Phase behavior study of crystallizing surfactant solutions

Anne Golematis, Norman J  Wagner, Prachi Thareja, and Carrie Street
Department of Chemical Engineering

We present the case study of an industrial skin-care cream. A variety of skin-care creams are complex systems whose behavior under flow directly affects their tactile properties or their sensory perception. The goal of this research is to study the phase behavior of a skin cream formulation, which is essentially a mixture of surfactants in water. The components are Sodium Dodecyl Sulfate(SDS), Tegobetaine, Palmitic acid(PA), and water. Palmitic acid is known to exist as crystals at room temperature and is therefore expected to affect the macroscopic properties of the cream formulation.    The phase behavior of various surfactant formulations in water were studied, to find a region of composition where the formulation exists as one phase. Pseudo ternary phase diagrams were constructed at different temperatures to show the one or two phase regions containing various palmitic acid crystal morphologies. The one phase region is of most practical interest and depending on the surfactant compositions, the microstructure will correlate with the observed macroscopic properties. Research funded by: Unilever and The University of Delaware’s Undergraduate Research Program.

Metabolic flux analysis (MFA) has become a key concern in biotechnology and medicine. Metabolic fluxes describe what the cell does and, as such, provide an informative description of the overall cellular physiology. Stable-isotope MFA is the most comprehensive and effective means of characterizing metabolic fluxes in vivo. In this project, we applied a novel analytical technique, tandem mass spectrometry (GC-MS/MS), to quantify isotopic labeling distributions of glucose and amino acids to study metabolic fluxes. Specifically, we quantified stable-isotope labeling (13C and 2H) by GC-MS/MS in order to better identify the position of labeled atoms and increase the number of independent measurements. Using linear mapping techniques, we constructed isotopomer coefficient matrices from which the number of independent measurements for particular parent ions and amino acids were determined. These independent measurements could be used to estimate metabolic flux rates from labeling experiments. We started by analyzing 11 amino acids previously determined to be useful and identifiable. Then, having well-established methodology and analytical procedure, we moved to examine glucose labeling. We validated our techniques with labeled standards and applied these methods to analyze samples collected from tracer experiments with rat hepatocytes (i.e., liver cells). It is believed that by being able to more accurately measure and observe hepatocyte pathways, we will eventually be able to understand more about metabolic diseases such as type II diabetes. This project was funded by the University of Delaware's Science Scholars Program.

G-protein coupled receptors (GPCRs) are a large family of integral membrane proteins which mediate signaling across the cellular membrane. The diverse structure and function of these receptors has motivated efforts to characterize this class of proteins in academia and in the pharmaceutical industry. Despite these efforts, GPCRs prove difficult to express and purify in quantities sufficient for structure determination.  A system has been developed to express and purify peptides of GPCRs in Escherichia coli for nuclear magnetic resonance (NMR). Included in this system are a ketosteroid isomerase domain to enable high expression, Strep affinity tags for purification, a thrombin cleavage site, and a 6xHis-tag to facilitate purification.  Repeated GSG-linker sequences between the ketosteroid isomerase and the thrombin cleavage site as well as between the thrombin cleavage site and the GPCR peptide have been included to improve thrombin cleavage yields. The aim of this work has been to determine the minimal required length of these linker sequences for efficient cleavage of fusion protein from the adenosine receptor peptides hA₁R 260 and hA₂aR 259.

First-Principle Calculations of Adsorbate-Metal Interactions for Catalyst Design

Stanley Herrmann, Giannis Mpourmpakis and Dionisios G. Vlachos
Department of Chemical Engineering

The metal-adsorbate bond formation is a fundamental step in catalysis that controls activity and selectivity.  By tailoring the strength of this bond, one can control catalyst performance.  A possible way to do this is to use novel metals as catalysts.  Herein, density functional theory (DFT) was employed to determine the binding energy of several adsorbates participating in the water-gas shift reaction (H, O, C, CO, OH) on 10-atom metal clusters.  The d9 (Cu, Ag, Au) pure metals and their bimetallic combinations (CuAg, CuAu, AgCu, AgAu, AuCu, AuAg) were used.  A novel optimization scheme was applied to allow a small cluster to mimic the properties of the bulk metal structure.  We investigated the H and C interaction on a hollow site of the metal; the O and OH interaction on the bridge site; and the CO interaction on the top site.  We used spin unrestricted calculations and we accounted for a series of different spin multiplicities.  A database of binding energy values was constructed.  On top of this, the effect of bimetallic catalyst selection on the adsorbate binding energy was explored and trends among metals were extracted that can be used toward catalyst design.  Funding for this research was provided by Delaware’s NSF ESPCoR Program.

Expression of short non-coding RNA and their role in acid tolerance in Escherichia coli
(First place McNair Research Award)

Manuel R. Jimenez¹, Shawn W. Jones, and Eleftherios T. Papoutsakis
¹Northwestern University

In order to create an Escherichia coli strain more tolerant to short-chain fatty acids, like butyric and acetic acid, we applied the results from previous acid stress studies on Clostridium acetobutylicum and Streptococcus mutans. These studies indicated a role of two non-coding RNA (ncRNA) sequences in acid tolerance: the antisense sequence of the 16S promoter and the 4.5S small RNA component of the signal recognition particle (SRP). Both of these ncRNA sequences were overexpressed in separate strains of E. coli and then stressed with lactic acid (pH 5.82), acetic acid (pH 6.28), and butyric acid (pH 6.5). Each strain, along with a plasmid control, were grown in media with 0.00%, 0.25%, 0.50%, and 0.75% (v/v) of each acid, and optical density measurements were taken after 12 hr and 24 hr of growth. In general, the strain overexpressing the 4.5S ncRNA grew to a higher optical density than the plasmid control with significant advantage under lactic and butyric acid after 24 hr of growth. Overexpressing the antisense of the 16S promoter did not perform as well as the 4.5S strain but did display a slight growth advantage under lactic and acetic acid after 12 hr of growth. Overall, the data indicate that overexpressing these ncRNAs gives a slight growth advantage but further investigation is needed. The 4.5S strain may benefit from the overexpression of the Ffh protein, since the studies in Streptococcus mutans indicate this ncRNA and protein interact. We have also begun the characterization of the 5’ end of the 16S rRNA gene using the 5’ Rapid Amplification of cDNA Ends (RACE) technique. Both the sense and antisense strand are being probed to see if any ncRNAs are encoded and expressed in wild-type E. coli. We are also developing a 3’ RACE methodology, which polyadenylates the RNA for amplification, to further investigate the 5’ end of the 16S rRNA.

Testing Model Assumptions for Non-Native Protein Aggregation Kinetics

Brendan J. MacLean and Christopher J. Roberts
Department of Chemical Engineering

A mathematical model of non-native protein aggregation is developed and analyzed that explicitly accounts for the dynamics of pre-nucleation which were neglected in earlier generations of the Lumry-Eyring Nucleated-Polymerization (LENP) model (Andrews, J.M.; Roberts, C.J.  J. Phys. Chem. B 2007, 111, 7897-7913 and Li, Yi; Roberts, C.J.  J. Phys. Chem. B 2009, 113, 7020-7032).  The new model is able to capture an expanded range of experimentally observable behaviors, in terms of the kinetics of monomer loss and aggregate growth.  The model is first solved for cases in which folding-unfolding equilibration is assumed to be much faster than aggregation rates and aggregate-aggregate condensation does not occur.  As a result, only two limiting kinetic behaviors are found:  association-limited aggregation, where dimerization is rate limiting; and nucleation-and-growth limited aggregation, analogous to the earlier LENP models.   A global search of parameter space indicates that for most experimentally relevant cases, self-association or pre-nucleation can be assumed to be pre-equilibrated.  Although experimental examples for aggregation in which association may realistically be rate-limiting appear to be scarce, the current model is sufficiently general to hold under these conditions.  Ongoing efforts and preliminary results are directed at testing the assumption of pre-equilibrated folding-unfolding transitions in the LENP model.  Funding from the Howard Hughes Medical Institute is gratefully acknowledged.

Development of a Peptide Nucleic Acid Based siRNA Delivery System

Tejal U. Naik, Abbygail A. A. Palmer, Peter G. Millili, Millicent O. Sullivan
Department of Chemical Engineering

Small interfering RNA (siRNA) is a double stranded RNA molecule that has a major role in gene silencing by splicing the mRNA of a gene of interest, thus inhibiting expression.  The ability to harness the therapeutic benefit of siRNA can have a widespread impact on a variety of therapies ranging from cancer to HIV.  However, the efficacy of such treatments is limited due to the many extracellular and intracellular barriers associated with delivery.  One approach, involving siRNA functionalized surfaces, can improve the cellular response and specificity by creating a tunable release system. In this work, we propose the utilization of a peptide nucleic acid based siRNA surface-mediated delivery system.  Peptide nucleic acids are nucleic acid analogs that hybridize with complementary DNA or RNA sequences, enabling direct attachment of various macromolecules such as peptides.  Conjugation of targeting and protective moieties can potentially enhance the delivery of siRNA.  In order to test this, a cell transfection model utilizing stably transfected B16FO mouse melanoma cells (producing Green Fluorescent Protein (GFP)) was established.  Anti-GFP siRNA was designed and its efficacy was evaluated via Flow Cytometry and fluorescence microscopy.  Optimization of cell seeding density, siRNA concentration, use of antibiotics, and time of transfections was accomplished, effectively demonstrating gene silencing capability.  In preparation for future siRNA modifications, peptide nucleic acid-peptide conjugates were purified through Reverse Phase High Performance Liquid Chromatography (RP-HPLC).  Development of various chemical linkages and optimization of a cell transfection model lay the groundwork for future development of this nucleic acid delivery system.

Equilibrium Morphologies of Block Copolymer Thin Films as Determined by Self-Consistent Field Theory.

Jason Papandrea, Julie N. L. Albert, and Thomas H. Epps, III
Department of Chemical Engineering

Block copolymers are chained molecules consisting of covalently bonded homopolymer subunits. They self-assemble into periodic nanoscale structures which are potentially useful in applications such as membranes and templating. The phase behavior of bulk diblock copolymer systems is well-established in the literature, and now there is much interest in understanding more complicated systems involving multiblock copolymers and thin films. Self-consistent field theory (SCFT) allows prediction of equilibrium block copolymer morphologies and serves as a useful guide for experiments with these systems. We ran one-dimensional (1-D) SCFT simulations, which are useful for modeling lamellar structures, to determine equilibrium morphologies and corresponding free energies of bulk diblock copolymers of various compositions; the results matched expectations.  In addition, we ran thin film simulations with symmetrical diblock copolymers which yielded lamellar structures that varied according to surface preferentiality. These results will provide the foundation for future 2-D and 3-D simulations of other block copolymer morphologies.  

Using Zeolite-L for the Production of Dye Sensitized Photoluminescent Solar Concentrators

Tracy L. Powell, Raul Lobo, and Dustin Fickel
Department of Chemical Engineering

The ability to harness sunlight for energy production provides an alternative energy source which is both renewable and adaptable. The most common solar cells on today's market operate off of the semi-conductive properties of doped silicon wafers which, due to the high cost of production, cannot compete with coal or natural gas. The goal of this project was to produce a less expensive, yet efficient, photoluminescent solar concentrator using the structural properties of zeolite-L, a nanoporous crystalline material, in combination with fluorescent dye molecules for the collection and use of electromagnetic radiation. In the end, the solar concentrator will be coupled to a traditional (but smaller) solar cell.  The dyes, 4-dimethylamino-4'-nitrostilbene and nile red, were selected due to their size (they fit in the zeolite) and minimal overlap of absorption and emission wavelengths, and were incorporated into the pores of the zeolite which serves to align the dye molecules and ensure efficient capture of emitted photons. Dye loaded zeolite-L was suspended in a polymethylmethacrylate-toluene solution and then spin-coated onto a square inch glass slide. Due to the structure of the zeolite platelets, the pores of the zeolite should align and direct the photons to the edges of the cell where they can be collected. Testing of the device is in progress. This work was funded by Delaware's Science and Engineering Scholars Program.

Mathematical Modeling of Mesenchymal Stem Cell Differentiation Into Osteoblasts

Sven A. Saldanha, Prasad S. Dhurjati, Anja Nohe, and Gilberto Schleiniger
Departments of Chemical Engineering, Biological Sciences, and Mathematical Sciences

According to NIH, 10 million Americans have osteoporosis and an additional 34 million are at high risk with low bone density.  Osteoporosis is caused by an imbalance of osteoclasts, which break down bone, and osteoblasts, which are responsible for bone formation.  Research has shown that BMP-2 (bone morphogenetic protein) stimulation of the stem cell line C2C12, mouse myoblast cells, results in differentiation into osteoblasts, adipocytes, and myotubes dependent on the concentration of the ligand present in the serum.  This complicates matters as selective differentiation is highly favored for treatment of low bone density.  Additionally, the formation of other cells, for example adipocytes (fat cells), can actually be problematic as research has shown that fat cells have a detrimental effect upon osteoporosis. The current project seeks to determine the mechanism of mesenchymal stem cells differentiation.  To gain insight into the mechanism responsible for the differentiation process, a differential equation based mathematical model has been developed.  This mathematical model primarily considers the activation of the SMAD signaling pathway upon stimulation with BMP-2 and resulting osteoblast differentiation.  This project was funded by the University of Delaware Science and Engineering Scholars Undergraduate Research Program.

Mathematical Simulation of Drug Concentration in the Human Body: Application to Nanomedicine and Pediatrics
(Recieved third place in Sigma Xi competition.)

Sharon Weaver, Prasad Dhurjati, and Anja Nohe
Departments of Chemical Engineering and Biological Sciences

The knowledge of the concentration of a drug in the internal organs of the human body is crucial in ensuring its maximum efficacy. The goal of the project is to estimate the concentration of drugs in different organs using a mathematical model.  The major tasks included the creation and modification of a mathematical model of the human body. The starting point for the research was a model developed by Bischoff.  The model can be used for calculating the concentration of  methotrexate (a cancer drug), in the blood, liver, kidney, muscle and gut of four different animals: rats, mice, dogs and humans. The model equations were coded in MATLAB and the previous simulation results were validated. The model was then modified for two new applications. The first model modification was to show how methotrexate concentration differs in children and adults. Work was done to adjust the volume and blood flow rates to discover why children have more severe side effects when administered methotrexate. The second application deals with  target-specific drugs, where a nanoparticle can be directed to accumulate in specific organs based upon charge and size.  Initial graphs were made to model how the charge of a drug administered intravenously affects the concentration of the drug in various organs. Further research will allow for the addition of drug size as well as expanding the model to include additional organs.  The results of the modified models will have important implications for the design and delivery of pediatric drugs and nanomedicine. This project was funded by Howard Hughes Medical Institute.

The Electrochemical Properties of Tungsten Monocarbide (WC) Over Wide pH and Potential Ranges

Mark Weidman, Dan Esposito, and Jingguang G. Chen
Department of Chemical Engineering

Tungesten Monocarbide (WC) has been shown to exhibit catalytic properties similar to those of Platinum (Pt).  Due to the limited supply and high cost of Pt, WC has attracted interest for its potential application as a catalyst for the hydrogen oxidation reaction and methanol oxidation reaction in fuel cells.  Thin foils made of Tungsten (W), WC, and Pt were studied in an electrochemical cell to determine their stability and activity over a wide range of pH values.  WC thin films were synthesized on polycrystalline W foil substrates via physical vapor deposition (PVD) or in a furnace in the presence of methane.  Cyclic voltammetry (CV), chronoamperometry (CA), chronopotentiometry (CP), and linear scans (LS) were used to measure the electrochemical properties of the foils in electrolytes of varying pH.  X-ray photoelectron spectroscopy (XPS) and x-ray diffraction (XRD) confirmed the surface and bulk compositions, respectively.  WC was able to catalyze the hydrogen evolution reaction at lower overpotentials than W.  WC also showed a greater resistance to surface oxidation than W.  While W and WC behaved similarly at low pHs, their behavior diverged at higher pHs as a result of different surface oxide formation and different solubility of these oxides.  Funding is provided by the Department of Energy and the University of Delaware’s Undergraduate Research Program. 

Altering Recombinant Protein Production Using RNA Interference

Jinming Xing¹, Stephanie Hammond², and Kelvin Lee^2
1Delaware Technical & Community College, ²Delaware Biotechnology Institute

Mammalian cells are widely used to produce recombinant therapeutic proteins. In a previous study of Chinese hamster ovary (CHO) cells producing secreted alkaline phosphatase (SEAP), several proteins showed altered expression in a high-producing cell line. Cofilin, a key regulator of actin cytoskeleton dynamics in mammalian cells, was down-regulated in a high-producing cell line. Here, RNA interference is used to decrease cofilin levels, aiming to improve SEAP secretion. Previous experiments showed a 50% transient knock-down of cofilin by short interfering RNA (siRNA) along with 40% increase in SEAP production. To generate long-term knock-down of cofilin, short hairpin RNA (shRNA) vectors targeting the cofilin gene were made and transfected into CHO-SEAP cells.  The knock-down of cofilin was determined by western blotting while SEAP production was measured using a SEAP activity assay.  Preliminary results show a 25% knock-down of cofilin by shRNA and a 20% increase of SEAP production.  These results suggest that these shRNA vectors may be used to generate stable cell lines with reduced cofilin levels and enhanced SEAP production.  This project was supported by Grant No. 2P20RR016472-09 under the INBRE Program of the National Center for Research Resources (NCRR), a component of the National Institutes of Health (NIH), and Genentech.