The molecular interaction of quaternary ammonium compounds (QAC) with bacterial cells was investigated using various techniques to obtain the knowledge on what kinds of damage are done to the cell that results in cell death. Electron microscopy and flow cytometry were used to evaluate direct effects of didecyldimethyl ammonium chloride (DDAC) to Pseudomonas aeruginosa cells. The characteristic effects of DDAC observed on the bacterial structure were the formation and pinching off of blebs from the outer membrane and the formation of multilamellar structures within the nucleoid area. This was followed by coagulation of the entire nucleoid region and cytoplasm. The morphological changes observed in the intact gram-negative bacterium probably accounted for the bactericidal action of DDAC. An extensive theoretical discussion was provided to explain the observed effects of DDAC on the bacterial envelopes and cytoplasm. A flow cytometric approach was evaluated to determine its use for assessing antibacterial activity of a variety of QAC on P. aeruginosa ATCC 15442. LIVE/DEAD BacLight{dollar}\rm \sp{lcub}TM{rcub}{dollar} viability kit and the SYTOX{dollar}\rm \sp{lcub}TM{rcub}{dollar} kit were used as vital fluorescent stains. The DDAC had the greatest activity by converting the most number of bacteria in the population from green to red, followed by tetradecyl benzalkonium chlorides (BKC), hexadecyl BKC and dodecyl BKC. The flow cytometer provided an excellent means to assess the antibacterial activity of individual QAC. In order to decipher the interaction of DDAC with E. coli lipids, a steady-state and time-resolved emission spectra of 2-dimethylamino-6-lauroyinaphthalene (Laurdan), and steady-state anisotropy decay of 1,6-diphenyl-1,3,5,-hexatriene (DPH) were performed. Finally, {dollar}\sp{lcub}31{rcub}{dollar}P NMR spectroscopy allowed for convenient and qualitative measurements of the effects of DDAC on lipid polymorphism which seems relevant in the DDAC effects on membrane perturbation and possibly bactericidal activity. A schematic model for DDAC-membrane interactions that takes into account Laurdan spectral data, and turbidimetric data in conjunction with {dollar}\sp{lcub}31{rcub}{dollar}P NMR results was presented, which attempts to explain the different modes of action of DDAC on lipid supramolecular structures with respect to molar % of DDAC to lipids.

Human cytomegalovirus (HCMV) can cause severe sight-threatening and/or life-threatening disease, particularly in immunocompromised hosts. A 92.5 kDa cell membrane protein was identified as a putative receptor for HCMV envelope glycoprotein H (gH) that mediates virus/cell fusion. Four steps were taken to identify this protein and verify its function: (1) identification of a fusion-negative cell line, (2) isolation of cDNA clones that encode the receptor, (3) isolation of receptor peptides, and (4) transfection of fusion-negative cells with receptor cDNA. Identification of fusion-incompetent cells was performed using immunofluorescence (IFA) and immunoblot assays employing anti-idiotypic antibodies that mimic gH, and a fusion assay with R18-labeled HCMV. CEM, HEL, HFF, CHO and Vero cells showed relatively high receptor densities consistent with their ability to fuse with HCMV, whereas MOLT-4 cells had very low receptor density and were not able to fuse with HCMV. Clones containing cDNA of the receptor were identified from HEL lambdagtl1 expression libraries by screening with the anti-idiotypic antibodies. Inserts from two (131 and 611) of several identified clones were subsequently isolated and fused in frame with the glutathione S-transferase (GST) gene in a pGEX-4T-1 vector, and proteins were expressed in E. coli. The purified proteins (FR131 and FR611) were shown to bind specifically to the antibodies, and inhibit HCMV/cell fusion and viral plaque formation in a specific and dose-dependent manner. The cDNA encoding the partial receptor contained an open reading frame with a predicted 345 amino acids consisting of two potential transmembrane spanning domains and several possible nuclear localization signals. A region of similarity was also identified between the predicted protein and two DNA binding proteins. Transfection of fusion-incompetent MOLT-4 cells with a eucaryotic expression vector containing the partial receptor cDNA rendered these cells susceptible to HCMV/cell fusion. In addition, fusion blocking studies using synthetic receptor peptides indicated the HCMV binding domain is contained within an 18 amino acid segment from the extracellular region in the middle of the protein. This work not only identifies a novel cell membrane protein, but provides additional evidence for the role of the 92.5 kDa cell membrane protein in HCMV/cell fusion.

Enteropathogenic Escherichia coli (EPEC), an agent of severe infant diarrhea in the less developed world, causes attaching and effacing lesions on epithelial cells of the human intestine and on cultured human cells. These lesions are marked by localized rearrangements of the actin cytoskeleton, resulting in microvillus destruction and the formation of shallow protrusions from the apical surface that intimately bind to the bacteria. The genetics of the attaching and effacing phenotype were investigated by mapping previously-generated transposon insertions that abolish the phenotype. This mapping resulted in the discovery of the locus of enterocyte effacement (LEE), a {dollar}\sim{dollar}35.5-kb cluster of DNA, containing two genes previously shown necessary for AE lesion formation and many hitherto undescribed sequences. DNA probes derived from sequences throughout the LEE hybridized under high stringency to a panel of EPEC and other enteric pathogens that cause AE lesions, but not to non-AE strains of the same species. The nucleotide sequence of 23 kb of the LEE was determined and found to contain 32 novel open reading frames, including nine similar to genes encoding components of specialized secretory systems that secrete virulence factors in a variety of other bacterial pathogens. The {dollar}\rm G+C{dollar} nucleotide composition of the LEE is significantly lower than the E. coli genome as a whole, suggesting that it was acquired by horizontal gene transfer. To see if the LEE contains all genes necessary for the AE phenotype, its 35-kb sequence and approximately 750 bp of flanking DNA were isolated on a cosmid cloning vector. Addition of the LEE-containing cosmid, but not the vector alone, to a variety of avirulent laboratory E. coli strains conferred the ability to secrete at least one EPEC protein from the bacterium, trigger host signal transduction pathways, and induce AE lesions on cultured epithelial cells. These results show that the LEE is a functional genetic unit containing all factors necessary to confer AE lesion formation in vitro.

Bacteriophage T4 is unique among phages in that its growth only requires the Escherichia coli GroEL (cpn60) protein but not the GroES (cpn10) co-chaperonin. Our findings indicate that bacteriophage T4 synthesizes its own co-chaperonin, gp31, which in conjunction with GroEL, is strictly required for the folding of the T4 major capsid protein, gp23. Overexpression work demonstrates that GroEL and gp31 are both necessary and sufficient for the proper folding and oligomerization of the major capsid protein in vivo. Although gp31 and GroES have no significant sequence homology, either on the primary amino acid level or on the predicted secondary structural level, the two proteins appear to be functional homologues. Cryoelectron microscopy of gp31 purified to 90% indicates gp31 forms oligomeric ring structures that are comparable to those formed by GroES. Studies using expression vectors and genetic analysis indicate that although the GroEL-gp31 complex can fold proteins normally folded by GroEL-GroES, the reverse is not true. Thus, even when GroES is overproduced it is unable to mediate chaperonin-assisted folding of the T4 major capsid protein. In addition, co-immunoprecipitation experiments using antisera directed against GroEL and GroES, suggest that GroES protein is functional during T4 infection and able to bind GroEL protein. These studies demonstrate that gp31 is specifically required to fold the T4 major capsid protein because of a difference in co-chaperonin function and not because GroES protein is inactivated or limiting during the course of T4 infection. In addition, co-immunoprecipitation experiments demonstrate that although gp31 is required for the folding of the T4 major capsid protein, gp31 is not required for the binding of the major capsid protein to GroEL. T4bypass31 are mutant phage that can bypass the strict chaperonin requirement for growth. The presence of both the bypass 31-1 and bypass 31-2 mutations in gp23 (BY23) allows the major capsid protein to fold, although less efficiently, in a GroEL-gp31 chaperonin independent mode. The mutations that confer the bypass phenotype have now been sequenced to two specific regions of the gene encoding the major capsid protein which appear to be critical folding sites of the polypeptide as judged by previous genetic analysis. The bypass31-1 mutation is a single missense mutation near the 3 prime end of gene that results in the conversion of Ala-455 to Val-455. The bypass31-2 site, located in the center of gene 23, consists of three independent missense mutations that convert Gly-292 to Ser-292, Val-306 to Ile-306, and Val-307 to Ile-307. Characterization of these mutations by site-directed mutagenesis and genetic studies show that all three mutations at the bypass31-2 site are required for optimal phage growth in the absence of GroEL and gp31, but that Ile-306 and Ile-307 are essential to maintain the bypass phenotype. In addition, all four bypass mutations behave additively at elevated temperatures. Mutational analysis also suggests that the bypass phenotype is inherent to the gp23 polypeptide and is not due to translational pausing. (Abstract shortened by UMI.)

Although the density of the HIV envelope protein (gp160) can vary on cells and virions, there has been little investigation into how the levels affect HIV infection and cell-to-cell fusion. To address this several cell lines stably producing infectious, replication defective virions of the T-cell line adapted HIV isolate, HXB2, were made. The cell lines varied from one another in gp160 surface expression over a five-fold range, while expressing similar levels of the Gag polyprotein. As a consequence, the cell lines collectively produced virus that varied in gp160 content over a six-fold range. With the cell lines it was observed that increasing gp160 surface expression increases cell-to-cell fusion efficiency and syncytium size. With the virus preparations it was found that gp160 levels had a biphasic affect on infection efficiency. Virus preparations with gp160 levels of 10-30 molecules per virion showed that increased gp160 content correlated with increased infection efficiency. However, at gp160 levels above 50 molecules per virion there was a decrease in infection efficiency associated with increasing gp160 per virion content. Therefore, it appears that increasing gp160 surface density exerts a qualitatively different affect on infection and cell-to-cell fusion. Based upon functional analysis, several cell lines producing the gp160 molecule from the primary isolate, JR-FL were also made. These cell lines were less characterized, and yet have the potential to be useful reagents in HIV research as well.

Herpes simplex virus (HSV) genes are regulated in a cascade of immediate early (IE), delayed early (DE), late (L). The large subunit of ribonucleotide reductase (RR) from HSV-2, designated ICP10, has been grouped with DE proteins. The amino-terminal domain of ICP10 has protein kinase (PK) activity and properties similar to growth factor receptor kinases that can be activated to transforming potential. The studies described in this dissertation sought to develop a better understanding of regulatory aspects of ICP10 regulation as well as the role of ICP10 expression in neoplastic transformation. Regulation of expression of the ICP10 gene was studied by immunofluorescence with the intact ICP10 gene or by chloramphenicol acetyltransferase (CAT) analysis with hybrid ICP10 promoter constructions containing the wild type ICP10 promoter or site-directed mutants deficit in specific cis-response motifs. Co-transfection of these constructions with DNA encoding an HSV nonspecific transactivator (IE110) or an IE gene-specific transactivator (Vmw65), enhanced expression at least 10-fold, regardless of the assay system. In contrast, expression was minimally enhanced by DNA encoding a DE gene transactivator (IE175) at low doses and slightly reduced at high doses. Sequence analysis of the ICP10 promoter revealed the presence of both herpesvirus IE gene-specific (TAATGARAT, GA-rich, and {dollar}\alpha{dollar}H2-{dollar}\alpha{dollar}H3 motifs), as well as cellular cis-response motifs (potential SP-1, consensus AP-1, and octamer transcription factor-1 (OTF-1) binding elements). Factors that bind to the ICP10 promoter were identified by gel retardation analysis with mixtures of uninfected cell nuclear extracts and virion lysates or in vitro synthesized OTF-1 and Vmw65. The Oct-1 motif (ATGCAAAT) was necessary for optimal Vmw65 binding to, but not for transactivation of the ICP10 promoter as evidenced by competition experiments with oligonucleotides overlapping the consensus IE110 promoter virion response element and by site-directed mutagenesis of the motif. The 3{dollar}\sp\prime{dollar} portion of the TAATGARAT motif (GARAT) was dispensible for binding but necessary for activation. These data suggest that ICP10 behaves as an IE gene and could therefore affect host gene regulation independent of lytic infection. ICP10-PK has neoplastic transforming potential in vitro. Anchorage independent growth was observed in cells transfected with the vectors that express the entire ICP10 protein or just the PK domain, but not a frameshift (not expressing ICP10) mutant or a carboxy-terminus (RR domain) expression vector.