Characteristics Of RdxA Gene Mutations In Metronidazole-resistant Helicobacter Pylori And The Related Application

Background/Objective:Because antibiotic resistance has become one of the overriding factors reducingthe success rates of H.pylori eradication therapy in recent years, elucidating theresistance mechanisms is increasingly vital. In the field of research about resistancemechanisms of H.pylori to metronidazole, the researchers seem to agree that mutationsin rdxA encoding an oxygen-insensitive NADPH nitroreductase and frxA geneencoding a NADPH-flavin oxidoreductase are the main mechanism in MTZ resistance.However, the detail contributation of frxA mutation remains controversy. Recentstudies showed mutations in rdxA was necessary for MTZRstrains to obtain theresistant phenotype, while frxA mutations alone was not sufficient to induce resistance.Reports from UK and Japan demonstrated that the inactivation of frxA was also foundin MTZSstrains, and inferred inactivation of rdxA may be the real confer. Unfortunately,the precise relation between rdxA mutational sites or types and resistant level isunknown. Moreover, many clinical trials and Maastricht IV/Florence ConsensusReport explicitly indicated that the consistency between clinical outcomes and in vitrosusceptibility tests for MTZ in H.pylori was poor. Currently, CLSI (The Clinical andLaboratory Standards Institute) and EUCAST (European Committee on AntimicrobialSusceptibility Testin) has defined8μg/ml as the MIC breakpoint of Mtz for H.pylori.Whether this cut-off value is the most appropriate to predict clinical outcome or notstill requires a further investigation. To understanding these doubts, we’ll analyse thecharacteristics of rdxA mutations in induced MTZRstrains and MIC distribution of1439strains isolated from China during2008and2011. Together with results ofcomparison in MIC distribution between Chinese strains and11168strains fromEUCAST database, we will discuss the MTZ resistance mechanism in H.pylori. Andon the base of these information, in combination with pharmacokinetics of MTZ, therationality of current breakpoint would be discussed.Methods:(1) Induction of resistance in vitro to select single clones: MTZ-susceptive H.pylori26695strains were incubated on MTZ-free plates for2days. Then inoculumsof each suspension(108colony forming units, cfu) were transferred onto surface ofmedium containing concentrations of MTZ (group A,8μg/ml; group B,32μg/ml) for3-5days to screen resistant single clonies.(2) The clones picked from each culture were subsequently grown oncorresponding resistance flats in large scale. Collecting the induced MTZ-resistantstrains in vitro would be achieved to extract genome DNA.(3) Design of specific primers according to the sequences of H.pylori26695rdxAand frxA genes from GeneBank.(4) Amplification and sequencing frxA and rdxA genes.(5) Bioinformatics analysis: using AlignX and MEGA to analyse the RdxAmutational features at nucleotides and amino acids level. SMART analysis service andDomPred were used to predict the function domain, while Swiss-model server andPymol were used to analyse the spatial structures differences of RdxA.(6) Phenotypic validation experiment: The strains screened from group A wereinoculated on mediums with various concentrations of MTZ (8,16,32,64μg/ml and128μg/ml), while the substrate of strains from group B involved32μg/ml,64μg/ml and128μg/ml of MTZ. All strains were incubated under microaerophilic condition at37℃for2-4days.(7) Statistical analysis of rdxA sequence of clinical strains in previous studies: Aliterature search was done for articles published from2000to2012(abstracts were notincluded) by performing a Pubmed and Medline search with the keywords:‘Helicobacter pylori’,‘metronidazole’,‘resistant’,‘rdxA’,‘minimum inhibitoryconcentration’. Study eligibility was limited to those that have sequenceing data ofrdxA gene and detailed MIC values.Association between MIC and truncated RdxA was quantified by chi-square test.A P value of less than0.05was accepted as statistically significant. Data were analyzedusing SPSS statistical software package version16.0(SPSS Inc., Chicago, IL).(8) Determination of MICs: The MICs of MTZ in induced in vitro and1439clinical H.pylori strains were determined by E-test method (AB Biodisk, Sweden).Experiments were performed according to the protocols of the CLSI and repeated2 times. H.pylori26695and ATCC43504were used as control strains.(9) Calculating the steady state peak plasma concentrations of MTZ in currentHelicobacter pylori eradication regimens using formula as follows:Results:(1)38MTZRsingle clonies were picked. Of that total,31were screened fromgroup A, while7were born of group B. The genomic DNAs of the38resistant cloniesand their sensitive parental H.pylori26695were extracted for further determining andanalysising the sequences of frxA and rdxA. Sequencing analysis showed the mutationrate of rdxA was100%, while only2strain obtained mutant frxA genotype (themutation rate was5.3%).(2) Characteristics of RdxA amino-acid sequences from induced strains: Theamino-acid sequences were compared with RdxA of H.pylori26695. Three hot regionstermed α (10-20th Aa), β (30-80th Aa) and γ (140-180th Aa) were identified withinamino acid sequence of RdxA,and the rates were13.2%,68.4%and18.4%respectively.No changes occured in80-140th Aa region. RdxA in67.7%(21/31) of strains fromgroup A were premature, and32.3(10/31) of strains obtainedone-amino-acid-substitution. And yet,7strains from group B were all have a truncatedRdxA, and the truncation sites were all in α region, close to N-terminal. Overall,truncated sites predominantly located in region β, approximately82.1%(23/28).(3) Prediction analysis of function for mutant RdxA: RdxAs with truncation sitesbefore56th Aa (include56) have been shown absent of nitroreductase domain. The3Dmodels of38mutational peptides were built using the X-ray structure of RdxA (PDB3qdl) from H.pylori26695as the structural template which is composed of two chainsand has two FMN (flavin mononucleotide) binding domains. FMN is linked toflavodoxins by non-covalent binding. The binding amino acids were mainly located inα, β and γ regions. From the view of structures, affinity between FMN and RdxAswhose fracture positions were before the80th amino acid (termed type A mutation)may decrease more than long truncated RdxAs with fracture positions after the140thamino acid (termed type B mutation). Pockets of full-length RdxAs (termed type C mutation) were unbroken, and the structures may be influenced slightly by oneamino-acid substitution.(4) The resistant level of strains with mutant RdxA: Strains with short truncatedRdxA all possess ability to grow on medium containing higher concentration MTZ(more than64μg/ml) and strains with long truncated RdxA can grow normally underthe MTZ concentration of32μg/ml and a few under64μg/ml of MTZ. Some otherstrains owning full-length RdxA solely from groupA can live only on the agar platewith less than32μg/ml of MTZ, perhaps this is because these missense mutationsdecrease but not entirely eliminate the activity of encoded protein, except one strainwith C148Y which obtained higher resistant level (≥256μg/ml).(5) The relevance of RdxA mutation pattern and MIC level in previous researches:A total of194clinical isolates described in previous ten researches during2000and2012from different countries were reviewed.We divided them into two parts:81strainswith MIC less than or equal to16μg/ml and other113strains with MIC more than16μg/ml. The statistical analysis showed that the rate of truncated RdxA in these twogroups has significant difference (9.88%vs56.64%, P<0.001).(6) Results of MIC: MICs range of strains containing type A, B and C mutationsare≥64μg/ml,≥32μg/ml and8-32μg/ml (except the strain with C148Y,MIC≥256μg/ml), respectively. In1439strains isolated from China, a continuousspectrum of MIC values was observed and there were977(67.89%) resistant strainsjudged by current breakpoint. By comparing this with data from the EUCAST MICdistribution website (11168strains), we found H.pylori with high MIC level in Chinesestrains in recently years occupied a larger proportion, and the number of strains whoseMIC were256μg/ml increased sharply.(7) Oral administration of400mg tid or500mg tid (common administrationmethods and dose in current regimens) produced about20.1μg/ml and26.0μg/ml ofsteady state peek plasma concentrations (Css max) respectively, and this doesn’tcorrespond to current breakpoint (8μg/ml). The mismatched percentage were over12.5%(>124/977) based on MIC distribution of H.pylori recently from China. Conclusions:(1) Lower antibiotic pressure could induced resistant phenotype more easily, andthis indicates that insufficient bactericidal dosage may function as selective pressurefor resistant H.pylori in vivo.(2) frxA mutation contribute little to MTZ resistance of H.pylori, while rdxA is theprime confer.(3)3hot mutation regions were identified in RdxA, namely that α (10-20th Aa), β(30-80th Aa) and γ (140-180th Aa). Region β is the major mutation region in RdxA.Loss or mutation of FMN-binding residues would greatly affected protein activity. Inaddition, the148cysteine residue are also crucial to MTZ reduction activity of RdxA.(4) The mutation pattern of rdxA is related with MIC level: type A mutation resultsin high level resistance; type B pattern leads to middle resistant phenotype; while typeC may mainly generate low MIC, except some key amino acids were effected. AndRdxA truncation tends to happen in strains with middle-high resistant level.(5) The distribution of MIC for MTZ showed a continuous spectrum rangingbetween8ug/ml and <256ug/ml, reflecting the production difference in bactericidalintermediate of MTZ caused by diverse mutations in RdxA. A centralized MICdistribution at256μg/ml, suggesting it was related to full deactivation of RdxA.Accordingly, we proposed that MTZ resistant mechanisms in H.pylori may becategorized into two types: qualitative change-deactivation of RdxA (termed type)and quantitative change-partial inactivation of RdxA (termed type). Consequently,for cases of H.pylori with MIC≥256μg/ml infection, the eradication regimens shouldnot include MTZ.(6) Together with pharmacokinetic data, characteristic of MIC distribution andputative resistance mechanism, we conclude that the current breakpoint for MTZ inH.pylori is not most appropriate.(7) Two multi-centre randomized controlled clinical trials are needed to establisha rational breakpoint:①Comparing the efficacy of H.pylori eradication of single largeonce-daily dose administration and low dose serial MTZ administered dosage regimen,such as1.5g, qd and500mg, tid, in the same drug combinations, total dose and standardtreatment duration; efficacy of H.pylori eradication in different drug combinations with the same MTZ administration.②Designing clinical trials to evaluate the relationshipbetween level of MTZ resistance and success of H.pylori eradication in triple orquadruple therapies containing MTZ.