1997 Summer HHMI-supported Participants-HHMI Undergraduate Program - Carnegie Mellon University

1997 Summer HHMI-supported Participants


Brian Bluth, Carnegie Mellon University
(Advisor: Dr. Javier Lopez)

Physical Mapping of the Dfur1 Locus

Drosophila furin 1 (Dfur1) was isolated as a target gene controlled by the homeotic gene Ultrabihorax (Ubx) during midgut development. It encodes a kexin-like endoprotease with specificity for cleavage at dibasic sites, such as are found in the precursors of secreted signaling factors like TGF-b. UBX positively regulates expression of Dfur1 in the midgut, where both genes are expressed in parasegment 7, just anterior to the site where the central midgut constriction will form. Ubx is required for formation of this constriction, which is also known to require cell interactions between mesoderm and endoderm. A putative P element enhancer trap insertion in Dfur1 has been isolated. This insertion is recessive lethal and causes a failure in formation of the central midgut constriction, suggesting that Dfur1 is required for this process. To confirm that this insertion is within Dfur1 and to map its location accurately, a collection of genomic clones spanning the Dfur1 locus has been obtained and mapped physically by restriction endonuclease mapping. The clones and physical map will now be used to determine the site of P element insertion and the extent of several deletion alleles generated by imprecise excision.

Image of Patricia A. DiGiuseppe

Patricia A. DiGiuseppe, Carnegie Mellon University
(Advisor: Dr. James F. Williams)

Natural Killer Cell Sensitivity of Adenovirus Transformants

Human adenoviruses are double-stranded DNA human respiratory viruses that are non-oncogenic in humans. Some are oncogenic in rats and all adenoviruses can transform rodent cells in vitro. Such characteristics make adenoviruses a valuable model in the study of viral oncogenesis. E1A and E1B are early adenovirus genes that play a role in transformation and oncogenicity. The E1A gene of Ad12, an oncogenic serotype, contains unique regions not found in Ad5, a non oncogenic serotype. One such unique region is a 60 b.p. spacer located between conserved regions 2 and 3 of E1A . Experiments using Ad5/Ad12 chimeric viruses (Ad12 viruses in which the spacer region was exchanged for the corresponding sequence in Ad5) showed that the spacer plays a role in tumorigenicity and oncogenicity. The mechanism by which it influences these processes is not known, but it was thought that it might play a role in the resistance of Ad12 transformants to NK cells. Prior experimentation utilizing spleen cell mixtures as an NK source and 51Cr release assays has shown that rodent cell lines transformed by Ad12 are NK resistant, while lines transformed by Ad5 are NK sensitive. However, by testing cell lines transformed by chimeric viruses it was determined that while the spacer region is linked to tumorigenicity, it is apparently not related to NK activity. The objective of my research was to purify spleen cell extracts from Lister Hooded rats utilizing four step discontinuous Percoll gradients with a view of enriching NK cells from spleen cell mixtures, and to determine the cytotoxic activity of these NK cells on adenovirus-transformed cell lines HRE5L and HRK47B utilizing non-radioactive methods such as Trypan Blue Exclusion (TBE) Assays and Lactate Dehydrogenase (LDH) Cytotoxicity Assays. Preliminary TBE Assays proved inconclusive due to low effector (NK):target cell ratios. Preliminary LDH assays proved inconclusive due to high LDH levels in effector cell controls. Future research will include more work with the LDH assays, and 3H thymidine release assays. In addition, more chimeric genes have been made in which regions of E1A are exchanged at the C terminus, which may be linked to NK sensitivity, and cells transformed by these new chimeric viruses will be tested.


Sarah E. Frew, Carnegie Mellon University
(Advisor: Dr. Elizabeth W. Jones)

High Copy Suppression of pep5 Temperature-Sensitive Mutants in Saccharomyces cerevisiae

pep mutants of the yeast Saccharomyces cerevisiae. exhibit decreased vacuolar protease activity. In particular, pep5 mutants accumulate inactive precursors to the vacuolar hydrolases. Recently, an insertion mutation of PEP5 was shown to be suppressed by a mutation of the VPS8 gene. The direct interaction between these two gene products was considered unusual since the two proteins were thought to act at different steps of the secretory pathway. This led us to propose that Pep5p is a bifunctional protein that functions at two steps in the vacuolar branch pathway. The purpose of this project is to identify proteins that interact with Pep5p. If separate domains of the protein function at the two different steps, we might expect to define two different, possibly overlapping, sets of interacting proteins that are required at the two different steps.

High copy suppression was the method used to identify interacting proteins. The premise behind this approach is that a partially active or unstable protein may be stabilized or activated through interaction with a normal partner and that the chances of a successful interaction will be enhanced if the normal partner is present in substantial excess. Several pep5 temperature-sensitive (ts) mutations were generated by PCR mutagenesis, cloned onto plasmids, and their associated phenotypes characterized. At the nonpermissive temperature, these pep5 ts mutants show a range of phenotypes, including a sensitivity to growth on media containing excess divalent cations and a lack of Carboxypeptidase Y (a vacuolar hydrolase) activity. After the ts mutations were each integrated into the yeast genome using the pop-in/pop-out replacement method, a consequent temperature-sensitive yeast strain was transformed with a high copy genomic library. In order to continue this project in the future, candidate suppressors of the ts phenotypes will be screened at the nonpermissive temperature. Colony hybridization will be used to eliminate any candidate suppressors that are PEP5 after the candidate plasmids have been shuttled into E. coli. The remaining candidates will then be analyzed with restriction enzymes and the ends of the insert containing the suppressing gene will be sequenced and checked for identity in the Saccharomyces Genome Database. Any gene likely to be the suppressor will be subcloned from the insert and retested for suppression, as well as tested for suppression of the remaining ts mutations.

Elaine Pinheiro

Elaine Pinheiro, Carnegie Mellon University
(Advisor: Dr. John L. Woolford)

Identification of Novel Genes Necessary for 60S Subunit Biogenesis in Saccharomyces cerevisiae

My primary objective was to search for an RNA or protein molecule that functionally or physically interacts with the Drs1p during ribosomal assembly. Drs1p plays an important role in the processing of ribosomal RNA in the formation of the 60S ribosomal subunit. Drs1p is a member of a family of proteins that are believed to use the energy generated from ATP hydrolysis to funtion as putative ATP- dependent RNA helicases or ATPases. It also contains acidic and charged regions that may interact with other proteins or with RNA. Drs1p might promote a conformational rearrangement of rRNA or small nucleolar RNAs that is necessary for the processing of rRNA and ribosome assembly. A genetic screen has been implemented to find cases where a mutation in the RNA binding or RNA unwinding motif of the DRS1 gene and a second mutation in another gene prove to be lethal to the cell. This is referred to as a synthetic lethal interaction.

From this screen three mutants have been identified, sld1, sld2, sld3, (synthetic lethal with drs1) which appear to interact with the Drs1p. All synthetic lethal strains show deficit in 60S ribosomal subunit when analyzed in a wild type DRS1 background. Thus, this genetic screen has identified relevant genes and the mutant phenotypes suggest that the SLD gene products are involved in the assembly of the 60S ribosomal subunit. Work on cloning the SLD2 gene has identified a complementing region of chromosome XVI. This region includes genes encoding a ribosomal protein and an ATPase.


Alex Proekt, Carnegie Mellon University
(Advisor: Dr. John A. Pollock)

Characterization of a Polyclonal anti-helmsman (N63) Antibody

Research in Dr. Pollock's lab is targeted towards understanding the molecular events that lead to the formation of the visual system in Drosophila. Thorough characterization of the genes expressed in the visual system during development is crucial in understanding the molecular events that underlie formation of the neural network required for vision. One project involves the characterization of a novel gene helmsman (hlm). hlm was first identified because it is expressed in the adult retina and was later shown by Northern blot analysis to be expressed during the earlier stages of development of the visual system and formation of the neural network. Through in situ hybridization experiments it was shown that hlm is expressed not only in the developing retina, but also in the trachea during embryogenesis. Analysis of the hlmcDNA shows that the protein that is encoded by the gene is 533 amino acids long, contains three N-linked glycosylation sites and a signal sequence. Presence of the glycosylation sites and signal sequence indicate that the protein functions extracellularly, but says nothing about the precise localization of the gene product. Knowledge of the subcellular localization of HLM would provide information about possible functions of the gene. The project that I have been working on involves subcellular localization of the gene product of HLM. The overall goal of my project is creation of the spatial and temporal map of the expression of the gene product. This goal can be accomplished by tagging the gene product with a fluorescently labeled antibody. hlmgene product was overexpressed as a fusion protein with a maltose binding protein in E. Coli and purified by affinity chromatography. Polyclonal antibodies against the fusion construct have been made and tested by Western analysis and tissue staining. In the Western analysis the antibody was found to recognize a band of approximately 70 kD as expected from the sequence. Signal was found to be more intense in the head than in the body. Western analysis of adult eya flies, a strain that lacks eyes, showed no signal as expected. Tissue staining of the adult visual system was performed and staining was localized to the retina as was expected from the in situ hybridization analysis.

Image of Chandra K. Wheeler

Chandra K. Wheeler, Carnegie Mellon University
(Advisor: Dr. Adam Linstedt)

Targeting of Giantin to the Golgi Complex

The goal of my project was to determine what part of giantin, a C-terminally anchored protein, is responsible for Golgi targeting. The targeting of C-terminally anchored proteins is not very well understood for two reasons. First, these proteins are not targeted in a srp dependent manner as most other integral membrane proteins are. Second, the mechanism of vesicle transport with proteins that have such large cytoplasmic domains is unknown. Using restriction enzymes and the polymerase chain reaction (PCR), constructs were generated with and without the transmembrane domain of giantin (which was hypothesized to be responsible for Golgi targeting). Once the constructs were created, bacteria were transformed and mini preps were done. Then the constructs were transfected into mammalian cells. Finally, immunofluorescence was used to localize the new constructs within the cells. I tried to create two different constructs this summer. The first construct consisted of the transmembrane domain from giantin and a hemagglutinin epitope for performing immunofluorescence. The plan was to use PCR to copy the transmembrane domain of giantin but, unfortunately, the PCR would not work. The second construct consisted of full length giantin with the transmembrane domain from syntaxin. Syntaxin is another C-terminally anchored protein but it is localized to the plasma membrane rather than the Golgi complex. This construct was created but the results from the transfection were inconclusive. If I had had more time this summer, I would have liked to have tried a few different things on my constructs. For the first construct, perhaps the PCR would have worked with a new primer. For the second construct, I would have tried another method of transfection.