Thesis (Ph.D.)--University of California, San Francisco, 2009. / Source: Dissertation Abstracts International, Volume: 70-04, Section: B, page: 2016. Adviser: Kevan M. Shokat.

Thesis (Ph.D.)--University of California, San Francisco, 2009. / Source: Dissertation Abstracts International, Volume: 70-04, Section: B, page: 2010. Adviser: Thomas B. Kornberg.

Inactivation of the tumor suppressor protein p53 is a very important and common step in the process of carcinogenesis. The overall purpose of this project was to gain a better understanding of the mechanisms involved in the regulation of p53 function. To gain insight into these mechanisms, we chemically mutagenized A1-5 cells expressing high levels of temperature sensitive p53 val135 (tsp53) and selected for clones that were capable of growth at the permissive temperature for p53 activation. The clones generated, called ALTR (for A&barbelow;1-5 L&barbelow;ow T&barbelow;emperature R&barbelow;esistant), could grow at the permissive temperature. Using the ALTR cell system and the parent A1-5 cells, we determined that nuclear translocation of p53 could result in a change in the conformation from mutant to wild-type but that these may be two separable events. We also investigated, in depth, the mechanism by which p53 was inactivated in one ALTR cell line, ALTR 9. We identified calpain mediated degradation of p53 as a partial mechanism of p53 inactivation in these cells. Our results suggest that degradation of p53 by calpain can lead to the functional inactivation of p53 and that this degradation can be regulated by genomic stress. To gain insight into the significance of cytoplasmically sequestered p53 protein in tumors, we chose a neuroblastoma derived cell line, SK-N-SH, that expresses a wild-type but cytoplasmically sequestered p53 protein. We report here, that down regulation of p53 by HPV-16 E6 resulted in the morphological conversion of SK-N-SH cells to substrate-adherent fibroblast-like S-type cells. The morphologic conversion was accompanied by a loss of neurofilament expression, a marker for the neuronal N-type cells, an increase in the expression of vimentin, a lack of responsiveness to RA induced neuronal differentiation, and loss of anchorage independent growth. These results suggest that p53 is required for the maintenance of the neuroblastic tumorigenic phenotype. Both the ALTR cell system and the SK-N-SH cells provided us with insight into the mechanisms involved in p53 inactivation resulting in tumor formation.

Cells of the cardiovascular system express multiple connexins (Cx) with Cx40 and Cx43 being commonly coexpressed in many tissues. The expression levels of connexins are dynamic and can vary in response to a growth stimulus. It is not clear why cells express multiple connexins, or what advantage such dynamic regulation of expression patterns have on cell function. These issues are further complicated by the ability of some connexins to interact to form heterogeneous gap junction channels, with little being known regarding functional properties of such channels. The purpose of these experiments was threefold: (1) To determine whether Cx40 and Cx43 are capable of interacting to form heteromeric/heterotypic gap junction channels; (2) To characterize the functional properties of Cx40/Cx43 heteromeric/heterotypic channels; and (3) To determine the effect that changing Cx40:Cx43 expression ratio has on functional properties of heteromeric/heterotypic channels. Cell lines were developed that express only Cx43 (Rin43), Cx40 (Rin40), and Cx40 and Cx43 in varying Cx40:Cx43 expression ratios (6B5n, A7r5, A7r540C1, and A7r540C3). The Cx40:Cx43 expression ratios in the 6B5N, A7r5, A7r540C1, and A7r540C3 cells are approximately 1:1, 3:1, 5:1, and 10:1, respectively. Functional properties of the gap junction channels formed between these cells were determined using both electrophysiological and dye coupling techniques. Pairing of Rin43 and Rin40 cells demonstrated that Cx40 and Cx43 are capable of forming homomeric/heterotypic gap junctions with unique voltage-dependent gating and single channel behaviors. Rin43/A7r5 cell pairs displayed voltage-dependent gating and single channel conductance profiles that could only be explained by the presence of heteromeric/heterotypic gap junction channels between these cells. Pairing Rin43 cells with coexpressing cells of high Cx40:Cx43 expression ratio resulted in channel activities that were not predicted by the gating and conductance patterns of Cx40/Cx43 heterotypic channels. However, the dye coupling characteristics of these same cells in coculture demonstrated that the permeability of the channels formed between these cell types reflected that of Cx40 channels. In summary, Cx40 and Cx43 are capable of forming heteromeric/heterotypic gap junction channels. Increasing the Cx40:Cx43 ratio in coexpressing cells results in channels with unique gating and conductance properties, however dye permeability of these cells is predicted by their relative Cx40 content. Therefore, varying Cx40:Cx43 expression ratio provides cells with a mechanism to finely control the types of molecules shared between cells.

We previously identified a novel structural variant of the alpha6 integrin called alpha6p. This variant was produced on the cell surface and was missing the β-barrel extracellular domain. Using several different concentrations of amiloride, aminobenzamidine and PAI-1 and the urokinase-type plasminogen Activator (uPA) function blocking antibody (3689) we showed that uPA, acting as a protease, is responsible for production of α6p. We also showed that addition of uPA in the culture media of cells that do not produce α6p, resulted in a dose dependent α6p production. In contrast, the addition of uPA did not result in the cleavage of other integrins. Using α2-antiplasmin and plasmin depleted media, we observed that uPA cleaves the alpha6 integrin directly. Further, 12-O-tetradecanoyl-phorbol-13-acetate (TPA) induced the production of alpha6p, and this induction was abolished by PAI-1 but not α2-antiplasmin. Using site directed mutagenesis we have identified the site of cleavage to be at arginines 594 and 595. We have also shown that while a fraction of α6 integrin is normally associated with CD151, the α6p form is not. In order to determine whether α6 integrin clipping occurs in tissue, we have found that α6p is present in human prostate cancer tissue, in normal mouse epidermis, in mouse papillomas and squamous cell carcinomas induced by DMBA, TPA and MNNG treatments and in mouse melanomas induced by activated ras. Interestingly, subcutaneous injection into athymic nude mice of a malignant mouse keratinocyte derived cell line (6M90) that is α6p negative, results in the development of tumors that contain α6p integrin. Furthermore, we have shown that PC3N cells transfected with an uncleavable mutant of the α6 integrin grew smaller tumors when injected subcutaneously in SCID mice compared to wildtype α6 transfected cells. In addition, the tumors from the uncleavable mutant alpha6 transfected PC3N cells had higher levels of activated caspase 3 indicating higher levels of apoptosis. This finding suggests that the α6 integrin clipping is important for integrin signaling for survival. Collectively, all these data suggest that the cell surface clipping of the α6 integrin is a novel mechanism for altering integrin-laminin interactions during skin tissue remodeling and carcinogenesis.

The (140 kD) α6 integrin is an essential gene product in epithelial cell maintenance and remodeling of the stratified epithelium. The prostate gland is an example of a glandular epithelium. In prostate cancer, alterations of integrins are observed. Specifically, a shift from α6β4 to persistent expression of α6β1 integrin occurs. Accompanying the loss of polarized α6β4 is loss of its extracellular ligand, laminin-5. Using immunofluorescence staining human prostate, breast and colon tissues, were examined for β4 integrin and laminin-5 expression. Loss of β4 and laminin-5 was apparent beginning in PIN lesions and was absent in prostate carcinoma, differing from retained expression in breast and colon carcinoma. These data suggested progressive loss of β4 integrin and laminin-5 occurs and that this combined defect is unique to prostate cancer progression. A novel 70 kD (non-reduced) variant of the α6 integrin, called α6p for the latin word parvus (small), was identified on the cell surface of normal epithelial and carcinoma cell lines. The α6p variant paired with either β1 or β4 subunits and retained sequences corresponding to the extracellular 'stalk region' and the cytoplasmic tail of the α6 integrin. The β-propeller domain postulated to mediate ligand binding, was missing from this variant. Protein levels of α6p increased three fold during calcium-induced terminal differentiation in a normal mouse keratinocyte model system. Production of the α6p variant was dependent upon an intact actin cytoskeleton. Cell surface α6p was less responsive to changes in the actin cytoskeleton, relative to that observed for α6 and β1 integrins, suggesting α6p did not participate in the focal contact. Additionally, inhibition of serine/threonine phosphatases decreased α6 integrin protein levels, but not α6p integrin, again suggesting the variant functioned as an inactive subunit for signaling. Finally, α6, but not α6p integrin co-immunoprecipitated with hemidesmosome components: laminin-5 and CD151. Preliminary data demonstrated adhesion to synthetic peptide integrin antagonists resulted in a 65 kD form of the alpha6p variant with no alteration of α6 integrin. Together the presented data were consistent with differential regulation of alpha6 and α6p integrins and suggested the α6p variant functioned as an inactive receptor.

Vesicle (exosome) externalization is a mechanism for removal of obsolescent membrane proteins, such as the transferrin receptor (TFR), during reticulocyte maturation. This thesis involves further characterization of the exosomal phenomenon using avian (nucleated) and mammalian erythroid systems. Chicken reticulocytes release exosomes, indicating independence of this process from enucleation. Exosome release during chicken erythroleukemic cell (HD3) differentiation (representing pre-reticulocyte development) demonstrates that this process commences prior to the reticulocyte stage. / The heat shock protein, hsp70, has been found in exosomes from every species examined (four species, including the chicken). It is shown that hsp70 is physically, but non-covalently, associated with exosomal TFR, suggesting its possible involvement in targeting proteins into exosomes during their formation. / Exosome formation is an energy-dependent process. Our observations indicate that the primary energy substrate for avian red cells is not glucose, but glutamine, inosine and guanosine. Work with the HD3 cells (capable of glucose transport), suggests there is an early differentiation-associated switch in the chicken red cell's ability to use glucose as a major energy substrate.

Transforming growth factor-beta (TGF-beta) plays an important role in angiogenesis and wound healing. The major cell type involved in angiogenesis are microvascular endothelial cells. Endoglin, a transmembrane glycoprotein, is highly expressed on vascular endothelial cells and has been shown to bind TGF-beta1 and TGF-beta3 and inhibit TGF-beta-induced responses. Mutation in the endoglin gene has been implicated in hereditary haemorrhagic telangiectasia type 1 (HHT1), a dominantly inherited vascular disorder. The role of endoglin in regulating TGF-beta function in the microvasculature is not well understood. The initial interactions on the cell surface between endoglin and the TGF-beta receptors may be an important mechanism by which endoglin modulates TGF-beta signaling and responses. In this study, we show for the first time that on human microvascular endothelial cells, endoglin forms a heteromeric association with betaglycan, a TGF-beta receptor with which endoglin shares limited homology. This complex formation can occur in both a ligand-dependent and ligand-independent manner. In addition, we demonstrate the occurrence of three higher order complexes which contain endoglin and the TGF-beta type II and/or type I receptors in different numbers or ratios. Our results suggest that endoglin may modulate TGF-beta signal transduction by interacting with betaglycan and the TGF-beta signaling receptors on the cell surface in different ratios.

The cardinal feature of apoptosis is the activation of a family of proteases termed caspases, and in some forms, loss of mitochondrial integrity and release of cytochrome c into the cytoplasm are viewed as necessary components of cell death. In receptor-mediated apoptosis, more specifically Somatostatin (SST) receptor (SSTR)- and Fas-signaled apoptosis, an additional feature is intracellular acidification (pHi) and Protein Tyrosine Phosphatase (PTP) activity. Increased translocation and activity of PTP, more specifically SHP-1 at the cell membrane as well as the functional activation of caspase-8 are required to reduce pHi. Also, a decrease in mitochondria) membrane potential (Deltapsim) and release of cytochrome c (cyt c) are concomitant with a decrease in pHi suggesting that mitochondrial dysfunction is a consequence and not a cause of acidification. Inhibition of caspase-8 activation and prevention of SHP-1 translocation abrogates the ability of SST to induce acidification and mitochondrial disruption, whereas inhibition of other caspases such as caspase-9, or -3/7 are ineffective. Protection from cell death can also be accomplished by Bcl-2. In this model, Bcl-2 elevates resting pHi and attenuates acidification and apoptosis only when the overexpressed Bcl-2 is targeted to the endoplasmic reticulum (ER) but not the mitochondria. In conclusion, we established that mitochondrial breakdown follows acidification and that acidification is dependent on caspase-8 and SHP-1 activation. Additionally, effector caspases are induced only when there is concomitant acidification. Bcl-2 cannot prevent acidification-dependent apoptosis at the level of the mitochondria despite protecting it revealing that mitochondrial disruption is a potentiating, but not essential event in the apoptotic pathway.

Previous studies have demonstrated that receptor tyrosine kinases (RTKs) use common intracellular signal transduction pathways. This is remarkable due to the divergent cellular responses elicited from different RTKs. The insulin receptor (IR) is involved in blood glucose homeostasis whereas the epidermal growth factor receptor (EGFR) has been linked to liver regeneration. We therefore used rat liver, an organ enriched in both EGF and insulin receptors, to study the specificity of signal transduction in vivo. / Upon EGF administration, the EGFR at the plasma membrane (PM) was tyrosine phosphorylated on the major in vivo activation site, Y 1173, and recruited the adaptor proteins SHC and GRB2. Following ligand-mediated endocytosis, the activated EGFR was associated on endosomal membranes with the signaling complex SHC/GRB2 and the guanine nucleotide exchange factor, SOS. EGF administration also led to tyrosine phosphorylation of the cytosolic proteins focal adhesion kinase (FAK) and SHC. These observations were associated with increased MAP kinase activity and the transcription of c-myc, c-fos and c-jun. / In response to insulin, IR kinase activity and autophosphorylation was observed at the PM but not after IR internalization into endosomes. This is postulated to be due to rapid degradation of insulin in the endosomal lumen allowing for the dephosphorylation of the IR by protein tyrosine phosphatase(s). To evaluate the role of endosomal degradation in IR signaling, an insulin analog, termed H2, was characterized for its clearance and processing in liver. Although liver uptake of H2 was similar to that of insulin, its clearance was slower. This correlated with reduced ligand dissociation from internalized IR as well as slower degradation kinetics. / In response to H2, IR traffic was delayed in the endosomal apparatus. This occurred with increased IR autophosphorylation and tyrosine kinase activity in this compartment. H2 but not insulin induced JNK activity and c-jun transcription. Insulin stimulated MAP kinase and glycogen synthase (GS) activities in rat liver. H2, however, was less efficient in inducing MAP kinase activity than insulin and GS activity was not observed. Impaired GS activity in response to H2 correlated with increased PKC activity. The above observations of insulin and H2 were not due to changes in SHC nor IRS-1 signaling. / These studies indicate that the modification of RTKs at the level of the endosome alters receptor traffic and specificity of signal transduction pathways and support the hypothesis that RTK endocytosis plays a role in the regulation of RTK signaling.