• Angiogenesis (2003-2008)

Angiogenesis is the formation of new blood vessels that occurs during tumor initiation and growth and its inhibition is considered as an important therapeutic target.

Computational Modeling of Angiogenesis: As a graduate student in the laboratory of Prof. Aleksander Popel I developed computational models describing in molecular level detail the extracellular matrix proteolysis by endothelial cells and their migration in physiologically relevant in silico 3D tissues during angiogenesis. We were the first group to construct such models using experimental information.

Bioinformatic Identification of Antiangiogenic Peptides: Furthermore, I developed bioinformatics tools to accelerate the discovery of novel bioactive antiangiogenic peptides, biomolecules that inhibit the formation of new blood vessels. Such biomolecules were discovered primarily using proteomic approaches on biopsy samples, something that was slow and labor intensive. We were the first group to construct computational bioinformatic tools and discover more than 120 of such biomolecules, 4 times more than previously identified during the last 30 years of research in this field. 

In vitro and In vivo Peptide Screening in Tumor and Macular Degeneration Animal Models: As part of my graduate work I synthesized and screened for anti-angiogenic efficacy these peptides in vitro, and in vivo in various animal tumor models. My discovery has been co-developed by a number of collaborators at the Wilmer Eye Institute at Johns Hopkins as an anti-angiogenic treatment for Macular Degeneration and clinicians at the Sidney Kimmel Comprehensive Cancer Center as a tumor anti-angiogenic therapy. The impact of the discovery remains currently significant especially suppressing ocular neovascularization, and it is the focus of a start-up company in the Baltimore area (Asclepix Therapeutics LLC).

• Drug Delivery and Tissue Engineering (2009-2013)

As a post-doctoral research associate in the lab of Prof. Langer, I concentrated my research efforts in developing novel drug delivery vehicles for a variety of therapeutic molecules including peptides, proteins and oligonucleotides. The focus of such efforts was to establish a platform to integrate the material selection aspect of the drug delivery formulations with biodistribution and drug efficacy and accelerate the data driven, rational development and optimization of such formulations.

Development of Cell Penetrating Peptide Libraries for Drug Delivery Applications: I designed and synthesized libraries of novel cell penetrating peptides; such peptides are “magic bullets” that can deliver a variety of drugs in different tissues. In these studies I combined in silico molecular modeling techniques with experimental screening to design peptides that are more potent, biocompatible and less immunogenic than the so far known cell penetrating peptides. The efficacy of the peptides to deliver a variety of therapeutics including plasmid DNAs and small interfering RNAs was tested in vitro and in vivo. In combination with efficacy screening I developed novel techniques to study the biodistribution of the formulations using in vivo fluorescent imaging and x-ray based, micro Computed Tomography (CT) methods. That required the development of novel molecular probes as well as image processing algorithms to quantify their accumulation in tissues and correlate it with the activity of the delivered drugs in vivo. Our group was the first to create DNA origami based formulations for drug delivery and use the cell penetrating peptides to deliver the formulations intracellularly.

Application of Systems Biology Methods to Drug Delivery and Tissue Engineering Problems: Another significant problem that I had the opportunity to explore in the laboratory of Prof. Langer was evaluating the way that cells perceive and interact with synthetic polymer scaffolds and drug delivery vehicles, like nanoparticles, using systems biology approaches. In one example, I used graph theory as applied to protein-protein interaction networks in combination with high throughput screening of the genes expressed in human embryonic stem cells that grow on synthetic scaffolds of varying mechanical properties, to understand the major signaling pathways during the differentiation of the cells towards different lineages as a dynamical function of their mechanical microenvironment. In addition, such system biology computational tools provided a platform to recreate signaling pathways, in silico, associated with the endocytosis of nanoparticles and identify key contributors in such pathways using high-throughput imaging methods.

Molecular Dynamics Simulations-based Design of Nanoparticles: Furthermore, I started exploring how other computational tools like molecular dynamics simulations can help us understand the way that nanoparticles interact with the cellular plasma membranes. I established a collaboration with Accelrys, a leading company developing molecular dynamics software, to create tools that can be used specifically for studying the interaction of nanoparticles with cellular plasma membranes. Such approaches will eventually help us studying the mechanisms driving such interactions and identify a set of principles to design novel materials to formulate nanoparticles based on first principles. 

• High Resolution Optical Microscopy Methods for Reconstructing the Brain Connectome (2014-now)

Optical microscopy is fast and allows the high-throughput collection of multiplexed data, though it suffers from relatively low resolution occurring at the common diffraction limited optical microscopes, limiting it to ~250nm. Expansion Microscopy, a new super-resolution optical microscopy modality developed at Prof. Boyden's lab addresses the resolution limitation issue of optical microscopes by utilizing the physical magnification, the expansion, that occurs in a sample that is embedded in a swelling polyelectrolyte gel. Current ExM protocols can achieve up to 10-15nm resolution in a common diffraction limited optical microscope.

As a research scientist in the laboratory of Prof. Ed Boyden I am developing tools to enable all scientists to routinely image entire shapes and architectures (structural information) of single neurons, or neuronal networks using Expansion Microscopy (ExM). In addition I am developing methods to preserve the total proteomic information in the ExM polyelectrolyte gels to allow the combinatorial imaging of structural and functional information with nanometer resolution. 

ExM Compatible Lipid Labeling: To take advantage of the high resolution achieved with ExM using a common confocal microscope and observe lipids in the nanometer, I have developed two methods for labeling lipids in fixed tissues. The first is based on a new family of lipid intercalating small cationic peptides. The second is a method that achieves direct chemical modifications of lipids to facilitate the covalent conjugation of labels compatible to Expansion Microscopy. Chemical modification of lipids yields denser labeling of the cellular membranes for ExM, similarly to osmium tetraoxide labeling for electron microscopy.

Conservation of Total Proteomic Information in the ExM hydrogels: The current ExM protocol requires tissue processing with an enzyme that non-specifically cleaves all of the proteins. Such protein cleavage eliminates the mechanical stress induced by the mechanical forces transduced by the protein structures and facilitates the isotropic expansion of the gel (the quantified distortion in the expanded gel after proteolysis is in the nanometer order of magnitude). Although such step is critical for the preservation of spatial information in the expanding gel, it practically eliminates most of the proteomic information as all of the proteins are cleaved. To overcome the limitations related to isotropic expansion of hydrogels in ExM and preserve the total proteomic information, I have synthesized cleavable fixatives that facilitate reversible tissue fixation. By reversing the tissue fixation there is no need for enzymatic treatment of the hydrogels to achieve isotropic expansion, and the total proteomic information is preserved in the expanded gel.