Simian virus 40 (SV40) entry through caveolae represents a novel form of endocytosis. SV40 translocation to caveolae was mainly examined by cell fractionation and electron microscopy (EM) in CV-1 African green monkey kidney fibroblast cells. SV40 translocated to the caveolin-enriched low density Triton-insoluble complexes as identified by Western blotting. SV40 partitioning into this fraction was inhibited by nystatin. EM analysis indicated that SV40 translocation to caveolae was prevented by the tyrosine kinase inhibitor genistein (thin sections), but SV40 was not blocked from entering the Triton-insoluble low density fraction (negative staining). Northern analysis suggested that an SV40 signal that upregulates c-myc is transmitted through caveolae. The role of the major histocompatibility complex (MHC) class I proteins was not clear in SV40 translocation to caveolae. This is partly because of the high background level of class I molecules in the low density fraction isolated by sodium carbonate sucrose gradients (LDF-C). Class I proteins were enriched in the LDF-C relative to the whole cell extract or to the plasma membrane isolated by silica coating. The clustering of class I molecules was independent of any induction. Indirect immunofluorescence labeling of class I proteins and caveolin indicated that class I molecules colocalize with caveolae. MHC class I signaling and its physiological relevance were analyzed using a monoclonal antibody (mAb) W6/32. By immunoprecipitation and thin layer chromatography, ras was found to be activated by W6/32. W6/32 did not seem to increase the tyrosine-phosphorylation level in phospholipase C-$\gamma1.$ Engagement of class I molecules did not seem to induce apoptosis, but resulted in a distinct physiological response, the adhesion strengthening in CV-1 cells to Jurkat cells. The adhesion strengthening seemed to be sensitive to the protein kinase C inhibitor calphostin C, but enhanced by the tyrosine kinase inhibitor genistein. Furthermore, the mAb-induced adhesion strengthening was transient, antibody-concentration dependent, but independent of the Fc portion of the mAb. A serum factor was found to inhibit the basal level of adhesion in CV-1 cells to Jurkat cells. These findings represent one of the first investigations of MHC class I signaling in non-lymphoid cell lines.

U14 small nucleolar RNA (snoRNA) is required for processing of 18S ribosomal RNA. It was hypothesized that U14 might base pair with 18S RNA through two highly conserved U14 sequence elements known as domains A and B. Using Saccharomyces cerevisiae as the experimental system, I showed that: (1) the domain A and B elements are functionally interdependent, and (2) single-point mutations in domain A combined with complete substitution of domain B causes lethality while either mutation alone does not. Direct interaction of U14 with 18S RNA was shown by demonstrating that a lethal mutation in U14 domain A can be suppressed with a mutation which restores complementarity in the corresponding region of 18S RNA. Y-domain in yeast U14 was postulated to serve as a recognition element for vital intermolecular or intramolecular interactions. Consistent with this assumption, mutations in several conserved nucleotides of the loop cause growth defects. In contrast, alterations to the stem have little or no effect. Using a lethal mutation in the loop, three different intragenic suppressor mutations were mapped to three positions adjacent to the primary mutation, and are predicted to influence the structure of the loop. An extragenic suppressors (UF1) able to rescue a cold-sensitive mutation in the loop encodes an essential putative ATP-dependent RNA helicase. Loss of UF1 gene expression caused a reduction in 18S rRNA production, without affecting accumulation of 25S rRNA or U14 snoRNA. Pulse-chase analysis showed that depletion of UF1 protein impaired pre-18S rRNA processing. Finally, an effort was made to define minimum pre-rRNA substrates that can be used to produce functional 18S and 25S rRNAs in vivo. The rDNA operon was split either between the 18S RNA and 5.8S/25S coding units, or between the 18S/5.8S RNA and 25S RNA coding units. The test fragments were expressed from GAL7 promoters. The results showed that functional rRNAs could be produced in trans, but only when the operon was divided between the 18S RNA and 5.8S/25S RNA coding sequence.

Senescence in plants is a genetically controlled developmental event that enables the plant to eliminate individual cells, tissues, organs and even the entire plant in order to adapt to change in environmental conditions, reproductive needs, or pathogenic attack. An integral part of the cell death occurring during senescence is the selective degradation of proteins. Proteins may be degraded via the activity of several classes of enzymes, characterized by the different amino acids present at their catalytic sites, known as proteinases. Alternatively, protein degradation in eukaryotic cells can be accomplished through the action of an ATP dependent pathway known as the ubiquitin system, which is capable of selecting specific proteins for degradation by creating ubiquitin-protein conjugates that are then hydrolyzed by the 26S proteasome, a multisubunit proteinase. As a first step in determining whether the ubiquitin pathway was involved in the death of cells during vascular tissue development, experiments were conducted to localize the presence of ubiquitin and ubiquitin-protein conjugates to differentiating vascular tissues. Immunolabeling on tissue prints from cross sections of Phaseolus vulgaris petioles, Gossypium hirsutum hypocotyls and Coleus x hybridus internodes showed accumulation of ubiquitin-protein conjugates in regions of vascular tissues. Immunohistochemical labeling confirmed the presence of elevated levels of ubiquitin-protein conjugates in differentiating and regenerating xylem tissues of Coleus. In addition, labeling of ubiquitin-protein conjugates was observed in parenchymatous tissues and elevated levels of ubiquitin-protein conjugates was also observed in vascular cambia. The senescence of daylily (Hemerocallis hybrid cv Stella d'Oro) petals is accompanied by a rapid decline in protein over a 24 h time period. To determine how this loss of protein is regulated we conducted experiments to characterize and measure the activity of several classes proteinases. Treatment with class specific inhibitors indicated the presence of serine, cysteine, and metalloproteinases. The activity of these proteinases increased after flower opening and this increase in activity could be eliminated by treatment with the protein synthesis inhibitor cycloheximide. Ion leakage was delayed by treatment with inhibitors of the 26S proteasome, suggesting a link between protein hydrolysis and membrane permeability. Immunoblots labeling ubiquitin-protein conjugates and ubiquitin system enzymes revealed that overall profiles of ubiquitinated proteins do not change during senescence but that several enzymes of the ubiquitin system decline after flower opening. Northern analysis showed slight accumulation polyubiquitin message during petal senescence. We propose that senescence in daylily petals is controlled in part by the appearance of at least three classes of proteinases which may work in concert with the ubiquitin system.

Apoptosis is a type of programmed cell death that is essential for the development and maintenance of tissue homeostasis in multicellular organisms. Because apoptosis is important in many biological situations, laboratories are working to delineate the signaling pathways responsible for this process. This thesis work attempts to identify new components of the signaling pathways by examining a role for mitochondria and proteasomes in T cell apoptosis. Interest in mitochondria was initiated when a library screen performed in the laboratory revealed that portions of the mitochondrial genome, called apt-1 and apt-3, were downregulated in thymocytes during negative selection. In this thesis work, two aspects of mitochondrial function were examined in the T cell hybridoma, DO11.10: the integrity of the mitochondrial genome as assessed by levels of apt-l and apt-3 and the integrity of the mitochondrial membrane as assessed by membrane potential. Both the steady-state levels of apt-l and apt-3 RNA and mitochondrial transmembrane potential declined rapidly in apoptotic cells. Recent studies from other laboratories indicate that these changes are the result of the active participation of mitochondria in apoptosis. Apoptosis relies heavily on proteolysis, and this dependence has encouraged investigators to analyze the significance of many proteases and proteolytic pathways. The second part of this thesis work focused on describing a role for the ubiquitin-proteasome pathway in apoptosis. Proteasome inhibitors were used to determine whether the proteasome is required in thymocyte apoptosis. Treatment of thymocytes with acetyl-leu-leu-methioninal (LLM), acetyl-leu-leu-norleucinal (LLnL), carbobenzoxyl-leu-leu-leucinal (MG132), or lactacystin inhibited death induced by dexamethasone, ionizing radiation, or phorbol 12-myristate 13-acetate (PMA). These results suggest that proteasome activity is necessary for the progression of many apoptotic pathways in thymocytes. Inhibitor studies also indicate that the proteasome functions upstream of many apoptotic events, including: PARP cleavage, caspase-3 activation, and mitochondrial transmembrane potential depolarization. Although a precise role for the proteasome in apoptosis cannot be assigned without an identification of its proteolytic targets, the work described in this thesis provides the first definitive link between proteasomes and apoptosis in mammalian systems.

The antioxidant activities of compounds endogenous to bovine rod outer segments (ROS) were investigated by measuring the loss of polyunsaturated fatty acids (PUFA's) from membranes exposed to the water-soluble oxidant $2,2\sp\prime$-azobis(2-amidinopropane) dihydrochloride (AAPH). Osmotically intact ROS, ROS membranes, and unilamellar liposomes prepared from ROS phospholipids (PL) were compared. Intact ROS were most resistant to oxidative loss of PUFA's, followed by ROS membranes and then PL liposomes. The development of a model membrane system allowed the investigation of putative antioxidants singly and in combination. Lipid-soluble compounds (vitamin E, free fatty acids, retinol, retinaldehyde) were incorporated into PL liposomes. Water-soluble compounds (reduced glutathione, taurine) were dissolved in incubation buffer. It was found that normal physiological concentrations of alpha-tocopherol (vitamin E), free fatty acids (16:0, 18:0, 18:1, 22:6), retinol, and glutathione significantly decreased oxidative loss of PUFA's. When the major free fatty acids were added to PL liposomes at the same concentrations found when ROS phospholipase A is stimulated, the oxidative loss of PUFA's was reduced by 31%. The antioxidant effect of free fatty acids suggests that endogenous phospholipase A's may act to protect membranes by releasing fatty acids from phospholipids in proportions and concentrations that afford protection to membrane lipids. The antioxidant activities of the two major retinoid compounds involved in the visual cycle in ROS, all-trans retinaldehyde and all-trans retinol, were compared. The addition of retinol partially protected PUFA's in ROS PL liposomes, whereas retinaldehyde promoted lipid peroxidation. When isolated ROS were stimulated to produce endogenous retinol, PUFA loss was inhibited by up to 17%. These findings suggest that there is an antioxidant function for the enzymatic reduction of retinaldehyde to retinol during photoreception. Water-soluble antioxidants, taurine and reduced glutathione (GSH), were investigated individually and in combination with retinol in ROS PL liposomes. GSH significantly protected PUFA's in ROS PL liposomes. Taurine alone showed little antioxidant activity, but in combination with retinol protected lipids by up to 47.9% (an increase in antioxidant protection of 24.8% over retinol alone). These data support previous findings that taurine protects ROS lipids during exposure to cyclic light conditions.

The generation of site-directed mutants within the actin binding domain of the EGF receptor modulates receptor function in internalization and ligand binding. In addition, truncation of the EGFr at residue 996 results in a loss of high affinity ligand binding, inhibited internalization and reduced signaling capacity. Mutation of tyrosine 992 to phenylalanine (Y992F) and glutamate 991 to glutamine (E991Q) increases the rate at which receptors are internalized. The presence of a phenylalanine residue eliminates EGFr-mediated phosphorylation at Tyr992 while the E991Q mutation might also eliminate phosphorylation at this position due to a disruption of the kinase recognition motif. Thus, phosphorylation of Tyr992 appears to function in the regulation of receptor internalization. The mutation of tyrosine 992 to a glutamate residue (Y992E) causes a three-fold increase in receptor affinity for its ligand and demonstrates the existence of novel third and potentially fourth affinity states for the EGFr. A very high affinity EGFr state with a K$\sb{\rm d}$ of approximately 10 pM has been identified as has an intermediate state of 1.5 nM. The deletion of the C-terminal 190 amino acids of the EGFr causes a complete abolition of the previously observed high affinity state of the EGFr and also causes a significant reduction in the affinity of the low affinity state of the EGFr. Phorbol ester treatment of wild type and mutant EGFr causes a loss of the high affinity receptors, and also a decrease in the overall affinity of the receptor for its ligand which is similar to the loss seen in the deletion mutant. This suggests that control of the affinity state of the EGFr is mediated through the C-terminal 190 amino acids of the receptor. In addition, the C-terminal 190 amino acids of the receptor have been identified as containing a domain which regulates the phorbol ester induced conversion of receptor affinity. The amino acid composition in the vicinity of tyrosine 992 has been shown to play a role in the internalization of the EGF receptor and in the regulation of receptor affinity for its ligand.

Long nanotubes of fluid-bilayers were used to create templates for photochemical polymerization into solid-phase conduits and networks. Micromechanical methods were developed which allowed each nanotube to be pulled from a micropipette-held feeder vesicle by mechanical retraction of the vesicle after molecular bonding to a rigid substrate. The caliber of the tube was controlled precisely in a range from 20 to 200 nanometers by setting the suction pressure in the micropipette. Branched conduits were formed by coalescing separate nanotubes drawn serially from the feeder vesicle surface. Single nanotubes and nanotube junctions could be linked together between bonding sites on a surface to create a functionalized network. After assembly, the templates were stabilized by photoinitiated radical cross-linking of hydrophilic monomer contained in the aqueous solution confined by the lipid bilayer boundary. Nanometer-sized vesicles that were prepared by extrusion were also used as templates for photopolymerization. Results from dynamic light scattering and electron microscopy experiments suggest that UV initiated, free-radical polymerization of vesicle-encapsulated monomer resulted in the formation of cross-linked polymer networks that were surrounded by a bilayer lipid membrane. Using fluorescence spectroscopy to monitor the release of an initially entrapped marker, it was determined that lumenal polymerization neither disrupted the semi-permeable membrane, nor did it effect the osmotic release of encapsulated solutes. The addition of detergent to a suspension of polymerized vesicles completely dissolved the bilayer membrane, leaving behind a rigid gel replica of the vesicular template. Two polymerizable amphiphiles with reactive headgroups were prepared and incorporated into the phospholipid bilayer, so as to provide a means to copolymerize the bilayer sheath with the liposome-encapsulated monomer. By monitoring the release of entrapped solutes, it was discovered that photopolymerization of vesicles constructed with either one of the amphiphilic monomers in combination with encapsulated monomer resulted in membrane destabilization, and the complete release of the entrapped solutes. This represented a new approach to the photoinduced release of vesicle contents. In fact, vesicles that contained the polymerizable lipid but not the entrapped monomer also exhibited contents release upon polymerization. During polymerization, the propagating membrane-bound polymer destabilized the membrane. Two simple models were put forth to explain the photoinduced release from polymerized vesicles. One takes into consideration the mechanical stresses that develop in the membrane asymmetric polymerization; the second suggests that release of contents was caused by formation of a membrane-bound, polymeric surfactant which forms “pores” in the membrane.

Cytokinesis, the process of physically separating cells for division, requires the precise orchestration of numerous physical, mechanical, chemical and biological processes. For these processes to function well, complex coordination of various proteins, with crosstalk between them, either as signaling molecules or as just plain structural components that contribute to the physical separation must exist. Actin, a structural polymer and myosin, a motor are two proteins that contribute to this process significantly. Both proteins are assembled in the contractile ring and together are responsible for the process of constriction. A thorough understanding of the behavior of these proteins, in the contractile ring as well as outside in a cell undergoing cytokinesis is therefore important to prevent possible defects that might lead to deleterious diseases. In this dissertation research, a combination of techniques are made use of, that involve live imaging of fluorescently labeled proteins in cells undergoing cytokinesis along with the use of drugs that either disrupt the structure (integrity) or function of cytokinetic proteins. I generated two LLCPK1 (pig epithelial cell lines); one that stably expresses GFP-actin and the other that stably expresses Tandemn Dimer RFP-myosin regulatory light chain (TDRFP-MRLC). Live imaging and analysis of cells expressing GFP-actin shows that actin in the contractile is highly dynamic and need to be dynamic. Evidence is presented for new roles of Myosin II, in addition to generating the force for cytokinesis. Myosin not only contributes to disassembly of actin in the contractile ring but is also required to maintain actin in the equatorial region. Live imaging of the cell lines that expresses TDRFP-MRLC or GFP-actin helped in the better understanding of the role of microtubules in simultaneously regulating actin and myosin dynamics, not only in the contractile ring to allow ingression, but also in preventing contractile activity outside in the contractile ring. Cytokinesis involves other proteins besides actin and myosin, which help in their recruitment, assembly, ingression and subsequent disassembly. Decreasing the accumulation of actin in the contractile ring, by treatment with Latrunculin B facilitated the examination of spatial and temporal events involved in building the ring. Actin, myosin and other proteins organized as nodes that coalesce during ingression, similar to the fission yeast. We conclude that this mode of cytokinesis a highly conserved feature of cytokinesis.

At the adult frog neuromuscular junctions (Rana pipien and Rana catesbiana), F-actin microfilaments are enriched in the nonrelease domains of nerve terminals, outside the vesicle-rich release sites. The development of this defined F-actin cytoskeleton may be critical for nerve terminal function as microfilaments may play a role in synaptic vesicle release and recycling and/or synaptic maintenance. I used cutaneous pectoris muscles of adult frogs (Rana pipiens) and bullfrog larvae (Rana catesbiana) stained with markers of synaptic and cytoskeletal components to ask how elements of the neuronal cytoskeleton, microfilaments, microtubules, and neurofilaments, are organized at the frog neuromuscular junction and how they organize during development of presynaptic motor nerve terminals. The presynaptic actin cytoskeleton stained by β-actin antibody extended the length of the nerve terminal in a series of interconnected rings that surrounded clusters of synaptic vesicles. At developing neuromuscular junctions β-actin stain is initially concentrated at growth cones and intermittently along the lateral surfaces of the nerve terminal. The assembly of the β-actin cytoskeleton appeared secondary to clustering of synaptic vesicles. I compared the stability of the presynaptic actin cytoskeleton of developing and adult neuromuscular junctions after treatment with latrunculin A. The β-actin cytoskeleton was noticeably less stable at larval neuromuscular junctions than at adult synapses. These data support a role for the actin cytoskeleton in presynaptic maturation and stability. I tested whether the perisynaptic Schwann cells (PSCs) have a role in maintaining the actin cytoskeleton of the nerve terminal using complement-mediated cell lysis to selectively ablate PSCs in vivo. At various time points after ablation, I examined the actin organization at denuded motor nerve terminals. I report here that the stability and long-term maintenance of the nerve terminal actin cytoskeleton is dependent at least in part on the presence of PSCs. Following ablation of PSCs, a significant decrease in the intensity of presynaptic actin stain was observed and remained altered for several weeks. After PSC ablation, terminals also displayed reduced staining for synaptic vesicles.

Autoreactive, immature T cells (thymocytes) are deleted from the thymus during development by the process of negative selection. This mechanism occurs when the thymocyte, T cell receptor (TCR) recognizes self-antigen, causing the cell to die by an apoptotic pathway. This mechanism results in the deletion of autoreactive T cells. Thymocyte development proceeds through several stages, determined by the differential expression of the T cell co-receptor molecules CD4 and CD8. The developmental stage where negative selection occurs is one in which thymocytes are expressing a functional TCR on the cell surface along with both CD4 and CD8, termed the double positive stage. These thymocytes are in a quiescent, G0 state and make up greater than 80% of the total population. The demonstration that cell cycle progression plays a role in the apoptotic process of several quiescent cell types, as well as the requirement for mature T cells to be in the late G1 phase during activation induced cell death, suggested that thymocytes may also advance to the G1 stage of the cell cycle prior to apoptosis. It has been established that the early cell cycle genes c-fos, c-jun, and c-myc are induced in thymocytes after stimulation, indicating possible entry into the cell cycle. The hypothesis that thymocytes enter the cell cycle before undergoing apoptosis was tested by examining expression levels of the various G1 cyclins and cyclin dependent kinase inhibitors at the mRNA and protein levels. Several indications of an early G1 cell cycle transition occurring during thymocyte apoptosis were observed, such as the downregulation of p27KIP1 and p130, the upregulation of cyclin D3, and the phosphorylation of the retinoblastoma protein. Finally, the requirement for the activation of the cyclin-dependent kinase 2 (CDK2) in negative selection was examined. It was shown that the phosphorylation and expression of the TCR-mediated apoptosis-related transcription factors Nur77 and Egr1 are downstream of CDK2 activation. In addition, a protein associated with the transcription factor Egr-1 was identified as a possible target of CDK2 kinase activity.