Because of the well established epidemiologic and biologic interactions between human immunodeficiency virus (HIV) transmission and genital ulcer disease, measures to control chancroid could have a significant impact on the epidemic of HIV. I developed an IgG and IgM antibody serologic enzyme linked immunosorbent assay (ELISA) using pooled sera from clinically and microbiologically proven cases of chancroid as positive controls, and pooled sera from normal individuals without a history of sexually transmitted diseases as negative controls. Cross reactivity was minimized by absorption of serum samples with a sorbent prepared from three other Haemophilus species. Performance of the ELISA was enhanced in the period of early convalescence from acute primary chancroid and was not diminished in the presence of HIV coinfection, reflecting the tempo of the natural serologic response to and acute infection. I also developed an inhibition ELISA to determine antigenicity of potential vaccine candidates in human H. ducreyi infection, using lipooligosaccharide (LOS) as the test antigen. In a panel of 10 sera from cases of primary natural H. ducreyi infection reactive to H. ducreyi 35000 soluble antigen, only 4 samples were identified with reaction to H. ducreyi 35000 LOS. Inhibition ELISA confirmed that pre-adsorption with LOS substantially diminished reactivity of these sera to LOS but not to soluble antigen. Using the ELISA to detect immunogenicity by measuring the serologic response to vaccine candidates in rabbits, we further tested the feasibility of three bacterial antigen preparations to induce protective immunity against infection and disease. LOS carbohydrate and a pilus preparation were purified from H. ducreyi 35000 and used in a booster immunization procedure. Virulence was assayed by intra-epithelial challenge and measurement of disease for homologous strain 35000 or the virulent clinical isolate RO-34. LOS and the pilus preparation both induced humoral responses to the corresponding antigen, but the carbohydrate preparation did not. Immunization with LOS or carbohydrate did not modify virulence of infection with H. ducreyi 35000. Immunization with the strain 35000 pilus preparation significantly reduced the severity of disease, and the duration of infection and disease compared with controls. I conducted passive immunization experiments, and characterized the inflammatory infiltrate of chancroidal lesions. Naive rabbits were passively immunized with 24 or 48 mg of purified polyclonal IgG intravenously and 24 hours after infusion, challenged with the homologous strain 35000. No significant difference in disease resulting from infection with the homologous strain was observed. I then comparatively evaluated the serial immunohistology of lesions produced by infectious challenge with the homologous strain in sham-immunized or rabbits immunized with the pilus preparation. Flow cytometric analysis of rabbit peripheral blood leucocytes with CD5 and CD4 rabbit lymphocyte markers prior to infectious challenge revealed two T lymphocyte populations: CD5+CD4+ and CD5+CD4$-$. Pilus preparation vaccinee lesions showed significant quantitative acceleration and increase in T lymphocyte infiltration, and similar early increased recruitment of the CD4+ T lymphocyte subset preceding early lesion sterilization and early healing without ulceration. (Abstract shortened by UMI.)

Detection of $\beta$-lactamase was performed by the Masuda Test, the Double Disc Synergy Test (DDST), and a novel test developed during the course of this work for ESBL as follows: Serial dilutions of antibiotics against which enzyme activity is to be demonstrated are incorporated into Mueller-Hinton agar containing methylumbeliferyl-$\beta$-D-glucuronide (MUG) 70 $\mu$g/mL. Growth of Escherichia coli on this medium gives fluorescence but growth of K. pneumoniae does not. These plates are seeded with an E. coli lawn and K. pneumoniae strains are spot inoculated by Steers Replicator onto the lawn. At antibiotic concentrations above the MIC of E. coli, fluorescence on the lawn is inhibited. On the same plates, satellite growth of E. coli occurs around enzyme-producing K. pneumoniae isolates and this is manifested as a ring of fluorescence. The incorporation into the medium of clavulanic acid indicates if the enzyme is neutralized. In the Series I the K. pneumoniae isolates positive by DDST were also positive by the satellite fluorescence test. None of the isolates were positive by DDST and negative by fluorescence testing. Conclusions are (1) Routine susceptibility tests do not detect ESBL in K. pneumoniae isolates which may be clinically important. (2) The satellite fluorescence test is a more sensitive test than the DDST. (3) This fluorescence test is a convenient, flexible, and sensitive test for screening ESBL production in K. pneumoniae isolates. (4) A small number of K. pneumoniae isolates in Ottawa show a decreased susceptibility to third generation $\beta$-lactams with evidence of enzyme degradation but are susceptible by zone diameter measurements. (Abstract shortened by UMI.)

Peptostreptococcus species are anaerobic Gram-positive cocci associated with a variety of female genital tract infections that includes bacterial vaginosis and pelvic inflammatory disease. One of the accepted indicators in bacterial vaginosis is the production of proteases which involves Peptostreptococcus spp. One hundred Peptostreptococcus isolates were screened for aminopeptidase activities and for proteolytic activity. Five ATCC Peptostreptococcus strains and 95 clinical isolates comprising P. anaerobius, P. asaccharolyticus, P. magnus, P. micros, and P. prevotii were included. Statistical analysis showed that the prevalence of gelatin hydrolysis is significantly correlated to the number of positive aminopeptidase reactions to synthetic peptides among the 5 Peptostreptococcus species. The rapid ID32A kit was also used to identify the 100 Peptostreptococcus isolates as compared to conventional biochemical tests and gas-liquid chromatography, the current "gold" standard method. Overall, the rapid ID32A kit is a useful method for the rapid differentiation of P. anaerobius and P. asaccharolyticus from other Peptostreptococcus spp. Alternate methods should be used to speciate isolates of P. magnus, P. micros, and P. prevotii. Protein L (L) and albumin-binding (A) proteins have been identified as putative virulence factors of P. magnus isolates. When 32 P. magnus isolates from 4 different countries were screened for protein L gene sequences, the only positive isolates were three which had been previously identified. In order to determine whether these putative virulence factors might be clonal, protein L-containing (L$\sp+$ A$\sp-$) or albumin-binding isolates (L$\sp-$ A$\sp+$) and unrelated isolates (L$\sp-$ A$\sp-$) were analyzed using three genotyping methods. Based on patterns from PFGE and ribotyping, cluster analysis showed that the three L$\sp+$A$\sp-$ isolates were more closely related to each other than with other isolates. Six of eight L$\sp-$A$\sp+$ isolates formed two indistinguishable groups of three isolates each, but these two groups were genetically distant from the other two related L$\sp-$A$\sp+$ isolates. L$\sp-$A$\sp-$ isolates were not related, either among themselves, or to any other isolates. Thus, L$\sp+$ and A$\sp+$ phenotypes and genotyping methods (PFGE and ribotyping) are useful for differentiating P. magnus isolates and identifying specific strain types. In conclusion my studies have found that the proteolytic activity of Peptostreptococcus spp. is actually due in part to an array of aminopeptidase reactions. As well, certain isolates of Peptostreptococcus species require further taxonomic work which will likely lead to changes in classification and nomenclature. All 11 clinical isolates originally identified as P. prevotii by the "gold" standard methodology were other identified as P. magnus (n = 4) or could not be identified by the rapid ID32A kit. Protein L-containing and some albumin-binding P. magnus isolates are clonal in structure. This is in agreement with past studies that most pathogenic isolates of bacteria are descendent from one parental strain. (Abstract shortened by UMI.)

The stability and mechanism of the action of the enzyme Bacillus circulans xylanase (1,4-$\beta$-D-xylanohydrolase, EC 3.2.1.8) were investigated by various biophysical techniques. On the basis of the sequence homology between B. circulans xylanase and xylanase A from Schizophyllum commune, a disulphide bond was introduced between the residues 100 and 148, S100C/N148C (DS1 mutant). The presence of this covalent cross-link leads to a 5$\sp\circ$C increase in the melting point of the protein, as verified by differential scanning calorimetry. Introduction of another disulphide bond in a similar position, V98C/A152C, the DS2 mutant also enhances the stability of the protein by as much as 4$\sp\circ$C. On the basis of the notion that the increase in the stability of a protein is proportional to the number of residues encompassed by the cross-link, the N and C-termini were joined, A1GC/G187,C188, to form the circular xylanase (cXI) mutant. This mutant also acquired a 3.8$\sp\circ$C elevated melting point. A combination of S100C/N148C and A1GC/G187,C188 mutations is accompanied by a 12.3$\sp\circ$C increase in the melting point, which is 3.5$\sp\circ$C more than expected. Thus, the stabilization effect of the two disulphide bonds appears to be cooperative rather than additive. Moreover, the thermodynamic data obtained from differential scanning calorimetry under reversible conditions support previous findings that the stabilizing effect of disulphide bonds is a consequence of a decreased conformational entropy of the unfolded state. Furthermore, the results of this study support the notion that the introduction of a disulphide bond can be used as a strategy to prevent the aggregation of proteins by restricting the exposure of some elements required for this process. The active site of Bacillus circulans xylanase contains two acidic residues, glutamic acids 78 and 172, which are crucial for the catalytic activity of the enzyme. Fourier-transform infrared and near-UV circular dichroism spectroscopies were used to determine the pK$\rm\sb{a}$ of these residues. For the wild type enzyme, a titration of one of carboxylate groups occurs at pH 6.8, as evidenced by FT-IR spectroscopy. This titration is absent in the E78Q of E172Q variants of the enzyme. This, taken with crystallographic data, indicates that glutamic acid at position 172 has an abnormally high pK$\rm\sb{a}$ of 6.8. The high pK$\rm\sb{a}$ value of Glu 172 is caused largely by electrostatic interactions of this residue with a proximal glutamic acid at position 78. Furthermore, the circular dichroism spectrum of the wild type xylanase shows a structural transition at pH 4.9 in the near-UV region which is absent for the E78C mutant. This taken with the fact that glutamic acid 78 forms a network of hydrogen bonds with tyrosine 69 and tryptophan 71 indicates that the transition with pK$\rm\sb{a}$ 4.9 should be attributed to glutamic acid 78. The presence of two proximal carboxyl groups, from glutamic acids 78 and 172, results in a pH-dependent destabilisation of the protein structure as evidenced by differential scanning calorimetry experiments, with the wild type xylanase and a number of mutant proteins.

Commercial poultry operations continuously test for the presence of avian leukosis virus (ALV) in their flocks. Tests currently in use for detecting ALV are ELISA (enzyme-linked immunosorbent assay)-based and detect either specific viral proteins or antibodies that are raised against ALV upon infection. Regions of the ALV genome have been identified which differentiate endogenous from exogenous ALV and these areas have been targeted with PCR primers. Using semi-nested PCR amplification of the LTR (long terminal repeats), we were able to detect all four subgroups of exogenous viruses affecting chickens (A, B, C and D). Two other sets of semi-nested primers targeted within the variable regions of the env (envelope) gene are able to determine the subgroup, A or B (the predominant subgroups infecting North American flocks), of the infecting virus. These sets of primers were first tested on chick embryo fibroblasts (CEF) not carrying any ev genes. The cells were infected with viral isolates of subgroups A (RAV-1), B (RAV-2), C (RAV-49) and D (RAV-50). The LTR primers detected all four subgroups, while the env primers detected only the virus of the subgroup for which they were designed. (Abstract shortened by UMI.)

v-rel is the transforming gene of the avian reticuloendotheliosis virus, strain T (REV-T), which causes fatal lymphomas in young birds and also transforms avian lymphoid cells in vitro. The v-rel oncogene codes for a protein of 503 amino acids. The v-rel protein exhibits extensive homology within its N-terminal 300 amino acids with several mammalian $\kappa$B enhancer binding proteins including 50-kDa and 65-kDa subunits of NF-$\kappa$B transcription factor. In this study, an array of biochemical functions of mutant v-rel proteins was investigated in order to understand the mechanism of transformation by v-rel oncogene. Since the v-rel protein shows striking homology with NF-$\kappa$B subunits, wild-type v-rel or deletion and linker insertion mutants of v-rel were expressed in insect cells to determine the $\kappa$B-specific DNA binding activity. C-terminal deletions of up to 100 amino acids, or deletion of 11 N-terminal amino acids had no effect on DNA binding activity of v-rel protein, whereas N-terminal deletion of 99 amino acids abolished DNA binding activity completely. All of the mutant v-rel proteins with transforming activity also retained the DNA binding activity like wild-type v-rel protein, whereas all of the mutant v-rel proteins which were nontransforming lost DNA binding activity. Therefore, it is clear that the $\kappa$B-specific DNA binding activity is necessary for transformation by v-rel. All of the non-DNA binding, non-transforming mutants of v-rel exhibited a transdominant negative effect. These results suggest that the formation of DNA binding complexes by v-rel protein with NF-$\kappa$B subunits is necessary for transformation. In order to determine the possible role of pp40 association in transformation, pp40 association with v-rel protein was examined. Association of pp40 with the DNA binding mutant v-rel proteins including wild-type v-rel protein resulted in inhibition of DNA binding activity. These results imply that the physical association with pp40 itself might not be sufficient for transformation by v-rel. Since the v-rel protein directly binds to the $\kappa$B site, the $\kappa$B-dependent transcriptional activity of v-rel protein was investigated. The non-DNA binding, non-transforming mutants function as transdominant inhibitors of the rel family of transcription factors. These results support a model in which activation of gene expression directly by v-rel protein is required for its ability to induce oncogenic transformation. (Abstract shortened by UMI.)

Defective interfering particles (DIPs) of vesicular stomatitis virus generated and/or amplified during serial high multiplicity passage in BHK-21 cells. Two IND-DIPs (IND-ST and IND-LT) were derived from the Indiana (Ind) serotype and two NJ-DIPs (Ogd-DI and Haz-DI) were generated from Ogden and Hazelhurst strains, respectively, of the New Jersey (NJ) serotype of VSV. The biological activity of each DIP was examined by the interference assay. Two IND-DIPs interfered homotypically and heterotypically. In contrast, two NJ-DIPs interfered homotypically but could not interfere heterotypically. Intracellular RNA synthesis in standard virus infected cells and in cells infected with both standard virus and DIPs revealed that the level of RNA synthesis of the standard virus was markedly reduced if DIP interfered with the production of the standard virus. In contrast, the level of mRNA transcription was not affected if DIPs did not interfere with replication of the standard virus. Furthermore, DIP genomes were replicated only if the DIP interfered with the replication of the standard virus. These data strongly suggest that the first step of DIP-mediated viral interference is at the level of viral RNA synthesis. In an effort to understand how these DIPs are generated, which genetic elements are involved in viral interference and how the DIPs interfere with the replication of standard virus, the complete sequence analyses of cloned cDNAs from the four different DIPs were carried out by the dideoxy chain termination method. The size of each DIP genome was 2,660 bases (IND-ST), 5,313 bases (IND-LT), 2,984 bases (Ogd-DI), and 6,110 bases (Haz-DI). The DIP genomes of IND-ST, Ogd-DI, and Haz-DI were derived from the 5$\sp\prime$ end of the standard virus genome and retained different lengths of inverted complementary termini (54 bp, 18 bp, and 71 bp, respectively), to form panhandle stem structures. In contrast, IND-LT retained the 3$\sp\prime$ half of the standard virus genome (4,974 bases) and 530 bases of the 5$\sp\prime$ end of the standard virus genome. This DIP genome represented a single deletion of a part of the L gene and thus contained the same 3$\sp\prime$ and 5$\sp\prime$ termini as the standard virus genome. Examination of the DIP genomes revealed that they all retained the 5$\sp\prime$ end of the standard virus. These findings suggest that the 5$\sp\prime$ terminus may be essential for control of DIP replication and interference activity. To understand the mechanism(s) of viral interference mediated by DIPs, a reverse genetic system for VSV was employed. All five viral genes from each of the three different strains of VSV were cloned and placed under the control of the T7 RNA polymerase promoter. Furthermore, cDNAs of four different DIPs were subcloned into a specialized transcription vector containing the T7 RNA polymerase promoter, hepatitis delta virus ribozyme, and the T7 terminator to generate DIP RNAs with the exact 3$\sp\prime$ and 5$\sp\prime$ termini identical to those of the authentic DIP genomic RNAs. The T7 RNA polymerase expressed by a recombinant vaccinia virus successfully rescued the cloned DIP genomes, which were amplified by subsequent passages. Furthermore, the rescued recombinant DIPs had the same biological activities as the original DIPs. Future experiments will help to identify cis-acting sequences in the DIP genomes and the roles of individual viral proteins involved in viral interference.

Cells of the obligate halophile Halobacterium cutirubrum were grown to the stationary phase in the presence of 32P-orthophosphate, 35S-sulfate, 14C-labelled acetate, malonate, mevalonate and glycerol and in the presence of 1,(3)-3H- and 1,3- 14C-glycerol mixture and 2-3H- and 1,3-14C-glycerol mixture as precursors. 32P was incorporated into the phosphatide components in proportion to their relative concentrations, thus: phosphatidyl glycerol phosphate (diether analogue) > unidentified minor phosphatide > phosphatidyl glycerol (diether analogue). 32S-sulfate was incorporated mostly into the glycolipid sulfate ester component, but small amounts of 35S also appeared in a more polar sulfolipid and in a minor unidentified sulfolipid. 14C-Mevalonate was the most efficient precursor for the phytanyl chains, followed in decreasing order by acetate and glycerol. Only traces of 14C from the malonate precursor were incorporated, and these appeared entirely in the phytanyl groups. A considerable proportion of 14C from the glycerol precursor was found in the glycerol and sugar moieties of the lipids. Little or no labelling of the water-soluble moieties of the lipids was observed with the other carbon precursors. Only traces of 14C-labelled fatty acids were detected with any of the 14C precursors; with acetate in particular, only 0.3% of the 14C in the total lipids was found in fatty acids. The synthesis of fatty acids was confirmed by the use of cell-free enzyme system and 14C-malonyl-CoA as a precursor. Although 0.5 - 0.6% of the 14C in the precursor was incorporated in fatty acids, it is believed that the fatty acid synthetase system is depressed by the high salt concentrations. It is concluded that the predominant bio-synthetic route for the lipid chains is the mevalonate pathway for synthesis of isoprenoid groups, and that the malonyl-CoA pathway for synthesis of fatty acids is operating at a very low level of activity. Studies on the metabolism of 14C and 32P-glycerol phosphate by whole cells and cell-free enzyme preparations showed that H. cutirubrum does not utilize alpha-GP as such, but it contains an active glycerol phosphatase which catalyzes the hydrolysis of alpha-GP into glycerol and Pi. The latter two components then pass into the cells where glycerol is phosphorylated by a glycerol kinase specifically to form sn-glycerol-3-phosphate. sn-Glycerol-3-phosphate was formed not only by glycerol phosphorylation, but also by reduction of dihydroxyacetone phosphate by L,alpha-glycerol phosphate dehydrogenase. Using tritiated glycerol, the incorporation of 3H into lipids of H. cutirubrum cells varied with the position of the 3H label in the precursors supplied. Total lipids and lipid moieties from cells grown in 2-3H-glycerol lost almost completely the 3H activity. In contrast, when cells were grown in 1(3)- 3H-glycerol, the phytanyl groups and the sugar residues of the lipids retained about 50% of the 3H activity while the glycerol moiety of the diphytanyl glycerol ether retained about 100% of the 3H activity. It is concluded that glycerol is metabolized via three separate pathways: (1) by the reverse of the Embden-Meyerhof cycle, to sugars; (2) by glycolysis, to acetate and then to isopentenyl pyrophosphate through the mevalonate pathway; (3) by a new pathway, in which glycerol is converted to diphytanyl glycerol ether by an as yet unknown series of reactions, probably, involving dihydroxyacetone. Direct studies on the biosynthesis of phosphatides of H. cutirubrum by short-term labelling of total lipids with 32P i was undertaken. During the first minute of labelling, an unidentified phospholipid was initiated which bears a precursor-product relationship between it and PGP-diether analogue. It is concluded that the unidentified phospholipid may be the immediate phosphorus-containing precursor of the major phosphatide PGP-diether analogue.

A temperature-dependant rabbit model of Haemophilus ducreyi infection was used as a quantitative virulence assay to evaluate the effect of host factors upon ulcerative cutaneous disease production. New Zealand White rabbits underwent inoculation with H. ducreyi strain #35000 either after iron loading, dexamethasone immunosuppression, prior infection or immunization. Chancroid-like disease was scored for severity, time, culture of lesions, and serologic response. In primary infections, culture-positive ulcerative lesions were consistently produced at and above 105 colony forming units inocula. Iron and dexamethasone treatment increased lesion severity and duration of culture positivity. Infection of previously infected animals produced sterile lesions of greater size and higher cumulative disease score at 105 colony forming units inocula. Infection of immunized animals produced sterile lesions of lesser severity at 105 colony forming units, with protection from ulcer formation. Efficacy of ceftriaxone treatment was tested in naive control and iron loaded rabbits. Antibiotic was injected as a single intramuscular dose of 0.1 mg/Kg and 5 mg/Kg, four days following inoculation with Haemophilus ducreyi #35000. In naive control rabbits, antibiotic treatment at each dose sterilized the lesions within 24 hours with attenuation of disease effect. In naive iron loaded rabbits lesions were sterilized later on day 10 with 0.1 mg/Kg ceftriaxone and on day 5 with 5 mg/Kg ceftriaxone. Virulence scores corroborated the lessened microbiologic efficacy of antibiotic treatment. We conclude that quantitative assay of infection and disease in iron loaded animals, with relative prolongation of disease effect and culture positivity of lesions may be a sensitive model in which to comparatively measure virulence related to bacterial factors. As well, limitations of efficacy or synergy of antibiotic treatments may be evaluated. We conclude that prior antigenic exposure and immune response through infection or immunization attenuates subsequent homologous infection. Disease produced in re-infection is amplified, while disease produced in inoculation after immunization is attenuated at lower inocula. This suggests that while there is inducible immunity to H. ducreyi infection and disease, the disease effect may be in part related to host inflammatory response to bacterial antigen or effect of bacterial toxin. This system of measurement of virulence may be useful towards understanding pathogenesis, and identifying strategy for vaccine development in chancroid.

This investigation had two objectives: (a) to determine the relative susceptibility of several animal species to L. pomona infection and (b) to study the immunological reactivity of guinea pigs and calves to killed suspensions of the organism and fractions thereof. An examination was also made into some of the methods presently used to detect infection or immunity. The relative susceptibility of several animal species to L. pomona infection was compared with that of guinea pigs. Therefore it was first necessary to assay the infectivity of the organism in this species. Bacterial counts were made of young leptospiral culture which was then diluted to provide appropriate dose levels. Sixty guinea pigs in groups of four animals each were injected with infective material ranging down from 100,000 to 13 organisms per inoculum. The level of 50 leptospirae per inoculum was selected as the guinea pig infective dose since it was the lowest in which all animals in the group became infected and showed a regular response. The larger doses of leptospirae did not produce more severe disease as judged by the temperature response. In fact, disease appeared to be less severe at the higher dose but occurred sooner after exposure. Ultimate agglutination-lysis titres not affected by the number of organisms in the infective dose. (Abstract shortened by UMI.)