J Antimicrob Chemother 2023; 78: 975–982 https://doi.org/10.1093/jac/dkad031 Advance Access publication 10 February 2023 Increasing trend of antimicrobial resistance in Shigella associated with MSM transmission in Barcelona, 2020–21: outbreak of XRD Shigella sonnei and dissemination of ESBL-producing Shigella flexneri Albert Moreno-Mingorance1,2, Alba Mir-Cros1,3, Lidia Goterris1,4, Virginia Rodriguez-Garrido1,4, Elena Sulleiro1,2,3,4, M. Jesús Barberà5, Mireia Alberny6, Yannick Hoyos-Mallecot1,4, Vicente Descalzo5, Albert Bravo4, Josep Roca-Grande1,2,3, Belén Viñado1,4, Tomàs Pumarola1,2,3,4, M. Nieves Larrosa1,2,3,4† and Juan José González-López 1,2,3,4*† 1 Vall d’Hebron Institut de Recerca (VHIR), Vall d’Hebron Hospital Universitari, Vall d’Hebron Barcelona Hospital Campus, Barcelona, Spain; 2 Department of Genetics and Microbiology, Universitat Autònoma de Barcelona, Bellaterra, Spain; 3CIBER de Enfermedades Infecciosas (CIBERINFEC), Instituto de Salud Carlos III, Madrid, Spain; 4Department of Clinical Microbiology, Vall d’Hebron Hospital Universitari, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain; 5Drassanes-Vall d’Hebron Sexually Transmitted Infections Unit, Vall d’Hebron Hospital Universitari, Barcelona, Spain; 6Primary Health-Care Division, Catalan Institute of Health, Barcelona, Spain *Corresponding author: E-mail: juanjo.gonzalez@vallhebron.cat †These senior authors contributed equally to this article. Received 6 September 2022; accepted 23 January 2023 Background: Several countries have recently reported the detection of ESBL-producing Shigella sonnei asso­ ciated with transmission among MSM. In a previous study by our group, 2.8% of Shigella spp. obtained from MSM in Barcelona between 2015 and 2019 were ESBL producers. Objectives: To describe and characterize the emerging ESBL-producing Shigella spp. associated with sexual transmission among MSM detected from 2020 to 2021 in Barcelona, elucidating their connectivity with contem­ poraneous ESBL-producing Shigella spp. from other countries. Results: From 2020 to 2021, we identified that among MSM, 68% of S. sonnei were XDR harbouring blaCTX-M-27 and 14% of Shigella flexneri were MDR harbouring blaCTX-M-27. WGS analysis showed that the ESBL-producing S. sonnei were part of a monophyletic cluster, which included isolates responsible for the prolonged outbreak oc­ curring in the UK. Our data also reveal the first emergence and clonal dissemination of ESBL-producing and fluoroquinolone-resistant S. flexneri 2a among MSM. Conclusions: We report an increasing trend of antimicrobial resistance in Shigella spp. among MSM in Barcelona since 2021, mainly as a consequence of the dissemination of XDR ESBL-producing S. sonnei, previously reported in the UK. These results highlight the importance of international collaborative surveillance of MDR/XDR S. sonnei and S. flexneri for rapid identification of their emergence and the prevention of the transmission of these pathogens. Introduction as a first/second-line treatments.7,8 During the last decade, an in­ crease in the number of cases of shigellosis in MSM, especially by In high-income countries, shigellosis is mainly found in travellers isolates resistant to oral treatments, has been documented. returning from high-risk regions and as a sexually transmitted in­ Recently, the emergence of ESBL-producing S. sonnei has been fection (STI) in MSM. Outbreaks of Shigella sonnei and Shigella reported in several countries.1,9–11 In a rapid risk assessment flexneri among MSM have been reported worldwide.1–6 released in February 2022, the ECDC raised the level of risk of According to international guidelines, the antibiotic therapy re­ shigellosis due to MDR S. sonnei for MSM to moderate, based on commended for shigellosis includes ciprofloxacin, azithromycin, the high probability of infection but the low impact of such trimethoprim/sulfamethoxazole (co-trimoxazole) or ceftriaxone infection.10 © The Author(s) 2023. Published by Oxford University Press on behalf of British Society for Antimicrobial Chemotherapy. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https:// creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 975 Increased AMR in MSM-associated Shigella spp., Barcelona In a previous study conducted by our group, we showed that pKSR100 (accession no. LN624486), p183660 and p893916 (accession no. among MSM-associated Shigella spp. collected in Barcelona from MW396858), which have previously been identified to encode antimicrobial 2015 to 2019 only 2.8% were ESBL producers as they carried resistance determinants in MSM-associated Shigella isolates using BLAST blaCTX-M-27.12 The aim of this study was to describe and character­ Ring Image Generator (BRIG) and Mauve.14,15 Maximum-likelihood phylogeny was carried out with Snippy (v4.3.6) ize the emerging ESBL-producing Shigella isolates associated with S. flexneri 2a strain 2457T (AE014073) and S. sonnei 53G with sexual transmission among MSM detected from 2020 to (NC_016822) as the reference strains, removing recombination events 2021 in Barcelona, elucidating their connectivity with contem­ with Gubbins (v2.3.4). Finally, the single-nucleotide polymorphism poraneous ESBL-producing Shigella spp. from other countries. (SNP)-based phylogenetic tree was constructed after 1000 bootstrap re­ plicates with the IQ-Tree (v1.6.10) and tree annotation with relevant metadata was performed with the interactive tree of life (iTOL) (http:// Methods itol.embl.de). The genome of representative Shigella spp. isolates previ­ ously studied by our group and obtained from MSM from Barcelona in Study population and bacterial isolates the period 2015–19 as well as the genome of international isolates ob­ From January 2020 to December 2021, 37 S. sonnei and 43 S. flexneri tained in previous studies were included to add context (Table S1, avail­ were obtained from stool cultures of patients with diarrhoeal disease able as Supplementary data at JAC Online). who attended primary healthcare units in Barcelona and different de­ partments of the Vall d’Hebron Hospital, including the Sexually Transmitted Infections Unit, the Tropical Medicine and International Data analysis Health Unit, and other medical departments. Duplicate isolates from Categorical variables were compared using the chi-squared test (χ2 test) the same individual were excluded. Patients’ medical records were re­ or Fisher’s exact test when appropriate. viewed to collect their sociodemographic characteristics and to identify if pathogen acquisition was related to a sexually transmitted infection Sequence data deposition among MSM or to other sources. Reads used for this study were deposited in the National Center for Biotechnology Information (NCBI) under the BioProject number Antibiotic susceptibility testing (AST) PRJNA864648. AST using selected antimicrobials was performed by disc diffusion or the gradient diffusion method (EtestTM, bioMérieux). MICs were interpreted Ethics according to the EUCAST clinical breakpoints, except for azithromycin for which epidemiological cut-off values were used (https://eucast.org/). The study was approved by the Ethics Committee of the Vall d’Hebron MDR isolates were defined as those with resistance to ≥3 antibiotic cat­ Hospital, reference number PR(AG)456/2020. The study was conducted egories and XDR as those with susceptibility to only one or two categories in accordance with the principles laid out in the Declaration of Helsinki of antibiotics.13 and in accordance with the principles of Good Clinical Practice. WGS and bioinformatic analysis Results WGS was performed in all MSM-associated S. flexneri and S. sonnei as well Bacterial isolates and antimicrobial resistance as all ESBL-producing Shigella obtained, regardless of the route of acqui­ sition. Short-read sequencing for each isolate was performed by Nextera From 2020 to 2021, 37 S. sonnei and 43 S. flexneri were obtained DNA Flex and MiSeq (Illumina) according to the manufacturer’s instruc­ from 80 patients in the Clinical Microbiology Department of the tions. Trimmomatic (v0.38), Unicycler (v0.4.7) and SPAdes (v3.14.1) Vall d’Hebron Hospital. Of these, 59.5% of S. sonnei (22/37) and were used to perform raw-read trimming and de novo genome assembly. 65.1% of S. flexneri (28/43) were considered to have been ac­ Shigatyper (v1.0.5), Mykrobe (v0.9.0) and Resistance Gene Identifier (RGI, quired through sexual contact among MSM (n = 50, median age v4.2.2) were used for in silico S. flexneri serotyping, S. sonnei genotyping 36.2 years, range 20–63) with no history of recent travel to and for the identification of acquired antimicrobial resistance genes and low- or middle-income countries. point mutations associated with antimicrobial resistance, respectively. AST showed that 29% of Shigella spp. were ESBL producers. The presence of the plasmid p183660 (accession no. KX008967), By species, 49% of S. sonnei (18/37) and 12% of S. flexneri which has previously been identified as simultaneously encoding anti­ microbial resistance determinants to third-generation cephalosporins (5/43) produced this resistance mechanism. By years, 33% of (3GCs) (blaCTX-M-27), macrolides [mph(A) and erm(B)], sulphonamides S. sonnei (5/15) were ESBL producers in 2020, and 59% in (sul1), trimethoprim (dfrA17) and aminoglycosides (aadA5) in Shigella 2021 (13/22). Regarding S. flexneri, no ESBL producers were was investigated by mapping the trimmed reads to the plasmid as refer­ identified in 2020; however, 23% of the isolates obtained in ence with Snippy (v4.3.6).14,15 2021 (5/22) possessed this resistance mechanism. According Based on the results from the short-read sequencing, representative to the source of acquisition, 83% (19/23) of ESBL-producing MSM Shigella isolates encoding blaCTX-M-27 and showing different resistance Shigella could be identified as acquired among MSM. The sus­ phenotypes were sequenced to characterize blaCTX-M-27-harbouring plas­ ceptibility to the remaining first-/second-line antimicrobials is mids by long-read sequencing using MinION technology (Oxford shown in Table 1. MDR was found in 95% and 14% of Nanopore Technologies). After DNA extraction using DNeasy PowerSoil MSM-associated S. sonnei and S. flexneri, respectively, while Pro Kit (QIAGEN, Hilden, Germany), libraries were prepared using Native Barcoding Kit 96 v14 and sequenced on R10.4.1 flow cell (Oxford XDR was only detected among S. sonnei as 68% were Nanopore Technologies). Basecalling was carried out with Guppy (v6.3.8) co-resistant to azithromycin, ciprofloxacin, co-trimoxazole in standard mode. After trimming with Filtlong (v0.2.1), a hybrid assembly and cephalosporins (Table 1). with short reads was performed with Unicycler (v0.4.8) and SPAdes Antimicrobial resistance gene content analysis revealed that (v3.14.1). Plasmids encoding blaCTX-M-27 were compared with the plasmids among the ESBL-producing Shigella isolates, 87% (20/23) 976 Moreno-Mingorance et al. Table 1. Antimicrobial resistance rates of S. sonnei and S. flexneri isolates obtained in this study from 2020 to 2021, classified according to the source of acquisition S. sonnei S. flexneri Non-MSM/Unk Non-MSM/Unk MSM (N = 22) (N = 15) (N = 15) n (%) n (%) Total (N = 37) n (%) MSM (N = 28) n (%) n (%) Total (N = 43) n (%) Antimicrobial agent Ceftriaxone 15 (68) 3 (20) 18 (49) 4 (14) 1 (7) 5 (12) Cefixime 15 (68) 3 (20) 18 (49) 4 (14) 1 (7) 5 (12) Ertapenem 0 0 0 0 0 0 Ciprofloxacin 21 (95) 4 (27) 25 (68) 11 (39) 5 (33) 16 (37) Co-trimoxazole 22 (100) 9 (60) 31 (84) 10 (36) 6 (40) 16 (37) Azithromycin 21(95) 1 (7) 22 (59) 17(61) 5 (33) 22 (51) MIC 16–48 mg/L 0 2 (13) 2 (5) 3 (11) 1 (7) 4 (9) MIC ≥ 64 mg/L 21 (95) 1 (7) 22 (59) 14 (50) 4 (27) 18 (42) MDR 21 (95) 3 (20) 24 (65) 5 (18) 1 (6) 6 (14) XDR 15 (68) 1 (7) 16 (43) 0 0 0 Unk, unknown. P values <0.05 were considered statistically significant at the 95% CI level. Significance found between acquisition groups within the same species, indicated by bold type. encoded blaCTX-M-27 and 13% (3/23) blaCTX-M-15. All (19/19) non-ESBL producer S. flexneri 3a from the UK, with p183660 (iden­ MSM-ESBL-producing isolates encoded blaCTX-M-27, while tity >99.97%; coverage 92%) previously identified from a among the non-MSM/unknown origin isolates, 25% (1/4) har­ CTX-M-27-producing S. sonnei isolate in the UK in 2015 and with boured blaCTX-M-27 and 75% (3/4) blaCTX-M-15. Interestingly, three plasmid p893916 (identity >99.98%; coverage 100%) described MSM-associated S. sonnei encoded blaCTX-M-27 despite being sus­ in a CTX-M-27-producing CipR.MSM5 genotype S. sonnei from the ceptible to all the β-lactams tested and not showing evidence of UK in 2020. On the other hand, the blaCTX-M-27-encoding plasmid ESBL production (i.e. synergy between amoxicillin/clavulanate from the S. flexneri isolates (pSH242 and pSH282) also showed and 3GCs). The genomic analysis of these isolates showed a dis­ high identity with pKSR100, p183660 and p893916, but lower ruption of the −35 (TTGAAA) and −10 (TACAAT) promoter regions coverage than that found within the S. sonnei isolates (identity of blaCTX-M-27 due to the inversion of an IS26 element located up­ >97.9%; coverage 83%–90%; identity >99.9%; coverage 76%– stream (Figure 1, pSH280). Azithromycin-resistant isolates en­ 86% and identity >99.89%; coverage 82%–93%, respectively).14,15 coded mph(A) and/or erm(B), the co-trimoxazole resistant The p183660 backbone was conserved in all plasmids, showing showed different combinations of the dfrA1, dfrA17, sul1 and structural differences mainly due to deletions and genomic reor­ sul2 genes and the ciprofloxacin-resistant isolates presented tri­ ganizations of mobile genetic elements containing different com­ ple mutations in the QRDRs of the gyrA (S83L and D87G/N) and binations of antimicrobial resistance determinants. Specifically, parC (S80I) genes. Other single mutations in QRDRs as well as pSH242 and pSHS282 from S. flexneri did not harbour the plasmid-mediated quinolone resistance mediated by the Qnr pKSR100 integron containing sulphonamide, trimethoprim and family were also detected (Table S1). aminoglycoside resistance genes (sul1, dfrA17 and aadA5), and additionally pSHS282 did not carry the genomic region containing the macrolide resistance determinants mph(A) and erm(B). Genomic analysis of blaCTX-M-27-encoding plasmids Interestingly, blaCTX-M-27 in S. sonnei was flanked by IS26 and Plasmid alignment of the trimmed short reads revealed that all IS15 while in S. flexneri it was flanked by two IS26 (Figure 1 and MSM-associated S. sonnei encoding blaCTX-M-27 showed 95% Figure S1). coverage of p183660, while the MSM-associated S. flexneri en­ coding blaCTX-M-27 exhibited a coverage of between 79% and Genotyping, serotyping and genomic epidemiology 89%. Consequently, two MSM-associated S. sonnei (SH212 and SH280) and two S. flexneri (SH242 and SH282) encoding of Shigella spp. blaCTX-M-27 and showing different resistance phenotype were se­ Genotyping of XDR ESBL-producing S. sonnei isolates according to lected for long-read sequencing to perform a detailed genomic the Hawkey’s classification showed that 100% (15/15) of analysis of blaCTX-M-27-harbouring plasmids and to compare MSM-associated isolates belonged to the CipR.MSM5 genotype them with previously identified Shigella MSM-associated plas­ (genotype number 3.6.1.1.2) [Figure 2(a)].16 The three non-MSM/ mids pKSR100, p183660 and p893916. unknown origin ESBL-producing S. sonnei belonged to genotypes The blaCTX-M-27-encoding plasmid from the S. sonnei isolates CipR.MSM5, MSM4 (genotype number 3.7.25) and Central Asia III (pSH212 and pSH280) showed a high identity and coverage with (genotype number 3.6), respectively [Figure 2(a)]. It was of note plasmid pKSR100 (identity 98.11%; coverage 90%) described in a that the non-MSM-associated S. sonnei belonging to genotype 977 Increased AMR in MSM-associated Shigella spp., Barcelona pSH282 pSH242 pKSR100 p183660 p893916 pSH212 pSH280 Figure 1. Structural organization of the region encoding antibiotic resistance determinants of blaCTX-M-27-harbouring plasmids of S. sonnei and S. flex­ neri from this study compared with MSM-associated Shigella plasmids pKSR100, p183660 and p893916. pSH282 and pSH242 are plasmids of S. flexneri SH282 and SH242, respectively; pSH212 and pSH280 are plasmids of S. sonnei SH212 and SH280, respectively. Light grey shading denotes shared re­ gions of identity. DUF denotes domain-containing proteins. * denotes partial features. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. CipR.MSM5 was obtained from a female in the context of an intra­ Discussion familial transmission event involving an MSM positive for S. sonnei (SNP distance = 1) (Figure 2). Regarding the MSM-associated We describe an increase and dissemination of XDR non-ESBL-producing S. sonnei isolates, 86% (6/7) belonged to ESBL-producing S. sonnei and MDR ESBL-producing S. flexneri the genotype CipR.MSM5 and 14% (1/7) to the CipR.parent (geno­ among MSM in Barcelona during 2020–21. Specifically, 68% of type number 3.6.1). S. sonnei and 14% of S. flexneri acquired by MSM were ESBL pro­ The phylogeny of the S. sonnei isolates belonging to the 3.6 ducers and simultaneously resistant to ciprofloxacin, azithromy­ clade showed that all the XDR ESBL-producing CipR.MSM5 isolates cin and/or co-trimoxazole. These findings differ from the data obtained in this study nested together and were genetically clo­ previously obtained by our group in 2015–19, in which only ser to the representative XDR CipR.MSM5 isolates obtained in the 3.8% of S. sonnei and 2.3% of S. flexneri acquired among MSM UK (SRR11206407) and Australia (SRR12132003) (range 6–11 were ESBL producers.12 SNPs) than to the non-ESBL-producing CipR.MSM5 S. sonnei ESBL-producing S. sonnei among MSM has been detected in isolates obtained in Barcelona between 2015 and 2019 several countries. Australia reported a prolonged outbreak of (range 12–42 SNPs) [Figure 2(b)].1,15 On the contrary, the ESBL-producing S. sonnei belonging to the VN2.KH1.Aus geno­ non-ESBL-encoding CipR.MSM5 isolates obtained between 2020 type (genotype number 3.7.29.1.4.1) from 2019 to 2020, mainly and 2021 were genetically closer to the CipR.MSM5 isolates ob­ associated with males.1 In addition, the MSM-related tained in Barcelona in the period 2015–19. Finally, the three iso­ ESBL-producing isolates detected in Canada and in Belgium be­ lates encoding blaCTX-M-27 but not showing an ESBL phenotype tween 2017–19 and 2018–19, respectively, and the unique also belonged to CipR.MSM5 and were clustered together (range ESBL-producing S. sonnei isolate detected in Barcelona during 1–3 SNPs) [Figure 2(b)]. 2015–19 also belonged to this genotype.4,12,19 Nevertheless, The in silico serotyping of MSM-associated S. flexneri showed sporadic detection of XDR ESBL-producing CipR.MSM5 isolates that 46% (13/28) belonged to the 2a serotype, 27% (8/28) to were also identified in Australia and in the UK before the 3a, 21% (6/28) to 1b and 4% (1/28) to 1c. All the ESBL-producing COVID-19 epidemic.1,11 However, an outbreak of XDR S. flexneri isolates (5/5) belonged to the 2a serotype. The four ESBL-producing CipR.MSM5 S. sonnei was detected in the UK be­ MSM-associated ESBL-producing S. flexneri isolates formed a tween September 2021 and February 2022.11 In our previous closely related independent cluster within the phylogeny (range study, we identified that the most prevalent MSM-related S. son­ 0–15 SNPs) and no epidemiological link was identified among nei genotype detected during 2015–19 in Barcelona was the patients from whom these isolates were obtained. The re­ CipR.MSM5.12 Despite not being ESBL-producers, isolates belong­ maining non-MSM-associated ESBL-producing isolates showed ing to this genotype have shown great expansion in recent years, a higher genetic distance from them (range 200–204 SNPs) as they have also been found as being the most prevalent among (Figure 3). The four MSM-associated ESBL-producing isolates pos­ the ciprofloxacin- and azithromycin-resistant isolates in sessing a triple mutation in the QRDR [gyrA (S83L and D87N) and Australia, the UK and the USA during 2018–19 and also the parC (S80I)] were genetically close to the non-ESBL-producing most prevalent in MSM-associated shigellosis in Belgium.4,16 fluoroquinolone-resistant S. flexneri identified in Barcelona in This may demonstrate successful worldwide dissemination of the 2015–19 and 2020–21 periods, in Australia in 2016 this genotype, even before ESBL acquisition. (SRR7885970) and in the UK in 2020 (SRR12175226) In our present study, three MSM-associated XDR (Figure 3).17,18 ESBL-producing CipR.MSM5 isolates were sporadically detected 978 Moreno-Mingorance et al. (a) Country Genotype ESBL Tree scale: 0.001 This study isolates MRD Australia Central Asia III CTX-M-15 QRDR ESBL Genotype Belgium CipR.MSM1 CTX-M-27 Country Egypt CipR.MSM5 QRDR France CipR.parent gyrA (D87Y) Korea Global III gyrA (S83L) Spain VN2.KH1.Aus gyrA (S83L, D87G), parC (S80I) United Kingdom VN2.MSM2 MRD Uruguay VN2.MSM2.Aus mphA Latin America III mphA, ermB MSM4 This study isolates MSM-associated non-ESBL producer MSM-associated ESBL producer Non-MSM/Unk ESBL producer (b) Country Genotype ESBL QRDR MRD This study isolates Tree scale: 0.00010000100001000009 Figure 2. Phylogenomic analysis of S. sonnei isolates. (a) Maximum-likelihood tree of the ESBL-producing S. sonnei and MSM-associated S. sonnei iso­ lates collected in Barcelona in this study from 2020 to 2021 (n = 25), as well as the MSM-associated S. sonnei isolates previously identified in the 2015– 19 period in Barcelona (n = 26) and the selected international representative S. sonnei isolates (n = 27). The tree was rooted with S. sonnei 53G. (b) Phylogeny of the S. sonnei isolates belonging to the 3.6 clade based on the Hawkey et al. classification. Isolate metadata and molecular characteristics are colour coded, as detailed in the key. MRD, macrolide resistance determinant. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. 979 Increased AMR in MSM-associated Shigella spp., Barcelona Country Serotype ESBL This study isolates Australia 1b CTX-M-15 MRD QRDR ESBL Bangladesh 1c CTX-M-27 Serotype Country China 2a QRDR France 3a gyrA (S83L) India 3b gyrA (D87G) Iraq 4av gyrA (S83L, D87N), parC (S80I) Korea 5a MRD South Africa X ermB Spain Y mphA United Kingdom mphA, ermB Vietnam This study isolates MSM-associated non-ESBL producer MSM-associated ESBL producer Non-MSM/Unk ESBL producer Tree scale: 0.01 Figure 3. Phylogenomic analysis of S. flexneri isolates. Maximum-likelihood tree of the ESBL-producing S. flexneri and MSM-associated S. flexneri iso­ lates collected in Barcelona in this study from 2020 to 2021 (n = 29), as well as the MSM-associated S. flexneri isolates previously identified in the 2015– 19 period in Barcelona (n = 44) and the selected international representative S. flexneri isolates (n = 31). The tree was rooted with S. flexneri 2a strain 2457T. Isolate metadata and molecular characteristics are colour coded, as detailed in the key. MRD: macrolide resistance determinant. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. in December 2020 in Barcelona; however, in June 2021 a re­ this clone has been introduced in our country, replacing the appearance and rapid outbreak expansion was detected. The S. non-ESBL-producing CipR.MSM5 clone, which previously circu­ sonnei phylogeny revealed that all the outbreak-related XDR lated in our area, and originating a local outbreak in Barcelona. CipR.MSM5 isolates from Barcelona were part of the same mono­ The successful dissemination of this clone among different coun­ phyletic cluster in which the representative XDR CipR.MSM5 iso­ tries may be the consequence of the interconnected internation­ lates from the UK and Australia were also found. This fact al sexual networks among MSM after the relaxation on the reinforces the international transmission hypothesis by which COVID-19 epidemic measures in the summer of 2020.11,12,20 980 Moreno-Mingorance et al. On the other hand, the genomic analysis of blaCTX-M-27-encoding countries by the cases included was not collected, and conse­ plasmids in this study showed a high similarity of these mobile quently, it was not possible to establish epidemiological links be­ genetic elements with previous plasmids encoding resistance to tween the patients and/or identify possible transmission chains. macrolides, co-trimoxazole, aminoglycosides and/or β-lactams in In summary, in this work we report an increasing trend of anti­ MSM-associated Shigella (pKSR100, p183660 and p893916). The microbial resistance in Shigella spp. among MSM in Barcelona pKSR100 plasmid, which among others encodes macrolide resistant since 2021, mainly as a consequence of the emergence of an determinants, was proposed to have had a role in driving the epi­ outbreak of XDR ESBL-producing S. sonnei belonging to the demics of azithromycin-resistant S. sonnei and S. flexneri among CipR.MSM5 genotype. Our data demonstrate a close genetic link MSM.21 Thus, the combination of pKSR100-type plasmids that with the XDR ESBL-producing S. sonnei isolates responsible for have acquired blaCTX-M-27 with a successfully disseminated geno­ the prolonged outbreak detected in the UK also in 2021, confirm­ type of S. sonnei could have facilitated the expansion and in­ ing its successful international dissemination coinciding with the crease of XDR ESBL-producing Shigella among MSM during the increase in social interactions after the COVID-19 lockdown. last 2 years. Additionally, the location of several resistance genes Additionally, to our knowledge, we have detected the first clonal within the same plasmid could also be an element that may con­ dissemination of MDR ESBL-producing S. flexneri among MSM, tribute to its co-selection and dissemination, especially in con­ which also represents a significant health threat. These results re­ texts of high antibiotic pressure such as MSM receiving inforce the importance of collaborative international surveillance treatment for other STIs such as Chlamydia trachomatis and of MDR/XDR Shigella spp. for rapid identification of its emergence Neisseria gonorrhoeae infection, as has previously been docu­ as well as the need to implement measures that allow early diag­ mented in the UK.11 nosis and contact tracing of cases of shigellosis among MSM to Unlike S. sonnei, no MSM-associated outbreaks of reduce its dissemination. ESBL-producing S. flexneri have previously been reported. In our study, we identified four MSM-associated CTX-M-27-producing S. flexneri 2a isolates. As no epidemiological link was detected Acknowledgements among the patients from whom these isolates were obtained We would like to thank the technical staff of the Clinical Microbiology and they showed a close genetic distance, the existence of sec­ Laboratory of Hospital Vall d’Hebron for their contribution to the study. ondary non-detected cases could be considered. blaCTX-M-27 in S. flexneri isolates was found in a pKSR100-related plasmid and showed a high identity with plasmids harbouring blaCTX-M-27 in S. sonnei. Consequently, the hypothesis of the acquisition by Funding This work was partially supported by the ‘Ministerio de Economía y S. flexneri of pKSR100-type encoding the blaCTX-M-27 plasmid cir­ Competitividad’, ‘Instituto de Salud Carlos III’, and co-financed by the culating among S. sonnei would be plausible. For this to occur, European Regional Development Fund (ERDF) ‘A Way to Achieve Europe’ several genomic reorganizations involving the loss of several (Spanish Network for Research in Infectious Diseases, grant number antimicrobial resistance genes should have happened. Another RD16/0016/0003) and by the Centro de Investigación Biomédica en Red possible scenario could have been the acquisition of a genomic (CIBER de Enfermedades Infecciosas, grant no. CB21/13/00054). A.M.M. fragment containing blaCTX-M-27 by a circulating pKSR100-type is supported by a grant from the ‘Fondo de Investigación Sanitaria’ plasmid in S. flexneri mediated by the IS26 elements flanking (Contratos Predoctorales de Formación en Investigación, grant number this structure. These hypotheses have yet to be confirmed. In FI19/00315). any case, it is necessary to continue monitoring the evolution of ESBL production in S. flexneri isolates in MSM in our environ­ ment, as well as their possible emergence or dissemination in Transparency declarations other countries. All authors declare that they have no conflicts of interest to disclose. In this study we identified a significant increase of quinolone resistance in S. flexneri in MSM compared with what we observed in the 2015–19 period (4.5% versus 38%) as a consequence Supplementary data of the introduction of fluoroquinolone-resistant S. flexneri 2a Figure S1 and Table S1 are available as Supplementary data at JAC Online. (P < 0.05).12 Although the MSM-associated ESBL-producing S. flexneri detected in Barcelona should not be considered XDR, as simultaneous resistance to co-trimoxazole and azithromycin has not yet been found, these results should warn of the possibil­ References ity that S. flexneri could also acquire additional resistance deter­ 1 Ingle DJ, Andersson P, Valcanis M, et al. Prolonged outbreak of minants, thereby following the same antibiotic resistance multidrug-resistant Shigella sonnei harboring blaCTX-M-27 in Victoria, evolution as the MSM-related S. sonnei. This could be a plausible Australia. Antimicrob Agents Chemother 2020; 64: e01518-20. https:// phenomenon that might happen by horizontal gene acquisition doi.org/10.1128/AAC.01518-20 of such resistance determinants in a population with high anti­ 2 Baker KS, Dallman TJ, Field N et al. Genomic epidemiology of Shigella in microbial exposure. the United Kingdom shows transmission of pathogen sublineages and One limitation of this study is that due to the retrospective determinants of antimicrobial resistance. Sci Rep 2018; 8: 7389. https:// design of the research, some demographic and epidemiological doi.org/10.1038/s41598-018-25764-3 data are missing. Specifically, data regarding sexual contacts 3 Tsai C-S, Lin K-Y, Liou B-H et al. Changing epidemiology of shigellosis in or information about travel to non-low- or -middle-income Taiwan, 2010-2019: an emerging threat to HIV-infected patients and 981 Increased AMR in MSM-associated Shigella spp., Barcelona men who have sex with men. Emerg Microbes Infect 2022; 11: 498–506. 13 Magiorakos A-P, Srinivasan A, Carey RB et al. Multidrug-resistant, ex­ https://doi.org/10.1080/22221751.2022.2031309 tensively drug-resistant and pandrug-resistant bacteria: an international 4 Fischer N, Maex M, Mattheus W et al. Genomic epidemiology of persist­ expert proposal for interim standard definitions for acquired resistance. ently circulating MDR Shigella sonnei strains associated with men who Clin Microbiol Infect 2012; 18: 268–81. https://doi.org/10.1111/j.1469- have sex with men (MSM) in Belgium (2013–19). J Antimicrob 0691.2011.03570.x Chemother 2021; 77: 89–97. https://doi.org/10.1093/jac/dkab377 14 Mook P, McCormick J, Bains M et al. ESBL-producing and 5 Campbell D, Bowen A, Bhatnagar A et al. Identification and character­ macrolide-resistant Shigella sonnei infections among men who have ization of Shigella with decreased susceptibility to azithromycin in the sex with men, England, 2015. Emerg Infect Dis 2016; 22: 1948–52. United States, 2005 to 2014. J Glob Antimicrob Resist 2020; 21: 417–9. https://doi.org/10.3201/eid2211.160653 https://doi.org/10.1016/j.jgar.2019.12.005 15 Locke RK, Greig DR, Jenkins C et al. Acquisition and loss of CTX-M plas­ 6 Wilmer A, Romney MG, Gustafson R et al. Shigella flexneri serotype 1 in­ mids in Shigella species associated with MSM transmission in the UK. fections in men who have sex with men in Vancouver, Canada. HIV Med Microb Genom 2021; 7: 000644. https://doi.org/10.1099/mgen.0.000644 2015; 16: 168–75. https://doi.org/10.1111/hiv.12191 16 Hawkey J, Paranagama K, Baker KS et al. Global population structure 7 Williams PCM, Berkley JA. Guidelines for the treatment of dysentery and genotyping framework for genomic surveillance of the major dysen­ (shigellosis): a systematic review of the evidence. Paediatr Int Child tery pathogen, Shigella sonnei. Nat Commun 2021; 12: 2684. https://doi. Health 2018; 38: S50–65. https://doi.org/10.1080/20469047.2017. org/10.1038/s41467-021-22700-4 1409454 17 Ingle DJ, Easton M, Valcanis M et al. Co-circulation of 8 Shane AL, Mody RK, Crump JA et al. 2017 Infectious Diseases Society of multidrug-resistant Shigella among men who have sex with men in America Clinical Practice Guidelines for the Diagnosis and Management of Australia. Clin Infect Dis 2019; 69: 1535–44. https://doi.org/10.1093/cid/ Infectious Diarrhea. Clin Infect Dis 2017; 65: e45–80. https://doi.org/10. ciz005 1093/cid/cix669 18 Dallman TJ, Charles H, Prochazka M et al. Emergence of novel strains 9 Campos-Madueno EI, Bernasconi OJ, Moser AI et al. Rapid increase of of Shigella flexneri associated with sexual transmission in adult men in CTX-M-producing Shigella sonnei isolates in Switzerland due to spread England, 2019–2020. J Med Microbiol 2021; 70: 001437. https://doi.org/ of common plasmids and international clones. Antimicrob Agents 10.1099/jmm.0.001437 Chemother 2020; 64: e01057-20. https://doi.org/10.1128/AAC.01057-20 10 ECDC. Rapid Risk Assessment: Increase in Extensively-drug Resistant 19 Gaudreau C, Bernaquez I, Pilon PA et al. Clinical and genomic investi­ Shigella sonnei Infections in Men Who have Sex With Men in the EU/EEA gation of an international ceftriaxone- and azithromycin-resistant and the UK. 2022. https://www.ecdc.europa.eu/en/publications-data/ Shigella sonnei cluster among men who have sex with men, Montréal, rapid-risk-assessment-increase-extensively-drug-resistant-shigella-sonnei. Canada 2017–2019. Microbiol Spectr 2022; 10: e02337-21. https://doi. org/10.1128/spectrum.02337-21 11 Charles H, Prochazka M, Thorley K et al. Outbreak of sexually transmit­ ted, extensively drug-resistant Shigella sonnei in the UK, 2021–22: a de­ 20 Andrés C, Piñana M, Borràs-Bermejo B et al. A year living with scriptive epidemiological study. Lancet Infect Dis 2022; 22: 1503–10. SARS-CoV-2: an epidemiological overview of viral lineage circulation by https://doi.org/10.1016/S1473-3099(22)00370-X whole-genome sequencing in Barcelona city (Catalonia, Spain). Emerg 12 Moreno-Mingorance A, Espinal P, Rodriguez V et al. Circulation of Microbes Infect 2022; 11: 172–81. https://doi.org/10.1080/22221751. multi-drug-resistant Shigella sonnei and Shigella flexneri among men 2021.2011617 who have sex with men in Barcelona, Spain, 2015–2019. Int J 21 Baker KS, Dallman TJ, Field N et al. Horizontal antimicrobial resistance Antimicrob Agents 2021; 58: 106378. https://doi.org/10.1016/j. transfer drives epidemics of multiple Shigella species. Nat Commun 2018; ijantimicag.2021.106378 9: 1462. https://doi.org/10.1038/s41467-018-03949-8 982