Large multi-component structures play an essential role in many crucial cellular processes. The morphogenetic pathway of the long, non-contractile tail of bacteriophage λ provides a superb paradigm for studying the assembly of macromolecular complexes. This thesis describes the structural and functional characterization of two λ tail proteins, gpU and gpV, with the aim of improving our understanding of phage tail assembly and evolution, while also providing a starting point to answering some of the fundamental questions surrounding the assembly and function of other supramolecular structures. Tail Terminator Proteins (TrPs) play an essential role in regulating the length of phage tails, and serve as the interaction surface for phage heads. To provide insight into the mechanisms by which TrPs exert their functions, I have determined the X-ray crystal structure of gpU, the TrP from phage λ, in its biologically relevant hexameric state. The gpU hexamer displays several flexible loops that are involved in head and tail binding. By comparing the hexameric crystal structure of gpU to its previously determined NMR solution structure I was able to identify large structural rearrangements in the protein, which are likely induced upon oligomerization. In addition, I have shown that the hexameric structure of gpU is very similar to the structure of a putative TrP from a contractile phage tail even though they display no detectable sequence similarity. This finding implies that the TrPs of non-contractile tailed phages are evolutionarily related to those of contractile-tailed phages. To determine the mechanism by which tail tubes self-assemble prior to termination, I have determined the NMR solution structure of the N-terminal domain of gpV (gpVN), the protein comprising the major portion of the phage λ tail tube. I found that approximately 30% of gpVN is disordered in solution and that some of these disordered regions are biologically important. Intriguingly, my gpVN structure is very similar to a previously solved tail tube protein from a contractile-tailed phage, once again suggesting an evolutionary connection between these two distinct tail types. A remarkable structural similarity is also seen to the hexameric structure of Hcp1, a component of the bacterial type VI secretion system. This finding, coupled with other similarities between phage and type VI secretion proteins support an evolutionary relationship between these systems. Using Hcp1 as a model, I proposed a mechanism for the oligomerization and polymerization of gpV involving several disorder-to-order transitions. Further supporting the importance of unstructured regions, I have shown that the unstructured linker between the N- and C-terminal domains of gpV is crucial for protein function and that a complete truncation of the C-terminal domain (gpVC) results in a 100-fold decrease in activity compared to full-length gpV (gpVFL). To provide insight into the role of gpVC, I determined its NMR solution structure and showed that it possesses an Ig-like fold, however the function of gpVC remains unknown. Interestingly, the gpVC structure revealed the location of two residues that when mutated were previously shown to either abrogate (G222D) or restore (G222D/P227L) function of gpVFL. In addition to being inactive, I demonstrated that the G222D mutation also exerts a temperature dependent dominant negative phenotype. My preliminary NMR data suggests that G222D causes gpVC to partially unfold and that this destabilized form of the domain interacts with gpVN in a region that is likely involved in both oligomerization and hexamer-hexamer interactions. To further our understanding of how these mutations exert their effect, I determined the NMR solution structure of gpVC-P227L. My structure reveals that the β7-β8 region of gpVC-P227L is altered compared to gpVC-WT and suggests that the conformational changes in gpVC-P227L may protect the domain from protein-folding defects induced by the G222D mutation.

This thesis develops and utilizes in vitro approaches to study ciliate/fish interactions. The thesis is divided into six chapters. Chapter one reviews the literature on culturing ciliates and fish cells. Chapter two develops methods for culturing the ciliate Tetrahymena thermophila in media developed and used for mammalian and piscine cells. Chapter three explores the interactions of T. thermophila with monolayers of epithelial cells from fish and mammals. Chapter four studies the interactions of T. corlissi, T. thermophila, and T. canadensis with monolayers of epithelial and fibroblasts from a wide range of animals. The interactions of T. thermophila with the fish viruses are described for the rhabdovirus, viral hemorrhagic septicemia virus (VHSV), in chapter five and for the aquareovirus, Chum salmon reovirus (CSV) in chapter six. The summaries for these six chapters are presented in the following six paragraphs. How the ciliates of fish can be cultured and used to study ciliate/fish interactions are reviewed. The culturing of ciliates is done in media based on either freshwater, seawater, or vertebrate bodily fluids together either with bacteria, fish cells, or organic matter, which can be undefined, such as proteose peptone, or defined. Some ciliates can be pathogenic but with a variable dependency on the fish host. The most dependent and difficult to culture has been Ichthyophthirius multifiliis. Cryptocaryon irritans has been maintained successfully in co-cultures with fish cells. Pathogenic scuticociliates and tetrahymenas can be cultured axenically. Established cultures have been used to screen drugs for their potential chemotherapeutic value and to study pathogenic mechanisms. As well as being pathogens, ciliates interact with fish in other ways. Free-living forms can modulate the activities of other fish microbial pathogens and be food for fish larvae. Tetrahymena spp. have been shown in culture to phagocytose pathogenic bacteria and microsporidia spores. Large-scale cultures of both freshwater and marine ciliates have been achieved and could be a source of feed for fish larvae. In the future cell cultures should be invaluable in studying these and other possible relationships between fish and ciliates. The transfer of Tetrahymena thermophila from normosmotic solutions (~20 to 80 mOsm/kg H2O) to hyperosmotic solutions (> 290 mOsm/kg H2O) was investigated. During the first 24 h of transfer from proteose-peptone yeast extract (PPYE) to either 10 mM HEPES or PPYE with added NaCl to give ~300 mOsm/kg H2O, most ciliates died in HEPES but survived in PPYE. Supplementing hyperosmotic HEPES or PPYE with fetal bovine serum (FBS) enhanced survival. When ciliates were transferred from PPYE to a basal medium for vertebrate cells, L-15 (~320 mOsm/kg H2O), only a few survived the first 24 h but many survived when the starting cell density at transfer was high (100,000 cells/mL) or FBS was present. These results suggest that nutrients and/or osmolytes in either PPYE or FBS helped ciliates survive the switch to hyperosmotic solutions. FBS also stimulated T. thermophila growth in normosmotic HEPES and PPYE and in hyperosmotic L-15. In L-15 with 10 % FBS the ciliates proliferated for several months and could undergo phagocytosis and bacterivory. These cell culture systems and results can be used to explore how some Tetrahymena species function in hyperosmotic hosts and act as opportunistic pathogens of vertebrates. Although several species of Tetrahymena are often described as histophagous and opportunistic pathogens of fish, little is known about ciliate/fish cell interactions, but one approach for studying these is in vitro with cell lines. In this study T. thermophila, B1975 (wild type) and NP1 (temperature sensitive mutant for phagocytosis) were cultured on monolayers of three fish epithelial cell lines, CHSE-214, RTgill-W1, and ZEB2J, and of the rabbit kidney epithelial cell line, RK-13. Generally the ciliates flourished, whereas the monolayers died, being completely consumed over several days. The destruction of monolayers required that the ciliates be able to make contact with the animal cells through swimming, which appeared to dislodge or loosen cells so that they could concurrently be phagocytosed. The ciliates internalized into food vacuoles ZEB2J from cell monolayers as well as from cell suspensions. Phagocytosis was essential for monolayer destruction as monolayers remained intact under conditions where phagocytosis was impeded, such as 37 °C for NP1 and 4 °C for B1975. Monolayers of fish cells supported proliferation of ciliates. These results show for the first time that T. thermophila can ‘eat’ animal cells or be histophagous in vitro, with the potential to be histophagous in vivo. The activities of T. corlissi, T. thermophila, and T. canadensis were studied in co-culture with cell lines of insects, fish, amphibians, and mammals. These ciliates remained viable regardless of the animal cell line partner. All three species could engulf animal cells in suspension. However, if the animal cells were monolayer cultures, the monolayers were obliterated by T. corlissi and T. thermophila. Both fibroblast and epithelial monolayers were destroyed but the destruction of human cell monolayers was done more effectively by T. thermophila. By contrast, T. canadensis was unable to destroy any monolayer. At 4 °C T. thermophila and T. corlissi did not undergo phagocytosis and did not destroy monolayers, whereas T. canadensis was able to undergo phagocytosis but still could not destroy monolayers. Therefore, monolayer destruction appeared to require phagocytosis, but by itself this was insufficient. Additionally the ciliates expressed a unique swimming behavior. Tetrahymena corlissi and T. thermophila swam vigorously and repeatedly into the monolayer, which seemed to loosen or dislodge cells, whereas T. canadensis swam above the monolayer. Therefore differences in swimming behavior might explain why T. corlissi has been reported to be a pathogen but T. canadensis has not. Incubating the fish pathogen VHSV with the ciliate T. thermophila, inactivated the virus, depending on the incubation temperature. Without the ciliates, the VHSV titre declined significantly over 72 h at 30 °C, but remained unchanged at 22 °C and 14 °C. At 30 °C, the ciliates only slightly enhanced the heat inactivation of VHSV. At 22 °C, the ciliates inactivated a substantial proportion of the VHSV by 24 h but no inactivation had occurred by 72 h at 14 °C. The ciliates vigorously phagocytosed fluorescent beads at 22 °C but not at 14 °C. When VHSV were labeled with the nucleic acid stain SYBR Gold, internalization of the virus into food vacuoles was seen at 22 °C. Thus phagocytosis was one possible mechanism for VHSV inactivation by ciliates. However, another VHSV/ciliate interaction was revealed by immunofluorescent staining and might contribute to inactivation. After being incubated for 24 h with VHSV, washed, and stained at various times afterwards for VHSV G protein, the ciliates stained transitorily. The strongest staining was seen at approximately 30 minutes after washing and was confined largely to the cilia but after 60 minutes this staining was lost. Tetrahymena thermophila strains B1975, wild type, and NP1, a temperature sensitive mutant, activated the fish aquareovirus CSV, depending on the temperature. CSV caused fish cells to form syncytia. This cytopathic effect (CPE) was used to titre CSV in the fish cell line, CHSE-214. The CSV titre remained stable during incubations of up to 96 h in Leibovitz’s L-15 with FBS at 4, 14, 22 and 30 °C. When CSV was incubated with B1975 or NP1 at 22 °C in the same medium for between 24 and 96 h, the virus titre increased approximately 3 log. At 4 °C, the titre was unchanged by ciliates and T. thermophila was unable to phagocytose beads. At 30 °C, B1975 enhanced CSV infectivity and underwent phagocytosis, whereas NP1 did neither. When CSV were labeled with the nucleic acid stain SYBR Gold, internalization of the virus into B1975 food vacuoles was seen. Therefore the viral activation pathway likely involved phagocytosis. Tetrahymena canadensis were incubated with CSV-infected CHSE-214, washed by centrifugation through a step gradient of polysucrose, and transferred to new CHSE-214 cultures, which developed the characteristic CSV CPE. Thus as well as activating CSV, ciliates could transport CSV.

Graduation date: 1992

Although induction of innate immune responses during viral infection is essential, it can cause acute inflammation and lead to devastating results. The deleterious effect of innate immune responses has been demonstrated in gene therapy where administration of a replication deficient adenoviral vector (Ad) caused fatality during a clinical trial. Despite recent advances in our understanding of the innate immunity, there is a lack of understanding on how different cell types interact to mount inflammatory responses, which may play an important role in regulating immune responses in vivo. In this study, we investigated the interaction between macrophages and epithelial cells, the two major cell types capable of sensing and responding to viral infection in the airway, in induction of inflammatory responses against replication deficient Ads. We show in Chapter 2 that Ad infection of the macrophage-epithelial cell co-culture resulted in synergistic induction of inflammatory responses. Ad infection of the co-culture compared to macrophages alone resulted in higher cytotoxicity and induction of significantly higher levels of inflammatory mediators including pro-inflammatory cytokines, chemokines, nitric oxide, and reactive oxygen species. We found that these synergistic responses require macrophages and epithelial cells to be in close proximity suggesting that a novel mechanism regulates the inflammatory responses. In Chapter 3, we studied whether ATP plays a role in regulating inflammatory responses during acute Ad infection. Using the co-culture system, we found that ATP signaling through P2X7 receptor (P2X7R) is critical as inhibition or deficiency of P2X7R resulted in reduced inflammatory responses. We demonstrate that ATP-P2X7R signaling regulates inflammasome activation and IL-1β secretion. Furthermore, intranasal administration of Ad resulted in high mortality in mice but inhibition of ATP-P2X7R signaling enhanced survival and reduced inflammatory responses. These results suggest that ATP released by the infected cells plays an important role in regulating inflammatory responses during acute viral infection.

Thymidine kinase is a key enzyme in the nucleotide salvage pathway, catalyzing the production of dTMP from thymidine and ATP. In order to identify the structural features of the TK protein and/or primary amino acid sequences which contribute to the catalytic and regulatory activities of this enzyme, an in vitro transcription and translation system has been used in concert with protein engineering techniques for the production and phenotypic characterization of mutant and wild-type TK enzymes. Because of discrepancies in the literature regarding the quaternary structure of the VVTK, the native molecular weight and quaternary structure was determined to be an 80kDa homotetrameric enzyme by glycerol gradient fractionation, gel filtration and glutaraldehyde cross-linking analyses. Computer-assisted alignment of the predicted amino acid sequences derived from cellular and poxvirus TK genes identified seven highly-conserved domains distributed throughout the VVTK polypeptide (domains I-VII). Domain I (amino acid residues 11-18 ) exhibits a high degree of similarity to both ATP and GTP binding site consensus sequences, although the VVTK utilizes only ATP as a phosphate donor. Site directed mutagenesis and ATP-agarose affinity chromatography techniques were employed to confirm that this region was responsible for ATP binding and to determine which amino acids were essential for efficient binding. The TK gene (tdk) from E. coli was isolated and sequenced to serve as a prokaryotic enzyme with which to compare VVTK. The alignment revealed only 23% shared identity with VVTK and, interestingly, the identical and similar residues were clustered into three of the seven domains identified previously (domains I, III and VII). Preliminary evidence supports domain III (residues 78-90) as a putative magnesium binding site and that a highly conserved cysteine residue (cysteine 170) within domain VII (residues 168-171) may be a component of the catalytic site. Secondary structure alignment between Herpes Simplex Virus (HSV) TK and monkeypox TK (a close relative of VVTK) revealed that the putative nucleoside binding site of HSVTK aligns with residues within domain IV. Replacement of a VVTK residue (Q114) with the corresponding residue of HSVTK (an aspartic acid) greatly alters the substrate specificity and dTTP sensitivity of VVTK. / Graduation date: 1991

Large multi-component structures play an essential role in many crucial cellular processes. The morphogenetic pathway of the long, non-contractile tail of bacteriophage λ provides a superb paradigm for studying the assembly of macromolecular complexes. This thesis describes the structural and functional characterization of two λ tail proteins, gpU and gpV, with the aim of improving our understanding of phage tail assembly and evolution, while also providing a starting point to answering some of the fundamental questions surrounding the assembly and function of other supramolecular structures. Tail Terminator Proteins (TrPs) play an essential role in regulating the length of phage tails, and serve as the interaction surface for phage heads. To provide insight into the mechanisms by which TrPs exert their functions, I have determined the X-ray crystal structure of gpU, the TrP from phage λ, in its biologically relevant hexameric state. The gpU hexamer displays several flexible loops that are involved in head and tail binding. By comparing the hexameric crystal structure of gpU to its previously determined NMR solution structure I was able to identify large structural rearrangements in the protein, which are likely induced upon oligomerization. In addition, I have shown that the hexameric structure of gpU is very similar to the structure of a putative TrP from a contractile phage tail even though they display no detectable sequence similarity. This finding implies that the TrPs of non-contractile tailed phages are evolutionarily related to those of contractile-tailed phages. To determine the mechanism by which tail tubes self-assemble prior to termination, I have determined the NMR solution structure of the N-terminal domain of gpV (gpVN), the protein comprising the major portion of the phage λ tail tube. I found that approximately 30% of gpVN is disordered in solution and that some of these disordered regions are biologically important. Intriguingly, my gpVN structure is very similar to a previously solved tail tube protein from a contractile-tailed phage, once again suggesting an evolutionary connection between these two distinct tail types. A remarkable structural similarity is also seen to the hexameric structure of Hcp1, a component of the bacterial type VI secretion system. This finding, coupled with other similarities between phage and type VI secretion proteins support an evolutionary relationship between these systems. Using Hcp1 as a model, I proposed a mechanism for the oligomerization and polymerization of gpV involving several disorder-to-order transitions. Further supporting the importance of unstructured regions, I have shown that the unstructured linker between the N- and C-terminal domains of gpV is crucial for protein function and that a complete truncation of the C-terminal domain (gpVC) results in a 100-fold decrease in activity compared to full-length gpV (gpVFL). To provide insight into the role of gpVC, I determined its NMR solution structure and showed that it possesses an Ig-like fold, however the function of gpVC remains unknown. Interestingly, the gpVC structure revealed the location of two residues that when mutated were previously shown to either abrogate (G222D) or restore (G222D/P227L) function of gpVFL. In addition to being inactive, I demonstrated that the G222D mutation also exerts a temperature dependent dominant negative phenotype. My preliminary NMR data suggests that G222D causes gpVC to partially unfold and that this destabilized form of the domain interacts with gpVN in a region that is likely involved in both oligomerization and hexamer-hexamer interactions. To further our understanding of how these mutations exert their effect, I determined the NMR solution structure of gpVC-P227L. My structure reveals that the β7-β8 region of gpVC-P227L is altered compared to gpVC-WT and suggests that the conformational changes in gpVC-P227L may protect the domain from protein-folding defects induced by the G222D mutation.

INTRODUCTION: Since the eradication of smallpox in 1977, mass vaccination efforts against it have been discontinued. Thus, the majority of the younger population is susceptible to both smallpox virus and vaccinia virus (VV). The re-emergence or intentional release of smallpox will present a serious threat to global health. There are limited supplies of smallpox vaccine, which is associated with significant complications, and pooled anti-VV human immune globulin (VIG) that can be used as prophylaxis or to treat smallpox-exposed individuals. We are developing murine monoclonal antibodies (MAbs) able to neutralize VV. The developed MAbs may be useful in establishing a rapid diagnostic test for the detection of VV infection or providing the genetic materials needed for developing recombinant antibodies suitable for human use. METHODS: VV Western Reserve (WR) strain was propagated in HeLa or Chicken Embryo Fibroblast (CEF) cell lines, purified through a 36% sucrose cushion and inactivated by binary ethyleneimine (BEI). Female BABL/c mice were immunized with inactivated VV. Hybridoma cell lines (HCLs) were developed from spleen cells of the mice with high neutralizing antibody titers. Tissue culture supernatants from the developed HCLs were screened by Enzyme-Linked Immunosorbent Assay (ELISA) and Plaque Reduction Assay (PRA) for their abilities to produce neutralizing antibodies against VV. HCLs producing neutralizing antibodies were sub-cloned by limiting dilution method. Highly neutralizing MAbs were isotyped and purified. The effect of using increasing microgram amounts of each MAb or mixtures of two MAbs on VV neutralization has been determined. Specific target proteins recognized by MAbs were detected by western blot assay (WB). The abilities of the developed MAbs to neutralize other three VV strains, Large-variant (L-variant), IHD-W and New York City Board of Health (NYCBH), were measured. RESULTS: We have developed 261 HCLs producing anti-VV antibodies; 65 of them neutralized VV. Twelve HCLs were sub-cloned. We developed 79 sub-clones producing neutralizing MAbs. The majority of them were immunoglobulin IgG1/κ isotype. Four highly neutralizing MAbs were concentrated and purified. They were able to neutralize 50% of VV infection at 0.01-0.1 µg in PRAs. Synergistic effects on VV neutralization were observed when mixing two MAbs from clones, 1-E9-1-E4 and 2-B7-9-E6, at the amounts giving about 20% and 40% VV neutralization. Based on the WB results, the developed MAbs are recognizing 75 kilodalton (kDa), 45 kDa, 35 kDa or 8 kDa WR VV proteins. The abilities of the developed MAbs to neutralize other strains of VV varied. CONCLUSIONS: Several HCLs producing antibodies against VV were developed. Highly neutralizing MAbs against WR VV have been produced and purified. Virus neutralization is dose dependent and some of MAbs have synergistic neutralization effects on each other. Most of the MAbs were targeting the same three virus envelope proteins indicating that these proteins contain important epitope(s) responsible for the neutralizing effects by the developed MAbs. Variable neutralization abilities were observed on three other VV strains indicating their immunobiologic differences with WR VV strain. The developed MAbs may be used as a research tool to study VV pathogenesis or for the development of chimeric antibodies for clinical applications. / October 2006

Satellite panicum mosaic virus (SPMV) depends on its helper Panicum mosaic virus (PMV) for replication and movement in host plants. The positive-sense single-stranded genomic RNA of SPMV encodes a 17-kDa capsid protein (CP) to form 16-nm virions. Previous studies showed that SPMV CP has multiple functions during infection including encapsidation, symptom exacerbation, inhibiting the accumulation of SPMV DIs, and facilitating systemic movement. This dissertation confirms and extends the results of our previous reports with new biological and biochemical evidence. For example, the dosage effect of SPMV CP on symptom severity supports its function as a pathogenicity factor. Biological assays also demonstrate compensatory effects of SPMV CP on virus mutants defective in systemic movement. In addition, it is shown for the first time that SPMV CP is involved in cellto- cell movement of SPMV RNA and associated with the cell wall and membranes, a signature property of plant virus movement proteins. However, SPMV CP in the cytosol exists exclusively as virions and is dispensable for symptom exacerbation. SPMV CP contains a distinctive N-terminal arginine-rich motif (N-ARM), which is required for the in vitro binding of SPMV and PMV genomic RNAs by SPMV CP. Mutations of this region impair all known functions of SPMV CP. Interestingly, manipulation of the C-terminus of SPMV CP resulted in the same phenotypes as alterations in the N-ARM except that this does not affect the RNA binding activity of SPMV CP. Biological experiments demonstrate that virions are not required for the properties of SPMV CP to facilitate local and systemic movement and inhibit the accumulation of SPMV DIs, suggesting that SPMV CP and RNA form alternative complexes for these purposes. This dissertation study reveals the nucleolar localization of SPMV CP and its interaction with PMV CP in the form of virions. The identification of distinct functional domains of SPMV CP and its complex subcellular localization profile resulted in the proposal of a tentative model on how the functions of SPMV CP are coordinated for a robust infection. This dissertation provides a foundation for further understanding of the complex interactions among host plants, helper viruses, and satellites.

The coat protein of satellite panicum mosaic virus (SPMV) was used to stabilize viral vector gene inserts in planta. A Potato virus X (PVX) vector carrying the SPMV capsid protein (CP) gene was successfully stabilized through three serial passages in Nicotiana benthamiana from the upper non-inoculated leaves following rub inoculation. The presence of SPMV CP expression from the PVX vector was confirmed by necrotic lesions that occur only when SPMV CP is present and by western blot and reversetranscription PCR analyses. In addition, PVX-SPCP was co-inoculated onto N. benthamiana with a Tomato bushy stunt virus vector carrying a green fluorescent protein gene, which normally does not yield GFP expression in upper tissue due to loss of the insert. However, upon co-inoculation with PVX-SPCP, upper non-inoculated leaves exhibited GFP accumulation based on green fluorescence by UV illumination at 488 nm and western blot analysis. GFP expression was more abundant in upper non-inoculated N. benthamiana leaves as well as systemic tissues when the co-inoculation experiments were performed at 20°C compared to 25°C. These results suggest that SPMV CP is a viable molecular tool for stabilizing viral vector gene inserts in planta.

No description available.