Dr. Joel Sunshine

Dr. Sunshine


This thesis discusses the development of biodegradable polymers, nanoparticles, and microparticles for gene delivery and immune activation. The bulk of this thesis focuses on trying to understand basic principles important to the development of polymer-based gene delivery nanoparticles and acellular artificial antigen presenting cells (aAPC) for CD8+ T cell activation, In addition to the basic work, these technologies can be applied to many areas of human health and I have focused on applications in cancer and ophthalmology.

The root cause of many diseases has a genetic component, from single gene disorders like hemophilia to multifactorial disorders like cancer. As a result, gene therapy has enormous therapeutic potential if it can be done safely and efficiently. The vast majority of the effort into gene therapy has been directed at co-opting viruses as vectors for gene delivery, as viruses have evolved to be supremely efficient at getting their genetic information into foreign cells, but viral gene therapy has been hampered by high profile setbacks in clinical trials, owing to the potential for insertional mutagenesis and overly aggressive host immune responses to the vector. Theoretically, non-viral gene therapy should overcome these limitations by reducing the host response, enabling unlimited cargo capacity, and should be easier to produce and standardize. However, non-viral gene therapy has thus far been unable to achieve the required high transfection efficacies seen with viruses. A class of synthetic cationic polymers, poly(ß-amino)esters (PBAEs), have shown promise, but in order to develop next-generation synthetic polymer vectors for gene delivery, a deeper understanding of the relationship between polymer design and functional outcomes is required. In this thesis, I investigated structureiii function relationships within a library of PBAEs that we developed, with an eye towards investigating the impact of polymer properties on critical barriers to intracellular delivery. To extend this work, we looked to develop PBAEs for non-viral gene delivery to the eye, by investigating how different particle formulations might enable differential delivery to ocular cell types, and by performing a pilot study looking at in vivo gene delivery to the mouse retina via subretinal injection. We found that polymer hydrophobicity was a critical dimension that significantly effected polymer vector performance. We also found that the amine termini of PBAEs were critical to vector function at the level of nanoparticle uptake even though they did not substantially alter any putative key nanoparticle properties. We found that polymer formulations that worked well for one cell class worked well for another cell type within that class but may not work well for a different cell class, and demonstrated that PBAEs could engender high levels of gene expression in the mouse retina.

Tumor immunotherapy requires the activation of cytotoxic T lymphocytes (CTLs) against tumor-specific targets. This activation process occurs in vivo through the interaction of activated antigen presenting cells (APCs) with CD8+ T-cells in the lymph nodes. As an alternative to inducing biological APCs to create the targeted response of interest in the CD8+ T cell population, acellular artificial antigen presenting cells (aAPC) have been designed that mimic biological APC by presenting proteins for signal 1 (antigen specificity) and signal 2 (costimulation) on the surface of spherical particles. When activated, biological APCs undergo significant changes to surface composition and surface morphology; however, in the quest to mimic this process and engender immune responses with aAPCs, the focus has been squarely on the changes to surface protein composition. We hypothesized that artificial antigen presenting cell (aAPC) shape (or morphology) is a critical parameter that modulates T-cell activation and proliferation. We hypothesized that high aspect ratio ellipsoidal aAPCs, rather than spherical aAPCs, might enable increased contact between aAPCs and T-cells, result in enhanced T-cell activation in vitro, and mediate enhanced aAPC-based tumor killing in vivo in melanoma mouse models. To this end, we fabricated ellipsoidal aAPCs using PLGA microparticles, and tested the effects of shape on T-cell activation and tumor prevention by aAPCs in vitro and in vivo. We found that ellipsoidal aAPCs were substantially more efficient than their spherical counterparts at CTL activation, that increased aspect ratio resulted in increased activation, and that ellipsoidal aAPCs reduced / delayed tumor growth and increased mouse survival as compared to spherical and non-cognate controls in a melanoma tumor prevention study in vivo.

May, 2013

Green Lab Holiday Party


Football-Shaped Particles Bolster The Body’s Defense Against Cancer

Researchers at Johns Hopkins have succeeded in making flattened, football-shaped artificial particles that impersonate immune cells. These football-shaped particles seem to be better than the typical basketball-shaped particles at teaching immune cells to recognize and destroy cancer cells in mice.

“The shape of the particles really seems to matter because the stretched, ellipsoidal particles we made performed much better than spherical ones in activating the immune system and reducing the animals’ tumors,” according to Jordan Green, Ph.D., assistant professor of biomedical engineering at the Johns Hopkins University School of Medicine and a collaborator on this work. A summary of the team’s results was published online in the journal Biomaterials on Oct. 5.

According to Green, one of the greatest challenges in the field of cancer medicine is tracking down and killing tumor cells once they have metastasized and escaped from a tumor mass. One strategy has been to create tiny artificial capsules that stealthily carry toxic drugs throughout the body so that they can reach the escaped tumor cells. “Unfortunately, traditional chemotherapy drugs do not know healthy cells from tumor cells, but immune system cells recognize this difference. We wanted to enhance the natural ability of T-cells to find and attack tumor cells,” says Jonathan Schneck, M.D., Ph.D., professor of pathology, medicine and oncology.

In their experiments, Schneck and Green’s interdisciplinary team exploited the well-known immune system interaction between antigen-presenting cells (APC) and T-cells. APCs “swallow” invaders and then display on their surfaces chewed-up protein pieces from the invaders along with molecular “danger signals.” When circulating T-cells interact with APCs, they learn that those proteins come from an enemy, so that if the T-cells see those proteins again, they divide rapidly to create an army that attacks and kills the invaders.

According to Schneck, to enhance this natural process, several laboratories, including his own, have made various types of “artificial APCs” – tiny inanimate spheres “decorated” with pieces of tumor proteins and danger signals. These are then often used in immunotherapy techniques in which immune cells are collected from a cancer patient and mixed with the artificial APCs. When they interact with the patient’s T-cells, the T-cells are activated, learn to recognize the tumor cell proteins and multiply over the course of several days. The immune cells can then be transferred back into the patient to seek out and kill cancer cells.

The cell-based technique has had only limited success and involves risks due to growing the cells outside the body, Green says. These downsides sparked interest in the team to improve the technique by making biodegradable artificial APCs that could be administered directly into a potential patient and that would better mimic the interactions of natural APCs with T-cells. “When immune cells in the body come in contact, they’re not doing so like two billiard balls that just touch ever so slightly,” explains Green. “Contact between two cells involves a significant overlapping surface area. We thought that if we could flatten the particles, they might mimic this interaction better than spheres and activate the T-cells more effectively.”

To flatten the particles, two M.D./Ph.D. students, Joel Sunshine and Karlo Perica, figured out how to embed a regular batch of spherical particles in a thin layer of a glue-like compound. When they heated the resulting sheet of particles, it stretched like taffy, turning the round spheres into tiny football shapes. Once cooled, the film could be dissolved to free each of the microscopic particles that could then be outfitted with the tumor proteins and danger signals. When they compared typical spherical and football-shaped particles – both coated with tumor proteins and danger signals at equivalent densities and mixed with T-cells in the laboratory – the T-cells multiplied many more times in response to the stretched particles than to spherical ones. In fact, by stretching the original spheres to varying degrees, they found that, up to a point, they could increase the multiplication of the T-cells just by lengthening the “footballs.”

When the particles were injected into mice with skin cancer, the T-cells that interacted with the elongated artificial APCs, versus spherical ones, were also more successful at killing tumor cells. Schneck says that tumors in mice that were treated with round particles reduced tumor growth by half, while elongated particles reduced tumor growth by three-quarters. Even better, he says, over the course of a one-month trial, 25 percent of the mice with skin cancer being treated with elongated particles survived, while none of the mice in the other treatment groups did.

According to Green, “This adds an entirely new dimension to studying cellular interactions and developing new artificial APCs. Now that we know that shape matters, scientists and engineers can add this parameter to their studies,” says Green. Schneck notes, “This project is a great example of how interdisciplinary science by two different groups, in this case one from biomedical engineering and another from pathology, can change our entire approach to tackling a problem. We’re now continuing our work together to tweak other characteristics of the artificial APCs so that we can optimize their ability to activate T-cells inside the body.”

This work was supported by grants from the Johns Hopkins University Institute for NanoBioTechnology, the National Institute of Allergy and Infectious Diseases (AI072677, AI44129), the National Cancer Institute (CA108835) and the National Institute of Biomedical Imaging and Bioengineering (EB016721).



Dr. Stephany Tzeng

Dr. Tzeng


The fields of biomaterials, nanobiotechnology, and gene and drug delivery have all progressed over the past decades and have rapidly become a focus of research for many applications. In particular, gene delivery, with cargoes including DNA, small interfering RNA (siRNA), and short hairpin RNA (shRNA), is a very attractive tool for research purposes as well as clinical application. The ability to change a cell’s expression at the genetic level affords researchers great flexibility in studying a cell’s behavior in relation to its gene expression in a laboratory setting. More translational applications for which gene therapy can be a useful tool include cancer therapy and regenerative medicine. For the latter, cells must be directed to grow or behave in strictly defined manners, an issue that is often addressed via administration of soluble factors or spatial or mechanical cues during the cell culture period. While these are by no means strategies to be disregarded, cells can be guided more directly using gene therapy. For example, in some cases, stem cell differentiation is controlled primarily by or inhibited by known factors. While designing drugs to target the specific proteins of interest is dependent on protein structure as well as the ability to deliver the drug, knowing the gene sequence could allow us to deliver or suppress the gene directly, bypassing undruggable protein targets.

In the case of disease treatment, many diseases, including inherited and some acquired diseases like cancer, are genetic in origin or are affected by the patient’s genetic background. The biodegradable polymer nanoparticles we have designed are able to combat such diseases by changing the gene expression of cancer cells, such as by decreasing their expression of survival factors or causing them to overexpress apoptotic factors that cause cell death. Unlike traditionally studied viral methods, our synthetic nanoparticle system avoids many of the safety concerns surrounding viruses, including toxicity, severe inflammatory or immune response, and the potential for insertional mutagenesis. Non-viral gene delivery is an exceedingly versatile tool; as will be shown below, many types of therapeutic nucleic acids can be delivered using our polymers, and once delivery to a given cell type is optimized, the sequence of the genes being delivered do not affect the properties of the nanoparticles, therefore allowing for essentially any gene to be delivered using our system.

Poly(beta-amino ester)s (PBAEs), a newer class of biomaterials effective in gene delivery, have been developed for DNA and siRNA delivery to human GBM cells. Importantly, specific chemical structures that have a strong effect on delivery of one nucleic acid type over the other have been identified, and overall trends in polymer structures have been correlated with transfection efficacy. PBAEs that can transfect 2-D and 3-D brain cancer cultures while having minimal effect on fetal brain cells have also been found. This phenomenon was verified in a different species and tissue type, namely rat liver cancer and healthy hepatocytes transfected in a co-culture system. Tumor cell death was caused in vitro after transfection with functional DNA coding for suicide genes or other methods of causing cancer cell apoptosis. DNA and/or siRNA delivery of functional genes was also used for stem cell engineering and regenerative medicine by causing overexpression or knockdown of transcription factors. In this series of applications, siRNA delivery to bone marrow-derived mesenchymal stem cells caused enhanced osteogenic differentiation; DNA delivery to embryonic stem cell-derived neural stem cells caused enhanced neuronal differentiation and maturation; and DNA delivery to adipose-derived mesenchymal stem cells induced secretion of growth factors to enhance vascularization of these cultures for ischemia treatment. Finally, with an eye toward eventual translation of this technology to the clinic, we assessed the ability of PBAE/DNA nanoparticles to overcome in vivo barriers. Procedures to make this technology more translatable are detailed in this work, and we have also shown in a proof-of-concept experiment the ability of our nanoparticles to transfect cancer cells in vivo.

The work presented in this thesis can serve as a starting point for future transfections and for rationally designing PBAEs with the most effective structures. In addition, the PBAE conditions discovered can be used for eventual in vivo studies of cancer. These methods may be used on their own for regenerative medicine and cancer therapy applications or could serve as a complementary tool along with conventional strategies and treatments in the future.

Jan 2014

Dr. Nupura Bhise

Dr. Bhise


Gene therapy involves the delivery of deoxyribonucleic acid (DNA) into cells to override or replace a malfunctioning gene for treating debilitating genetic diseases, including cancer and neurodegenerative diseases. In addition to its use as a therapeutic, it can also serve as a technology to enable regenerative medicine strategies. The central challenge of the gene therapy research arena is developing a safe and effective delivery agent. Since viral vectors have critical immunogenic and tumorogenic safety issues that limit their clinical use, recent efforts have focused on developing non-viral biomaterial based delivery vectors. Cationic polymers are an attractive class of gene delivery vectors due to their structural versatility, ease of synthesis, biodegradability, ability to self-complex into nanoparticles with negatively charged DNA, capacity to carry large cargo, cellular uptake and endosomal escape capacity.

In this thesis, we hypothesized that developing a biomaterial library of poly(betaamino esters) (PBAE), a newer class of cationic polymers consisting of biodegradable ester groups, would allow investigating vector design parameters and formulating effective non-viral gene delivery strategies for cancer drug delivery, tissue engineering and stem cell engineering. Consequently, a high-throughput transfection assay was developed to screen the PBAE-based nanoparticles in hard to transfect fibroblast cell lines. To gain mechanistic insights into the nanoparticle formulation process, biophysical properties of the vectors were characterized in terms of molecular weight (MW), nanoparticle size, zeta potential and plasmid per particle count. We report a novel assay developedfor quantifying the plasmid per nanoparticle count and studying its implications for co-delivery of multiple genes. The MW of the polymers ranged from 10 kDa to 100 kDa, nanoparticle size was about 150 run, zeta potential was about 30 mV in sodium acetate buffer (25 mM, pH 5) and 30 to 100 plasmids were associated with a single polymeric nanoparticle.

To develop PBAE vectors for application in cancer drug delivery and 3-D tissue engineered cultures, the gene delivery efficacy of PBAE nanoparticles was evaluated in mammary epithelial cells used as a model for studying normal development of mammary gland as well as the events that lead to development of breast cancer. We investigated how small molecular changes to the end-capping terminal group of the polymer and changes to the polymer MW affect gene delivery in 2-D mammary cell culture compared to 3-D primary organotypic cultured mouse mammary tissue. We reported that the polymers synthesized here are more effective for gene deliverythan FuGENE ® HD, one of the leading commercially available reagents for non-viral gene delivery. We also highlighted that transfection of the 3-D organotypic cultures is more difficult than transfection of 2-D cultures, but likely models some of the key challenges for in vivo gene therapy more closely than 2-D cultures. Finally, we evaluated the use of PBAE nanotechnology for genetic manipulation of stem cell fate for regenerative medicineapplications. We developed a PBAE nanoparticle based non-viral protocol and compared it with an electroporation based approach to deliver episomal plasmids encoding reprogramming factors for derivation of human induced pluripotent stem cells (hiPSC). The hiPSCs generated using these approaches can be differentiated into specific cell types for in vitro disease modeling and drug screening, specifically to study retinal degeneration.

Dec 2014