Genomewide overexpression screen for fosfomycin resistance in Escherichia coli: Mura confers clinical resistance at low fitness cost

Genomewide O e exp ession Sc een o Fos omycin Resis ance in Esche ichia coli: Mu A Con e s Clinical Resis ance a Low Fi ness Cos Alejand o Couce, a Alejand a B iales, b Alexand o Rod íguez-Rojas, a Coloma Cos as, a Ál a o Pascual, b and Jesús Blázquez a Cen o Nacional de Bio ecnología, Consejo Supe io de In es igaciones Cien í icas, Mad id, Spain, a and Depa amen o de Mic obiología, Uni e sidad de Se illa, Se ille, Spain b To de e mine whe he he o e exp ession o ch omosomal genes can con e os omycin esis ance, genomewide sc eening o a comple e se o 5,272 plasmid-exp essed open eading ames o Esche ichia coli (ASKA collec ion) was pe o med. Majo esul s a e ha (i) no clinical le el o esis ance is achie ed by o e exp essing ch omosomal genes, excep mu A; (ii) his le el is eached a a low i ness cos ; and (iii) his cos is much lowe han ha imposed by o he mu a ions con e ing os omycin esis ance. The eme gence o an ibio ic esis ance mu an s in a bac e ial popula ion is shaped by se e al ac o s, and bo h he mu a ion a e and he i ness cos o esis ance a e pa icula ly ele an (2). I esis ance came a a high i ness cos , he g ow h a e would no be enough o o se he clea ance imposed by he hos o o p e en he bac e ia om being ou compe ed by i e suscep ible bac e ia once he an ibio ic is emo ed (2). Fos omycin (Fos) is a b oad-spec um bac e icidal an ibio ic ac i e agains bo h G am-posi i e and G am-nega i e bac e ia (7). Con enien ly, Fos ea men s ha e shown a ela i ely low likelihood ha esis an mu an s will pe sis in i o, p obably due o he high biological cos o esis ance mu a ions, leading o good he apeu ic e ec i eness (19). Fos esis ance is acqui ed mainly by educing he cell’s d ug up ake (4, 10, 11, 22), al hough ac i e e lux, a ge al e a ions, and plasmid-encoded esis ance ha e also been desc ibed in Esch- e ichia coli and o he species (18, 21, 23, 24). Resis an mu a ions usually en ail a mode a e o high i ness cos (1, 15, 19), as well as educed i ulence (8, 13, 15), which has been in oked o explain he low p e alence o esis an s ains (19). Despi e his, a signi i- can inc ease in esis ance has been ecen ly desc ibed a e an i- bio ic p essu e in he communi y (20), sugges ing ha he e may be o he , uniden i ied, ways o a ain less cos ly high-le el Fos esis ance. The genome o E.coli ha bo s a subs an ial ese oi o esis- ance genes whose o e exp ession can dec ease suscep ibili y (22). The comple e E.coli open eading ame (ORF) ASKA lib a y has al eady been sc eened o esis ance o 237 oxins and an ibio ics. Howe e , Fos was no included among he d ugs used o sc een o esis ance. Gi en he enewed in e es in Fos ea men , we explo ed he capaci y o he o e exp ession o ch omosomal E. coli genes o con e clinical le els o Fos esis ance by sc eening he comple e ASKA lib a y (12). Genomewide o e exp ession sc eening o Fos esis ance. The comple e ASKA lib a y was eplica ed in duplica e in 96-well pla es con aining Lu ia-Be ani (LB) b o h plus chlo amphenicol (50 ␮ g/ml) wi h and wi hou isop opyl- ␤ -D- hiogalac opy ano- side (IPTG) o a inal concen a ion o 100 ␮ M and incuba ed o e nigh a 37°C. The ansc ip ion o he cloned genes o he ASKA collec ion is unde he con ol o he P ac p omo e , which is induced by he addi ion o IPTG (12). A 5- ␮ l sample om each well was spo ed on o an LB aga pla e con aining ei he 32 ␮ g/ml Fos ( he EUCAST b eakpoin o En e obac e iaceae [h p://www .eucas .o g/clinical_b eakpoin s/]) o no an ibio ic. Pla es we e incuba ed o 24 h a 37°C. Only 1 o he 5,272 clones was able o g ow on Fos. The clone, con aining plasmid pCA24N-mu A, was isola ed om he o iginal ozen 96-well pla e and in oduced by ans o ma ion in o s ain MG1655. Th ee independen ans o - man s we e isola ed, and Fos esis ance was e i ied as indica ed abo e. The p esence o he wild- ype mu A gene in he plasmid was e i ied by sequencing. The e o e, unde ou expe imen al condi ions, no clinical le el o Fos esis ance can be achie ed by he o e exp ession o ch omosomal genes, excep mu A. E ec o mu A o e exp ession on Fos esis ance le el. MICs o Fos o s ains MG1655(pCA24N) and MG1655(pCA24N- mu A) we e de e mined by he b o h mic odilu ion me hod as ecommended by he CLSI (3), excep ha LB b o h was used ins ead o Muelle -Hin on medium. Table 1 shows how inc eases in mu A ansc ip ion, caused by ising IPTG concen a ions, p o- mo e s ong inc eases in he MIC o Fos. E ec o mu A o e exp ession on g ow h a e. To gain a quan i a i e insigh in o he i ness cos imposed by his ype o esis ance, g ow h cu es in he p esence o di e en concen a- ions o IPTG (3.3, 10, 33, and 100 ␮ M) we e eco ded. O e nigh cul u es we e dilu ed 1:100, and g ow h was esumed o 4 h, wi h and wi hou di e en concen a ions o IPTG. Cul u es o he con ol and mu A-ca ying s ains we e hen dilu ed 1:36 in LB b o h plus IPTG in a inal olume o 180 ␮ l pe well in a la - bo om 96-well pla e. The pla e was co e ed wi h a lid o p e en e apo a ion and incuba ed a 37°C in a mul iwell luo ime e . Op ical densi y a 595 nm was eco ded e e y 10 min a e 10 s o o bi al shaking. Two con ols we e included, s ain MG1655 ha - bo ing he ec o pCA24N and he s ain exp essing mu A, MG1655(pCA24N-mu A), wi hou IPTG. Figu e 1A shows ha he addi ion o IPTG, e en a he highe concen a ion, did no p oduce any e ec on he g ow h o he con ol s ain in LB b o h. On he con a y, when he exp ession o mu A was inc eased by Recei ed 11 No embe 2011 Re u ned o modi ica ion 25 Decembe 2011 Accep ed 18 Feb ua y 2012 Published ahead o p in 27 Feb ua y 2012 Add ess co espondence o Jesús Blázquez, [email p o ec ed]. Copy igh © 2012, Ame ican Socie y o Mic obiology. All Righ s Rese ed. doi:10.1128/AAC.06122-11 0066-4804/12/$12.00 An imic obial Agen s and Chemo he apy p. 2767–2769 aac.asm.o g 2767 on July 26, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://aac.asm.o g/Downloaded om he addi ion o di e en concen a ions o IPTG, he g ow h o MG1655(pCA24N-mu A) was a ec ed (Fig. 1B). These esul s indica ed ha o e p oduc ion o Mu A con e s a i ness cos p opo ional o he le el o ansc ip ion (Fig. 2), which opens he possibili y o e olu iona y adjus men o he esis ance- i ness adeo . In pa icula , clinical esis ance le els (32 ␮ g/ml) can be achie ed a a low i ness cos (abou 5%, he esul o in e pola ion in Fig. 2), whe eas he i ness cos o he commonly ound pe meabili y mu an s (wi h al e ed GlpT and/o UhpT unc ionali y) has been es ima ed o be ⬃20% (19). A sim- ple calcula ion can illus a e he meaning o his i ness cos s, in e ms o how long i will ake o a wild- ype s ain o domina e he popula ion. Assuming ha bo h wild- ype and mu an popula- ions g ow exponen ially and excluding de no o mu a ions, i can be easily shown ha he ime needed o a y he a io o mu an o wild- ype bac e ia by ac o C( C ) is equal o log W C, whe e wis he ela i e i ness o he mu an ( a io o he g ow h a es). Calcula - ing C wi h i ness cos s o 5% and 20%, espec i ely, shows ha i will ake mu A o e exp ession mu an s mo e han ou imes as long as pe meabili y mu an s o be ou compe ed by he wild ype in he absence o an ibio ic p essu e. The exac numbe o gene - a ions depends on C. Fo example, i will ake he wild- ype 51.6 gene a ions o inc ease i s equency om 0.001% o 90% agains pe meabili y mu an s, whe eas i will ake 224.5 gene a ions agains o e exp ession mu an s. Due o he cu en sca ci y o new an ibio ics, Fos has been p oposed as an al e na i e ea men o in ec ions caused by a wide a ie y o bac e ia (7). Ou genomewide sc eening has shown ha o he 5,272 ch omosomal genes es ed, only 1, mu A, is able o con e clinical le els o esis ance when o e exp essed. E en TABLE 1 E ec o mu A ansc ip ion le el on os omycin MIC o s ain MG1655 con aining ei he pCA24N ( ec o alone) o pCA24N- mu A (exp essing mu A) IPTG concn ( ␮ M) MIC o os omycin ( ␮ g/ml) pCA24N pCA24N-mu A 014 3.3 1 8 10 1 16 33.3 1 64 100 1 4,096 FIG 1 G ow h cu es a di e en IPTG concen a ions. The alues shown a e a e ages o se en independen expe imen s. Di e en shades o g ay ep esen di e en concen a ions o IPTG. (A) G ow h o con ol s ain MG1655 ca ying he cloning ec o pCA24N. (B) G ow h o he s ain ca ying mu A-exp essing plasmid pCA24N-mu A. FIG 2 T adeo be ween i ness and esis ance. Fi ness and MICs we e es i- ma ed o s ain MG1655 wi h a plasmid ha bo ing he mu A gene a di e en le els o induc ion (exp essed as concen a ions o IPTG). The g ow h a e was es ima ed as he maximum slope o he na u al loga i hm o op ical densi ies e sus ime. Cu e i ing was done using a eg ession spline me hod (4, 24), and i ness was calcula ed as he a io o he g ow h a e a each concen a ion o IPTG o ha wi hou IPTG (16). These measu emen s we e de e mined om se en independen cul u es, and a single mean ela i e i ness le el was calcula ed o each condi ion. E o ba s ep esen s anda d de ia ions. Couce e al. 2768 aac.asm.o g An imic obial Agen s and Chemo he apy on July 26, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://aac.asm.o g/Downloaded om hough he e ec o o e exp ession o mu A on Fos esis ance has been desc ibed al eady (9, 16, 19), ou esul s indica e ha no o he ch omosomal genes can p oduce Fos esis ance by o e ex- p ession. In addi ion, we show ha o e p oduc ion o Mu A o a le el high enough o p oduce clinical Fos esis ance exac s a i ness cos signi ican ly lowe han ha imposed by o he mu a ions ound in clinical isola es (19). How likely a e mu an bac e ia o e exp essing mu A o be se- lec ed in i o? Apa om he i ness cos , he mu a ion a e also plays a majo ole in de e mining his p obabili y (2). O e exp es- sion mu an s could easily a ise unde na u al condi ions by in- c easing RNA polyme ase binding a ini y h ough p omo e mu- a ion, by dis up ing gene con ol exp ession as desc ibed o o he esis ance de e minan s (6, 17), o by acqui ing o eign DNA segmen s ia, o ins ance, inse ion sequences wi h s ong p omo e ac i i y (5, 14). In any case, u he clinical s udies a e necessa y o de e mine he ela i e impo ance o his mechanism o esis ance. ACKNOWLEDGMENTS This wo k was suppo ed by g an s PI10/00105 and REIPI RD06/0008 om he Minis e io de Ciencia e Inno ación, Ins i u o de Salud Ca los III ( he la e was co inanced by he Eu opean De elopmen Regional Fund “A Way o Achie e Eu ope” ERDF); by he Spanish Ne wo k o Resea ch on In ec ious Diseases (REIPI RD06/0008); and by he PAR p ojec (241476) om he EU 7 h F amewo k P og amme. REFERENCES 1. Alós JI, Ga cia-Pena P, Tamayo J. 2007. Biological cos associa ed wi h os omycin esis ance in Esche ichia coli isola es om u ina y ac in ec- ions. Re . Esp. Quimio e . 20:211–215. 2. Ande sson DI, Le in BR. 1999. The biological cos o an ibio ic esis- ance. Cu . Opin. Mic obiol. 2:489–493. 3. CLSI. 2006. Me hods o dilu ion an imic obial suscep ibili y es s o bac e ia ha g ow ae obically. App o ed s anda d—7 h edi ion. CLSI M7-A7. Clinical and Labo a o y S anda ds Ins i u e, Wayne, PA. 4. Co da o JC, e al. 1976. Fos omycin esis ance: selec ion me hod o in e nal and ex ended dele ions o he phosphoenolpy u a e:suga phos- pho ans e ase genes o Salmonella yphimu ium. J. Bac e iol. 128:785– 793. 5. Co ec S, e al. 2003. AmpC cephalospo inase hype p oduc ion in Acin- e obac e baumannii clinical s ains. J. An imic ob. Chemo he . 52:629– 635. 6. Depa dieu F, Cou alin P, Kolb A. 2005. Binding si es o VanRB and sigma70 RNA polyme ase in he anB ancomycin esis ance ope on o En e ococcus aecium BM4524. Mol. Mic obiol. 57:550–564. 7. Falagas ME, Giannopoulou KP, Kokolakis GN, Ra ailidis PI. 2008. Fos omycin: use beyond u ina y ac and gas oin es inal in ec ions. Clin. In ec . Dis. 46:1069–1077. 8. Gismondo MR, e al. 1994. Esche ichia coli: e ec o os omycin ome a- mol on some u o i ulence ac o s. J. Chemo he . 6:167–172. 9. Ho ii T, Kimu a T, Sa o K, Shibayama K, Oh a M. 1999. Eme gence o os omycin- esis an isola es o Shiga-like oxin-p oducing Esche ichia coli O26. An imic ob. Agen s Chemo he . 43:789–793. 10. Kadne RJ, Winkle HH. 1973. Isola ion and cha ac e iza ion o mu a- ions a ec ing he anspo o hexose phospha es in Esche ichia coli.J. Bac e iol. 113:895–900. 11. Kahan FM, Kahan JS, Cassidy PJ, K opp H. 1974. The mechanism o ac ion o os omycin (phosphonomycin). Ann. N. Y. Acad. Sci. 235:364– 386. 12. Ki agawa M, e al. 2005. Comple e se o ORF clones o Esche ichia coli ASKA lib a y (a comple e se o E. coli K-12 ORF a chi e): unique e- sou ces o biological esea ch. DNA Res. 12:291–299. 13. Klein U, Pawelzik M, Op e kuch W. 1985. In luence o be a-lac am an ibio ics, os omycin and ancomycin on he adhe ence (hemagglu ina- ion) o Esche ichia coli-con aining di e en adhesins. Chemo he apy 31: 138–145. 14. Maki H, Mu akami K. 1997. Fo ma ion o po en hyb id p omo e s o he mu an llm gene by IS256 ansposi ion in me hicillin- esis an S aph- ylococcus au eus. J. Bac e iol. 179:6944–6948. 15. Ma chese A, Gualco L, Debbia EA, Schi o GC, Schi o AM. 2003. In i o ac i i y o os omycin agains g am-nega i e u ina y pa hogens and he biological cos o os omycin esis ance. In . J. An imic ob. Agen s 22(Suppl. 2):53–59. 16. Ma qua d JL, Siegele DA, Kol e R, Walsh CT. 1992. Cloning and sequencing o Esche ichia coli mu Z and pu i ica ion o i s p oduc , a UDP-N-ace ylglucosamine enolpy u yl ans e ase. J. Bac e iol. 174: 5748–5752. 17. McAleese FM, Fos e TJ. 2003. Analysis o mu a ions in he S aphylococ- cus au eus cl B p omo e leading o inc eased exp ession. Mic obiology 149:99–109. 18. Mendoza C, e al. 1980. Plasmid-de e mined esis ance o os omycin in Se a ia ma cescens. An imic ob. Agen s Chemo he . 18:215–219. 19. Nilsson AI, Be g OG, Aspe all O, Kahlme e G, Ande sson DI. 2003. Biological cos s and mechanisms o os omycin esis ance in Esche ichia coli. An imic ob. Agen s Chemo he . 47:2850–2858. 20. O eo J, e al. 2010. Pa allel inc ease in communi y use o os omycin and esis ance o os omycin in ex ended-spec um be a-lac amase (ESBL)- p oducing Esche ichia coli. J. An imic ob. Chemo he . 65:2459–2463. 21. Poole K. 2005. E lux-media ed an imic obial esis ance. J. An imic ob. Chemo he . 56:20–51. 22. Soo VW, Hanson-Man ul P, Pa ick WM. 2011. A i icial gene ampli i- ca ion e eals an abundance o p omiscuous esis ance de e minan s in Esche ichia coli. P oc. Na l. Acad. Sci. U. S. A. 108:1484–1489. 23. Takaha a S, e al. 2010. Molecula mechanisms o os omycin esis ance in clinical isola es o Esche ichia coli. In . J. An imic ob. Agen s 35:333– 337. 24. Wachino J, Yamane K, Suzuki S, Kimu a K, A akawa Y. 2010. P e a- lence o os omycin esis ance among CTX-M-p oducing Esche ichia coli clinical isola es in Japan and iden i ica ion o no el plasmid-media ed os omycin-modi ying enzymes. An imic ob. Agen s Chemo he . 54: 3061–3064. Fos omycin Resis ance a Low Fi ness Cos May 2012 Volume 56 Numbe 5 aac.asm.o g 2769 on July 26, 2017 by USE/BCTA.GEN UNIVERSITARIAh p://aac.asm.o g/Downloaded om