2009 Research Experiences for Undergraduates (REU) Participants-Department of Biological Sciences - Carnegie Mellon University

2009 Participants

National Science Foundation Research Experiences for Undergraduates (REU)

Undergraduate Research Experiences in Cellular and Molecular Biosciences

Cortlyn BrownCortlyn Brown, University of Chicago

Mentor: Dr. Eric Ahrens

In Vitro Evaluation of Multimodal Nanoemulsions for 19F MRI with Biosensing Properties

The prevalence of complex medical diseases, such as autoimmune diseases, cancers, and neurodegenerative diseases, has recently led to an increased interest in the application of cellular therapies. This has prompted the need for a non-invasive way to image the cellular activity and trafficking patterns following injection or transplantation. Recently, 19F MRI cell tracking approaches have been developed (Ahrens et al., Nat. Biotech. 2005) that employ ex vivo cell labeling with perfluorocarbon (PFC). Our lab has recently developed dual mode reagents for 19F MRI and fluorescent detection using perfluoropolyether (PFPE) nanoemulsions (Janjic et al., JACS, 2008). Using these reagents the 19F MRI data can be confirmed through microsocopy or immunocytochemistry in biopsied tissues. In this project we expanded these methods and implemented biosensing (pH, protease, and esterase) cyanine dyes (CypHer5 and Cy3) on new PFPE nanoemulsions to gain an understanding of their in vitro behavior and stability. We determined that CypHer5 fluorescence was unstable at 37oC while Cy3 maintained fluorescence stability. Furthermore, when exposed to acidic temperatures the nanoemulsion diameter, measured by Dynamic Light Scattering, increased, indicating a surfactant rearrangement. We also determined that sucrose prohibited nanoemulsion cellular uptake in vitro while serum significantly increased cellular uptake in vitro.

Richard Decal

Richard Decal, New College of Florida

Mentor: Dr. Brooke McCartney

Salt-Bridges and Phosphorylation are Not Essential for APC2 Function in the Destruction Complex

Human Adenomatous polyposis coli (APC) is mutated in both familial and sporadic colon cancer. APC proteins are part of a destruction complex that targets the Wnt signaling effector ß-catenin for degradation. Although it is clear that inappropriate activation of Wnt signaling in APC mutant tissue plays a key role in colon cancer initiation, how APC functions in the destruction complex is not well understood. In vitro studies identified phospho-Ser residues in the 20 amino acid repeats (20Rs) of APC, as well as Glu/Asp residues N-terminal to each 20R, which appear to play essential roles in APC-ß-catenin binding. Specifically, biochemical and crystallographic data showed that phosphorylation of these Ser residues, in addition to the Glu/Asp salt-bridges to ß-catenin, significantly increased APC affinity for ß-catenin. To test the hypothesis that phosphorylation and salt-bridging are essential for APC function in the destruction complex in vivo, we use Drosophila as a model organism. The lab generated targeted mutants of the conserved residues of Drosophila APC2 that abolish or mimic phosphorylation (by mutating Ser to Ala or Asp, respectively), or abolish salt-bridging (by mutating Glu/Asp to Ala). These APC2 mutants, driven by the native APC2 promoter, have been incorporated into an APC2 null background. This allows us to ask whether these mutant proteins retain any wild-type function. We conducted four assays to test the function of the mutant proteins: (1) hatch rate analysis, (2) an embryonic cuticle analysis, (3) examination of APC2 and (4) Armadillo (Drosophila β-catenin) localization in the embryo.  We discovered that our mutants rescued most wild-type activity, indicating that phosphorylation and salt-bridging play an important, but non-essential, role in APC2 function in the destruction complex.

Alex EllisonAlex Ellison, Connecticut College

Mentor: Dr. Ziv Bar-Joseph

Large-scale Conservation Analysis Among Closely Related Species

Many biological processes operate in a similar manner across a large number of species or conditions, and many of the genes that participate in these systems are conserved across organisms (Wilkins, 2001). However, recent studies show that while the sequence information between orthologous genes is conserved, in many cases these genes do not show the same functional behavior. Differences in expression patterns (Yong et al., 2009), protein-protein interactions (Odom et al., 2007), genetic interactions (Roguev et al., 2008) and other large-scale biological measurements are observed.

In this study, we try to reconcile two apparently contradictory observations: the high level of similarity on the pathway level, and the low-level of behavioral similarity on the gene level. In order to explain this apparent contradiction, we are looking for minimal "conservation units" that retain most of the dynamic behavior via cross species analysis.

Various types of data (protein-protein interactions, protein-DNA interactions, genetic interactions, functional similarity, co-expression, and sequence data) were collected from model organisms C. elegans, S. pombe, and S. cerevisiae. Networks were constructed to model each data type such that nodes represent genes and edges represent biological interactions or pattern similarity. The immediate neighborhoods of all orthologous nodes are compared across the various data sources, and the overall conservation in neighborhood similarity is assessed using rigorous statistical measurements. By creating and cross-referencing these networks find “conservation units,” comprised of some combination of the data types, that link orthologous genes in such a way that the low level of similarity on the pathway level can be explained.

A graphical user interface was created that allows researchers to easily navigate through the vast amount of collected data and identify minimal "conservation units" of specific genes.

James Hopkins

James Hopkins, Wells College

Mentor: Dr. David Hackney

Development of Novel Protein Purification Techniques

The proteins InaD and NorP form disulfide bonds (DB) under appropriate oxidizing conditions. The PDZ1 domain within InaD forms this disulfide bond with a C-terminal five amino acid residue (Thr-Glu-Phe-Cys-Ala) that is found in NorP. Work by Kimple and Sondek in 2002 has shown that using InaD is a viable way to isolate proteins containing the NorP amino acid tag. A current issue is in the creation of the disulfide bond which requires an aggressive oxidizing agent that can modify thiol groups outside of the active site. We will want to match the entropy of formation of the disulfide for an easier oxidation to take place. The experiments conducted tested the time it took to form and cleave the disulfide as well as at what other conditions, i.e. temperature, were required for maximum DB formation. We found that 2mM Lipoic Acid was enough to create the DB within twenty minutes at 0°C. For maximum cleavage of the DB increasing concentrations of DTT could be used with decreasing incubation times. This method of purification is not fully developed as of yet, but offers a strong bases in which it can be utilized. I also compared the effectiveness and practicality of using His-tagged proteins (HTPs) binding to phosphocellulose (P11) and metal affinity NTA as purification methods. We found that by binding our HTPs to P11 at pH 6.4, washing with a weak salt concentration, and eluding with a pH shift to 8.0 allowed for a purification of HTPs rivaling that of metal affinity NTA.

Karen Kormuth

Karen Kormuth, Washington & Jefferson

Mentor: Dr. Veronica Hinman

Evolutionary History of Polyketide Synthase in Asterina miniata

Polyketide synthase (Pks) is an enzyme commonly found in bacteria, fungi, and plants. Genes for pks have also been found in echinoderms such as the sea urchin, Strongylocentrotus purpuratus. Previous research has shown that the pks in the sea urchin is more closely related to pks in a slime mold. One possibility for this phenomenon is that this gene was acquired by echinoderms through horizontal gene transfer events. Sea urchins actually have two non-FAS Pks genes. SpPks, found in the sea urchin, S. purpuratus, is necessary for the production of pigment in some of its embryonic cells. The function and expression of SpPks2 have not been characterized. A partial sequence for pks2 has also been discovered in another echinoderm, the sea star, Asterina miniata. It is unknown whether sea stars have a true ortholog of SpPks though their embryos do not produce pigment. Therefore, the purpose of this project was to compare sea star pks to its orthologs in S. purpuratus. Specific aims of this project were to identify additional pks orthologs in A. miniata and determine if the sea star has a gene orthologous to SpPks. Methods employed in this project included degenerate polymerase chain reaction (PCR) to amplify additional sequences of pks from adult A. miniata cDNA derived from pigmented cells. A maximum parsimony method was used in a phylogenetic analysis that included the new AmPks sequences. Two pks orthologs were cloned from A. miniata one of which is orthologous to SpPks and the other to SpPks2. This finding shows that, as with S. purpuratus, A. miniata has at least two genes for Pks, but further research must be done to understand the expression and function of pks in sea star embryos.

Lindsay McCullough

Lindsay McCullough, Wilkes Honors College, Florida Atlantic University

Mentor: Dr. David Hackney

Developing Methods for Long Term Observation of Motor Movement

Kinesin is a motor protein that transports cargo inside the cell. GFP is used to see these motor proteins but is known to photobleach quickly making the actual movement and transport that kinesin performs unable to be seen. Fluorogen activating proteins (FAP) uses a normally flexible genetically encoded molecule that binds to a protein dye. The binding of the dye makes the FAP protein rigid and effectively causes fluorescence. Although FAP photobleaches as well, the dye is constantly renewed therefore the photobleaching is not detected. This constant renewal also prevents blinking, a problem that comes across when using other molecular markers such as Quantum Dots. Using PCR to attach the restriction sites, the H6 disulfide free version of FAP was cloned into a thioredoxin plasmid (TC) and to a K412 motor domain of kinesin1 plasmid. The K412-FAP was cloned and transformed successfully but proved very difficult to purify. The TC-FAP was transformed into E. coli and the protein was effectively purified and coupled with malachite green and a dyedron to produce fluorescence. A dyedron consists of a malachite green molecule with 3 Cy3s attached to increase fluorescence and alter the excitation wavelength to that of Cy3. The fluorescence produced by the coupling of the FAP and the dye was viewed using a PTI fluorometer showing increase in fluorescence when using the FAP with the dye rather than just the dye alone. A titration in this fluorometer was done to discover that this FAP must be dimerized in E. coli to emit fluorescence. Stopped-Flow fluorescence revealed an extremely slow disassociation rate. This leads us to believe that this particular FAP may not be as useful in detecting motor movement, but may be more useful in viewing non-motor proteins in the cell for an extended period of time.

Michelle Peck

Michelle Peck, Grove City College

Mentor: Dr. Peter Berget

Development of A Disulfide Bond-Free FAP, HL4-MG (MG13), for use as an Intracellular Biosensor

Recent research has demonstrated the use of fluorogen activating proteins (FAPs) in imaging cell surface proteins. These FAPs come from a library of human single-chain antibodies (scFvs) that have internal disulfide bonds that are only stable in non-reducing environments. Thus, in the reducing environment of the cytoplasm, scFvs cannot properly fold and are usually non-functional. The goal of this project is to extend the ability of the FAP HL4-MG (MG13) so it can function as an intracellular biosensor to measure protein-protein interactions. We carried out site-directed mutagenesis through PCR primer design and overlap PCR to replace a disulfide bond forming cysteine codon with alanine in each of the variable heavy (VH) and variable light (VL) domains. These mutants, Cys2 and Cys4, were verified by restriction fragment length polymorphism (RFLP) analysis and sequencing. Restriction digestions and ligations were carried out to place these mutant domains into the galactose-inducible FAP yeast surface display expression vector, pPNL6. A double mutant was also created by combining the mutant VH and VL domains in the same pPNL6 vector. Yeast cells transformed with these mutant plasmids were induced to express MG13 on the cell surface. Flow cytometry then measured both the surface expression of MG13 and the fluorescence of its cognate fluorogenic dye, malachite green. The mutant MG13 clones showed reduced fluorescence compared to the WT, while still maintaining some FAP activity. Cys2 maintained about 50% of the WT activity, while Cys4 and Cys2-Cys4 only kept about 5% of the WT activity. Currently, affinity maturation is being carried out on Cys2-Cys4 MG13 to identify clones that have increased FAP activity compared to the mutant parent. Identified clones will be enriched and then used in testing to detect protein-protein interactions. If FAP activity of this disulfide-bond free MG13 can be regained so it is able to activate malachite green in the context of a reducing environment, MG13 will be ready for use as an intracellular biosensor.

Rachel Stamateris

Rachel Stamateris, Allegheny College

Mentor: Dr. Charles Ettensohn

The Role of Wnt Signaling Components in the Polarized Degradation of β-catenin in the Sea Urchin Embryo (Strongylocentrotus purpuratus)

The Wnt signaling pathway is an important mechanism conserved among most animal species. When the Wnt pathway is activated, the Axin/GSK-3/APC “destruction complex” is inhibited, which normally degrades the protein β-catenin in the absence of the Wnt signal. Stabilized β-catenin is able to enter the nucleus and act as a transcription factor, activating certain target genes. In the model organism Strongylocentrotus purpuratus (sea urchin), the Wnt pathway is responsible for activating genes that control the formation of the endomesoderm during embryogenesis. The early sea urchin embryo has nuclear β-catenin concentrated at the vegetal pole, while the animal pole has no detectable nuclear β-catenin. The animal pole eventually forms ectoderm, while the vegetal pole forms endomesoderm, largely due to this β-catenin gradient seen along the embryo. Therefore, the overall goal of the project is to determine why β-catenin is localized in the vegetal pole of the embryo by determining which Wnt components may play a role in this polarized degradation of β-catenin. To accomplish this, I examined several maternal components of the Wnt signaling to see if they too are present in some sort of gradient along the AV axis of the embryo.  Specifically, I used in situ hybridization to determine if mRNAs that encode the most abundant maternal Wnt ligands and Wnt receptors (including Wnt16, Fz and LRP6) are present in a gradient in the egg or early embryo. The results of the in situ experiments indicate that these Wnt components are not asymmetrically distributed.  I also used quantitative PCR (QPCR) to quantify the number of transcripts of many of the Wnt components in the egg, as well as to detect possible differences in Wnt mRNA levels between the animal and vegetal poles of the 16-cell stage embryo. The most abundant transcripts in the egg (greater than 500 transcripts per egg) were LRP6, Fz5/8, Fz and Wnt16. Fz9/10 was present at an intermediate level (67 transcripts per egg), while Wnt7, Wnt3, Wnt8, WntA, Wnt6, Wnt4, Wnt5, and Fz4 were present at extremely low levels (less than 20 transcripts per egg), making them highly unlikely candidates for contributing to β-catenin stabilization. Of the transcripts tested by QPCR in the 16-cell stage, all increased in transcript number from their respective levels in the egg, suggesting zygotic transcription is occurring. The QPCR data also did not indicate any signs of asymmetry between the animal and vegetal poles of the most abundant Wnt components in the 16-cell stage. Overall, asymmetry in the localization of maternal Wnt components is unlikely to contribute to the stabilization of β-catenin in the sea urchin embryo.

Kevin Watkins

Kevin Watkins, Mount Union College

Mentor: Dr. Aaron Mitchell

Gene Function Analysis in Candida albicans Using the UAU1 Cassette

Candida albicans is a fungus that naturally grows in the human body. However, upon environmental changes, C. albicans can become pathogenic. This is particularly relevant to those who are immunocompromised, including hospitalized patients. The goals of my research were to determine the function of nonessential genes that do not have homologues in nonpathogenic fungi and determine which genes are involved in reaction to stresses. We hypothesize that those genes that are nonessential and do not have homologues in nonpathogenic fungi may be involved in disease development. To elucidate this information, homozygous mutants were created and then assayed for potential defects. As C. albicans is an asexual diploid, research into the genetic basis of pathogenicity has required two successful rounds of transformation and, therefore, been slow. With the development of the UAU1 cassette, transformation is much easier and quicker as it requires only one round of transformation. Insertion mutations into three genes, ORF19.2003, ORF19.7056, and ORF19.943, individually yielded hypersensitivity to caspofungin, an antifungal drug that inhibits cell wall synthesis and repair. ORF19.2003 is most similar to a plasma membrane protein in Saccharomyces cerevisiae that serves as a choline/ethanolamine transporter. ORF19.7056 is most similar to a plasma membrane protein in S. cerevisiae that is a dicarboxylic amino acid permease. Lastly, ORF19.943 is most similar to a plasma membrane protein in S. cerevisiae that is a multicopper ferro-O2-oxidoreductase. As the enzyme, β-1,3-glucan synthase, that is a major component of cell wall synthesis is found in the plasma membrane, we believe that the three proteins are normally found in the neighborhood of or somehow interact with β(1-3) glucan synthase for it to properly carry out its function. Therefore, when these proteins are missing, cell wall synthesis is inhibited. This information furthers our understanding of how C. albicans reacts to a stress, namely caspofungin.

Please send inquiries about our Research Experiences for Undergraduates program to bio-reu@andrew.cmu.edu.