1. Doctoral research- Stony Brook University

I am currently working in Dr. Ira Cohen’s laboratory studying repair of myocardium. Specifically, I investigate mechanisms of atrial regeneration. I am learning a lot about cardiac physiology and pathology, stem cell molecular biology, biomaterials, surgical preparations and imaging. I am co-advised by Dr. Glenn Gaudette (a recent SB BME faculty member) who is now an Assistant Professor of Surgery at the University of Massachusetts. I am very excited about the work in my laboratory and look forward to spending more time on my projects.

2. Graduate laboratory rotations- Stony Brook University

In the summer before starting medical school, I performed a short rotation in Dr. Yi Xian Qin’s laboratory. Here I worked with a post-doctoral fellow to initiate the use of wavelet analysis to study ultrasound signals from bone scans. The following summer, I studied under Dr. Molly Frame, learning about microcirculation, fluid mechanics in a hamster cheek pouch model and angiogenesis. This background in vascular physiology nicely complemented my passion for pathology and atherosclerosis (see below). Since I had spent 4 years studying vascular disease at the University of Pennsylvania, I had a good sense of the literature in the field. Dr. Frame and I realized that almost every study of endothelial dysfunction in atherosclerosis focused on the role of these cells in large vessels (carotid arteries, aorta, renal branch arteries, etc.). Recognizing that microvascular endothelial cells (ECs) outnumber macrovascular ECs by over 1000-fold and that these cells are not protected from the upregulated inflammatory and oxidative state in the circulation, we proposed to study the role of microvascular ECs in atherosclerosis.

3. Undergraduate Research at the Institute for Medicine and Engineering- University of Pennsylvania

I began working at the Institute for Medicine and Engineering as a freshman at Penn in 1998. I was mentored initially by Dr. Victor Rizzo and later by Dr. Brian Helmke and Dr. Peter Davies. My major research activity at the University of Pennsylvania was a project measuring spatiotemporal deformation of cytoskeleton in living endothelial cells under hemodynamic flow. The study aimed to characterize the strain in the cytoskeletal network in response to apical shear stress. Endothelial dysfunction is believed to be the primary event in the formation of atherosclerotic plaques. These lesions form in predictable locations of the vascular tree, described by disturbed flow with low mean shear stresses. Regions of vessels where flow is laminar and mean shear stress is high are protected from atherogenesis. Thus, regional hemodynamics is a critical determinant in the specificity of the disease. Many studies have attempted to determine the molecular reasons for this, attributing much to mechanotransduction events. Our work was important because we were the first group to measure intracellular strain using an endogenous marker. Our studies and methods can now be paired with molecular approaches to form a cohesive picture of how endothelium responds to different mechanical environments.

We transfected bovine aortic endothelial cells with a green fluorescent protein (GFP) plasmid to visualize the cytoskeletal intermediate filament vimentin. GFP-vimentin was visualized at steady state and then flow was initated at 12 dyne/cm². Epifluorescence images were acquired of single cells at three different z-planes: 0.3 mm, 2.4 mm, and 4.6 mm above the cover slip. Deconvolution and image reconstruction were performed. One of my major contributions to the project was a robust image processing scheme to measure strain from the fluorescence image slices. This involved a series of filters, thresholding and binary operations in MatLab and IDL. We identified vertices in the cytoskeletal network and tracked the movement of these vertices in time.The algorithm implemented a Delaunay triangulation to connect neighboring points in a network of triangles. For each triad in the sub-region, over a time interval, the Lagrangian strain tensor, E_ij was computed. Principal values of strain E_I and E_II were computed as the eigenvalues of E_ij. The principal stretch ratios λ_I, and λ_II were computed from the principal strains. The product of the two stretch ratios was also computed to verify volume conservation in a given cell slice. The algorithm outputted color maps of regional magnitude and direction of strain for each of the z-sections and for each time interval (before flow onset and after terminating flow).

Our data showed that apical forces (in this case, fluid shear stresses) are transferred across the membrane to the cell lumen via changes in strain in the cytoskeletal network. We found alignment of strain vectors and strain focusing at cell-cell junctions. These are often sites of important molecular signaling events, emphasizing the role of mechanics in signal transduction.

Findings from this project were published in April 2003:

Helmke, B.P., Rosen, A.B., and Davies, P.F. Mapping mechanical strain of an endogenous cytoskeletal network in living endothelial cells. Biophysical Journal. 84:2691-2699, 2003.

3. Undergraduate Research at the Institute for Medicine and Engineering, Part II- University of Pennsylvania

Beginning in the summer of 2001, Drs. Davies and Helmke and I began using a 2-D video microscopy system to address the question of whether the cytoskeletal deformation was a passive or active process. We developed a novel algorithm to measure the time constant for deformation by studying temporal correlation maps of images at 1-second intervals after the onset of flow. Preliminary findings from that study showed that the deformation of vimentin cytoskeleton is a passive process, occurring well before the time constant for protein remodeling.

4. Senior Design Project in Bioengineering- University of Pennsylvania

For my Senior Design Project for the bioengineering major, I designed computer algorithms for image processing in a Radiology laboratory. In the fall of 2001, I began working on a project with Dr. Leon Axel, who was developing techniques to quantify cardiac mechanics from tagged magnetic resonance (MR) images. To consider the relationship between heart disease and myocardial mechanics, correlative techniques have been developed with the use of tagged MRI. This process creates a spatial modulation in the longitudinal magnetization of the heart (SPAMM), resulting in a tag on the myocardium. The process changes the magnetic spin of the tissue in a discrete region in space at an instantaneous time point; any subsequent motion of the material within that region is defined by the motion of the underlying tissue. Thus the tag, or region of diminished intensity in the 2-D image can be used to track motion of the heart wall and ultimately compute strain in the tissue. Dr. Axel and his graduate students had a challenge for me: they were trying to extract the image features (tags and heart wall contours) from the 2-D images, but they had no known model upon which to test their algorithms. I generated a computer phantom of a heart with user-defined geometry and dynamic properties in MatLab. Once the model was generated, the user could embed tags into the “heart,” define an imaging plane and generate synthetic images of tagged “myocardium.” The other members of the lab could then use these images to test their feature extraction algorithms against defined locations and movements of “heart” features.

5. Research at Brookhaven National Laboratories

As a junior in high school, I began working with Dr. Diane Cabelli in the Chemistry Department of Brookhaven National Labs. Dr. Cabelli mentored me as I performed work for which I was ultimately named a Semi-Finalist in the National Westinghouse Competition. The following is the abstract from this project:

Generation of OH Radicals from Superoxide Dismutase: A Potential Relationship to Amyotrophic Lateral Sclerosis

The genetic defect associated with familial Amyotrophic Lateral Sclerosis (fALS) is a single-site mutation on an enzyme, superoxide dismutase (SOD). Deleterious effects of fALS are suspected to be initiated by the generation of hydroxide free radicals from SOD and peroxide.Investigation of this reaction over various pH ranges and ionic concentrations may shed new light on SOD's role in ALS. This study revealed a difference between reactions of non-mutated and mutated SOD by suggesting that, in certain mutants, H2O2 is the reactant, while in native enzyme, it is HO2. This research also demonstrated that HO2 and O2 have similar ionic effects in the system.