Nilsson Holguin
Queens, NY
Trinity College
Incoming PhD Student
Biomedical Engineering
 
Faculty Advisor:
Partap Khalsa, Associate Professor
Department of Biomedical Engineering
 
Development of a scaling relationship to extrapolate neurophysiological data from feline to human spines during lateral bending.
Nilsson Holguin1(1), Allyson Ianuzzi(2), Partap S. Khalsa(2)
 
Rationale: Low back pain (LBP) is an epidemic in the U.S., with more than 80% of the adult populationexperiencing a severe episode during their lifetime. An etiology of LBP is the facet joint capsule (FJC), which is the ligament that encapsulates the bony, articulating joints of the spine. The lumbar FJC is innervated with mechanically sensitive neurons and functions to provide both proprioceptive and nociceptive (i.e., pain) information to the central nervous system. For ethical reasons, it is not possible to directly measure the responses of human FJC afferents, hence animal models are utilized. However, the neurophysiological data from a cat model cannot be directly extrapolated to humans because the relationship between the biomechanics of the two species is unknown. Therefore, the broad aim of this project is to develop a scaling relationship, based on the biomechanics between cat and human lumbar spines, for use in extrapolating neurophysiological data from cat to human.
 
Methods: Human (n=3) and cat (n=6) cadaveric lumbar spines were received, measured, potted and actuated in lateral bending (LB).The parameters measured were the yield force, torque (M) at L5-S1 in humans and L7-S1 in cats, moment of inertia (I) of the lumbar disc, and intervertebral angulations which was used to calculate the torsional joint stiffness (E). The radii of curvature (µ) of the spine specimens during LB at the torque limit were calculated as: µ = EI/M, and compared using regression techniques. If proven statistically viable, the radius of curvature would be used as a method of scaling the torque limits.
 
Results: A comparison of linear regression lines did not detect a significant difference (p= 0.84) between the cat and human data, suggesting that the way in which the radii of curvature (µ) change in both species is similar. The overall coincidence of the relationship had a low correlation when the data was pooled (R² = 0.07). The fit was significantly improved by comparing the data separately (p=0.05). The radius of curvature relationship (µ) was well correlated in the cat spines (R² = 0.73) but less so in the human spines (R² =0.25 ).
 
Discussion and Conclusions: The data suggest that while the radii of curvature (µ) of both species are not significantly different, one relationship alone cannot be used for both species, as the combined R² and p value suggest. This could be due to a difference in the curvature in the neutral posture between the species. However, if the difference in neutral position is introduced, a single relationship could be used to describe both data when the geometries of the specimens are introduced as a method for comparison. Future studies will include collecting data from additional human spines, which should improve the fit. The current study, along with further studies, will aid in the understanding and formulation of a scaling relationship between human and cat spines under LB and allow neurophysiological data from cats to be used to project how human nociceptors and mechanoreceptors respond to analogous biomechanical motions.
 
(1) Department of Biomedical Engineering, Trinity College, Hartford, CT 06106
(2) Department of Biomedical Engineering, Stony Brook University, Stony Brook, NY 11794-8018

Aluta Khanyile
Bronx, NY
Stony Brook University
Junior
Pharmacology
 
Faculty Advisor:
Howard Crawford, Assistant Professor
Department of Pharmacological Sciences
 
The Notch-1 pathway affects the expression of Matrix Metalloproteinase-7 in Pancreatic cancer
Aluta Khanyile, Howard Crawford
 
The expression of MMP-7 is seen in the early stages of pancreatic tumor progression. MMP-7 is found in significant amounts in 99‰ of patients suffering from pancreatic cancer. Notch, a protein that is critical for cell fate decisions and differentiation, has also been found to be significant in the formation of pancreatic cancer. Notch acts as a transcription factor via its interaction with the DNA-binding protein, CBF. We set out to test the hypothesis that MMP-7 is regulated by Notch. EBNA-2 is an adenoviral protein that activates promoter regions that have CBF binding sites. SUH-L is a mammalian protein that inhibits promoter regions that have CBF binding sites. By overexpressing EBNA-2 and SUH-L with a MMP-7 promoter/reporter construct we found that SUH-L has the ability to turn down the MMP-7 promoter, whereas EBNA-2 had no effect. While these data are consistent with Notch being a possible regulator of MMP-7 expression, the ineffectiveness of EBNA-2 needs to be examined further.
 
(1)Department of Pharmacological Sciences, Stony Brook University, Stony Brook, NY 11794-8651

Rachel A. Matundan
Queens, NY
College of Mount St. Vincent
Incoming PhD Student
Biochemistry
 
Faculty Advisor:
Caroline Kisker, Associate Professor
Department of Pharmacological Sciences
 
Determining the active site divalent cation coordination for the C-terminal endonuclease domain of UvrC
Rachel A. Matundan(1), James J. Truglio(2),Caroline Kisker(2)
 
The nucleotide excision repair (NER) mechanism involves the recognition and repair of many types of damaged DNA present in both eukaryotic and prokaryotic cells. In prokaryotes the three main proteins involved are UvrA, UvrB and UvrC. UvrA and UvrB are responsible for recognition of the damage, whereas the proper function of UvrC is imperative to the repair process because it involves the catalytic domains necessary for the incision of the damaged genetic material.
 
The structure of the C-terminal endonuclease domain of UvrC (amino acids 339-557 of UvrC from Thermotoga maritima) has been recently solved by x-ray crystallography. The active site is similar to that of RNAse H suggesting a similar enzyme mechanism. The active site pocket contains four highly conserved residues: D367, D405, D429 and H488. In the crystal structure, a Mn2+ ion is bound solely to H488. Mutation of this residue to Ala, Asp or Glu results in a UvrC variant that is still partially active but significantly impaired compared to wildtype UvrC. In this study, we sought to determine the crystal structures of mutants H488E and H488A in context of the UvrC C-terminal domain to obtain detailed information about the role of this residue in catalysis.
 
(1)Department of Biochemistry, College of Mount Saint Vincent, Riverdale, NY 10471
(2)Department of Pharmacological Sciences, Stony Brook University, Stony Brook, NY 11794-8651

Rosemary Orhue
Bronx, NY
Stony Brook University
Senior
Biology, Sociology
 
Faculty Advisor:
Robert S. Haltiwanger, Professor
Department of Biochemistry and Cell Biology
 
Are O-Glucose Modifications Important for the Mouse Notch1 Signaling Pathway?
Rosemary Orhue(1), Dr. Robert Haltiwanger (2), Dr. Aleksandra Nita-Lazar (2)
 
Complex carbohydrates play important roles in regulating signal transduction during development. Notch, a protein located at the plasma membrane that is responsible for significant cell fate decisions, is modified by sugars and these sugars affect its signaling pathway. Defects in Notch cause devastating diseases such as T-cell leukemia, CADASIL, and congenital heart defects. Notch has thirty-six tandem epidermal growth factor (EGF) repeats in its extracellular domain, many of which have sites where O-glucose sugars are attached through glycosylation. To determine if O-glucose modifications are important for Notch signaling, mutations were made at specific highly conserved EGF repeats (10,13, 14,16 and 27) containing O-glucose sites to prevent O-glucose modifications. Cos-7 cells were transfected with wild type or mutant Notch, co-cultured with ligand and control cells, and evaluated in signaling assays to determine if O-glucose was important for Notch to function. Here we show that when the O-glucose site at EGF repeat 27 was mutated, delta-activated Notch signaling decreased, while Jagged-activated Notch signaling remained the same.
 
(1)Department of Biology, Stony Brook University, Stony Brook, NY 11794-5110
(2)Department of Biochemistry and Cell Biology, Stony Brook University, Stony Brook, NY 11794-5215

Mayra B. Pastore-Patel
Martinez, CA
San Francisco State University
Senior
Biochemistry
 
Faculty Advisor:
Guangwei Du, Research Scientist
Department of Pharmacological Sciences
 
Role of Phospholipase D in Actin Cytoskeletal Reorganization of Epithelium Cells
Mayra B. Pastore-Patel(1), Michael A. Frohman(2) and Guangwei Du(2)
 
Phospholipase D (PLD) is involved in many intra/extra-cellular signaling transduction processes. The signaling relay is achieved through the production phosphatidic acid (PA) and choline from phosphodiester hydrolysis of phosphatidylcholine. There are two PLD isoforms in mammalian cells, PLD1 and PLD2. PLD activity has been reported to be involved in epithelium cell tumorogenesis, but the mechanism remains unclear. One possibility consists of promoting cytoskeletal rearrangement. Epithelium cells connect to one another through cellular junctions such as adherens and tight junctions. An important change that occurs in cancerous cells is their lost of cell-cell attachment through cytoskeletal remodeling. My current project was to investigate the potential role of PLD1 and PLD2 in actin cytoskeletal rearrangement during tumor progression.
 
The function of PLD in NMuMG epithelium cells was explored through overexpression and downregulation of the individual isoforms. First, I examined the subcellular localization of each protein using EGFP-tagged-PLD expression constructs, and assayed changes in cell-cell tight junctions as a consequence of the overexpression. My results revealed that PLD1 localizes to the perinuclear region and PLD2 to the plasma membrane, consistent with previous reports in other cell types. However, there is not enough evidence to conclude PLD overexpression alters actin cytoskeleton. To downregulate PLD expression, I adapted a new RNA interference (RNAi) technology. This technology uses short hairpin RNA (shRNA) expressed as a fusion with the human microRNA- 30 (mir-30) precursors and driven by a CMV promoter. To date, several constructs expressing PLD1 or PLD2 shRNA have been assayed with Western Blotting after transient cell transfection. Subsequent analysis allowed me to conclude that shRNAs successfully downregulated the protein's overexpression. My next step is to generate tetracycline-regulated constructs, using pTRE-Tight and PIRES2-EGFP, and producing RNAi to induce downregulation of PLD.
 
(1)Department of Chemistry and Biochemistry, San Francisco State University, 1600 Holloway Ave, San Francisco, CA 94132
(2)Department of Pharmacological Sciences and The Center for Developmental Genetics, Stony Brook University, Stony Brook, NY 11794-5140

Kestrel Perez
Bridgehampton, NY
Southampton College
Incoming PhD Student
Marine Biology
 
Faculty Advisor:
Stephan Munch, Assistant Professor
Marine Sciences Research Center
 
Estimating Heritability in the Atlantic Silverside (Menidia menidia)
Kestrel Perez(1), Stephan B. Munch(2)
 
Larval fish can face strong positive selection for fast growth as rapid growth leading to a larger size earlier is beneficial to survival. Environmental factors are commonly known to influence larval growth rates, however little is understood about how parental genotype affects variation in a phenotypic offspring trait, such as length. To investigate the level of heritability in growth rate variation, we created 55 families of Atlantic silversides (Menidia menidia) by strip- spawning adults. All environmental conditions were held constant to ensure minimal growth rate variation due to environmental factors. We measured lengths of larvae at 1, 5, 10, and 24 days post hatch. An estimate of heritability was determined from growth rate variation among full sib and half sib families. The maximum likelihood estimate of heritability (h²) was 0.07 with a 95‰ confidence limit from 0 to 0.55. The mean Bayesian estimate of heritability (E (h²/data) was 0.21 (? (h²/data) = 0.16) which corresponds with the expected result for organisms facing positive selection. This estimate of heritability in Atlantic silversides has important implications for better understanding larval growth.
 
(1)Department of Marine Biology, Southampton College, Southampton, NY 11968
(2)Marine Sciences Research Center, Stony Brook University, Stony Brook, NY 11794-5000

Kacey-Ann Thompson
Queens, NY
SUNY Old Westbury
Junior
Biology
 
Faculty Advisor:
Caroline Kisker, Associate Professor
Department of Pharmacological Sciences
 
Crystal Structure of Xanthine Dehydrogenase Variants from Rhodobacter Capsulatus
Kacey-Ann Thompson(1), Jennifer Doebbler(2), Caroline Kisker(2)
 
The oxidation of hypoxanthine to xanthine followed by the oxidation of xanthine to uric acid is catalyzed by Xanthine Dehydrogenase (XDH). The enzyme is inhibited by the clinical drug allopurinol, which becomes oxidized to alloxanthine and cannot be further oxidized, thus acting as a tight inhibitor. Mutations at one of the two key electron transfer residues, E232Q or E730Q hinders the functionally-important electron transfer mechanism, as does the exchange of tungsten for the molybdenum in the cofactor, and prevents the enzyme from turning over the substrate. Crystal structures of the XDH variants (E232Q, E730Q, and the Tungsten derivative) having either hypoxanthine or xanthine bound to them, will give a clearer illustration of how the enzyme functions, as well as giving further insight to developing a new and safer clinical drug to treat hyperuricemia, that is, the build up of uric acid.
 
Pseudo two-dimensional diamond shaped XDH variant crystals were grown by vapor diffusion and soaked in hypoxanthine or xanthine solutions. These crystals were taken to the National Synchrotron Light Source at Brookhaven National Laboratory for x-ray diffraction analysis. We were unable to determine the crystal structure as the crystals did not give sufficient diffraction results, however future attempts are in progress.
 
(1)Department of Biological Sciences, SUNY College at Old Westbury, PO Box 210, Old Westbury, NY 11568-0210
(2)Department of Pharmacological Sciences, Stony Brook University, Stony Brook, NY 11795-5115

Wasnard Victor
Westbury, NY
Columbia University
Incoming Master´s Student
Biomedical Engineering
 
Faculty Advisor:
Stefan Judex, Assistant Professor
Department of Biomedical Engineering
 
An Automated Method for Quantification of Trabecular Bone Architecture via Micro-Computed Tomography
Wasnard Victor(1) Stefan Judex(2)
 
INTRODUCTION: Quantification of trabecular bone architecture is essential for the study, diagnosis, and ultimate treatment of bone disease. A state-of-the-art method for bone analysis is micro-computed tomography. As the number of bones to scan increases, bone analysis can become a lengthy process. Trabecular bone must be separated from cortical bone to distinctly analyze its quantity and microarchitecture by manually contouring each slice, in a cumbersome manner. The goal of this study was to modify the Image Processing Language (IPL) program of the Scanco micro-ct system to automatically segment trabecular bone from cortical bone faster than the manual way. MATERIALS AND METHODS: Two-dimensional bone slices of the metaphyseal region of the femur from female mice were obtained via a micro-ct scanner (vivaCT 40, Scanco Medical, Switzerland) from a genetic bone study¹. IPL was used to construct three-dimensional images of the bone and provide quantitative data. Both manual and automatic segmentation procedures were performed for a basis of comparison. For the automatic segmentation, different parameters for a Gauss kernel filter such as standard deviation, and support, were varied to obtain optimal bone architecture. RESULTS: Each high resolution bone scan consisted of 144 slices for each bone. For these 144 slices the separation of trabecular bone from cortical bone by manually drawing contour lines for each slice took approximately one hour. Analysis of the mice consisted of three different bone scans for the left and right femur, totaling 864 scans, resulting in six hours for manual segmentation per mice. For the automatic segmentation, on the other hand, a contour line was drawn for the first slice and then iterated forward through the slices. This iteration process took approximately thirty minutes, resulting in three hours for automatic segmentation per mice. The automatic segmentation produced best results when segmentation parameters were increased from default values of 2 and 4 to a standard deviation value of 50, and support value of 14 respectively. As the values increased, more trabecular bone regions were visualized, and the images produced by the manual and automatic segmentation procedures became more comparable. Increasing these values too far pass 50 and 14 however, overestimated the amount of cortical bone, and we visualized less trabecular bone. DISCUSSION: In this study an automated method for segmenting bone was investigated as a means to quantify trabecular bone architecture faster than current manual methods of segmentation. Results indicate that certain parameters for automatic segmentation can realize comparable segmentation to that of current manual methods. However, quantified data via automatic segmentation must be investigated further.
 
(1)Department of Biomedical Engineering, Columbia University, New York, NY 10027-8904
(2)Department of Biomedical Engineering, Stony Brook University, Stony Brook, NY 11794-2580
*Gene Expression Patterns in Bone After 4 Days of Hind-limb Unloading in Two Inbred Strains of Mice. Zhong N.

 
Akua Bonsra
Ph.D. Student in Pharmacological Sciences
 
Activities Coordinator

 
Joseph Ortiz
Ph.D. Student in Applied Mathematics and Statistics
 
Instructor of the Intensive Mathematics Seminar (IMS)

 
Eihab "Harry" Jaber
Ph.D. Student in Materials Science and Engineering
 
Instructor of the Research Method Seminar

 
Daniel Stoebel
Ph.D. Student in Ecology and Evolution
 
Instructor of the Writing Workshop

 
Glenn Roach
Incoming Master Student in Mechanical Engineering
 
Residential Liaison Officer

 
Roberto Colson
Ph.D. Student in Applied Mathematics and Statistics
 
Instructor of IMS

 
Teresa Shakespeare
Ph.D. Student in Physiology and Biophysics
 
Instructor of Research Methods Seminar

 
Dr. Erwin George
Ph.D. in Applied Math and Statistics, 2003
 
Supervisor of IMS

 
Rose Saint Fleur
Ph.D. Student in Applied Mathematics and Statistics
 
Head Liaison