Trends in Helicobacter pylori resistance to clarithromycin: from phenotypic to genomic approaches

1 T ends in Helicobac e pylo i esis ance o cla i h omycin: om pheno ypic o genomicapp oaches And eia T.Ma ques1, Jo ge M. B.Ví o 1,2, And eaSan os3, MónicaOleas o3 and Filipa F.Vale1,* MINI REVIEW Ma ques e al., Mic obial Genomics 2020;6 DOI 10.1099/mgen.0.000344 Recei ed 09 Augus 2019; Accep ed 10 Feb ua y 2020; Published 02 Ma ch 2020 Au ho a ilia ions: 1Hos –Pa hogen In e ac ions Uni , Resea ch Ins i u e o Medicines (iMed- ULisboa), Facul y o Pha macy, Uni e sidade de Lisboa, Lisboa, Po ugal; 2Depa men o Biochemis y and Human Biology, Facul y o Pha macy, Uni e sidade de Lisboa, 1649 003 Lisbon, Po ugal; 3Na ional Re e ence Labo a o y o Gas oin es inal In ec ions, Depa men o In ec ious Diseases, Na ional Ins i u e o Heal h D Rica do Jo ge, Lisbon, Po ugal. *Co espondence: Filipa F. Vale, ale. ilipa@ gmail. com Keywo ds: Helicobac e pylo i; cla i h omycin; esis ance; 23S ibosomal RNA subuni ; nex - gene a ion sequencing; poin mu a ions. Abb e ia ions: CLA, cla i h omycin; EF- Tu, elonga ion ac o he mo uns able; EPI, e lux pump inhibi o ; FISH, luo escence in si u hyb idiza ion; FRET, luo escence esonance ene gy ans e ; MIC, minimal inhibi o y concen a ion; NGS, nex - gene a ion sequencing; PPI, p o on pump inhibi o ; RND, esis ance- nodula ion- cell di ision; 23S RNA, 23S ibosomal RNA; WHO, Wo ld Heal h O ganiza ion. Accession numbe s o an um isola es o H. pylo i CLA- esis an sequenced genomes: SRR9930173–SRR9930179. 000344 © 2020 The Au ho s This is an open- access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License. Abs ac Fo a long ime Helicobac e pylo i in ec ions ha e been ea ed using he mac olide an ibio ic, cla i h omycin. Cla i h omycin esis ance is inc easing wo ldwide and is he mos common cause o H. pylo i ea men ailu e. He e we e iew he mecha- nisms o an ibio ic esis ance o cla i h omycin, de ailing he indi idual and combina ions o poin mu a ions ound in he 23S RNA gene associa ed wi h esis ance. Addi ionally, we conside he me hods used o de ec cla i h omycin esis ance, empha- sizing he use o high- h oughpu nex - gene a ion sequencing me hods, which we e applied o 17 newly sequenced pai s o H. pylo i s ains isola ed om he an um and co pus o a ecen colonized paedia ic popula ion. This se o isola es was com- posed o six pai s o esis an s ains whose pheno ype was associa ed wi h wo poin mu a ions ound in he 23S RNA gene: A2142C and A2143G. O he poin mu a ions we e ound simul aneously in he same gene, bu , acco ding o ou esul s, i is unlikely ha hey con ibu e o esis ance. Fu he , among suscep ible isola es, genomic a ia ions compa ible wi h mu a ions p e iously associa ed wi h cla i h omycin esis ance we e de ec ed. Exposu e o cla i h omycin may selec low- equency a i- an s, esul ing in a p og essi e inc ease in he esis ance a e due o selec ion p essu e. Da a SuMMaRy The genome accession numbe s and me ada a a e p esen ed in he Reposi o ies sec ion and in Table1. IN RoDuc IoN Helicobac e pylo i is a long- las ing human (s omach) a - elle companion causing gas i is, pep ic ulce and gas ic ca cinoma [1, 2]. Cla i h omycin (CLA) has been he basis o H. pylo i ea men because o i s low minimal inhibi o y concen a ion (MIC), good mucosal di usion and ela i ely small e ec on gas ic acidi y [3]. Fo yea s, his ea men consis ed o iple he apy ha combined CLA wi h ei he amoxicillin o me onidazole, and a p o on pump inhibi o (PPI) [4]. Howe e , he e icacy o iple he apy is in decline, mos ly due o H. pylo i esis ance o CLA, con ibu ing o he inc easing bu den o mul id ug- esis an G am- nega i e in ec ion. This global an ibio ic c isis was ecognized by he Wo ld Heal h O ganiza ion (WHO), who in 2017 published a lis o an ibio ic- esis an p io i y pa hogens o esea ch and de elopmen o new an ibio ics, including CLA- esis an H. pylo i [5]. CLA is a mac olide de i ed om e y h omycin, whose bac e ios a ic ac i i y depends on i s abili y o inhibi he bac e ial p o ein syn hesis [6]. P o ein syn hesis is c i ical o li e and is pe o med in a e y old nanomachine, he ibo- some. This nanomachine a ose om he molecula e olu ion ha p eda ed he i s li ing cells [7, 8]. Ribosomes ha e wo subuni s, he la ge and he small, made o RNA and p o eins, p esen ing basic machine y o syn he ize polypep ides: bo h subuni s join, emb acing an mRNA molecule, a e ecog- ni ion o a s a e sequence, and bo h ha e a simila si e o ecei e/dock he RNA, and inally a simila exi ing unnel o he nascen p o ein chain [9]. Bu , as hey a e e y old, enough ime has passed o he e olu ion o mechanisms and compounds o s all his c ucial nanomachine. O ganisms who had ha abili y in s a a ion imes could s op o he s om consuming esou ces ha a e undamen al o hei 2 Ma ques e al., Mic obial Genomics 2020;6 own su i al. Acco dingly, he e a e se e al molecules ha a ge di e en pa s o he ibosome, all o hem p oduced by bac e ia, such as chlo amphenicol, e acyclines, s ep o- mycin and e y h omycin [10]. E y h omycin is a na u al an ibio ic p oduced by Saccha- opolyspo a e y h aea ( o me ly S ep omyces e y h aeus) whose o iginal s ain p oduced ou ypes o molecules: e y h omycin A, B, C and D [11]. E y h omycin A is a mac olide (a mac ocyclic lac one wi h a ing o 12 o mo e membe s de i ed om a polyke ide) ha inhibi s bac e ial p o ein syn hesis by in e e ing wi h aminoacyl ansloca- ion, p e en ing he ans e o he RNA bound a he A si e o he RNA complex o he P si e o he RNA complex (Fig.1a) [12]. E y h omycin may be bac e ios a ic o bac e- icidal, depending on he o ganism and d ug concen a ion. Howe e , he ac ion o mac olides is no simple. A ecen e iew poin s o o he impo an mac olide ac ions, such as being modula o s o pep ide bond o ma ion, no only ibo- some unnel plugs [10]. This e iew is abou H. pylo i esis ance o a small man- made al e a ion o e y h omycin: he 6- O- me hyl e he o e y h- omycin A, named cla i h omycin (CLA) (Fig.1B). CLA inhibi s bac e ial p o ein syn hesis by e e sibly binding o he 50S ibosomal subuni [6]. I s mechanism o ac ion is o bind o he pep idyl ans e ase loop o he V domain o 23S ibosomal RNA (23S RNA) gene, which esul s in s uc u al changes and dissocia ion o pep idyl- RNA om ibosome, in e e ing wi h nascen pep ide chain elonga ion [13]. Misuse o an ibio ics accele a es esis ance, namely when a ge ing ibosome, which is e y lexible and a mino con o ma ional change could be enough o a oid he ac ion o na u al o semi- syn he ic an ibio ics [14]. Bac e ial popula- ions ha e a a iabili y o RNA sequences ha assu es he su i al o some o i s membe s, which we will show using ou own esul s. cLa esis ance mechanism H. pylo i’s an imic obial esis ance is mainly acqui ed by poin mu a ions, which a e ansmi ed e ically by bina y ission, esul ing in a p og essi e inc ease in he esis ance a e due o selec ion p essu e. Se e al PCR- based s udies ha e demons a ed ha poin mu a ions in he pep idyl ans e ase loop o he V domain o 23S RNA gene a e esponsible o he CLA esis - ance pheno ype in clinical H. pylo i s ains om a ious geog aphical loca ions [15–20]. These mu a ions a e able o dis up he pep idyl ans e ase loop con o ma ion and inhibi he binding be ween CLA and he 23S RNA, educing i s e iciency and leading o a esis ance pheno ype [18]. Two copies o he 23S RNA ope on a e p esen in he H. pylo i genome [19, 21] and, o mos H. pylo i s ains, mu a- ions a e gene ally ound in bo h copies; ne e heless, a he e ozygo e pheno ype is su icien o con e in e media e esis ance o CLA [16, 22]. Resis an s ains can be di ided in o wo g oups: a high le el o esis ance (MIC >64 mg l−1) and a low le el o esis ance (0.5≤MIC≤1 mg l−1). The mos p e alen and well- documen ed mu a ions in H. pylo i occu in wo speci ic adjacen nucleo ide posi ions, an adenine- o- guanine ansi ion a ei he posi ion 2142 (A2142G) o 2143 (A2143G), o , less equen ly, an adenine- o- cy osine ans e sion a posi ion 2142 (A2142C), and hese mu a ional e en s a e esponsible o mo e han 90 % o CLA esis ance in de eloped coun ies [23]. In pa icula , mu a ion a posi ion 2143 is usually associa ed wi h di e en le els o esis ance (MICs anging om 2 o 256 mg l −1 ), while s ains wi h mu a ion a posi ion 2142 equen ly exhibi a mo e es ic ed esis ance (MIC o 64 mg l −1 ). These wo mu a- ions (2142 and 2143) we e o iginally desc ibed as posi ions 2058 and 2059, based on Esche ichia coli 23S RNA sequence [15, 16, 18, 24] and la e changed o 2143 and 2144 acco ding o he H. pylo i 23S RNA sequence GenBank U27270 [20]. A e Taylo e al. de e mined he DNA sequences o he wo copies o he 23S RNA gene om H. pylo i s ain UA802 and compa ed he sequences om CLA- esis an s ains, hey p oposed ha he posi ions associa ed wi h CLA esis ance we e nucleo ides 2142 and 2143 [19] and mos in es iga o s choose o use his nomencla u e. Fu he mo e, i has been epo ed ha o he mu a ions a e able o con e CLA esis - ance, including mu a ions A2115G, G2141A, A2144T and T2289C [25–28], whils C2694A and T2717C ha e been associa ed wi h low esis ance le els [29, 30]. Fig.2 illus a es poin mu a ions ha con e CLA esis ance in he model o he 23S RNA domains V and VI. The con e sion T2182C is one con o e sial mu a ion and has been epo ed as no being equi ed o CLA esis ance, as well as con e ing low- le el o high- le el esis ance, wi h an MIC >64 mg l −1 [30–33]. O he mu a ions ha e been desc ibed in he li e a u e (i.e. G1939A, C2147G, G2172T, T2215C and C2245T), bu hei ole in ailu e o CLA- based he apy is s ill no p o en o is no being consis en ly epo ed [34–38]. Table2 summa izes a comple e lis wi h desc ibed single and combined poin mu a ions con e ing CLA esis ance (sea ch ca ied ou in Impac S a emen An ibio ic- esis an Helicobac e pylo i s ains a e inc easing in p e alence, as acknowledged ecen ly by he Wo ld Heal h O ganiza ion when H. pylo i was included in a lis o bac e ia ha pose he g ea es h ea o human heal h. We e iewed he mechanisms o esis - ance o cla i h omycin, p oducing a comple e compen- dium o all mu a ions ound in he 23S RNA gene ha ha e been associa ed wi h esis ance o his mac olide an ibio ic. Addi ionally, we desc ibed he pheno ypic and geno ypic me hods used o de ec ion o cla i h omycin esis ance, placing pa icula emphasis on he use ul- ness o whole- genome sequencing in de ec ing he esis ance and p og ession owa ds esis ance ound in mino i y genomic a ian s. 3 Ma ques e al., Mic obial Genomics 2020;6 Table 1. Associa ion be ween poin mu a ions in he 23S RNA and he cla i h omycin- esis an pheno ype o 17 pai s o H. pylo i isola es om he an um and co pus Genome Pheno ype Posi ion o he mu a ion Cla i h omycin MIC (mg l−1) 10 087A Resis an A2143G 8 10087C Resis an A2143G 8 10 103A Resis an A2142C, G2212A >256 10103C Resis an A2142C, G2212A >256 10 120A Resis an A2143G, C2759T 12 10120C Resis an A2143G, C2759T 12 10 198A Resis an A2143G, C2772T 24 10198C Resis an A2143G, C2772T 24 10 211A Resis an A2143G 12 10211C Resis an A2143G 12 10 212A Resis an A2143G 2 10212C Resis an A2143G 2 10 104A Suscep ible – – 10104C Suscep ible – – 10 127A Suscep ible T2182C – 10127C Suscep ible T2182C – 10 128A Suscep ible – – 10128C Suscep ible – – 10 133A Suscep ible – – 10133C Suscep ible – – 10 144A Suscep ible – – 10144C Suscep ible – – 10 147A Suscep ible* – – 10147C Suscep ible* – – 10 201A Suscep ible – – 10201C Suscep ible – – 10 210A Suscep ible – – 10210C Suscep ible – – 10 215A Suscep ible – – 10215C Suscep ible – – 10 218A Suscep ible – – 10218C Suscep ible – – 10 222A Suscep ible – – 10222C Suscep ible – – *Wi h esis an clones. A, an um isola e; C, co pus isola e. Fig. 1. Two- dimensional chemical s uc u e o e y h omycin A (a) and CLA (b). CLA is he 6- O- me hyl e he o e y h omycin A. S uc u es om PubChem (accessed Augus 2019) [67]. July 2019, using he keywo ds CLA esis ance, H. pylo i, 23S RNA and e lux pumps, and a ‘snowball’ sea ch – pu suing e e ences o e e ences). o he esis ance mechanisms Ano he possible mechanism o CLA esis ance is mul id ug e lux pump sys ems. E lux o an imic obial compounds is commonly obse ed in bac e ia, educing in acellula an imic obial concen a ion [39, 40]. E lux pumps o he esis ance- nodula ion- cell di ision (RND) amily a e espon- sible o mac olide esis ance in G am- nega i e bac e ia and his mechanism has also been p oposed o H. pylo i [41, 42]. Th ee pu a i e ope ons ha e been desc ibed as an RND e lux sys em in H. pylo i: he ABC (hp0605–hp0607), he DEF (hp0969–hp0971, ecen ly denomina ed cznABC) and he GHI (hp1327–hp1329, ecen ly denomina ed czcAB- c dB), wi h he he ABC ope on being mos simila o mul id ug e lux pumps. The he A, he D and he G genes encode an ou e - memb ane p o ein TolC homologue o E. coli, while he B/he C, he E/he F and he H/he I genes a e homologues o he ac A/ac B genes, encoding a memb ane usion and RND cy oplasmic pump p o eins, espec i ely [39, 43, 44]. A ou h gene clus e including ORF hp1489–1487 was desc ibed, whe e hp1489 is a TolC homologue and hp1488 shows simila i ies wi h he ac A gene [45]. In H. pylo i a syne gis ic e ec be ween 23S RNA mu a ions and e lux pumps is likely o be p esen in esis an s ains; he o me lowe s he CLA a ini y o he ibosome and he la e exc e es he an ibio ic. In some CLA- esis an s ains p esen ing 23S RNA mu a ions, i was shown ha he p esence o e lux pumps inhibi o s (EPIs) was able o dec ease he MIC o mos o he s udied s ains by ou old. The EPIs lead o an augmen ed in acellula CLA concen a- ion, which binds o he ibosome e en in he p esence o mu a ions. Howe e , in hese cases, he inal MICs we e s ill in he esis an ange [46]. Examina ion o he gene ic a i- an s o hese ou e lux pumps e ealed ha CLA esis an 4 Ma ques e al., Mic obial Genomics 2020;6 Fig. 2. Seconda y s uc u e model o he pep idyl ans e ase cen e in he domain V (g een) and domain VI (blue) o he H. pylo i 23S RNA. The indica ed poin mu a ion posi ions co espond o single mu a ions (no equi ing combina ion wi h o he mu a ions) con e ing low- le el o high- le el CLA esis ance. Mu a ions 2141 (G2141A), 2144 (A2144T), 2182 (T2182C), 2190 (T2190C), 2195 (C2195T), 2244 (T2244C), 2694 (C2694A) and 2711 (T2711C) a e indica ed by black ci cles and he mos p e alen mu a ions 2142 (A2142G) and 2143 (A2143G) a e indica ed by ed ci cles. Image a ailable a (and modi ied om) h p://www. na.icmb.u exas.edu (accessed July 2019) [68]). The ci cled nucleo ides indica e he posi ions o mu a ions ha con e CLA esis ance in H. pylo i (de ails and e e ences a e gi en in Table2). s ains a e mo e p one o single- nucleo ide a ian s in all ou clus e s o e lux genes, wi h signi ican di e ences o clus e he ABC [47]. Al hough he exac unde lying mechanism is unclea , i may in ol e an ibio ic e lux. To he bes o ou knowledge, he e lux pumps alone (ei he by mu an a ian s o by di e en ial exp ession p o ile) ha e no been associa ed wi h H. pylo i CLA esis ance. O he no el candida es o CLA esis ance likely p esen a syne gis ic e ec wi h 23S RNA poin mu a ions. Indeed, a e he exposu e o a suscep ible s ain o low doses o CLA, mu a ions in genes in B ( ansla ion ini ia ion ac o IF-2) and pl22 ( ibosomal p o ein L22) con e ed low- le el esis ance o CLA (low MIC alue), while mu a ions in hese genes plus mu a ions in 23S RNA inc eased he MIC alue. Addi ionally, compa a i e p o eomics analysis highligh ed he possible in ol emen o ou e - memb ane p o eins in CLA esis ance. Resis an s ains in compa ison o suscep ible s ains p esen ed up egula ed U easeB subuni and EF- Tu (elonga ion ac o he mo uns able) and down egula ed Ho C (e lux pump) and OMP31 [48]. Finally, he spoT [bi unc- ional (p)ppGpp syn hase] gene is in ol ed in ole ance o CLA, up egula ing anspo e genes (HP0939, HP1017, HP0497 and HP0471) [49]. O e all, he mechanisms besides mu a ions in 23S RNA appea o po en ia e he e ec o ibo- somal mu a ions by in e e ing ei he wi h o he ansla ion machine y o wi h an ibio ic anspo , educing i s in acel- lula concen a ion. De ec ion me hods An imic obial suscep ibili y es ing should be pe o med whene e possible o guide he apy selec ion. Fo CLA, o which he in i o esis ance is p edic i e o unsuccess ul ea men s, his es assumes pa icula ele ance in he managemen o H. pylo i in ec ion, especially in high p ima y esis ance egions [50]. Pheno ypic and geno ypic me hods can be used o es suscep ibili y o CLA. Se e al pheno ypic 5 Ma ques e al., Mic obial Genomics 2020;6 Table 2. Poin mu a ions in he 23S RNA egion and MIC alues associa ed wi h CLA- esis an H. pylo i s ains Posi ion* Mu a ion MIC (mg l−1)Re e ences 1592† A1592G+T2182C 1 [69] A1592G+C2195T 2 [69] A1592G + T1644C + A1821G + G1826A + T1830C+T2182C 1 [69] 1592† A1592T + A1821G + G1826A + T1830C+T2182C 4 [69] 1652† A1652G 2 [69] 1694† G1694A + A1738G + C1953T 4 [69] 1738† A1738G + G1826C+C2195T 1 [69] 1769† C1769T 1 [69] 1821 A1821G+T2182C 1 [31] A1821G + G1826A + T1830C+T2182C 4 [31] 1826† G1826A+A2143G 4 [69] 1939 G1939A + T1942C+A2142G+C2147G 1 [35] G1939A+A2306G 0.064 o 0.5 [34] 1944 C1944T+G2212A 0.064 o 0.5 [34] 1953 C1953T+A2143G+T2182C+A2223G+T2244C 1.5 o 256 [70] 2115 A2115G 1.24 o 4 [25, 26] A2115G+A2141G 4 [22] A2115G+A2142G 4 [26] A2115G+A2143G 4 [26] A2115G+A2144T 4 [26] 2141 G2141A 1.25 o 28 [25–27] 2142 A2142C 64 o >256 [15, 27, 71–73] A2142C+A2142G na [73] A2142C+A2143G na [71] A2142C+G2212A >256 This wo k A2142C+A2142G+A2143G na [71] 2142 A2142G 0.1 o 256 [16, 22, 26, 27, 35, 38, 72–76] A2142G+A2143G 4 [26, 71, 73] A2142G+A2144T na [71] A2142G+T2182C 1 o 256 [76, 77] A2142G+C2195T ≤32 [74] A2142G+A2223G >32 [74] A2142G+A2143G+A2144T 256 [26] A2142G+T2182C+C2195T >32 [74] Con inued 6 Ma ques e al., Mic obial Genomics 2020;6 Posi ion* Mu a ion MIC (mg l−1)Re e ences 2143 A2143G 0.1 o 256 [16, 22, 26, 27, 32, 35, 38, 71–76, 78] A2143G+A2174G ≤32 [74] A2143G+T2182C 0.5 o 45 [36, 74, 76–78] A2143G+A2223G ≤32 [74] A2143G+G2224A 4 [27] A2143G+T2244C 3 o 48 [70] A2143G+C2245T 16 [27] A2143G+C2759T 12 This wo k A2143G+C2772T 24 This wo k A2143G+T2182C+T2190C 52 [78] A2143G+T2182C+C2195T 53.3 [78] A2143G+T2182C+A2223G 80 [78] A2143G+T2182C+T2244C 1.5 o 96 [70] A2143G+C2195T+A2223G ≤32 [74] A2143G+A2223G+T2244C 1.5 o 96 [70] A2143G+T2182C+A2223G+T2244C 1.5 o 256 [70] A2143G+T2182C+T2244C+A2302G 1.5 o 256 [70] 2143 A2143T+T2182C+G2172T+G2254T na [36] 2144 A2144T 3 o 134 [25–27] 2182 T2182C 1 o >64 [31, 32, 72, 74, 76, 78] T2182C+T2244C 3 o 48 [70] T2182C+T2190C+C2694A >0.5 [30] 2190 T2190C na [79] 2195 C2195T na [79] C2195T+T2182C+A2223G+T2244C 1.5 o 256 [70] 2224 G2224A+T2289C 64 [28] G2224A+C2245T+T2289C >256 [28] 2244‡ T2244C 1.5 o 5 [70, 80] 2289 T2289C 8 [28] 2694 C2694A 1 [30] 2711§ T2711C 0.5 o 1 [29] na, no a ailable. *Consolida ed posi ion acco ding he nomencla u e published by Taylo e al.[19] (1997), which co esponds o he e e ence sequence H. pylo i 23S RNA gene (GenBank: U27270.1) nucleo ide. †Ma a e al. [69] desc ibe hese posi ions as 1593, 1653, 1695, 1739, 1770 and 1827, espec i ely ( e e ence sequence GenBank: U27270.1). ‡Khademi e al. [80] desc ibe his posi ion as 2243. §Fon ana e al. [29] desc ibe his posi ion as 2717. Table 2. Con inued 7 Ma ques e al., Mic obial Genomics 2020;6 Table 3. Geno ypic me hods used o de ec mac olide esis ance in H. pylo i Based on 23S RNA gene Me hod Re e ence Wi h ampli ica ion RFLP [55] PCR ollowed by e e se hyb idiza ion [57, 58] Oligonucleo ide liga ion assay [81] DNA enzyme immunoassay [82, 83] P e e en ial homoduplex o ma ion assay [84] Real- ime PCR [56, 85] 3’-misma ched e e se p ime PCR [86] Mic oelec onic chip a ay [87] Dual- p iming oligonucleo ide‐based mul iplex PCR [88] Mic oa ay [89] D ople digi al PCR [90] Wi hou ampli ica ion Pep ide nucleic acid- luo escence in si u hyb idiza ion [59, 91] Nex - gene a ion sequencing [92] me hods ha e been de eloped, such as he aga dilu ion me hod, which is conside ed o be he e e ence me hod in compa ison o o he echniques. The MIC b eakpoin o CLA, which is based on epidemiological cu - o alues, is 0.25 mg l−1 o suscep ible and 0.5 mg l−1 o esis an s ains [51]. This me hod is e y ime- consuming and is a ely pe o med in ou ine labo a o ies. The E- es me hod, based on g adien di usion, wi h he abili y o p oduce an MIC esul , is cu en ly he me hod o choice in mos o he clinical labo a o ies pe o ming an imic obial suscep ibili y es ing o H. pylo i, since is adap ed o slow- g owing bac e ia. A good co ela ion has been ound be ween his me hod and he aga dilu ion me hod [51]. B eakpoin suscep ibili y es ing is a simpli ied e sion o he aga dilu ion me hod. I consis s o inocula ing a line o he s ain o be es ed on an aga pla e con aining an an ibio ic concen a ion equal o he b eakpoin concen a ion ha de ines esis ance. The b o h dilu ion me hod is seldom used due o he di icul y o g owing H. pylo i in b o h media. Howe e , i is possible o use i and ob ain accep able MIC esul s i he b o h is supple- men ed wi h se um o de ib ina ed blood [52–54]. Finally, he simples and mos economical me hod o ou ine suscep i- bili y es ing is he disk di usion me hod, which is gene ally no ecommended o slow- g owing bac e ia, al hough i has been alida ed o de ec mac olide esis ance accu a ely in H. pylo i, while e y h omycin is he ecommended an ibio ic o mac olide suscep ibili y es ing. In H. pylo i, gi en he low numbe o ch omosomal poin mu a ions con e ing CLA esis ance (Table 2), accu a e geno ypic me hods ha e de eloped exponen ially. These es s a e as e han he pheno ypic me hods, and a e easy o es ablish in ou ine p ac ice. The mos common is based on ampli ica ion o he 23S RNA by PCR, ollowed by di e en de ec ion me hods. PCR- RFLP was one o he i s me hods o be de eloped, making use o h ee di e en es ic ion enzymes, one o each o he h ee mos common poin mu a- ions (A2142/3G and A2142C) [55]. This labo ious me hod, based on a p o ile band esol ed in an aga ose gel, was apidly eplaced by as e and mo e sensi i e me hods, such as eal- ime PCR, which allows he de ec ion o H. pylo i as well as CLA esis ance- associa ed poin mu a ions in a single eac- ion. This es is usually based on a bip obe and luo escence esonance ene gy ans e (FRET), allowing easy disc imina- ion o he h ee mu a ions by mel ing cu e analysis, wi h he po en ial o being used di ec ly on gas ic biopsies o on s ool specimens, inc easing he sensi i i y o he me hod o de ec mixed suscep ible and esis an popula ions in a single sample [56]. O he PCR- based me hods can a y in hei de ec ion me hod, o example combining mul iplex con en ional PCR and amplicon de ec ion ia a e e se hyb idiza ion and alkaline phospha ase eac ion on a memb ane s ip coa ed wi h highly speci ic p obes complemen a y o he selec i ely ampli ied nucleic acid sequences [57, 58]. Using PCR- based molecula me hods on non- in asi e samples such as s ools makes suscep ibili y es ing easie and mo e p ac ical o pe o m, and he e o e such me hods a e widely a ailable. Geno ypic me hods no in ol ing DNA ampli ica ion ha e also been de eloped, using luo escence in si u hyb idiza ion (FISH), which is based on p obes ha hyb idize wi h speci ic RNA sequences o mic o- o ganisms. The bes pe o mance is achie ed using pep ide nucleic acid p obes, which a e usually smalle han ypical DNA p obes, inc easing hei abili y o pene a e he bac e ial cell wall, and a e mo e esis an o deg ada ion by nucleases and p o eases [59]. Finally, he de elopmen o high- h oughpu nex - gene a ion sequencing (NGS) me hods has allowed he use o whole- genome sequencing as a genome- based yping me hod, addi ionally enabling an ibio ic esis ance de e minan s o be in e ed. Table3 p esen s an o e iew o some o he geno- ypic me hods used o de ec mac olide esis ance in H. pylo i. NGS o de ec ing esis ance: an example As p e iously s a ed, i has been epo ed ha o he mu a- ions a e able o con e CLA esis ance, bu hei ole is s ill con o e sial. We he e o e used a se o an um and co pus pai s o H. pylo i clinical isola es in o de o cla i y he ole o hese less common mu a ions, as well as he powe o NGS o de ec esis ance. The genomes o 17 pai s o isola es om a pool o colonies om he an um and co pus o a paedia ic popula ion we e sequenced, Illumina MiSeq assembled wi h SPAdes3.13 [60] and analysed. An imic obial suscep ibili y es ing was pe o med by disk di usion o e y h omycin, and he E- es was used o de e mine he MIC o CLA o he esis an s ains. Six pai s o isola es we e CLA- esis an (12/34 isola es) and a o al o six poin mu a ions we e de ec ed (Table 1). All pai s o CLA- esis an isola es o 8 Ma ques e al., Mic obial Genomics 2020;6 Table 4. Genomes o Helicobac e pylo i s ains suscep ible o CLA p esen ing a ian s ma ching desc ibed mu a ions in 23S RNA o CLA esis ance Genome Posi ion Change Va ian equency (%) Co e age 10 222A 2142 A -> G 6.3 191 10222C 2142 A -> G 6.9 275 10 147A 2142 * A -> G 2.5 204 2301 A -> G 4.7 235 10147C 2142 A -> G 12.9 240 2301 A -> G 12.0 259 *Al hough ewe han 10 eads we e obse ed wi h mu a ion A -> G, he esul is p esen ed because o he consis ency be ween an um and co pus isola es. an um and co pus o he same pa ien displayed he same mu a ions. All hese isola es ha e mu a ions a posi ions 2142 and 2143, wi h A2143G (83.3 % o isola es, 10/12) being p edominan , ollowed by A2142C (16.7 % o isola es, 2/12). In de ail, h ee pa ien s had he single mu a ion A2143G o bo h an um and co pus isola es (6/12 isola es), and h ee pa ien s had double mu a ions, A2142C+G2212A o one case (2/12 isola es), A2143G+C2759T o he second case (2/12 isola es) and A2143G+C2772T o he las case (2/12 isola es). The MICs o CLA a ied be ween 2 and 24 mg l−1 o isola es ha bou ing he mu a ion A2143G, while he MIC was >256 mg l−1 o he wo isola es om he same pa ien wi h he mu a ion A2142C (Table1). These alues a e in he ange o hose desc ibed ex ensi ely in he li e a u e o hese mu a- ions, making i di icul o p edic he ole o he emaining mu a ions ound simul aneously. Two addi ional mu a ions ound in combina ion wi h A2143G, C2759T and C2772T, bo h loca ed a 23S RNA domain VI, a e epo ed o he i s ime (Fig.2 and Table2). A T2182C mu a ion was ound in wo suscep ible isola es (an um and co pus om same pa ien ), which is in ag eemen wi h i s con o e sial ole in CLA esis ance [30–33]. Based on he esul s om ou se ing, we conclude ha he esis ance pheno ype is ela ed o he p esence o he well- desc ibed 2142 and 2143 poin mu a ions, while he p esence o he o he epo ed mu a ions pe se canno con e a esis - ance pheno ype, co obo a ing p e ious s udies. Rega ding he p edic ion o esis ance le els, we canno make a conclu- si e s a emen abou he ole o hese less common o new mu a ions, since he MICs desc ibed a e all in he ange o he p e iously desc ibed alues o isola es ha bou ing he well- desc ibed 2142 and 2143 poin mu a ions. Ne e heless, his example o he applica ion o NGS o CLA esis ance de e mina ion shows he powe o genome sequencing o co ec ly de ec ing esis ance in clinical s ains. In ac , all o he cases wi h MIC alues ha we e compa ible wi h esis ance o CLA p esen ed he mos common mu a ions in he 23S RNA gene associa ed wi h esis ance (Table1). The con inuous educ ion o he cos o genome sequencing, coupled wi h he mul i- asking pe o mance o NGS (e.g. i is also use ul o molecula yping, popula ion gene ics o epidemiology), makes his echnique a s ong candida e o in oduc ion in o clinical labo a o ies o he es ing o an i- bio ic esis ance. Genomic a iabili y o he 23S RNa gene: genome-based p edic ion o e olu ion owa ds cLa esis ance H. pylo i genomes a e highly di e se and no in equen ly s ains in ec ing he same hos a e gene ically di e en i- a ed [1, 61]. Mo eo e , H. pylo i genomes e ol e apidly du ing ch onic in ec ion [62, 63] and labo a o y cul u e [64], e idencing i s ema kable adap i e capaci y o i s niche. The genomes a ailable o each bac e ial s ain a e ypically he canonical genomes, i.e. he mos equen esi- dues o nucleo ides a each posi ion. The a ia ion wi hin a bac e ial genome is commonly dis ega ded so ha a consensus wo king sequence is a ailable and also due o he di icul y o asce aining i i ep esen s a ue a ia ion o a sequencing e o [64]. Impo an ly, he 23S RNA gene is p esen in wo copies in he H. pylo i genome, which due o hei simila i y a e assembled in he same locus. In ac , a common cause o he unde es ima ion o gene numbe is he collapse o gene copies in o a single locus ha due o highly simila i y in sequence a e challenging o assemble as sepa a e loci [65]. To unde s and he di e - si y wi hin H. pylo i clinical isola es, we ha e wo ked wi h he same se o H. pylo i s ains and looked o genomic a ian s in he 23S RNA gene p esen ing a co e age le el >100×, and a leas 10 copies o he a ian . Sequence eads we e mapped o a canonical genome and a BAM ile was p oduced using SAM ools [66]. Va ian calling o e BAM iles was pe o med wi h Geneious 8.1.9, selec ing a ian s a a equency a ian >1 % and maximum a ian P- alue 10E-6. Two ou o 11 pai s o s ains (18.1%) pheno ypically suscep ible o CLA p esen ed a ian s ma ching mu a ions associa ed wi h CLA esis ance (Table 4). Exposu e o CLA may selec low- equency a ian s ha lead o he a- peu ic ailu e and he eme gence o a esis an s ain. This appea ed o be he case o he pai 10 147A/10147C, which is suscep ible o CLA bu p esen ed a ew colonies g owing a high CLA concen a ion (Tables1 and 4), sugges ing a geno ype e ol ing owa ds CLA esis ance. Fo one o he pai s o esis an isola es (10 103A and 10103C), a genomic a ia ion was also obse able: mos o he eads had he mu a ion A2142C, bu he mu a ion A2142G was also ound wi h a equency o 28.5 and 28.2 %, espec i ely (Table1). Bo h mu a ions a e associa ed wi h CLA esis ance. High- dep h whole- genome sequencing is hus a powe ul echnique no only o de e mine he mu a ion associa ed wi h CLA esis ance, bu also o p edic he e olu ion in e ed by a ian call analysis, examining he dep h o eads mapped o each nucleo ide a ia ion posi ion known o be associa ed wi h esis ance. In addi ion, al hough he co ela ion be ween CLA esis ance de ec ed by pheno ypic 9 Ma ques e al., Mic obial Genomics 2020;6 and geno ypic me hods is in gene al good, he la e is much mo e accu a e in de ec ing low numbe s o mu a ed bac e- ial cells wi hin a pool o o he wise suscep ible bac e ia. Acco ding o ou da a, he genomic a iabili y showed ha abou one i h o he suscep ible isola es coming om a pool o colonies is likely o ha e expe ienced mic oe o- lu ion e en s, i.e. sha e a common ances o and p esen dis inc genomic al e a ions wi hin pa ien . In hese cases, CLA may ac as a selec i e agen o he mino i y and esis an a ian s. The e o e, he in oduc ion o NGS o CLA esis ance es ing is addi ionally impo an as in addi ion o iden i ying s ains ha a e esis an , i may also con ibu e o he iden i ica ion o a ian s in which he esis an geno ype is in mino i y, a oiding he selec ion o his esis an subpopula ion. conclusion The esis ance mechanism o CLA is mainly insc ibed on h ee mu a ions ound on he 23S RNA gene, impai ing a ge ecogni ion by he an ibio ic. In his s udy, mu a- ions A2142C and A2143G we e also de ec ed in he 23S RNA gene V domain o CLA- esis an H. pylo i. Mu a ion T2182C was ound in suscep ible isola es. Se e al me hods allow he de ec ion o CLA esis ance. NGS no only allows us o e i y he p esence o mu a ions in canonical genomes associa ed wi h esis ance, bu is also use ul o asce ain e olu ion owa ds esis ance e idenced by genomic a i- a ion ma ching known mu a ions associa ed wi h CLA esis ance. The e will be always an ibio ic esis ance o old and new molecules, because ha is how bac e ia had su i ed o millions o yea s. The e o e, he s a egy o o e come CLA esis ance should be o con inue o s udy H. pylo i biology o ind new a ge s o elimina e he bac e ia and o design new molecules. Funding in o ma ion F. F. V. is he ecipien o a p ojec g an (PTDC/BTM- SAL/28978/2017) om he Fundação pa a a Ciência e a Tecnologia (FCT), which suppo ed his wo k. J. V.’s esea ch g oup was inanced by New England Biolabs, Inc. (USA). Acknowledgemen s Whole- genome sequencing was pe o med a Unidade de Tecnologia e Ino ação (Depa amen o de Gené ica Humana, Ins i u o Nacional de Saúde Dou o Rica do Jo ge, Lisbon, Po ugal). Con lic s o in e es The au ho s decla e ha he e a e no con lic s o in e es . Da a Bibliog aphy 1. Na ional Ins i u e o Heal h, Po ugal. Genbank, SRR9930173– SRR9930179 (2020). Re e ences 1. Még aud F, Lehou s P, Vale FF. The his o y o Helicobac e pylo i : om phylogeog aphy o paleomic obiology. Clin Mic obiol In ec 2016;22:922–927. 2. Vale FF, Oleas o M. O e iew o he phy omedicine app oaches agains Helicobac e pylo i. WJG 2014;20:5594. 3. Még aud F. Cu en ecommenda ions o Helicobac e pylo i he apies in a wo ld o e ol ing esis ance. Gu Mic obes 2013;4:541–548. 4. Mal e heine P, Meg aud F, O'Mo ain C, Hungin APS, Jones R e al. Cu en concep s in he managemen o Helicobac e pylo i in ec- ion- he Maas ich 2-2000 consensus epo . Alimen Pha macol The 2002;16:167–180. 5. WHO. Global p io i y lis o an ibio ic- esis an bac e ia o guide esea ch, disco e y, and de elopmen o new an ibio ics 2017. 6. Goldman RC, Zakula D, Flamm R, Beye J, Capobianco J. Tigh binding o cla i h omycin, i s 14- (R)- hyd oxy me aboli e, and e y h- omycin o Helicobac e pylo i ibosomes. An imic ob Agen s Chem- o he 1994;38:1496–1500. 7. Robe son MP, Joyce GF. The o igins o he RNA wo ld. Cold Sp ing Ha b Pe spec Biol 2012;4:a003608. 8. WSV Y, Vincen NG, Base ga SJ. Ribonucleop o eins in a chaeal p e- RNA p ocessing and modi ica ion. A chaea 2013;2013:614735. 9. A enz S, Wilson DN. Bac e ial p o ein syn hesis as a a ge o an i- bio ic inhibi ion. Cold Sp ing Ha b Pe spec Med 2016;6:a025361. 10. Vázquez- Laslop N, Mankin AS. How mac olide an ibio ics wo k. T ends Biochem Sci 2018;43:668–684. 11. Kibwage IO, Hoogma ens J, Roe s E, Vande haeghe H, Ve bis L e  al. An ibac e ial ac i i ies o e y h omycins a, B, C, and D and some o hei de i a i es. An imic ob Agen s Chemo he 1985;28:630–633. 12. E molenko DN, Co nish PV, Ha T, Nolle HF. An ibio ics ha bind o he a si e o he la ge ibosomal subuni can induce mRNA anslo- ca ion. RNA 2013;19:158–166. 13. Menninge JR. Func ional consequences o binding mac olides o ibosomes. J An imic ob Chemo he 1985;16 Suppl A:23–34. 14. Ves e B, Dou hwai e S. Mac olide esis ance con e ed by base subs i u ions in 23S RNA. An imic ob Agen s Chemo he 2001;45:1–12. 15. S one GG, Sho idge D, Flamm RK, Ve salo ic J, Beye J e  al. Iden i ica ion o a 23S RNA gene mu a ion in cla i h omycin- esis an Helicobac e pylo i. Helicobac e 1996;1:227–228. 16. Ve salo ic J, Sho idge D, Kible K, G i y MV, Beye J e al. Mu a- ions in 23S RNA a e associa ed wi h cla i h omycin esis ance in Helicobac e pylo i. An imic ob Agen s Chemo he 1996;40:477–480. 17. Vianna JS, Ramis IB, Ramos DF, VON G oll A, Sil a PEAda. D ug esis ance in Helicobac e pylo i. A q Gas oen e ol 2016;53:215–223. 18. Occhialini A, U daci M, Douce - Populai e F, Bébéa CM, Lamouli- a e H e  al. Mac olide esis ance in Helicobac e pylo i: apid de ec ion o poin mu a ions and assays o mac olide binding o ibosomes. An imic ob Agen s Chemo he 1997;41:2724–2728. 19. Taylo DE, Ge Z, Pu ych D, Lo T, Hi a suka K. Cloning and sequence analysis o wo copies o a 23S RNA gene om Helicobac e pylo i and associa ion o cla i h omycin esis ance wi h 23S RNA mu a- ions. An imic ob Agen s Chemo he 1997;41:2621–2628. 20. Ve salo ic Je al. Poin mu a ions in he 23S RNA gene o Heli- cobac e pylo i associa ed wi h di e en le els o cla i h omycin esis ance. J An imic ob Chemo he 1997;40:283–286. 21. Bukano NO, Be g DE. O de ed cosmid lib a y and high- esolu ion physical?gene ic map o Helicobac e pylo i s ain NCTC11638. Mol Mic obiol 1994;11:509–523. 22. Hul én K, Gib eel A, Sköld O, Engs and L. Mac olide esis ance in Helicobac e pylo i: mechanism and s abili y in s ains om cla i h omycin- ea ed pa ien s. An imic ob Agen s Chemo he 1997;41:2550–2553. 23. Meg aud F. H pylo i an ibio ic esis ance: p e alence, impo ance, and ad ances in es ing. Gu 2004;53:1374–1384. 24. Wang G, Taylo DE. Si e- speci ic mu a ions in he 23S RNA gene o Helicobac e pylo i con e wo ypes o esis ance o mac olide- lincosamide- s ep og amin B an ibio ics. An imic ob Agen s Chem- o he 1998;42:1952–1958. 25. Kocazeybek B, Sakli MK, Yuksel P, Demi ci M, Caliskan R e  al. Compa ison o new and classical poin mu a ions associa ed wi h cla i h omycin esis ance in Helicobac e pylo i s ains isola ed om dyspep ic pa ien s and hei e ec s on pheno ypic cla i h o- mycin esis ance. J Med Mic obiol 2019;68:566–573.