Molecular epidemiology of multi- and extensively-drug-resistant Mycobacterium tuberculosis in Ireland, 2001–2014

Accepted Manuscript Title: Molecular epidemiology of multi- and extensively-drug-resistant mycobacterium tuberculosis in ireland, 2001-2014 Author: E. Roycroft, R.F. O'Toole, M.M. Fitzgibbon, L. Montgomery, M. O'Meara, P. Downes, S. Jackson, J. O'Donnell, I.F. Laurenson, A.M. McLaughlin, J. Keane, T.R. Rogers PII: DOI: Reference:

S0163-4453(17)30315-8 https://doi.org/doi:10.1016/j.jinf.2017.10.002 YJINF 3997

To appear in:

Journal of Infection

Accepted date:

3-10-2017

Please cite this article as: E. Roycroft, R.F. O'Toole, M.M. Fitzgibbon, L. Montgomery, M. O'Meara, P. Downes, S. Jackson, J. O'Donnell, I.F. Laurenson, A.M. McLaughlin, J. Keane, T.R. Rogers, Molecular epidemiology of multi- and extensively-drug-resistant mycobacterium tuberculosis in ireland, 2001-2014, Journal of Infection (2017), https://doi.org/doi:10.1016/j.jinf.2017.10.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Molecular Epidemiology of Multi- and Extensively-DrugResistant Mycobacterium tuberculosis in Ireland, 2001-2014

16

Highlights

17 18 19 20 21 22 23

E Roycroft1,2, RF O’Toole2,3, MM Fitzgibbon1,2, L Montgomery1, M O’Meara4, P Downes4, S Jackson5, J O’Donnell5, IF Laurenson6, AM McLaughlin7, J Keane7, TR Rogers1,2 1. Irish Mycobacteria Reference Laboratory, Labmed Directorate, St. James’s Hospital, Dublin, Ireland 2. Department of Clinical Microbiology, Trinity Translational Medicine Institute, Trinity College, Dublin, Ireland 3. School of Medicine, Faculty of Health, University of Tasmania, Hobart, Australia 4. Department of Public Health, Dr. Steeven’s Hospital, Dublin, Ireland 5. Health Protection Surveillance Centre, Dublin, Ireland 6. Scottish Mycobacteria Reference Laboratory, Edinburgh, UK 7. Department of Respiratory Medicine, St. James’s Hospital and Trinity Translational Medicine Institute Trinity College Dublin, Ireland

    

Prevalence of MDR/XDR-TB in Ireland, while low, still poses a threat to Public Health High lineage diversity was found among MDR/XDR-TB strains ‘Cross-border’ European Union strains have been found in Ireland Evidence of in vivo micro-evolution of strains was found during the study Putative transmission between an Irish-born patient and a non-Irish born patient was discovered

24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

Abstract

45 46 47 48 49

Introduction

Objectives: The primary objective of this work was to examine the acquisition and spread of multidrug resistant (MDR) tuberculosis (TB) in Ireland. Methods: All available Mycobacterium tuberculosis complex (MTBC) isolates (n=42), from MDRTB cases diagnosed in Ireland between 2001 and 2014, were analysed using phenotypic drugsusceptibility testing, Mycobacterial-Interspersed-Repetitive-Units Variable-Number Tandem-Repeat (MIRU-VNTR) genotyping, and whole-genome sequencing (WGS). Results: The lineage distribution of the MDR-TB isolates comprised 54.7% Euro-American, 33.3% East Asian, 7.2% East African Indian, and 4.8% Indo-Oceanic. A significant association was identified between the East Asian Beijing sub-lineage and the relative risk of an isolate being MDR. Over 75% of MDR-TB cases were confirmed in non-Irish born individuals and 7 MIRU-VNTR genotypes were identical to clusters in other European countries indicating cross-border spread of MDR-TB to Ireland. WGS data provided the first evidence in Ireland of in vivo microevolution of MTBC isolates from drug-susceptible to MDR, and from MDR to extensively-drug resistant (XDR). In addition, they found that the katG S315T isoniazid and rpoB S450L rifampicin resistance mutations were dominant across the different MTBC lineages. Conclusions: Our molecular epidemiological analyses identified the spread of MDR-TB to Ireland from other jurisdictions and its potential to evolve to XDR-TB. Keywords: tuberculosis; molecular epidemiology; drug resistance

Drug resistance threatens the global management of tuberculosis [1, 2]. MDR-TB occurs when an isolate displays resistance to rifampicin and isoniazid. Extensively-drug-resistant TB displays resistance to the above plus a fluoroquinolone and a second-line injectable agent. One hundred and five countries in the world, including low-prevalence countries like Ireland, have reported XDR-TB to

Page 1 of 23

50 51 52 53 54 55 56

date [3-9]. Both MDR- and XDR-TB require complicated, lengthy treatment which can be complicated by many side-effects and may not be successful (50% global success rate reported in 2015) [10]. Treatment has been estimated to be approximately €10,000 for susceptible TB, €57,000 for MDR-TB, and over €170,000 for XDR-TB [11]. In 2015, there were an estimated 580,000 cases of multi- or extensively-drug-resistant tuberculosis (MDR/XDR-TB) worldwide. Immigration of people from high TB prevalence settings to the EU, as well as free-movement policies within the EU have been associated with its spread.

57 58 59 60 61 62 63

Due to the slow and fastidious growth patterns of TB, drug resistance can take many weeks to confirm in the diagnostic laboratory [12]. Phenotypic drug susceptibility testing is not always reliable, especially in the case of pyrazinamide [13, 14]. Various rapid molecular tests have been developed to determine drug resistance (e.g. line probe assays Hain GenoType MTBDRplus and MTBDRsl for direct respiratory specimens or cultures, and Cepheid GenXpert MTB/RIF for direct specimens) [1517]. While extremely useful, rapid molecular assays cannot cover all possible mutations, and do not take into account the development of novel mutations across the entire genome.

64 65 66 67 68 69 70 71 72 73

WGS is becoming more accessible for the diagnostic laboratory, although data analysis remains challenging [12, 18, 19]. Studies have shown that WGS using Illumina Next Generation Sequencing (NGS) is a 'rapid' and 'comprehensive' method for drug resistance profiling and epidemiology of TB [12]. The TB genome is relatively stable, therefore lends itself well to WGS [20-22]. Several international studies have been undertaken to determine the correlation between WGS for genotypic detection of drug-resistance associated mutations present in TB genomes and phenotypic drug susceptibility testing (DST) results [23-26], to the benefit of the wider scientific community working with TB. The better the correlation between genotypic resistance prediction and phenotypic DST, the more confidence can be placed in the variant, either as a resistance determinant, a benign polymorphism, or a phylogenetic (or lineage-defining) variation.

74 75 76 77 78 79 80 81

Walker and Kohl et al designed an algorithm to predict anti-tuberculous drug resistance in a retrospective cohort study which involved collation of a resistance mutation catalogue, 2,099 MTBC isolates used as a training set on which to test that catalogue, and a validation set of 1,552 further genomes on which the final set of drug-resistance-associated could be verified [27]. With their final set of resistance determinants, they were able to predict phenotypic resistance (for all drugs that had phenotypic DST available) in the validation set with sensitivity ranging from 45.5% (95% CI 16.776.6) for ofloxacin to 96.8% (CI 94.1-98.5) for rifampicin, and specificity ranging from 94.2% (CI 92.7-95.4) for ethambutol to 100% (99.4-100) for ofloxacin.

82 83 84 85 86 87

The origin and spread of MDR/XDR-TB in Ireland is largely unknown. An earlier study on TB drug resistance in an Irish hospital (1991-2001) found eight cases of MDR-TB among 864 culture-positive TB cases treated during the period 1991-2001 [28]. A previous history of TB and ‘foreign-national’ status were risk factors associated most with development of MDR-TB. Iatrogenic resistance due to treatment mismanagement was cited as a more important issue than non-Irish-nationals emigrating from the newer EU accession countries at the time.

88 89 90 91 92 93

Although the proportion of TB cases that are MDR is still relatively low, MDR/XDR-TB cases increased year on year in Ireland from 2003 (n=1) to a high of seven cases in 2007, and have been present in variable numbers since then [29]. Here, we performed the first molecular epidemiological study of MDR-TB in Ireland to determine the genotypes present and how they compare to those circulating in Europe and worldwide, to establish whether clusters of cases are present, and to examine if MDR strains are prevalent in both Irish-born and non-Irish-born patients, and whether

Page 2 of 23

94 95 96 97

there has been transmission between these two cohorts. It was hypothesised that MDR-TB is not readily transmitted in Ireland but is predominantly introduced from other jurisdictions, and that WGS could be extremely useful for rapid diagnosis of MDR/XDR-TB cases in the routine diagnostic laboratory.

98 99 100 101

Methods

102 103 104 105 106 107 108

Isolates were grown in MGITTM 960 liquid culture and, when the required threshold was reached, DST was performed as previously described using the reference standard method [30-32]. Neat concentrations of isolates (at 1-2 days after reaching their growth threshold) were grown in the presence of a critical concentration of the desired drug compared to a 1:100 dilution of each neat isolate (growth control). Table 2 details the drugs tested and their critical concentrations. Second-line DST was performed at reference laboratories in the UK. WHO-endorsed MGITTM DST was used in the majority of cases (15-17).

109 110 111 112

MIRU-VNTR genotyping DNA extraction was performed using crude or GenoLyse® (Hain Lifescience GmbH, Germany), according to the manufacturers’ instructions. Whole-genome DNA extraction was performed on previously-frozen, four-week-old MGITTM cultures, using an adapted, previously-published Autogen QuickGene protocol (Kurabo Bio-medical, Osaka, Japan) [33].

113 114 115 116

The Quant-iT Qubit™ dsDNA HS Assay Kit (Invitrogen, Carlsbad, California, USA) was used for DNA quantification of double-stranded DNA in both extracts and libraries, according to the manufacturers’ instructions. DNA library preparation was performed using an adapted Illumina Nextera-XT library preparation protocol [12].

117 118 119

For MIRU-VNTR genotyping, twenty-four informative loci were analysed according to the manufacturer’s instructions (Genoscreen, Lille, France, and GeneMapper software, Applied Biosystems, USA) [34].

120 121 122 123 124 125 126 127 128 129 130 131

WGS was performed using Illumina MiSeq 300-cycle Next Generation Sequencing (NGS) technology following the manufacturer’s instructions (Illumina, San Diego, California). Fastq files were imported into Geneious-R9 software, where 5’ and 3’ ends were trimmed based on > 5% chance of base-call error. Paired-end reads were mapped to the H37Rv (GenBank accession AL123456.3) reference genome [35]. All variants (inside and outside coding regions, synonymous and non-synonymous) were extracted, at a minimum read-depth of 5x, to a Microsoft Excel file. Minimum variant frequency (vf) was set at 0.25, while maximum variant P-value was set at 10-6 and minimum strand-bias at 10-5 when exceeding 65% bias. The Single Nucleotide Variants (SNVs) were filtered for those that were non-synonymous with a minimum vf of 90%. Filtering was performed manually according to an algorithm designed by Walker and Kohl et al using a custom macro to extract all variants from 25 candidate drug-resistance-associated genes (Table 1-2) [23]. Discrepancies were investigated and repeated in as far as it was possible. WGS could not be repeated due to resource constraints.

132 133 134 135 136

A neighbour-joining phylogeny, using MIRU-VNTR genotypes of the MDR/XDR-TB cohort, was built using the MIRU-VNTRplus database (Figure 1) [36]. A maximum likelihood phylogenetic inference tree was built in PhyML using the generalised time reversible (GTR) substitution model using the whole genomes of the cohort mapped to H37Rv (Figure 2) [37]. Isolates were analysed while taking into account a previously-calculated threshold for recent transmission. Five or fewer

Isolates were selected on the basis of their Drug Susceptibility Testing (DST) profile (Tables 1, 2). Thirty nine isolates were sequenced using WGS; three had insufficient DNA for sequencing, however DST data was available (IEMDR37-39).

Page 3 of 23

137 138 139

SNVs differentiating isolates indicated recent transmission, 5-12 SNVs indicated possible transmission, > 12 - < 20 SNVs represented a grey-zone, and > 20 SNVs ruled out recent transmission based on the estimated MTBC mutation rate of 0.5 SNV per genome per year [33].

140 141 142 143 144

Results

145 146 147 148 149 150 151 152 153

MDR/XDR-TB Patient Demographics Forty two MDR/XDR-TB isolates from 41 individual patients were collected nationally, from 20012014. A second case of MDR-TB in one patient was determined to be re-infection with a new strain rather than reactivation of latent MDR-TB. Seventeen isolates had been processed as specimens at the IMRL, while the remainder were referred from external laboratories. It cannot be guaranteed that all isolates from the time period are included since no laboratory is obliged to send isolates to the national reference laboratory, however since the number of isolates exceeds that reported by the HPSC, it could be assumed that the study is representative of at least the majority of cases. Patient demographics are detailed in Table 1. Whole genome sequence was not obtained for three isolates.

154 155 156 157 158 159 160

Cases were 58.5% male, 39% female (2.5% unknown). Median age at presentation was 34 years (IQR 29-40). Seventy-three per cent (n=30) of cases were pulmonary and 5% (n=2) were extra-pulmonary in nature (remainder unknown). Thirty-one cases (75.6%) were confirmed non-Irish-born and 4 cases were confirmed Irish-born (9.8%). Country of origin for the remaining cases was unknown (n=6, 14.6%). Countries (known) were extremely diverse: 48.6% Russia or former Soviet State, 11.4% Ireland, 22.9% Africa, and 17.1% Asia. No isolates originated in the North- or South-American continents.

161 162 163

Forty two MDR/XDR-TB isolates were MIRU-VNTR genotyped (Table 1). The distribution of lineages was as follows: 54.7% Euro-American Lineage 4 (n=23), 33.3% East Asian Lineage 2 (n=14), 7.2% East African Indian Lineage 3 (n=3) and 4.8% Indo Oceanic Lineage 1 (n=2).

164 165 166 167 168 169 170 171

EU ‘Cross-Border’ Clusters Identified Seven Multi-locus Variant Analysis (MLVA) MtbC15-9 clusters were identified: East Asian Lineage 1, sub-lineage Beijing 100-32 (n=7 isolates), 94-32 (n=3) and 1773-32 (n=2); Euro-American Lineage 4, sub-lineage Ghana 67-25 (n=2), sub-lineage LAM (Latin American Mediterranean) 843-52 (n=3) and sub-lineage Ural 163-15 (n=2); and EAI (East African Indian) Lineage 3, sub-lineage Delhi/CAS (Central Asian Strain) 8958-32 (n=2) (Figure 1, Table1). The remainder were unique (n=20). All seven of the clustered genotypes were found to be identical to genotypes that have been implicated in cross-border clustering by the European Centre for Disease Control (ECDC) from 2003-2014 [38].

172 173 174 175 176 177

WGS versus MIRU-VNTR Genotyping WGS analysis matched conventional genotyping results, with only slight variation (Figure 2 for further details). For example, Mycobacterial Locus Variant Analysis code (MLVA) MtbC15-9 895832 clustered IEMDR22 and 25 together (Figure 1). The WGS phylogeny showed that they were separated by just 2 SNVs (≤5 SNVs), confirming the MIRU-VNTR genotyping result within this household outbreak setting.

178 179

Conventional genotyping grouped IEMDR02/IEXDR1, IEMDR04, 07, 10, 15, 20 and 32 together, denoted by MtbC15-9 100-32 (Figure 1). When the whole genomes of these isolates were analysed,

From 2001-2014 there were 5,874 cases of TB reported in Ireland by the HPSC (Health Protection Surveillance Centre), of which 37 were MDR/XDR-TB, representing 0.6% of cases. The first HPSCrecorded case of MDR-TB was in 1998, although drug resistance was documented prior to this, and the first XDR-TB was reported in 2005 [12, 40, 41].

Page 4 of 23

180 181 182 183 184 185 186

14 SNVs was the most that distinguished any of them, which indicates that these lie in a grey-zone (i.e. >12 - <20 SNVs). Epidemiological information would be required to link the furthest cases. A sub-cluster which included IEMDR02/IEDXDR1 (collected 2005), IEMDR10 (collected 2007), and IEMDR32 (collected 2013) were within 5 SNVs of each other, which would suggest recent transmission. Any discrepancies that occurred were where WGS grouped isolates closer together on the tree than MIRU-VNTR genotyping had (e.g. IEMDR40 has grouped more closely to IEMDR01 and 41, and IEMDR29 has grouped more closely to IEMDR19 and 30 on the WGS tree).

187 188 189 190 191 192 193 194 195 196 197

Putative MDR-TB Transmission MtbC15-9 67-25 clustered IEMDR01 and IEMDR41 as identical (Ghana sub-lineage). WGS also indicates transmission between these two cases whereby the isolates differed by < 5 SNVs (Figure 2). IEMDR01EARLY represents an earlier isolate of IEMDR01 (collected 2001) while IEMDR41 was collected in 2004 from an asylum seeker from the Ivory Coast. IEMDR01 (Irish-born) was a longterm patient in a healthcare facility in Dublin, reported anecdotally to be non-compliant with medication. On further investigation, it was discovered that IEMDR41 also spent time at this facility. Since no other MIRU-VNTR genotypes matched these two in the Irish Mycobacteria Reference Laboratory (IMRL) database, no other individual who may have been involved in the chain of transmission was identified. Lineage could not be used to indicate direction of transmission since Ghana strains have been seen in Irish TB cases previously.

198 199 200 201 202 203 204

WGS showed that the isolates shared high-confidence mutations for isoniazid (Rv1908c katG S315T) and rifampicin (Rv0667 rpoB H445Y, or H526Y with E.coli numbering). The major difference between IEMDR01EARLY and IEMDR01 was that the latter isolate harboured a high-confidence mutation associated with phenotypic ethambutol resistance Rv3795 embB M306V that was not present in the earlier isolate. IEMDR41 did not harbour embB M306V. If transmission could be confirmed between the IEMDR01 and IEMDR41 patients, it would represent the first known transmission of MDR-TB in Ireland.

205 206 207 208 209 210 211 212

Phenotypic DST Table 2 describes DST results for the MDR/XDR-TB cohort. All isolates were resistant to rifampicin and isoniazid except one (IEMDR26, rifampicin resistant only). Apart from rifampicin and isoniazid, rifabutin (19/21, 90.5%), streptomycin (28/42, 67%), ethambutol (22/42 intermediate or resistant, 52.4%) and pyrazinamide (18/42, 42%) showed the highest frequency of resistance. IEXDR1/IEMDR02 [7], IEMDR27 and IEMDR32 displayed XDR-TB resistance patterns. Others exhibited pre-XDR-TB resistance patterns (IEMDR09 and IEMDR35). IEMDR26, although not a classical MDR-TB, displayed resistance to kanamycin, capreomycin and low-level moxifloxacin.

213 214 215 216 217

Drug-Resistance-Associated Mutations present Fastq files were mapped to the H37Rv reference genome (AL123456, NC_000962.3) with a mean coverage of 157, a mapping quality of 59, and with a mean mapped read percentage of 95.9% [39]. Resistance-associated mutations found within twenty-five candidate genes (and their promoter regions, i.e. 100bp upstream) are shown in Table 3.

218 219 220 221 222 223

The katG S315T mutation was by far the most common SNV found associated with isoniazid resistance (n=34/37, 92%), followed by fabG1 C-15T (n=9/37, 24%) which was seen in conjunction with katG S315T in seven cases. RpoB S450L (S531L with E.coli numbering) was the most common SNV associated with rifampicin resistance (n=25/39, 64%), followed by SNVs at rpoB codon 445 (position 526 with E.coli numbering, n=9, 23%). Ethambutol resistance was dominated by SNVs at embB codon 306 (n=19/27, 71%). Pyrazinamide mutations were scattered across the pncA gene.

Page 5 of 23

224 225 226 227

Streptomycin resistance was mostly associated with rpsL K43R (18/26, 69%). Fluoroquinolone (FQ) resistance-associated mutations were observed at a lower frequency than resistance mutations for any of the other drugs (n=6/39, 15%). Aminoglycoside resistance was mainly associated with kanamycin resistance (eis mutations, n=9/17, 53%).

228 229 230 231 232 233 234 235 236

Iatrogenic Progression from Susceptible to MDR-TB A pan-susceptible (i.e. susceptible to all first-line anti-TB agents) respiratory isolate was recovered from a sample taken in February 2004 from a 49-year-old Irish man with no known risk factors for TB. A second respiratory isolate (dated September 2004) was phenotypically resistant to isoniazid. In March 2005, a third isolate (IEMDR03, Table 1, Figures 1-2) displayed phenotypic resistance to both isoniazid and rifampicin. An rpoB mutation (S450L, or S531L with E.coli numbering) confirmed the presence of rifampicin resistance, while no isoniazid-resistance-associated mutation was found. Line probe assays and WGS were used to confirm the above results. Clinically, the patient was confirmed as having developed MDR-TB based on failure of first-line empiric therapy.

237 238 239 240 241 242 243 244 245 246

WGS of a third sequential isolate failed to discover a known mutation that corresponded with phenotypic isoniazid resistance. The second isolate had acquired a mutation in katG Q439R (coverage 137, vf 96.4%), which was not found in the first. However, the third isolate did not harbour that particular mutation, but instead a different katG mutation, D381A (coverage 177, vf 99.4%), as well as high-confidence rpoB S450L (coverage 243, vf 100%). A SNV in Rv3728 was also found in the latter isolate, V488I (coverage 151, vf 91.4%). The former katG Q439R mutation has been reported previously however has not been found to be of ‘high-confidence’ for drug resistance, and seems to have been lost prior to the development of classical MDR-TB in this case [40]. Neither katG D381A nor Rv3728 V488I was found in the literature. However, it has been hypothesised that any novel nonsynonymous katG variant could potentially play a role in resistance to isoniazid [41].

247 248 249 250 251 252 253 254

Iatrogenic Progression from MDR- to XDR-TB In February 2014, IEMDR27 was received in the IMRL from a 40-year-old, Latvian man who exhibited clinical symptoms of pulmonary TB, was a smoker, had a history of alcohol misuse, and originated from a high-TB-burden country (Table 1, Figures 1-2). The isolate was resistant to all firstline agents, as well as rifabutin, prothionamide, PAS and capreomycin (Table 2). The patient was diagnosed with MDR-TB. A second respiratory isolate recovered 9 months later was found to be resistant to the above plus fluoroquinolones. This represents the first known case of transition from MDR- to XDR-TB while on treatment.

255 256 257 258 259

Both isolates shared high-confidence mutations such as rpoB S450L (S531L E.coli numbering), katG S315T, embB Q497R, and pncA H51Y. One significant mutation was acquired over time in the gyrA gene D94Y (coverage 153, vf 98.7%), which has been strongly-associated with fluoroquinolone resistance. No high-confidence mutation associated with aminoglycoside resistance was found; an XDR-TB phenotype was diagnosed due to the presence of capreomycin resistance.

260 261 262 263

In each sequential case, only non-synonymous SNVs with a read depth above 20 and vf above 95% were considered. MIRU-VNTR genotypes were identical and no more than 5 SNVs separated the strains when 68 genes were analysed, therefore, these cases were considered recurrent infection rather than relapse.

264 265

MDR/XDR-TB cohort isolate raw NGS data has been submitted to the European Nucleotide Archive - Project PRJEB15076 (ERP016769), and ReSeqTB database.

Page 6 of 23

266 267 268 269 270 271 272 273 274 275 276

Discussion

277 278 279 280 281 282

In a comparison between the Irish MDR/XDR-TB cohort and European clusters reported by the ECDC from 2014, we share 7 'cross-border cluster' genotypes (MtbC15-9). This provides the first molecular evidence to support the hypothesis that for Ireland, as with other countries in Europe, movement of people within and from outside the EU has augmented the spread of MDR/XDR-TB [38]. This finding may provide an impetus for improved screening for TB, including MDR-TB, at points of entry and follow-up care post migration.

283 284 285

For the most part, the WGS phylogeny matched MIRU-VNTR genotyping clustering. Discrepancies between WGS and MIRU-VNTR genotyping have previously been documented and are most likely due, in part, to the latter’s significantly lower resolution [47-49].

286 287 288 289 290 291 292 293 294 295 296 297

When compared to worldwide studies on MDR/XDR-TB, the current study cohort is strikingly similar in its mutation make-up to worldwide strains. One study examined rifampicin (rpoB), isoniazid (katG, inhA), fluoroquinolone (gyrA, gyrB), and aminoglycoside (rrs, eis) resistance in 417 isolates from India, Moldova, the Philippines and South Africa [50]. The most common mutations found were also found in the current study. They noted little regional variation among strains, borne out by the current study. One variation they did see was with rrs and eis promoter mutations associated with kanamycin resistance. In India and South Africa, kanamycin resistance was mainly caused by rrs mutations, while in Moldova, it was mainly associated with eis mutations. Isolates in the current patient cohort who originated in India or South Africa were not resistant to kanamycin, or were not tested for kanamycin. The only Moldovan isolate did have an rrs mutation but was not phenotypically tested for kanamycin susceptibility. Perhaps these variations are to do with regional empirical drug regimens used and consequent selection pressure.

298 299 300 301 302 303 304 305 306 307 308

Another multi-national study analysed mutations in rifampicin (rpoB), isoniazid (katG, mabA-inhA promoter region), and fluoroquinolones (gyrA) in MDR/XDR-TB isolates from Belarus, China, Iran/Iraq, Honduras, Romania and Uganda (n=117) [51]. Researchers found a large regional difference in mutations and saw a narrower set of mutations and more fluoroquinolone resistance in the higher TB prevalence countries. The main rpoB mutation found in the current cohort (S450L, or S531L E. coli numbering) was also found across all sites in the multi-national study, followed by H445Y (or H526Y E. coli numbering) which was mainly found in Romanian isolates in the multinational study. KatG S315T was also found across the board, similar to the current study. They found that combined mutations in katG S315T and inhA C-15T were most commonly seen in Romanian isolates. This combination was found in our study in 7 isolates, but was not found in the only Romanian isolate.

This study, spanning 2001 to 2014, is the first documented in-depth analysis of the molecular epidemiology and drug resistance of MDR/XDR-TB using WGS in Ireland [7]. A detailed picture of MDR/XDR-TB present could be useful in the fight to control its spread. It is clear that most of the cases arose in those born outside Ireland, from a diverse range of high-burden TB areas. This corresponds with previous studies on drug resistance in Ireland and the UK [28, 42]. The high diversity of MDR/XDR-TB lineages matches the overall diversity of susceptible MTBC strains found here to date [43, 44]. For the Beijing sub-lineage, the relative risk of isolating an MDR/XDR-TB strain is 3.8, which is considered significant (p-value <0.0001). Association between this lineage and drug resistance has been observed elsewhere [45, 46]. No other lineage was found to be associated with drug resistance.

Page 7 of 23

309 310 311 312 313 314 315 316 317

Even though there has been evidence that cross-border clusters of MTBC genotypes have reached Ireland, there has been no confirmed transmission of MDR/XDR-TB to the native population. There has been transmission within immigrant families, and possibly between individuals living together who were from the same geographical area, but no transmission to an Irish individual, although reactivation of latent TB infection (LTBI) could prove otherwise in years to come. The only possible transmission events that could have occurred seem to be from Irish to non-Irish patients, which may be a consequence of non-compliance or ineffective treatment regimen, proving the hypothesis of the previous Irish study, that iatrogenic resistance could pose a greater threat to MDR-TB control and prevention than the import of MDR-TB from areas of high prevalence [28].

318 319 320 321 322 323 324 325 326 327 328 329 330 331 332

Selection pressure due to empiric drug regimens, ineffective drug penetration, and non-compliance with treatment, as with other types of bacteria, have been part of the reason that drug resistance has developed. IEMDR03 is an example of microevolution from susceptible to drug-resistant over 15 months. Although most of the MDR/XDR-TB isolates did not develop resistance within Ireland according to the evidence available, it is a risk worth bearing in mind for the Irish cohort. More research is required into sequential isolates in order to find out what occurs in vivo over time. There is a possibility that the third sequential isolate that had essentially ‘lost’ the katG Q439R mutation is a case of sampling bias, i.e. there was a sub-population with the mutation, and a sub-population without, and only the sub-population without was sampled in that particular isolate. There was no evidence of hetero-resistance found, however. IEMDR27 developed the mutations required to be defined as XDR-TB while on what clinicians deemed to be the correct regimen in hospital (therefore non-compliance was not a factor). Microevolution within a high-prevalence strain, such as the Haarlem sub-lineage, could be a more significant threat to TB control than import of MDR/XDR-TB. Sequential isolates in the current study has shown, for the first time, that microevolution can, and has, occurred in vivo. Further work should focus on sequential isolates from recurrent infection.

333 334 335 336 337 338 339 340

One Lithuanian patient had contracted MDR-TB twice within 3 years. The first was a Beijing strain and the second a LAM strain. Fourteen out of twenty-four MIRU-VNTR loci were distinct and their MtbC15-9 codes differed significantly (100-32 and 121-52, respectively). Their whole genomes were at least 48 SNVs apart, indicating that they were un-related. The patient had visited his home country within this timeframe and it is hypothesised that he contracted both strains in that region, a known high MDR-TB burden country [10]. The two isolates do not seem to have transferred resistance from one to the other. Nevertheless, it must be considered a possibility, when analysing MTBC, that there could be mixed infection, co-infection and/or hetero-resistance at play.

341 342 343 344 345 346 347 348 349 350

One Irish patient (IEMDR01), anecdotally associated with non-compliance, had drug-resistanceassociated mutations for rifampicin and isoniazid to begin with. An ethambutol mutation was subsequently acquired (embB M306V). This mutation has been associated with both susceptible and resistant strains. Perhaps selection pressure of varying drug concentrations encouraged the genetic change. IEMDR01 could have been involved in transmission with IEMDR41. IEMDR41 shares the same drug-resistance-associated mutations as IEMDR01, excluding the embB M306V. This could be seen as an indication of the direction of transmission, i.e. from non-Irish-born to Irish-born patient. This hypothesis is strengthened by the fact that the strain was designated ‘Ghana’, associated geographically with the Ivory Coast. However, epidemiological evidence suggests that the Irish-born patient presented first which would refute that hypothesis.

351 352 353

MDR/XDR-TB strains collected in Ireland from 2001 to 2014 were found to be highly diverse and similar to those found circulating in Europe (7 ‘cross-border’ clusters found). Drug-resistanceassociated mutations were largely similar despite lineage diversity, and also strikingly similar to those

Page 8 of 23

354 355 356 357 358 359 360

found in MDR/XDR-TB strains worldwide. While conclusive evidence of MDR/XDR-TB transmission within Ireland was not found, putative transmission was suggested, both by WGS and MIRU-VNTR genotyping. Microevolution of strains has occurred, as has reinfection with two different lineages of MTBC. Beijing lineage was found to be significantly-associated with multi-drug resistance. Our findings have important implications for future prevention and monitoring of MDR/XDR-TB in Ireland and affirm the benefits of WGS and in-depth molecular characterisation of clinical isolates.

361 362 363 364 365 366 367 368 369 370

Acknowledgements The Authors would like to thank IMRL colleagues who performed the identification, DST, and MIRU-VNTR genotyping on this MDR/XDR-TB cohort. Thank you also to the members of staff at the Scottish Mycobacteria Reference Laboratory and the National Mycobacterium Reference Unit, London, (especially Dr. Tim Brown) for second- and third-line DST performed on some of the isolates. We would also like to thank our Users from External Laboratories who sent the isolates to the IMRL for further analysis. Finally, we would like to acknowledge the patients involved in the study. Funding for this study was provided by the Clinical Microbiology Department of Trinity College Dublin and the LabMed Directorate at St. James’s Hospital, Dublin.

371 372

References

373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400

1. 2.

3.

4. 5.

6. 7. 8.

9. 10. 11. 12.

Dheda, K., et al., Global control of tuberculosis: from extensively drug-resistant to untreatable tuberculosis. Lancet Respir Med, 2014. 2(4): p. 321-38. Couvin, D. and N. Rastogi, Tuberculosis - A global emergency: Tools and methods to monitor, understand, and control the epidemic with specific example of the Beijing lineage. Tuberculosis (Edinb), 2015. 95 Suppl 1: p. S177-89. Singhal, P., et al., A study on pre-XDR & XDR tuberculosis & their prevalent genotypes in clinical isolates of Mycobacterium tuberculosis in north India. Indian J Med Res, 2016. 143(3): p. 341-7. Zhao, L.L., et al., Molecular characterisation of extensively drug-resistant Mycobacterium tuberculosis isolates in China. Int J Antimicrob Agents, 2015. 45(2): p. 137-43. Caceres, O., et al., Characterization of the Genetic Diversity of Extensively-Drug Resistant Mycobacterium tuberculosis Clinical Isolates from Pulmonary Tuberculosis Patients in Peru. PLoS One, 2014. 9(12): p. e112789. O'Toole, R.F., et al., Draft Genome Sequence of the First Isolate of Extensively DrugResistant Mycobacterium tuberculosis in New Zealand. Genome Announc, 2014. 2(3). Roycroft, E., et al., Draft Genome Sequence of the First Isolate of Extensively Drug-Resistant Mycobacterium tuberculosis in Ireland. Genome Announc, 2014. 2(5). Silva, C., et al., Mycobacterial interspersed repetitive unit typing and mutational profile for multidrug-resistant and extensively drug-resistant tuberculosis surveillance in Portugal: a 3year period overview. Int J Antimicrob Agents, 2014. 44(6): p. 546-51. Poudel, A., et al., Characterization of extensively drug-resistant Mycobacterium tuberculosis in Nepal. Tuberculosis (Edinb), 2013. 93(1): p. 84-8. WHO, World Health Organisation Global Tuberculosis Report 2015. 2015, World Health Organisation: Online. p. 1-192. Diel, R., et al., Costs of tuberculosis disease in the European Union: a systematic analysis and cost calculation. Eur Respir J, 2014. 43(2): p. 554-65. Pankhurst, L.J., et al., Rapid, comprehensive, and affordable mycobacterial diagnosis with whole-genome sequencing: a prospective study. Lancet Respir Med, 2016. 4(1): p. 49-58.

Page 9 of 23

401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455

13.

14.

15.

16.

17. 18. 19. 20. 21. 22. 23.

24. 25.

26.

27.

28. 29.

30. 31.

32.

Chedore, P., et al., Potential for erroneous results indicating resistance when using the Bactec MGIT 960 system for testing susceptibility of Mycobacterium tuberculosis to pyrazinamide. J Clin Microbiol, 2010. 48(1): p. 300-1. Zhang, Y., S. Permar, and Z. Sun, Conditions that may affect the results of susceptibility testing of Mycobacterium tuberculosis to pyrazinamide. J Med Microbiol, 2002. 51(1): p. 429. Hillemann, D., et al., Use of the genotype MTBDR assay for rapid detection of rifampin and isoniazid resistance in Mycobacterium tuberculosis complex isolates. J Clin Microbiol, 2005. 43(8): p. 3699-703. Hillemann, D., S. Rüsch-Gerdes, and E. Richter, Feasibility of the GenoType MTBDRsl assay for fluoroquinolone, amikacin-capreomycin, and ethambutol resistance testing of Mycobacterium tuberculosis strains and clinical specimens. J Clin Microbiol, 2009. 47(6): p. 1767-72. Boehme, C.C., et al., Rapid molecular detection of tuberculosis and rifampin resistance. N Engl J Med, 2010. 363(11): p. 1005-15. Didelot, X., et al., Transforming clinical microbiology with bacterial genome sequencing. Nat Rev Genet, 2012. 13(9): p. 601-12. Reuter, S., et al., Rapid bacterial whole-genome sequencing to enhance diagnostic and public health microbiology. JAMA Intern Med, 2013. 173(15): p. 1397-404. Cole, S.T. and B.G. Barrell, Analysis of the genome of Mycobacterium tuberculosis H37Rv. Novartis Found Symp, 1998. 217: p. 160-72; discussion 172-7. Cole, S.T., et al., Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature, 1998. 393(6685): p. 537-44. Eldholm, V. and F. Balloux, Antimicrobial Resistance in Mycobacterium tuberculosis: The Odd One Out. Trends Microbiol, 2016. Walker, T.M., et al., Whole-genome sequencing for prediction of Mycobacterium tuberculosis drug susceptibility and resistance: a retrospective cohort study. Lancet Infect Dis, 2015. 15(10): p. 1193-202. Coll, F., et al., Rapid determination of anti-tuberculosis drug resistance from whole-genome sequences. Genome Med, 2015. 7(1): p. 51. Feuerriegel, S., et al., PhyResSE: a Web Tool Delineating Mycobacterium tuberculosis Antibiotic Resistance and Lineage from Whole-Genome Sequencing Data. J Clin Microbiol, 2015. 53(6): p. 1908-14. Schito, M. and D.L. Dolinger, A Collaborative Approach for 'ReSeq-ing' Mycobacterium tuberculosis Drug Resistance: Convergence for Drug and Diagnostic Developers. EBioMedicine, 2015. 2(10): p. 1262-5. Walker, T.M., et al., Whole-genome sequencing for prediction of Mycobacterium tuberculosis drug susceptibility and resistance: a retrospective cohort study. The Lancet Infectious Diseases, (0). Min, S.M., et al., Antibiotic resistant tuberculosis and bovine tuberculosis in an Irish hospital population (1991 to 2001). Ir Med J, 2005. 98(2): p. 38-40. HPSC. Reports on the Epidemiology of TB in Ireland. [Report (Online), available at: http://www.hpsc.ie/A-Z/VaccinePreventable/TuberculosisTB/Epidemiology/AnnualReports/] 1998-2014; Reports on the Epidemiology of TB in Ireland from 1998 to 2014]. Available from: http://www.hpsc.ie/AZ/VaccinePreventable/TuberculosisTB/Epidemiology/AnnualReports/. Tortoli, E., et al., Use of BACTEC MGIT 960 for recovery of mycobacteria from clinical specimens: multicenter study. J Clin Microbiol, 1999. 37(11): p. 3578-82. Tortoli, E., et al., Evaluation of automated BACTEC MGIT 960 system for testing susceptibility of Mycobacterium tuberculosis to four major antituberculous drugs: comparison with the radiometric BACTEC 460TB method and the agar plate method of proportion. J Clin Microbiol, 2002. 40(2): p. 607-10. Rüsch-Gerdes, S., et al., Multicenter laboratory validation of the BACTEC MGIT 960 technique for testing susceptibilities of Mycobacterium tuberculosis to classical second-line drugs and newer antimicrobials. J Clin Microbiol, 2006. 44(3): p. 688-92.

Page 10 of 23

456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501

33. 34.

35. 36.

37. 38. 39. 40.

41. 42. 43. 44. 45. 46. 47.

48.

49. 50. 51.

Walker, T.M., et al., Whole-genome sequencing to delineate Mycobacterium tuberculosis outbreaks: a retrospective observational study. Lancet Infect Dis, 2013. 13(2): p. 137-46. Allix-Beguec, C., M. Fauville-Dufaux, and P. Supply, Three-year population-based evaluation of standardized mycobacterial interspersed repetitive-unit-variable-number tandem-repeat typing of Mycobacterium tuberculosis. J Clin Microbiol, 2008. 46(4): p. 1398406. Kearse, M., et al., Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics, 2012. 28(12): p. 1647-9. Allix-Beguec, C., et al., Evaluation and strategy for use of MIRU-VNTRplus, a multifunctional database for online analysis of genotyping data and phylogenetic identification of Mycobacterium tuberculosis complex isolates. J Clin Microbiol, 2008. 46(8): p. 2692-9. Guindon, S., et al., New algorithms and methods to estimate maximum-likelihood phylogenies: assessing the performance of PhyML 3.0. Syst Biol, 2010. 59(3): p. 307-21. ECDC, Molecular Typing for Surveillance of Multidrug-resistant tuberculosis in the EU/EEA. January 2016, European Centre for Disease Control: Online. Langmead, B., Aligning short sequencing reads with Bowtie. Curr Protoc Bioinformatics, 2010. Chapter 11: p. Unit 11.7. Jeeves, R.E., et al., Mycobacterium tuberculosis Is Resistant to Isoniazid at a Slow Growth Rate by Single Nucleotide Polymorphisms in katG Codon Ser315. PLoS One, 2015. 10(9): p. e0138253. Torres, J.N., et al., Novel katG mutations causing isoniazid resistance in clinical M. tuberculosis isolates. Emerg Microbes Infect, 2015. 4(7): p. e42. Abubakar, I., et al., Drug-resistant tuberculosis: time for visionary political leadership. Lancet Infect Dis, 2013. 13(6): p. 529-39. Fitzgibbon, M.M., et al., A snapshot of genetic lineages of Mycobacterium tuberculosis in Ireland over a two-year period, 2010 and 2011. Euro Surveill, 2013. 18(3). Ojo, O.O., et al., Molecular epidemiology of Mycobacterium tuberculosis clinical isolates in Southwest Ireland. Infection, Genetics and Evolution, 2010. 10(7): p. 1110-1116. Merker, M., et al., Evolutionary history and global spread of the Mycobacterium tuberculosis Beijing lineage. Nat Genet, 2015. Glynn, J.R., et al., Worldwide occurrence of Beijing/W strains of Mycobacterium tuberculosis: a systematic review. Emerg Infect Dis, 2002. 8(8): p. 843-9. Stucki, D., et al., Standard Genotyping Overestimates Transmission of Mycobacterium tuberculosis among Immigrants in a Low-Incidence Country. J Clin Microbiol, 2016. 54(7): p. 1862-70. Roetzer, A., et al., Whole genome sequencing versus traditional genotyping for investigation of a Mycobacterium tuberculosis outbreak: a longitudinal molecular epidemiological study. PLoS Med, 2013. 10(2): p. e1001387. Gardy, J.L., et al., Whole-genome sequencing and social-network analysis of a tuberculosis outbreak. N Engl J Med, 2011. 364(8): p. 730-9. Rodwell, T.C., et al., Predicting extensively drug-resistant Mycobacterium tuberculosis phenotypes with genetic mutations. J Clin Microbiol, 2014. 52(3): p. 781-9. Rosales-Klintz, S., et al., Drug resistance-related mutations in multidrug-resistant Mycobacterium tuberculosis isolates from diverse geographical regions. Int J Mycobacteriol, 2012. 1(3): p. 124-30.

502 503 504

Page 11 of 23

505 506 507 508

Figure 1. MIRU-VNTRplus database neighbour joining phylogeny constructed with MIRU-VNTR genotyping data from MDR/XDR-TB isolates collected in Ireland from 2001-2014.

509 510

Colours indicate the lineage assigned according to MIRU-VNTR genotyping pattern detected. Details on labels include isolate name, lineage, country of origin (‘?’ = estimated) and MtbC 15-9 code.

Page 12 of 23

511 512

513 514 515

Figure 2. Polar radial WGS Maximum likelihood phylogeny constructed using PhyML from the whole genomes of MDR/XDR-TB strains isolated in Ireland from 2001-2014.

516 517 518 519

MtbC15-9 codes are incorporated into the taxa labels. Taxa in blue represent those where WGS and MtbC15-9 match. Minor differences occurred where WGS provided higher resolution differentiation of strains. IEMDR08 not included due to low coverage. WGS not peformed on IEMDR37-39 inlcusive.

520

Page 13 of 23

521

Table 1. MDR/XDR-TB Patient Cohort Demographics. Study No.

M/F

Address (County)

Country of Origin

Age

TB Risk Factors

Specimen Type

Collected

Lineage

MIRU-VNTR Genotype

MtbC 15-9

MDR01EARLY

M

TIPPERARY

IRELAND

34

NK

SPUTUM

2001

GHANA

223263342334425143233613

67-25

IEMDR42

NK

NK

ANGOLA

18

NK

SPUTUM

2001

BEIJING

244233352644425153353823

94-32

IEMDR15

F

DUBLIN

ALBANIA

27

NK

NK

2003

BEIJING

244233352644425173353723

100-32

IEMDR01

M

TIPPERARY

IRELAND

37

NK

SPUTUM

2004

GHANA

223263342334425143233613

67-25

IEMDR04*

M

DUBLIN

LITHUANIA

34

1,2

SPUTUM

2004

BEIJING

244233352644425173353723

100-32

IEMDR41

M

LOUTH

IVORY COAST

33

NK

SPUTUM

2004

GHANA

223263342334425143233613

67-25

IEXDR1/IEMDR02 (7)

F

DUBLIN

LITHUANIA

26

1

SPUTUM

2005

BEIJING

244233352644425173353723

100-32

IEMDR03

M

DUBLIN

IRELAND

49

NK

SPUTUM

2005

EAI

215834372943266223342713

1740-44

IEMDR05

M

DUBLIN

INDIA

27

1,3

NK

2005

EAI

22433438236314a223374613

1741-212

IEMDR16

M

NK

NK

19

NK

NK

2005

LAM

132275332224126153322_22

?-51

IEMDR36

M

NK

NK

67

NK

SPUTUM

2005

EURO-AMERICAN

214213322434226153334122

?-62

IEMDR37 (WGS NP)

F

NK

NK

80

NK

SPUTUM

2005

X

224224342234425153332832

?-15

IEMDR38 (WGS NP)

M

KILDARE

NK

40

NK

SPUTUM

2005

DELHI/CAS

232236442244225143353743

9045-32

IEMDR39 (WGS NP)

M

NK

NK

76

NK

SPUTUM

2005

EURO-AMERICAN

224243122424225153335522

1286-15

IEMDR40

M

DUBLIN

NK

53

NK

SPUTUM

2005

HAARLEM

224224332334425153333732

?-15

IEMDR14

F

NK

MONGOLIA

28

1

NK

2006

BEIJING

244233362544425173353823

1773-32

IEMDR06

F

DUBLIN

NIGERIA

31

1

SPUTUM

2007

S

233343312434225143233a22

1742-17

IEMDR07

M

SLIGO

LITHUANIA

24

NK

SPUTUM

2007

BEIJING

244233352644425173353723

100-32

Page 14 of 23

IEMDR08

F

DUBLIN

GEORGIA

33

1,4,5

SPUTUM

2007

LAM

132254332224125153322622

843-52

IEMDR09*

M

DUBLIN

LITHUANIA

37

1,2,4

SPUTUM

2007

LAM

132244332224125153322622

121-52

IEMDR10

M

MEATH

LITHUANIA

37

1

NK

2007

BEIJING

244233352644425173353723

100-32

IEMDR11

F

NK

ZIMBABWE

29

1,6

URINE

2008

LAM

244213232424116143522102

1743-54

IEMDR12

F

DUBLIN

AZERBAIJAN

29

1,4,5

SPUTUM

2009

BEIJING

244233352644425153353623

1065-32

IEMDR13

M

DUBLIN

ZIMBABWE

42

1,6

NK

2009

S

233343312334225143233a22

1772-17

IEMDR17

M

DUBLIN

ROMANIA

19

1

SPUTUM

2010

EURO-AMERICAN

222253122434225143335522

1655-15

IEMDR18

M

WESTMEATH

CHINA

26

1,6

BAL

2010

BEIJING

244233342__4425173353323

?-32

IEMDR26

F

NK

IRELAND

45

NK

NK

2010

HAARLEM

213226332434425153333732

?-15

IEMDR19

M

DUBLIN

LATVIA

39

1,6

SPUTUM

2011

URAL

235237232244425113323632

163-15

IEMDR20

M

DUBLIN

UKRAINE

30

1,4,5,6,7

SPUTUM

2011

BEIJING

244233352644425173353723

100-32

IEMDR21

F

DUBLIN

SOUTH AFRICA

34

1,6

NK

2011

LAM

134254332224122143322722

8075-482

IEMDR28

F

MEATH

MOLDOVA

29

1,4,6

SPUTUM

2011

LAM

132253(4)3322241251533223(6)22

?-52

IEMDR22

M

GALWAY

INDIA

16

1

NK

2012

DELHI/CAS

222236452244225173353623

8958-32

IEMDR23

F

DUBLIN

MONGOLIA

29

1

SPUTUM

2012

BEIJING

244233362544425173353823

1773-32

IEMDR24

M

MEATH

NIGERIA

37

1

SPUTUM

2012

LAM

22421333154422515333_522

?-26

IEMDR25

F

GALWAY

INDIA

38

1

PLEURAL BX

2012

DELHI/CAS

222236452244225173353623

8958-32

IEMDR33

M

NK

LITHUANIA

50

NK

LUNG

2012

BEIJING

244233352644425153353823

94-32

IEMDR34

M

CORK

RUSSIA

49

1

SPUTUM

2012

BEIJING

244233352644425153353823

94-32

IEMDR29

F

DUBLIN

SOMALIA

24

1

BAL

2013

EURO-AMERICAN

214225132134425113333a22

12428-15

IEMDR30

M

LAOIS

LITHUANIA

44

1,4,9

SPUTUM

2013

URAL

235237232244425113323632

163-15

Page 15 of 23

IEMDR31

F

DUBLIN

ZIMBABWE

32

1,4,6

SPUTUM

2013

LAM

244214132324116152442822

12880-443

IEMDR32

F

DUBLIN

RUSSIA

37

1,6

SPUTUM

2013

BEIJING

244233352644425173353723

100-32

IEMDR27

M

DUBLIN

LATVIA

40

1,2,4,8

SPUTUM

2014

LAM

132254332224125153322622

843-52

IEMDR27LATER

M

DUBLIN

LATVIA

40

1,2,4,8

SPUTUM

2014

LAM

132244332224125153322622

843-52

IEMDR35

M

ROSCOMMON

LATVIA

53

6,8

SPUTUM

2014

LAM

132254332224125153322622

843-52

522 523 524 525 526 527 528

Demographics (in chronological order) included gender, address (county), country of origin, age at presentation, risk factors for TB, specimen type, year of collection, sub-lineage, MIRU-VNTR genotype and MtbC15-9 code. * indicates two different MDR strains separately isolated from the same patient. TB risk factors: 1-From a high TB burden country, 2-smoker, 3-occupation (Health Care Worker), 4-previous history of TB, 5-non-compliance with treatment, 6immunocompromised, 7-intra-venous drug user (IVDU), 8-alcohol misuse and 9-incarcerated in prison. Age corresponds to age in years at time of presentation. WGS NP – whole genome sequencing not performed, BAL – bronchoalveolar lavage, NK – not known. Sequential isolates were included (IEMDR01 and IEMDR01EARLY, MDR27 and MDR27LATER).

529 530

Page 16 of 23

531

Table 2. MDR/XDR-TB conventional phenotypic drug susceptibility testing (DST) results, 2001-2014. MDR no.

S M

S M

RI F

IN H

IN H

EM B

PZ A

CY C

PA S

CF Z

ET I

PR O

AM K

KA N

CA P

CI P

MX F

MX F

OF X

RF B

CL A

LZ D

CRITICAL CONCENTRATION µg/ml

1.0

4.0

1.0

0.1

0.4

5.0

100

40.0

2.0

4.0

5.0

2.5

1.0

2.5

2.5

1.0

0.25

2.0

2.0

0.5

0.5

1.0

IEMDR03

S

_

R

R

_

S

S

_

_

_

_

_

S

S

S

_

S

S

S

_

_

_

IEMDR05

R

_

R

R

_

R

S

S

S

S

S

_

_

_

_

_

_

_

_

_

S

_

IEMDR19

R

_

R

R

_

R

S

_

_

S

_

S

S

_

S

S

_

_

_

R

S

_

IEMDR30

R

R

R

R

R

R

S

_

_

_

_

S

S

S

S

_

S

S

S

_

_

S

IEMDR22

S

_

R

R

R

S

S

_

_

_

_

S

S

S

S

_

S

S

S

_

_

_

IEMDR25

S

_

R

R

R

S

S

_

_

_

_

S

S

S

S

_

S

S

S

_

_

_

IEMDR38

S

_

R

R

_

S

_

_

_

_

_

_

_

_

_

_

_

_

_

_

_

_

IEMDR14

R

_

R

R

_

R

R

_

_

S

_

R

S

_

S

S

_

_

_

R

S

_

IEMDR23

R

R

R

R

R

S

R

_

_

_

_

R

S

S

S

_

S

S

S

_

_

_

IEMDR18

R

_

R

R

_

R

R

_

S

S

_

S

S

S

S

S

S

S

_

R

R

S

IEMDR12

S

_

R

R

R

R

S

_

S

S

_

R

S

S

S

R

R

R

R

R

S

_

IEMDR33

R

R

R

R

R

R

S

_

_

_

_

_

R

R

S

_

S

S

S

_

_

_

IEMDR34

R

_

R

R

_

S

R

_

_

_

_

_

S

R

S

_

S

S

S

_

_

_

IEMDR42

R

_

R

R

_

R

S

_

_

_

_

_

S

S

S

_

S

S

S

_

_

_

IEXDR1/IEMDR02

R

_

R

R

_

R

R

S

R

S

S

S

R

_

S

S/R

_

_

_

R

S/R

_

IEMDR04

R

_

R

R

R

S

S

S

S

S

R

S

_

_

_

_

_

_

_

_

S

_

IEMDR07

R

_

R

R

R

R

R

_

_

S

_

S

S

_

S

S

_

_

_

R

R

_

Page 17 of 23

IEMDR10

R

_

R

R

R

R

R

_

_

S

_

S

S

_

S

S

_

_

_

R

R

_

IEMDR15

R

_

R

R

_

I

R

_

_

_

_

_

S

R

S

_

S

S

S

_

_

_

IEMDR20

R

_

R

R

_

R

R

_

_

S

_

R

S

R

S

S

S

S

S

R

_

S

IEMDR32

R

R

R

R

R

R

R

S

S

R

S

S

R

R

R

_

R

R

R

R

_

S

IEMDR01

R

_

R

R

R

R

R

S

S/R

_

R

_

_

_

_

_

_

_

_

_

_

_

IEMDR41

R

_

R

R

_

S

R

_

_

_

_

_

S

S

S

_

S

S

S

_

_

_

IEMDR06

R

_

R

R

R

R

R

_

_

_

_

S

_

_

_

_

_

_

_

S

_

_

IEMDR13

R

_

R

R

_

I

S

_

S

R

_

S

S

_

S

S

_

_

_

R

S

_

IEMDR08

R

_

R

R

R

S

S

_

_

_

_

R

S

_

S

_

_

_

_

_

_

_

IEMDR27

R

R

R

R

R

R

R

S

R

_

S

R

S

S

R

_

R

R

R

R

_

S

IEMDR35

R

R

R

R

R

R

R

_

_

_

_

R

S

_

R

_

S

S

S

R

_

S

IEMDR28

R

_

R

R

R

S

S

_

S

S

_

R

S

_

S

S

_

_

_

S

_

S

IEMDR09

R

_

R

R

_

R

R

_

S

_

S

_

_

S

R

_

_

_

R

R

_

IEMDR16

R

_

R

R

_

S

R

_

_

_

_

_

R

R

S

_

S

S

S

_

_

_

IEMDR21

R

_

R

R

R

R

S

_

_

S

_

S

S

_

S

S

S

S

_

R

_

_

IEMDR11

S

_

R

R

R

R

R

_

_

S

_

S

S

_

S

S

_

_

_

R

R

_

IEMDR31

S

S

R

R

R

S/R

S

_

_

_

S

S

S

S

S

_

S

S

S

R

_

S

IEMDR17

S

_

R

R

R

S

S

_

_

S

_

S

_

_

S

_

_

_

R

R

_

IEMDR39

S

_

R

R

_

S

S

_

_

_

_

_

_

_

_

_

_

_

_

_

_

_

IEMDR36

S

_

R

R

_

S

S

_

_

_

_

_

S

S

S

_

R

S

S

_

_

_

IEMDR24

S

_

R

R

R

S

S

_

R

_

_

S

S

S

S

S

S

S

S

R

_

S

Page 18 of 23

IEMDR26

S

_

R

S

_

S

S

_

_

_

_

_

S

R

R

_

R

S

S

_

_

_

IEMDR37

S

_

R

R

_

S

S

_

_

_

_

_

_

_

_

_

_

_

_

_

_

_

IEMDR40

S

_

R

R

_

S

S

_

_

_

_

_

S

R

S

_

S

S

S

_

_

_

IEMDR29

R

R

R

R

R

S

R

_

S

_

_

S/R

R

R

R

S

S

S

S

R

_

S

532 533 534 535 536 537

Isolates are grouped according to their MIRU-VNTR clustering pattern (Figure 1). IEXDR1 was previously published in 2014 [1]. S susceptible, R resistant, S/R may have been susceptible or resistant, depending on method or laboratory, I intermediate, INH isoniazid, RIF rifampicin, EMB ethambutol, PZA pyrazinamide, SM streptomycin, AMK amikacin, KAN kanamycin, CAP capreomycin, ETI ethionamide, PAS para-amino-salicylic acid, LZD linezolid, CFZ clofazimine, CYC cycloserine, PRO prothionamide, CIP ciprofloxacin, MXF moxifloxacin, OFX ofloxacin, RFB rifabutin, CLA clarithromycin.

538 539

Page 19 of 23

540 541 542

Table 3. Single Nucleotide Variations (SNVs), that have been associated with resistance to isoniazid, rifampicin, ethambutol, streptomycin, pyrazinamide, fluoroquinolones and aminoglycosides, found in MDR/XDR-TB isolates collected in Ireland from 20012014. STUDY NO.

INH

RIF

EMB

PZA

SM

FQ

AG

Rv

Rv2428, 1483, 1484, 1908c, 1854c, 2245

Rv0667, 0668

Rv3794, 3795, 3793, 1267c, 0342, 0343, 3264c, 3266c

Rv163 0, 2043c

Rv1630, 3919, 1694, Rvnr01/MTB00 0019

Rv0006,00 05

Rv3919, Rvnr01/MTB000 019, Rv1694, 2416c,

GENE

ahpC, fabG1 (mabA), inhA, katG, ndh, kasA

rpoB, rpoC

embA, embB, embC, embR, iniA, iniC, manB, rmlD

rpsA, pncA

rpsL, gidB, tlyA, rrs

gyrA, gyrB

gidB, rrs, tlyA, eis

IEMDR03

_

rpoB S450L (S531L E.coli)

_

_

_

_

_

IEMDR05

katG S315T

rpoB S450L (S531L E.coli)

embB D328Y

pncA C14G

rrs A514C

_

rrs A514C

IEMDR19

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli)

embB S297A

_

rpsL K43R

_

eis C-10T

IEMDR30

katG S315T

rpoB S450L (S531L E.coli)

embB M306V

_

rpsL K88R

_

_

IEMDR22

katG S315T

rpoB D435Y (D516Y E.coli)

_

_

_

_

_

IEMDR25

katG S315T

rpoB D435Y (D516Y E.coli)

_

_

_

_

_

IEMDR14

fabG1 C-15T, inhA S94A

rpoB S450L, D435Y (S531L, D516Y E.coli)

embB M306V

pncA D12E

rpsL K43R

_

_

IEMDR23

fabG1 C-15T, inhA S94A

rpoB S450L, D435Y (S531L, D516Y E.coli)

embB M306V

pncA D12E

rpsL K43R

_

_

IEMDR18

katG S315T

rpoB S450L (S531L E.coli)

embB M306I, embA C-12T

pncA Q10P

rpsL K43R

_

_

IEMDR12

katG S315T

rpoB S450L (S531L E.coli)

embB D354A

pncA A-12G

rrs C517T

gyrA S91P

rrs C517T

Page 20 of 23

IEMDR33

katG S315T

rpoB S450L (S531L E.coli)

embB M306V

_

rpsL K43R, rrs C517T

_

rrs C517T, eis G14A

IEMDR34

katG S315T, fabG1 C-15T

rpoB D435V, D435A (D516V, D516A E.coli)

_

pncA Y103H

rpsL K43R

_

eis C-10T

IEMDR42

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli)

embA C-12T, embB Y334H

pncA C14St op

rpsL K88R

_

_

IEXDR1/IEMD R02

katG S315T

rpoB H445Y (H526Y E.coli)

embB M306V

pncA G132 C

rpsL K43R, rrs A1401G

gyrA D94A

rrs A1401G

IEMDR04

katG S315T

rpoB H445Y (H526Y E.coli)

embA C-8T

pncA D12N

rpsL K43R

gyrB N499T

eis C-10T

IEMDR07

katG S315T

rpoB S450L (S531L E.coli)

embB M306V

pncA D136 N

rpsL K43R

_

eis C-10T

IEMDR10

katG S315T

rpoB S450L (S531L E.coli), rpoC F452S

embB M306V

_

rpsL K43R

_

eis C-10T

IEMDR15

katG S315T

rpoB S450L (S531L E.coli), rpoC F452S

embB M306V

_

rpsL K43R

_

eis C-10T

IEMDR20

katG S315T

rpoB S450L (S531L E.coli)

embB M306I

pncA M175 V

rpsL K43R

_

_

IEMDR32

katG S315T

rpoB S450L (S531L E.coli)

embB G406A

pncA I31S

rpsL K43R, rrs A1401G

gyrA A90V

rrs A1401G

IEMDR01

katG S315T, fabG1 T-8C

rpoB H445Y (H526Y E.coli)

embB M306V, D328Y

pncA C14St op

_

_

_

IEMDR01EAR LY

katG S315T, fabG1 T-8C

rpoB H445Y (H526Y E.coli)

embB M306V, M306I

pncA C14St op

_

_

_

Page 21 of 23

IEMDR41

katG S315T, fabG1 T-8C

rpoB H445Y (H526Y E.coli)

_

_

_

_

_

IEMDR06

katG S315T

rpoB I491F (I572F E.coli)

embB M306I

_

rpsL K43R

_

_

IEMDR13

katG S315T

rpoB S450L (S531L E.coli)

embB M306I

_

rpsL K43R

_

_

IEMDR08

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli), rpoC F452S

_

_

_

_

_

IEMDR27

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli)

embB Q497R

pncA H51Y

rrs A514C

_

rrs A514C

IEMDR27LAT ER

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli)

embB Q497R

pncA H51Y

rrs A514C

gyrA D94Y

rrs A514C

IEMDR35

katG S315T, fabG1 C-15T

rpoB S450L (S531L E.coli)

embB Q497R

pncA H51Y

rrs A514C

_

rrs A514C

IEMDR28

katG S315T, fabG1 C-15T

rpoB H445L (H526L E.coli)

_

_

rrs A514C

_

rrs A514C

IEMDR09

katG S315T

rpoB S450L (S531L E.coli)

embB M306I, D328Y

_

rpsL K43R

gyrA A90V

eis C-10T

IEMDR16

katG S315T

rpoB S450L (S531L E.coli), rpoC F452S, G332R

embB M306I

_

rpsL K43R

_

eis C-10T

IEMDR21

katG S315T

rpoB H445Y (H526Y E.coli)

embB M306V

_

rpsL K88M

_

_

IEMDR11

katG S315T

rpoB S450L (S531L E.coli)

embB M306I, Q497R

_

_

_

_

IEMDR31

katG S315T, ahpC G-48A

rpoB S450L (S531L E.coli)

embB M306V

_

_

_

_

IEMDR17

ahpC C-52T, katG Q295P

rpoB H445D (H526D E.coli)

_

_

_

_

_

IEMDR36

katG S315T

rpoB D435V (D516V E.coli)

_

_

_

_

_

IEMDR24

katG S315T, P241P

rpoB S450L (S531L E.coli)

_

_

_

_

_

IEMDR26

_

rpoB H445Y (H526Y E.coli)

_

_

_

_

_

IEMDR40

katG S315T

rpoB H445R (H526R E.coli)

_

_

_

_

_

Page 22 of 23

IEMDR29

katG S315T

rpoB S450L (S531L E.coli)

embB Y319S

pncA G108R

rpsL K43R, rrs A1401G

_

rrs A1401G

543 544 545 546

Twenty five genes were analysed using different resistance mutation catalogues as well as manual detection. Isolates are grouped together according to their MIRU-VNTR genotyping pattern (Figure 1.) INH isoniazid, RIF rifampicin, EMB ethambutol, PZA pyrazinamide, SM streptomycin, FQ fluoroquinolones, AG

aminoglycosides.

547 548

Page 23 of 23