UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Liposomal encapsulation of the anti-leukemic small molecule UNC0642 for increased tolerability Cottle, Andrew Gregory 2018

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2018_september_cottle_andrew.pdf [ 11.26MB ]
JSON: 24-1.0369269.json
JSON-LD: 24-1.0369269-ld.json
RDF/XML (Pretty): 24-1.0369269-rdf.xml
RDF/JSON: 24-1.0369269-rdf.json
Turtle: 24-1.0369269-turtle.txt
N-Triples: 24-1.0369269-rdf-ntriples.txt
Original Record: 24-1.0369269-source.json
Full Text

Full Text

 LIPOSOMAL ENCAPSULATION OF THE ANTI-LEUKEMIC SMALL MOLECULE UNC0642 FOR INCREASED TOLERABILITY by  Andrew Gregory Cottle  B.Sc. (Hons.), The University of Calgary, 2013  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Biochemistry and Molecular Biology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2018  © Andrew Gregory Cottle, 2018   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis/dissertation entitled:  Liposomal encapsulation of the anti-leukemic small molecule UNC0642 for increased tolerability   Submitted by Andrew Gregory Cottle  in partial fulfillment of the requirements for the degree of Master of Science in Biochemistry and Molecular Biology  Examining Committee: Pieter Cullis, Biochemistry and Molecular Biology  Supervisor  Fabio Rossi, Medical Genetics Supervisory Committee Member  Leonard Foster, Biochemistry and Molecular Biology Supervisory Committee Member Shyh-Dar Li, Pharmaceutical Sciences Additional Examiner     Additional Supervisory Committee Members:  Supervisory Committee Member  Supervisory Committee Member    iii  Abstract  Acute Myeloid Leukemia (AML) is the most poorly treated subtype of leukemia, suffering from dated chemotherapeutic regimens and poor long-term survival rates. New therapies are needed for this aggressive disorder. Epigenetic inhibitors targeting G9a/GLP, crucial histone methyltransferases for AML progression, are a promising avenue for treatment. Unfortunately, one of the most potent small molecule inhibitors of these enzymes, UNC0642, suffers from dose-limiting toxicities. It was hypothesized that encapsulation of UNC0642 in liposomal nanoparticles would help attenuate these toxicities, better control drug release and biodistribution of the small molecule, and ultimately enable the therapeutic use of UNC0642.   The work presented here demonstrates that UNC0642 can be successfully loaded within liposomal systems composed of DSPC/Cholesterol or POPC/Cholesterol to a maximum drug:lipid ratio of 0.2. These particles show stable drug release patterns, which may be due to precipitation of the drug within the nanoparticle interior as observed using cryo-TEM. Furthermore, the particles exhibit a dose-dependent uptake into HoxA9/Meis1 AML cells in vitro and an anti-proliferative effect that is 10-times lower than the free drug in cell culture. When injected in a single dose tolerability study using healthy C57Bl/6 mice, liposomal encapsulation is shown to completely abolish the acute toxicity associated with free UNC0642 injection. Even at doses 8-times higher than the maximum tolerated dose of the free small molecule, minimal toxicity is observed for liposomal UNC0642. In addition, no major signs of chronic toxicity are observed when the encapsulated drug is injected in a multiple-dose regimen (2 injections/week for 4 weeks), as determined by detailed clinical iv  scoring of animal health, body weight, and blood chemistry. These reductions in acute and chronic toxicity are attributed to the nanoparticle carriers’ ability to alter the biodistribution of the drug and prevent exposure of the nervous system to UNC0642.   The data presented in this thesis form the basis of a promising and novel therapeutic strategy for treatment of AML through epigenetic inhibition. These systems are well suited for pre-clinical studies in animal models of disease and offer an easily modifiable nanoparticle carrier for tuning biodistribution and drug release to better treat this aggressive form of leukemia.     v  Lay Summary   Acute Myeloid Leukemia (AML) is a form of cancer originating from the body’s blood cells and is very poorly treated at present. However, a novel way of combatting AML has been discovered. This method uses drugs that prevent the growth and spread of leukemic cells by inhibiting the function of DNA-modifying proteins crucial to leukemic cell function. One of these drugs, UNC0642, is well suited to treating cancer cells in a laboratory setting but has adverse side effect when injected. To address these side effects, we loaded the drugs into nanoparticle carriers called liposomes. In this study, we demonstrate an efficient way of making UNC0642-loaded liposomes, study their structure, and ultimately determine that they can effectively nullify the toxicity associated with the free drug in animal models. Overall, these nanoparticles are a promising step forward towards the development of a new and effective treatment for AML.     vi  Preface   Formulation of all nanoparticles, loading with UNC0642, and subsequent in vitro characterization, including sizing and drug release studies, were performed by myself. UNC0642-LNP samples imaged using cryo-TEM imaging were prepared by myself and imaged by Dr. Jayesh Kulkarni with the assistance of the UBC BioImaging Facility. Cell culture experiments and in vivo tolerability studies, including both single and multiple-dose experiments, were performed by myself and Phuong Nguyen in collaboration with the lab of Dr. Fabio Rossi. FACS analysis for particle uptake experiments was performed with the assistance of Dr. Genc Basha. Quantification of blood biomarkers was performed by Idexx Reference Laboratories (Delta, BC).   I was responsible for the experimental design of UNC0642-LNP preparation and characterization with assistance and contributions from Drs. Yuen Yi C. Tam, Sam Chen, Ying Tam, Jayesh Kulkarni, Genc Basha, and Ms. Nisha Chander. Experimental design for cell culture and in vivo experiments was performed by myself and Phuong Nguyen in collaboration with the lab of Dr. Fabio Rossi, with assistance, input, and feedback of the above-mentioned individuals. All data analysis was performed by myself.    All chapters of this thesis were prepared and written by myself. Drs. Pieter Cullis, Sam Chen, Yuen Yi C. Tam, and Jayesh Kulkarni were involved in editing various iterations and sections of this thesis.   All work and procedures involving animals presented in this dissertation have been approved by the Animal Care Committee at The University of British Columbia and were vii  performed in accordance with guidelines established by the Canadian Council on Animal Care.  Animal Care and Ethics Protocol: A16-0275.  Online Animal Care Training Program Certificate Number: 6861-14         viii  Table of Contents  Abstract ................................................................................................................................... iii	Lay Summary .......................................................................................................................... v	Preface ..................................................................................................................................... vi	Table of Contents ................................................................................................................. viii	List of Tables .......................................................................................................................... xi	List of Figures ........................................................................................................................ xii	List of Abbreviations ........................................................................................................... xiv	Acknowledgements ............................................................................................................. xvii	Dedication ............................................................................................................................. xix	Chapter 1: Introduction ......................................................................................................... 1	1.1	 Liposomal nanoparticles for small molecule encapsulation ............................................ 1	1.1.1 Overview of lipid nanoparticle technology ........................................................................... 1	1.1.2 History of liposomal nanoparticle development ................................................................... 2	1.1.3 Liposome production ............................................................................................................ 5	1.1.4 Techniques for encapsulating therapeutics within liposomes ............................................. 12	1.1.5 Liposome interactions with biological components in vivo ................................................ 15	1.1.6 Liposome characteristics influencing blood clearance ....................................................... 18	1.1.7 Liposome characteristics influencing drug release rates ..................................................... 22	1.1.8 Application of liposomal therapeutics in cancer treatment ................................................. 24	ix  1.2	 Acute myeloid leukemia and its treatment ...................................................................... 28	1.2.1 Overview of leukemia ......................................................................................................... 28	1.2.2 Treatment of acute myeloid leukemia ................................................................................. 31	1.2.3 Epigenetic regulators and their application to AML treatment ........................................... 32	1.3 Thesis objectives ....................................................................................................................... 35	Chapter 2: Materials and Methods ..................................................................................... 37	2.1 Materials .................................................................................................................................... 37	2.2 Preparation of liposomes with pH gradient ........................................................................... 37	2.3 Remote loading of UNC0642 into preformed liposomes ....................................................... 38	2.4 Characterization of UNC0642-LNP ........................................................................................ 38	2.5 Drug release assay of UNC0642-LNP incubated in PBS (37 °C) ......................................... 39	2.6 Drug release assay of UNC0642-LNP incubated in 50% FBS (37 °C) ................................ 39	2.7 Cryogenic transmission electron microscopy (cryo-TEM) of UNC0642-LNP ................... 40	2.8 In vitro anti-proliferation assays ............................................................................................. 41	2.9 In vitro cell uptake assays ......................................................................................................... 42	2.10 Single dose in vivo tolerability tests ....................................................................................... 42	2.11 Multiple dose in vivo tolerability tests .................................................................................. 43	Chapter 3: Design, Characterization, and Tolerability of UNC0642-LNP ..................... 45	3.1 Synopsis ..................................................................................................................................... 45	3.2 Results ........................................................................................................................................ 46	3.2.1 UNC0642 can be efficiently encapsulated in DSPC/Chol and POPC/Chol liposomes via remote loading .............................................................................................................................. 46	3.2.2 Limited drug release is observed from UNC0642-LNP upon incubation in 37 °C PBS or 50% FBS ...................................................................................................................................... 50	x  3.2.3 Cryo-TEM imaging of DSPC/Chol UNC0642-LNP reveals amorphous spot patterns of electron density within the liposome interior ............................................................................... 54	3.2.4 In vitro studies in HoxA9/Meis1 murine leukemia cells show dose-dependent liposome uptake and lower anti-proliferative effect in UN0642-LNP compared to free UNC0642 ........... 56	3.2.5 Liposomal encapsulation completely abolishes the acute toxicity of UNC0642 and results in no substantial long-term toxic after a single injection ............................................................. 60	3.2.6 When administered using a multiple-dose regimen (2 injections/week, 4 weeks), low doses of free drug (1.25 mg/kg) and high doses of UNC0642-LNP (20 mg/kg) are tolerated with only minor signs of toxicity ................................................................................................................. 67	3.3 Discussion .................................................................................................................................. 77	Chapter 4: Conclusion and Future Directions ................................................................... 88	References .............................................................................................................................. 92	   xi  List of Tables Table 1. Overview of lipid-based therapeutics approved by the FDA. .................................... 2	Table 2. Leukemia subtypes and associated survival statistics. .............................................. 31	Table 3. Particle size parameters for liposomes used to generate UNC0642-LNP. ............... 48	     xii  List of Figures Figure 1. Structure of the G9a/GLP inhibitor UNC0642 ........................................................ 35	Figure 2. Chemical structure of liposome lipid components of UNC0642-LNP. .................. 48	Figure 3. UNC0642 can be encapsulated in DSPC/Chol or POPC/Chol liposomes to a maximum 0.2 drug:lipid (molar) via remote loading. ................................................. 50	Figure 4. Limited drug release is obtained upon incubation of DSPC/Chol or POPC/Chol UNC0642-LNP in PBS at 37 °C. ................................................................................ 51	Figure 5. Incubation in 50% FBS in PBS results in increased but still limited drug release rates. ............................................................................................................................ 53	Figure 6. Cryo-TEM imaging of DSPC/Chol particles show amorphous and electron dense structures on the liposome interior. ............................................................................. 56	Figure 7. HoxA9/Meis1 AML cells incubated with fluorescent liposomes display a dose-dependent increase in mean cell fluorescence. ........................................................... 57	Figure 8. UNC0642-LNP causes a dose-dependent inhibition of proliferation in in HoxA9/Meis1 cells at a scale 10x less than free UNC0642. ...................................... 60	Figure 9. Liposomal encapsulation eliminates the dose-dependent acute toxicity of UNC0642. ................................................................................................................... 64	Figure 10. UNC0642-LNP and free UNC0642 show no signs of significant longitudinal toxicity following a single dose. ................................................................................. 66	Figure 11. Multiple dose regimens of UNC0642 or UNC0642-LNP injection show limited indications of long-term toxicity based on body weight and clinical health score. .... 71	xiii  Figure 12. The ability of liposomal encapsulation to abolish the acute toxic effects of UNC0642 injection is maintained through multiple injections. ................................. 74	Figure 13. Blood chemistry analysis of animals measured after completion of a multiple dose regimen shows no clear signs of substantial chronic toxicity for low doses of free drug (1.25 mg/kg) or high doses of UNC0642-LNP (20 mg/kg). .............................. 77	     xiv  List of Abbreviations A/G   Albumin/globulin (ratio) ALL  Acute lymphocytic leukemia ALP  Alkaline phosphatase ALT  Alanine transaminase AML  Acute myeloid leukemia Apo   Apolipoprotein AS  Ammonium sulfate AST  Aspartate transaminase BD  Biodistribution BUN  Blood urea nitrogen Chol  Cholesterol  CK  Creatine kinase  CLL  Chronic lymphoblastic leukemia CMC  Critical micelle concentration  CML  Chronic myeloid leukemia  Cryo-TEM Cryogenic transmission electron microscopy DiI  1,1'-dioctadecyl- 3,3,3',3'-tetramethylindocarbocyanine perchlorate  DSPC  1,2-distearoyl-sn-glycero-3-phosphocholine DSPE   1,2-distearoyl-sn-glycero-3-phosphoethanolamine EtOH  Ethanol  EPC  Egg phosphatidylcholine xv  EPR  Enhanced permeability and retention effect FACS  Fluorescence-activated cell sorting FBS  Fetal bovine serum  GGT  Gamma-glutamyl transferase  IP   Intraparatoneal  IV   Intraveinous  LNP   Lipid nanoparticles  LDH  Lactate dehydrogenase MFI  Mean fluorescence intensity MLV   Multilamellar vesicles  MPS   Mononuclear phagocyte system  Pb  Protein binding parameter  PBS   Phosphate buffered saline  PC   Phosphatidylcholine PDI   Polydispersity index  PEG  Polyethylene glycol  PI  Propidium iodide  PK  Pharmacokinetics pKa  Modified acid dissociation constant (Ka): -log(Ka) POPC  1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine RES  Reticuloendothelial system SD  Standard deviation  SDMA  symmetric dimethylarginine xvi  SUV  Small unilamellar vesicles UHPLC  Ultra high performance liquid chromatography WFI  Water for injection  xvii  Acknowledgements   I would like to sincerely thank my supervisor, Dr. Pieter Cullis, for giving me the opportunity to work in his lab. Becoming a part of his research team has evolved from an engaging thesis project to an all-encompassing experience that has taught me unbelievable amounts about research, being part of a dedicated team, and the many ways science can be applied in and outside of the lab.  A huge thank you to my committee members, Drs. Fabio Rossi and Leonard Foster for supporting me through the ups and downs of my project. The mentalities with which you approach not only science but life in general are truly inspiring and reflect brightly in the members of your research groups I have had the pleasure of meeting.   I would have been absolutely lost over the past years if it weren’t for the other members of the Cullis lab that have grown into my family during my time here. Thank you sincerely to Drs. Chris Tam and Sam Chen for your guidance and leadership throughout my time here. You have been phenomenal mentors and have taught me almost everything I know about this field and so much more. To our lab manager, Cayetana Schluter, a gigantic thank you for keeping the entire lab from bursting into flames and helping with absolutely everything. Thank you also to Drs. Ying Tam and Paulo Lin, your support in addressing random study problems and providing (supposedly) witty banter has not gone unappreciated. I have had the pleasure of working alongside many wonderful post-docs, visiting scientists, and research associates in the lab including Drs. Terry Allen, Souvik Biswas, Lizzy Wang, Chen Wan, Dominik Witzigmann, Roy van der Meel, Genc Basha, Igor Zhigaltsev, Ismail Hafez, Yan Liu, and Josh Zaifman, and thank them all for their support and teachings. Thank xviii  you to my fellow grad students (past and present), including Dr. Alex Leung, Dr. Justin Lee, Mina Ordobadi, Nisha Chander, and especially Joslyn Quick and Dr. Jayesh Kulkarni. Joslyn, thank you for your wonderfully dark sense of humor, dance/walk moves, and impeccable taste in music. Jay thank you for your charmless personality, the wonderful drawings you have given us all, and for always having my back inside and out of the lab.  A huge thank you to my collaborator on this project, Phuong (Joey) Nguyen from the Rossi Lab, for your contributions to the cell and animal studies, without which my project would not have been possible.  Thank you to the CIHR Banting and Best CGS Program as well as UBC for financial scholarship support throughout my project.  To everyone at Molecular You, especially Dr. Robert Fraser, Solveig Johannessen, Ash Anwar, and Nadya Calderon, thank you for giving me a second life outside of my studies and showing me the delights of growing a small team into something amazing.   A heartfelt thanks to the friends, new and old, that have always been there for me. Thanks to the many new friends I have made in the Biochemistry program, especially Dr. Craig Kerr, Michael Carlson, and Sean Workman for keeping life ridiculous and entertaining. To Michael Daw, Heather Worthington, and Geri Ruissen, thanks for suffering my friendship for the longest and sticking by my side, always. Thank you to Julia Crimeni for being ridiculous, kind, and unconditionally loving. You are the most wonderful escape when life gets overwhelming, I will never stop loving you for it.   Of course, a final thank you to my family for making me who I am today, always giving me a safe haven of warmth and acceptance to return to, and teaching me what is truly important in life.   xix  Dedication  To my family   1  Chapter 1: Introduction 1.1 Liposomal nanoparticles for small molecule encapsulation 1.1.1 Overview of lipid nanoparticle technology Lipid nanoparticles (LNP) encompass a diverse class of sub-micron scale carriers that have seen a rapid expansion in design and application since their original discovery in the 1960s. From humble beginnings in model membrane studies, lipid nanoparticles have evolved into sophisticated systems with numerous applications in the medical field and beyond. Numerous FDA-approved drugs utilize LNP technology, with many more in various stages of clinical trials. Though a complex range of LNP have been developed, each with a unique structure and capable of delivering distinct therapeutic payloads, the first developed and best characterized are liposomal carriers. Such systems are composed of one or more spherically aligned lipid bilayers surrounding an aqueous core. Liposomal bilayers are comprised of various mixtures of natural and/or synthetic lipids, the exact ratio of which can be modulated to obtain physiochemical properties in line with the desired application of the formulation. For example, lipid composition has been shown to affect important pharmacological properties such as pharmacokinetics and biodistribution, as discussed in the following sections. In addition to lipid composition, modification of the internal aqueous compartment relative to the external environment can produce important characteristics aiding in encapsulation of a variety of small molecules. Once prepared, liposomal nanoparticles offer unique advantages for drug delivery, including protecting encapsulated therapeutics from degradation, altering biodistribution and circulation halftimes of their payloads upon injection, and helping to decrease toxic side effects of free drugs upon 2  injection. Numerous lipid-based therapeutics have been approved for use by the FDA across a variety of applications (see Table 1). Table 1. Overview of lipid-based therapeutics approved by the FDA.  Product Name Encapsulated Agent Indication Year of Approval Doxil Doxorubicin AIDS-related kaposi’s sarcoma, ovarian, and breast cancer 1995 Abelcet  Amphotericin B Fungal infection 1995 DaunoXome Daunorubicin Aids-related kaposi’s sarcoma 1996 Amphotec Amphotericin B Fungal infection 1996 Ambisome Amphotericin B Fungal infection 1997 Depocyt Cytarabine  Neoplastic meningitis 1999 Myocet Doxorubicin Breast cancer  (combination therapy) 2000 Visudyne Verteporfin Macular degeneration with subfoveal choroidal neovascularization 2000 Mepact  Mifamurtide  Osteosarcoma  (combination therapy) 2004 DepoDur Morphine  Analgesic  2004 Exparel Bupivacaine Regional analgesic  2011 Marqibo Vincristine Acute lymphoblastic leukemia 2012 Onivyde Vincrisitne  Metastatic pancreatic cancer  (combination therapy)  2015 Vyxeos Cytarabine and Daunorubicin  Acute myeloid leukemia  2017  1.1.2 History of liposomal nanoparticle development   The first documented precursor to modern liposomes was presented in a study by Bangham and Horne in 1964, in which dispersions of lecithin were observed to form multi-lamellar structures comprised of concentric rings of lipid bilayers. These rings were referred to as multilamellar smectic mesophases 1,2.  Following their discovery, key qualities of these 3  dispersions were established that would lay the groundwork for the utility of modern day liposomes. These qualities included the ability of the structures to maintain ion gradients across their membranes, which could be disrupted upon the introduction of detergents 3,4. As interest in these structures grew, they were formally classified as “liposomes”, defined as single or multilamellar microscopic vesicles comprised of one or more lipid bilayer 2,5. In order to generate reproducible liposomes, major strides were made in refining their production and structure. It was found that sonication of multilamellar dispersions resulted in the formation of more uniform, unilamellar liposomes 6. Shortly after, it was found that similar unilamellar structures could also be generated through rapid injection of an ethanolic lipid solution into buffer 7. These were some of the first steps towards the production of monodisperse liposome populations with uniform and consistent properties, a key feature for their future therapeutic applications.  A crucial functionality of these systems was established by Gregoriadis and Ryman in 1971, who showed that liposomes were capable of encapsulating therapeutic agents 8. Specifically, they encapsulated the proteins amyloglucosidase and albumin and showed stable retention of these payloads upon injection into animals. Once this precedent was set, further discoveries concerning the loading of therapeutics into liposomes were made, with encapsulation of actinomycin D and penicillin demonstrated in 1973 9. This study also noted the ability of liposomes to alter the pharmacokinetics of injected molecules, with entrapped penicillin showing approximately 10-fold higher plasma concentrations compared to free penicillin 60 minutes post injection. The entrapped penicillin showed a notable increase in blood circulation time as well as a greater biodistribution to the liver and kidney compared to the free drug. These early studies helped to establish that liposomes could be used alter and 4  control a therapeutic agent’s bioavailability and pharmacokinetic properties. In the following years, many further demonstrations of drug encapsulation were documented, along with clear evidence of altered biodistribution. Kimelberg et al. encapsulated the chemotherapeutic and anti-inflammatory agent methotrexate into positively charged liposomes and showed a 100-fold increase in circulating drug 4 hours post injection 10. This study demonstrated two key features of liposomes. First, encapsulation prevented degradation of the drug upon introduction to the bloodstream. Second, the size of the liposomes affected their pharmacokinetic properties, with larger particles showing only a 6-fold increase in plasma methotrexate relative to free drug injection, while smaller particles showed a 100-fold increase.  Further discoveries identifying fundamental properties of liposomes that would later turn out to be vital in their application to drug delivery were made in the years following. The effect of changing liposome lipid composition on their uptake into cells was shown in 1976 by Poste and Papahadjopoulos 11. This work showed that varying the degree of saturation of the component lipids determined not only the total level of particle uptake into the cells but also the route through which they were taken up. Specifically, lipid compositions including more saturated species that raised the phase transition temperature of the mixture above 37 °C (referred to as solid lipid vessels) showed uptake via endocytic pathways. In contrast, “fluid” vessels with transition temperatures below 37 °C showed uptake via non-endocytic pathways. Though this study only considered in vitro uptake of particles, similar observation were soon made in vivo 12. This study also indicated the growing interest in the application of liposomes to improving chemotherapeutic treatment, with demonstrated encapsulation of vinblastine, actinomycin D, cytosine arabinoside, and 5  daunomycin. In the following years, many groups focused on the benefits of encapsulating chemotherapeutics within liposomes, with increased pre-clinical efficacy and decreased toxicity arising as general outcomes for many drugs. Some of the first demonstrations of increased efficacy was shown in 1975 and 1976, where encapsulated cytarabine was shown to be significantly more effective than the free drug in treating leukemic mice 13,14. These promising initial studies in the field of chemotherapeutic encapsulation paved the way for the first FDA approved liposomal therapeutic: Doxil 15. Composed of liposome-encapsulated doxorubicin, Doxil was approved in 1995 for the treatment of AIDS-related kaposi’s sarcoma and ovarian cancer. Since then, over 14 liposomal therapeutics have been approved for use in humans across a wide variety of applications (see Table 1).   1.1.3 Liposome production   The first unilamellar liposomes were generated by dissolving their components in chloroform or a similarly volatile solvent and then drying this mixture on a glass surface to produce a thin lipid film. This film was then hydrated under rapid mixing to cause swelling of the film and spontaneous formation of multilamellar vesicles (MLV) due to the amphipathic nature of the lipid components. Though these MLV were originally useful as model membranes (see Section 1.1.2 History of liposomal nanoparticle development), their lack of uniformity and reproducibility made them poorly suited to any therapeutic application. As such, there have been many advances in optimizing the production of liposomes, with modern day methods enabling tight control over final liposome size and uniformity.  6  Initially, unilamellar liposomes were made from MLV with the use of sonication (ultrasonic irradiation). Though this method of generating small unilamellar vesicles (SUV) allows for control over particle lamellarity, the size range of the vesicles produced is typically very broad, ranging from 20-150 nm 16. To address this problem, size exclusion chromatography was originally used to isolate a subpopulation of the particles with a particular diameter 17,18.  However, only a small percent of the original MLV are successfully converted into the target SUV, with those outside the target size distribution filtered out and discarded. In addition to the poor overall yield, size exclusion chromatography results in significant dilution of the sample generated, requiring a time consuming concentration step 16. Differential high-speed centrifugation has been used to avoid size exclusion chromatography and serves as a far more efficient method of isolating the desired SUV. This technique offers improvements in the time investment necessary and also provides better separation of the various subpopulations of SUV and MLV 16. In spite of the improvements in post-sonication processing, there are major shortcomings associated with the sonication step itself. Firstly, since sonication relies on mechanical disruption of MLV to form SUV, there is an inherent generation of heat during the process that can lead to lipid degradation. Secondly, if probe tip sonicators are used, samples can become contaminated with metals sloughed from the probe during extended use 19. Finally, and perhaps most importantly, sonication is unable to produce a monodisperse population of SUV directly and requires the use of post-processing steps as previously detailed.       An alternative method of size reducing MLV to liposomes of a controlled diameter, which also relies on mechanical dispersion, is homogenization. Early approaches to this technique utilized a French Press to force MLV suspensions through a single small aperture 7  under high pressure, producing sufficient forces to break MLV into consistently sized SUV 20. To address the volume limitations of this initial technique (approximately 1-40 mL), a variation on the concept was designed in the form of microemulsificaiton 21. This process also relies on high pressure to size reduce MLV to liposomes but separates the sample into multiple streams using microfluidic channels. These streams are ultimately brought back together in a controlled collision with one another inside an interaction chamber, generating turbulent forces sufficient to break apart MLV into smaller liposomes. The resulting samples are then cycled continuously back through the apparatus for a set number of passes allowing for thorough size reduction. A variation on this concept, relying on similar high pressure size reduction, employs a single sample stream ejected from an aperture at high speed before collision with a metal surface within the interaction chamber 22. Regardless of the collision technique, these forms of homogenization are compatible with large sample volumes (>50 mL) and sterile preparation conditions 23.  Despite the advances made in the homogenization techniques, one of the most widely used methods of MLV size reduction is extrusion. The concept relies on forcing samples through polycarbonate membranes containing pores of controlled diameter, a process that is repeated for a number of passes. Initially demonstrated by Olson et al. in 1979, the first iterations of this technique used very low pressures and required multiple passes through membranes of consecutively decreasing pore size 24. Major improvements were made with the invention of a custom-built device allowing for extrusion at increased pressures by Hope et al. in 1985 25. Later brought to market as the Lipex Extruder, this device allowed for reliable size reduction of MLV preparations into unilamellar liposomes at rapid timescales without the need of organic solvents. Furthermore, it allowed for control over particle size 8  through variation of the pore diameter of the polycarbonate membranes used, allowing for generation of liposomes from 30-400 nm 26. The high pressures enabled by the extruder are key to the success of the technique, as mechanistic studies have shown that samples must reach a certain pressure threshold before they are able to pass through the polycarbonate membrane pores. This threshold is dependent on the lipid composition used and represents the minimum pressure necessary to induce rupture of the bilayers present in the MLV processed 27,28. Importantly, these studies showed that extruded MLV undergo rupture and reformation of their bilayers as opposed to only deformation when passing through membrane pores.  To further improve uniformity, pre-formed MLV can also be converted into liposomes expressing higher internal volumes (and thus decrease multilamellarity) using freeze-thaw techniques. This method employs repeated cycles of freezing the MLV sample using liquid nitrogen and then quickly thawing it at 40 °C 29. Particles generated in this way are often not completely unilamellar and exhibit a vesicle within vesicle structure. This method is often used in combination with other techniques, such as extrusion, to generate uniform unilamellar liposomes.   The above described techniques for liposome preparation rely on mechanical disruption of previously prepared (or pre-formed) MLV systems. However, several alternate processes relying on the combination of aqueous and organic solutions to generate liposomes have also been developed. These techniques allow for the direct production of unilamellar systems, without the need of separate MLV production and size reduction steps.   The first reported method of aqueous/organic mixture production was published by Batzri and Korn in 1973 7. Here, lipid components dissolved in ethanol were rapidly injected 9  into an aqueous buffer solution (KCl). Due to the rapid increase in solvent polarity upon mixing, the component lipids spontaneously arrange into liposomes of small size (~26 nm). This was the first step in the development of numerous organic/aqueous mixture techniques and offered a valid alternative to sonication, the only other method of liposome preparation at the time. Unlike sonication, no degradative stresses were placed on the lipid components and there was no risk of metal contamination. However, the method offered little control over the size of particles generated. In fact, the relatively small liposomes made using this technique showed low entrapped volumes and high degrees of membrane curvature, which would later be shown to be highly undesirable when preparing liposomes for drug delivery purposes 30.  Improvements have been made on this initial method through introducing water to a concentrated ethanol stock of lipid components followed by rotary evaporation to eliminate the ethanol component. Additionally, modification of the ethanol:water ratio used during the initial combination step allowed for control over generated particle size. By decreasing the percent ethanol within the range of 60% to 20%, the method was able to generate particles between 1280 nm to 160 nm in diameter. A similar production method, reverse phase evaporation, was developed in 1978 by Szoka and Papahadjopoulos 31. Using this method, lipids were initially dissolved in organic solvents with relatively poor water miscibility (ethyl-ether, isopropyl-ether, halothane, or trifluorotrichloroethane) prior to the addition of aqueous solution. The mixture is sonicated to form an emulsion and then rotary evaporated to gradually remove the organic phase. This leads to the production of a gel that, upon agitation and/or dilution with additional aqueous buffer, converts into a liposome suspension. The proposed mechanism behind this technique is that sonication of the aqueous/organic mixture 10  generates inverted micelles composed of phospholipids surrounding an aqueous core. As the organic solvent is removed, these inverted micelles cluster closer and closer together, beginning the formation of the aforementioned gel. At a critical organic:aqueous solvent threshold, a certain percentage of the inverted micelles collapse and their lipophilic components reform around the remaining inverted micelles to form bilayers, thus producing unilamellar liposomes 32. This technique offers several advantages, including that much of the organic solvent is removed during rotary evaporation and thus requiring limited, if any, dialysis to eliminate residual solvent. Secondly, it allows for high encapsulation efficiencies of numerous compounds simply by introducing them to the aqueous buffer prior to its initial mixing with the organic phase. Encapsulation efficiencies of up to ~60% were obtained for carboxyfluorescin and cytarabine, with entrapment of other compounds possible but at lower levels 31.   Liposomes can also be prepared with the aid of detergents. Generally, these techniques rely on solubilizing lipid components within mixed detergent micelles followed by gradual removal of the detergent to promote liposome formation. The composition and physiochemical properties of the detergents utilized impact the resulting particles generated, with important considerations being those impacting their critical micelle concentration (CMC). Detergents and lipophilic components may initially be combined in organic solvents, rotary evaporated to produce a detergent/lipid film, and then rehydrated in aqueous buffer to generate mixed micelles. Alternatively, detergents may be directly added to MLV preparations to disrupt these structures and produce mixed micelles 33 The differentiating factor between the many techniques that fall under this category is the exact method used to eliminate the detergent. Common approaches include dialysis, gel chromatography, and 11  absorption. Dialysis involves dilution of the mixed micelles to near or below their CMC, resulting in gradual removal of the detergents and subsequent promotion of liposome formation 34. Gel chromatography involves passing the mixed micelles through a column matrix such as Sephadex 4B, G-50, or G-200. This results in sequential removal of the detergent as the mixed micelles pass through the column, generating liposomes before they reach the bottom of the matrix 33,35. This method of removal is highly efficient in eliminating residual levels of detergent when compared to dialysis. Finally, absorption techniques utilize synthetic beads coated with hydrophobic resins that bind and remove detergents from solution. These beads may be used to generate a detergent exchange column matrix, which then has the mixed micelle solution passed over it 36.   Relatively recent advances in liposome production technology have yielded techniques that require neither the high pressure of homogenization nor laborious detergent removal. In 2002, a novel “heating method” was reported, capable of generating uniform liposomes encapsulating DNA 37. This technique hydrates each lipid component separately and then rapidly combines them at high temperatures (120 °C) in the presence of 3% glycerol. This technique does not require any subsequent processing step, as the glycerol used is biocompatible and has no deleterious effects on the final liposomes. Additionally, due to the high temperatures used in manufacturing the particles, the solution generated is largely sterile and appropriate for in vitro or in vivo studies without additional filtering 38. Another recently developed technique is microfluidic mixing, which uses a staggered herringbone mixing device with microscale geometry that promotes rapid, millisecond-scale mixing of two input streams. When lipids dissolved in ethanol in one stream are mixed with the aqueous buffer in the second stream, the lipids experience a rapid increase in polarity that 12  promotes liposome formation. This allows for the production of limit-size liposomes in the range of 20-50 nm 39-41.   Overall, a diverse range of techniques have been developed and thoroughly optimized to produce uniform liposomes of a variety of sizes and compositions. Such advances have enabled the use of these systems as potent therapeutic delivery vessels. The technology for loading therapeutics into liposomes has developed alongside production methods and a similarly broad range of methods has been developed for encapsulating a variety of drug types.    1.1.4 Techniques for encapsulating therapeutics within liposomes   One of the most important considerations in drug loading of liposomes is the physiochemical nature of the compound to be entrapped. Traditionally, hydrophilic molecules have shown a much greater degree of compatibility with such loading than hydrophobic ones. This is largely due to the difference in bilayer permeability of these two drug categories. Since hydrophilic drugs are comparatively less able to pass through the lipid bilayer, they can remain effectively entrapped within the liposomes’ aqueous core. Hydrophobic molecules are more membrane-permeable and are better able to escape the liposomes, resulting in unstable loading and low encapsulation efficiencies. In spite of these challenges, a variety of techniques have been developed that achieve encapsulation of both types of molecules.   The first demonstrated techniques for generating drug-loaded liposomes took advantage of passive loading methods. These rely on incorporation of the therapeutic into either the lipophilic or aqueous components during initial liposome formation. Thus, as the 13  liposomes are formed, the drug is passively incorporated into the interior of the particle along with the aqueous buffer. However, since these techniques have no method of preferentially accumulating drug within the particle, they typically suffer from low encapsulation efficiencies, ranging from 5-60% 42. An exception exists in passive entrapment using freeze-thaw techniques at high lipid concentrations, which have shown efficiencies of over 85%, though this percentage was reported for entrapping only radioactive sodium 43. In the event of passive entrapment of charged drugs, oppositely charged lipids can be utilized to improve encapsulation efficiencies. This was demonstrated with doxorubicin (positively charged) and liposomes incorporating phosphatidylserine (PS) or cardiolipin (negatively charged) 44.  A key turning point in drug encapsulation was the discovery of active loading techniques, which utilize pre-formed liposomes with specific characteristics that promote accumulation of the drug within their aqueous interior. This was first demonstrated using liposomes exhibiting a pH gradient, with an acidic interior and neutral exterior, to load weak base molecules. Though this technique was first used to study the response of model membranes to pH gradients 45,46, it was quickly adopted for loading weak base drugs into liposomes 47,48. By using this technique, it was possible to obtain a 100-fold increase in the concentration of doxorubicin on the interior of the liposomes relative to the exterior concentration 48.   The generation of pH-gradient liposomes can be achieved in a number ways. The most simplistic method is to initially produce liposomes with an acidic aqueous component,  making particles with an acidic interior and exterior. Subsequent replacement of the exterior medium with a neutral buffer, through techniques such as dialysis, will generate liposomes with the desired pH gradient. When such liposomes are combined with a weak base molecule 14  with a pKa that is higher than the particles’ internal pH, spontaneous influx of the drug into the liposomes occurs. When present in the neutral exterior buffer (with pH greater than or near to its pKa), the molecule exists in equilibrium with a significant proportion present in its uncharged state. In this neutral form, the drug is more readily able to pass through the liposomal bilayer into the interior aqueous core. Upon doing so, it encounters the acidic environment at a pH lower than its pKa and becomes protonated. Due to this introduction of one or more charged groups, the weak base becomes unable escape the particle by passing back through the lipid bilayer and is effectively trapped. Loading in this manner can continue until the pH gradient is exhausted and insufficient protons are available to protonate newly arrived neutral weak base species 49.  A more involved method of pH gradient generation is through the use of ionophores, chemical species that act to exchange ions across membranes. If liposomes can be generated with an ionic gradient, the introduction of an ionophore that exchanges the ions for protons will enable efficient acidification of the internal compartment. This strategy has been effectively employed using a variety of ionophores, including nigericin (H+/K+ antiporter) and A23187 (H+/divalent cation antiporter) 50,51.   The pH gradient can also be established by preparing liposomes with an ammonium sulfate gradient (or any other ammonium salt). This method of active loading has been used with great success to load weak base therapeutics into liposomal systems approved for use in humans, including Doxil and Marqibo 52. Using this method, liposomes are prepared using ammonium sulfate as the aqueous component and then dialyzed against a neutral buffer such as PBS, generating particles with ammonium sulfate present only in the interior aqueous compartment. Once established this gradient converts into a pH gradient that replenishes 15  itself upon drug loading. The ammonium ions (NH4+) present in the liposome interior naturally exists in equilibrium with ammonia (NH3) and protons (H+). Since NH3 is much more membrane permeable, it can exit the liposome while the proton remains entrapped, acidifying the interior space. When a weak base drug is introduced into the solution, it collects within the acidified interior as detailed previously. By doing so, the intra-liposomal protons are consumed, promoting further dissociation of the remaining ammonium ions and enabling additional loading of the drug. This cycle repeats until the ammonium gradient is exhausted and high drug:lipid ratios are achieved 49. This method offers additional benefits in that the counter-ion to ammonium (sulfate in the above description) can enhance the stability of the encapsulated drugs. This can be achieved in one or both of two ways. First, the anion can complex with the protonated weak base to form a precipitate, effectively decreasing the intraliposomal concentration of soluble weak base and preventing associated osmotic stress. Secondly, the presence of the anion can contribute to an internal aqueous environment that decreases the membrane/buffer partition coefficient of the weak base, decreasing its ability to escape across the liposome membrane 53. The ability of the counter-ion to stabilize the loaded drug varies depending on the nature of both the species being loaded and the anion itself. As such, loading has been examined using a variety of ammonium salts such as citrate, phosphate, and sulfate 54.   1.1.5 Liposome interactions with biological components in vivo  The study of the pharmacokinetics and biodistribution of liposomal systems following administration in vivo has evolved in parallel with the technologies for producing and loading the particles. Key advances in the understanding of how the carriers interact with 16  the immune system, the inherent architecture of various organs, and blood components have allowed for increasing control over where particles accumulate post-injection and how quickly they release their therapeutic payload.   The primary challenge for utilizing liposomes in vivo is their strong interactions with the body’s immune system. Although the particles are typically composed of biocompatible and natural lipids, they are still foreign entities introduced into the circulation, which are exactly what the immune system has evolved to combat. As such, it comes as no surprise that most traditional liposomes are rapidly taken up by the body’s macrophages and other phagocytic cells upon injection 55. These cells are largely found in the mononuclear phagocyte system (MPS), which consists of the body’s phagocytic cells found in the liver, spleen, bone marrow, and lymphatic system. The aforementioned organs typically show the greatest degree of liposome accumulation post-injection. The highest particle concentrations are typically found within the liver (which comprises approximately 90% of the MPS 56) and, to a lesser extent, the spleen 57. Though these sites contain extremely high densities of phagocytic cells, their unique anatomy also contributes to the observed nanoparticle accumulation. In the case of the liver, the organ’s capillaries contain full fenestrations that allow for direct diffusion or extravasation of nanoparticles from the circulation into the perisinusoidal space. These fenestrations typically range from ~100-150 nm and completely lack a basement membrane, allowing for uninhibited diffusion 58. Following extravasation from the capillaries, nanoparticles are readily available to the phagocytic Kupffer cells of the liver 59. The spleen also exhibits enhanced potential for nanoparticle diffusion from the circulation, with a portion of its vasculature opening directly into its parenchyma. Nanoparticles that exit the capillaries through this route are exposed to the organ’s interior 17  red pulp, which contains high amounts of phagocytic cells involved in erythrocyte recycling and pathogen clearance 60.   It should be noted that liposome uptake by phagocytic cells, regardless of their location in the body, is typically preceded by interactions between the liposomes and proteins in circulation. Opsonization, the binding of a diverse set of serum components to the surface of liposomes or other foreign bodies as part of immune response that promotes their phagocytic clearance, is a well-established factor in particle clearance in vivo. The serum components involved in this process include apolipoproteins, immunoglobulins, and complement proteins, among others 61. In studying the general amount of total proteins bound to the liposome surface (Pb or Protein Binding Parameter) in vitro, inverse correlations between this value and circulation clearance times have been found for systems composed of phosphatidylcholine (PC) and Cholesterol (Chol) 62. Similar trends have been found for each subset of serum components that contribute to opsonization. Regarding apolipoproteins (Apo), various forms of Apo A, B, C, and E have been shown to associate with liposomes, contributing to MPS and liver uptake in two ways. Firstly, they are thought to act in the opsonization of the particles, supporting uptake via macrophages. Additionally, adsorption of increasing levels of Apo B and E have been correlated to increased liposome uptake in hepatic cells, likely through receptor-mediated endocytosis 63. Similar trends have been observed for immunoglobulins, with antibody coated liposomes showing significant increases in macrophage uptake in vitro and in vivo 64. In the case of complement activation, studies have shown that liposomal membranes of varying composition are capable of binding complement proteins and their fragments 65. Increased binding of complement proteins, specifically C3, has been observed in anionic liposomes composed of PC/Chol/cardiolipin 18  that show rapid clearance from the body. In contrast, neutral liposomes composed of 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC)/Chol show significantly lower levels of C3 binding and exhibit extended circulation times 64.   Overall, though the rapid accumulation of traditional liposomes in the MPS is beneficial for therapies targeting the immune system, it is a distinct challenge for formulations that need to reach other cell types or parts of the body. As such, a great deal of effort has been devoted to characterizing ways to evade the MPS. Specifically, studies establishing which liposome characteristics that most strongly affect interaction with plasma components and MPS uptake have been crucial in understanding nanoparticle clearance.    1.1.6 Liposome characteristics influencing blood clearance   Liposomes can be generated from a vast range of potential lipid components, which ultimately allow for the production of particles with a diverse set of physiochemical characteristics. Through extensive research, a subset of characteristics that have the largest impact on blood retention times and MPS clearance have been identified. These include particle charge, size, membrane fluidity, and surface coating.   The inclusion of charged lipids into liposomes can distinctly affect their circulation halftimes, largely through influencing the amount of plasma proteins bound to the nanoparticle surface. Positively charged liposomes show high Pb values, indicating a high degree of opsonization upon injection 66. Furthermore, varying indications of toxicity have been reported for cationic liposomes in vitro and in vivo. Incubation of positively charged liposomes with plasma results in increased turbidity and formation of a clot-like structure, while incubation with erythrocytes leads to limited hemolysis 66. Injection of cationic 19  liposomes in rats has also led to increased DNA strand breakage and cytokine expression in the liver and spleen, though the morphology and clinical biomarker levels related to the health of these organs were unaltered 67. Incorporating anionic lipids into liposomes has also been shown to decrease their circulation times, though the extent of this effect varies depending on the negative lipid species used 68. Though introduction of cardiolipin, phosphatidic acid, or phosphatidylserine leads to drastic reductions in circulation half-life, similarly negatively charged particles made with phosphatidylglycerol or phosphatidylinositol show relatively moderate reductions in retention times.   Particle size was one of the first evaluated parameters affecting blood retention, with numerous studies since 1975 showing that, as a general rule, smaller particles are cleared more slowly than large ones 69,70. Though this trend was originally observed when comparing larger MLV to smaller LUVs, it has remained valid for nanoparticles of identical degrees of lamellarity and lipid composition generated at different sizes. Specifically, levels of particle opsonization decrease in a manner proportional to liposome size for diameters of 800-200 nm 71. For particles ranging between 200-70 nm, previous studies have reported no major changes in clearance rate. However, particles smaller than 70 nm showed increased liver clearance 72. It was noted that these trends changed based on lipid composition, with more negatively charged particle showing decreased size-dependent clearance. This reinforces that size is just one contributor to a complex suite of factors determining particle clearance.  In particles of comparable surface charge and size, the fluidity of the lipid bilayer has been shown to impact circulation profiles as well. Fluidity is affected by several factors, some of the most prominent being lipid chain length, degree of acyl chain saturation, and cholesterol content. In liposomes composed purely of phosphatidylcholine, longer lipid chain 20  lengths and increased acyl chain saturation are associated with increased clearance rates and association with plasma membranes. More specifically, liposomes with membranes in the liquid crystalline state (injected at temperatures above their phase transition temperature) show increased circulation while those with gel-state membranes show a sharp decrease in this value. Increased membrane fluidity obtained through using shorter or unsaturated acyl chains both result in similar increases in blood circulation times and decreases in protein binding 62. The study also showed that the introduction of cholesterol to liposomes with gel-state membranes results in dramatic decreases in protein binding and increases in circulation time. In membranes with high transition temperatures, such as those consisting of lipids with long and saturated acyl chains, cholesterol functions to increase membrane fluidity. By doing so, it is thought to decrease the inherent packing defects present in gel-state lipid membranes, which are hypothesized to promote increased interaction with plasma components involved in opsonization. Indeed, when imaged through freeze-fracture microscopy, pure DSPC liposomes show angular fracture planes indicative of non-spherical morphology, likely caused by the aforementioned packing defects of the gel-state membrane. In contrast, introduction of 45% cholesterol content (mol:mol) to the same liposomes results in smooth, spherical liposomes without any apparent imperfections 73.  Due to the clear link between plasma component absorption and decreased blood retention times, it was hypothesized that introduction of a surface coating on a liposome could help decrease these unfavorable interactions. The first implementation of surface coating occurred in 1987, where liposomes were generated with an exterior layer of sialic acid 74. This was accomplished by incorporating the ganglioside GM1 into the lipid composition of the particles. Inspired by the naturally occurring sialic acids found in the 21  extracellular matrix of cell membranes, gangliosides were hypothesized to be a biocompatible yet effective method of preventing opsonization. By optimizing the concentration of GM1 incorporated, a 10-fold increase in circulating levels of PC/Chol liposomes was obtained. Substitution of sphingomyelin for PC produced even further increases in circulation time and a decreased distribution to the liver and spleen. Building on this first demonstration, rapid improvements were made in the field of surface coating. The most significant of these was the discovery of the shielding effects of polyethylene glycol (PEG) when used to decorate the liposome surface. Identified as a highly hydrophilic synthetic polymer that could shield the liposome surface from protein interaction, PEG was initially linked to several lipophilic moieties to allow it to stably incorporate into the liposome bilayer. Stearyl-PEG showed relatively moderate improvements to circulation half-time but conjugating the polymer to variations of phosphatidylethanolamine (PE) lipids proved to be an effective route for introducing stable PEG shielding 75,76. Using this strategy, PEG coatings allowed for far improved shielding efficiency, with 5-7% PEG- distearoylphosphatidylethanolamine (DSPE) particles showing similar circulation kinetics as 10% GM1 liposomes 77. Furthermore, it was shown that by increasing the amount of PEG-lipid included in a formulation, the circulation lifetime could be extended even further, allowing for tight control over formulation pharmacokinetics 75. Successful surface shielding using PEG or other polymers allowed for the production of “stealth” liposomes that exhibited far longer circulation lifetimes and decreased MPS uptake compared to traditional vessels. These qualities allowed for new biodistribution profiles to be obtained and opened new avenues for disease treatment. By avoiding substantial and rapid uptake by the MPS, PEGylated particles can accumulate within sites of tumor growth or inflammation. This is 22  largely due to the enhanced permeability and retention (EPR) effect, which refers to the observed increase in blood vessel permeability and decrease in lymphatic drainage in tumor or inflammation sites 78. This discovery was crucial in the development of numerous liposomal formulations for cancer treatment and demonstrations of increased treatment efficacy through use of PEGylated liposomes quickly followed.   1.1.7 Liposome characteristics influencing drug release rates   With the major challenge of MPS evasion addressed by surface modifications, the next major challenge to efficacious liposomal therapeutics is controlling drug release rates. Major factors influencing release rates from loaded nanoparticles include interactions with plasma components, particle size, and drug precipitation. It should be noted that a formulation’s drug release rate and the degree to which the above factors influence it are highly dependent on the nature of the encapsulated drug.   Many of the liposome-plasma component interactions described in the preceding section will contribute not only to the clearance rate of the particles but also their rate of drug release. For example, the insertion of complement proteins and other opsonins can compromise membrane integrity, resulting in greater drug leakage. In extreme cases, complement activation can result in to formation of a membrane attack complex, which functions much like a pore in the membrane that facilitates content efflux 79. Lipoprotein interactions, including those with LDL and HDL have also been shown to increase drug leakage rates upon in vitro incubation 80. In addition to destabilization associated with apolipoprotein exchange between endogenous lipoproteins and liposomes, direct lipid component exchange between these two species has also been demonstrated. Pure 23  phosphatidylcholine particles injected into mice show a time dependent increase in cholesterol content, with the majority of the lipid arising from lipoprotein donation. This exchange occurs until there are roughly equimolar concentrations of phospholipid and cholesterol in the particles 81. As such, inclusion of cholesterol in the initial liposome composition can help to avoid any membrane disruption associated with lipid exchange.   Particle size has also shown to be an effector of drug retention in liposomes. Smaller liposomes exhibit higher degrees of membrane curvature, which is hypothesized to induce structural strain preventing ideal lipid packing. Due to these induced packing defects, the membrane becomes leakier overall 30,82.   As previously stated, the physiochemical properties of the encapsulated drug will greatly influence drug release rates. In particular, the ability of the drug to precipitate on the interior of the liposome will typically lead to significant increases in overall retention. A number of drugs have been observed to precipitate at suitably high intraliposomal concentrations, as indicated by distinct structures visualized using cryo-TEM. For example, doxorubicin forms long, bar-shaped precipitates resulting in a “coffee-bean” like structure observed in loaded liposomes 54 whereas vincristine shows an amorphous precipitate appearing as distinct spots of increased electron density 83. In both cases, this precipitation has been hypothesized to contribute to the stability of drug retention in these systems. In contrast, drugs that do not show precipitation typically exhibit rapid drug release upon plasma incubation in vitro or in vivo. For example, the antibiotic ciprofloxacin shows extremely rapid release rates upon introduction to plasma components when loaded into liposomes 84. In spite of the drug being concentrated to twice its solubility limit in the liposomal interior, no precipitation is observed, resulting in extremely high intraliposomal 24  concentrations of the soluble drug. Since drug leakage is proportional to the concentration of soluble drug on the liposome interior (according to Fick’s law), this allows liposome-encapsulated ciprofloxacin to respond very quickly to any membrane destabilization affecting permeability, facilitating rapid drug release. In contrast, liposomes with drug precipitations have a significant portion of their encapsulated content present as insoluble precipitates, which exists in equilibrium with the soluble drug. This effectively lowers the concentration of soluble drug in the liposomal interior, resulting in more stable retention and less responsiveness to membrane destabilizing effects 83.   Overall, the work done to characterize how these factors influence drug release rates have been a major part of translating liposomal therapeutics from the bench to pre-clinical experiments to clinical applications in humans. Together with an understanding of how liposomes can be produced, mechanisms for loading them with drug, and how they behave upon injection, this knowledge has led to the successful generation of numerous FDA-approved liposomal formulations.   1.1.8 Application of liposomal therapeutics in cancer treatment   As discussed previously (Section 1.1.2 History of liposomal nanoparticle development), lipid-based therapeutics have received FDA approval for a broad range of applications. A large proportion of these are focused on cancer therapy. Doxil, one of the first FDA approved liposomal therapeutics, encapsulates doxorubicin and allows for drastic increases in therapeutic efficacy compared to the free drug through several mechanisms. Firstly, liposomal encapsulation allows for a more desirable biodistribution profile, greatly increasing doxorubicin tumor accumulation 85. This is possible due to the inclusion of PEG 25  2000-DSPE, allowing for surface shielding and tumor accumulation via the EPR effect. This was one of the first demonstrations of successful “passive” targeting of liposomes to tumor sites in humans. Specifically, passive targeting refers to accumulation at a target site through naturally occurring processes as opposed to the inclusion of targeting agents on the liposome surface. Examples of natural processes influencing passive targeting include the aforementioned EPR effect, post-injection adsorption of apolipoproteins, and the inherently fenestrated vasculature of the liver or spleen. By enabling passive tumor targeting via stealth liposome technology, Doxil was able to increase not only the amount of drug that reached the tumor site but also the duration of drug exposure the tumor experiences 86.  The second way in which Doxil improves on the efficacy of free doxorubicin is through significant reduction of adverse effects and toxicity. Free doxorubicin exhibits dose-limiting cardiotoxicity, with induced cardiomyopathy being one of the most common effects. This prevents the free drug from being administered at high concentrations or in long-term regimens 87. In contrast, Doxil exhibits dramatically reduced cardiotoxicity. This is largely due to its ability to 1) Decrease the acute peak plasma concentration of doxorubicin following free drug injection; and 2) Limit the biodistribution of the drug to cardiomyocytes 88.  Together, the ability of Doxil to improve the pharmacokinetics/biodistribution of doxorubicin and decrease its inherent toxicity allow it to serve as a greatly enhanced treatment for numerous cancer types 15. In AIDS-related kaposi sarcoma, Doxil treatments enhance drug accumulation in skin lesions and allow for reductions in total doxorubicin dose while maintaining efficacy. When used to treat breast cancer, the liposomal therapeutic is similarly efficacious to free doxorubicin but the large decreases in cardiotoxicity make it a much safer alternative. Additionally, when used in combination chemotherapy regimens, 26  Doxil has been shown to result in reduced severe immune response to repeated treatments. Various types of solid tumor cancers have shown similar improvements in effectiveness and decreases in toxicity when treated with Doxil, including ovarian cancer, sarcoma, brain cancers such as glioma, prostate cancer, and head and neck cancer 88.   The general principles resulting in increased efficacy compared to free drug observed with Doxil, namely improved pharmacokinetics/biodistribution and decreased toxicity, form the basis for the utility of liposomal encapsulation of many other cancer therapeutics. DaunoXome consist of daunorubicin encapsulated in DSPC/Chol liposomes. While the formulation does not contain PEG-lipid, it still results in significantly increased circulation time and shows a reduction in cardiotoxicity and myelosuppression when compared to free drug 89. This is possible due to the relatively small size of the particles (~45 nm) and the inclusion of cholesterol to extend its plasma half-life. Non-PEGylated liposomal formulations like DaunoXome have been shown to be effective in treating numerous cancer types. Myocet, a non-PEGylated formulation of doxorubicin (egg PC/Chol), has shown to be effective in treating breast cancer when implemented as part of combination chemotherapy strategies. In spite of its lack of surface shielding and its larger particle size, Myocet shows similar trends in reducing myelosuppression, gastrointestinal side effects, and cardiotoxicity while maintaining anti-tumor efficacy 90,91.   As demonstrated by the previously described formulations, liposomal therapeutics show great promise in the treatment of solid tumor cancers. Of course, this is partially due to their ability to decrease toxicity but also to their ability to accumulate drug in the target tumor sites via the EPR effect. Nonetheless, liposomes have also shown utility in treating other cancer subtypes that do not primarily manifest in traditional solid tumors. For example, 27  the chemotherapeutic vincristine, a vinca alkaloid that functions in preventing cell division via inhibition of microtubule formation, has shown favorable encapsulation in sphingomyelin (SM)/Chol liposomes. This formulation is marketed as Marqibo and has primary applications in acute lymphoblastic leukemia (ALL). Since ALL is a hematological cancer, it does not immediately form solid tumors, thus preventing liposomal therapeutics from taking advantage of the EPR effect. Despite this, clinical trials have shown Marqibo to be effective in treating Philadelphia chromosome-negative ALL 92. As expected, when compared to free vincristine, Marqibo allowed for higher drug doses to be regularly administered without any subsequent increases in toxicity or side effects.  Furthermore, the liposomal therapeutic showed meaningful clinical outcomes in patients that had relapsed after previously receiving multiple anti-cancer therapies.   Liposomal formulations have also been developed for the treatment of other forms of leukemia. Acute myeloid leukemia (AML) has recently had a breakthrough liposomal treatment approved by the FDA in 2017. Marketed as Vyxeos, this is a liposomal formulation of daunorubicin and cytarabine 93. These are two traditional chemotherapeutics that target distinct pathways in cancer progression. Daunorubicin acts via topoisomerase II inhibition and subsequent DNA strand breakage while cytarabine is a toxic nucleoside mimic. Currently, the combination of daunorubicin and cytarabine is the standard of care treatment for AML. Delivered together, they obtain synergistic therapeutic effects. However, due to their unique physiochemical profiles, they are cleared at different rates from the body, preventing optimal drug:drug ratios from being maintained at target sites. Furthermore, as with any unprotected chemotherapeutic, they exhibit off-target effects including toxicity in cardiac, myeloid, and gastrointestinal cells 94. Encapsulation of both agents allows for not 28  only the standard bioprotective effect through altered biodistribution but also the ability maintain the synergistic daunorubicin:cytarabine ratio that is most effective in treating AML. Furthermore, though no traditional solid tumors are present to exhibit the EPR effect, the sinusoidal vessels of the bone marrow, in which a proportion of the leukemic blast cells reside, allows for increased accumulation of the nanoparticles. As such, Vyxeos allows for increased delivery of both drugs to the diseased sites at the optimal drug:drug ratio 95. When tested in phase III clinical trials, Vyxeos provided significant increases in overall survival, decreases in 60-day mortality, and improvements in a number of other disease-related endpoints compared to standard combination therapy with the two free drugs 96.   Overall, liposomal therapeutics are well suited to treat both cancers that manifest solid tumors as well as others that do not. Their inherent ability to prolong circulation lifetimes of their payload and decrease associated toxicity impact almost all liposomal cancer therapeutics. The ability to facilitate accumulation in solid tumors and disease sites of hematopoietic cancers provides an added benefit to these nanoparticle therapies. In particular, the development of LNP formulations to treat numerous types of leukemia are providing promising new avenues for combatting this disease, particularly in its most ill-addressed subtypes including AML.   1.2 Acute myeloid leukemia and its treatment 1.2.1 Overview of leukemia  Leukemia is a hematological cancer that arises from oncogenic mutations in immature blood cell progenitors produced in the bone marrow. These progenitor cell populations are separated into two general categories: myeloid and lymphoid progenitors. 29  This distinction arises from the cell types the stem cells ultimately mature into: myeloid progenitors differentiating to monocytes, granulocytes, erythrocytes, and megakaryocytes; lymphoid cells into B and T cell. As such, leukemic mutations can lead to two corresponding categories of leukemia depending on which of these progenitor types becomes cancerous: myeloid leukemia and lymphocytic leukemia 97. Additionally, depending on the aggressiveness or rate at which the disease progresses, each of these two cancer types can be classified as acute or chronic. Due to the development of oncogenic mutations, leukemic blast cells become arrested in their immature state, prevented from normal maturation into their fully functional blood cell varieties. As with any cancer, they also become capable of rapid and uncontrolled replication 98. Due to their unregulated division, the non-functional leukemic cells eventually crowd out normal blood cells in all their stages of development. This results in impairment in blood cell production and function. As such, leukemic patients commonly experience inhibited blood clotting (coagulopathy) and compromised immune function due to insufficient platelet and white blood cell (WBC) production. As the disease progresses, the spread of leukemic blasts out of the bone marrow and into circulation can directly or indirectly lead to increased risk of hemorrhage, vulnerability to severe infection, and blood viscosity 99. Additionally, the cancer cells can easily travel to other organs and manifest metastases or interfere with normal organ function due to their immense quantities 100. This can lead to compromised organ function and eventual organ failure in untreated patients 101.   The treatment of leukemia has seen many important advances since the disease was first reported in the 19th century. Early treatments using relatively crude and non-specific techniques showed abysmal success rates and were capable of inducing only short-lived 30  decreases in the number of circulating blast cells. These included arsenic solutions (1860 and onwards) and unrefined radiotherapy (1890s) 102,103. The first complete, though temporary, remission of the disease was obtained in pediatric ALL patients treated with aminopterin in 1948, though these individuals largely experienced subsequent relapse of the cancer 104. The two decades following this breakthrough showed marked discoveries of new chemotherapeutic agents and the implementation of combination chemotherapy regimens, which utilize two or more anti-cancer agents together for synergistic effect 105. By combining prednisone, vincristine, methotrexate, and 6-mercaptopurine, a combination regimen producing the first lasting remissions in ALL patients was discovered in 1965 106. Since this discover, many new combination chemotherapy regimens have been designed and tested, with some form of chemotherapeutic mixture typically serving as the standard of care for most leukemia subtypes. This modern approach to disease treatment has provided very promising long-term remission and cure rates for certain subtypes of the disease, with pediatric ALL serving as one of the most treatable forms. As shown in Table 2, combination chemotherapy treatments in this form of leukemia produce long term survival rates of over 90% 107. However, other subtypes of the disease show relatively low treatment success, with AML standing as one of the most poorly addressed subtypes in terms of long-term and median survival statistics 108. As such, the need for innovative treatments for the disease is still high, as contemporary therapies are insufficient in providing versatile cure rates.      31  Table 2. Leukemia subtypes and associated survival statistics.  Subtype Long-Term Survival (%)  Median Survival (Years)  Pediatric Acute Lymphocytic >90 >10 Chronic Myeloid >80 >10 Adult Acute Lymphocytic 40-50 >3 Chronic Lymphocytic 30-40 6-7  Acute Myeloid 30-40 1-2    1.2.2 Treatment of acute myeloid leukemia   As mentioned previously, AML is an aggressive subtype of the disease arising from myeloid blast cells and is the most poorly treated of the four classic leukemia subtypes. Though the treatment of AML has shown some success in obtaining steady remissions in younger patients, reliable and effective therapies for the more prevalent older patient populations are still lacking 109. Combination chemotherapy of cytarabine and an anthracycline chemotherapeutic (typically daunorubicin) currently serves as the standard of care for AML. This treatment is referred to as “7+3”, indicating the dosing schedule of seven days of continuous cytarabine infusion with three concomitant days of anthracycline injection at the start of treatment 110. The continued use of the 7+3 treatment is somewhat discouraging, as this strategy was first tested in 1973 111 and does not result in high long-term survival or cure rates as seen in Table 2. Numerous variations on this strategy including testing the effects of different anthracyclines (such as idarubicin) and adjusted drug:drug ratios have shown little improvement 108,112. Because of the high doses of the free drugs 32  necessary and the associated toxicities, this treatment strategy is particularly dangerous for older patients 109. As detailed previously (Section 1.1.8 Application of liposomal therapeutics in cancer treatment), major improvements in the tolerability of various chemotherapeutics have been made through liposomal encapsulation. As such, it is not surprising that this strategy has been applied to cytarabine/daunorubicin therapies. Vyxeos, a liposomal formulation of these drugs shows increased treatment efficacy by delivering the drugs at the synergistic ratio while also reducing toxicity. Though these innovations have allowed for significant improvements in two year survival rates (31.2% as opposed 12.3% for the free drug combination regimen) this long term survival percentage is still quite low in comparison to other leukemia subtypes 93. Overall, though major improvements have been made to existing chemotherapeutic regimens, these strategies still ultimately rely on the same combination strategies that have been used since their first application to AML in 1973. Even with state of the art drug delivery systems, cytarabine/anthracycline regimens are unable to produce long term survival rates above 50% in the most common AML patient populations. As such, there is a clear need for new and more effective AML treatment strategies.   1.2.3 Epigenetic regulators and their application to AML treatment  In recent years, the importance of epigenetic regulation in the manifestation and progression of leukemia, including AML has become evident. Key chromatin modifying enzymes have become appealing therapeutic targets, as they have been shown to play a role in the development of the two mutation pathways that are critical for leukemia development: arrest of differentiation and rapid replication. Of particular interest are the histone 33  methyltransferases G9a and GLP. In addition to playing a role in normal hematopoietic pathways in healthy cells, they have been shown to be vital to the progression of numerous types of leukemia 113. In vitro studies show that deletion, active site mutation, or inhibition of G9a results in significantly reduced leukemic cell replication rates. Additionally, inhibition of the enzyme results in clear signs of increased myeloid differentiation in the treated cells. Taken together, these two findings indicate that G9a inhibition can address both the arrested development and rapid replication characteristics shown to be key for leukemia manifestation and progression.  Though these in vitro results were initially obtained using murine leukemic cells, similar results were soon obtained using a variety of human cell lines. This demonstrated the wide applicability of G9a inhibition across different species as well as different oncogenic variants of AML. The effectiveness of G9a inhibition was also demonstrated in vivo. Mice inoculated with G9a knockout leukemic cells showed significantly improved survival rates, as did those inoculated with cells pre-treated with G9a inhibitors 114. In light of these promising results in cell lines and animal models, G9a/GLP inhibition presents an encouraging therapeutic target for the treatment of AML. To this end, several G9a/GLP inhibitors have been iteratively developed to maximize their potency.  A promising series of small molecules have been developed by Dr. Jin’s group at the University of North Carolina (UNC), with the inhibitors UNC0638 and UNC0646 tested against leukemic cells in the aforementioned in vitro and in vivo G9a inhibition experiments 114. This line of inhibitors was originally developed using structure-function predictions based on structural data of G9a’s catalytic site, which resulted in the discovery of BIX-01294 115. Based on this initial scaffold, iterative chemical screens were used to further develop the 34  inhibitors and optimize their in vitro and in vivo properties. This resulted in the development of UNC0638 and UNC0646 116. Recently, an even more potent iteration of the UNC compound line has been developed, with a G9a IC50 (the dose to obtain 50% inhibition of activity) below 2.5 nM 117. This small molecule, UNC0642, thus serves as a prime candidate for further G9a inhibition studies in the context of AML treatment. However, though the molecule shows promising in vitro efficacy, efforts to apply it to animal models have been compromised due to an inherent acute toxicity associated with the drug (unpublished results).  Intraperitoneal (IP) injection of the drug at 5 mg/kg results in a severe adverse neurological response in the form of seizures. Though animals can recover after a single dose, a second injection within 24 hours leads to an increase in the severity of the acute toxic response observed, with animals reaching the human endpoint soon after due to severe seizure. Furthermore, even a single 5 mg/kg IV injection results in lethal toxicity. Overall, though UNC0642 stands as a promising anti-leukemic drug, its inherent toxicity prevents translation of the agent into pre-clinical studies.   35    Figure 1. Structure of the G9a/GLP inhibitor UNC0642  Developed by Dr. Jin’s lab at the University of North Carolina, UNC0642 represents one of the most potent small molecule inhibitors (G9a IC50<2.5 nm) developed by this group.   1.3 Thesis objectives  Liposomal nanoparticles have a well-documented history of enabling effective anti-cancer therapies by attenuating drug-related toxicity and controlling drug biodistribution (as reviewed in Section 1.1.8 Application of liposomal therapeutics in cancer treatment). As such, these systems are well suited to addressing the inherent toxicity of UNC0642.  The ultimate objective of this thesis is to is to examine the ability of liposomal encapsulation to decrease the inherent toxicity of UNC0642. To evaluate this, the following sub-objectives must be accomplished: 1) Optimization of an efficient loading strategy for encapsulating UNC0642 within liposomal carriers, generating UNC0642-LNP; 2) In vitro characterization of UNC0642-LNP, including structural, drug retention, and anti-NNOO NCH3NHNCH3H3CNFF36  proliferation cell culture studies; and 3) In vivo characterization of the effect of encapsulation on toxicity via tolerability studies using murine systems.  By systematically addressing these sub-objectives in turn, we show that UNC0642-LNP systems can indeed be generated and characterized. Furthermore, we show that liposomal encapsulation effectively addresses the inherent toxicity of the drug by comparing free vs liposomal UNC0642 injections in terms of their acute and chronic toxicity at varying doses. The capacity for longitudinal administration of the systems is also evaluated by determining the compatibility of these liposomal systems with multiple-dose regimens that mimic anti-leukemic therapy schedules. Overall, it is shown that UNC0642-LNP are a safe and promising anti-AML formulation that are well suited to future pre-clinical testing in animal models of the disease.     37  Chapter 2: Materials and Methods 2.1 Materials Phospholipids used for liposome preparation, including 1,2-distearolyl-sn-glycero-3-phosphocholine (DSPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) were purchased from Avanti Polar Lipids (Alabaster, AL). Cholesterol was purchased from Sigma-Aldrich (Saint Louis, MO). 1,1’-dilinoleyl-3,3,3’,3’-tetramethylindocarbocyanine perchlorate (DiI) was purchased from Invitrogen (Eugene, OR). UNC0642 was provided by the lab of Dr. Jian Jin (University of North Carolina).  Ammonium sulfate, Dulbecco’s phosphate buffered saline (PBS), and fetal bovine serum (FBS) were purchased from Sigma-Aldrich (Saint Louis, MO).   2.2 Preparation of liposomes with pH gradient  Powdered lipid stocks were combined and dissolved in anhydrous ethanol at 65 °C. MLV were generated from dissolved lipids through rapid addition 300 mM ammonium sulfate (65 °C) to a final concentration of 30 mg/mL total lipid and 10% EtOH. MLV were extruded at 65 °C at 350 psi through two stacked polycarbonate filters with 0.1 µm pore size for a total of 10 passes. Following extrusion, particles were dialyzed against 300 mM ammonium sulfate overnight to remove residual EtOH. Prior to drug loading (see Section 2.3 Remote loading of UNC0642 into preformed liposomes), particles were dialyzed against Dulbecco’s phosphate buffered saline (PBS, Sigma-Aldrich, Saint Louis, MO) overnight using 12-14 kDa regenerated cellulose membranes (Spectrum Labs, Rancho Dominguez, 38  CA). Cholesterol concentration of the particles was measured using the Wako Cholesterol E assay (Mountain View, CA) and used to determine the total lipid concentration.   2.3 Remote loading of UNC0642 into preformed liposomes   Prepared liposomes with pH gradient (see previous section) were combined with UNC0642 dissolved in PBS to a final concentration of 5 mM total lipid and an initial drug:lipid (molar) ratio of 0.1, 0.2, or 0.3. Loading mixture was incubated at 65 °C for 20 minutes before being dialyzed against PBS overnight to remove any unencapsulated UNC0642. Dialyzed particles were sterile-filtered using a 0.2 µm syringe filter (Pall, Ville St. Laurent, QB).   2.4 Characterization of UNC0642-LNP   Particle size and polydispersity index (PDI) were determined through dynamic light scattering (DLS) using the Malvern Zetasizer NanoZS (Worcestershire, UK). Reported values correspond to number mean diameters. Cholesterol concentration of the UNC0642-LNP particles was determined using the Wako Cholesterol E assay (Mountain View, CA) and converted to total lipid concentration (assuming a 55:45 molar ratio of phospholipid:cholesterol). The concentration of UNC0642 in the loaded particles was measured using absorbance at 250 nm (A250 in particles diluted to ~0.25 mM in 70% isopropanol). Encapsulation efficiency (percent encapsulation) was determined through comparison of the drug:lipid ratio of UNC0642-LNP pre- and post-dialysis to remove unencapsulated UNC0642 (see Section 2.3 Remote loading of UNC0642 into preformed 39  liposomes). Drug:lipid ratios were determined using the molar concentrations of UNC0642 and total lipid determined through A250 and Wako Cholesterol E assay, respectively.   2.5 Drug release assay of UNC0642-LNP incubated in PBS (37 °C)  POPC/Cholesterol or DSPC/Cholesterol particles diluted to a final concentration of 2.5 mM total lipid in PBS were added to 12-14 kDa dialysis membranes and incubated against a 2000-times excess of PBS at 37 °C. At the indicated time points (0, 2, 8, 24, and 48 hours), an aliquot of the incubated sample was removed from the dialysis membrane for analysis. Aliquots were analyzed to determine drug:lipid ratio as detailed in the previous section and percent retention calculated relative to the t = 0 hours time point.    2.6 Drug release assay of UNC0642-LNP incubated in 50% FBS (37 °C)    POPC/Cholesterol or DSPC/Cholesterol UNC0642-LNP (0.2 drug:lipid ratio) diluted to a final concentration of 2.5 mM total lipid in 50% FBS (in PBS) were added to 12-14 kDa dialysis membranes and incubated in 100 mL of 50% FBS (diluted in PBS) with constant stirring. 1 mL aliquots of the incubation solution on the exterior of the dialysis membranes was removed at 0, 2, 8, and 24 hour time points. After each aliquot was taken, 1 mL of 50% FBS in PBS was added to the incubation solution in order to keep the overall incubation volume constant. After the 24-hour time point aliquot was collected, the dialysis membrane was cut to fully release the UNC0642-LNP. An aliquot of this “full release” time point was collected for analysis as well. The concentration of UNC0642 in the collected aliquots was measured using ultra high pressure liquid chromatography (UHPLC) run on a Waters Acquity H-Class UHPLC instrument equipped with an Acquity photodiode array detector. 40  Prior to injection into the UHPLC instrument, samples were diluted by a factor of 4 with UPLC-grade methanol, vortexed for 30 seconds at maximum, let sit 5 minutes at room temperature, vortexed for 30 seconds at maximum and centrifuged (5 min/16 000 xg) to precipitate a portion of the serum proteins present. The supernatant of the centrifuged samples were run on an BEH C18 column at a flow rate of 0.306 mL/min, column temperature of 55 °C, and injection volume of 5 µL. The mobile phase used was comprised of a mixture of solvents A (0.1% acetic acid in water) and B (0.1% acetic acid in methanol). The ratio of solvents A:B proceeded in three stepwise linear gradients: 1) 100:0 to 50:50 (0-10 minutes); 2) 50:50 to 0:100 (10-12 min); and 3) 0:100 to 100:0 (12-15 minutes). Absorbance at 250 nm was used to quantify the concentration of UNC0642 in each sample through use of a standard curve generated in 50% FBS. Percent drug release at each time point was calculated relative to the amount of UNC0642 present in the full release aliquots.   2.7 Cryogenic transmission electron microscopy (cryo-TEM) of UNC0642-LNP  UNC0642-LNP (0.2 drug:lipid, molar) were concentrated (Amicon Ultra-15 Centrifuge Filter Units, Millipore, Billerica, MA) to a total lipid concentration of ~20 mg/mL prior to analysis. Prior to concentration, a 1:1 dilution in water for injection (WFI) grade water  was made to avoid vesicle deformation upon concentration. Concentrated samples were aliquoted (2-4 µL) onto glow-discharged copper grids and a FEI Mark IV Vitrobot (FEI, Hillsboro, OR) was used to plunge-freeze loaded grids to generate vitreous ice. Frozen samples were imaged using a FEI Titan Krios (FEI, Hillsboro, OR) at 300 kV and a Falcon III direct electron detector. A nominal under-focus of 1-2 µm at a magnification of 47 000x 41  was used to image samples with enhanced contrast. All samples were stored in liquid nitrogen prior to imaging at the UBC Life Sciences Centre (UBC, Vancouver, BC).   2.8 In vitro anti-proliferation assays  Assays were performed using HoxA9/Meis1 leukemic cells harvested from leukemic C57Bl/6 mice. Cells were originally generated through viral transduction as described previously 114. Growth media was composed of DMEM with FBS (15%), penicillin, streptomycin, L-glutamine, β-mercaptoethanol (1/100 000), and the following cytokines: rmSCF (100 ng/mL), rmIL-3 (10 ng/mL), rhIL-6 (10 ng/mL) (Stem Cell Technologies). Cells were plated in 6-well non-cell culture treated plates (Falcon/Corning Inc., Corning, NY) at a density of 50 000 cells/well approximately 4 hours prior to treatment. Either free UNC0642 or UNC0642-LNP (55:45 molar ratio of DSPC/Cholesterol, 0.2 drug:lipid ratio) dissolved in PBS were diluted as necessary with PBS and added to the appropriate volume of media to obtain final treatment concentrations of 0, 0.5, 1, 2, 4, or 6 µM UNC0642 (free drug) or 0, 15, 30, 60, 90, 120, or 150 µM UNC0642 (UNC0642-LNP). Free drug or UNC0642-LNP were diluted with PBS to 7.5% of the total treatment volume prior to further dilution with media in order to maintain a constant PBS:media volume ratio. The final per well treatment volume was 4 mL. Treated cells were incubated at 37 °C and 5% CO2 for a total of 48 hours. Post-treatment, cells were harvested, pelleted and resuspended in a total volume of 2 mL in fluorescent-activated cell sorting (FACS) buffer. An aliquot of this resuspension was diluted 1:1 with CountBright Absolute Counting Beads (Molecular Probes/Invitrogen, Eugene, OR) at an initial concentration of 1 million beads/ml prior to running on BD LSR II flow cytometer and analysis using associated BD FACSDiva 42  software. Cells were counted until a total of 1000 beads were recorded and final cell concentration calculated based on the ratio of cells:beads counted.  2.9 In vitro cell uptake assays   Assays were performed using ex-vivo HoxA9/Meis1 leukemic cells harvested from leukemic C57Bl/6 mice. In all treatments, empty DSPC/Cholesterol/DiI (55:44.8:0.2, molar) liposomes were diluted with PBS to 7.5% of the total treatment volume prior to further dilution with media in order to maintain a constant PBS:media volume ratio. Final treatment concentrations used were 0, 46.38, 92.76, or 463.80 µg/mL total lipid. Treated cells were incubated at 37 °C and 5% CO2 for a total of 48 hours. After treatment, cells were pelleted, washed with FACS buffer once, re-pelleted, and resuspended in FACS buffer containing PI stain. Cells were analyzed via FACS using a BD LSR II flow cytometer . Cells were analyzed for their PI and DiI signals. Live cells (PI-negative) were assayed for LNP uptake (DiI signal), with a total of 5 000 events collected per sample. Analysis was performed using FlowJo software.   2.10 Single dose in vivo tolerability tests  All tolerability tests used female C57Bl/6 mice aged 10-12 weeks at injection (Jackson Laboratory). Animals were injected with PBS, free UNC0642 dissolved in PBS, or UNC0642-LNP (DSPC/Cholesterol, 55:45 molar ratio, loaded at 0.2 molar drug:lipid ratio) in PBS. Free UNC0642 in PBS was sterile filtered using a 0.2 µm filter (Pall, Ville St. Laurent, QB). Injections were delivered intravenously (IV) through tail vein at a volume of 10 µL per gram of body weight. Doses were based on mg of UNC0642 (free or LNP-43  encapsulated) per kg animal body weight. Free UNC0642 injections were given at 5.0, 2.5, and 1.25 mg/kg. UNC0642-LNP injections were given at 5.0, 7.5, 10, 15, and 20 mg/kg. Following injection, animals were monitored closely for signs of acute toxicity for 30 minutes, monitored at 60 minutes, and again at 4-8 hours post injection. Animal health and body weight was measured/scored using the Animal Monitoring Clinical Score Reference and Recording Sheet for Drug Testing prior to injection and at all the above indicated timepoints. Animals were weighed and their health was scored daily for 14 days following injection to detect any long-term toxicity associated with a single dose. After the final health and weight measurement, animals were euthanized via CO2 inhalation and a necropsy performed to assess organ appearance. All procedures performed were approved by the Animal Care Committee of the University of British Columbia and were performed in accordance with guidelines established by the Canadian Council on Animal Care.   2.11 Multiple dose in vivo tolerability tests  Female CD57Bl/6 mice aged 10-12 weeks were injected with PBS, free UNC0642, or UNC0642-LNP using a dose regimen of 2 injections per week for 4 weeks. Free drug was injected at 1.25 mg UNC0642/kg body weight while the liposomal drug was administered at 5, 10, or 20 mg UNC0642/kg body weight. Injection details were identical to those of the single dose tolerability experiments (see Section 2.10 Single dose in vivo tolerability tests). Animals were monitored, weighed, and their health scored as detailed in the single dose tolerability experiments (see Section 2.10 Single dose in vivo tolerability tests). 14 days after the final injection, animals were euthanized via CO2 inhalation and their blood collected through cardiac puncture. Whole blood was transferred to BD Microtainer SST tubes 44  (Becton, Dickson, and Co., Franklin Lake, NJ), allowed to clot at room temperature for 1 hour, and centrifuged at 1300 xg for 3 minutes. Serum was isolated post-centrifugation and stored at 4 °C prior to blood chemistry analysis (Chemistry Screen CHMSCW and LDH tests, performed by IDEXX, Delta, BC). A necropsy was performed on euthanized animals to assess general organ status via macroscopic observation. All procedures performed were approved by the Animal Care Committee of the University of British Columbia and were performed in accordance with guidelines established by the Canadian Council on Animal Care.    45  Chapter 3: Design, Characterization, and Tolerability of UNC0642-LNP  3.1 Synopsis   Three stages of work are presented, corresponding to 1) Development and characterization of a loading strategy to generate liposomes encapsulated with UNC0642 (UNC0642-LNP); 2) In vitro characterization of the UNC0642-LNP including drug retention, structural, and cell culture studies; and 3) In vivo comparison of the tolerability of free vs liposome-encapsulated UNC0642 using single- and multiple-dose regimens.   UNC0642 was successfully loaded into DSPC/Chol and POPC/Chol liposomes using an ammonium sulfate/pH gradient technique to a maximum drug:lipid ratio (mol:mol) of 0.2. Both particle compositions showed stable retention of the drug upon in vitro incubation in PBS and 50% FBS at 37 °C with a maximum of ~10% and 20% drug release observed after 24 hours in particles containing DSPC or POPC, respectively. Cryo-TEM imaging of UNC0642-LNP showed evidence of drug precipitation within the liposomes, likely contributing to their stable release kinetics. In cell culture experiments, UNC0642-LNP expressed a dose-dependent uptake into HoxA9/Meis1 leukemic cells and a 10-fold decrease in anti-proliferative effect relative to the free drug. Single dose in vivo tolerability tests of UNC0642-LNP injection showed a complete elimination of the acute toxicity response associated with the free drug, even at doses 8-times higher than the maximum tolerated dose of free UNC0642 (20 mg/kg vs 2.5 mg/kg). Furthermore, no substantial long-term toxicity was observed following single injection of either free or encapsulated drug at these doses. 46  The UNC0642-LNP showed compatibility with multiple-dose regimens (2 IV injections/week for 4 weeks) with no acute toxicity observed and no major signs of chronic adverse effects detected (based on animal health scores, body weight, and blood chemistry).   Overall, this work presents a promising liposomal formulation of UNC0642 that is well suited for further pre-clinical tests in animal models of AML.   3.2 Results 3.2.1 UNC0642 can be efficiently encapsulated in DSPC/Chol and POPC/Chol liposomes via remote loading  Due to the weak base characteristics of UNC0642, which contains multiple ionizable basic nitrogens as seen in Figure 1, it was hypothesized that a remote loading strategy could be used to obtain efficient loading of the drug into pre-formed liposomes. To this end, DSPC/Chol (55:45, molar) or POPC/Chol (55:45, molar) liposomes were prepared containing 300 mM ammonium sulfate in their aqueous core and PBS as an exterior buffer (see Figure 2 for structures of lipid components). DLS analysis of the particles showed a monodisperse population (PDI <0.1) of uniformly-sized ~80 nm particles (Table 3). These liposomes were loaded at 5 mM total lipid at 65 °C for 20 minutes, after which any unentrapped UNC0642 was removed via dialysis. Particles were loaded at initial target drug:lipid ratios of 0.1, 0.2, or 0.3 (mol:mol). Drug:lipid ratios were assayed for samples taken before and after dialysis, and used to calculate percent entrapment.  As shown in Figure 3A, efficient loading was achieved for both DSPC/Chol and POPC/Chol particles at an initial target drug:lipid ratio of 0.1 (encapsulation efficiencies of 95.5% ±1.65% and 96.9% ± 1.60%, respectively) as well as 0.2 (encapsulation efficiency of 47  92.1% ±1.44% and 86.0% ± 1.94%, respectively). Increasing the initial target ratio to 0.3 resulted in marked decreases in encapsulation efficiency and values of 66.5% ± 1.93% and 56.9% ±1.23% for DSPC/Chol and POPC/Chol particles, respectively. When evaluating the final drug:lipid ratio obtained after dialysis to remove unencapsulated drug (Figure 3B), the final ratios were similar to the initial loading ratios for DSPC/Chol and POPC/Chol particles when loaded at 0.1 or 0.2 drug:lipid. However, when particles were loaded at a target ratio of 0.3, the final ratio obtained after dialysis to remove unencapsulated drug was only slightly higher than 0.2 (0.22 ± 0.0063 and 0.21 ± 0.0045 for DSPC/Chol and POPC/Chol particles, respectively). This indicated a maximum final ratio achievable of ~0.2, regardless of whether the initial ratio is increased past this point.    48     Figure 2. Chemical structure of liposome lipid components of UNC0642-LNP.  Liposomes used to generate UNC0642-LNP were composed of a 55:45 molar ratio of phospholipid:cholesterol. The phospholipids DSPC and POPC were used in this study.  Table 3. Particle size parameters for liposomes used to generate UNC0642-LNP.  Lipid Composition Mean Diameter ± SD (nm) PDI DSPC/Chol (55:45) 79.66 ± 5.71 0.050  POPC/Chol (55:45) 78 ± 3.26 0.048  49       0.0 0.1 0.2 0.3 0.405060708090100Loading Drug:Lipid (mol:mol)Percent Encapsulation (%)DSPCPOPCA0.0 0.1 0.2 0.3 Drug:Lipid (mol:mol)Final Drug:Lipid (mol:mol)DSPCPOPCB50  Figure 3. UNC0642 can be encapsulated in DSPC/Chol or POPC/Chol liposomes to a maximum 0.2 drug:lipid (molar) via remote loading.  Percent encapsulation (A) and final drug:lipid (mol:mol) ratios achieved (B) as a function of initial drug:lipid (mol:mol) during loading for DSPC/Chol and POPC/Chol (55:45, molar) liposomes loaded with UNC0642. Particles were loaded at 65 °C, 5 mM total lipid for 20 minutes followed by overnight dialysis against PBS to remove any unencapsulated drug. Encapsulation determined through quantification of UNC0642 and total lipid and represented as a percentage of the pre-dialysis loading drug:lipid ratio. Results represent the mean of three replicates ± SD.   3.2.2 Limited drug release is observed from UNC0642-LNP upon incubation in 37 °C PBS or 50% FBS  We began characterization of the UNC0642-LNP by analyzing drug release rates from the particles in a basic mimic of biological conditions. Conditions included PBS to mirror natural salinity upon venous injection and a temperature of 37 °C to mimic natural body temperature. DSPC/Chol or POPC/Chol particles (55:45, molar) loaded with UNC0642 at a drug:lipid ratio of 0.2 were incubated in the above conditions at a concentration of 2.5 mM total lipid in dialysis tubing. This allowed for any drug released from the particles to escape into the vast excess of incubation buffer while retaining the loaded particles. Aliquots of the incubated solution (interior of dialysis tubing) were removed at 0, 2, 8, 24, and 48 hours and assayed for concentration of UNC0642 and total lipid. As shown in Figure 4, limited release was observed for the DSPC/Chol particles, with 91.2% ± 0.2% and 89.3% ± 1.8% percent retention observed after 24 and 48 hours incubation, respectively. A relatively 51  greater degree of release was observed for the POPC/Chol particles, which showed 85.5% ± 2.3% and 67.7% ± 2.7% retention at 24 and 48 hours respectively. Overall, the POPC/Chol particles released 5.7% more drug than the DSPC/Chol particles after 24 hours and 21.6% more drug after 48 hours of incubation.      Figure 4. Limited drug release is obtained upon incubation of DSPC/Chol or POPC/Chol UNC0642-LNP in PBS at 37 °C.  Drug retention of UNC0642-LNP composed of DSPC/Cholesterol and POPC/Cholesterol (55:45, mol:mol) as a function of time (hours) incubated against PBS at 37 °C. Particles initially containing 0.2 molar drug:lipid ratio were incubated at a concentration of 2.5 mM total lipid. Encapsulation was determined through quantification of UNC0642 and total lipid concentration at each time point and presented as a percentage of the drug:lipid ratio measured at t = 0 hours. Results represent the mean of three replicates ± SD.  0 8 16 24 32 40 4805060708090100Time (hours)Drug Retention (%)DSPCPOPC52   To obtain release data using a more accurate mimic of physiological conditions post venous injection, release experiments were repeated using 50% FBS in PBS at 37 °C. This allowed for the introduction of lipoproteins to the incubation conditions, which have previously been shown to play a key role in nanoparticle destabilization in vitro and in vivo 80,81. Despite these changes to the incubation conditions, only a small increase to the drug release rate was observed for either DSPC/Chol or POPC/Chol formulations compared to PBS incubation. As shown in Figure 5, after 24 hours of incubation, drug retentions of 91.1% ± 2.4% and 79.6% ± 2.0% were observed for the DSPC/Chol and POPC/Chol particles, respectively. Compared to the values observed for PBS after 24 hours (Figure 4, 91.2% ± 0.2% and 85.5% ± 2.3% for particles containing DSPC and POPC, respectively), there was only a minor change in the amount of drug released (5.9% greater release in 50% FBS for POPC/Chol) or no change (overlapping error bars for PBS vs 50% FBS incubations of DSPC/Chol particles).  Since the ultimate objective of this thesis was to decrease the inherent toxicity of free UNC0642 through stable liposomal encapsulation, the DSPC/Chol formulations at 0.2 drug:lipid (mol:mol) were selected for further characterization and testing. These particles exhibited the highest degree of drug retention in both PBS and 50% FBS at 37 °C, indicating that they were able to stably encapsulate the drug even after prolonged exposure to simulated physiological conditions. This stable retention was desirable in further experiments evaluating the effect of encapsulation on toxicity as it was expected to help eliminate the potentially confounding influences of free drug leaking from less stable systems.     53   Figure 5. Incubation in 50% FBS in PBS results in increased but still limited drug release rates.  Drug retention rates of UNC0642-LNP composed of DSPC/Cholesterol and POPC/Cholesterol (55:45, mol:mol) after incubation against 50% FBS in PBS at 37 °C. Particles initially contained 0.2 molar drug:lipid ratio and were incubated at a concentration of 2.5 mM total lipid. Drug retention was quantified using UHPLC assay to determine released drug concentration using a fixed total volume. Percent encapsulation was reported relative to [UNC0642] measured at 100% release. Results represent the mean of three replicates ± SD.    0 4 8 12 16 20 24060708090100Time (hours)Drug Retention (%)DSPCPOPC54  3.2.3 Cryo-TEM imaging of DSPC/Chol UNC0642-LNP reveals amorphous spot patterns of electron density within the liposome interior  Due to the high degree of drug retention observed in the previous release studies, we performed structural evaluations of the UNC0642-LNP formulations. As shown in Figure 6, the DSPC/Chol particles, which previously exhibited the highest drug retention, showed a typical bilayer structure, with sizes in agreement with those obtained via DLS (~80 nm). Interestingly, amorphous and electron dense structures were observable in the aqueous core of most liposomes (red arrows). This structure was observable in the vast majority of the UNC0642-LNP particles visualized and appeared only within the liposome interior. A small number particles (white arrows) did not show the amorphous spots of electron density.       55        AB56   Figure 6. Cryo-TEM imaging of DSPC/Chol particles show amorphous and electron dense structures on the liposome interior.  Two representative cryo-TEM images of DSPC/Cholesterol (55:45, mol:mol) liposomes containing UNC0642 at a molar drug:lipid ratio of 0.2. Red arrows indicate loaded liposomes containing amorphous precipitate, while the white arrow indicates a liposome without amorphous precipitate. Panels A and B were imaged using the same sample of UNC0642-LNP. Scale bar = 100 nm.   3.2.4 In vitro studies in HoxA9/Meis1 murine leukemia cells show dose-dependent liposome uptake and lower anti-proliferative effect in UN0642-LNP compared to free UNC0642   To further characterize the in vitro properties of the UNC0642-LNP, we performed cell culture experiments using an AML cell line overexpressing HoxA9/Meis1 oncogenes. The first series of experiments focused on evaluation of the ability of these cells to take up the nanoparticles of interest. To quantify this, liposomes labeled with a small amount of fluorescent lipophilic dye (DiI) were prepared at a molar composition of 55:44.8:0.2 DSPC/Cholesterol/DiI. These fluorescent particles were added to cells in culture at concentrations of 46.38, 92.76, and 463.80 µg/mL total lipid (approximating in vivo injections at 0.5, 1.0, and 5.0 mg/kg). Treatments were incubated for 48 hours and uptake quantified via fluorescence analysis using FACS. The results of this study (Figure 7) show a dose-dependent increase in the mean fluorescence intensity (MFI) observed per cell. For the 46.38, 92.76 and 463.80 µg/mL treatments, a relative increase of approximately 4, 5, and 16-57  fold was observed, respectively. This corresponds to a near linear correlation between treatment concentration and corresponding mean fluorescence intensity per cell.   	 	 Figure 7. HoxA9/Meis1 AML cells incubated with fluorescent liposomes display a dose-dependent increase in mean cell fluorescence.  Quantification of UNC0642-LNP uptake (percent relative increase of mean fluorescence intensity per cell) as a function of particle concentration (µg/mL total lipid). in HoxA9/Meis1 leukemia cells. Cells were treated with indicated concentrations of DSPC/Cholesterol/DiI liposomes (55:44.8:0.2, molar) for 48 hours prior to quantification of mean fluorescence intensity per cell via FACS. 5 000 events were collected per sample, selecting for live cells PBS46.3892.76463.8005001000150020002500[Total Lipid] (µg/mL)Relative Increase of MFI (%)58  using PI staining. Mean fluorescence intensity was normalized to that of the PBS control. Results represent the mean of three replicates ± SD.   Further in vitro studies were performed to evaluate the anti-proliferative effects of UNC0642 in its free vs liposome encapsulated forms. To this end, HoxA9/Meis1 cells were treated for 5 days with increasing concentrations of either free UNC0642 dissolved in PBS (Figure 8A) or UNC0642-LNP composed of DSPC/Chol (55:45, mol:mol) (Figure 8B). Both treatments showed a dose-dependent decrease in the number of viable cells counted after five days of treatment. Approximately 10-times higher concentrations of UNC0642-LNP were required to obtain the same degree of anti-proliferative effect as the free UNC042 treatments. As an example, 60 µM of UNC0642-LNP yielded an average final cell count of 421 ± 29, while a similar final cell count was observed for 6 µM of free UNC0642 (391 ± 97 cells). The UNC0642-LNP treatments appeared to reach a plateau in anti-proliferative effect at approximately 120 µM encapsulated UNC0642, after which further increases resulted in only limited reductions in final cell counts.    59     	  	  0 1 2 3 4 5 60200400600800100012001400[UNC0642] (µM)Final Cell Count (x1000 cells)A0 25 50 75 100 125 1500200400600800100012001400[UNC0642] (µM)Final Cell Count (x1000 cells)B60    Figure 8. UNC0642-LNP causes a dose-dependent inhibition of proliferation in in HoxA9/Meis1 cells at a scale 10x less than free UNC0642.   Cells were treated at with either free UNC0642 (A) or DSPC/Cholesterol (55:45, molar) liposomes containing UNC0642 at a drug:lipid (molar) ratio of 0.2 (B). Free drug was administered at concentrations of 0.5, 1, 2, 4, or 6 µM UNC0642. UNC0642-LNP was dosed at concentrations of 15, 30, 60, 90, 120, or 150 µM of encapsulated UNC0642. At the time of treatment, 50 000 cells were plated per 4 mL well. The number of live cells (screened using PI staining) were quantified after 5 days of treatment using FACS. Results represent the mean of three replicates ± SD.   3.2.5 Liposomal encapsulation completely abolishes the acute toxicity of UNC0642 and results in no substantial long-term toxic after a single injection   The ultimate test of determining the effect of liposomal encapsulation on the toxicity of UNC0642 was to perform a series in vivo tolerability experiments comparing free vs encapsulated drug. To this end, the first test performed was a single dose tolerability study in which C57Bl/6 mice were injected IV with varying doses of the two drug formats. Although a single dose does not closely mirror the more regular dosing schedules that would be used for leukemia treatment, this test offered an informative baseline for determining tolerated doses for future experiments. Injections began at a UNC0642 concentration of 5.0 mg/kg and were either increased (if the 5.0 mg/kg injection was well tolerated) or decreased (if the 5.0 mg/kg injection was poorly tolerated). Acute response to injection was scored using a 61  detailed animal health monitoring sheet (Animal Monitoring Clinical Score Reference and Recording Sheet for Drug Testing), which included dedicated scores to appearance, activity, neurological response, etc. In all categories, higher scores indicated a decline in animal health.  An injection of 5.0 mg/kg free UNC0642 led to a severe acute response, as indicated by the sharp increase in the observed cumulative health score shown in Figure 9A. Within 10 minutes of injection, clear signs of toxicity were observed including intense piloerection, hunching, and a complete lack of movement. Furthermore, a major contributor to the increase in cumulative health score was the dramatic increase in neurological score, characterized by sever seizures (Figure 9B). These seizures progressed in severity to a humane endpoint at 11 minutes post-injection and the animal was sacrificed immediately. According to standard tolerability study protocol, the injected dose was halved to 2.5 mg/kg free UNC0642 in the next experimental group. Similar to the 5.0 mg/kg free drug injection, an acute response was observed within minutes after injection but to a lesser degree. All injected animals again exhibited numerous adverse side effects upon injection, including seizure. However, after a peak at 8 minutes post-injection, this response began to subside. Between 30-60 minutes post injection, recovery was complete with no adverse signs of poor health. When the injected dose of free drug was again lowered to 1.25 mg/kg, the acute response was even less intense with full recovery after 30 minutes. The response at 1.25 mg/kg free drug did include a neurological component of lesser severity, with slightly inhibited motor coordination and an unnatural rapid nodding of the head akin to a minor seizure.   62   In contrast to the free drug, UNC0642-LNP administered at 5.0 mg/kg showed a complete lack of acute response to the injection. There was an exact overlap of the average cumulative score curves of the 5.0 mg/kg UNC0642-LNP treatment and the PBS control. The same overlap was observed in the average neurological score curve, demonstrating a complete lack of observable neurotoxicity. When doses were increased to 7.5, 10.0, 15.0, and 20.0 mg/kg of UNC0642-LNP, the complete lack of an acute clinical response was maintained, with no negative impacts on any of the scored health categories. Importantly, this included a complete lack of any increase in average neurological score in any of the UNC0642-LNP doses.   After the above detailed single injections of free or liposomal UNC0642, animals were monitored for 14 days for signs of chronic toxicity. Monitoring included daily weighing and scoring using the same health monitoring system as for the acute observations. As shown in Figure 10A, single injections of either form of the drug did not result in major increases in the cumulative health scores of the animals. Though both the highest concentrations of the UNC0642-LNP (20 mg/kg) and free UNC0642 (2.5 mg/kg) curves do show significant increase compared to the PBS control, the highest magnitude of this increase was 1.0 for UNC0642-LNP and 1.3 for free UNC0642. Lower injections of either form of the drug resulted in similar trends to the highest concentration, with near complete overlap between the average cumulative score curves. When considering the effects of injection on body weight (Figure 10B), no significant difference was observed between the injections of the PBS control, highest concentration of UNC0642-LNP (20 mg/kg), or highest concentration of free drug (2.5 mg/kg). This overlap was maintained at all injected doses for free and encapsulated drug but these curves were excluded from Figure 10 for clarity. 63       0 10 20 30 40 50 600. Post Injection (minutes)Avg. Cumulative Clinical ScoreAPBS2.5 mg/kg Free1.25 mg/kg Free 5.0 mg/kg Free20 mg/kg LNP0 10 20 30 40 50 600. Post Injection (minutes)Avg. Neurological ScoreBPBS2.5 mg/kg Free1.25 mg/kg Free 5.0 mg/kg Free20 mg/kg LNP64  Figure 9. Liposomal encapsulation eliminates the dose-dependent acute toxicity of UNC0642.  Average cumulative clinical score (A) and average neurological score (B) vs time post injection (min) for C57BL/6 mice (female, 10-12 weeks old) injected with various concentrations of free (“Free”) or liposome-encapsulated (“LNP”) UNC0642. Liposomal formulations were composed of DSPC/Cholesterol (55:45, molar) and were loaded with UNC0642 at a molar drug:lipid ratio of 0.2. All injected compounds were dissolved in PBS and delivered IV through tail vein injection. Doses were given at mg of UNC0642 per kg of animal body weight. Injection volumes were given at 10 µL per gram of body weight. UNC0642-LNP were also injected at 5.0, 7.5, and 10.0 mg/kg but curves overlapped significantly with 20 mg/kg UNC0642-LNP curves and were excluded for clarity. Results reflect the mean of three replicate animals ± SD for all injections with the exception of the 5.0 mg/kg free drug treatment, which reflects the results from one injection.    65      0 2 4 6 8 10 12 Post Injection (days)Avg. Cumulative Clinical ScoreAPBS20 mg/kg LNP2.5 mg/kg Free0 2 4 6 8 10 12 14-6.0-4.0- Post Injection (days)Avg. Change in Body Weight (%)BPBS20 mg/kg LNP2.5 mg/kg Free66  Figure 10. UNC0642-LNP and free UNC0642 show no signs of significant longitudinal toxicity following a single dose.  Average cumulative clinical score (A) and average percent change in body weight (B) as a function of time post injection (days) for C57BL/6 mice (female, 10-12 weeks old) injected once with various concentrations of free or liposome-encapsulated UNC0642 (“Free” and “LNP” treatments, respectively). Liposome-encapsulated formulations were composed of DSPC/Cholesterol (55:45, molar) loaded with UNC0642 at a molar drug:lipid ratio of 0.2. All injected compounds were dissolved in PBS and delivered IV through tail vein injection. Doses were given at mg of UNC0642 per kg of animal body weight. Injection volumes were given at 10 µL per gram of body weight. Results reflect the mean of three replicate animals ± SD.    67  3.2.6 When administered using a multiple-dose regimen (2 injections/week, 4 weeks), low doses of free drug (1.25 mg/kg) and high doses of UNC0642-LNP (20 mg/kg) are tolerated with only minor signs of toxicity    To more accurately replicate a dosing regimen that would be used in leukemia treatment, tolerability studies were performed using 2 injections/week for a total of 4 weeks. Injections were made at 1.25 mg/kg free drug and 5, 10, or 20 mg/kg UNC0642-LNP. Daily monitoring of both clinical health score and body weight was recorded as in the single dose tolerability study for 4 weeks of injections followed by a 14-day observation period to check for signs of long-term toxicity.  As shown in Figure 11A, even the control group receiving PBS injections showed an increase in clinical score after the first week of injections. Though the cumulative score fluctuated throughout the duration of the experiment the overall increase was relatively minor and ranged between 0-1.83. Injections with 1.25 mg/kg led to significantly increased average clinical scores throughout the study. Though the score curves for PBS and the free drug injections overlapped in a small number of days, the majority of the days showed a significant difference between the two.  Repeated injection of UNC0642-LNP also led to an increase in average clinical score. This increase was not as large as was observed for the free UNC0642 treatments group when considering the magnitude of the average increase. However, the majority of the days show no significant difference between the free UNC0642 and UNC0642-LNP curves (i.e. overlapping error bars). For both the free and encapsulated drug treatments, the largest deviation from the PBS control score was observed on days 23 and 24. During these days, both the 1.25 mg/kg free drug and 20 mg/kg UNC0642-LNP showed a score that was ~2 68  units higher than the PBS control. It should be noted that the curves for 5.0 and 10.0mg/kg UNC0642-LNP overlapped almost completely with the 20 mg/kg UNC0642-LNP curve and were therefore excluded from Figure 11 and Figure 12 for clarity. Overall, similar average cumulative clinical score curves were obtained for multiple-dose injections of free drug at 1.25 mg/kg and encapsulated UNC0642 at a dose 16-times higher (20 mg/kg).  Figure 11B shows that no significant negative impact was observed on animal body weight upon multiple injections with either free or liposome-encapsulated UNC0642. As with the average cumulative health score data, the treatment groups receiving UNC0642-LNP injections at 5.0 and 10.0 mg/kg were comparable to that of the 20.0 mg/kg group and were excluded for clarity.  As with the single dose tolerability studies, animals were monitored closely for 60 minutes following each injection to detect any signs of acute toxicity. Results of these acute observations for the 8th and final injection of the multiple-dose study (Figure 12) were similar to those obtained after a single dose (Figure 9). Notably, the ability of liposomal encapsulation to completely abolish the severe acute response caused by the free drug, even at concentrations 16-times higher, was maintained from injection 1 to injection 8 of the multiple dose study. Again, as seen in the single dose experiments, this amelioration of the acute response resulted in a complete lack of any negative impact on neurological health. In contrast, animals treated with free drug continued to experience severe neurological and other clinical signs of compromised health following each injection, though they were shown to recover within 30-60 minutes post-injection.   Blood samples were collected from animals 14 days following their final injection to look for any signs of chronic toxicity using blood chemistry analysis. The results of these 69  analyses (Figure 13) show very limited differences between the PBS control and any of the experimental treatments including the 1.25 mg/kg free drug injection or any of the three UNC0642-LNP injections. 2-way ANOVA comparison of each experimental group to the PBS control were performed for each of the four biomarker sets (Figure 13 A-D). When considering biomarkers of liver health, (Figure 13A), no significant differences were observed for alanine transaminase (ALT), alkaline phosphatase (ALP) or gamma-glutamyl transferase (GGT). The only significant increase seen for aspartate transaminase (AST) levels was the 5.0 mg/kg UNC0642-LNP treatment, in which there was an approximately 3-fold increase in biomarker levels. All other treatments showed no significant change in AST levels.  Biomarkers of general protein concentration (Figure 13B) showed the most significant differences between the experimental groups and the PBS control. A significant decrease in total protein concentration was seen for all four experimental groups. The largest decrease, 6.0 g/L, was seen for 5.0 mg/kg UNC0642-LNP, while the smallest decrease (3.7 g/L) was observed in the 20 mg/kg UNC0642-LNP animals. Globulin levels were quite similar, but a small significant difference was observed for the 1.25 mg/kg free drug and 5.0 mg/kg UNC0642-LNP treatments (decreases of 3.0 and 3.5 g/L, respectively). No significant changes were seen for Albumin or albumin/globulin (A/G) ratio for any of the experimental groups.  No significant differences were seen for either of the tissue damage biomarkers, lactate dehydrogenase (LDH) and creatine kinase (CK), as shown in Figure 13C. Similarly, there was neither a significant increase or decrease in the levels of the kidney biomarkers, blood urea nitrogen (BUN) and symmetric dimethylarginine (SDMA) (Figure 13D). 70  Significantly decreased levels of creatinine were observed for the two highest UNC0642-LNP treatments, with a decrease of 4.8 and 4.3 µmol/L observed for the 10 and 20 mg/kg UNC0642-LNP treatments, respectively.    71     Figure 11. Multiple dose regimens of UNC0642 or UNC0642-LNP injection show limited indications of long-term toxicity based on body weight and clinical health score.  0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 400. Post Injection (days)Avg. Cumulative Clinical ScoreAPBS20 mg/kg LNP1.25 mg/kg Free 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 400. Post Injection (days)Avg. Change in Body Weight (%)BPBS20 mg/kg LNP1.25 mg/kg Free 72  Average cumulative clinical health score (A) and average percent change in body weight (B) as a function of days post injection for C57BL/6 mice (female, 10-12 weeks old) injected 2 times per week for 4 weeks with various concentrations of free or liposome-encapsulated UNC0642 (“Free” or “LNP” treatments, respectively). Liposome-encapsulated formulations were composed of DSPC/Cholesterol (55:45, molar) loaded with UNC0642 at a molar drug:lipid ratio of 0.2. All injected compounds were dissolved in PBS and delivered IV through tail vein injection. Doses were given at mg of UNC0642 per kg of animal body weight. Injection volumes were given at 10 µL per gram of body weight and injections were administered on days 1, 4, 8, 11, 15, 18, 22, and 25. Results reflect the mean of three replicate animals ± SD (PBS and 1.25 mg/kg free UNC0642 treatments) or four replicate animals ± SD (all UNC0642-LNP treatments).  73      0 10 20 30 40 50 600. Post Injection (minutes)Avg. Cumulative Clinical ScoreAPBS1.25 mg/kg Free 20 mg/kg LNP0 10 20 30 40 50 600. Post Injection (minutes)Avg. Neurological ScoreBPBS1.25 mg/kg Free 20 mg/kg LNP74  Figure 12. The ability of liposomal encapsulation to abolish the acute toxic effects of UNC0642 injection is maintained through multiple injections.  Average cumulative clinical score (A) and average neurological score (B) vs time post injection (min) for C57BL/6 mice (female, 10-12 weeks old) after the 8th injection of a 4-week regimen (2 injections/week) using various concentrations of free or liposome-encapsulated UNC0642 (“Free” or “LNP” treatments, respectively). Liposomal formulations were composed of DSPC/Cholesterol (55:45, molar) and were loaded with UNC0642 at a molar drug:lipid ratio of 0.2. All injected compounds were dissolved in PBS and delivered IV through tail vein injection. Doses were given at mg of UNC0642 per kg of animal body weight. Injection volumes were given at 10 µL per gram of body weight. UNC0642-LNP were also injected at 5.0 and 10.0 mg/kg but curves overlapped with 20 mg/kg UNC0642-LNP curves and were excluded for clarity. Results reflect the mean of three replicate animals ± SD (PBS and 1.25 mg/kg free UNC0642 treatments) and four replicate animals ± SD (all UNC0642-LNP treatments).    75         ALT (IU/L)AST (IU/L)ALP (IU/L)GGT (IU/L)050100150200250Concentration (IU/L)APBS1.25 mg/kg Free 5.0 mg/kg LNP10 mg/kg LNP20 mg/kg LNP***Total Protein (g/L)Albumin (g/L)Globulin (g/L)A/G Ratio01020304050Concentration (g/L)  or RatioPBS1.25 mg/kg Free 5.0 mg/kg LNP10 mg/kg LNP20 mg/kg LNP************ **B76      LDH CK0100200300400500Concentration (IU/L) PBS1.25 mg/kg Free 5.0 mg/kg LNP10 mg/kg LNP20 mg/kg LNPCUrea (BUN) (mmol/L )Creatinine (µmol/L )SDMA (µg/dL)0510152025Concentration (mmol/L, µmol/L, or µg/dL) PBS1.25 mg/kg Free 5.0 mg/kg LNP10 mg/kg LNP20 mg/kg LNP** *D77  Figure 13. Blood chemistry analysis of animals measured after completion of a multiple dose regimen shows no clear signs of substantial chronic toxicity for low doses of free drug (1.25 mg/kg) or high doses of UNC0642-LNP (20 mg/kg).  Blood chemistry results for C57BL/6 mice (female, 10-12 weeks old) injected 2 times per week for 4 weeks with various concentrations of free UNC0642 or liposome-encapsulated UNC0642 (“Free” or “LNP” treatments, respectively). Analytes measured include biomarkers of liver health (A), general protein concentration (B), tissue damage (C), and kidney function (D).  Measures were taken from serum harvested via cardiac puncture performed on animals sacrificed 14 days following final injection. Liposome-encapsulated formulations were composed of DSPC/Cholesterol (55:45, molar) loaded with UNC0642 at a molar drug:lipid ratio of 0.2. All injected compounds were dissolved in PBS and delivered IV through tail vein injection. Doses were given at mg of UNC0642 per kg of animal body weight. Results reflect the mean of three replicate animals ± SD (PBS and 1.25 mg/kg free UNC0642 treatments) and four replicate animals ± SD (all UNC0642-LNP treatments). Abbreviations used: ALT (Alanine Transaminase), AST (Aspartate Transaminase), ALP (Alkaline Phosphatase), GGT (Gamma-Glutamyl Transferase, A/G (Albumin/Globulin), LDH (Lactate Dehydrogenase), CK (Creatine Kinase), BUN (Blood Urea Nitrogen), SDMA (Symmetric Dimethylarginine).   3.3 Discussion  The findings from this work strongly support the initial hypothesis that liposomal encapsulation of UNC0642 would decrease the inherent in vivo toxicity of the small molecule. These results provide thorough characterization of the UNC0642-LNP, including 78  initial loading strategy, in vitro structure and retention characteristics, and clear demonstrations of reduced toxicity in vivo using both single and multiple-dose regimens. Two topics warrant further discussion after evaluation of these results. The first relates to the loading and retention of UNC0642 into the liposomes and how these processes may be influenced by drug precipitation within the interior of the particles. The second concerns the mechanism through which nanoparticle encapsulation is able to alter the bioavailability and, consequently, the toxicity of UNC0642.   With respect to development of a loading strategy to generate UN0642-LNP, remote loading has been well established as an efficient method for the encapsulation of weak base small molecule therapeutics (see Section 1.1.4 Techniques for encapsulating therapeutics within liposomes). As such, the observations that UNC0642 was successfully loaded into DSPC/Chol and POPC/Chol LNP using this method to final drug:lipid ratios of 0.1 and 0.2 were not surprising and largely expected (Figure 3). As initial loading ratios were pushed past this point, a drop in percent encapsulation was observed, which corresponded to a plateau in achievable final drug:lipid ratios. This drop in percent encapsulation was also expected and can be rationalized based on previous studies. An initial justification for a maximum loading capacity for pH gradient liposomes is based on exhaustion of the proton gradient on which the systems rely on for effective drug encapsulation, a phenomenon that has been proposed previously 118. As more and more drug is introduced to a fixed concentration of pH gradient liposomes, it logically follows that there will come a point at which the free protons present on the liposome interior will effectively be depleted. At this point, any non-protonated free drug would be able to exit the liposome and avoid effective encapsulation. However, though this gradient exhaustion rationale has been supported by 79  other small molecule-liposome systems, the observation of amorphous, electron dense spot patterns on the interior of UNC0642-LNP upon cryo-TEM imaging may suggest additional factors contributing to the plateau in loading capacity at 0.2 drug:lipid. These patterns are thought to be precipitation of the drug into amorphous granular structures caused by the high local concentration of the drug upon remote loading. Similar patterns of amorphous electron density spots have been observed in other small molecule-loaded liposomes, specifically Sphingomyelin/Chol systems loaded with the chemotherapeutic vincristine 83. In these systems, increasing drug:lipid ratios yielded observations of a general increase in electron density in the liposome interior, suggesting precipitation into an amorphous gel with no clear structure. However, increasing the drug:lipid ratio past 1.03 (wt/wt) led to visualization of distinct granular structures with high electron density. Though not as prominent as the structures observed in vincristine loaded liposomes, the UNC0642-LNP imaged in Figure 6 showed a consistent granular structure of electron density in the majority of the liposome cores. This structure was distinct from the small subpopulation of liposomes which showed no such pattern (example indicated by white arrow), and was likely caused by a granular precipitate of UNC0642. Though drug precipitation has been observed for a variety of small molecules upon remote loading, including topotecan, doxorubicin, and mitoxantrone, the precipitates are most commonly seen as clearly defined, long, and thin bundles or crystals 54. Vincristine, vinorelbine, and potentially UNC0642 are therefore unique in that precipitation occurs in a less structured, granular form. All of the aforementioned precipitate types (including both rod and granular forms) have been shown to disrupt the ammonium/pH gradient driving drug encapsulation 118. More specifically, the precipitates have been shown to lead to a disruption of the liposomal membrane leading to leakage of the aqueous 80  liposome interior. Though such leakage is seen at lower levels in liposomes encapsulating drugs that precipitate amorphously, this phenomenon could help to explain the loss of percent encapsulation seen for the UNC0642-LNP at drug:lipid ratios above 0.2. Further studies would be required to confirm and characterize the precipitation of UNC0642. These would entail preparing UNC0642-LNP at increasing drug:lipid ratios for 1) Imaging them via cryo-TEM to look for a corresponding increase in precipitate formation; and 2) Performing release experiments to look for a relationship between drug:lipid ratio and rate of drug release. These experiments are discussed in more detail in Chapter 4: Conclusion and Future Directions.  Precipitation has also been shown to affect drug release rates from loaded liposomes, increasing the overall stability of the drug within the liposome interior and leading to slower release rates over time at higher drug:lipid ratios. These trends have been observed in doxorubicin, vincristine, and vinorelbine loaded liposomes, all of which exhibit drug precipitation 119. In short, precipitation of these drugs within the liposomes leads to an equilibrium being formed between the soluble and precipitated forms of the small molecule. This effectively decreases the amount of drug in its soluble form compared to a scenario in which there is no precipitation. Since drug release rate from liposomes, according to Fick’s law, is proportional to the concentration gradient of drug across the liposomal membrane and thus the amount of soluble drug contained within the liposome interior, an effective reduction of the amount of soluble drug present leads to a slower release rate. This could help to explain the high degree of stability observed in the UNC0642 particles in both PBS (Figure 4) and 50% FBS (Figure 5).  81  Though a high degree of retention was seen in both DSPC/Chol and POPC/Chol particles, the latter showed an increase rate of release. This was an expected result and can be rationalized due to the greater degree of unsaturation in POPC’s acyl chains compared to those of DSPC (see Figure 2). It is fundamentally understood that introducing units of unsaturation into lipid acyl chains leads to increased membrane fluidity and a lower phase transition temperature (Tm) for a given species or mixture of lipid species 120. In the context of drug-loaded liposomes, this increase in membrane fluidity can lead to the interior components of a liposome being more easily able to leak from the nanoparticle. This leakage can include the entrapped drug molecules, ammonium ions or protons necessary for the pH gradient driving drug encapsulation, as well as other potentially stabilizing species such as sulfate ions. In all three of these cases, this increase in membrane fluidity leads to a compromised ability to effectively retain any encapsulated drug. For these reasons, it was expected that the particles containing POPC (phospholipid Tm = -2 °C for pure phospholipid) showed greater drug release than those with DSPC (phospholipid Tm = 55 °C) 121.  With respect to the second topic of discussion, namely the mechanism through which liposomal encapsulation affects the inherent toxicity of UNC0642, clear indications of altered bioavailability are present in the results obtained. It was hypothesized that nanoparticle entrapment would lead to altered bioavailability of the drug and, by doing so, would help eliminate its off-target side effects. The in vitro anti-proliferation experiment results showed a clear difference in the activity of free vs liposomal UNC0642 that likely arose from such changes in bioavailability (Figure 8). The approximately 10-fold difference in anti-proliferative effect of the UNC0642-LNP compared to free drug was probably caused by the cells being less able to access the drug in its encapsulated form. This reduced 82  availability could have arose due to a number of factors. As seen in the drug retention studies, a portion of the drug was released from the DSPC/Chol particles (approximately 9% per day) when incubated in similar conditions (Figure 5). As such, a major component of the activity of UNC0642-LNP was likely due to slow leakage of the payload from the nanoparticles and subsequent uptake of the leaked free drug by the cells. Another important contribution to UNC0642-LNP activity was the direct uptake of the particles themselves. As indicated by the results in Figure 7, there was a linear increase in uptake of DSPC/Chol/DiI particles in response to increasing treatment concentration in HoxA9/Meis1 cells (indicated by an increase in mean fluorescence intensity per cell). The correlation between increased cell fluorescence and particle uptake has been established previously, with past studies showing stable association of DiI with liposomal carriers under similar cell culture conditions 122. Furthermore, studies performed in our group characterizing lipid exchange upon incubation in 100% mouse plasma have shown a similar stable association of DiI with liposomes (data not published). However, though the results in Figure 7 indicate that there was some degree of UNC0642-LNP uptake by the AML cells, the magnitude of this uptake in terms of the number of particles taken up per cell and intracellular concentration cannot be exactly determined. Furthermore, particle uptake by the AML cells did not guarantee that the encapsulated drug was bioavailable. If taken up into the endosomal compartment, the UNC0642 payload would have to have escaped not only its liposomal carrier but also the endosome itself. Without the aid of any lipid component that facilitates such endosomal release, it is possible that particles in the endosome were ultimately trafficked to lysosomal compartments. Degradation in these sites would have prevented the drug from acting intracellularly as desired. Overall, the decreased anti-proliferative effect upon liposomal 83  encapsulation observed in vitro provides evidence of the desired alteration of the drug’s bioavailability. If encapsulation was able to alter the in vitro distribution and availability of the drug, it should be expected to do the same in vivo. The results from both the single and multiple-dose tolerability studies provide strong evidence that this was the case.  The complete abolishment of UNC0642’s severe acute toxicity upon liposomal encapsulation provides the strongest evidence of the particles effectively altering the drug’s biodistribution and bioavailability as desired (Figure 9 and Figure 12). Though we can only speculate as to the exact cause of the free drug’s toxicity, the strong neurological component of the severe acute response to free drug injection, manifesting in short-term seizures, suggests a neurotoxic element. Furthermore, past characterizations of UNC0642 have shown moderate brain penetration upon IP injection at 5 mg/kg 117. If the side effects of free-drug injection are largely due to off-target effects in the brain or nervous system, it is largely expected that liposomal entrapment would reduce toxicity. It has been previously established that DSPC/Chol or other similar lipid nanoparticles have poor biodistribution to the brain and spinal column. Specifically, similar PC/Chol particles have shown less than 0.1% of the injected dose reaching these compartments in timeframes ranging from 5-180 minutes to 21 hours after injection 123,124. In addition to likely preventing exposure to the brain or nervous system, encapsulation likely affected the pharmacokinetics of UNC0642 significantly. Following IV injection, most small molecules are effectively cleared from circulation on the scale of minutes to hours 125. In contrast, liposomal encapsulated versions of the same molecules often exhibit extended plasma circulation at higher overall concentrations of the drug. Though pharmacokinetic studies would be required to confirm that this is the case with UNC0642 and UNC0642-LNP, the pattern of increased plasma retention times in similar 84  small molecule liposomal systems strongly suggests that the same holds true for this drug. Past studies using DSPC/Chol particles of identical composition to those used here have shown detectable levels of the liposomes 24 hours post injection and a circulation half-life of approximately 12 hours 126. Though loading of the particles with UNC0642 could have minor effects on plasma retention times (compared to empty particles), it is likely that the UNC0642-LNP exhibit pharmacokinetic properties similar to those of unloaded DSPC/Chol liposomes.  Due to this likely increase in plasma retention time of the UNC0642-LNP, there is a concern of longitudinal drug release from the particles in circulation that could potentially lead to a delayed toxicity response. Fortunately, there was very limited evidence of any such long-term toxicity observed in either the single or multiple dose tolerability studies. With respect to the single dose study (Figure 10), only mild increases in the average cumulative score were observed at the highest dose UNC0642-LNP, with the greatest discrepancy between the PBS control and the 20 mg/kg UNC0642-LNP treatment being a cumulative score of 1.0. This corresponds to only a minor effect on animal health and stands in stark contrast to the increases of 5-10 units seen in the acute toxicity response to the free drug injections. Furthermore, no adverse effects were observed to animal body weight at the highest UNC0642-LNP dose, again indicating a lack of longitudinal toxicity. When considering the multiple-dose tolerability study, the potential for long-term toxicity is increased due to a more consistent presence of the UNC0642-LNP in the bloodstream, leading to more prolonged drug release and extended exposure of tissues to the drug. In addition, there is the potential for incomplete clearance of the UNC0642-LNP between injections, which could lead to steady accumulation of the particles with each consecutive 85  injection. The dosing regimen (2 injections/week) was selected based on the previously reported clearance rates of DSPC/Chol liposomes that suggest near complete clearance within 48 hours post injection 126. However, since pharmacokinetic studies of the UNC0642-LNP were not performed, there was no guarantee that they would exhibit PK parameters identical to those previously observed. The longitudinal observations of cumulative health scores and animal body weight from the multiple dose study (Figure 11) indicated that any sort of long-term particle accumulation, if it occurred, had no major negative impact on animal health. Though higher health scores were seen for certain days during the study (Figure 11A), the largest difference between the scores of the PBS control and the highest UNC0642-LNP dose (2.0 units) was much smaller than any of the acute clinical responses observed. Once again, the lack of any significant negative impact on body weight (Figure 11B) also indicated a lack of toxic nanoparticle accumulation or, if accumulation did occur, a lack of any significant negative impact on animal health. These conclusions were further supported by the blood chemistry results (Figure 13), which showed limited signs of chronic toxicity. The panel of liver damage relevant biomarkers measured (Figure 13A) only showed significantly increased levels of AST in the 5 mg/kg UNC0642-LNP treatment. The fact that there was no increase in AST in the 10 or 20 mg/kg UNC0642-LNP treatments suggests that this effect was likely not caused by the particles, as a dose-dependent increase in the enzyme levels would be expected if this were the case. Additionally, any increases in AST caused by liver damage are typically associated with changes in the levels of other liver damage markers, which was not the case for the 5.0 mg/kg treatment (ALT, ALP, and GGT concentrations were comparable between the 5.0 mg/kg and PBS treatments) 127. This evident lack of chronic liver damage provides support for the UNC0642-LNP particles’ 86  compatibility with dosing regimens that could be used in leukemia treatment. Since similar nanoparticles have shown preferential accumulation within the liver upon IV injection, the lack of any clear signs of liver toxicity indicate that neither accumulation of the encapsulated UNC0642 or the lipid carrier contribute to any long-term hepatic toxicity. Similarly, the lack of any significant changes in CK or LDH indicated a lack of skeletal muscle damage or general tissue damage, respectively (Figure 13C) 128,129. Elevated levels of LDH can also be induced by leukemia or other sources of anemia 130. The lack of any elevation in LDH levels in response to any dose of UNC0642-LNP is encouraging for future experiments on leukemic animals, in which the animals’ hematopoietic capabilities would be compromised, making them more vulnerable to complications arising from toxicity-induced anemia. The protein-related biomarkers measured also show limited signs of chronic toxicity (Figure 13B). The uniform decrease in total protein observed in response to all treatments could have been indicative of kidney or liver dysfunction 131. However, the lack of a corresponding uniform drop in albumin or globulin levels fails to provide additional evidence for these conditions. The lack of any clear indications of liver toxicity from the liver-associated biomarkers also makes the possibility of liver dysfunction unlikely. Kidney function biomarkers showed little change with increasing UNC0642-LNP doses, though a small decrease in creatinine was observed for the two highest nanoparticle treatments (Figure 13D). Though an increase in creatinine would be indicative of compromised kidney function, a decrease is more ambiguous 132. Potential causes include muscle loss, though the lack of significant weight loss in either of these groups indicates that, if this were the cause, such loss was relatively minor 133. Taken together, the kidney function biomarker levels show no indication of chronic nephrotoxicity.    87  Overall, the ability of liposomal encapsulation to completely abolish the acute toxicity of free UNC0642 while producing only minor signs of chronic toxicity was likely due to altered drug biodistribution. Ultimately, this allowed for injection of UNC0642 at concentrations 8-fold higher than the maximum tolerated dose of the free drug (20 mg/kg vs 2.5 mg/kg) while remaining compatible with single or multiple dose regimens. It should be noted that the complete lack of acute response to the 20 mg/kg UNC0642-LNP and the associated mild long-term toxicity suggests that even higher doses could be well tolerated in healthy animals.   88  Chapter 4: Conclusion and Future Directions  The work presented has effectively documented the generation, characterization, and in vivo validation of a versatile LNP-based system for epigenetic inhibition of G9a/GLP for future application in AML treatment. A reliable and efficient loading method has been demonstrated for two distinct lipid compositions and will likely remain effective for other compositions as long as they are compatible with ammonium sulfate gradient-mediated loading. Furthermore, this is the first time a nanoparticle formulation of UNC0642 has been reported. The in vitro characterization performed has revealed an intriguing indication of drug precipitation that warrants further examination. To confirm the presence of precipitate, two further experiments could be pursued. First, liposomes loaded with ascending drug:lipid ratios could be imaged using cryo-TEM and observed for a corresponding increase in the number and intensity of electron dense structures present in their interior. Previous work with liposomal vincristine shows a dose-dependent darkening of the intraliposomal space along with the amorphous, electron dense spots becoming more numerous and defined 83. If UNC0642 is indeed precipitating in the liposomes in a manner similar to vincristine, it would be expected to behave in a similar fashion. Such experiments could be extended through exploration of the effect of increasing the concentration of ammonium sulfate included in the liposome interior to see if this could enhance the particles’ drug loading capacity (effectively increasing the maximum possible drug:lipid ratios achievable). If such increased loading were possible, imaging of particles loaded at ratios of 0.3 or higher could allow for visualization of more clearly defined drug precipitates. Secondly, additional release experiments could be performed using UNC0642-LNP containing varying drug:lipid ratios 89  incubated against a low concentration of ammonium chloride. Such an incubation is designed to collapse the ammonium gradient that is involved in drug retention. Previously characterized liposomal drug formulations showing precipitation uniformly demonstrate drug release rates proportional to their drug:lipid ratio 83,118,119. If UNC0642-LNP showed a similar relationship between these two values, it would lend strong support to the hypothesis that the drug precipitates within the liposomes.   The most prominent future study direction would be to test the validated UNC0642-LNP systems in animal models of AML. Such models are readily available, with C57Bl/6 mice having been successfully irradiated and inoculated with the same HoxA9/Meis1 cell line used for the in vitro cell culture studies in this work 114. Since the UNC0642-LNP have already been tested against these cells, this provides the ideal animal model for future efficacy experiments. The mouse breed is also identical between the AML animal model and the strain used in the tolerability studies performed in this work. One important consideration is that the leukemic animals will likely be comparatively fragile relative to the healthy animals used in the toxicity studies presented. As such, efficacy studies should be performed at a variety of UNC0642-LNP doses in the event that 20 mg/kg falls above the maximum tolerated dose for the diseased mice. The leukemic animals would be dosed at a range of UNC0642-LNP concentrations along with the maximum tolerated dose of free drug to compare efficacy. By monitoring the impact of treatment on the rate of disease progression (monitored by leukemic blast counts from regularly collected blood samples), a clear indication of the UNC0642-LNP’s anti-cancer efficacy could be obtained. To give a comprehensive overview of how the particles acted in vivo, it would be prudent to perform pharmacokinetic and biodistribution studies in healthy and, ideally, AML animals. Such 90  experiments would provide in vivo drug release rates (measured by tracking plasma drug:lipid ratio over time using radiolabeled lipid and UNC0642) and organ distribution patterns. Of particular interest would be distribution rates to the bone marrow, the major site of hematopoiesis from which leukemic cells originally arise. If the treatments were given at early stages of cancer progression, the majority of the leukemic cells would likely be situated in this compartment, making increased nanoparticle accumulation at the site ideal.  The combination of the anti-cancer efficacy, pharmacokinetic, and biodistribution data would serve as a launching point for future experiments aiming to optimize the anti-leukemic efficacy of the formulation. The DSPC/Chol particles tested here are one of the more simplistic systems used for liposomal drug delivery and many modifications to lipid composition could help maximize the particles’ potency. Circulation time could be extended if necessary by introducing PEG-lipids to the liposome surface, a common feature of many anti-cancer formulations as detailed in previous sections (see Section 1.1.7 Liposome characteristics influencing drug release rates). Increased circulation time would likely elevate the proportion of the injected dose reaching the bone marrow by decreasing MPS uptake. Modifications to the acyl chains of the phospholipid used would help optimize rate of release, which could be necessary to find a balance between retention preventing drug distribution to sites of toxicity and release to diseased areas. Changing the charge of the particles as well as the cholesterol concentration could also help optimize biodistribution. Negatively charged liposomes with lower cholesterol content have demonstrated increased accumulation in the bone marrow, offering another avenue for increasing accumulation in this target site 134.  91  A final direction for future studies would be expansion of the inhibitors encapsulated within the nanoparticle systems. Other iteration in the UNC inhibitor line show similar chemical scaffolds that would likely be amenable to pH-gradient loading due to their weak base characteristics. This would allow for the nanoparticle formulations to remain up to date with the latest G9a/GLP inhibitors developed.  Overall, several intriguing future directions are available for the work presented. This study’s experiments provide the foundation for a potent anti-AML formulation that, first and foremost, can be safely administered. The utilization of liposomal carriers for UNC0642 not only decreases the inherent toxicity as desired but also establishes an easily modifiable platform for therapeutic delivery that can be tailored as necessary to optimize drug release and biodistribution. Encapsulation of an inherently toxic anti-cancer drug is a well-established story, with many liposomal chemotherapeutics approved by the FDA today and more on the horizon. As such, it is not unreasonable to expect a similar path forward for nanoparticle-encapsulated epigenetic inhibitors of AML.     92  References 1 Bangham, A. D. & Horne, R. W. Negative Staining of Phospholipids and Their Structural Modification by Surface-Active Agents as Observed in the Electron Microscope. J Mol Biol 8, 660-668 (1964). 2 Deamer, D. W. From "banghasomes" to liposomes: a memoir of Alec Bangham, 1921-2010. FASEB J 24, 1308-1310, doi:10.1096/fj.10-0503 (2010). 3 Bangham, A. D., Standish, M. M. & Watkins, J. C. Diffusion of univalent ions across the lamellae of swollen phospholipids. J Mol Biol 13, 238-252 (1965). 4 Bangham, A. D., Standish, M. M. & Weissmann, G. The action of steroids and streptolysin S on the permeability of phospholipid structures to cations. Journal of Molecular Biology 13, 253-IN228, doi:https://doi.org/10.1016/S0022-2836(65)80094-8 (1965). 5 Sessa, G. & Weissmann, G. Phospholipid spherules (liposomes) as a model for biological membranes. Journal of lipid research 9, 310-318. 6 Johnson, S. M. & Bangham, A. D. Potassium permeability of single compartment liposomes with and without valinomycin. Biochimica et Biophysica Acta (BBA) - Biomembranes 193, 82-91, doi:https://doi.org/10.1016/0005-2736(69)90061-3 (1969). 7 Batzri, S. & Korn, E. D. Single bilayer liposomes prepared without sonication. Biochimica et Biophysica Acta (BBA) - Biomembranes 298, 1015-1019, doi:https://doi.org/10.1016/0005-2736(73)90408-2 (1973). 8 Gregoriadis, G. & Ryman, B. E. Liposomes as carriers of enzymes or drugs: a new approach to the treatment of storage diseases. Biochemical Journal 124, 58P (1971). 9 Gregoriadis, G. Drug entrapment in liposomes. FEBS Letters 36, 292-296, doi:10.1016/0014-5793(73)80394-1 (1973). 10 Kimelberg, H. K., Tracy, T. F., Biddlecome, S. M. & Bourke, R. S. The effect of entrapment in liposomes on the in vivo distribution of [3H]methotrexate in a primate. Cancer Research 36, 2949-2957 (1976). 11 Poste, G. & Papahadjopoulos, D. Lipid vesicles as carriers for introducing materials into cultured cells: influence of vesicle lipid composition on mechanism(s) of vesicle incorporation into cells. Proceedings of the National Academy of Sciences of the United States of America 73, 1603-1607 (1976). 12 Juliano, R. L. & Stamp, D. Pharmacokinetics of liposome-encapsulated anti-tumor drugs: Studies with vinblastine, actinomycin D, cytosine arabinoside, and daunomycin. Biochemical Pharmacology 27, 21-27, doi:https://doi.org/10.1016/0006-2952(78)90252-6 (1978). 13 Kobayashi, T., Tsukagoshi, S. & Sakurai, S. Enhancement of the Cancer Chemotherapeutic Effect of Cytosine Arabinoside Entrapped in Liposomes on Mouse Leukemia L-1210. GANN Japanese Journal of Cancer Research 66, 719-720, doi:10.20772/cancersci1959.66.6_719 (1975). 14 Mayhew, E., Papahadjopoulos, D., Rustum, Y. M. & Dave, C. Inhibition of Tumor Cell Growth in Vitro and in Vivo by 1-β-d-Arabinofuranosylcytosine Entrapped within Phospholipid Vesicles. Cancer Research 36, 4406 (1976). 93  15 Barenholz, Y. Doxil® — The first FDA-approved nano-drug: Lessons learned. Journal of Controlled Release 160, 117-134, doi:https://doi.org/10.1016/j.jconrel.2012.03.020 (2012). 16 Barenholz, Y. et al. A simple method for the preparation of homogeneous phospholipid vesicles. Biochemistry 16, 2806-2810, doi:10.1021/bi00631a035 (1977). 17 Litman, B. J. Lipid model membranes. Characterization of mixed phospholipid vesicles. Biochemistry 12, 2545-2554, doi:10.1021/bi00737a028 (1973). 18 Newman, G. C. Structural studies on phosphatidylcholine-cholesterol mixed vesicles. Biochemistry (Easton) 14, 3363-3370, doi:10.1021/bi00686a012. 19 Akbarzadeh, A. et al. Liposome: classification, preparation, and applications. Nanoscale Research Letters 8, 102, doi:10.1186/1556-276X-8-102 (2013). 20 Barenholzt, Y., Amselem, S. & D, L. A new method for preparation of phospholipid vesicles (liposomes) - french press. FEBS Letters 99, 210-214, doi:https://doi.org/10.1016/0014-5793(79)80281-1 (1979). 21 Mayhew, E., Lazo, R., Vail, W. J., King, J. & Green, A. M. Characterization of liposomes prepared using a microemulsifier. Biochimica et Biophysica Acta (BBA) - Biomembranes 775, 169-174, doi:https://doi.org/10.1016/0005-2736(84)90167-6 (1984). 22 Brandl, M., Bachmann, D., Drechsler, M. & Bauer, K. H. Liposome Preparation by a New High Pressure Homogenizer Gaulin Micron Lab 40. Drug Development and Industrial Pharmacy 16, 2167-2191, doi:10.3109/03639049009023648 (1990). 23 Wagner, A. & Vorauer-Uhl, K. Liposome technology for industrial purposes. J Drug Deliv 2011, 591325, doi:10.1155/2011/591325 (2011). 24 Olson, F., Hunt, C. A., Szoka, F. C., Vail, W. J. & Papahadjopoulos, D. Preparation of liposomes of defined size distribution by extrusion through polycarbonate membranes. Biochimica et Biophysica Acta (BBA) - Biomembranes 557, 9-23, doi:https://doi.org/10.1016/0005-2736(79)90085-3 (1979). 25 Hope, M. J., Bally, M. B., Webb, G. & Cullis, P. R. Production of large unilamellar vesicles by a rapid extrusion procedure. Characterization of size distribution, trapped volume and ability to maintain a membrane potential. Biochimica et Biophysica Acta (BBA) - Biomembranes 812, 55-65, doi:https://doi.org/10.1016/0005-2736(85)90521-8 (1985). 26 Mayer, L. D., Hope, M. J. & Cullis, P. R. Vesicles of variable sizes produced by a rapid extrusion procedure. Biochimica et Biophysica Acta (BBA) - Biomembranes 858, 161-168, doi:https://doi.org/10.1016/0005-2736(86)90302-0 (1986). 27 Hunter, D. G. & Frisken, B. J. Effect of Extrusion Pressure and Lipid Properties on the Size and Polydispersity of Lipid Vesicles. Biophysical Journal 74, 2996-3002, doi:https://doi.org/10.1016/S0006-3495(98)78006-3 (1998). 28 Mui, B., Chow, L. & Hope, M. J. in Methods in Enzymology Vol. 367    3-14 (Academic Press, 2003). 29 Bally, M. B., Hope, M. J., Mayer, L. D., Madden, T. D. & Cullis, P. R. in Liposomes as Drug Carriers (ed G. Gregoriadis) Ch. 60, 841-851 (John Wiley and Sons, 1988). 30 Yamauchi, M. et al. Release of Drugs from Liposomes Varies with Particle Size. Biological and Pharmaceutical Bulletin 30, 963-966, doi:10.1248/bpb.30.963 (2007). 94  31 Szoka, F. & Papahadjopoulos, D. Procedure for preparation of liposomes with large internal aqueous space and high capture by reverse-phase evaporation. Proceedings of the National Academy of Sciences of the United States of America 75, 4194-4198 (1978). 32 Gharib, R., greige-gerges, H., Fourmentin, S., Charcosset, C. & Auezova, L. Liposomes incorporating cyclodextrin–drug inclusion complexes: Current state of knowledge. Vol. 129 (2015). 33 Brunner, J., Skrabal, P. & Hausser, H. Single bilayer vesicles prepared without sonication physico-chemical properties. Biochimica et Biophysica Acta (BBA) - Biomembranes 455, 322-331, doi:https://doi.org/10.1016/0005-2736(76)90308-4 (1976). 34 Alpes, H. et al. Formation of large unilamellar vesicles using alkyl maltoside detergents. Biochimica et Biophysica Acta (BBA) - Biomembranes 862, 294-302, doi:https://doi.org/10.1016/0005-2736(86)90231-2 (1986). 35 Allen, T. M., Romans, A. Y., Kercret, H. & Segrest, J. P. Detergent removal during membrane reconstitution. Biochimica et Biophysica Acta (BBA) - Biomembranes 601, 328-342, doi:https://doi.org/10.1016/0005-2736(80)90537-4 (1980). 36 Ueno, M., Tanford, C. & Reynolds, J. A. Phospholipid vesicle formation using nonionic detergents with low monomer solubility. Kinetic factors determine vesicle size and permeability. Biochemistry 23, 3070-3076, doi:10.1021/bi00308a034 (1984). 37 Mozafari, M., J Reed, C., Rostron, C., Koçum, İ. & Piskin, E. Construction of stable anionic liposome-plasmid particles using the heating method: A preliminary investigation. Vol. 7 (2002). 38 Mozafari, M. R. Liposomes: an overview of manufacturing techniques. Cellular & molecular biology letters 10, 711-719 (2005). 39 Zhigaltsev, I. V. et al. Bottom-up design and synthesis of limit size lipid nanoparticle systems with aqueous and triglyceride cores using millisecond microfluidic mixing. Langmuir : the ACS journal of surfaces and colloids 28, 3633-3640, doi:10.1021/la204833h (2012). 40 Zhigaltsev, I. V., Tam, Y. K., Leung, A. K. K. & Cullis, P. R. Production of limit size nanoliposomal systems with potential utility as ultra-small drug delivery agents. Journal of Liposome Research 26, 96-102, doi:10.3109/08982104.2015.1025411 (2016). 41 Kastner, E. et al. High-throughput manufacturing of size-tuned liposomes by a new microfluidics method using enhanced statistical tools for characterization. International Journal of Pharmaceutics 477, 361-368, doi:https://doi.org/10.1016/j.ijpharm.2014.10.030 (2014). 42 Cullis, P. R., Mayer, L. D., Bally, M. B., Madden, T. D. & Hope, M. J. Generating and loading of liposomal systems for drug-delivery applications. Advanced Drug Delivery Reviews 3, 267-282, doi:https://doi.org/10.1016/0169-409X(89)90024-0 (1989). 43 Mayer, L. D., Hope, M. J., Cullis, P. R. & Janoff, A. S. Solute distributions and trapping efficiencies observed in freeze-thawed multilamellar vesicles. Biochim Biophys Acta 817, 193-196 (1985). 95  44 Herman, E. H., Rahman, A., Ferrans, V. J., Vick, J. A. & Schein, P. S. Prevention of chronic doxorubicin cardiotoxicity in beagles by liposomal encapsulation. Cancer Res 43, 5427-5432 (1983). 45 Deamer, D. W., Prince, R. C. & Crofts, A. R. The response of fluorescent amines to pH gradients across liposome membranes. Biochimica et Biophysica Acta (BBA) - Biomembranes 274, 323-335, doi:https://doi.org/10.1016/0005-2736(72)90180-0 (1972). 46 Nichols, J. W. & Deamer, D. W. Catecholamine uptake and concentration by liposomes maintaining pH gradients. Biochimica et Biophysica Acta (BBA) - Biomembranes 455, 269-271, doi:https://doi.org/10.1016/0005-2736(76)90169-3 (1976). 47 Mayer, L. D., Bally, M. B. & Cullis, P. R. Uptake of adriamycin into large unilamellar vesicles in response to a pH gradient. Biochim Biophys Acta 857, 123-126 (1986). 48 Haran, G., Cohen, R., Bar, L. K. & Barenholz, Y. Transmembrane ammonium sulfate gradients in liposomes produce efficient and stable entrapment of amphipathic weak bases. Biochim Biophys Acta 1151, 201-215 (1993). 49 Barenholz, Y. in Liposome Technology Vol. 2  (ed G. Gregoriadis) Ch. 1, 1-26 (Informa Healthcare 2007). 50 Fenske, D. B. et al. Ionophore-mediated uptake of ciprofloxacin and vincristine into large unilamellar vesicles exhibiting transmembrane ion gradients. Biochimica et Biophysica Acta (BBA) - Biomembranes 1414, 188-204, doi:https://doi.org/10.1016/S0005-2736(98)00166-7 (1998). 51 Gubernator, J. Active methods of drug loading into liposomes: recent strategies for stable drug entrapment and increased in vivo activity. Expert Opinion on Drug Delivery 8, 565-580, doi:10.1517/17425247.2011.566552 (2011). 52 Bulbake, U., Doppalapudi, S., Kommineni, N. & Khan, W. Liposomal Formulations in Clinical Use: An Updated Review. Pharmaceutics 9, doi:10.3390/pharmaceutics9020012 (2017). 53 Barenholz, Y. Liposome application: problems and prospects. Current Opinion in Colloid & Interface Science 6, 66-77, doi:https://doi.org/10.1016/S1359-0294(00)00090-X (2001). 54 Fritze, A., Hens, F., Kimpfler, A., Schubert, R. & Peschka-Süss, R. Remote loading of doxorubicin into liposomes driven by a transmembrane phosphate gradient. Biochimica et Biophysica Acta (BBA) - Biomembranes 1758, 1633-1640, doi:https://doi.org/10.1016/j.bbamem.2006.05.028 (2006). 55 van Etten, E. W. M., ten Kate, M. T., Snijders, S. V. & Bakker-Woudenberg, I. A. J. M. Administration of Liposomal Agents and Blood Clearance Capacity of the Mononuclear Phagocyte System. Antimicrobial Agents and Chemotherapy 42, 1677-1681 (1998). 56 Baas, J., Senninger, N. & Elser, H. [The reticuloendothelial system. An overview of function, pathology and recent methods of measurement]. Zeitschrift fur Gastroenterologie 32, 117-123 (1994). 96  57 Allen, T. M. & Cullis, P. R. Liposomal drug delivery systems: from concept to clinical applications. Adv Drug Deliv Rev 65, 36-48, doi:10.1016/j.addr.2012.09.037 (2013). 58 Wisse, E., Jacobs, F., Topal, B., Frederik, P. & De Geest, B. The size of endothelial fenestrae in human liver sinusoids: implications for hepatocyte-directed gene transfer. Gene Therapy 15, 1193+ (2008). 59 Bertrand, N. & Leroux, J.-C. The journey of a drug-carrier in the body: An anatomo-physiological perspective. Journal of Controlled Release 161, 152-163, doi:https://doi.org/10.1016/j.jconrel.2011.09.098 (2012). 60 Mebius, R. E. & Kraal, G. Structure and function of the spleen. Nature reviews. Immunology 5, 606-616, doi:10.1038/nri1669 (2005). 61 Sercombe, L. et al. Advances and Challenges of Liposome Assisted Drug Delivery. Frontiers in Pharmacology 6, 286, doi:10.3389/fphar.2015.00286 (2015). 62 Semple, S. C., Chonn, A. & Cullis, P. R. Influence of Cholesterol on the Association of Plasma Proteins with Liposomes. Biochemistry 35, 2521-2525, doi:10.1021/bi950414i (1996). 63 Bisgaier, C. L., Siebenkas, M. V. & Williams, K. J. Effects of apolipoproteins A-IV and A-I on the uptake of phospholipid liposomes by hepatocytes. The Journal of biological chemistry 264, 862-866 (1989). 64 Semple, S. C., Chonn, A. & Cullis, P. R. Interactions of liposomes and lipid-based carrier systems with blood proteins: Relation to clearance behaviour in vivo. Adv Drug Deliv Rev 32, 3-17 (1998). 65 Chonn, A. & Cullis, P. R. Ganglioside GM1 and Hydrophilic Polymers Increase Liposome Circulation Times by Inhibiting the Association of Blood Proteins. Journal of Liposome Research 2, 397-410, doi:10.3109/08982109209010217 (1992). 66 Senior, J. H., Trimble, K. R. & Maskiewicz, R. Interaction of positively-charged liposomes with blood: implications for their application in vivo. Biochimica et Biophysica Acta (BBA) - Biomembranes 1070, 173-179, doi:https://doi.org/10.1016/0005-2736(91)90160-A (1991). 67 Knudsen, K. B. et al. In vivo toxicity of cationic micelles and liposomes. Nanomedicine: Nanotechnology, Biology and Medicine 11, 467-477, doi:10.1016/j.nano.2014.08.004 (2015). 68 Chonn, A., Semple, S. C. & Cullis, P. R. Association of blood proteins with large unilamellar liposomes in vivo. Relation to circulation lifetimes. The Journal of biological chemistry 267, 18759-18765 (1992). 69 Juliano, R. L. & Stamp, D. The effect of particle size and charge on the clearance rates of liposomes and liposome encapsulated drugs. Biochemical and Biophysical Research Communications 63, 651-658, doi:https://doi.org/10.1016/S0006-291X(75)80433-5 (1975). 70 Magin, R. L., Hunter, J. M., Niesman, M. R. & Bark, G. A. Effect of vesicle size on the clearance, distribution, and tumor uptake of temperature-sensitive liposomes. Cancer drug delivery 3, 223-237 (1986). 71 Harashima, H., Sakata, K., Funato, K. & Kiwada, H. Enhanced hepatic uptake of liposomes through complement activation depending on the size of liposomes. Pharmaceutical research 11, 402-406 (1994). 97  72 Liu, D., Mori, A. & Huang, L. Role of liposome size and RES blockade in controlling biodistribution and tumor uptake of GM1-containing liposomes. Biochim Biophys Acta 1104, 95-101 (1992). 73 Nayar, R., Hope, M. J. & Cullis, P. R. Generation of large unilamellar vesicles from long-chain saturated phosphatidylcholines by extrusion technique. Biochimica et Biophysica Acta (BBA) - Biomembranes 986, 200-206, doi:https://doi.org/10.1016/0005-2736(89)90468-9 (1989). 74 Allen, T. M. & Chonn, A. Large unilamellar liposomes with low uptake into the reticuloendothelial system. FEBS Letters 223, 42-46, doi:10.1016/0014-5793(87)80506-9 (1987). 75 Blume, G. & Cevc, G. Liposomes for the sustained drug release in vivo. Biochim Biophys Acta 1029, 91-97 (1990). 76 Klibanov, A. L., Maruyama, K., Torchilin, V. P. & Huang, L. Amphipathic polyethyleneglycols effectively prolong the circulation time of liposomes. FEBS Lett 268, 235-237 (1990). 77 Allen, T. M., Hansen, C., Martin, F., Redemann, C. & Yau-Young, A. Liposomes containing synthetic lipid derivatives of poly(ethylene glycol) show prolonged circulation half-lives in vivo. Biochim Biophys Acta 1066, 29-36 (1991). 78 Maeda, H., Wu, J., Sawa, T., Matsumura, Y. & Hori, K. Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review. Journal of Controlled Release 65, 271-284, doi:https://doi.org/10.1016/S0168-3659(99)00248-5 (2000). 79 Malinski, J. A. & Nelsestuen, G. L. Membrane permeability to macromolecules mediated by the membrane attack complex. Biochemistry 28, 61-70 (1989). 80 Comiskey, S. J. & Heath, T. D. Serum-induced leakage of negatively-charged liposomes at nanomolar lipid concentrations. Biochemistry 29, 3626-3631, doi:10.1021/bi00467a006 (1990). 81 Rodrigueza, W. V., Haydn Pritchard, P. & Hope, M. J. The influence of size and composition on the cholesterol mobilizing properties of liposomes in vivo. Biochimica et Biophysica Acta (BBA) - Biomembranes 1153, 9-19, doi:https://doi.org/10.1016/0005-2736(93)90270-A (1993). 82 Nagayasu, A., Shimooka, T. & Kiwada, H. Effect of Vesicle Size on in Vivo Release of Daunorubicin from Hydrogenated Egg Phosphatidylcholine-Based Liposomes into Blood Circulation. Biological & Pharmaceutical Bulletin 18, 1020-1023, doi:10.1248/bpb.18.1020 (1995). 83 Johnston, M. J. et al. Therapeutically optimized rates of drug release can be achieved by varying the drug-to-lipid ratio in liposomal vincristine formulations. Biochim Biophys Acta 1758, 55-64, doi:10.1016/j.bbamem.2006.01.009 (2006). 84 Maurer, N., Wong, K. F., Hope, M. J. & Cullis, P. R. Anomalous solubility behavior of the antibiotic ciprofloxacin encapsulated in liposomes: a 1H-NMR study. Biochimica et Biophysica Acta (BBA) - Biomembranes 1374, 9-20, doi:https://doi.org/10.1016/S0005-2736(98)00125-4 (1998). 85 Gabizon, A. et al. Prolonged circulation time and enhanced accumulation in malignant exudates of doxorubicin encapsulated in polyethylene-glycol coated liposomes. Cancer Res 54, 987-992 (1994). 98  86 Gabizon, A., Shmeeda, H. & Barenholz, Y. Pharmacokinetics of pegylated liposomal Doxorubicin: review of animal and human studies. Clinical pharmacokinetics 42, 419-436, doi:10.2165/00003088-200342050-00002 (2003). 87 Chlebowski, R. T. Adriamycin (Doxorubicin) Cardiotoxicity: A Review. Western Journal of Medicine 131, 364-368 (1979). 88 Solomon, R. & Gabizon, A. A. Clinical Pharmacology of Liposomal Anthracyclines: Focus on Pegylated Liposomal Doxorubicin. Clinical Lymphoma and Myeloma 8, 21-32, doi:https://doi.org/10.3816/CLM.2008.n.001 (2008). 89 Forssen, E. A. The design and development of DaunoXome® for solid tumor targeting in vivo. Advanced Drug Delivery Reviews 24, 133-150, doi:https://doi.org/10.1016/S0169-409X(96)00453-X (1997). 90 Cowens, J. W. et al. Initial Clinical (Phase I) Trial of TLC D-99 (Doxorubicin Encapsulated in Liposomes). Cancer Research 53, 2796 (1993). 91 Harris, L. Liposome-encapsulated doxorubicin compared with conventional doxorubicin in a randomized multicenter trial as first-line therapy of metastatic breast carcinoma. Cancer 94, 25-36, doi:10.1002/cncr.10201 (2002). 92 Silverman, J. A. & Deitcher, S. R. Marqibo® (vincristine sulfate liposome injection) improves the pharmacokinetics and pharmacodynamics of vincristine. Cancer Chemotherapy and Pharmacology 71, 555-564, doi:10.1007/s00280-012-2042-4 (2013). 93 Mayer, L. D. et al. in 2016 CRS Annual Meeting. 94 Stentoft, J. The toxicity of cytarabine. Drug safety 5, 7-27 (1990). 95 Tolcher, A. W. & Mayer, L. D. Improving combination cancer therapy: the CombiPlex((R)) development platform. Future oncology (London, England), doi:10.2217/fon-2017-0607 (2018). 96 Lancet, J. E. et al. Final results of a phase III randomized trial of CPX-351 versus 7+3 in older patients with newly diagnosed high risk (secondary) AML. Journal of Clinical Oncology 34, 7000-7000, doi:10.1200/JCO.2016.34.15_suppl.7000 (2016). 97 Bullinger, L. & Fröhling, S. in Genomic and Personalized Medicine (Second Edition)   (ed Huntington F. Willard)  675-690 (Academic Press, 2013). 98 Rowley, J. D. in Encyclopedia of Genetics   (ed Jefferey H. Miller)  1088-1091 (Academic Press, 2001). 99 Rose-Inman, H. & Kuehl, D. Acute leukemia. Emergency medicine clinics of North America 32, 579-596, doi:10.1016/j.emc.2014.04.004 (2014). 100 Trendowski, M. The inherent metastasis of leukaemia and its exploitation by sonodynamic therapy. Critical Reviews in Oncology/Hematology 94, 149-163, doi:https://doi.org/10.1016/j.critrevonc.2014.12.013 (2015). 101 Chang, H. Y. et al. Causes of death in adults with acute leukemia. Medicine 55, 259-268 (1976). 102 Minot, G. R., Buckman, T. E. & Isaacs, R. Chronic myelogenous leukemia: Age incidence, duration, and benefit derived from irradiation. Journal of the American Medical Association 82, 1489-1494, doi:10.1001/jama.1924.02650450001001 (1924). 103 Antman, K. H. Introduction: the history of arsenic trioxide in cancer therapy. The oncologist 6 Suppl 2, 1-2 (2001). 99  104 Farber, S. & Diamond, L. K. Temporary remissions in acute leukemia in children produced by folic acid antagonist, 4-aminopteroyl-glutamic acid. N Engl J Med 238, 787-793, doi:10.1056/nejm194806032382301 (1948). 105 Thomas, A. Joe Burchenal and the birth of combination chemotherapy. British journal of haematology 133, 493-503, doi:10.1111/j.1365-2141.2006.06072.x (2006). 106 Frei, E., 3rd et al. The effectiveness of combinations of antileukemic agents in inducing and maintaining remission in children with acute leukemia. Blood 26, 642-656 (1965). 107 Pui, C.-H. & Evans, W. E. A 50-Year Journey to Cure Childhood Acute Lymphoblastic Leukemia. Seminars in hematology 50, 185-196, doi:10.1053/j.seminhematol.2013.06.007 (2013). 108 Freireich, E. J., Wiernik, P. H. & Steensma, D. P. The leukemias: a half-century of discovery. J Clin Oncol 32, 3463-3469, doi:10.1200/JCO.2014.57.1034 (2014). 109 Estey, E. & Dohner, H. Acute myeloid leukaemia. Lancet 368, 1894-1907, doi:10.1016/s0140-6736(06)69780-8 (2006). 110 Lichtman, M. A. A historical perspective on the development of the cytarabine (7days) and daunorubicin (3days) treatment regimen for acute myelogenous leukemia: 2013 the 40th anniversary of 7+3. Blood cells, molecules & diseases 50, 119-130, doi:10.1016/j.bcmd.2012.10.005 (2013). 111 Yates, J. W., Wallace, H. J., Jr., Ellison, R. R. & Holland, J. F. Cytosine arabinoside (NSC-63878) and daunorubicin (NSC-83142) therapy in acute nonlymphocytic leukemia. Cancer chemotherapy reports 57, 485-488 (1973). 112 Carella, A. M., Berman, E., Maraone, M. P. & Ganzina, F. Idarubicin in the treatment of acute leukemias. An overview of preclinical and clinical studies. Haematologica 75, 159-169 (1990). 113 Casciello, F., Windloch, K., Gannon, F. & Lee, J. S. Functional Role of G9a Histone Methyltransferase in Cancer. Frontiers in Immunology 6, 487, doi:10.3389/fimmu.2015.00487 (2015). 114 Lehnertz, B. et al. The methyltransferase G9a regulates HoxA9-dependent transcription in AML. Genes Dev 28, 317-327, doi:10.1101/gad.236794.113 (2014). 115 Kubicek, S. et al. Reversal of H3K9me2 by a small-molecule inhibitor for the G9a histone methyltransferase. Molecular cell 25, 473-481, doi:10.1016/j.molcel.2007.01.017 (2007). 116 Liu, F. et al. Optimization of cellular activity of G9a inhibitors 7-aminoalkoxy-quinazolines. J Med Chem 54, 6139-6150, doi:10.1021/jm200903z (2011). 117 Liu, F. et al. Discovery of an in vivo chemical probe of the lysine methyltransferases G9a and GLP. J Med Chem 56, 8931-8942, doi:10.1021/jm401480r (2013). 118 Johnston, M. J., Edwards, K., Karlsson, G. & Cullis, P. R. Influence of drug-to-lipid ratio on drug release properties and liposome integrity in liposomal doxorubicin formulations. J Liposome Res 18, 145-157, doi:10.1080/08982100802129372 (2008). 119 Zhigaltsev, I. V. et al. Liposome-encapsulated vincristine, vinblastine and vinorelbine: a comparative study of drug loading and retention. J Control Release 104, 103-111, doi:10.1016/j.jconrel.2005.01.010 (2005). 100  120 Anderson, M. & Omri, A. The effect of different lipid components on the in vitro stability and release kinetics of liposome formulations. Drug Deliv 11, 33-39, doi:10.1080/10717540490265243 (2004). 121 Avanti. Phase Transition Temperatures for Glycerophospholipids, https://avantilipids.com/tech-support/physical-properties/phase-transition-temps/ (2018). Accessed Date: 5/30/2018. 122 Snipstad, S. et al. Labeling nanoparticles: Dye leakage and altered cellular uptake. Cytometry A 91, 760-766, doi:10.1002/cyto.a.22853 (2017). 123 Chen, D. & Lee, K. H. Biodistribution of calcitonin encapsulated in liposomes in mice with particular reference to the central nervous system. Biochim Biophys Acta 1158, 244-250 (1993). 124 Gaillard, P. J. et al. Pharmacokinetics, brain delivery, and efficacy in brain tumor-bearing mice of glutathione pegylated liposomal doxorubicin (2B3-101). PLoS One 9, e82331, doi:10.1371/journal.pone.0082331 (2014). 125 Mayer, L. D., Bally, M. B. & Cullis, P. R. Strategies for Optimizing Liposomal Doxorubicin. Journal of Liposome Research 1, 463-480, doi:10.3109/08982109009036008 (1990). 126 Wehbe, M. et al. A simple passive equilibration method for loading carboplatin into pre-formed liposomes incubated with ethanol as a temperature dependent permeability enhancer. J Control Release 252, 50-61, doi:10.1016/j.jconrel.2017.03.010 (2017). 127 Ozer, J., Ratner, M., Shaw, M., Bailey, W. & Schomaker, S. The current state of serum biomarkers of hepatotoxicity. Toxicology 245, 194-205, doi:https://doi.org/10.1016/j.tox.2007.11.021 (2008). 128 Danpure, C. J. Lactate dehydrogenase and cell injury. Cell Biochemistry and Function 2, 144-148, doi:10.1002/cbf.290020305 (1984). 129 Komulainen, J. & Vihko, V. Exercise-induced necrotic muscle damage and enzyme release in the four days following prolonged submaximal running in rats. Pflügers Archiv 428, 346-351, doi:10.1007/BF00724517 (1994). 130 Kornberg, A. & Polliack, A. Serum lactic dehydrogenase (LDH) levels in acute leukemia: marked elevations in lymphoblastic leukemia. Blood 56, 351-355 (1980). 131 MayoClinic. Test ID TP: Protein, Serum, Total https://www.mayomedicallaboratories.com/test-catalog/Clinical+and+Interpretive/8520 (2018). Accessed Date: 5/30/2018. 132 Breyer, M. D. & Qi, Z. Better nephrology for mice--and man. Kidney international 77, 487-489, doi:10.1038/ki.2009.544 (2010). 133 NIH. Creatinine blood test https://medlineplus.gov/ency/article/003475.htm (2018). Accessed Date: 5/30/2018. 134 Tardi, P., Wan, C. P. L. & Mayer, L. Passive and semi-active targeting of bone marrow and leukemia cells using anionic low cholesterol liposomes. Journal of Drug Targeting 24, 797-804, doi:10.1080/1061186X.2016.1184669 (2016).  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items