dendritic Polymers, John 
Wiley & Sons, West Sussex, 2001. 
[116] T.D. McCarthy, P. Karelas, S.A. Henderson, M. Giannis, D.F. O'Kefe, G. 
Hery, J.R. Paull, B.R. Mathews, G. Holan, Dendrimers as drugs: discovery and 
preclinical and clinical development of dendrimer-based microbicides for HIV and STI 
prevention. Mol. Pharm. 2(4) (2005) 312-318. 
 
197 
 
 
[117] H.L. Crampton, E.E. Simanek, Dendrimers as drug delivery vehicles: non-
covalent interactions of bioactive compounds with dendrimers. Polym. Int. 56 (2007) 
489-496. 
[118] P. Couck, R. Claeys, E. Vanderstraeten, F.K. Gorus, Evaluation of the Stratus CS 
fluorometer for the determination of plasma myoglobin. Acta Clin. Belg. 60(2) (2005) 
75-78. 
[119] A. D'Emanuele, D. Atwood, Dendrimer-drug interactions. Adv. Drug Delivery 
Rev. 57(15) (2005) 2147-2162. 
[120] M. Na, C. Yiyun, X. Tongwen, D. Yang, W. Xiaomin, L. Zhenwei, C. Zhichao, 
H. Guanyi, S. Yunyu, W. Longping, Dendrimers as potential drug cariers. Part I. 
Prolonged delivery of ketoprofen by in vitro and in vivo studies. Eur. J. Med. Chem. 
41(5) (2006) 670-674. 
[121] U. Gupta, H.B. Agashe, A. Asthana, N.K. Jain, A review of in vitro-in vivo 
investigations on dendrimers: the novel nanoscopic drug cariers. Nanomedicine 2(2) 
(2006) 66-73. 
[122] R. Wiwatanapatapee, L. Lomlim, K. Saramunee, Dendrimers conjugates for 
colonic delivery of 5-aminosalicylic acid. J. Control. Release 88(1) (2003) 1-9. 
[123] C. Yiyun, X. Tongwen, Dendrimers as potential drug cariers. Part I. 
Solubilization of non-steroidal anti-inflamatory drugs in the presence of 
polyamidoamine dendrimers. Eur. J. Med. Chem. 40(11) (2005) 1188-1192. 
[124] A.S. Chauhan, N.K. Jain, P.V. Diwan, A.J. Khopade, Solubility enhancement of 
indomethacin with poly(amidoamine) dendrimers and targeting to inflamatory regions 
of arthritic rats. J. Drug Target. 12(9-10) (2004) 575-583. 
 
198 
 
 
[125] A. Quintana, E. Raczka, L. Piehler, I. Le, A. Myc, I. Majoros, A.K. Patri, T. 
Thomas, J. Mule, J.R. Baker, Jr., Design and function of a dendrimer-based therapeutic 
nanodevice targeted to tumor cels through the folate receptor. Pharm. Res. 19(9) (2002) 
1310-1316. 
[126] G. Wu, R.F. Barth, W. Yang, S. Kawabata, L. Zhang, K. Gren-Church, Targeted 
delivery of methotrexate to epidermal growth factor receptor-positive brain tumors by 
means of cetuximab (IMC-C225) dendrimer bioconjugates. Mol. Cancer Ther. 5(1) 
(2006) 52-59. 
[127] S.D. Konda, M. Aref, S. Wang, M. Brechbiel, E.C. Wiener, Specific targeting of 
folate-dendrimer MRI contrast agents to the high afinity folate receptor expresed in 
ovarian tumor xenografts. Magn. Reson. Mater. Phys., Biol. Med 12(2-3) (2001) 104-
113. 
[128] A. Florence, Dendrimers: a versatile targeting platform. Adv. Drug Delivery Rev. 
57(15) (2005) 2101-2286. 
[129] Y. Koyama, V.S. Talanov, M. Bernardo, Y. Hama, C.A. Regino, M.W. Brechbiel, 
P.L. Choyke, H. Kobayashi, A dendrimer-based nanosized contrast agent dual-labeled for 
magnetic resonance and optical fluorescence imaging to localize the sentinel lymph node 
in mice. J. Magn. Reson. Imaging 25(4) (2007) 866-871. 
[130] H. Kobayashi, S. Kawamoto, R.A. Star, T.A. Waldmann, Y. Tagaya, M.W. 
Brechbiel, Micro-magnetic resonance lymphangiography in mice using a novel 
dendrimer-based magnetic resonance imaging contrast agent. Cancer Res. 63(2) (2003) 
271-276. 
 
199 
 
 
[131] S. Langereis, Q.G. de Lussanet, M.H. van Genderen, E.W. Meijer, R.G. Bets-
Tan, A.W. Grifioen, J.M. van Engelshoven, W.H. Backes, Evaluation of Gd(II)DTPA-
terminated poly(propylene imine) dendrimers as contrast agents for MR imaging. NMR 
Biomed. 19(1) (2006) 133-141. 
[132] Y. Choi, T. Thomas, A. Kotlyar, M.T. Islam, J.R. Baker, Synthesis and functional 
evaluation of DNA-asembled polyamidoamine dendrimer clusters for cancer cel-
specific targeting. Chem. Biol. 12(1) (2005) 35-43. 
[133] M. Mamede, T. Saga, T. Ishimori, T. Higashi, N. Sato, H. Kobayashi, M.W. 
Brechbiel, J. Konishi, Hepatocyte targeting of 111In-labeled oligo-DNA with avidin or 
avidin-dendrimer complex. J. Control. Release 95(1) (2004) 133-141. 
[134] C.J. Hawker, J.M. Frechet, Preparation of polymers with controlled molecular 
architecture. A new convergent approach to dendritic macromolecules. J. Am. Chem. 
Soc. 112 (1990) 7638-7647. 
[135] F. Vogtle, M. Gorka, R. Hese, P. Ceroni, M. Maestri, V. Balzani, Photochemical 
and photophysical properties of poly(propylene amine) dendrimers with peripheral 
naphthalene and azobenzene groups. Photochem. Photobiol. Sci. 1(1) (2002) 45-51. 
[136] R. Esfand, D.A. Tomalia, Poly (amidoamine)(PAMAM) dendrimers: from 
biomimicry to drug delivery and biomedical applications. Drug Discov. Today 6(8) 
(2001) 427-426. 
[137] V.K. Yelepeddi, A. Kumar, S. Palakurthi, Surface modified poly (amido amine) 
dendrimers as diverse nanomolecules for biomedical applications. Expert Opin. Drug 
Deliv. 6(8) (2009) 835-850. 
 
200 
 
 
[138] V.J. Vendito, C.A. Regino, M.W. Brechbiel, PAMAM dendrimer based 
macromolecules as improved contrast agents. Mol. Pharm. 2(4) (2005) 302-311. 
[139] R. Duncan, L. Izo, Dendrimer biocompatibility and toxicity. Adv. Drug Delivery 
Rev. 57(15) (2005) 2215-2237. 
[140] N. Malik, R. Wiwatanapatapee, R. Klopsch, K. Lorenz, H. Frey, J.W. Wener, 
E.W. Meijer, W. Paulus, R. Duncan, Dendrimers: relationship betwen structure and 
biocompatibility in vitro, and preliminary studies on the biodistribution of 125I-labeled 
polyamidoamine dendrimers in vivo. J. Control. Release 65(1-2) (2000) 133-148. 
[141] N. Malik, E.G. Evagorou, R. Duncan, Dendrimer-platinate: a novel approach to 
cancer chemotherapy. Anticancer Drugs 10(8) (1999) 767-776. 
[142] S. Kannan, P. Kolhe, V. Raykova, M. Glibatec, R.M. Kannan, M. Lieh-Lai, D. 
Baset, Dynamics of celular entry and drug delivery by dendritic polymers into human 
lung epithelial carcinoma cels. J. Biomater. Sci. Polym. Ed. 15(3) (2004) 311-330. 
[143] I.J. Majoros, A. Myc, T. Thomas, C.B. Mehta, J.R. Baker, PAMAM dendrimer-
based multifunctional conjugate for cancer therapy: synthesis, characterization, and 
functionality. Biomacromolecules 7(2) (2006) 572-579. 
[144] R. Wiwatanapatapee, B. Care?o-G?mez, N. Malik, R. Duncan, Anionic 
PAMAM dendrimers rapidly cross adult rat intestine in vitro: a potential oral delivery 
system? Pharm. Res. 17(8) (2000) 991-998. 
[145] M. El-Sayed, M. Ginski, C. Rhodes, H. Ghandehari, Transepithelial transport of 
poly(amido amine) dendrimers across Caco-2 cel monolayers. J. Control. Release 81(3) 
(2002) 355-365. 
 
201 
 
 
[146] M. El-Sayed, M. Ginski, C. Rhodes, H. Ghandehari, Influence of surface 
chemistry of poly (amido amine) dendrimers on Caco-2 cel monolayers. J. Bioact. 
Compat. Polym. 18(1) (2003) 7-22. 
[147] M. El-Sayed, C.A. Rhodes, M. Ginski, H. Ghandehari, Transport mechanism(s) of 
poly (amido amine) dendrimers across Caco-2 cel monolayers. Int. J. Pharm. 265(1-2) 
(2003) 151-157. 
[148] K.M. Kitchens, R.B. Kolhatkar, P.W. Swan, H. Ghandehari, Endocytosis 
inhibitors prevent poly (amido amine) dendrimer internalization and permeability across 
Caco-2 cels. Mol. Pharm. 5(2) (2008) 364-369. 
[149] R.B. Kolhatkar, K.M. Kitchens, P.W. Swan, H. Ghandehari, Surface acetylation 
of poly (amido amine) (PAMAM) dendrimers decreases cytotoxicity while maintaining 
membrane permeability. Bioconjug. Chem. 18(6) (2007) 2054?2060  
[150] R.B. Kolhatkar, P.W. Swan, H. Ghandehari, Potential oral delivery of 7-ethyl-
10-hydroxy-camptothecin (SN-38) using poly (amido amine) dendrimers. Pharm. Res. 
25(7) (2008) 1723-1729. 
[151] W. Ke, Y. Zhao, R. Huang, C. Jiang, Y. Pei, Enhanced oral bioavailability of 
doxorubicin in a dendrimer drug delivery system. J. Pharm. Sci. 97(6) (2008) 2208-2216. 
[152] American Cancer Soceity, Cancer Facts & Figures 2010. Atlanta: American 
Cancer Society; 2010. 
[153] Z.F. Gelad, D. Provenzale, Colorectal cancer: national and international 
perspective on the burden of disease and public health impact. Gastroenterology 138(6) 
2177-2190. 
 
202 
 
 
[154] D. Cunningham, W. Atkin, H.J. Lenz, H.T. Lynch, B. Minsky, B. Nordlinger, N. 
Starling, Colorectal cancer. Lancet 375(9719) 1030-1047. 
[155] C. Kurkjian, S. Kummar, Advances in the treatment of metastatic colorectal 
cancer. Am. J. Ther. 16(5) (2009) 412-420. 
[156] P. Comela, A review of the role of capecitabine in the treatment of colorectal 
cancer. Ther. Clin. Risk Manag. 3(3) (2007) 421-431. 
[157] C.F. Stewart, W.C. Zamboni, W.R. Crom, P.J. Houghton, Disposition of 
irinotecan and SN-38 following oral and intravenous irinotecan dosing in mice. Cancer 
Chemother. Pharmacol. 40(3) (1997) 259-265. 
[158] Y. Kawato, M. Aonuma, Y. Hirota, H. Kuga, K. Sato, Intracelular roles of SN-
38, a metabolite of the camptothecin derivative CPT-11, in the antitumor efect of CPT-
11. Cancer Res. 51(16) (1991) 4187-4191. 
[159] H. Zhao, B. Rubio, P. Sapra, D. Wu, P. Reddy, P. Sai, A. Martinez, Y. Gao, Y. 
Lozanguiez, C. Longley, L.M. Grenberger, I.D. Horak, Novel prodrugs of SN38 using 
multiarm poly(ethylene glycol) linkers. Bioconjug. Chem. 19(4) (2008) 849-859. 
[160] F. Meyer-Losic, C. Nicolazi, J. Quinonero, F. Ribes, M. Michel, V. Dubois, C. 
de Coupade, M. Boukaisi, A.-S. Ch?n?, I. Tranchant, V. Aranz, I. Zoubaa, J.-S. 
Fruchart, D. Ravel, J. Kearsey, DTS-108, a novel peptidic prodrug of SN38: in vivo 
eficacy and toxicokinetic studies. Clin. Cancer Res. 14(7) (2008) 2145-2153. 
[161] N. Vijayalakshmi, A. Ray, A. Malugin, H. Ghandehari, Carboxyl terminated 
PAMAM-SN38 conjugates: synthesis, characterization, and in vitro evaluation. 
Bioconjug. Chem. 21(10) (2010) 1804-1810. 
 
203 
 
 
[162] R. Jevprasesphant, J. Penny, R. Jalal, D. Atwood, N.B. McKeown, A. 
D'Emanuele, The influence of surface modification on the cytotoxicity of PAMAM 
dendrimers. Int. J. Pharm. 252(1-2) (2003) 263-266. 
[163] A. Ivanov, Exocytosis and Endocytosis, Humana Pres, 2008, pp. 15-33. 
[164] E. Macia, M. Ehrlich, R. Masol, E. Boucrot, C. Brunner, T. Kirchhausen, 
Dynasore, a cel-permeable inhibitor of dynamin. Dev. Cel 10(6) (2006) 839-850. 
[165] T. L?hmann, M. Rimann, A.G. Bitermann, H. Hal, Celular uptake and 
intracelular pathways of PL-g-PEG-DNA nanoparticles. Bioconjug. Chem. 19(9) 
(2008) 1907-1916. 
[166] H. Inokuchi, T. Takei, K. Aikawa, M. Shimizu, The efect of hyperosmosis on 
paracelular permeability in Caco-2 cel monolayers. Biosci. Biotechnol. Biochem. 73(2) 
(2009) 328-334. 
[167] Y. Phonphok, K.S. Rosenthal, Stabilization of clathrin coated vesicles by 
amantadine, tromantadine and other hydrophobic amines. FEBS Let. 281(1-2) (1991) 
188-190. 
[168] K. Sato, J. Nagai, N. Mitsui, Y. Ryoko, M. Takano, Efects of endocytosis 
inhibitors on internalization of human IgG by Caco-2 human intestinal epithelial cels. 
Life Sci. 85(23-26) (2009) 800-807. 
[169] A.E. Gibson, R.J. Noel, J.T. Herlihy, W.F. Ward, Phenylarsine oxide inhibition of 
endocytosis: efects on asialofetuin internalization. Am. J. Physiol. Cel Physiol. 257(2) 
(1989) C182-C184. 
 
204 
 
 
[170] Z. Ma, L.-Y. Lim, Uptake of chitosan and asociated insulin in Caco-2 cel 
monolayers: a comparison betwen chitosan molecules and chitosan nanoparticles. 
Pharm. Res. 20(11) (2003) 1812-1819. 
[171] E. Van Hame, H.L. Dewerchin, E. Cornelisen, B. Verhaselt, H.J. Nauwynck, 
Clathrin- and caveolae-independent entry of feline infectious peritonitis virus in 
monocytes depends on dynamin. J. Gen. Virol. 89(Pt 9) (2008) 2147-2156. 
[172] M.L. Torgersen, G. Skreting, B. van Deurs, K. Sandvig, Internalization of 
cholera toxin by diferent endocytic mechanisms. J. Cel Sci. 114(Pt 20) (2001) 3737-
3747. 
[173] E. Roger, F. Lagarce, E. Garcion, J.-P. Benoit, Lipid nanocariers improve 
paclitaxel transport throughout human intestinal epithelial cels by using vesicle-mediated 
transcytosis. J. Control. Release 140(2) (2009) 174-181. 
[174] Y. Kurtoglu, M. Mishra, S. Kannan, R. Kannan, Drug release characteristics of 
PAMAM dendrimer-drug conjugates with diferent linkers. Int. J. Pharm. 384(1-2) 
(2010) 189-194. 
[175] C.A. Lipinski, Drug-like properties and the causes of poor solubility and poor 
permeability. J. Pharmacol. Toxicol. Methods 44(1) (2000) 235-249. 
[176] G. Xu, W. Zhang, M.K. Ma, H.L. McLeod, Human carboxylesterase 2 is 
commonly expresed in tumor tisue and is correlated with activation of irinotecan. Clin. 
Cancer Res. 8(8) (2002) 2605-2611. 
[177] M.L. Rothenberg, Irinotecan (CPT-11): recent developments and future 
directions-colorectal cancer and beyond. Oncologist 6(1) (2001) 66-80. 
 
205 
 
 
[178] H. Zhao, B. Rubio, P. Sapra, D. Wu, P. Reddy, P. Sai, A. Martinez, Y. Gao, Y. 
Lozanguiez, C. Longley, L.M. Grenberger, I.D. Horak, Novel prodrugs of SN38 using 
multiarm poly(ethylene glycol) linkers. Bioconjug. Chem. 19(4) (2008) 849-859. 
[179] E. Roger, F. Lagarce, J.-P. Benoit, The gastrointestinal stability of lipid 
nanocapsules. Int. J. Pharm. 379(2) (2009) 260-265. 
[180] M. Ahlmark, J. Veps?l?inen, H. Taipale, R. Niemi, T. J?rvinen, Bisphosphonate 
prodrugs: synthesis and in vitro evaluation of novel clodronic acid dianhydrides as 
bioreversible prodrugs of clodronate. J. Med. Chem. 42(8) (1999) 1473-1476. 
[181] R. Kimsey, E. Harding, A spectrophotometric asay optimizing conditions for 
pepsin activity. Amer. Biol. Teach. 60(3) (1998) 200-201. 
[182] M. Mullaly, D. OCalaghan, R. FitzGerald, W. Donnely, J. Dalton, Proteolytic 
and peptidolytic activities in commercial pancreatic protease preparations and their 
relationship to some ehey protein hydrolysate characteristics J. Agric. Food Chem. 42 
(1994) 2973-2961. 
[183] C.E. Wheelock, T.F. Severson, B.D. Hamock, Synthesis of new 
carboxylesterase inhibitors and evaluation of potency and water solubility. Chem. Res. 
Toxicol. 14(12) (2001) 1563-1572. 
[184] W. Yamamoto, J. Verweij, P. de Bruijn, M.J. de Jonge, H. Takano, M. 
Nishiyama, M. Kurihara, A. Spareboom, Active transepithelial transport of irinotecan 
(CPT-11) and its metabolites by human intestinal Caco-2 cels. Anticancer Drugs 12(5) 
(2001) 419-432. 
 
206 
 
 
[185] M.E. Fox, S. Guilaudeu, J.M. Frechet, K. Jerger, N. Macaraeg, F.C. Szoka, 
Synthesis and in vivo antitumor eficacy of PEGylated poly(l-lysine) dendrimer-
camptothecin conjugates. Mol. Pharm. 6(5) (2009) 1562-1572. 
[186] T. Sakthivel, I. Toth, A.T. Florence, Distribution of a lipidic 2.5 nm diameter 
dendrimer carier after oral administration. Int. J. Pharm. 183(1) (1999) 51-55. 
[187] D. Pisal, V. Yelepeddi, A. Kumar, R. Kaushik, M. Hildreth, X. Guan, S. 
Palakurthi, Permeability of surface-modified poly (amido amine) (PAMAM) dendrimers 
across Caco-2 cel monolayers. Int. J. Pharm. 350(1-2) (2008) 113-121. 
[188] S. Parveen, S.K. Sahoo, Nanomedicine: clinical applications of polyethylene 
glycol conjugated proteins and drugs. Clin. Pharmacokinet. 45(10) (2006) 965-988. 
[189] T. Okuda, S. Kawakami, T. Maeie, T. Nidome, F. Yamashita, M. Hashida, 
Biodistribution characteristics of amino acid dendrimers and their PEGylated 
derivatives after intravenous administration J. Control. Release 114 (2006) 69-77. 
[190] T. Okuda, S. Kawakami, N. Akimoto, T. Nidome, F. Yamashita, M. Hashida, 
PEGylated lysine dendrimers for tumor-selective targeting after intravenous injection in 
tumor-bearing mice J. Control. Release 116 (2006) 320-336. 
[191] H. Yang, W.J. Kao, Dendrimers for pharmaceutical and biomedical applications. 
J. Biomater. Sci., Polym. Ed. 17(1-2) (2006) 3-19. 
[192] S. Chong, S.A. Dando, R.A. Morrison, Evaluation of biocoat intestinal epithelium 
diferentiation environment (3-day cultured Caco-2 cels) as an absorption screning 
model with improved productivity. Pharm. Res. 14(12) (1997) 1835-1837. 
 
207 
 
 
[193] A.B. Foraker, R.J. Walczak, M.H. Cohen, T.A. Boiarski, C.F. Grove, P.W. 
Swan, Microfabricated porous silicon particles enhance paracelular delivery of insulin 
across intestinal Caco-2 cel monolayers. Pharm. Res. 20(1) (2003) 110-116. 
[194] P.W. Swan, K.M. Hilgren, F.C. Szoka, Jr., S. Oie, Enhanced transepithelial 
transport of peptides by conjugation to cholic acid. Bioconjugate Chem. 8(4) (1997) 520-
525. 
[195] D. Shcharbin, J. Mazur, M. Szwedzka, M. Wasiak, B. Palecz, M. Przybyszewska, 
M. Zaborski, M. Bryszewska, Interaction betwen PAMAM 4.5 dendrimer, cadmium and 
bovine serum albumin: a study using equilibrium dialysis, isothermal titration 
calorimetry, zeta-potential and fluorescence. Colloids Surf. B iointerfaces 58(2) (2007) 
286-289. 
[196] H. Ghandehari, P.L. Smith, H. Elens, P.Y. Yeh, J. Kopecek, Size-dependent 
permeability of hydrophilic probes across rabbit colonic epithelium. J. Pharmacol. Exp. 
Ther. 280(2) (1997) 747-753. 
[197] P.M. Paulo, J.N. Lopes, S.M. Costa, Molecular dynamics simulations of charged 
dendrimers: low-to-intermediate half-generation PAMAMs. J. Phys. Chem. B 111(36) 
(2007) 10651-10664. 
[198] M. Moriya, M.C. Linder, Vesicular transport and apotransferin in intestinal iron 
absorption, as shown in the Caco-2 cel model. Am. J. Physiol. Gastrointest. Liver. 
Physiol. 290(2) (2006) G301-309. 
[199] T. Xuan, J.A. Zhang, I. Ahmad, HPLC method for determination of SN-38 
content and SN-38 entrapment eficiency in a novel liposome-based formulation, LE-
SN38. J. Pharm. Biomed. Anal. 41(2) (2006) 582-588. 
 
208 
 
 
[200] M. Ramesh, P. Ahlawat, N.R. Srinivas, Irinotecan and its active metabolite, SN-
38: review of bioanalytical methods and recent update from clinical pharmacology 
perspectives. Biomed. Chromatogr. 24(1) (2010) 104-123. 
[201] J. Peterson, V. Alikma, J. Subbi, T. Pehk, M. Lopp, Structural deviations in poly 
(amido amine) dendrimers: a MALDI-TOF MS analysis. Eur. Polym. J. 39 (2003) 33-42. 
 
 
 
  
 
  
 
Curriculum Vitae 
 
I. PERSONAL DATA 
Addres: 6219 Suton Court 
 Elkridge, MD 21201 
Telephone: 301-325-690 
Education  
2010 Completion of Ph.D. in Bioengineering with emphasis on 
polymeric drug delivery, University of Maryland, College 
Park, MD 
2006 Bachelors of Science, Chemical Enginering, University of 
Maryland, College Park, MD, Suma Cum Laude 
  
I. PROFESIONAL EXPERIENCE 
2006-2010 Graduate Research Asistant, Fischel Department of 
Bioengineering, University of Maryland, College Park, MD 
and Department of Pharmaceutical Sciences, University of 
Maryland, Baltimore, MD. 
2009 Summer Intern, Department of Formulation Sciences, 
MedImune LC, Gaithersburg, MD 
2007-2008 Graduate Teaching Asistant, Fischel Department of 
Bioengineering, University of Maryland, College Park, MD. 
 
  
II. PUBLICATIONS IN REFERED JOURNALS 
1. D. Goldberg, N. Vijayalakshmi, P. Swan and H. Ghandehari. G3.5 PAMAM 
Dendrimer-SN38 Conjugates Enhance Transepithelial Transport of SN38 while 
minimizing Gastrointestinal Toxicity and Release. J Control Release, Submited. 
2. D. Goldberg, S. Bishop, A. Shah and H. Sathish, Formulation Development of 
Therapeutic Monoclonal Antibodies using High-throughput Fluorescence and 
Static Light Scatering Techniques: Role of Conformational and Colloidal 
Stability, J Pharm Sci, Epub ahead of print, Oct 19, 2010. 
3. D. Goldberg, H. Ghandehari and P. Swan, Celular Entry of G3.5 PAMAM 
Dendrimers by Clathrin- and Dynamin-dependent Endocytosis is Required for 
Tight Junctional Opening in Intestinal Epithelia, Pharm Res, 27(8) (2010) 1547-
57. 
4. D. Swet, R. Kolhatkar, A. Ray, P. Swan, and H. Ghandehari, Transepithelial 
Transport of PEGylated Anionic Poly (Amido Amine) Dendrimers: Implications 
for Oral Drug Delivery. J Control Release 138(1) (2009) 78-85. 
5. M. Al-Sheikhly, D. Swet, L. Salamanca-Riba, B. Varughese, J. Silverman, A. 
Christou, and W. Bentley, Radiation-induced Failure Mechanisms of GaAs-
based Biochips, IEE, Device and Materials Reliability, 4, (2004) 192-197. 
 
IV. BOK CHAPTER 
1. R. Kolhatkar, D. Swet, and H. Ghandehari, Functionalized Dendrimers as 
Nanoscale Drug Cariers, in Multifunctional Pharmaceutical Nanocariers, 
Springer 201-232, 2008. 
 
  
V. ABSTRACTS FOR NATIONAL AND INTERNATIONAL CONFERENCES 
1. D. Goldberg, N. Vijayalakshmi, P. Swan and H. Ghandehari. G3.5 PAMAM 
Dendrimer-SN38 Conjugates Enhance Transepithelial Transport of the Drug 
while minimizing Gastrointestinal Toxicity, 8
th
 Meting of the Globalization of 
Pharmaceutics Education Network, Chapel Hil, NC, November 10-12, 2010. 
(Podium Presentation) 
2. B. Avarit, D. Goldberg, H. Ghandehari and P. Swan, PAMAM Dendrimers as 
Potent Tight Junctional Modulators, 8
th
 Meting of the Globalization of 
Pharmaceutics Education Network, Chapel Hil, NC, November 10-12, 2010. 
(Poster) 
3. D. Goldberg, N. Vijayalakshmi, P. Swan and H. Ghandehari, G3.5 PAMAM 
Dendrimer-SN38 Conjugates Enhance Transepithelial Transport of the Drug 
while Minimizing Gastrointestinal Toxicity and Release, 8
th
 International 
Nanomedicine and Drug Delivery Symposium (NanoDDS?10), Omaha, NE, 
October 3-5, 2010. (Poster) 
4. D. Goldberg, P. Swan and H. Ghandehari, Mechanisms of PAMAM 
Dendrimer Transepithelial Transport and Tight Junction Modulation, 37
th
 
Annual Meting and Exposition of the Controlled Release Society, Portland, OR, 
July 10-14, 2010. (Podium Presentation) 
5. D. Goldberg, P. Swan and H. Ghandehari, Mechanisms of PAMAM 
Dendrimer Transepithelial Transport and Tight Junction Modulation, The 26
th
 
Southern Biomedical Engineering Conference, College Park, MD, April 30- May 
2, 2010. (Podium Presentation) 
 
  
6. D. Swet, R. Kolhatkar, P. Swan, and H. Ghandehari, Mechanisms of Transport 
of PEGylated Anionic PAMAM Dendrimers Acros Caco-2 Cel Monolayers, 
14
th
 International Symposium on Recent Advances in Drug Delivery Systems, 
Salt Lake City, UT, February 15-18, 2009. (Poster) 
7. D. Swet, R. Kolhatkar, and H. Ghandehari, PEGylation of Anionic PAMAM 
Dendrimers: Implications for Oral Delivery, 35
th
 Annual Meting and Exposition 
of the Controled Release Society, New York, NY, July 12-16, 2008. (Poster). 
8. R.B. Kolhatkar, D. Swet, and H. Ghandehari, PAMAM Dendrimers: Surface 
Modification and Potential in Oral Delivery of SN-38, 3rd Annual Mountain 
West Biomedical Engineering Conference, Salt Lake City, Utah, September 14-
15, 2007. 
9. H. Ghandehari, R. Kolhatkar, A. Nan, S.B. Le, and D. Swet, Transcelular 
Transport and Toxicity of Dendritic and Silica-Based Nanoconstructs, The 5th 
International Nanomedicine and Drug Delivery Symposium (NanoDDS?07), 
Boston, MA, November 1-2, 2007. 
 
VI. HONORS AND AWARDS 
2010 Southern Biomedical Engineering Conference Graduate Student Paper 
Award 
2009 Best Graduate Student Poster, MedImune Intern Poster Competition  
2009 Fischel Felowship in Bioengineering: Awarded for ?DendriPharm 
Systems? busines plan detailing comercialization opportunity of 
graduate research 
 
  
2006-2009 National Science Foundation Graduate Research Felowship 
2006 A. James Clark School of Engineering Deans? Award  
2006 Chemical Engineering Outstanding Senior Award 
2005-2006 Bary M. Goldwater Scholarship (National Scholarship) 
  
VI. LEADERSHIP ACTIVITIES 
2008-2010 National Science Foundation Graduate Research Felowship Contact 
2007 High School Summer Student Mentor 
 
VII. PROFESIONAL MEMBERSHIPS 
1. Controlled Release Society 
2. Tau Beta Pi Engineering Honors Society