published_or_final_version / Dentistry / Doctoral / Doctor of Philosophy

Από 3 πρότυπα στελέχη Pseudomonas aeruginosa το Smooth στέλεχος PAC-IR και τα Rough στελέχη PAC-557 και PAC-605, που δώρησε ευγενώς στο Εργαστήριο Μικροβιολογίας του Πανεπιστημίου Πατρών η Dr. ΡΜ Meadow του University Co 1 lege, London παρελήφθησαν με ειδική μεθοδολογία ο λιποπολυσακχαρίτης L.P.S, η εξωκυττάρια ουσία (Slime) και οι εξωτερικές μεμβράνες. Έγιναν ηλεκτροφορήσεις με αποδιατακτικούς παράγοντες (SDS-PAGE) και χρώση των πηκτωμάτων για πρωτεΐνες με Coomassie blue, και για L.P.S με AgNÖ3. Οι ηλεκτροφορητικές εικόνες του Slime, του L.P.S και των εξωτερικών μεμβρανών διαφέρουν μεταξύ τους και στα 3 στελέχη. Ανάλυση των ουδετέρων σακχάρων έδειξε ότι και τα 3 Slime περί έχουν με ποσοτικές διαφορές 6 σάκχαρα, τη ραμνόζη, τη φουκόζη, τη ξυλόζη, τη μαννόζη και τη γαλακτόζη. Αντίθετα μαννόζη και γαλακτόζη δεν περιέχονται στο Smooth L.P.S, ενώ οι Rough L.P.Ss περί έχουν μόνο ραμνόζη και γλυκόζη. Ανάλυση του λιπιδικού στοιχείου των 3 Slime και των 3 L.P.Ss έδειξε ότι ποιοτικά περιέχουν 5 ίδια λιπαρά οξέα. Δωδεκανοικό οξύ (~Cii), 20Η-δωδεκανοικό (-Ci?), δεκατετρανικό οξύ (~Ci 4), δεκαεξανικό οξύ (~Cit), δεκαοκτανικό οξύ (-Ci s) και τα ισομερή του. Μετά από χρωματογραφία μοριακής διήθησης στα 3 Slime ο πολυσακχαρίτης που παραλαμβάνεται από την υδατανθρακική κορυφή δεν περιέχει L πρωτεΐνη και είναι μοριακού βάρους 40.000-100.000 daltons. Τα αποτελέσματα αυτά αποδεικνύουν ότι η παραγωγή του Slime είναι κοινή ιδιότητα Smooth και Rough στελεχών Ρ.aeruginosa και ότι το υδατανθρακικό στοιχείο του Slime είναι αυτοτελής ουσία, ελεύθερη πρωτεϊνών, που δεν αποτελείται από τις πλευρικές αλυσίδες του L.P.S. / Lipopolysaccharide (L.P.S), slime and outer membranes were obtained from a smooth, nonmucoid Pseudomonas aeruginosa strain, PAC-IR and its two rough mutants, PAC~ 557 and PAC-605 (Kindly provided by Dr. PM Meadow, University College, London). Electrophoretic analysis on SDS-polyacrylamide gels and staining with silver nitrate and Coomassie blue showed different profiles between L.P.S, Slime and outer membranes in all three strains. Comparative analysis of the saccharide moiety between L.P.S and Slime of each strain by H.P.L.C demonstrated that the saccharide moiety of slime has different composition from that of L.P.S. Six neutral sugars, rhamnose, fucose, xylose, mannose, galactose and glucose were identified in all three slimes though in different amounts. Mannose and galactose were not found in the smooth L.P.S, whereas only rhamnose and glucose were identified in the L.P.Ss of the rough strains. Comparative analysis of the lipid moiety between L.P.S and Slime in all three strains with Gas Chromatography and Mass Spectroscopy indicated that lipid moiety had no quaiitative differences concerning the lipid acids. Five lipid acids were identified in all three slimes and L.P.Ss dodecanoic acid (~Ci2)> 20H~dodecanoic acid (-C2?), tetradecanoic (-C14), exadecanoic (Ci(,), octadecanoic (- Cis) and its isomeric forms. After gel filtration of all three slimes the polysaccharide obtained from the carbohydrate peak fraction was found to be protein free. By gel filtration the mo 1 ecular size of the polysaccharide was estimated to be about 40000-100000 daltons. Whether the polysaccharide moiety is a glycolipid is not clear at the present. This problem is currently under investigation. Our results suggest that slime production is a common property shared by both Smooth and Hough Ρ.aeruginosa strains. The saccharide moiety of slime does not represent the side chains of L.P.S and it is protein free.

Cystic Fibrosis is one of the most common genetic recessive diseases among Caucasians and is caused by mutations in the Cystic Fibrosis Transmembrane conductance Regulator (CFTR) gene on chromosome 7. There are different classes of CFTR mutation, leading to differences in disease severity among patients. In addition to the CFTR genotype, secondary genetic factors, modifier genes, also influence CF phenotypes. Due to the dysfunction of CFTR protein and production of thickened mucus, bacterial infection in the lungs is favored and can lead to further clinical complications in CF patients. Pseudomonas aeruginosa is one of the most common bacteria detected among patients. The aim of this project was to investigate four candidate modifier genes, Factor B, Complement Factor 3, Toll-like Receptor 4 and Heme oxygenase-1, which might affect the status of Pseudomonas aeruginosa infection. A total of 22 single nucleotide polymorphisms (SNPs) were selected in these four genes and they were tested against five phenotypic traits, including age of diagnosis, FEV1% predicted andstandard deviation value, age of first Pseudomonas aeruginosa infection and Pseudomonas aeruginosa infection status. Among the selected SNPs, both case-control studies and family-based analysis were performed in order to establish any correlation between the genotypes and the phenotypes. In addition, haplotype analysis was performed to determine whether there was interaction between SNPs or whether there were unidentified SNPs in the vicinity of the selected ones that might contribute to the observed phenotypic traits. Among the 22 chosen SNPs, 13 of them were found to be significantly linked to one or more of the tested phenotypes. The three most significant associations were BF_2557 with lung function, HMOX1_9531 with lung function and BF_7202 with age of diagnosis. Several haplotypes were significantly associated with one of the five phenotypes. There was no evidence for the presence of unidentified SNPs or interaction between SNPs. Most of haplotype associations were likely due to the presence of a single SNP which was found to be significantly linked to the phenotype. Conclusively, both SNPs and haplotype analyses suggest that the four candidate genes are modifiers of disease severity in CF.

This thesis is an investigation of Pseudomonas aeruginosa bacterial biofilms. The objective of the first study was to evaluate the biofilm-forming capacity of canine otitis isolates of P. aeruginosa and to compare the minimum inhibitory concentrations (MICs) of antimicrobials for planktonic versus biofilm-embedded bacteria. Biofilm forming ability was assessed using a microtitre plate assay. Broth microdilution was used to assess the MICs of neomycin, polymyxin B, enrofloxacin and gentamicin for the planktonic and biofilm-embedded bacteria of eighty-three isolates. Thirty-three (40%) isolates were biofilm producers and MICs for biofilm-embedded bacteria were significantly higher than their planktonic counterparts for all antimicrobials (all P<0.05). The objective of the second study was to evaluate the impact of Tromethamine edetate disodium dihydrate (Triz-EDTA®) in combination with antimicrobials on antimicrobial susceptibility of P. aeruginosa biofilm-embedded bacteria. MICs of the four antimicrobials for the biofilm embedded bacteria and biofilm-embedded bacteria with added Triz-EDTA® were assessed with broth microdilution for thirty-one biofilm-producing isolates. Addition of Triz-EDTA® significantly reduced MICs for neomycin (P < 0.008) and gentamicin (P < 0.04) but not enrofloxacin (P = 0.7), or polymyxin B (P = 0.5). The objective of the third study was to determine the presence of biofilm-associated genes in biofilm forming and non-biofilm forming isolates. Four genes involved with carbohydrate matrix production (pelA), irreversible attachment (sadB) and quorum sensing (lasB, rhlA) were selected. DNA was extracted and polymerase chain reaction (PCR) was performed for all isolates. All isolates possessed lasB and sadB, 74 (90%) possessed pelA and 74 (90%) possessed rhlA. All thirty-two (100%) isolates that were classified as biofilm producers contained all genes. There was an association between the presence of pelA and rhlA and biofilm production (P < 0.017) and between the presence of rhlA and pelA and the quantity of biofilm produced (both P < 0.001). These results highlight that biofilm formation of Pseudomonas aeruginosa otic isolates does occur and can impact antimicrobial therapy. Certain compounds can also influence antimicrobial susceptibility of biofilm-embedded bacteria. Genetics may also play a role in biofilm formation.

Expression of the mexCD-oprJ multidrug efflux operon is enhanced by the presence of membrane damaging agents [e.g., the biocide chlorhexidine (Chx)] or mutations in the nfxB gene encoding a repressor of efflux gene expression, both dependent on the AlgU envelope stress response sigma factor. Details of mexCD-oprJ regulation are, however, lacking. In examining the mexCD-oprJ locus, a gene, PA4596, encoding a homologue of NfxB (61% identity) was identified downstream of oprJ, a location conserved in all sequenced Pseudomonas aeruginosa isolates and in Pseudomonas putida. Opposite to mexCD-oprJ, PA4596 expression was reduced by Chx exposure, as assessed using RT-PCR; although like mexCD-oprJ, this was AlgU-dependent (i.e., lost in a ΔalgU strain). Deletion of PA4596 had no impact on Chx resistance indicating that it is not required for Chx-inducible mexCD-oprJ expression/ MexCD-OprJ-dependent Chx resistance. In contrast, mexCD-oprJ expression and the attendant multidrug resistance of nfxB deletion mutants were compromised upon deletion of PA4596, indicating that PA4596 plays a positive role in mexCD-oprJ expression in such mutants. Consistent with this, PA4596 expression increased in nfxB deletion and missense mutants in parallel with mexCD-oprJ. Intriguingly, mexCD-oprJ expression and multidrug resistance were observed in a mutant lacking an nfxB mutation (demonstrating an NfxB-like phenotype) and in an nfxB missense mutant and these were not compromised upon deletion of PA4596. Thus, mexCD-oprJ hyperexpression can be both PA4596-dependent and -independent. A bacterial 2-hybrid assay revealed a PA4596-PA4596 interaction, consistent with the protein forming dimers as NfxB has been shown to do. Two-hybrid assays also demonstrated that NfxB and PA4596 interact. While the functional significance of this remains to be elucidated, it is consistent with their common role in regulating mexCD-oprJ expression and is suggestive of a complex and possibly novel regulatory mechanism. These data highlight the complexity of mexCD-oprJ regulation and the apparently multiple pathways to efflux gene expression, suggestive of multiple roles for this efflux system in P. aeruginosa independent of antimicrobial efflux. / Thesis (Master, Microbiology & Immunology) -- Queen's University, 2009-08-18 14:25:18.107

Pseudomonas aeruginosa is a common soil organism as well as an opportunistic human pathogen. Treatment of P. aeruginosa infections is often complicated by innate resistance to a variety of antimicrobials mediated by multidrug efflux systems. The MexAB-OprM efflux system is constitutively expressed in wildtype strains and contributes to innate antimicrobial resistance, while hyperexpression of the system results in acquired high levels of resistance. MexAB-OprM is hyperexpressed in nalC mutants containing mutations in the gene encoding NalC, a repressor of a two-gene operon, PA3720-armR. armR encodes a protein modulator of MexR, a repressor of mexAB-oprM expression. Previous reports showed that genes encoding the MexAB-OprM efflux system are upregulated in response to pentachlorophenol (PCP), a phenolic compound that is a common environmental contaminant. Induction of mexAB-oprM and PA3720-armR by PCP was confirmed using RT-PCR, and MexAB-OprM was shown to be involved in PCP resistance. An electromobility shift assay (EMSA) showed that PCP interacts with NalC, interfering with its binding to the PA3720-armR promoter region and thereby promoting PA3720-armR expression. Nonetheless, the increase in ArmR did not drive mexAB-oprM expression suggesting that PCP induction of this efflux operon occurred via a different mechanism. A direct PCP-MexR interaction could not be demonstrated using an EMSA. PCP exposure did, however, reduce expression of nalD, encoding a second repressor of mexAB-oprM, which might explain the PCP-promoted increase in mexAB-oprM expression. PCP is unlikely to be the intended inducer(s)/substrate(s) for this system but probably resembles these. Several compounds related to PCP were tested for an interaction with NalC but all were negative in EMSAs. Plants produce a variety of phenolic compounds, which are often antimicrobial and, so, root extracts of various plants were tested for an ability to interact with NalC and interfere with promoter binding. Extracts from Boehmeria tricuspis, Uncaria tomentosa and Ixiolirion tataricum were shown to interact with NalC, suggesting that plant compounds may be the intended inducers/substrates for NalC/MexAB-OprM. / Thesis (Master, Microbiology & Immunology) -- Queen's University, 2010-09-10 10:35:16.271

Pseudomonas aeruginosa is resistant to many clinically-relevant antibiotics and this is partly attributable to multidrug efflux systems in this organism. One of these, MexEF-OprN, exports chloramphenicol, fluoroquinolones and trimethoprim and is positively regulated by MexT, encoded by the mexT gene that is located upstream of the efflux genes. MexT also positively regulates the mexS gene that encodes an oxidoreductase of unknown function and whose loss increases expression of mexEF-oprN. A null mutation in the mexS gene of P. aeruginosa increased expression of two 3-gene operons, PA4354-PA4355-PA4356 and PA2813-2812-2811. This increased expression, which paralleled an increase in mexEF-oprN expression in the same mutant, was, like mexEF-oprN, MexT-dependent. The PA4356 (xenB) gene product is homologous to a xenobiotic reductase in P. fluorescens shown to remove nitro groups from trinitrotoluene and nitroglycerin. As nitration is a well-known result of nitrosative stress, a role for xenB (and the co-regulated mexEF-oprN and PA2813-2812-2811) in a nitrosative stress response was hypothesized and tested. Using S-nitrosoglutathione (GSNO) as a source of nitrosative stress, expression of mexEF-oprN, xenB and PA2811 was shown to be GSNO-inducible, though in the case of xenB and PA2811 this was only seen in a mutant lacking MexEF-OprN and only for xenB, this GSNO-inducible expression was dependent upon MexT. Consistent with MexT being required for mexEF-oprN and PA4354-PA4355-PA4356 expression, MexT bound to their promoter regions. Chloramphenicol, a nitroaromatic antimicrobial that is a substrate for MexEF-OprN and resembles a nitrosated nitrosative stress product accommodated by this efflux system induced expression of mexEF-oprN, but not xenB, or PA2811, and this was dependent upon the MexT regulator. In addition to MexT positive-regulating activity of genes in response to nitrosative stress, MexT also negatively regulates the expression of mexAB-oprM through direct binding to its promoter region and the oprD gene, encoding an outer membrane porin that provides a portal of entry for basic amino acids and carbapenem, whose expression was reduced only by GSNO. / Thesis (Ph.D, Microbiology & Immunology) -- Queen's University, 2011-05-15 20:59:57.018

The characteristics and expression of laboratory derived chloramphenicol (CM) resistance in P. aeruginosa were examined. Resistant strains exhibiting single cell resistance of 1.5 to 2 mg/mL were readily isolated following one passage in CM at 150 to 1000 (mu)g/mL. Isogenic strains, selected on CM at 150 and 500 (mu)g/mL were chosen for detailed study. Resistance was not a consequence of drug detoxification or altered sensitivity of the target site. The resistant strains exhibited unusual phenotypic properties including pronounced variations in growth rate, CM susceptibility and cell morphology as a function of the composition of the growth medium. Growth in CM also resulted in significant alterations in amino acid transport and respiratory capacity, the extent of which varied with the strain, the growth medium and the concentration of CM. These drug and medium-dependent alterations were determined to reside in an increased and highly specific requirement for Ca('2+), Mg('2+), Mn('2+) or Sr('2+). Manipulation of the divalent cation concentration of a variety of growth media resulted in dramatic alterations in growth rate, resistance and amino acid transport. Ca('2+) was significantly more effective than the latter three ions. The expression of native and plasmid-mediated CM resistance was also modified by the external concentration of divalent cations. In view of the nature and specificity of the cation requirement, it was concluded that (1) divalent cation-mediated alterations of outer membrane permeability are fundamental to the expression of native and acquired CM resistance in P. aeruginosa; (2) laboratory-derived CM resistance involves envelope changes, such that interaction with divalent cations promotes more effective exclusion of CM. The latter conclusion is supported by other divalent cation-dependent alterations in envelope function in the resistant strains.

The characteristics of the effects of chloramphenicol on Pseudomonas aeruginosa were examined. Resistant strains were easily isolated following a single passage in chloramphenicol at 150 $ mu$g/ml to 500 $ mu$g/ml. Drug detoxification or altered sensitivity of the target site could not be the mechanism of resistance. This resistance to chloramphenicol was correlated with the addition of an outer membrane protein with a molecular weight of 49 kDa and the loss of two outer membrane proteins, one with the molecular weight of 19 kDa and the other of about 10 kDa. The highly specific requirement of the resistant strains for Ca$ sp{2+}$, Mg$ sp{2+}$, Mn$ sp{2+}$ or Sr$ sp{2+}$ described by Irvin and Ingram (1982) was confirmed by the observation that the outer membrane of the resistant cells contained twice as much Mg$ sp{2+}$ cation as the sensitive cells. Many other experiments designed to observe the effects of chloramphenicol on the outer membrane of P. aeruginosa failed. It was concluded that the observations made in this study strongly suggested a "re-structuring" of the outer membrane of P. aeruginosa, rendering the resistant cells more impermeable to chloramphenicol.

<p>As an opportunistic Gram-negative pathogen, Pseudomonas aeruginosa must be able to adapt to changes and survive stressors in its environment during the course of infection. To aid survival in the hostile host environment, P. aeruginosa has evolved a myriad of virulence factors including the production of an exopolysaccharide capsule, as well as secretion of degradative proteases and lipases that also function as defense mechanisms. Outer membrane vesicles (OMVs) acts as a secretion system to disseminate virulence factors and function as a general bacterial stress response to remove accumulated periplasmic waste. Despite the growing insights of the field into the potential functions of OMVs, the mechanism for formation remains to be fully elucidated. The three proposed mechanisms for OMV formation in P. aeruginosa are mediated by the Pseudomonas quinolone signal PQS, the AlgU envelope stress response pathway, and the periplasmic chaperone MucD. This report investigates how P. aeruginosa responds to sublethal physiological stressors with regards to OMV production levels and whether the proposed mechanisms for OMV formation are required for stress-induced OMV formation. We concluded that exposure to cell wall directed stressors increased OMV production and activity of the sigma factor that controls MucD expression, AlgU. AlgU was shown to be sufficient to induced OMV production upon overexpression; however, stress-induced OMV production was not dependent on activation of AlgU as vesiculation could be induced in strains lacking AlgU. Furthermore, MucD levels were not inversely proportional to OMV production under acute stress, and the ability to produce PQS was not required for OMV production. Finally, an investigation of the response of P. aeruginosa to oxidative stress revealed that hydrogen peroxide-induced OMV production requires the presence of B-band but not A-band lipopolysaccharide. We also demonstrated that the ability for P. aeruginosa to sense oxidative stress via OxyR, was important for hydrogen peroxide-induced OMV production, by a yet to be determined method. Together these results demonstrate that current proposed mechanisms for OMV formation do not universally apply under all stress conditions, and that additional mechanisms for OMV formation are still to be identified and fully elucidated during acute stress in P. aeruginosa.</p> / Dissertation