2000 NSF Participants-Department of Biological Sciences - Carnegie Mellon University

2000 National Science Foundation (NSF)-supported Participants

Philip AndersonPhilip Anderson, Allegheny College
(Advisor: Dr. Alan Waggoner)

Identification of an near infrared-sensitive cyanine fluorochrome and an investigation of its potential for use in deep tissue imaging
Fluorescence imaging has emerged as an important technique in tissue visualization owing to superior resolution of the sample and high sensitivity to detail. Research using currently available fluorescent dyes is limited by fluorochromes (dyes) that can be visualized up to a depth of only a few hundred micrometers. Near infrared-sensitive fluorescent dyes hold the greatest potential for deep tissue imaging because biological materials tend not to absorb light in the near infrared region. The goal of the summer research project has been to identify and elucidate the properties of a fluorochrome that has potential applications for deep tissue imaging. The selected dye was inserted into muscle tissue of freshly prepared Gallus gallus and the maximum extent of detection was assessed using a CCD array. If detection can be improved, infrared-sensitive cyanine dyes may bring about important medical advances, enabling in vivo studies of tumor physiology and antibody targeting.

Clara BeardClara Beard, Durham Technical Community College
(Advisor: Dr. John Woolford)

Growing up the Hard Way: Ribosome Biogenesis
While ribosomes are necessary for life, very little is understood about the process by which they are synthesized in the nucleolus. Two ribosomal RNA primary transcripts (pre-rRNAâs) are made by two different RNA polymerases, RNA Pol I and RNA Pol III. One of these products, the 35S rRNA, is processed by exo- and endonucleolytic factors into 5.8S, 18S, and 25S rRNAs. The 5S rRNA is made by RNA Pol III. Proteins are attached to these rRNAs in some sequence to compose the ribosome. The 5.8S, 5S, and 25S rRNAs associate to form the 60S particle of the ribosome while the 18S rRNA composes the 40S particle. Evidence suggests that the pre-rRNAs are surrounded by proteins that assist the transition from pre-rRNA into mature protein/rRNA complexes. Recently, other labs have identified small nucleolar RNAs, snoRNAs, as guides for the sites of pre-rRNA that will be modified. SnoRNP proteins interact closely with snoRNAs to form small nucleolar ribonucleoprotein particles (snoRNPs). In addition to snoRNP proteins, there are many other nucleolar proteins necessary to make ribosomes. In this paper, I explored one snoRNP protein believed to be involved in this process of guiding the correct proteins onto the rRNA to create a ribosome: Nop1p, and one other nucleolar protein Nip7p. Nop1p is a box C+D snoRNP considered to be homologous to vertebrate fibrillarin in yeast. I planned on purifying these proteins and with them any other proteins that were attached in hopes of finding and characterizing more proteins that mediate the synthesis of ribosomes. Unforeseen problems with tagging the proteins did not allow the purification to take place. The information gained through the process; however, will be an asset to the lab and the next person to attempt this project. I also screened a yeast genomic library with ITS2, the second internally transcribed region of the pre-rRNA. This screen was carried out to identify genes and coding proteins that interact with the ITS2 RNA. These gene products would then be assumed to have some role in ribosomal biogenesis. Three genes were found using this screen; their roles will be examined in future experiments.

Kathleen BoyleKathleen Boyle, Immaculate College
(Advisor: Dr. Jonathan Minden)

The Effects of Caspases on Drosophila Melanogaster Embryonic Development
Caspases (cysteine proteases) are essential mediators of cell death, or apoptosis, which is important for normal development and tissue function. Caspases trigger cell death via an extensive caspase cascade. Here, as a result of the redundancy of caspase function, I have microinjected single caspase inhibitors and sets of caspase inhibitors and specific fluorescent dyes into the embryos to study development, morphogenesis and pattern repair resulting from these polypeptide inhibitors. These caspase inhibitors were also shown to effect macrophage function and differentiation in the Drosophila melanogaster embryo. Three-dimensional, time-lapse movies of Drosophila embryogenesis were performed after the microinjections to study the extent of cell death, macrophage function and differentiation and morphogenic movements, which can only be analyzed in vivo. Specifically the capsase inhibitors microinjected were Caspase 3 Inhibitor 1, Caspase 1 Inhibitor 1, Capase Inhibitor I and Group III Caspase Inhibitor 2. Abnormalities in the embryos that were taken into account were an unusual amount, time or distribution of cell death and macrophage, longer or shorter times of germ band elongation and retraction and the time between both, and segmentation defects. The embryos injected with Caspase 3 Inhibitor 1 showed little deviation from the wildtype, showing that the role of caspase 3 in embryogenesis can be compensated for by other caspases when caspase 3 is inhibited from functioning. Caspase 1 Inhibitor 1 injected embyos showed defects in cell death, macrophage function and germ band elongation and retraction, but also, unlike the embryos injected with other inhibitors, large amounts of cell death and phagocytic activity in the amnioserosa, indicating that these cells were phagocytizing each other. These results indicate that Caspase 1 plays a vital role, one that cannot be taken over by other caspases in its absence, in normal embryo development. Embryos injected with Caspase Inhibitor I, a broad spectrum caspase inhibitor, were shown to have all abnormalities I was searching for, proving that inhibition of several caspases at the same time causes extensive damage to embryonic development. Group III Caspase Inhibitor 2 injected embryos also showed great abnormalities in cell death, macrophage function, germ band elongation and retraction, and segmentation, showing that Group III caspases function in later morphogenic processes. In conclusion, caspases, to varying degrees when single and in specific combinations, have been proven to be essential elements in apoptosis, or programmed cell death that is needed for proper development, morphogenesis, macrophage function and pattern repair in the Drosophila melanogaster embryo.

Emily FaivreEmily Faivre, Augustana College
(Advisor: Dr. Jonathan Jarvik)

Analysis of CD-tagged mouse cells
Using a special CD-tagging retrovirus, an exon encoding green fluorescent protein (eGFP) was randomly incorporated into the genome of NIH 3T3 mouse fibroblasts. Because the integrated retroviral vector does not contain its own promoter, GFP expression only occurs when the vector sequence is present in an intron within a gene and is successfully transcribed, spliced, and translated as part of a fusion protein. Using GFP fluorescence as a marker, valuable information concerning a proteinÕs abundance and location in the cell can therefore be obtained by fluorescence microscopy. Ultimately, we would like to create cell libraries containing at least one tag for each gene in the genome. Such libraries would contain tens of thousands of different clones. Previously, two different retroviral constructs with different orientations of the GFP exon were used in separate infections of NIH 3T3 cells. In order to make a preliminary assessment of the success rate of tagging between the two constructs, I compared the numbers of GFP-positive cells in each transduced culture. Results suggested that one orientation is preferable to the other in terms of the yield of GFP-positive clones. Photomicrographs at both 40x and 100x magnification were taken; the images will be examined by artificial intelligence image-classification software in order to assign them to distinct morphological classes. Additionally, two cloning methods - cloning cylinders and fluorescence activated cell sorting (FACS) - were used in efforts to obtain pure clones of cells that stably express GFP and therefore contain the CD-tag in a stable state within the genome.

Megan KossMegan Koss, Wesleyan University
(Advisor: Dr. William Brown)

Engineering and Screening Single Chain Variable Fragments
Toluene Diisocyanate (TDI) and 4,4'-diphenylmethane diisocyanate (MDI) are highly reactive molecules that contain two isocyanate groups (R-N=C=O). These molecules are used in industrial settings to make synthetic polymers such as those that are in foam and paint. They are also toxic, inducing occupational asthma, however, the mechanism of toxicity is not yet understood. The isocyanate groups react in vivo to form conjugates, some of which are highly antigenic creating an immunological response that is implicated in the mechanism of toxicity.

A portion of our research was dedicated to screening a single chain variable (antibody) fragment (ScFv) library obtained from mice injected with TDI and MDI modified mouse serum albumin (MSA), one of the in vivo reaction targets of the isocyanates. Screening was performed by using a colony lift assay to isolate bacterial colonies that contain ScFv fragments that bind to TDI and MDI conjugated proteins. From this assay, five ScFvs were found that bind to MDI conjugated human serum albumin (HSA). The binding specificity of these proteins are being further characterized using Enzyme Linked Immunosorbent Assays (ELISAs) with the ultimate goal of using this information to further understand the mechanism of TDI and MDI toxicity using immunohistochemical localization of isocyanate targets in tissues from exposed animals.

Another portion of our research was dedicated to further characterizing ScFv #9, which is a ScFv that was selected for its ability to degrade MDI based polyurethanes. In order to study the mechanism of degradation, much of my research has focused on optimizing the techniques of purification. ScFv #9 is located in the insoluble fraction of cell extract and chemicals that denature the protein must be used to extract and solubilize the protein. Once denatured, it is difficult to return ScFv #9 to its native form and prevent it from aggregating. A very successful technique tested to optimize solubilization was the introduction of BSA as a carrier protein during protein renaturation. Also tested were the effects of differing salt concentrations and pHs on the solubility of ScFv #9 during purification. Further research is being done to determine how BSA stops ScFv #9 from aggregating and the effects of other carrier molecules, such as glycine, that may be more easily separated from ScFv #9.

Selena MellonSelena Mellon, Haverford College
(Advisor: Dr. Frederick Lanni)

Correlation of actin and a-actinin in lamellipodia
The rearrangement of cytoskeletal proteins and the subsequent formation of ultra-thin protrusions of cytoplasm called lamellipodia are essential for the continued mobility of amoeboid non-muscle tissue cells. Fundamental to understanding the dynamic forces involved in cell motility is the quantification of contractile proteins and their structural counterparts relative to their location in lamellipodia. Among these contractile proteins of interest are filamentous actin (F-Actin) and the respective Actin Binding Proteins (ABPs), such as the bivalent dimeric ABP a-actinin, that contribute to the structural integrity of the cell. Swiss Mouse 3T3 fribroblasts (SW3T3) and Chicken Embryo Fibroblasts (CEFs) were placed in an environment that facilitated the development of lamellipodia and were then subjected to 3.4% paraformaldehyde fixation and 0.05% Tritton X permeabilization so that the following immunofluorescence techniques could be applied, localizing the distributions of F-actin and a-actinin in lamellipodia: 1) F-actin staining using the high-affinity F-actin-binding toxin phalloidin conjugated to the fluorophore Rhodamine, and 2) anti-a-actinin indirect immunofluorescence using a secondary antibody conjugated to the fluorophore FITC (Fluorescein IsoThioCyanate). The resultant images were then used to generate profiles of the two fluorophores that demonstrated the densities of actin filaments and the ABP a-actinin in lamellipodia. Localizing densities of these two cytoskeletal components should help to elucidate the functional role of a-actinin in lamellipodial protrusion.

Gregory MosesGregory Moses, Harding University
(Advisor: Dr. A. Javier Lopez)

Genetic analysis of the hrp48-9G8 region in Drosophila melanogaster
A region of 11kb within polytene bands 27C2-5 of Drosophila contains three genes of special interest for studies of RNA processing. The most proximal gene, hrp48, encodes an hnRNP-A homolog involved in regulation of alternative splicing. 9G8, which resides only 300 bp distal to hrp48 and is transcribed divergently, also encodes a putative splicing regulator. Nested within the single intron of 9G8 is nop5, which encodes a protein required for processing ribosomal RNA. nop5 itself contains an intron and is transcribed in the same orientation as 9G8, raising the question of how their exons are sorted out appropriately. In addition, a P-element insertion (k0230) within the 9G8 intron but upstream of nop5 causes male-specific lethality, suggesting that 9G8 is required for dosage compensation and/or nop5 plays a special role in processing the ribosomal RNA transcripts from the Y chromosome. Consistent with either hypothesis, a suppressor of the male lethality has been isolated and mapped to the X chromosome. A mutation (IE27.1) generated by imprecise excision of the P-element in k0230 is recessive lethal in both sexes and fails to complement the male-specific lethality of k0230. Preliminary results indicate that the nop5 and 9G8 exons undergo inappropriate cross-splicing in transcripts from the IE27.1 allele.

To help analyze the biological roles and regulation of nop5 and 9G8, I carried out a screen for new mutations that fail to complement IE27.1; some of these may be specific alleles of one or the other gene. My strategy was to induce transposition of a nearby P-element inserted in the first intron of hrp48, screening for reversion of the associated recessive lethality and reinsertion of the P-element (marked with w+) into the second chromosome. I then tested these new insertions for failure to complement IE27.1. To date, I have isolated two new alleles of 9G8/nop5 and I am completing characterization of three additional candidates. In addition, I analyzed the IE27.1 allele more thoroughly by Southern blotting and by testing complementation against imprecise P-element excision alleles of hrp48. My results indicate that IE27.1 is a small deletion within the 9G8 intron. This mutation may be useful for understanding the mechanism that normally sorts out the exons of nop5 and 9G8.

Laura OppermanLaura Opperman, Alma College
(Advisor: Dr. Amy Csink)

Variability of Heterochromatic Satellite Sequences in Different Lines of D. melanogaster and D. simulans
A significant fraction of the genome of higher eukaryotes consists of heterochromatin. Earlier work has shown that heterochromatic satellite sequences vary greatly between species. We wished to quantitate the amount of satellite sequence variation within the two species D. melanogaster and D. simulans and see if the variation correlated with recent genomic events such as transposon movement and incipient speciation. In situ hybridization and quantitative fluorescence microscopy were employed and the satellite quantitations were made relative to a control; these quantitations served as an indicator of the amount of heterochromatic variability. Our results indicated that variability did exist between various lines of the two species, but not for all of the satellites examined. Variability was not correlated with increased transposition in D. simulans and incipient speciation in D. melanogaster. However, CM1 2b.0,a line of D. melanogaster that has very high transposition rates, did have an increased amount of satellite variability. Additionally, we found that the AAGAG sequence, thought to contain the centromere defining sequences on the X chromosome, was absent from the X chromosomes of the Canton-S and Zimbabwe D. melanogaster lines. This implies that a recent centromere shift may have taken place on the X chromosome of these lines.

Genelle WaltersGenelle Walters, LaSalle University
(Advisor: Dr. David Hackney)

The properties of kinesin's tail domain as a cause for an inhibited state
Kinesin is a motor protein that transports membrane-bound organelles and intermediate filaments. Through past experimentation, it has been proven that kinesin is in an inhibited state, mediated by kinesinÕs tail domain, when not bound to cargo. This inhibited state is characteristic of low microtubule-stimulated ATPase activity. The motility of kinesin on a microtubule will be further analyzed with a light chain insertion onto a heavy chain or cloning. Our research will focus on cloning various kinesin structures. By cloning, we can incorporate various types of kinesin and to see if there are any other factors in this inhibition. Other segments of the tail will be expressed/cloned to further discover what other factors that may inhibit ATPase. Every light chain in each cell used will be co expressed. Final results of our research will be used for work to study this phenomenon in the future.

The properties of protein fractionation will also be explored. Expected protein concentrations will be calculated and compared to numbers gathered in actual experimentation and analysis.

Adrienne WellsAdrienne Wells, Hendrix College
(Advisor: Dr. Robert F. Murphy)

Analysis of Protein Localization Patterns via Feature Based Classification of Fluorescence Images
Determination of subccellular localization can often provide important insight into protein function. Using fluorescent protein tagging and digital image collection via fluorescence microscopy, one can determine localization patterns. Computational methods applied to this process aid in the automation of microscopy and localization pattern recognition. Previously, our lab developed numerical features from digital images of proteins which formed the basis for protein localization pattern classification. To further increase the accuracy of this classification, additional features derived from wavelet transforms were tested with the aim of finding the most concise set of useful features. Using normalization and wavelet feature extraction algorithms from different sources, new features were added to the existing feature collection. After feature reduction with stepwise discriminant analysis, a set of "best" features (including some wavelet features) tested at 84% protein classification accuracy. In another application of feature extraction, computational methods were applied to fluorescent images collected from an experiment which tracked two different proteins' reaction to copper. The progression of the proteins' localization patterns was charted using correlation and overlap measures. As expected, the two proteins diverge as copper is added and converge after copper is removed.