The ubiquination pathway is an essential cell signaling process involved in protein regulation. This pathway uses three proteins in succession [an E1, E2 and E3] to attach the highly conserved 76 amino acid protein aptly named ubiquitin to a substrate. There are 2 E1s, 30-40 E2s and hundreds of E3s in eukaryotic cells. p53 is a substrate of the ubiquitin pathway and arguably the most important protein in the body for cancer prevention. Over 50% of all cancers have a mutated form of the p53 gene. p53 regulation is highly dependent on the E3 proteins MDMX and MDM2. The MDMX and MDM2 proteins have approximately 35% sequence homology and can function as either homodimers or a heterodimer. However, the interactions and differences between MDMX and MDM2 homodimers and heterodimer with E2 enzymes are poorly understood. The E2 plays an essential role in substrate regulation as the E2-E3 pair determines specificity of ubiquitin transfer onto the substrate. Using biochemical and structural techniques I have been able to demonstrate a difference in specificity for E2 enzymes depending on which form of MDMX and MDM2 protein is present. Enzymatic assays have provided evidence of specialized E2 interactions. Further developments in these interactions and assembly of E3-E2 pairs are being made possible by using Nuclear Magnetic Resonance spectroscopy [NMR]. This technique allows the exact orientation of atoms in a given environment to be determined. By comparing an E2 environment with and without the MDMX and MDM2 proteins, NMR provides insight into how and why only a subset of E2s interacts with these E3s.

Ribosome biogenesis is one of the most highly coordinated energy consuming cellular processes. Mutations of genes involved in ribosome biogenesis have been linked to developmental malformations, inherited bone marrow failure syndromes, cancer and aging. Ribosomal proteins are well documented structural components of the ribosome. Despite their ubiquitous expression, mutation of individual ribosomal protein genes results in cell and tissue specific defects. Phenotypic analysis of mice with mutations in various ribosomal protein genes demonstrates that some are not essential for global development or maintenance of tissues. Disruption of Rpl22, a component of the 60S ribosomal subunit, results in subtle defects in lymphocyte development. Our hypothesis is that this surprisingly mild phenotype is due to compensation by a recently identified ribosomal protein, Rpl22L1 (Rpl22-like 1), which has a protein sequence 69% identical to Rpl22. Previously Rpl22L1 was thought to be an unprocessed pseudogene but we and others have found that Rpl22L1 mRNA is translated. Rpl22L1 was also recently identified as a minor component of ribosomes in wild-type liver and mammary gland tissues. Western blot analysis of various mouse tissues showed differential expression of Rpl22 and Rpl22L1. By performing quantitative RT-PCR and western blot analysis, we detected an increase in Rpl22L1 mRNA and protein levels in Rpl22-null tissues. Quantitative RT-PCR analysis of mRNA isolated from T-cell progenitors showed that Rpl22L1 expression does not appear to be substantially increased in Rpl22-null lymphocytes, suggesting that Rpl22L1 is not compensating in this cell population. Western blot analysis performed on gradient fractions collected after polysome analysis of Rpl22-null mice liver also indicated that Rpl22L1 is present in actively translating ribosomes. Future studies include a plan to create and characterize Rpl22L1-null mice. Also, using cells in culture we will explore the mechanism by which Rpl22L1 is upregulated in the absence of Rpl22.

Ribosomes are essential multisubunit protein-RNA complexes responsible for translating mRNAs into proteins. In yeast (S. cerevisiae), the 78 ribosomal subunits are encoded by 137 genes due to the existence of duplicated paralogs in 59 cases. Although many of the duplicated genes encode the same or similar proteins, deletion of a single paralog often yields a specific phenotype such as lifespan extension, drug sensitivity, or altered cellular morphology. We discovered that a subset of ribosomal protein gene deletions gives rise to resistance to stress imposed on the endoplasmic reticulum (ER). We hypothesize that ribosomal protein deletions lead to decreased protein synthesis and translocation into the ER. Reduction in protein load may allow the ER to better deal with stress imposed by the glycosylation inhibitor tunicamycin. We further predict that while paralogs of the same gene encode similar protein products, each may contribute varying amounts of protein, causing phenotypic differences. Although ribosomal protein gene deletions exist, many have growth rate suppressors that make analysis difficult. As a result, we have regenerated the entire haploid ribosomal protein deletion set. Polysome profiling of select paralogs often confirmed decreased protein translation with the deletion of one paralog, but not the other, confirming that the amount of protein contribution from each gene to the ribosome varies. Moreover, analysis of the response to tunicamycin of all deletion strains indicated a strong positive correlation between slow growth and tunicamycin resistance.  Surprisingly, this resistance was independent of the unfolded protein response (UPR), which is a primary pathway to ER stress resistance. Together these findings suggest a model in which reduced protein translation is protective against ER stress. Current efforts are directed at testing other known phenotypes of ribosomal protein gene deletions with the new suppressor-free set to determine whether ribosome specificity exists and, if so, to what extent.

Mutations in the LMNA gene causes abnormalities in nuclear structure and alterations in gene expression which may lead to increased susceptibility to damage in response to stress, resulting in a variety of diseases termed laminopathies. Lamin A has been shown to interact with a variety of proteins, including heat shock factor (hsf) 2, which induces heat shock protein (hsp) transcription during the stress response. The purpose of this study was to determine if the interaction of lamin A and hsf2 is required for hsf2 activity, the subsequent induction of hsp70, and the activation of a cellular defense pathway. We propose that the decrease in this protective response is one mechanism which may cause LMNA-/- cells to be more susceptible to death. We show that LMNA is required for the nuclear localization of hsf2, as the absence of LMNA causes the translocation of hsf2 from the nucleus to the cytoplasm, where hsf2 is inactive. Additionally, LMNA-/- cells have impaired induction of hsp70 in response to the hsf2 activator, MG132, compared to LMNA+/+ fibroblasts. Subsequently, LMNA -/- fibroblasts are more susceptible to apoptotic death induced by MG132. These results demonstrate that the interaction of lamin A with hsf2 is required to maintain nuclear hsf2 and induce hsp70 expression. The inability of LMNA-/- cells to activate hsf2 and induce hsp70 may be one mechanism which causes LMNA-/- cells to be more susceptible to apoptosis.

Muscle creatine kinase (MCK) is an enzyme found in striated muscle cells and it is first detected at the onset of muscle differentiation. Transgenic mouse studies have shown that MCK gene expression is partially regulated by modulatory region-1 (MR-1) that resides within the gene’s first intron. In this study, we are using transgenic mice carrying test genes which are either “wildtype” or which are lacking the MR-1 region. Both test genes are linked to a beta-galactosidase reporter cDNA as a way of measuring their expression. The project aims to optimize a chemiluminescent assay for measuring the expression of MCK transgenes in different striated muscles and to develop a standard curve for these comparisons. An immediate technical problem in these studies was that the beta-galactosidase activity in the muscle extracts was unstable. To optimize this assay, lysis buffers containing either bovine serum albumin (BSA), dithiotheritol (DTT), or protease inhibitors were used to test their ability to help stabilize the beta-galactosidase activity. The lysis buffer containing protease inhibitors that were diluted in dimethyl sulfoxide (DMSO), and the lysis buffer containing DTT reduced the beta-galactosidase activity of the muscle samples. However, the lysis buffer containing BSA stabilized the beta-galactosidase activity. The BSA-lysis buffer is now being used for the analysis of transgenic muscle samples and the results of these assays will be reported.

"Bumped Kinase Inhibitors" (BKIs) are a type of small molecule inhibitor that have a large bulky aromatic group that binds selectively to the ATP-binding pocket of protein kinases that possess a small gatekeeper residue. The majority of human kinases have a large gatekeeper residue (Methionine, Phenylalanine, Leucine, Isoleucine) that does not permit the binding of this bulky aromatic “bump”, thus allowing us to selectively target non-human kinases that have smaller residues (Glycine, Alanine) at this position. Recent work has confirmed that the parasites Toxoplasma gondii and Cryptosporidium parvum each have a calcium-dependent protein kinase (CDPK-1) that has a Glycine gatekeeper, making these microorganisms potential targets for our BKIs. This discovery has exciting applications as both of these parasites are infectious to humans and current therapies have met higher levels of resistance, making the discovery of new drug targets a necessity. We have since screened our panel against these kinases and determined that they inhibit at low nanomolar concentrations. Also, we have validated that the inhibition of these kinases with our BKIs blocks the invasion of these parasites into human cells. Furthermore, screening the panel against human kinases with large gatekeeper residues has shown that the BKIs bind with very low affinity. However, a small subset of human kinases has Threonine gatekeepers and future work must be done on these kinases before the complete selectivity of BKIs can be guaranteed. Currently, we have focused on testing against human Threonine gatekeeper kinases and generating new compounds with even greater potency for the target kinases.

Cancer cells grow rapidly and uncontrollably, invading normal tissue and metastasizing throughout the entire body. The proliferative ability of cancer cells is reminiscent of the properties of earlier stages of development, such as embryonic stem cells. Some of the most aggressive tumors have a similar gene expression signature to embryonic stem cells. Furthermore, low oxygen concentration and hypoxic environments are common among aggressive tumors. We have shown that a link between hypoxia and the activation of key embryonic stem cell genes exist. Hypoxia inducible factor, HIF, a transcription factor that is stabilized in hypoxia, can change a cell’s metabolic state and induce expression of key stem cell markers. We are now testing whether human embryonic stem cells and cancer stem cells share a characteristic metabolic signature and further whether this signature is acquired by HIF activation. For this analysis, we established a quantitative real time polymerase chain reaction based assay to determine the number of mitochondria in a cell. We show that cancer cells have a similar level of mitochondria to human embryonic stem cells, and the overexpression of key HIF responsive stem cell markers such as microRNA 302 is sufficient to affect mitochondrial number. Furthermore, we show, in both human and mouse cells, that cells earliest in development have fewest mitochondria and as development progresses, mitochondria increase. We will proceed in testing the specific stage mitochondria are activated, and the specific role of HIF in creating this unique stem-cell-like metabolic state in pathological and normal conditions.