2006 REU Participants-Department of Biological Sciences - Carnegie Mellon University

2006 Participants

National Science Foundation
Mentored, Cutting Edge Research Experiences in the Molecular Biosciences
Research Experiences for Undergraduates (REU)

Sarah Beckman

Sarah Beckman, Clarkson University

Mentor: Dr. Peter Berget

Effect of Mutations in Single Chain Variable Fragment on Interaction with Thiazole Orange Dye

A biosensor is a molecule that can show biological activity, in this instance either by fluorescing or losing fluorescence. This is accomplished through the use of a single chain variable fragment (scFv) and a fluorogenic dye, thiazole orange (TO1).   The scFv is engineered in such a way that it can bind to the dye, thus causing the dye to fluoresce but only when bound to the scFv. An scFv is an engineered antibody containing both a heavy and light chain segment, which resembles the active part of an antibody. The scFvs I am working with contain three variable regions each in the heavy and light chains. If the scFv is altered in such a way that it contains a site for biological activity, such as phosphorylation, it can be acted upon by in vivo enzymes. This allows one to ascertain that phosphorylation of the engineered site on the scFv is taking place. However, this happens only if these sites alter the binding affinity to the dye; this process can be useful in elucidating biological pathways. In this specific case, three mutant plasmids have been affinity matured from scFv1 WT: scFv1 9c6, scFv1 6g2 and scFv1 4F5. All of these scFvs have shown increased fluorescence with binding to TO1 dye. My project is to assess the effects of individual mutations on dye binding and subsequent fluorescence by constructing an array of mutant scFvs carrying different combinations of the mutations harbored by 9c6, 6g2 and 4F5. They have been constructed by restriction digest manipulations and site directed mutagenesis, and then analyzed by flow cytometry. The analysis has shown that certain mutations are more important than others, specifically mutations in variable regions; however the scFv1 9c6 construct that achieved the greatest fluorescence is one that contains two mutations, each one in a different variable region. The affect on fluorescence contributed by each of these mutations is not simply additive.

Benjamin Blonder

Benjamin Blonder, Swarthmore College

Mentor: Dr. Alan Waggoner

Expression of a Thiazole Orange Dye-binding scFv in E. coli

Antibody single chain variable fragments (scFvs) are becoming important tools in the development of a new class of complex and sensitive biosensors. The ability to isolate small antibody fragments with high affinity for a given antigen is enabling the construction of novel fluorescent probes and signal transducing complexes. I am working with a scFv (NP29) comprised of a single heavy chain which binds tightly to a thiazole orange dye. It is hoped that by determining the structure of NP29 the nature of the dye binding can be determined. This will enable the scFv to be further modified and fused onto a complex sensor molecule that might enhance or shift dye fluorescence upon the binding of a target molecule. NP29 was originally expressed in a yeast strain unable to secrete the protein when grown a minimal media needed for structural determination by NMR. Thus, I have spent my summer transferring the protein to an E. coli expression system. I cloned the NP29 from S. cerevisiae and inserted it into a plasmid expressing the protein fused to tags for periplasmic expression (secretion) and purification by affinity chromatography. Cytoplasmic expression of NP29 has been successful in rich media and hopefully will also be successful in minimal media. Further work will be necessary to determine the proper conditions for periplasmic expression of the protein.

Nicole Bournival

Nicole Bournival, Fairfield University

Mentor: Dr. Tina Lee

Investigating the Mechanism of NDKB-Mediated Endoplasmic Reticulum Network Morphogenesis

Coat protein complex II (COP II) vesicles function in the transport of proteins between the endoplasmic reticulum (ER) and the Golgi apparatus in mammalian cells. NDKB was identified on the basis of its ability to facilitate COP II assembly. Recent work indicates that NDKB is a novel phosphoinositide-binding protein that also promotes ER network formation, most likely as a prerequisite to COPII assembly. In support of this, alteration of a cluster of positively charged residues on NDKB that leads to loss of phosphoinositide-binding simultaneously lead to the loss of ER network formation and COP II assembly functions. To identify the most relevant phosphoinositide on the ER that mediates NDKB function, a variety of phosphoinositide-binding domains were tested for their ability to competitively inhibit NDKB function. Surprisingly, the PH domain of oxysterol binding protein (OSBP), that binds both PtdIns(4)P and PtdIns(4,5)P2, did not inhibit NDKB function; rather, it mimicked NDKB’s ability to promote both ER network formation and COPII assembly. This suggested that phosphoinositide-binding accounts for NDKB function in both processes. Because the OSBP PH domain used in the assay is a GST fusion protein, it is likely to be dimeric. My research focused on testing whether the ability of OSBP PH to dimerize might be required for its ability to replace NDKB, possibly by inducing PtdIns(4)P or PtdIns(4,5)P2 clusters in the membrane. To test this idea, I generated a His-tagged monomeric OSBP PH protein and assayed it for COPII assembly function. Contrary to expectations, the monomeric OSBP PH domain also mimicked NDKB function, suggesting that lipid clustering is not required for NDKB function in ER network formation and COPII assembly. Finally, to test whether the most relevant NDKB target might be PtdIns(4)P or PtdIns(4,5)P2, I also generated a His-tagged PLCd1 PH protein that binds PtdIns(4,5)P2 but not PtdIns(4)P. In contrast to the OSBP PH domain, the PLCd1 PH domain was nonfunctional in COPII assembly, pointing to PtdIns(4)P as the most likely target for NDKB on the ER.

Kathryn Caperna

Kathryn Caperna, St. Mary's College of Maryland

Mentor: Dr. Charles Ettensohn

Analysis of the Localization of the Dishevelled Protein in Embryonic Development of Lytechinus variegatus

The dishevelled (Dsh) protein has been observed in the vegetal cortex of fertilized eggs and is highly conserved among many species. It has been previously discovered that this key signaling protein plays a role in early deuterostome embryo polarity, endomesoderm specification, and the stabilization of ß-catenin which is required for endoderm and mesoderm formation. We know that Dsh accumulates in small clusters, or puncta, within the vegetal cortex, but much remains unknown about its formation and localization within the developing sea urchin embryo. This project aimed to further characterize Dsh by looking at the dynamics of the formation and behavior of the cortical, vegetal puncta that contain Dsh. mRNAs that encode a form of Dsh tagged with Green Fluorescent Protein (LvDsh.GFP) as well as a Dsh mutant that is expressed at higher levels (LvDsh D NPBS.GFP) were synthesized by excising and purifying a plasmid from bacteria that contained the desired gene. Fertilized Lytechinus variegatus eggs were then injected with the mRNA and localization of Dsh was observed on a confocal laser scanning microscope. Although this approach results in overexpression of Dsh in the early embryo, previous studies have shown that no phenotypic changes result. Time-lapse movies revealed that dishevelled localizes in distinct puncta within the vegetal cortex and the perinuclear region of the early sea urchin embryo. Within each locale, the puncta move dynamically. In the focal plane recorded, the puncta also appear to coalesce within their respective regions. No clear cases of movement between the vegetal cortex and the perinuclear region were observed.

Sarah Clark

Sarah Clark, Colby College

Mentor: Dr. Brooke McCartney

Determination of Adenomatous Polypopsis coli 2 (APC2) Domain Function in Cytoskeletal Organization

Adenomatous polypopsis coli (APC), a colon cancer tumor suppressor, functions as a negative regulator of the Wnt/Wg signal transduction pathway. Additionally, APC family proteins are involved in cytoskeletal regulation, associating with actin and the plus ends of microtubules, promoting microtubule stability and forming a complex with the +TIP protein EB1 and the actin nucleator Diaphanous. To understand the role of APC2 in cytoskeletal regulation in vivo, we used Drosophila syncytial embryos, in which proper nuclear division relies on coordinated organization of actin and microtubules. During the four rounds of nuclear division of the syncytial blastoderm embryo, actin cycles from actin caps to rings and furrows, serves as a barrier against spindle collisions. APC2 localizes to actin structures in syncytial embryos. Using three mutant alleles of APC2 , we observed defects in actin rings and furrows during syncytial mitoses, suggesting a role for APC2 in actin organization and furrow extension. To continue investigation of the role of APC2 in cytoskeletal organization in the syncytial embryo, we examined actin defects exhibited by embryos expressing APC2 mutant alleles that affect different domains of the protein using immunofluorescence and confocal microscopy. We found that mutations in the Armadillo repeats of APC2, or C-terminal truncations of APC2, result in actin ring and furrow extension defects, suggesting that both domains are important in actin organization. Further, we found that conserved region B, a domain containing multiple phospho-acceptor residues, is important for actin organization. Similar to the regulation of APC proteins by GSK3 b in the Wnt/Wg signal transduction pathway, the role of APC2 in cytoskeletal organization may be regulated by the phosphorylation of amino acids in conserved region B. To understand the mechanisms by which APC2 carries out its cytoskeletal function, we are using live imaging to assess actin and microtubule behavior in wild type and mutant embryos.

Ashley Hurt

Ashley Hurt, Washington and Lee University

Mentor: Dr. Alison Barth

The Behavioral and Molecular Effects of Peritoneal Paxilline Injections on Wild Type Black 6 Mice

BK channels are functionally enhanced after seizure and contribute to abnormal excitability in the central nervous system (CNS). BK channel antagonists may reduce this abnormal excitability and thus decrease the likelihood of future seizures. This study investigated the effects of in vivo administration of BK channel antagonists on behavior, in order to determine what the side effects of BK channel antagonists might be as anticonvulsant agents. Animals were injected with either Paxilline (2.2mg/kg), a BK channel antagonist, or vehicle solution. Thirty minutes post injection, each animal was individually placed in a behavioral chamber and observed for a total of thirty minutes. The time to investigate foreign objects, time spent grooming, time stationary and total times reared were all accounted for in behavioral analysis. The animals were then allowed to sit for thirty additional minutes to ensure that the injected drug had ample time to reach the brain. Fos immunohistochemistry was used to compare Paxilline-induced changes in neural activity compared to vehicle injected controls. The nuclei of the amygdala, which is associated with fear, and the hippocampus, which correlates with learning and memory, were imaged to detect any changes in Fos immunohistochemistry. The behavioral and molecular data indicate that there is no significant distinction between Paxilline and vehicle solution injected animals, which is important because it implies that any differences seen between mice injected with Paxilline post status epilepticus and those that did not receive the drug can be attributed to molecular activation induced by the seizure.

Vilma Medrano

Vilma Medrano, Humboldt State University

Mentor: Dr. Brooke McCartney

Adenomatous Polyposis Coli: Phenotypes of Mutant Alleles in Imaginal discs and Its Role in the Activation of Wingless Signal Transduction Pathway

Mutations in the tumor suppressor gene, Adenomatous polyposis coli (APC), are linked to colon cancer. APC functions in part as a negative regulator of the Wnt/Wingless(Wg) signal transduction pathway. However, the precise mechanisms by which APC functions in signal transduction are not well understood. In Drosophila melanogaster , as in mammals, there are two APC s, APC1 and APC2 . Because the two APCs have redundant function in flies, we must mutate both APC s to investigate the cellular consequences of complete loss of APC function. However, flies mutant for both APC1 and APC2 die during larval development. Therefore, we used mitotic recombination to create clones of APC2 APC1 (double mutant) cells in the developing wing. In the wing imaginal discs, we observed that double mutant clone cells exhibit abnormal outpocketing and segregation from surrounding epithelial cells. By expressing a wild type APC2 transgene in double null clones, I have preliminary data that suggests suppression of the segregation phenotype. Previous data suggest that different APC2 alleles are associated with different phenotypic severities. I tested two different alleles of APC2 , together with the null allele of APC1, and found a range in severity of the phenotypes consistent with what has been observed in other developmental contexts. APC2 APC1 double null clones exhibit upregulation of Wg target genes due to the loss of negative regulation, as predicted. In contrast, I tested a weaker allele of APC2 and found little or no upregulation of Wg target genes in double mutant clones. These experiments have helped characterize APC2 allelic differences for a better understanding of APC function in the developing wing. These findings will ultimately lead to better understanding of APCs cellular functions and consequently its role in colon cancer.

Joseph Michie

Joseph Michie, Merrimack College

Mentor: Dr. Alison Barth

Spatial and Temporal Neuronal Firing Patterns After In Vivo Experience in Fos-drFP583 Mutant E5 Transgenic Mice

The project has sought to utilize the novel fluorescent clock drFP583 mutant E5 (timer) in transgenic mice that have timer fused to one of their immediate, early genes, c- fos , which is expressed after neuronal activity. The hope has been to create a way to resolve some temporal aspects of neural activity, perhaps at the circuit level, in addition to the useful method of differentiating between active and nonactive cells. The project has explored where timer is induced within the brain under basal states and also after status epilepticus, the primary means by which c- fos was induced. Discrete brain areas exhibited non-overlapping green and red cells. Initially, the timer mice exhibited status epilepticus resistance, but an alteration in dosage of pharmacological agents yielded an effective means to maximally induce the expression of timer in the mice. The possible time windows to observe the optimal expression of red fluorescence were narrowed to within a span of a few hours, which evinces the time period for the transition from green to red. Overall, the project will further our understanding of how discrete neuronal ensembles are activated during learning. Specifically, the project has addressed the question: where and for how long are different subclasses of neurons active after stimulation and how long c- fos is expressed. These questions have been addressed by analysis of the fos -timer mice through immunohistochemistry after status epilepticus induced by pharmacological agents. The project should provide insight into the spatio-temporal neuronal patterns associated with in vivo experience in fos -timer transgenic mice, of which the temporal aspect had until now largely been unexplored.

Ashlan Musante

Ashlan Musante, Wheaton College (MA)

Mentor: Dr. David Hackney

Cloning and Characterization of the Interaction of Kinesin Non-motor Microtubule Binding Sites with Microtubules

The stereospecific interaction of the motor domain globular "heads" of the kinesin motor protein superfamily members with microtubules (MTs) to produce processive movement is well studied. However, the binding behavior of non-motor domains of some kinesins and kinesin-associated proteins is less characterized; these interactions with MTs likely contribute to the net processivity of kinesin motors in conjunction with the activity of the primary MT-binding site in the motor domain. In the general interest of obtaining a greater understanding of the nature of kinesin non-motor MT binding activity and its contribution to net binding affinity, we cloned Drosophila Kinesin-1 and S. pombe EB-1 homologue, Mal-3, fusion proteins on the pET21 plasmid. Two of our three Kinesin-1 heavy chain clones, amino acids (aa's) 841-960 and 910-952, contained the auxiliary MT-binding sites while a control from aa's 841-910 did not. We also cloned the previously identified Mal-3 MT-binding site from aa's 1-119. All expressed MT-binding domains were preceded by an N-terminal His-tag and followed by a 14 aa C-terminal biotinylation tag. Constructs were purified using a nickel nitrilotriacetic acid (Ni-NTA) affinity column and analyzed for correct molecular weight using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Biotinylated constructs were bound to avidin-coated quantum dots, and their MT-binding behavior in vitro with sea urchin axonemes was observed using fluorescence microscopy to conduct single-molecule assays. While the S. pombe Mal-3 construct was not found to bind to MTs in this system, the 841-960 Drosophila kinesin construct showed axoneme binding suggested by difference interference contrast (DIC) microscopy and a negative control to be MT-specific. These constructs can be further utilized in future research to develop a more comprehensive understanding of the specific nature of kinesin non-motor MT binding behavior.

Adam Trexler

Adam Trexler, McDaniel College

Mentor: Dr. David Hackney

Green Fluorescent Protein as a Spectroscopic Probe for Head-tail Interaction in Kinesin Motor Protein Heavy Chain

Kinesin motor proteins appear to be regulated by a global folding event wherein the heavy chain tail domain folds back on the protein and binds the head domain. This blocks ATP binding and inactivates the motor ATPase activity. The purpose of this study was to create a reporter system using spectroscopic determination of head-tail binding to identify potential kinesin regulatory proteins. Our initial goal was to use fluorescence resonance energy transfer (FRET) between mantADP in the head active site and enhanced green fluorescent protein (eGFP) attached to the tail to assay head-tail binding. We cloned Drosophila kinesin tail fusion proteins TrxTev841-960GFP, TrxTev841-954GFP, TrxTev841-910GFP, and TrxTev910-952GFP into pET21 plasmids with His-tags. We expressed these in E. coli and purified the constructs on NTA columns. After purification our constructs produced a second absorbance peak at 400 nm in addition to the standard eGFP peak at 490 nm. The second peak magnitude was highly concentration dependent in weakly dimerized constructs such as TrxTev910-952GFP and weakly concentration dependent in strongly dimerized constructs such as TrxTev841-954GFP. These results are consistent with a model suggesting an eGFP dimer produces a second absorbance peak at 400 nm. Attempting to utilize this novel spectral feature, we tested binding of low concentration (monomer) tail-GFP constructs to head dimers, hoping to induce eGFP dimerization and a spectral change. The rate of mantADP release from dimeric DKH405 was reduced by TrxTev910-952GFP, but the appearance of a second peak was not observed. These preliminary results suggest our constructs bound but no eGFP dimer resulted. Furthermore, in preliminary experiments we were unable to find evidence of FRET between the eGFP and mantADP. A fluorescence correlation spectroscopic (FCS) experiment also verified head-construct complex formation. Given these preliminary results indicating head binding but no eGFP dimer formation, the head-tail complex structure may be such that eGFP attached to the tails cannot interact. This project has provided valuable structural clues regarding the autoinhibitory head-tail complex.

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