2002 Summer Scholar Participants-HHMI Undergraduate Program - Carnegie Mellon University

2002 Summer HHMI-supported Participants

Heather Bennett

Heather Bennett
Mentor: Dr. Peter Berget

Functionality Of CD-Tagged Mus Musculus Ribosomal Protein S11 In Saccharomyces Cerevisiae

The CD-tagging method has resulted in the tagging of many genes in mouse NIH 3T3 cells. My project included experiments to test whether the mouse ribosomal protein S11 remained functional after CD-tagging with green fluorescent protein. It also included experiments to determined if the mouse rpS11 could function in yeast cells to reverse the effects of an S11 deletion in yeast. Studying the expression of a CD-tagged protein in S. cerevisiae is a useful tool because this eukaryote is easy to grow and isolate ribosomes from. Yeast contain two copies of rpS11 gene (S11A and S11B) that make identical proteins. The cDNA of tagged and untagged mouse rpS11 gene was cloned into the yeast expression vector pCu416, whose promoter is induced by the presence of a trace amount of copper. These plasmids that contained either tagged or untagged mouse rpS11 were transformed into two different yeast strains: one that contained an S11A gene knockout, and its parent with both S11A and S11B functioning. If the mouse S11 protein was expressed in yeast, then the strains containing tagged S11 should fluoresce under the microscope. Fluorescence microscopy confirmed that the mouse protein was being expressed and was localized in the cytoplasm and nucleus. Further microscopy will be needed to determine if the CD-tagged protein is localized in the nucleolar caps. These results indicate that the protein is functioning normally, based on its cellular location. Sucrose gradient centrifugation was performed on extracts of the yeast strains to determine if the polysome profiles resembled those of the S11A knockout strain or the parental strain. It was concluded that the mouse S11 protein does not compensate for the S11A knockout, because the profiles of the tagged and untagged S11 gene in the knockout strain resembled the knockout's profile. If the mouse S11 did compensate for the deletion, the profile would be expected to shift toward the profile of the parental yeast strain. Fluorescence measurements of the gradient fractions were taken to see if the tagged mouse S11 protein was being incorporated into the yeast ribosomes. It was concluded that the protein was being inefficiently incorporated because there was only very slight fluorescence detected in the 40S and 80S polysome profile fractions. Growth curve analysis was performed to determine if the expression of mouse S11 changed the yeast strains' growth rates. Strains with uninduced rpS11 were expected to grow at a slower rate than induced rpS11. However, yeast cells without mouse S11 grew at a faster rate than those with S11, but when induced with copper, yeast with and without mouse S11 grew at approximately the same rate. This was an unexpected result because the presence of uninduced mouse S11 seemed to have a negative affect on the growth of both the S11A knockout and its parent strain. Future experiments should include polysome profiles for mouse S11 in the parental yeast strain and the creation of a diploid yeast strain with both S11A and S11B genes knocked out. This knockout, containing the mouse rpS11, would be sporulated and the products analyzed to see if a yeast cell can survive with only mouse S11 protein.

Jeffrey Clarke

Jeffrey Clarke
(Mentor: Dr. Jonathan Minden)

Caspase Inhibition in H99 Deficient Drosophila Embryos

Programmed cell death has a vital role in the development of Drosophila embryos. Deletion of the H99 region on both chromosomes of the Drosophila genome has been shown to inhibit apoptosis during all stages of embryogenesis. Proteins known as caspases, which are also required for apoptosis, have little or no effect when inhibited by short, peptide sequences in wild type embryos. A goal of this project was to determine the effect specific caspase inhibitors have in H99 deficient heterozygous embryos. It was hypothesized that the effects of the caspase inhibitors would be exaggerated due to the additional blocking of some caspases caused by the H99 deletion. To begin, the i>Drosophila fly stocks were tested for the deletion of the H99 region and balancer chromosomes using markers for cell death and the balancer. Due to the lethality of H99 homozygous embryos late in development, stocks must be heterozygous for the gene. Therefore, a 1:2:1 ratio is expected in the progeny. Acridine orange(AO), a fluorescent dye that marks apoptotic cells, was used to stain batches of embryos to reveal stocks that contained the H99 deletion. Next, a green fluorescent protein gene (GFP) was utilized to identify the presence of the balancer chromosome. Once Drosophila stocks were appropriately identified as heterozygous for H99 and containing the GFP marked balancer, the stocks were used for all future experiments. Embryos were collected and injected with caspase inhibitors 3 and 9, AO, and Vgal to show cell engulfment. Normally, homozygous H99 embryos have no cell death, while heterozygous H99 embryos have wild type levels of cell death. Fluorescent time-lapse movies taken during embryogenesis suggested that cell death and macrophage engulfment in heterozygous embryos were inhibited. An important conclusion that can be drawn is that the ability of the short peptide chain inhibitors to hinder cell death was amplified due to the heterozygous H99 background. Future experiments could focus on the testing of different combinations of short peptide sequence caspase inhibitors in a H99 background to further clarify which caspases are necessary and sufficient for apoptosis in Drosophila embryos.

Deborah DeJong

Deborah DeJong
(Mentor: Dr. Charles Ettensohn)

The Role of Aristaless in the Cell Fate Specification in the Sea Urchin Embryo

Aristaless is a paired-class homeodomain transcription factor that first appears in skeletogenic primary mesenchyme cells (PMCs) of the sea urchin embryo. In situ hybridization has revealed that aristaless mRNA can be detected as early as the 56 cell stage and is present in high concentrations through late cleavage and blastula. Morpholino experiments which blocked the translation of the aristaless message demonstrated that aristaless is necessary for the creation of the PMCs and the formation of the urchin skeleton. The exact role of aristaless in urchin development is still unknown. The goal of my project was to examine the temporal and spatial expression of the aristaless protein and compare these results to the in situ experiments. This was accomplished through immunostaining with an antibody that had been raised against the aristaless protein. Stainings with this antibody confirmed that the aristaless protein is contained only in the PMC lineage and is expressed from late cleavage through gastrula. Double staining experiments localized the protein to the nucleus of these cells. Immunostaining was also performed on trans-fate embryos in which non-PMC cells were experimentally induced to express a skeletal cell fate. Further antibody staining experiments will be completed on manipulated embryos to elucidate how aristaless functions in dictating the PMC cell fate.

Bucky Ferozan

Bucky Ferozan
(Mentor: Dr. William Brown)

Structural Characterization of Isocyanate Conjugates

Isocyanates are very reactive compounds that are used in commercial and biochemical applications. They are known to produce chronic allergic responses among some individuals that results in occupational asthma. Isocyanates can react with a variety of proteins in vivo. Human serum albumin (HSA) is one of the predominant proteins that are found to be modified when an individual is exposed to isocyanates. Depending on the pH of the reaction, an isocyanate like para-toluene monoisocyanate (TMI) can react with different amino acid side chains on HSA. An antibody single chain variable fragment (ScFv #12) developed in the lab can detect these differences and binds favorably to conjugates made at a pH greater than 8.4. Based on binding studies, we hypothesize that ScFv #12 recognizes the urea bond between TMI and lysine residues on HSA. To confirm this hypothesis, the objective of this research was to characterize the amount of lysine modification on TMI conjugates made at various pH values. A TNBS (2,4-trinitrobenzenesulfonic acid) assay was done to determine the free amine content on the protein. Mass spectrometry was used to determine the total number of TMI molecules that are bound to HSA. As a result, these two assays determine the extent of lysine modifications by TMI on pH variant conjugates.

Rakhi Jattani

Rakhi Jattani
Mentor: Dr. Javier Lopez)

Genetic Analysis of a Putative Splicing Regulator in Drosophila melanogaster

Alternative splicing of mRNA generates protein diversity. Mechanisms of alternative splicing regulation are understood poorly. Defects in splicing may lead to disease, including cancer. The purpose of this study was to use the Ultrabithorax (Ubx) gene of Drosophila as a model system to understand the regulation of splicing. Past work has suggested that the CG9373 gene plays an important role in the inclusion of alternatively spliced exon mI in Ubx mRNAs. This has been supported by two observations. First, this gene encodes a putative RNA binding protein that contains an unusual RNA binding domain similar to that in the B52ED mutant form of SR protein B52. The B52ED mutant interferes with the regulation of mI inclusion. Second, a genetic interaction between B52ED and a multigene deficiency that deletes CG9373 has been observed. The mI and mII exons contain a GC-rich element similar to a B52 binding site. This CG-rich region is required for inclusion of mI, suggesting that the CG9373 protein is involved in the splicing mechanism. The method of testing the role of CG9373 involves the use of double-stranded RNA interference (RNAi). The RNAi will be used to reduce expression of CG9373 in the Drosophila cell line SL2. The result should be increased resplicing of mI. Previous work showed that Ubx mRNA was expressed at extremely low levels in SL2 cells, so detection and quantitation of splice variants was difficult. I generated stable transformants of the SL2 cell line with Ubx minigene constructs and mutant variants, producing increased expression of Ubx for further testing on the splicing mechanism. Variations in levels of CG9373 expression could play a role in regulation of alternative splicing during development. To test this, I collected wild type Drosophila flies at different stages of development. Total mRNA was isolated and assayed using RT-PCR. The results showed CG9373 mRNA was expressed throughout embryonic development without significant variation in the level of expression. Male and female larvae were also collected during the 3rd instar. Results showed expression of CG9373 mRNA at similar levels among male and female larvae and wild type embryos. In addition, embryos were to be stained by in situ hybridization to detect spatial expression of CG9373, but the probe could not be generated successfully. Future research will involve analysis of the stably transformed SL2 cell lines for mI exclusion due to resplicing upon depletion of CG9373 product. Analysis of the ability of CG9373 protein to bind and bridge mI and mII will also aid in understanding the splicing mechanism.

Lisa Kwisnek

Lisa Kwisnek
(Mentor: Dr. Jonathan Minden)

DIGE studies of Gal-4 induced expression of the rpr gene in Drosophila to characterize a novel molecular pathway of apoptosis

During the development of Drosophila melanogaster, cell division and death is highly regulated so as to control morphogenesis of tissues and eliminate unneeded cells. Studies have shown that cell death can occur to correct induced events like hyperplastic growth or expanded regions of the fate map (Li et al., 1999). Systems of regulation, like apoptosis, are conserved throughout animal development. Because information gathered from Drosophila studies can be correlated to humans, fruit flies serve as a model organism for cell death research. Drosophila proteins, including Reaper, HID, and Grim, have been implemented in the processes leading to cell death by counteracting apoptotic inhibitor proteins called IAPs. Ultimately, this leads to the breakdown of the nuclear envelope and the fragmentation of DNA. However, many of the proteins downstream of Reaper that lead to this degradation are still unknown. The goal of this project was to collect embryos after Reaper expression begins but before apoptosis commences and to study their proteomes using Difference Gel Electrophoresis (DIGE), a 2-D method of separation. DIGE is unique in that samples are labeled then homogenized and run simultaneously on the first dimension, which separates based on charge. Then, the first dimension is subjected to PAGE for molecular mass separation. Imaging the gel at two different wavelengths allows one to visualize the labels separately. Using transgenic UAS-rpr flies, Reaper expression was controlled by an upstream activation sequence inducible by Gal-4. 3-D time-lapse microscopy was performed to determine the onset of cell death. Thus, embryos were collected and injected prior to cellularization (~stage 5) with Gal-4 and acridine orange (AO). AO is a fluorescence indicator that is activated when DNA begins to degrade in a common apoptotic response. As expected, induction with Gal-4 prior to cellularization caused rpr expression in UAS-rpr embryos and initiated cell death approximately 67.7 (+/- 14.5) min post-cellularization. The wildtype embryos were also injected, but cell death started 102.8 (+/- 12.7) min post-cellularization. Our wildtype results coincide with reported statistics; cell death is first detected about late stage 11, peaks, and fades by stage 14. To study the cytoplasmic molecular content downstream of rpr expression but prior to cell death, embryos were collected 75 min post-cellularization. Control (WT) and experimental (UAS-rpr) embryos were lysed then labeled with Cy5 and Cy3, respectively. Changes in the proteome of rpr-expressing embryos with respect to wildtype were observed. These results may lead to the characterization of novel proteins in the apoptotic pathway. MALDI-TOF and Q-TOF MS was performed to determine the molecule(s) involved. The MS results were submitted to a blast search called Mascot, which returned sequence possibilities. Unfortunately, the Mascot score was low due to low MS intensities. These low intensities are attributed to small protein concentrations recovered from the DIGE gel. Future studies will run two simultaneous reciprocal gels loaded with approximately 100 control and 100 mutant embryos per gel. This should increase the analyzable protein concentration sufficiently and shed more focused light on the ambiguous pathway to achieve cellular apoptosis.

Andrew Meyers

Andrew Meyers
(Mentor: Dr. Nathan Urban)

The Effects of Naris Occlusion Upon Odor Differentiation in Mice

The capacity of mammals to distinguish one smell from another is an ability, not unique to mice, that is dependent on the olfactory bulb. A common question among researchers in the field of olfaction is whether or not there is an anatomical parallel with the frequently observed behavioral changes and adaptations within the observable system of olfaction. The direct question examined here is if naris occlusion in mice will affect their ability to distinguish between similar odorants, and if it does, how long the natural process of neurogenesis will take to reestablish some degree of normal functioning in the olfactory bulb.

The first steps towards accomplishing these goals was find a reproducible behavioral discrimination test, one in which an animal is habituated to a unique odorant, introduced to a novel but similar odor, followed by a novel though unrelated odor. The animals in this experiment were habituated to an odorant over a series of ten minute timed trials, with different animals requiring from 30 to 50 minutes with the odor to become familiarized to it. They were then introduced to a similar odor, such as orange being similar to tangerine, and then an unrelated odor, such as ginger to tangerine. Times spent actively sniffing the odor introduction apparatus were recorded and compared over a series of mice and odorants. These numbers yielded varied results, with one of the conclusions being that the mice were able to distinguish accurately between two similar odorants and between habituated odors and novel odorants.

Following these behavioral experiments, a reliable method for the anatomical visualization of the activated periglomerular cells in the olfactory bulb was needed. Two methods were attempted, the "Faglu" method, using a formaldehyde and gluteraldehyde fixative, and the "SPG" method using a sucrose, phosphate, and glyoxylic acid fixative. In theory, these methods of fixation would allow researchers to visualize the active cells in the olfactory bulb, those actively expressing dopamine. These methods were however unreliable and inconsistent, and no dependable results could be obtained from the anatomical portion of the experiments.

The experiments were moderately successful overall. More research will need to be done however, in order to determine a dependable means of visualizing the dopaminergic cells in the olfactory bulb. Once this has been accomplished, the naris occlusion can be performed, the mice can be retested behaviorally, and the anatomical differences can be quantified.

Michael Palmer

Michael Palmer
(Mentor: Dr. Jonathan Minden)

Effects of Buffering YPD Media on the Growth of Yeast Cultures, and the Effect of a Heterozygous Diploid Deletion mutant of RPB-2 on the Yeast Proteome

In the past, variations in proteomes of identical yeast cultures grown on different days were found, while no changes were found for cultures in the same media on the same day. These inconsistencies made analysis of different proteomes difficult because the source of variation was unclear. The major goal of my work was to examine if buffering the growth media allowed for a more consistent and reproducible proteome from cultures grown on different days.

A total of eight cultures were grown, four a day on two days. On each day two wild type strains were grown under two separate conditions. Each strain was grown in a buffered (YPD + MOPS pH 6.5) medium and also unbuffered (YPD pH 6.5) medium. Cells were grown up to an optical density of 0.5, collected by a series of centrifugation steps, and then lysed. Proteins were collected and proteomes were analyzed using Difference Gel Electrophoresis (DiGE). DiGE is an alternate form of 2D PAGE that allows two samples to be fluorescently labeled with different dyes and run on the same gel, making comparison of the samples easier than with conventional 2D PAGE methods.

Another experiment was performed that was unrelated to the major focus of my research. The experiment compared a heterozygous diploid deletion mutant of RPB-2 and the parent wild type strain. A previous experiment was done exactly the same on a synthetic minimal media, pH 5.8, but no major changes were found. The goal was to analyze proteome changes caused by reducing the RPB-2 availability by 0.5 in a different, more competitive environment. Three parents and their subsequent mutant strains were all grown up on the same day in identical YPD media pH 5.8. The same collection and analysis as the aforementioned was performed. No major changes were seen.

Mary Ellen Wiltrout

Mary Ellen Wiltrout
(Mentor: Dr. Chien Ho)

Structural and Functional Characteristics of Recombinant Hemoglobins, (a29Leu ' Trp) and (a28Leu ' Trp)

Hemoglobin (Hb) is an important oxygen carrying protein in the human blood system. Due to the problem of shortages in the supply of donor blood, recombinant hemoglobins (rHbs) are being designed and researched as a possible Hb-based oxygen carrier in blood substitutes. Desirable characteristics of a blood substitute are low oxygen affinity, high cooperativity, and slow rates of auto- and NO-induced oxidation. rHb (aL29W) and rHb (aL28W) have been created to examine the oxygen-binding and cooperativity properties, oxidation rates, and structural features. Previously, it has been shown that oxidation of the heme iron atoms from the reduced Fe2+ state to the oxidized Fe3+state was inhibited for a mutation at a29 to phenylalanine in myoglobin, and the NO reaction was lower in both the deoxy and oxy forms of normal adult hemoglobin with the same mutation [Carver et al. (1992) J. Biol. Chem. 267, 14443-14450; Brantley et al. (1993) J. Biol. Chem. 268, 6995-7010]. This property is biologically significant to study, since the reduced (ferrous) state of hemoglobin is functional in humans but not the oxidized (ferric) state. In order to express the hemoglobin molecule with the desired mutation for this study, a system was implemented using Escherichia coli JM109 and plasmids designed with the rHb (aL29W) and rHb (aL28W) genes. Purification steps followed to isolate the hemoglobin. 1H-NMR data was collected to look into the structure of these particular mutants. Some distinguishable differences are visible on the spectra for rHb (aL29W) dealing with the distal heme pocket of the alpha chain. From 1H-NMR experimentation, evidence for the formation of met-hemoglobin on the spectra for rHb (aL28W) suggests it would have a less desirable, higher rate of autoxidation. Further studies on oxygen binding affinity illustrated rHb (aL29W) had a lower oxygen affinity and decreased cooperativity. Oxygen-binding experiments could not be performed on rHb (aL28W) due to significant met-hemoglobin formation. Future experiments will be conducted on the autoxidation rate of rHb (aL29W.)