The re-emergence of sexually transmissible multidrug resistant Shigella flexneri 3a, England, United Kingdom

doi:10.21203/rs.3.rs-3760116/v1

Download PDF

Article

The re-emergence of sexually transmissible multidrug resistant Shigella flexneri 3a, England, United Kingdom

https://doi.org/10.21203/rs.3.rs-3760116/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 02 Aug, 2024

Read the published version in npj Antimicrobials and Resistance →

You are reading this latest preprint version

Shigellosis is an enteric infection that transmits through the faecal-oral route, which can occur during sex between men who have sex with men (MSM). Between 2009 and 2014, an epidemic of sexually transmissible Shigella flexneri 3a occurred in England that subsequently declined. However, from 2018 to 2021, despite restrictions to control SARS-CoV-2 spread, S. flexneri 3a re-emerged. We explored possible drivers of the re-emergence by comparing host demographic features and pathogen genomics. The age and geographical distribution of cases in 2009–2014 (relative to 2018–2020) was comparable, with cases primarily among 35–64 year old men in London and the Southeast, a majority (54%) of which self-identified as gay or bisexual. Genomic analyses of 502 bacterial isolates showed that the majority (58%) of re-emerging MSM strains were a clonal replacement of the original lineage, with a reduced antimicrobial resistance profile. Specifically, more recent isolates had lost bla_TEM−1 and erm(B), likely in line with shifting antimicrobial selection pressures. The absence of major changes in the pathogen or host demographics suggest that other factors, such as host immunity, may have driven the re-emergence of S. flexneri 3a and highlight the need for further work in the area.

Biological sciences/Genetics/Microbial genetics/Bacterial genes

Health sciences/Diseases/Infectious diseases/Bacterial infection

Health sciences/Gastroenterology/Gastrointestinal diseases/Gastroenteritis

Biological sciences/Microbiology/Antimicrobials/Antimicrobial resistance

Biological sciences/Computational biology and bioinformatics

Bacteria in the Shigella genus are Gram-negative, non-motile, and highly pathogenic to humans. There are four species of Shigella: boydii, dysenteriae, flexneri and sonnei. As few as 10–100 colony forming units of Shigella bacteria can cause shigellosis,¹ a highly infectious enteric illness.² The low infectious dose, in combination with potential asymptomatic carriage, allows Shigella to propagate unbeknownst between hosts.³ Symptoms of shigellosis are fever, nausea, bloody diarrhoea, and stomach cramping.² Rarer and more severe complications of gastrointestinal shigellosis include haemolytic uremic syndrome (HUS),⁴ toxic megacolon,⁵ bacteraemia,⁶ and reactive arthritis.⁷ Shigella spp. are transmitted through the faeco-oral route, most commonly associated with person to person transmission. On a global basis, transmission is common in institutional,⁸ and community settings,⁹ and Shigella spp. are also transmitted through consumption of contaminated food and water,¹⁰ and is a major cause of travel-associated illness in higher-income nations.

In higher-income nations, since the 1970s,^11–13 shigellosis transmission has also been described as a sexually transmissible illness among men who have sex with men (MSM). The first UK outbreak of MSM-associated shigellosis transmission was identified in 2004.¹⁴ Since then, MSM-associated outbreaks have continued to increase in frequency in the UK, and other nations.¹⁵ This is associated with engagement in higher-risk sexual activities, such as chemsex, anal-digital contact, and condomless anal intercourse.¹⁶ In line with global trends, sexually transmissible shigellosis among MSM is most commonly associated with Shigella flexneri (S. flexneri) and Shigella sonnei (S. sonnei).¹⁷

Antimicrobial resistant (AMR) Shigella spp. have been described since the 1960s, when resistance to streptomycin, sulphonamides, tetracycline, and ampicillin became common.^18,19 Subsequently, resistance to ciprofloxacin,²⁰ and azithromycin,²¹ were identified in S. sonnei and S. flexneri. AMR Shigella spp. are a growing cause for concern, particularly in sexual transmission networks, as multiple drug resistant (MDR) and extensively drug resistant (XDR) strains of Shigella spp. are on the rise,²² with the bla_CTX−M family of genes - conferring ceftriaxone resistance - being described across Shigella spp.,²³ and recently causing an international outbreak of XDR S. sonnei.²² These resistance profiles are especially concerning, as The World Health Organization (WHO) currently recommends ciprofloxacin to be used as the first line of treatment, followed by azithromycin and ceftriaxone as second line treatments (alongside pivmecillinam).²⁴

In 2009 in the UK, transmission of Shigella spp. associated with domestically acquired infection in adult males, a proxy for MSM-associated sexual transmission, became increasingly common and was originally associated with increases in S. flexneri serotype 3a. Associated drivers were subsequently identified as sexual encounters through social media and smartphone applications, and engaging in higher-risk behaviours, such as chemsex, and condomless sexual intercourse.¹⁶ Following the 2009 epidemic of S. flexneri 3a, there were overlapping S. flexneri 2a and S. sonnei outbreaks,²⁵ driven by the acquisition of a plasmid encoding resistance to azithromycin.²¹ Since 2015, domestic acquisition of shigellosis among MSM in the UK has been dominated by ciprofloxacin resistant S. sonnei, thought to have been imported from Asia and containing triple mutations in the quinolone resistance-determining region (QRDR).²⁶ This S. sonnei lineage has latterly (2021 onwards) become XDR through acquisition of an extended-spectrum beta lactamase (ESBL) encoding plasmid,^22,27 that has also transferred into S. flexneri 2a.²⁸ The spread of AMR shigellosis may be exacerbated by behavioural changes associated with the use of pre-exposure prophylaxis (PrEP) for human immunodeficiency virus (HIV).^3,27

Between 2015 and 2018, the number of S. flexneri 3a cases fell to below pre-epidemic (2004–2008) levels. However, S. flexneri 3a increased again in the UK from 2019–2021, despite the non-pharmaceutical interventions associated with the SARS-CoV-2 pandemic that brought about a general reduction in gastrointestinal disease.²⁹ Here, we explored whether changes in host demographic profiles or pathogen evolution played a role in the re-emergence of sexually transmissible S. flexneri 3a. Understanding the drivers of the emergence of new subtypes of sexually transmissible shigellosis is key to intervening in the worsening AMR profile and likely increased transmission scenario. The aim of our study was to integrate whole genome sequencing (WGS) data with enhanced epidemiological data to explore how interactions between bacteria, host, and environmental factors may have initially driven the re-emergence of S. flexneri 3a in England, despite the impact of the SARS-CoV-2 pandemic on social interaction.

Overview of demographic data

Between 2004 and 2020, 40.5% (n = 6,864/16,952) of Shigella spp. diagnoses among adults processed by the Gastrointestinal Bacteria Reference Unit (GBRU) at the UKHSA were identified as S. flexneri. Following an increase in the proportion of S. flexneri cases that were serotype 3a between 2004 and 2013, and a decline between 2013 and 2018, the proportion of 3a more than tripled between 2018 and 2020 (from 7.8% [n = 7/90] to 30.0% [n = 85/287], Fig. 1a). Epidemiological comparisons of S. flexneri 3a cases among presumptive MSM in the emergence phase (2012–2013) and re-emergence phase (2019–2020) (Table 1), showed that the age and geographical distribution remained largely unchanged, with the greatest proportion of cases reported from London and the Southeast of England (Table 1). Hospitalisation data was not available for the cases identified during the emergence phase; however, enhanced surveillance questionnaires from the re-emergence phase indicate that 35% (n = 18/51) of presumptive MSM were hospitalised. WGS has revealed that the re-emergence of S. flexneri 3a has been driven mainly by a single 10 SNP cluster t10.1189, (n = 27/118, 22.8%) (Table 1).

Table 1

Summary of characteristics of the recent re-emergence of *S. flexneri* 3a cases among presumptive MSM and comparison to original emergence, England.
Characteristic	2012–2013 n (%)	2019–2020 n (%)
Age
16–34	95 (39.7)	54 (42.2)
35–64	136 (56.9)	67 (52.3)
65+	5 (2.1)	7 (5.5)
Unknown	3 (1.3)	0 (0.0)
Median age [IQR]^a	38.5 [30-46.5]	36 [29-45.5]
Region
London	140 (58.6)	91 (71.1)
South East	36 (15.1)	20 (15.6)
North West	28 (11.7)	2 (1.6)
West Midlands	8 (3.4)	1 (0.8)
East of England	13 (5.4)	8 (6.3)
South West	6 (2.5)	4 (3.1)
Yorkshire & Humber	2 (0.8)	1 (0.8)
North East	4 (1.7)	1 (0.8)
East Midlands	2 (0.8)	0 (0.0)
Unknown	0 (0.0)	0 (0.0)
Hospitalised
Yes	ND	18 (35.3)
No	ND	33 (64.7)
Median nights in hospital [IQR]^a		2 [2–4]
10 SNP^b cluster
t10.1177	27 (11.3)	0 (0.0)
t10.43	18 (7.5)	0 (0.0)
t10.138	2 (0.8)	0 (0.0)
t10.116	1 (0.4)	0 (0.0)
t10.1182	1 (0.4)	0 (0.0)
t10.1197	1 (0.4)	0 (0.0)
t10.1211	1 (0.4)	0 (0.0)
t10.1227	1 (0.4)	0 (0.0)
t10.1356	1 (0.4)	0 (0.0)
t10.1189	0 (0.0)	108 (84.4)
t10.1305	0 (0.0)	2 (1.6)
t10.1627	0 (0.0)	1 (0.8)
t10.1648	0 (0.0)	1 (0.8)
t10.1684	0 (0.0)	13 (10.2)
t10.1852	0 (0.0)	1 (0.8)
t10.1969	0 (0.0)	1 (0.8)
Unknown	186 (77.8)	1 (0.8)
^a Interquartile range (IQR)
^b Single nucleotide polymorphism (SNP)

S. flexneri 3a re-emergence featured clonal replacement

To determine the genomic epidemiology among 502 isolates of S. flexneri 3a from the emergent and re-emergent periods (2008–2013 n = 205; 2016–2020 n = 298), a phylogenetic tree was constructed and matched to epidemiological metadata (Fig. 2). This was complemented by Bayesian Analysis of Population Structure (BAPS) groups being computed to assign genetically similar clusters, which we called Global-BAPS clusters. This revealed that isolates clustered by patient travel destination, with isolates from cases reporting recent travel to Asia mostly (71%, n = 25/35) belonging to the Global-BAPS-3 cluster, while African-travel associated isolates typically (74%, n = 26/35) belonged to Global-BAPS-2 (Table 2, Fig. 2). Notably, more than half (52%, n = 257/502) of the isolates have no associated travel data. Global-BAPS-1 cluster was mostly (84%, n = 286/339) composed of males, aged 16–60, who were without a known history of international travel and was thus defined as MSM-associated. The majority (78%, n = 159/205) of the 2008–2014 S. flexneri 3a outbreak belonged to the Global-BAPS-1 cluster, as did the majority (58%, n = 180/312) of S. flexneri 3a isolated between 2016 and 2020.

Table 2

Temporal and demographic features of isolates in this study by Global-BAPS groups.
		Year (%)											Sex (%)		Age Group in Years (%)			Travel (%)
Global-BAPS Group	Total (n)	2008	2009	2010	2011	2012	2013	2016	2017	2018	2019	2020	Male	Female	< 16	16–60	> 60	Africa	Asia	Europe	N-America	Oceania	S-America
1	339	4	4	8	6	5	19	10	5	2	14	22	96	4	1	94	5	0	1	4	0	0	1
2	45	2	16	0	2	2	11	9	16	7	22	13	76	24	11	69	20	58	2	0	0	0	0
3	67	3	7	6	9	4	4	16	15	16	15	3	64	36	24	63	13	0	37	6	0	0	0
4	35	3	3	3	6	3	3	9	11	14	23	23	71	29	9	80	11	23	3	0	0	0	11
5	6	0	0	17	0	0	0	0	0	17	50	17	33	67	17	67	17	0	67	0	0	0	0
6	3	0	0	0	0	0	0	0	0	67	0	33	100	0	0	100	0	33	0	0	0	0	0
7	7	0	0	0	0	0	0	14	43	14	29	0	71	29	57	29	14	0	0	0	14	0	0

As we were focused on pathogen changes between the emergence and re-emergence of MSM-associated Global-BAPs-1 isolates, these isolates (n = 339) underwent a further round of BAPS grouping to facilitate study at a finer resolution. These more highly resolved groups are referred to as MSMA-BAPS groups (Table 3, Fig. 3). Correlation with the nomenclature from the previous study of the emergent event revealed that lineages, previously designated as sublineages B and C,³⁰ were associated with MSMA-BAPS-1, and sublineage A with MSMA-BAPS-3, while cases from 2019 onwards were now MSMA-BAPS-5. This revealed that the first outbreak was majorly associated with MSMA-BAPS-1 (23%) and BAPS-3 (39%), and the second outbreak was associated with MSM-BAPS-5 (42%), (Table 3). Thus, this subsequent analysis revealed further distinction between the two time periods, where the recent re-emergence (2016–2020) was associated with a single clonal expansion that replaced the earlier outbreak (Fig. 3). Thus, the clonal replacement for the re-emergence of S. flexneri 3a in England has evolved from MSMA-BAPS-1 to MSMA-BAPS 5.³⁰

Table 3

Temporal features of the Global-BAPS-1 cluster broken down into MSMA-BAPS groups.
		Year (%)
MSMA- BAPS Group	Total (n)	2008	2009	2010	2011	2012	2013	2016	2017	2018	2019	2020
1	79	3	1	3	4	8	42	27	8	4	1	1
2	21	24	33	38	0	0	0	0	5	0	0	0
3	90	2	4	17	18	12	36	8	2	0	1	0
4	17	29	12	12	12	6	0	18	6	0	0	6
5	132	0	0	0	0	0	0	2	5	3	35	56

Re-emergent S. flexneri 3a is less antimicrobial resistant

Owing to the importance of shifting resistance profiles in previous MSM-associated shigellosis outbreaks, we explored whether there were differences in the emergent and re-emergent subtypes, with respect to AMR. We identified the antimicrobial resistance genes (ARGs) across the isolates and correlated this with genomic groupings (Fig. 3, Table 4). This revealed that emergent MSMA-BAPS-1 isolates had highly conserved erm(B) (73%, n = 58/79), mph(A) (77%, n = 61/79) and bla_TEM−1 (80% n = 63/79) genes, whereas a loss of, first, bla_TEM−1 and then erm(B) ARGs in the re-emergent MSMA-BAPS-5 isolates was observed. Although bla_TEM−1 and then erm(B) genes were lost, mph(A) has been maintained in MSMA-BAPS-5 isolates (92%, n = 121/132). Previous work on this lineage revealed that most ARGs were carried on four main mobile genetic elements (MGEs); the Shigella resistance locus (bla_OXA−1, catA1, aadA1, and tet(B), pKSR100 (bla_TEM−1, erm(B), and mph(A)), pKSR100 integron (dfrA17, sul1 and aadA5) or pCERC1 (dfrA14 and sul2).³⁰ Grouping ARGs by these MGEs suggested that, in contrast to the emergent event associated with MSMA-BAPS-4 where 24% (n = 4/17) of isolates contain dfrA14 and sul2, indicative of pCERC1 presence, only a single MSMA-BAPS-5 isolate potentially contains the pKSR100 integron (indicated by presence of dfrA17, though absence of sul1 and aadA5, 0.75%, n = 1/133), and none contain genes associated with pCERC1 (indicated by genes dfrA14 and sul2, 0%, n = 0/133).

Table 4

Proportion of *S. flexneri* 3a MSMA-BAPS isolates containing various ARGs which demonstrated change between clusters.
	Proportion of MSMA-BAPS cluster containing gene (%)
Antimicrobial class	Gene	1 (n = 79)	2 (n = 21)	3 (n = 90)	4 (n = 17)	5 (n = 132)
Aminoglycosides	aph(3")-lb	4	0	37	24	0
	aadA5	3	0	17	0	0
	aph(6)-ld	4	0	39	24	0
B-lactams	bla_TEM−1	80	0	57	0	0
Trimethoprim	dfrA14	3	0	38	24	0
Trimethoprim	dfrA17	3	0	17	0	1
Macrolides	erm(B)	73	0	53	0	26
Macrolides	mph(A)	77	0	57	18	92
Sulphonamides	sul1	6	0	20	0	0
Sulphonamides	sul2	8	0	40	24	0
Antiseptics	qacE	1	0	17	0	0

Phenotypic testing (minimum inhibitory concentrations [MICs]) of the antimicrobials corresponding to the genotypic resistance profiles was performed to investigate whether changes in the genotype of the bacteria influenced phenotypic antimicrobial resistance. The findings largely corresponded (90%, n = 18/20 concordance statistic) with genotypic resistance (Supplementary Table 1). Presence of at least one of erm(B) or mph(A) and the isolate being resistant to azithromycin had a 94% (n = 16/17) concordance statistic. Presence of at least one of sul1, sul2, dfrA14, or dfrA1, and the isolate being resistant to trimethoprim-sulfamethoxazole had a 67% (n = 4/6) concordance statistic. This revealed that Global-BAPS-1 S. flexneri 3a in the UK has acquired resistance to azithromycin and trimethoprim-sulfamethoxazole, primarily through acquisition of mph(A), and sul1 and sul2. None of the isolates were resistant to ciprofloxacin, consistent with there being no relevant quinolone resistance determining region (QRDR) mutations (i.e. ParC S80I, nor GyrA S83L or D87G) present. Also, none of the isolates were resistant to ceftriaxone. While resistance to azithromycin correlated with the presence of either erm(B) or mph(A), higher levels of resistance (> 256mg/L) were typically associated with the presence of both genes (Supplementary Table 1), where the presence of both mph(A) and erm(B) genes and the isolate having a > 256mg/L resistance to azithromycin had a concordance statistic of 82% (n = 9/11). All isolates were sensitive to fosfomycin or mecillinam.

AMR gene loss on pKSR100 is facilitated by transposases

To investigate the structural basis for the changes in the ARG content of S. flexneri 3a over time, we performed long-read (Nanopore) sequencing on three representative isolates based on their AMR gene profiles, year of isolation, and BAPS-group assignment. We then investigated similarity among contiguous sequences containing mph(A), which revealed that the mph(A) genes were located on plasmids. We found that the pKSR100-like plasmids from: the pKSR100 reference plasmid from S. flexneri 3a, and 2016 and 2020 UKHSA isolates of S. flexneri 3a were overall genetically similar (Fig. 4a). However, the 2016 and 2020 pKSR100-like plasmids did have some notable differences, particularly around the region carrying AMR genes (Fig. 4b). The pKSR100-like plasmid from the 2020 isolate was shorter (69,055 bp) than the 2016 version (73,462 bp) and had lost the erm(B) and bla_TEM−1 genes, which were originally flanked by transposase genes. A qacE Δ-1-containing plasmid from a Global-BAPS-4 S. flexneri 3a isolate from 2020 was genetically distinct from the pKSR100 plasmids (Fig. 2), but contained a similar mph(A)-containing region also flanked by transposase genes. To explore the resolution more fully across the isolate collection (i.e. also in those isolates that were not long-read sequenced), we mapped isolates' sequencing reads against pKSR100 which showed that the more recent isolates mapped against a lower proportion of pKSR100, likely reflecting the loss of erm(B) and bla_TEM−1 captured in the individual isolate long-read comparisons (Supplementary Fig. 1), the decision to display percentage mapping of ≥ 60 was informed by class interval frequencies (Supplementary Fig. 1b) which allowed for a distinguishable colour gradient. This revealed that a newer, shorter plasmid with fewer AMR genes is now circulating in this MSM-associated S. flexneri 3a lineage.

We demonstrate distinct genomic subtypes of S. flexneri 3a associated with international travel, compared to MSM-associated shigellosis, in line with previous work.³⁰ This indicates that isolates associated with travel may be imported into the UK, which provides the basis for expansion of these initial travel-related clusters, in sexual transmission networks, hypothesised to create an environment of selection pressure from increased antimicrobial use for the treatment of other sexually transmissible infections e.g. gonorrhoea.^30,31 Other factors such as the host immune response and gut microbiota also likely play a role.

The changes in the retention of antimicrobial resistance genes between the pre- and post-2015 S. flexneri 3a isolates is intriguing. It appears that while the erm(B) and bla_TEM−1 genes have been dropped in the re-emergent isolates, mph(A) has been retained. We originally stipulated that this could suggest that the mph(A) gene has been transferred from the plasmid pKSR100 into the core chromosomal genome of these Shigella isolates due to its importance to the survival of the bacterium. However, our genomics analyses revealed that mph(A) was still being carried on plasmids. Perhaps erm(B) and bla_TEM1 have been dropped due to being of lesser importance for bacterial survival, or that resistance to clinically relevant concentrations of their corresponding antibiotics is sufficiently mediated by other genes. For example, the bla_OXA−1 chromosomal gene is highly conserved in Global-BAPS-1 S. flexneri 3a (Fig. 3), which confers resistance to a very similar substrate profile to that of bla_TEM−1,³² therefore, having both bla_OXA−1 and bla_TEM−1 is seemingly redundant, which would explain the loss of the latter, facilitated by its proximate transposases in the pKSR100-like plasmid. While the presence of both erm(B) and mph(A) is associated with a dual-action resulting in a higher MIC value for azithromycin,³³ it is clear that mph(A) alone can be sufficient in conferring a clinically resistant MIC value, which may explain the recent loss of erm(B) in S. flexneri 3a. The isolates remain susceptible to ciprofloxacin due to not having acquired relevant QRDR mutations, and are also susceptible to the third-generation cephalosporin ceftriaxone, having not acquired a gene in the bla_CTX−M family, unlike S. sonnei in the UK,²³ and internationally.²²

Although it is possible that stochastic population dynamics and other non-AMR changes in the organisms may have driven the re-emergence of S. flexneri 3a in the UK, we speculate that host immunity in the dense sexual transmission networks facilitating endemic shigellosis transmission may also play a role. In Israel, biennial peaks of S. sonnei have been described, alongside alternation between the S. flexneri 2a and S. flexneri 6 serotypes being the most common.³⁴ In comparison, the situation seen in England appears to be that there are more distanced peaks and shifts between S. flexneri 2a and S. flexneri 3a. Shifting dominance of S. flexneri serotypes has also been described in MSM in Montréal.³⁵ Negative frequency-dependent selection (NFDS) may also contribute to the re-emergence of rarer genotypes, as previously shown in Streptococcus pneumoniae.³⁶ NFDS is a type of natural selection, where the advantageous nature of a beneficial mutation or acquired gene diminishes over time, as the trait becomes more common in a population; particularly in the case of bacterial interactions with the human immune system, as a specific antigen becoming more common in a population of bacteria means that a greater proportion of human hosts will produce antibodies, which is especially noteworthy in interconnected sexual networks of MSM at a higher risk of shigellosis. Further research involving mathematical modelling and immunological studies (the usefulness of these reviewed elsewhere³⁷) are needed to elucidate the role of immunity and NFDS in the re-emergence of S. flexneri 3a and other serotypes and species of S. flexneri and Shigella, respectively.

To conclude, it is difficult to identify a single dominating factor in the re-emergence of S. flexneri 3a in the UK. Our research investigated a variety of variables which may have contributed to this re-emergence: human population demographic changes, host immunity, and changes in the genetic makeup of the S. flexneri 3a isolates. We showed that the population demographic of individuals contracting Shigella spp. is very similar to the previous outbreak, though the waning host immunity timelines that have been demonstrated in previous work may have contributed to this re-emergence. We see that there have been distinct genetic changes in the S. flexneri 3a pKSR100 plasmid over time, particularly in the loss of antimicrobial resistance genes erm(B) and bla_TEM−1, but retention of mph(A), which may confer a lower fitness cost to the bacteria and lead to its successful out-competition of other serotypes of S. flexneri and Shigella spp. We recommend that further research regarding the role of host immunity in the re-emergence of Shigella spp. is conducted through mathematical modelling and immunological studies to better predict and prevent future outbreaks of antimicrobial resistant, sexually transmissible enteric infections.

Data collection

Faecal samples of S. flexneri 3a were stored by the UK Health Security Agency (UKHSA) after being sent from hospitals and other diagnostic laboratories. Isolates with Sequence Read Run (SRR) numbers were sequenced by the UKHSA (2016–2020), while isolates with Experiment Run Archive (ERR) numbers were sequenced by the Wellcome Sanger Institute (2008–2013). The Second-Generation Surveillance System (SGSS) and Gastrointestinal Data Warehouse (GDW) were used as sources of the metadata in this study. More severe disease is more likely to be reported, which may represent a bias in case numbers and hospitalisation rates. All isolate accession numbers are available in Supplementary Data Table 1.

Data analysis

Where possible, samples from men who have sex with men (MSM) were identified from patient questionnaire data where sexual orientation was self-reported. Where this data was not available, samples from presumptive MSM were defined using previously validated parameters: males, aged 16 years and above, who had not declared any international travel in the last month.³⁸ Where travel history was not reported, this was assumed to mean that the patient did not travel.

Illumina whole-genome sequencing

Whole-genome-sequencing of Shigella isolates became a part of the standard surveillance routine at the UKHSA from August 2015 onwards. Where samples of S. flexneri 3a are sent to UKHSA from patients presenting with gastrointestinal illness, these samples had their DNA extracted for sequencing using an Illumina HiSeq 2500 machine. These Illumina reads were quality checked and trimmed and aligned to the S. flexneri serotype 2a strain 2457T reference genome (GenBank: AE014073.1). These reads were uploaded to the NCBI website for download by other individuals.

Nanopore sequencing

The two S. flexneri 3a isolates containing pKSR100-like plasmids and the one qacE Δ-1-containing plasmid underwent DNA extraction and Nanopore sequencing. Genomic DNA was extracted from overnight bacterial cultures using the MasterPure Complete DNA Purification Kit (Lucigen Corporations, Wisconsin, US) according to manufacturer’s instructions. DNA concentration, purity and quantity was assessed using Nanodrop spectrophotometry (DeNovix, Thermofisher, UK) and Qubit fluorometer (Invitrogen, US) according to manufacturer’s instructions. Sequencing libraries were prepared using the ligation sequencing kit - SQK-LSK109 - according to manufacturer’s instructions (Oxford Nanopore Technologies, Ltd, Oxford, UK). DNA libraries were sequenced using a MinION sequencer and FLO-MIN106 Flow Cell version R9.4.1 (Oxford Nanopore Technologies Ltd, Oxford, UK).

Genomic analysis

Genomic analysis of the S. flexneri 3a isolates sequenced by Wellcome Sanger Institute (2008–2013), and UKHSA (2016–2020), guided by data sharing agreements, was undertaken using a variety of methods, starting with downloading the FASTQ files available from the Sequence Read Archive (SRA) using the SRA toolkit (v 2.11.0) command: fastq-dump.³⁹ Trimmomatic (v 0.39)⁴⁰ was used to ‘trim’ the sequences used, with the following parameters: 2:30:10 LEADING : 20 TRAILING : 20 SLIDINGWINDOW : 4 : 20 MINLEN : 40. Afterwards, the quality of the trimmed reads were then checked by using fastQC,⁴¹ then combined with multiQC.⁴² The reference genome used was of S. flexneri 3a, (GenBank: GCA_904066025.1). Mapping the experimental genomes with the reference genome was undertaken by using Burrows-Wheel Aligner (BWA, v 0.7.17),⁴³ with PICARD (v 2.27.2)⁴⁴ being used to remove PCR duplicates. Quality of the mapping was checked by using qualimap (v 2.2.2)⁴⁵ bamqc.

Phylogenetic tree creation

Variant calling was performed using samtools (v 1.11)⁴⁶ and bcftools (v 1.9).⁴⁷ Using the parameters: -ntmax 25 -bb 1000 -m GTR + I + R + ASC, the phylogenetic tree was created using IQtree (v 2.2.0.3)⁴⁸ with the FASTA output generated by Gubbins (v 3.2.1)⁴⁹. The genome of each isolate was assembled using UNICYCLER (v 0.5.0),⁵⁰ where the quality of these assembled genomes was then checked using QUAST (v 5.02).⁵¹

Gene detection

Antimicrobial resistance (AMR), stress, and virulence genes were then identified using NCBI AMRFinder Plus (v 3.10.24),⁵² with the assembled genomes of each isolate and the following parameters: --nucleotide assembly.fasta --organism Escherichia --plus. Some (2.9%, n = 15/517) isolates were excluded from phylogenetic analysis due to quality control issues and not having an assigned SNP-address identifier.

Antimicrobial susceptibility testing

To investigate whether genotypic resistance translated to phenotypic resistance, 20 isolates from varying AMR profiles were subject to antimicrobial susceptibility testing using MIC Test Strips (MTS) manufactured by Liofilchem®, completed to obtain minimum inhibitory concentrations (MICs) as outlined by manufacturer’s directions. Results were interpreted according to clinical breakpoint definitions (v. 12) provided by the European Committee on Antimicrobial Susceptibility Testing (EUCAST).⁵³

Statistical analyses

Standard proportions were calculated.

Assembly and annotation

The isolates with the plasmids of interest were assembled and annotated. The fast5 read files generated from the MinION instrument were base-called and demultiplexed with Guppy (v 5.0, Oxford Nanopore Technologies Ltd. Oxford, UK). Processed read files were filtered using Filtlong (v 0.2.0), then assembled using Flye (v 2.9).⁵⁴ Racon (v 1.5.0)⁵⁵ was used to polish contigs with the Nanopore reads. Medaka (v 1.6.1) was used to polish Racon polished contigs, with Nanopore reads specifying the model r941_min_sup_g507. Polypolish (v 0.5.0)⁵⁶ was then used to polish with Illumina reads. The quality and statistics of each assembly was evaluated with QUAST (v 4.4.0)⁵¹ without a reference genome. Genomes were annotated using the Prokaryotic Genome Annotation Pipeline (PGAP).⁵⁷ The complete genome sequence data have been submitted to the National Centre for Biotechnology (NCBI) and have been deposited at GenBank under the BioProject number PRJNA999503.

Plasmid analyses

Plasmids (GBK files) were compared and visualised using Clinker,⁵⁸ embedded on the CompArative GEne Cluster Analysis Toolbox (CAGECAT, v 1.0) website, using default parameters. The 2020 pKSR100-like plasmid was assembled with Oxford Nanopore reads, and polished with Illumina reads from S. flexneri 3a (SRR12743036). The 2016 pKSR100-like plasmid was assembled with Oxford Nanopore reads, and polished with Illumina reads from S. flexneri 3a (SRR4897018). The 2014 pKSR100 reference plasmid was from S. flexneri 3a strain SF7955 (Plasmid accession LN624486.1). The 2020 qacE Δ-1-containing plasmid was assembled from Oxford Nanopore reads, and polished with Illumina reads from S. flexneri 3a (SRR1220418).

Ethics

No individual patient consent was required or sought as UKHSA has authority to handle patient data for public health monitoring and infection control under section 251 of the UK National Health Service Act of 2006 (previously section 60 of the Health and Social Care Act of 2001).

Code availability

No custom code was written to conduct this study. Details of bioinformatic packages and commands used can be found in the method section of this article.

Data availability

The sequence data for each of the 517 samples used in this study can be downloaded by using the SRA-toolkit and running a FASTQ-dump on the accession numbers of each isolate used in this study. See Supplementary Data 1 for the ERR and SRR accession numbers of the isolates used in this work, including anonymised patient and sample metadata. The Oxford Nanopore derived assemblies of isolates used in comparative plasmid analyses can be found under BioProject accession PRJNA999503.

Acknowledgements
This study is funded by the National Institute for Health and Care Research (NIHR) Health Protection Research Unit in Gastrointestinal Infections, a partnership between the UK Health Security Agency, the University of Liverpool, and the University of Warwick. The views expressed are those of the author(s) and not necessarily those of the NIHR, the UK Health Security Agency or the Department of Health and Social Care. This work was also supported by an MRC grant (MR/R020787/1, K.S.B) held and BBSRC grants (BB/V009184/1, K.S.B, M.D.S and BB/X003051/1 K.S.B).

Author Contributions

L.C.E.M, H.C, K.T, C.C, M.D.S, C.J and K.S.B contributed to the writing of the original draft. All authors contributed to reviewing and editing of subsequent drafts, formal analysis, the investigation, methodology, conceptualisation, validation, and visualisation. C.J and K.S.B contributed to funding acquisition, project administration, and providing resources. C.C, M.D.S, C.J and K.S.B contributed to supervision.

Competing Interests

The authors report no competing interests in this work.

Materials & Correspondence

Correspondence regarding isolates from the United Kingdom and the overall study should be directed to Dr. Claire Jenkins and Professor Kate S Baker.

DuPont, H. L., Levine, M. M., Hornick, R. B. & Formal, S. B. Inoculum Size in Shigellosis and Implications for Expected Mode of Transmission. The Journal of Infectious Diseases 159, 1126–1128 (1989). https://doi.org:10.1093/infdis/159.6.1126
Kotloff, K. L., Riddle, M. S., Platts-Mills, J. A., Pavlinac, P. & Zaidi, A. K. M. Shigellosis. Lancet 391, 801–812 (2018). https://doi.org:10.1016/s0140-6736(17)33296-8
Joyce, F. B. et al. Shigella is common in symptomatic and asymptomatic men who have sex with men visiting a sexual health clinic in Amsterdam. Sexually Transmitted Infections 98, 564 (2022). https://doi.org:10.1136/sextrans-2021-055274
Adams, C., Vose, A., Edmond, M. B. & Lyckholm, L. Shigella sonnei and hemolytic uremic syndrome: A case report and literature review. IDCases 8, 6–8 (2017). https://doi.org:https://doi.org/10.1016/j.idcr.2017.02.003
McGuire, E., Tiberi, S., Ciesielczuk, H. & Melzer, M. Shigellosis and toxic megacolon secondary to Shigella flexneri serotype 3a: The challenges of laboratory diagnosis. International Journal of Infectious Diseases 70, 104–106 (2018). https://doi.org:https://doi.org/10.1016/j.ijid.2018.02.020
Morduchowicz, G. et al. Shigella bacteremia in adults. A report of five cases and review of the literature. Archives of Internal Medicine 147, 2034–2037 (1987). https://doi.org:10.1001/archinte.147.11.2034
Kaeley, N., Kumar, M., Bhardwaj, B. & Nagasubramanyam, V. Shigella flexneri associated reactive arthritis - GI transmitted or sexually transmitted? Journal of Family Medicine and Primary Care 8, 1250 (2019). https://doi.org:10.4103/jfmpc.jfmpc_30_19
Thomas, M. E. & Tillett, H. E. Sonne dysentery in day schools and nurseries: an eighteen-year study in Edmonton. The Journal of Hygiene 71, 593–602 (1973). https://doi.org:10.1017/s0022172400046593
Rew, V. et al. Whole-genome sequencing revealed concurrent outbreaks of shigellosis in the English Orthodox Jewish Community caused by multiple importations of Shigella sonnei from Israel. Microbial Genomics 4 (2018). https://doi.org:10.1099/mgen.0.000170
Lampel, K. A., Formal, S. B. & Maurelli, A. T. A Brief History of Shigella. EcoSal Plus 8 (2018). https://doi.org:10.1128/ecosalplus.ESP-0006-2017
Bader, M. A. X., Pedersen, A. H. B., Williams, R., Spearman, J. & Anderson, H. Venereal Transmission of Shigellosis in Seattle-King County. Sexually Transmitted Diseases 4, 89–91 (1977).
Dritz, S. K. & Back, A. F. Letter: Shigella enteritis venereally transmitted. The New England Journal of Medicine 291, 1194 (1974).
Drusin, L. M., Genvert, G., Topf-Olstein, B. & Levy-Zombek, E. Shigellosis. Another sexually transmitted disease? The British Journal of Venereal Diseases 52, 348–350 (1976). https://doi.org:10.1136/sti.52.5.348
Morgan, O. et al. Shigella sonnei outbreak among homosexual men, London. Emerging Infectious Diseases 12, 1458–1460 (2006). https://doi.org:10.3201/eid1209.060282
Mohammed, S., Holly, O., Flanagan, Daniel, R. & Carrie, D. L. Factors associated with sexually transmitted Shigella in men who have sex with men: a systematic review. Sexually Transmitted Infections 99, 58 (2023). https://doi.org:10.1136/sextrans-2022-055583
Gilbart, V. L. et al. Sex, drugs and smart phone applications: findings from semistructured interviews with men who have sex with men diagnosed with Shigella flexneri 3a in England and Wales. Sexually Transmitted Infections 91, 598–602 (2015). https://doi.org:10.1136/sextrans-2015-052014
Daniel, R., John, D., Colin, F. & Nicolas, P.-S. Sexually transmitted Shigella flexneri and Shigella sonnei in men who have sex with men. Sexually Transmitted Infections 97, 244 (2021). https://doi.org:10.1136/sextrans-2020-054589
Farrar, W. E., Jr. & Eidson, M. Antibiotic Resistance in Shigella Mediated by R Factors. The Journal of Infectious Diseases 123, 477–484 (1971). https://doi.org:10.1093/infdis/123.5.477
Davies, J. R., Farrant, W. N. & Tomlinson, A. J. Further studies on the antibiotic resistance of Shigella sonnei. II. The acquisition of transferable antibiotic resistance in vivo. The Journal of Hygiene 66, 479–487 (1968). https://doi.org:10.1017/s0022172400041346
Chung The, H. et al. South Asia as a Reservoir for the Global Spread of Ciprofloxacin-Resistant Shigella sonnei: A Cross-Sectional Study. PLOS Medicine 13, e1002055 (2016). https://doi.org:10.1371/journal.pmed.1002055
Baker, K. S. et al. Horizontal antimicrobial resistance transfer drives epidemics of multiple Shigella species. Nature Communications 9, 1462 (2018). https://doi.org:10.1038/s41467-018-03949-8
Mason, L. C. E. et al. The evolution and international spread of extensively drug resistant Shigella sonnei. Nature Communications 14, 1983 (2023). https://doi.org:10.1038/s41467-023-37672-w
Mook, P. et al. ESBL-Producing and Macrolide-Resistant Shigella sonnei Infections among Men Who Have Sex with Men, England, 2015. Emerging Infectious Disease journal 22, 1948 (2016). https://doi.org:10.3201/eid2211.160653
World Health Organization (WHO). Guidelines for the control of shigellosis, including epidemics due to Shigella dysenteriae type 1. Report No. 924159330X, https://apps.who.int/iris/handle/10665/43252 (Geneva, 2005).
Simms, I. et al. Intensified shigellosis epidemic associated with sexual transmission in men who have sex with men - Shigella flexneri and S. sonnei in England, 2004 to end of February 2015. Eurosurveillance 20, 21097 (2015). https://doi.org:doi:https://doi.org/10.2807/1560-7917.ES2015.20.15.21097
Bardsley, M. et al. Persistent Transmission of Shigellosis in England Is Associated with a Recently Emerged Multidrug-Resistant Strain of Shigella sonnei. Journal of Clinical Microbiology 58,
10.1128
/jcm.01692 – 01619 (2020). https://doi.org:doi:10.1128/jcm.01692-19
Charles, H. et al. Outbreak of sexually transmitted, extensively drug-resistant Shigella sonnei in the UK, 2021-22: a descriptive epidemiological study. The Lancet Infectious Diseases 22, 1503–1510 (2022). https://doi.org:10.1016/s1473-3099(22)00370-x
Thorley, K. et al. Emergence of extensively drug-resistant and multidrug-resistant Shigella flexneri serotype 2a associated with sexual transmission among gay, bisexual, and other men who have sex with men, in England: a descriptive epidemiological study. The Lancet Infectious Diseases 23, 732–739 (2023). https://doi.org:https://doi.org/10.1016/S1473-3099(22)00807-6
Nicola, K. L. et al. Impact of the COVID-19 pandemic on gastrointestinal infection trends in England, February–July 2020. British Medical Journal Open 12, e050469 (2022). https://doi.org:10.1136/bmjopen-2021-050469
Baker, K. S. et al. Intercontinental dissemination of azithromycin-resistant shigellosis through sexual transmission: a cross-sectional study. The Lancet Infectious Diseases 15, 913–921 (2015). https://doi.org:10.1016/S1473-3099(15)00002-X
Tedijanto, C., Olesen, S. W., Grad, Y. H. & Lipsitch, M. Estimating the proportion of bystander selection for antibiotic resistance among potentially pathogenic bacterial flora. Proceedings of the National Academy of Sciences 115, E11988-E11995 (2018). https://doi.org:doi:10.1073/pnas.1810840115
Siu, L. K. et al. beta-lactamases in Shigella flexneri isolates from Hong Kong and Shanghai and a novel OXA-1-like beta-lactamase, OXA-30. Antimicrobial Agents and Chemotherapy 44, 2034–2038 (2000). https://doi.org:10.1128/aac.44.8.2034-2038.2000
Al-Marzooq, F., Ghazawi, A., Daoud, L. & Tariq, S. Boosting the Antibacterial Activity of Azithromycin on Multidrug-Resistant Escherichia coli by Efflux Pump Inhibition Coupled with Outer Membrane Permeabilization Induced by Phenylalanine-Arginine β-Naphthylamide. International Journal of Molecular Sciences 24 (2023). https://doi.org:10.3390/ijms24108662
Cohen, D. et al. Recent trends in the epidemiology of shigellosis in Israel. Epidemiology & Infection 142, 2583–2594 (2014). https://doi.org:10.1017/S0950268814000260
Ratnayake, R., Allard, R. & Pilon, P. A. Shifting dominance of Shigella species in men who have sex with men. Epidemiology & Infection 140, 2082–2086 (2012). https://doi.org:10.1017/S0950268812000738
Harrow, G. L. et al. Negative frequency-dependent selection and asymmetrical transformation stabilise multi-strain bacterial population structures. The International Society for Microbial Ecology Journal 15, 1523–1538 (2021). https://doi.org:10.1038/s41396-020-00867-w
Hay, J., Routledge, I., & Takahashi, S. Serodynamics: a review of methods for epidemiological inference using serological data. OSF Preprints (2023). https://doi.org:https://doi.org/10.31219/osf.io/kqdsn
Mitchell, H. D. et al. Use of whole-genome sequencing to identify clusters of Shigella flexneri associated with sexual transmission in men who have sex with men in England: a validation study using linked behavioural data. Microbial Genomics 5 (2019). https://doi.org:https://doi.org/10.1099/mgen.0.000311
SRAToolkitDevelopmentTeam. SRA Toolkit, <https://github.com/ncbi/sra-tools/wiki/01.-Downloading-SRA-Toolkit> (GitHub Repository, 2020).
Bolger, A. M., Lohse, M. & Usadel, B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120 (2014). https://doi.org:10.1093/bioinformatics/btu170
Andrews, S. FastQC: A Quality Control Tool for High Throughput Sequence Data < https://www.bioinformatics.babraham.ac.uk/projects/fastqc/> (2010).
Ewels, P., Magnusson, M., Lundin, S. & Käller, M. MultiQC: summarize analysis results for multiple tools and samples in a single report. Bioinformatics 32, 3047–3048 (2016). https://doi.org:10.1093/bioinformatics/btw354
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics 25, 1754–1760 (2009). https://doi.org:10.1093/bioinformatics/btp324
Broad Institute. Picard Toolkit v. 2.27.2, <https://broadinstitute.github.io/picard/> (Github Repository, 2019).
García-Alcalde, F. et al. Qualimap: evaluating next-generation sequencing alignment data. Bioinformatics 28, 2678–2679 (2012). https://doi.org:10.1093/bioinformatics/bts503
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009). https://doi.org:10.1093/bioinformatics/btp352
Danecek, P. et al. Twelve years of SAMtools and BCFtools. GigaScience 10 (2021). https://doi.org:10.1093/gigascience/giab008
Nguyen, L.-T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQ-TREE: A Fast and Effective Stochastic Algorithm for Estimating Maximum-Likelihood Phylogenies. Molecular Biology and Evolution 32, 268–274 (2014). https://doi.org:10.1093/molbev/msu300
Croucher, N. J. et al. Rapid phylogenetic analysis of large samples of recombinant bacterial whole genome sequences using Gubbins. Nucleic Acids Research 43, e15-e15 (2014). https://doi.org:10.1093/nar/gku1196
Wick, R. R., Judd, L. M., Gorrie, C. L. & Holt, K. E. Unicycler: Resolving bacterial genome assemblies from short and long sequencing reads. PLOS Computational Biology 13, e1005595 (2017). https://doi.org:10.1371/journal.pcbi.1005595
Gurevich, A., Saveliev, V., Vyahhi, N. & Tesler, G. QUAST: quality assessment tool for genome assemblies. Bioinformatics 29, 1072–1075 (2013). https://doi.org:10.1093/bioinformatics/btt086
Feldgarden, M. et al. AMRFinderPlus and the Reference Gene Catalog facilitate examination of the genomic links among antimicrobial resistance, stress response, and virulence. Scientific Reports 11, 12728 (2021). https://doi.org:10.1038/s41598-021-91456-0
European Committee on Antimicrobial Susceptibility Testing (EUCAST). Clinical Breakpoints and Dosing of Antibiotics-Bacteria (V 12.0) https://www.EUCAST.org/clinical_breakpoints/ (Vaxjo, 2022).
Kolmogorov, M., Yuan, J., Lin, Y. & Pevzner, P. A. Assembly of long, error-prone reads using repeat graphs. Nature Biotechnology 37, 540–546 (2019). https://doi.org:10.1038/s41587-019-0072-8
Vaser, R., Sović, I., Nagarajan, N. & Šikić, M. Fast and accurate de novo genome assembly from long uncorrected reads. Genome Research 27, 737–746 (2017). https://doi.org:10.1101/gr.214270.116
Wick, R. R. & Holt, K. E. Polypolish: Short-read polishing of long-read bacterial genome assemblies. PLOS Computational Biology 18, e1009802 (2022). https://doi.org:10.1371/journal.pcbi.1009802
Tatusova, T. et al. NCBI prokaryotic genome annotation pipeline. Nucleic Acids Research 44, 6614–6624 (2016). https://doi.org:10.1093/nar/gkw569
Gilchrist, C. L. M. & Chooi, Y.-H. clinker & clustermap.js: automatic generation of gene cluster comparison figures. Bioinformatics 37, 2473–2475 (2021). https://doi.org:10.1093/bioinformatics/btab007
Acknowledgements This study is funded by the National Institute for Health and Care Research (NIHR) Health Protection Research Unit in Gastrointestinal Infections, a partnership between the UK Health Security Agency, the University of Liverpool, and the University of Warwick. The views expressed are those of the author(s) and not necessarily those of the NIHR, the UK Health Security Agency or the Department of Health and Social Care. This work was also supported by an MRC grant (MR/R020787/1, K.S.B) held and BBSRC grants (BB/V009184/1, K.S.B, M.D.S and BB/X003051/1 K.S.B).

(Not answered)

SupplementaryData1.xlsx
Supplementary Data 1
SupplementaryInformation.docx

Download PDF

Journal Publication

published 02 Aug, 2024

Read the published version in npj Antimicrobials and Resistance →

Editorial decision: revise
20 Feb, 2024
Review #3 received at journal
20 Feb, 2024
Review #2 received at journal
16 Feb, 2024
Reviewer #3 agreed at journal
08 Feb, 2024
Review #1 received at journal
07 Feb, 2024
Reviewer #2 agreed at journal
07 Feb, 2024
Reviewer #1 agreed at journal
26 Jan, 2024
Reviewers invited by journal
25 Jan, 2024
Submission checks completed at journal
18 Dec, 2023
First submitted to journal
15 Dec, 2023
Editor assigned by journal
15 Dec, 2023

You are reading this latest preprint version

The re-emergence of sexually transmissible multidrug resistant Shigella flexneri 3a, England, United Kingdom

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Overview of demographic data

AMR gene loss on pKSR100 is facilitated by transposases

Discussion

Conclusion

Methods

Data collection

Data analysis

Illumina whole-genome sequencing

Nanopore sequencing

Genomic analysis

Phylogenetic tree creation

Gene detection

Antimicrobial susceptibility testing

Statistical analyses

Assembly and annotation

Plasmid analyses

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1