The Study On Persistence Mechanisms Of Uropathogenic Escherichia Coli

Two major problems have arisen in the field of antimicrobial therapy. One is the emergence of heritable antibiotic resistance and their spread by horizontal gene transfer. With the growing consumption burden and the irrational use or even abuse of antimicrobials, the increasingly serious problem of antimicrobial resistance has finally caught the global attention. Some pathogens have become resistant to most of the commonly used antibiotics. Therefore, when the infectious diseases caused by these "superbugs" occur, the patients and doctors are inevitably trapped into "no drug available" dilemma. On the other hand, the persistence of bacterial infections, in which the pathogens linger in the host for long periods of time in spite of prolonged treatment, has made another big deal to the antimicrobial therapy. Most of the persistent pathogens are susceptible to antibiotics in the in-vitro test of drug susceptibility, yet leaving the difficult problem of "no drug effective" in vivo. Some well-known examples are tuberculosis, lung infections in patients with cystic fibrosis, typhoid fever and urinary tract infections.Apart from the formation of biofilm and the virulence factors, such as adhesins as the most important ones, the persistence property of persister makes the great contribution to the bacterial long-term stay with the host, recalcitrant to therapeutic agents and host defense mechanisms. The persisters are a small fraction, of the order of10"6or less, of genetically identical bacterial cells that can survive the exposure to lethal concentrations of bactericidal antibiotics and regrow after removal of the antibiotics, with the descendants as sensitive to the same antibiotics as the original population, again leaving behind a tiny fraction of survivors.Urinary tract infections (UTIs), including cystitis and pyelonephritis, are the most common bacterial infections in women, which account for significant morbidity and substantial medical costs on individual, family and society. In terms of urinary anatomical differences between the genders, UTIs primarily affect women. It is estimated that the overall lifetime prevalence of UTIs in women is greater than60%. Further recurrent UTIs are reported in about25%of women within6months of an acute UTI episode and pose a major problem.Uropathogenic Escherichia coli (UPEC) is the causative pathogen in70-95%community acquired UTIs and over50%nosocomial UTIs. Many virulence factors made great contributions to the pathogenesis of UPEC. And the presence of adhesins is arguably the most important determinant of the pathogenicity for UPEC. Adhesive organelles, including type1, P and S/F1C pili along with afa/Dr adhesins, mediate bacterial attachment to host tissues within the urinary tract. E. coli expressing Dr/Afa adhesins may predispose the establishment of chronic or recurrent infections. Adhesin PapG is thought to be associated with bacterial adhesion to the kidney, so as to be related to the acute pyelonephoritis. Adhesin Ag43, encoded by flu gene, enables bacterial cells to adhere to each other and form biofilm over the host urothelial cells, leading to a hard clearance of the pathogens.Persisters also play an important role in the hard clearance of the UPEC. The mechanisms of persistence and persister generation are still to be understood. As pointed out early in1944, the suggestion that persisters are in a state of dormancy, which enables them to tolerate antibiotic exposure, has not changed with time. In2007, Zhang and his colleges identified in E. coli a new persistence gene phoU whose inactivation caused a metabolically hyperactive status of the cells with an increased expression of energy production genes, but a defect in persister formation when exposed to a diverse range of antibiotics and different kinds of stresses. In this study, we compared the phoU gene sequences of all the known E. coli strains published in the Genebank (http://www.ncbi.nlm.nih.gov/gene), and we identified an specific difference in the236th amino acid among some UPEC strains and the wild type W3110strains. In the W3110E. coli strain, there was glycine (G) in its236th amino acid locus, while glutamate(E) shown in the same locus within the noticed UPEC strains such as CFT073, UTI89,536and APEC01strains. We wondered if this amino acid mutation in the phoU gene of the UPEC strains evoked a variant role of PhoU played in the persistence of UPEC.In the present study, a total of47non-duplicate clinical isolates of UPEC were collected from clean midstream urine specimens of inpatients with urinary tract infections at Huashan Hospital, Shanghai from January2008to December2009, and26intestinal commensal E. coli strains, as control, were isolated from the anal swab specimens of the health volunteers. The objectives are to understand the UPEC drug suscepibility and adhesin genes pattern, and to investigate phoU related persistence mechanism of the clinical UPEC strains. The study included the following three parts.Part One Antimicrobial susceptibility testing and homology analysis of UPEC and intestinal commensal strains The47clinical UPEC isolates were collected according to standard laboratory protocols and isolated from urine with>105colony-forming units/ml. Together with the26intestinal commensal E. coli strains as control, the minimum inhibit concentration of15kinds of antimicrobial drugs to all of the73E.coli strains was detected using the agar dilution method according to the2010updated version of the CLSI (Clinical and Laboratory Standard Institute) files. The15antimicrobials included piperacillin, cefazolin, cefuroxime, cefotaxime, ceftazidime, cefepime, piperacillin/tazobactam, cefoperazone/sulbactam, imipenem, meropenem, gentamycin, amikacin, ciprofloxacin, levofloxacin and nitrofurantoin. E. coli ATCC25922was used as quality control strain.UPEC isolates demonstrated a remarkably lower susceptibility rates to most of the tested antimicrobials than the intestinal commensal strains. Especially, the sensitivity rate of UPEC to cefotaxime was only32%while89%of the commensal counterparts were susceptible to cefotaxime. Less than one quarter of the UPEC were sensitive to the fluoroquinolones (including ciprofloxacin and levofloxacin), yet the susceptibility rate of intestinal isolates were96%. Fortunately, there were still a relatively high susceptibility rate of UPEC to piperacillin/tazobactam, imipenem, meropenem, amikacin and nitrofurantoin (89%-100%).We carried out disk diffusion ESBL confirmatory test to the isolates with a MIC to cefotaxime≥4μg/ml, or a MIC to cefotazidime>16μg/ml according to the2009version of the CLSI files. We found28ESBLs-producing strains(60%) among the47UPEC stains, while1(4%) out of26commensal isolates was also proved ESBLs-producing. All the ESBLs-producing strains were cefotaxime resistant, and50%were resistant to cefotazidime. CTX-M-1,2,8,9,25/26group specific genes of28ESBLs-producing UPEC were determined by PCR with the primers reported previously. BlaCTX-M-1type genes were detected in14(50%) isolates.12(43%) isolates harbored BlaCTX-M-9type genes. The cefotazidime resistance rate (86%) of UPEC isolates carrying blaCTX-M-1type genes were significantly higher than that (17%) of the isolates with blaCTX-M-9type genes.The7housekeeping genes (adk、fumC、gyrB、icd、mdh、pur A and recA) for multilocus sequence typing (MLST) were selected from the E. coli database at the University College Cork E. coli MLST website (http://mlst.ucc.ie/mlst/dbs/Ecoli) and sequence type determination of the47UPEC and26commensal isolates was performed using the MLST PCR scheme accordingly. The based upon related sequence types (BURST) clustering algorithm (eburst.mlst.net) was used to analyze the allelic profiles and define clonal complexes.5UPEC and4commensal E. coli isolates shared the same type ST10, and one isolate respectively from the two of the groups belonged to ST453. In addition to that, the MLST of UPEC was thoroughly different from that of the intestinal commensal counterparts. The47UPEC isolates analyzed were assigned to18distinct STs. The most common STs were ST131(n=9,19.1%of isolates), followed by ST405(7,14.9%), ST10(5,10.6%), ST393(4,8.5%), ST69(4,8.5%), and ST95(3,6.4%). The six STs accounted for68.1%(32/47) of the isolates, demonstrating the diversity of lineages. The26isolates analyzed by MLST were grouped into22different STs, more diverse than the UPEC isolates. Of the22STs,18were already included in the E. coli website database, Among which ST10(n=4,15.4%) was the most popular group. And an unknown ST included2isolates, while the other20isolates were classified into the20STs, respectively.Part two. Comparison of adhesin genes between uropathogenic and intestinal commensal Escherichia coli strainsAmong the47UPEC isolates,21were from patients with recurrent lower UTIs. The recurrent lower UTIs were defined as at least two episodes of UTI within6month. And each time the patients had:a) typical clinical symptoms as urinary urgency and pain and frequent voiding, b) pyuria, c) without fever, chill or other systematic symptoms. Seven strains were isolated from acute pyelonephritis patients by the following criteria:a) the temperature>38.5℃, b) positive percussion pain on kidney area, c) an elevated leukocytes count and percentage of neutrophils in the blood routine test. The rest19strains were from patients with acute uncomplicated cystitis.The presence of adhesin genes afa, draE, daaE, papGI, papGⅡ, papGⅢ, flu, fimH, sfa and foc were investigated by PCR amplification using primers sets as described previously.①afa gene was present in6UPEC isolates, all of which belonged to the recurrent lower UTIs group (28.6%). None of the isolates from the acute pyelonephritis group or the acute uncomplicated cystitis group carried afa gene. And no afa gene was found in26intestinal commensal E. coli isolates. Neither draE nor daaE gene was detected among all the strains we screened.(2)11of the47UPEC isolates were papG gene positive (23.4%) including10papG Ⅱ and1papG Ⅲ, and1of the intestinal commensal isolates (3.8%) carried papG Ⅱ gene. Further more, the prevalence of papG in acute pyelonephritis (71.4%) was significantly higher than that of recurrent lower UTI group (14.3%) and that of acute uncomplicated cystitis group (15.8%).(3)80.9%of the UPEC isolates while50.0%of the intestinal commensal isolates were found biofilm related flu gene positive. However, the prevalence difference of the three UTI groups was not significant.(4)Gene fimH was screened positive in42of47(89%) UPEC isolates and20of26(76.9%) commensal isolates.⑤Sfa gene was present in one isolate each in the three UTI groups and one of the sfa positive isolates was also detected focG positive. But none isolate in the intestinal commensal group was screened sfa/foc positive.Most of the UPEC strains (72.3%,34/47) carried2to3kinds of adhesin genes, while80.8%(21/26) of the commensal isolates had1to2kinds of adhesin genes. The difference of the adhesin gene numbers between UPEC and commensal isolates were statistically significant.The ESBLs-producing UPEC carried fewer adhesin genes than the non-ESBLs-producing ones, of28ESBLs-producing UPEC,2(14%) were detected no adhesion genes,6(21%) carried1adhesin gene, and15(54%) were screened2kinds of adhesin genes. On the other hand, among the19non-ESBLs-producing UPEC isolates,1(5%) carried only one adhesin gene, and8(42%) and9(47%) isolates were respectively detected2and3kinds of adhesin genes, with the left carried5kinds of the detected genes. Therefore, most of the ESBLs-producing UPEC isolates carried no more than2adhesin genes, while over half of the non-ESBLs-producing UPEC were screened no less than3kinds.Part Three, phoU gene related persistence mechanism of the clinical UPEC isolatesWith the help of PCR and gene sequencing, we amplified and compared the complete phoU gene sequence of the47clinical UPEC isolates and26intestinal commensal E. coli strains.10of47(21%) UPEC was feathered with the same wild type phoU gene sequence as the strain W3110. The phoU gene sequences of79% (37/47) UPEC had amino acid differences from the wild type, so as to be called mutant phoU gene types in this study.20of the37mutant phoU UPEC carried the phoU G236E mutation and the rest17feathered with the same phoU S239P+D240K+K24JE mutation. Interestingly, all of the26intestinal commensal E. coli were phoU gene wild type.The MLSTs of the phoU wild type UPEC isolates were totally different from those of the phoU mutant ones.9ST131UPEC were all phoU G236mutant strains,7ST405and4ST393UPEC belonged to the phoU S239P+D240K+K241E mutation type, while the5ST10UPEC all carried wild type phoU gene. The UPEC isolates with papG gene positive were all included in the mutant phoU gene group. Furthermore, the UPEC isolates with mutant phoU gene carried more adhesin genes than those with wild type phoU gene. However, there were no significant differences between the wild type phoU UPEC and the mutant type phoU UPEC, whether referring to the urinary tract infection types or the antimicrobial susceptibilities.So as to study whether the phoU G236E mutation provoked a variant role of PhoU played in the persistence of UPEC, we performed the in vitro bactericidal test with phoU G236E mutant UPEC and phoU wild type UPEC. We chose3kinds of antimicrobials:cefotazidime, amikacin and ciprofloxacin. In the killing curve experiment, the UPEC with phoU G236E mutation showed increased ability of persistence to cefotazidime, amikacin and ciprofloxacin. Under the relatively low dose of the three antibiotics (2,4×MIC cefotazidime,4×MIC amikacin and8×MIC ciprofloxacin), the UPEC strain of wild type phoU gene was completely killed after20-30h, while3.5-7×101CFU/ml phoU G236E mutant isolate survived at the same time and remained no less than1×101CFU/ml until the56h after the administration of the antimicrobials. When shifted to the high concentration level of drugs (8×MIC cefotazidime,8×MIC amikacin and16×MIC ciprofloxacin), the colony counts of phoU gene wild type isolate sloped to zero as early as4h (20h for8×MIC cefotazidime), but yet the phoU G236E mutant isolate were not killed until30-44h exposure to the antimicrobials. Therefore, it was probable that the phoU G236E mutant isolates left more persisters and then more tolerant to antibiotics.In order to understand how the phoU G236E mutation might be involved in the enforcement in persister formation of the UPEC as shown by increased tolerance to various antibiotics, we performed a relatively quantitative PCR of phoU gene under the treatment of cefotazidime, amikacin and ciprofloxacin respectively (we chose the housekeeping gene mdh as the endogenous control). The result indicated that after12h exposure to cefotazidime, amikacin and ciprofloxacin, the relative phoU gene expression level of phoU G236E mutant UPEC isolate was upregulated6.2,20.6and6.8-fold higher than the phoU wild type isolate, which suggested the higher level of phoU gene expression in the phoU G236E mutant isolate leading to a more powerful formation of persisters to the threat of the various antimicrobials. However, the excact mechanisms are still to be understood.