Data & Analytics

2018 - Genetic Diversity of norA, Coding for a Main Efflux Pump of Staphylococcus aureus

Description
2018 - Genetic Diversity of norA, Coding for a Main Efflux Pump of Staphylococcus aureus
Published
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  ORIGINAL RESEARCH published: 09 January 2019doi: 10.3389/fgene.2018.00710  Edited by: Silvia Buroni,University of Pavia, Italy   Reviewed by: Maria José Saavedra,Universidade de Trás os Montes e Alto Douro, Portugal Teruo Kuroda,Hiroshima University, Japan *Correspondence: Taane G. Clark taane.clark@lshtm.ac.uk Isabel Couto icouto@ihmt.unl.pt  Specialty section: This article was submitted toEvolutionary and GenomicMicrobiology, a section of the journal Frontiers in Genetics  Received:  28 August 2018  Accepted:  18 December 2018  Published:  09 January 2019 Citation: Costa SS, Sobkowiak B,Parreira R, Edgeworth JD, Viveiros M,Clark TG and Couto I (2019) GeneticDiversity of norA, Coding for a MainEfflux Pump of Staphylococcus aureus. Front. Genet. 9:710.doi: 10.3389/fgene.2018.00710 Genetic Diversity of  norA , Coding fora Main Efflux Pump of Staphylococcus aureus Sofia Santos Costa 1  , Benjamin Sobkowiak   2  , Ricardo Parreira 1  , Jonathan D. Edgeworth  3  ,Miguel Viveiros 1  , Taane G. Clark   2,4 *  and  Isabel Couto 1 *  1 Global Health and Tropical Medicine, Instituto de Higiene e Medicina Tropical, Universidade NOVA de Lisboa, Lisbon,Portugal,  2 Faculty of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London,United Kingdom,  3 Department of Infectious Diseases, Centre for Clinical Infection and Diagnostics Research, Guy’s and St Thomas’ NHS Foundation Trust, King’s College London, London, United Kingdom,  4 Faculty of Epidemiology and PopulationHealth, London School of Hygiene and Tropical Medicine, London, United Kingdom NorA is the best studied efflux system of   Staphylococcus aureus  and thereforefrequently used as a model for investigating efflux-mediated resistance in this pathogen.NorA activity is associated with resistance to fluoroquinolones, several antiseptics anddisinfectants and several reports have pointed out the role of efflux systems, includingNorA, as a first-line response to antimicrobials in  S. aureus . Genetic diversity studiesof the gene  norA  have described three alleles;  norAI ,  norAII  and  norAIII . However,the epidemiology of these alleles and their impact on NorA activity remains unclear. Additionally, increasing studies do not account for  norA  variability when establishingrelations between resistance phenotypes and  norA  presence or reported absence,which actually corresponds, as we now demonstrate, to different  norA  alleles. In thepresent study we assessed the variability of the  norA  gene present in the genome of over1,000  S. aureus  isolates, corresponding to 112  S. aureus  strains with whole genomesequences publicly available; 917 MRSA strains sourced from a London-based studyand nine MRSA isolates collected in a major Hospital in Lisbon, Portugal. Our analysesshow that  norA  is part of the core genome of   S. aureus.  It also suggests that occurrenceof   norA  variants reflects the population structure of this major pathogen. Overall, thiswork highlights the ubiquitous nature of   norA  in  S. aureus  which must be taken intoaccount when studying the role played by this important determinant on  S. aureus resistance to antimicrobials. Keywords:  Staphylococcus aureus ,  norA , alleles, variability, efflux INTRODUCTION Staphylococcus aureus  is one of the major human pathogens in the hospital and community settings, causing a wide array of clinical manifestations, from mild skin infections to life-threatening systemic infections (Chambers and DeLeo, 2009; Tong et al., 2015). Development and acquisition of resistance to antibiotics and other antimicrobials is of paramountimportance in  S. aureus , as exemplified by the common occurrence and dissemination of strains displaying a phenotype of multidrug resistance (MDR), including methicillin-resistant Frontiers in Genetics | www.frontiersin.org  1  January 2019 | Volume 9 | Article 710  Costa et al. Diversity of   Staphylococcus aureus norA Staphylococcus aureus  (MRSA) strains (Chambers and DeLeo,2009).MRSAarespreadworldwideandareapublichealththreat,ranked amongst the major nosocomial pathogens (Lee et al.,2018; Tacconelli et al., 2018). MRSA epidemiology has revealed the local or global dissemination of a few number of clones,namely the lineages in clonal complexes CC1, CC5, CC8, CC22,CC30, CC45, CC59, and CC80 (Lee et al., 2018).In recent years, several studies have supported drug efflux asa player in the emergence of resistance toward antibiotics andother antimicrobials in  S. aureus  (DeMarco et al., 2007; Furi et al., 2013; Kwak et al., 2013; Costa et al., 2015). Of particular interest are multidrug efflux pumps (MDR EPs), which extrude a widerange of chemically dissimilar antimicrobials, being frequently associated with MDR phenotypes in bacteria (Piddock, 2006;Poole, 2007). In  S. aureus , more than twenty putative MDR EPsare encoded in the chromosome (Schindler et al., 2015), of which several have already been characterized (Costa et al., 2013b). Among these, NorA is the most well studied, being frequently used as a model for studying efflux-mediated resistance in thispathogen.NorA is a 388 aminoacid protein with 12 transmembranesegments(TMS)thatbelongstotheMajorFacilitatorSuperfamily (MFS) of secondary transporters. This MDR EP uses theproton motive force to extrude from the cell fluoroquinolones,ethidium bromide, quaternary ammonium compounds, andother antimicrobials (Yoshida et al., 1990; Kaatz et al., 1993; Neyfakh et al., 1993). Several reports have associated NorAactivity to low-level resistance to fluoroquinolones and reducedsusceptibility to biocides (DeMarco et al., 2007; Huet et al., 2008; Kosmidis et al., 2010; Costa et al., 2011, 2015; Furi et al., 2013). Other studies support a broader substrate range for NorA,including siderophores (Deng et al., 2012) and fusaric acid(Marchi et al., 2015). The NorA encoding gene,  norA,  was first identified in thechromosome of a fluoroquinolone-resistant  S. aureus  isolate,collected in 1986 at a Japanese hospital (Ubukata et al., 1989). Early studies have shown that genetic diversity of this gene canbe captured by three alleles, differing up to 10% at the level of the nucleotide sequence and 5% in the polypeptide sequence; norAI   (Yoshida et al., 1990),  norAII   (also described as  norA23 )(Noguchi et al., 2004) and  norAIII   (also described as  norA1199 )(Kaatz et al., 1993). The occurrence of   norA  variants is alsostrengthened by a recent study by  Brooks et al. (2018) that refers to the variability of this gene in a set of over 150  S. aureus  strains.Although some studies have been conducted to ascertain theimpact of this genetic diversity on the efflux activity of NorA(Schmitz et al., 1998; Sierra et al., 2000; Noguchi et al., 2004), this effect remains unclear.Despite this early characterization, there are still contradictory reports in literature on the role of NorA in  S. aureus  efflux-mediated antimicrobial resistance. Several studies have reportedon the putative absence of   norA , most probably due to failure toamplifythisgenewithprimersdirectedtoonlyoneofthepossible norA  alleles. The present study aims at clarifying some of theseaspects, by demonstrating that the  norA  gene is part of the coregenome of   S. aureus , reflecting on the genetic variability of thegene, its distribution amongst  S. aureus  clonal lineages, includingboth methicillin-resistant and -susceptible strains and possibleimpact on NorA function. MATERIALS AND METHODSStudy Datasets Four sets of nucleotide sequences of the  norA  structural gene andthe corresponding polypeptide sequences were used in this study,comprising (i) the sequences of the three  norA  alleles describedto date in literature; (ii) the  norA  sequences from 112  S. aureus whole genome sequences retrieved from the GenBank database;(iii) the  norA  sequences from 917 MRSA strains sourced froma London-based study (Auguet et al., 2018) (ENA accessionPRJEB11177); (iv) the  norA  sequences of nine MRSA isolatescollected in a major Hospital in Lisbon, Portugal, representativeof the circulating  norA  alleles in that hospital at the time (Costaet al., 2011, 2016), which were deposited in GenBank.A detailed list of all the sequences comprised in this study andrespective accession numbers can be found in  Supplementary Information S1 . The Lisbon MRSA isolates were previously characterized for their susceptibility toward fluoroquinolonesand biocides and presence of mutations in the quinolone-resistance determining region of   grlA /  gyrA  genes (Costa et al.,2011, 2013a) –  Supplementary Information S1 . Information onthe remaining  S. aureus  strains was gathered from the GenBank database and relevant published papers.Whenever sequence types (ST) were not provided,  in silico MLST was performed using the MLST 1.8 (Larsen et al., 2012) and SRST2 (Inouye et al., 2014) softwares. The datasets were used to construct a pan-genome (set of all genes within a givenspecies), and a core genome (set of genes found in all strains of that species) for  S. aureus  (Tettelin et al., 2005; van Tonder et al., 2014). Sequence Analysis of  norA  and MLST Alleles De novo  assemblies were performed for all London samplesusing Velvet (Zerbino and Birney, 2008) and VelvetOptimiser (Zerbino, 2010). Resulting contig FASTA files and FASTA files of the  S. aureus  samples obtained from GenBank were annotatedusingProkka(Seemann,2014).Thepan-genomeofthesesampleswas then constructed with Roary  (Page et al., 2015), and the sequences of the  norA  and MLST genes ( arcC, aroE, glpF, gmk, pta, tpi,  and  yqiL ) isolated using custom  R  scripts (R Core Team,2016).For the set of the nine Portuguese MRSA clinical isolates, norA  was amplified by PCR using three pairs of primers( Table 1 ). PCR reaction mixtures were prepared in 0.05 mLcontaining 2.5 U Taq Polymerase (Thermo Scientific, Waltham,MA, United States); 1X Taq buffer (Thermo Scientific); 30 pmolof each primer (Invitrogen, Carlsbad, CA, United States); 0.2 mMdNTPs (GE Healthcare, Chicago, IL, United States); 1.75 mMMgCl 2  (Thermo Scientific). The amplification conditions werethe following: initial DNA denaturation step at 95 ◦ C for 3 min,followedby35cyclesofdenaturationat94 ◦ Cfor1min,annealingat 52 ◦ C [norA(a)], 45 ◦ C [ norA (b)] or 50 ◦ C [ norA (c)] for 1 min, Frontiers in Genetics | www.frontiersin.org  2  January 2019 | Volume 9 | Article 710  Costa et al. Diversity of   Staphylococcus aureus norA TABLE 1 |  Primers used to sequence the  norA  promoter and coding region of  Staphylococcus aureus  clinical strains. Primer Sequence(5  – 3   ) AmpliconsizeLocation a Reference  norA _fw(a) TGTTAAGTCTT GGTCATCTGCA 761 706005–706026 Couto et al.,2008  norA _rv(a) CCATAAATCCA CCAATCCC706765–706747 This study b  norA _fw(b) TTCACCAAGCC ATCAAAAAG620 706671–706690 Sierra et al.,2000  norA _rv(b) CTTGCCTTTCT CCAGCAATA 707290–707271 Couto et al.,2008  norA _fw(c) GGTCATTATTA  TATTCAGTTGTTG419 707135–707158 Schmitz et al.,1998  norA _rv(c) GTAAGAAAAACGATGCTAAT 707553–707534  a Position of the primers in the genome of S. aureus DSM20231 (accession no.CP011526.1).  b The primer was designed with the aid of the free software Primer3(http://primer3.sourceforge.net/) and tested in silico (http://insilico.ehu.es/PCR/). andextensionat72 ◦ Cfor1min,followedbyafinalextensionstepat 72 ◦ C for 5 min. The PCR products were sequenced using thesame set of primers and the sequences analyzed and assembled tomake up the entire fragment using the software MAFFT v. 6 1 andBioEdit v. 7.0.9.0 2 .The  norA  and MLST gene sequences from all four datasetswere aligned using custom  R  scripts. Samples that sharedidentical  norA  and MLST sequences were grouped together andrepresentative sequences used for phylogenetic reconstruction.Maximum-likelihood (ML) tree construction was performedon the aligned  norA  and concatenated MLST gene sequencesseparatelyusingRAxML(Stamatakis,2014)withaGTRGAMMAevolutionary model. The resulting phylogenetic trees weredisplayed and annotated using FigTree 3 .  Analysis of the Impact of  norA  Variability on NorA Activity  The tridimensional structure of the NorA efflux pumpwas predicted via the  in silico  platform PHYRE2 (ProteinHomology/analogY Recognition Engine v2.0 4 ) (Kelley et al.,2015) based on the nucleotide sequences of the three  norA  variants described in literature. A mutational analysis based onthe predicted tridimensional structure was conducted with thePhyre2 Investigator tools, using the SuSPect algorithm (Yateset al., 2014). This algorithm produces a table of scores from 0to 100 according to predicted deleteriousness (0 = neutral to100 = deleterious). A score of 50 is recommended as a cut-off between neutral and deleterious variants, with extreme scoresbeing more confident predictions (Yates et al., 2014). In this work, a score of  ≥ 75 was used as cut-off value. 1 mafft.cbrc.jp/alignment/software/ 2 http://bioedit.software.informer.com/ 3 http://tree.bio.ed.ac.uk/software/figtree/ 4 http://www.sbg.bio.ic.ac.uk/phyre2 RESULTS AND DISCUSSION  norA  Is Part of the Core Genome of S. aureus Our study sought to establish if   norA  is part of the core genomeof   S. aureus.  This is a relevant question since a significant partof the studies on efflux-mediated resistance in this pathogenfocus mainly on the activity of NorA and some studies havereported the absence of the  norA  gene in several  S. aureus  strains(Monecke and Enricht, 2005; Vali et al., 2008; Conceição et al., 2015; Hasanvand et al., 2015; Liu et al., 2015; Ammar et al., 2016; Taheri et al., 2016; Antiabong et al., 2017; Hassanzadeh et al., 2017; Hadadi et al., 2018; Kernberger-Fischer et al., 2018). This reported absence could be explained by the genetic diversity  FIGURE 1 | (A)  The distribution of genes in the pan-genome of the Londonand GenBank   S. aureus  isolates (  n  = 1029). A total of 5,489 genes wereidentified with 1,551 (28.25%) genes found in more than 99% of strains. Themajority of genes (55.3%) were identified as ‘cloud’ genes, found only in asmall number of samples ( < 15%), demonstrating the diversity of the  S. aureus genome.  (B)  Rarefaction curves for total (red line) and core (blue line) genesidentified when increasing the sample size to reconstruct the pan-genome.Random sampling of samples was conducted 100 times and the mean size of the core genome and number of total genes [and standard deviation (dottedlines)] is shown as the sample size is increased by 1 to the total sample size(  n  = 1029). 90% of the total genes are identified after 514 samples (49.5% of total samples), and there is a plateau in size of the core genome ( ± 5% of thefinal genome size) at sample 256 (24.9% of total samples). The rarefactioncurve confirms that the gene diversity of the strains has been adequatelycaptured with the number of samples included in the analysis.Frontiers in Genetics | www.frontiersin.org  3  January 2019 | Volume 9 | Article 710  Costa et al. Diversity of   Staphylococcus aureus norA of the gene and a consequent primer failure during  norA  PCR screening. To test if   norA  is part of the core genome of   S. aureus ,we constructed the pan-genome from the 1029  S. aureus  isolateswhole genome sequence datasets (112 GenBank; 917 London S. aureus  isolates). The core genome of   S. aureus  consists of 1551genes (found in ≥ 99% of individuals with 95% BLAST identity; Figure1A ) with the genome of each individual isolate containinga median of 2196 genes (with a range of 1955 to 2469), marginally lower than the number of genes found in  S. aureus  referencestrains (Li et al., 2011). Rarefaction curves were calculated and determined that the number of samples used was adequate toaccurately reconstruct the pan-genome of the population; 90% of the total gene diversity was present when considering just 49.5%(514/1029) of the samples and there was a stabilization of thecore genome size after 24.9% (256/1029) of the total samples( Figure1B ).The norA genewasfoundtobepresentinallLondonand GenBank samples and thus is part of the core genome, witheach sample containing a single gene variant. Additionally, weperformed a Blastn search using as query the nucleotide sequenceof the  norAI   allele (GenBank accession no. D90119.1) againstall  S. aureus  genome sequences (complete, scaffold or contigformats) deposited in GenBank  5 , corresponding to more than8.000 (circa 8.150) sequences. All the Blastn searched sequencesretrieved hits with  ≥ 90% identity to  norAI   allele. This resultsupports the notion that  norA  is a  S. aureus  core gene, occurringin all  S. aureus  genomes. The finding that  norA  is part of   S. aureus core genome and thus is present in all  S. aureus  strains, impliesthat reporting the detection of this gene is not sufficient to makea direct association with a particular resistance phenotype. Tomake such a correlation, one must carry out expression analysisof the  norA  gene as well as of other efflux pump genes, as ithas been shown that  S. aureus  strains can display different effluxpump gene expression patterns (DeMarco et al., 2007; Kosmidis et al., 2010, 2012), even under pressure of the same antimicrobial(Huet et al., 2008; Costa et al., 2011, 2015). Genetic Diversity of the  norA  Gene One of the goals of this study was to determine the distributionof the  norA  variants amongst the several  S. aureus  clonallineages. A phylogenetic analysis was performed with the three 5 www.ncbi.nlm.nih.gov/genome/genomes/154 FIGURE 2 |  Maximum likelihood phylogeny of the  norA  sequences analyzed in this study. Identical  norA  sequences among all samples used in this study have beencollapsed and represented by a single label, which denotes a representative ST/CC of that group. The number of sequences is denoted by the last number of the tiplabel. Four major groups are found, clustering with the  norA  alleles described in literature,  norAI  (D90119.1),  norAII  (AB019536.1),  norAIII  (M97169.1), and the allelicvariant  norA -CC59/121. The  norAI  associated group is the most diverse, including samples belonging to CC1, CC5, CC8, CC15, CC80, and a wide range of sequence types. The  norAII  associated group contains samples belonging to clonal complexes CC22, CC30, CC36, and CC398. Major nodes supported withbootstrap values above 75% are denoted with a star.Frontiers in Genetics | www.frontiersin.org  4  January 2019 | Volume 9 | Article 710  Costa et al. Diversity of   Staphylococcus aureus norA norA  alleles and  norA  nucleotide sequences retrieved fromcomplete  S. aureus  genomes available in GenBank and from thePortuguese and London-based  S. aureus  collections ( Figure 2 ).As expected, a certain degree of genetic variability was revealedbythemaximum-likelihoodtreeobtained,despitethisgenebeingrelatively conserved.  Figure 3  shows the distribution of thepairwise genetic distance within  norA  across the samples usedto construct the phylogenetic tree, with a maximum of 143 SNPdifferences between samples (12.3% of the total gene length of 1164 bp).A global analysis showed a clear correlation betweenoccurrence of   norA  alleles and specific  S. aureus  clonal lineages.The  norAI   allele was mainly associated with genetic backgroundsof the clonal complexes CC1, CC5, CC8, and CC80 or closely related lineages. On the other hand,  norAII   and close variantsappear to be restricted to strains belonging to CC30, CC398, andCC22 ( Figure 2 ). The  norAIII   allele is only associated with the S. aureus  lineage CC45 and novel but related STs.Among the sequences retrieved from GenBank, the  norAI  allele is the most frequent (76 out of 112 strains) and the norAII   allele the second most frequent allele (21 out of 112strains), whereas  norAIII   occurs in just two CC45 strains. Wealso detected 13 more divergent variants, most closely related tothe  norAI   allele, that were associated with other lineages, suchas CC59 and CC121 ( Figures 2 ,  3 ). Although these findingsmay be biased as the dataset analyzed reflects the currentepidemiologically   S. aureus  relevant clones, with a frequency of  ∼ 70% MRSA strains (78 out of 112 genomes), it clearly showsthat  norAI   and related variants are the most prevalent allelesacross lineages, as suggested by earlier studies (Schmitz et al.,1998; Sierra et al., 2000; Noguchi et al., 2004). Interestingly, a sample isolated in Brazil in 2010 (FCFHV36) carried a  norAI  type variant of the  norA  gene that included a frameshift mutationin codon 129. This mutation caused a premature stop codon atcodon number 147, resulting in potential pseudogenization of  norA  in this isolate.Regarding the London isolates, we found the  norAII   alleleto be the most common allele, identified in 671/917 samples,due to the high number of CC22 strains in the study samples(539/917). As with the GenBank samples, the  norAIII   allelewas the least common variant found in the London isolates,with only the three CC45 and four novel ST strains carryingclosely related variants. The  norAI   allele was found in 220/917strains consisting of the broadest range of lineages, comprising21 different STs. Additionally, one ST22 sample possessed a  norA  variant with a frameshift mutation at codon 365 that results in apremature stop codon at codon number 367 and, again, possiblepseudogenization.Nine MRSA isolates, representative of a collection of 53 S. aureus  clinical isolates isolated at a major PortugueseHospital and previously characterized for efflux activity andclonal lineage (Costa et al., 2011, 2013a, 2015, 2016) were added to the study to ascertain their  norA  allele. The fullsequence of the  norA  gene was determined for these isolatesand compared with  norA  alleles from major clonal lineages( Figure 4 ). Both the  norAI   and  norAII   alleles were foundamongst the Portuguese isolates and their distribution reflectsthe findings for the other datasets analyzed. Eight isolates from FIGURE 3 |  Distribution of the pairwise SNP distance within  norA  sequences used in this study. The  norA  sequences include  norAI  (D90119.1),  norAII  (AB019536.1)and  norAIII  (M97169.1), the nine Portuguese isolates, 112 complete assemblies from GenBank, and 75 representative sequences of London isolates. The figure isannotated to show the pairwise SNP distance between isolates both within (blue) and between (red) each of the four major  norA  groups described in the main text:  norAI  group,  norAII  group,  norAIII  group, and CC121/CC59 group.Frontiers in Genetics | www.frontiersin.org  5  January 2019 | Volume 9 | Article 710
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks