Dr. Corey Bishop

dr bishop color final copy












Gene therapy is promising because nearly all inheritable diseases and cancer has an underlying genetic component, regardless of the cause. However, the development of a gene vector capable of delivering nucleic acids efficiently while remaining safe has been a challenge.

Generally speaking, viral vectors are highly adept at delivering nucleic acids intracellularly to self-replicate. Viral vectors, however, have been known to be associated with insertional mutagenesis and immunogenicity and in severe cases, death. Furthermore, viral vectors are difficult to chemically modify for optimization and to mass produce.

To date, despite that there have been more than 2000 gene therapy clinical trials worldwide, the Chinese FDA and the European Medicines Agency have been the only regulatory bodies to approve gene therapies; both approved therapies are viral-based. The Chinese FDA in 2003 approved an adenovirus delivering nucleic acid encoding P53 for head and neck squamous cell carcinoma.The European Medicines Agency approved an associated-adenovirus in 2012 encoding lipoprotein lipase, which is injected intramuscularly for hypertriglyceridemia. The U.S. FDA, however, has yet to approve a gene therapy application.

The lack of gene therapy approval by regulatory bodies delineates the need for safer and more effective vectors. A safer alternative to viral-based vectors are cationic, amine-containing polymeric vectors. Cationic polymer is capable of ionically complexing nucleic acid and forming nanoparticles on the order of 150-250 nm. This diameter, typically less than 400 nm or so, is capable of extravasating through mal-formed tumor vasculature. Tumor drainage is also typically compromised due to poorly formed vasculature. With improved extravasation and compromised drainage, these polymeric vectors are passively targeted to tumors – this is known as the enhanced permeability and retention effect.

The buffering ability of the polymer allows the nucleic acid cargo to escape the lysosomal degradation pathway post-endocytosis. As proton pumps within the phospholipid bilayer membrane endeavor to acidify the endosomes, the polymer buffers the protons, causing an imbalance in the charge and osmotic pressure due to the influx of chloride ions and water molecules, respectively. As the influx of water continues, the endosomes effectively burst, releasing the unscathed nucleic acid cargo from the degrading acidic environment of the lysosomes. This process of the cargo escaping endosomes through buffering of the polymer gene delivery vector is known as the proton sponge effect.

Ester-containing backbone polymers are capable of hydrolytic degradation which helps mitigate toxicity. Furthermore, polymeric vectors can be chemically modified for optimizing nucleic acid delivery and can be easily mass produced, in contrast to viral vectors.

However, it is unknown which polymer structures work well for gene delivery. To date, the polymer vectors which work well have been discovered through high throughput screening methods (although which polymer structure will work well is unknown, general desirable properties are known, i.e., degradable, capable of buffering and ionically complexing anionic nucleic acid). Understanding how polymer structure affects gene delivery function would allow for a rational approach to designing new vectors for gene delivery.

The objective of this work has been two-fold: the first objective has been to elucidate non-viral gene delivery barriers – in particular investigating polymer structure-cell function relationships for polymeric vectors; the second objective has been to develop a polymeric/inorganic hybrid, theranostic-enabling nanoparticle platform technology capable of co-delivering DNA and siRNA from a single formulation, as well as deliver two types of plasmids with differing expression time profiles.

More specifically, this work details how polymer structure affects polymer-DNA binding and how binding affects transfection levels, viability, zeta potential and diameter of an ester-, amine-containing polymer, poly(β-amino ester). We found transfection levels are biphasic with respect to binding in two human cancer cell lines. We found that binding constants in the range 4 -1 of (1-6)x10 M were necessary but not sufficient for optimal transfection. We also investigated the comparative binding strengths of branched and linear polyethyleneimine, poly(L-lysine) and PBAE with plasmid DNA and found PBAE has the weakest binding.

The development of a theranostic-enabling platform technology involved a layer-by-layer process which is also detailed herein. As mentioned these nanoparticles are polymer-inorganic material hybrids; the inorganic material of choice was gold nanoparticles. Gold nanoparticles are relatively biocompatible, are monodisperse, have interesting optical properties which can reveal the state of the gold nanoparticle, have photothermal capabilities, are easily chemically modified via thiol binds and are capable of passive and active targeting (thiolation of ligands which have receptors upregulated on cancer cells, i.e., folic acid and epidermal growth factor receptor, etc.).

Gold nanoparticles can be synthesized to absorb a desired wavelength of electromagnetic radiation. This phenomenon is known as the surface plasmon resonance. If the gold nanoparticles’ surface plasmon resonance is in the near infrared region, then the gold nanoparticles can be excited remotely, outside the body on demand. Near infrared light can penetrate biological tissue on the order of a couple of centimeters. Furthermore, the gold nanoparticles can be imaged through various modalities either directly or indirectly via photoacoustics or X-rays for diagnostic purposes.

The theranostic-enabling technology was capable of co-delivering DNA and siRNA; as well as delivering two layers of DNA with two different expression time profiles. As a potential application, co-delivering DNA and siRNA would allow for the knockdown of a dysfunctional aberrant protein which would also be replaced with a functional protein; or the knockdown of a vital growth factor and the expression of a protein which would mediate apoptosis (i.e., TNF-α, IL-2,4, or secretable TRAIL). The ability to express protein with different time profiles would have implications in controlling stem cell differentiation where the timing of protein presentation is critical.

This work also details a more high throughput method for assessing cellular and nuclear uptake rates using flow cytometry. This method may be used for elucidating structure-function relationships in various cell types. An auto-fitting, first order mass-action kinetic model was developed in MatLab to quantify the rate constants for comparing bottlenecks of various polymer structures in various cell lines. This model was used to assess rate differences between polymers which do not transfect well, tansfect mediocre, and transfect well in primary human glioblastoma in vitro. The model recapitulated the experimental data with good agreement without extrapolating data from literature.

Principal component analysis is a method to look at large data sets with unknown variable correlations and to quantify how each variable is correlated with another, as well as which and to what degree each variable may drive another. Principal component analysis was utilized to look at 27 physico-chemical properties and cell gene delivery outcomes (i.e., uptake, transfection levels, and viability). We found that LogP, the partition coefficient between two immiscible liquids, and an indirect measurement of hydrophobicity – the sum of the number of carbons between the esters in the backbone and the number of carbons between the amine and the alcohol of the sidechain, drive uptake and transfection.

May, 2015