UCSF

A plethora of drug delivery systems (DDS) have been designed over the years to maximize drug payload at tumor sites and reduce the off-target side effects associated with chemotherapeutics. DDS are generally classified by their mode of action and include: passive targeting, active targeting, and/or triggered release. Passive targeted DDS exploit the increased vascular permeability of hypoxic, inflamed, and cancerous tissues for tissue accumulation. Active targeted DDS incorporate targeting ligands that preferentially bind to over expressed receptors in cancer cells providing increased specificity. Nevertheless, off-target side effects still occur due to the lack of spatial control over drug release.

Triggered DDS provide a solution to uncontrolled drug release by eliciting release upon exposure to a disease associated stimuli. The degree of sophistication built into triggers varies by design, but the preference is for engineering simple and robust triggers that are disease-specific and can take advantage of the microenvironment cues of the diseased tissue. Although a number of triggerable systems have been engineered over the last two decades, there is ample opportunity for expanding the arsenal of triggers for drug delivery. In this thesis, I focused my efforts on designing triggerable delivery systems that are substrates for phosphatases, which are enzymes that can be over expressed in life-threatening diseases such as brain and prostate cancer. I report the design and engineering of novel phosphatase triggerable systems that mediate cytosolic delivery through the use of substrate masking and charge conversion.

The phosphatase triggerable systems were designed for cytosolic delivery of drugs. Two different triggering schemes were engineered. The first triggering mechanism was inspired by the fusion machinery of viruses. It was hypothesized that a phosphatase triggerable system could be engineered by masking the hydrophobic and membrane destabilizing segments of fusogenic peptides. Triggering would expose the full peptide sequence, which would partition into adjacent membranes. This membrane insertion event would provide enough enthalpic energy to drive membrane fusion and/or content release of drugs.

In order to design this mechanism, I first had to gain control over the membrane destabilizing properties of fusion peptides. In Chapter 2, it was hypothesized that phosphate substitutions would inactivate the membrane destabilizing properties of these peptides by masking their hydrophobicity. Moreover, phosphatase catalyzed dephosphorylation would regenerate the peptides and trigger lipid mixing and content release. Strategically positioned phosphate moieties were found to be effective at inactivating the fusion peptides, with two phosphates needed for complete loss of activity. The phosphopeptides were excellent substrates for phosphatases, which acted on them to regenerate their membrane destabilizing properties. This robust trigger mechanism was then used to design triggerable fusogenic liposomes.

In Chapter 3, it was hypothesized that fusogenic liposomes could be designed with surface-bound phosphopeptides, and that phosphatases would trigger these peptides to promote lipid mixing, cellular accumulation, and cytosolic content delivery. Fusogenic liposomes, termed FTP for fusogenic and triggerable phosphopeptide displaying liposomes, were found to be excellent substrates for phosphatases. Moreover, phosphatase triggering led to content release and lipid mixing of neighboring liposomes. In addition, FTP exhibited enhanced cellular accumulation as compared to non-triggerable liposomes and were able to mediate cytosolic delivery of their cargos. Although fusogenic liposomes derived from viral proteins provide an interesting template for triggerable DDS, they are complex in nature and might invoke an immune response upon repeated administration.

The second triggering mechanism exploited phosphatase catalyzed dephosphorylation as a method for converting an anionic liposome into a cationic liposome. Inspired by my electrical engineering background, surface charge converters (SCC) were engineered using inverse-phosphocholine lipids (CP) for triggered anionic-to-cationic charge conversion by phosphatases. Charge conversion was observed upon exposure of SCC to alkaline or acid phosphatases. A rapid increase in zeta potential was observed for SCC formulated with inverse-phosphocholine lipid DOCP. The net gain in zeta potential conversion was +80 mV, which is impressive for an enzyme triggered system. This surface charge conversion mechanism was exploited to design liposomes with enhanced content release and lipid mixing capabilities. SCC formulated with phosphatidylethanolamine and cholesterol were effective at lipid mixing with anionic membranes and delivering macromolecules into the cytoplasm of cells.

In Chapter 5, I provide mathematical insights for the use of infusion technique, such as convection and retro-convection, which can enable the delivery of phosphatase triggerable liposomes to the brain. I also provide mathematical justifications to support key assumptions that have been postulated for modeling these systems in the brain. I conclude by showing how multi-catheter arrays can be used for compartmentalizing drugs delivered into the brain interstitial area.

The thesis is divided into six chapters:

Chapter 1 provides background information on drug delivery systems, reviewing the major classes of delivery systems: passive, active, and triggerable. Furthermore, the chapter provides an overview of enzymatically-active systems, with a particular emphasis on phosphatase triggering.

Chapter 2 presents research that was completed to engineer fusogenic gp41 phosphopeptides. The biophysical and structural characterization of these peptides is presented.

Chapter 3 presents formulation and characterization studies for fusogenic and triggerable phosphopeptide displaying liposomes (FTP). I present formulation schemes as well as biophysical and in vitro cellular characterization.

Chapter 4 describes research performed to formulate and characterize surface charge converting liposomes. I present formulation schemes as well as biophysical and in vitro cellular characterization.

Chapter 5 provides insights into how mathematical modeling can set the limits for using advanced drug infusion and fluid removal techniques.

Chapter 6 provides a summary of this thesis, as well as recommendations for future designs of phosphatase triggerable systems.