Emerging resistant clones of Mycobacterium tuberculosis in a spatiotemporal context

Emerging resistant clones of Mycobacterium tuberculosis in a spatiotemporal context Abstract Objectives We assessed the genetic structure of the Mycobacterium tuberculosis population in Estonia with a special focus on major epidemic/endemic clones and drug resistance determinants. We investigated the hypothesis of the decisive impact of massive human influx on the locally circulating genotypes. Estonia received a mass immigration from Russia during 1945–90 followed by enhanced interaction with the EU since 1991. Methods The study sample included M. tuberculosis isolates from patients newly diagnosed with TB in 2014 in North Estonia (including the capital Tallinn). The isolates were subjected to first- and second-line drug susceptibility testing, detection of mutations in rpoB, katG, inhA, rrs, embB and gyrA and lineage/clone-specific genotyping. Results Of the M. tuberculosis isolates, 39.8% were assigned to the Beijing genotype; 56.8% of them were MDR. In contrast, all three major non-Beijing genotypes (LAM, Haarlem and Ural) were mainly drug susceptible. MDR was more prevalent among Beijing B0/W148-cluster isolates (81.8%) compared with other Beijing isolates (20.0%; P = 0.0007). The pre-XDR phenotype was found in eight isolates, of which six belonged to Beijing B0/W148. All rifampicin-resistant and ofloxacin-resistant and 97% of isoniazid-resistant isolates harboured resistance mutations in rpoB, gyrA and katG. The rpoB S531L, katG S315T and embB M306V mutations were the most prevalent. Conclusions The major pool of the Beijing isolates was brought to Estonia before 1990. However, an active circulation of the most hazardous MDR-associated Beijing B0/W148-cluster started only in the last 20 years and its significantly increased circulation presents the major threat to TB control in Estonia. The overwhelming prevalence of the rpoB531 and katG315 mutations in the MDR-associated Beijing isolates requires attention. Introduction Human immigration is an important factor in transmission of MDR-TB in Western European countries of the EU where such isolates are mainly imported from high-burden regions. In contrast, transmission of local isolates is a major concern for those EU members that used to be a part of the Former Soviet Union (FSU).1 Estonia is a prime example. According to the WHO estimation for 2015, the median (range) TB incidence (per 100000 persons) in Estonia was 18 (15–21), compared with 80 (69–92) in Russia and 41 (35–47) in Latvia. The incidence of MDR-TB (per 100000) in Estonia was 5.4 (3.8–7), which is similar to Latvia [5.0 (3.8–6.1)], much lower than in Russia [42 (34–49)], but much higher than in Poland [0.18 (0.13–0.24)] or Finland [0.31 (0.11–0.51)] (http://www.who.int/tb/country/data/profiles/en/). A high prevalence of the Mycobacterium tuberculosis isolates of the Beijing genotype (and consequently, a high MDR-TB rate) in Baltic countries was previously linked to massive immigration from Russia during 1945–90.2 Estonia restored its independence in 1991, following which the human influx from Russia and other post-Soviet countries was reduced, whereas human exchange with the EU increased. This situation may have led to a certain shift in the population structure of major human pathogens, including M. tuberculosis. Previous molecular M. tuberculosis studies in Estonia are separated by almost 20 years.3,4 Toit et al.4 concluded that recent TB transmission in Estonia is particularly associated with MDR- and XDR-TB and the Beijing lineage. However, 24-loci typing used in that study may not be discriminatory enough for Beijing types 100-32 and 94-32 to provide unambiguous proof of recent transmission, and some caution in interpretation is required. In this study, we assessed the genetic structure of the M. tuberculosis population in Estonia, with a special focus on the major epidemic/endemic clones and drug resistance determinants. The results were compared with the reanalysed earlier Estonian studies and were placed within the broader context of emerging epidemic clusters of M. tuberculosis in this part of northern Europe and northern Eurasia, i.e. Russia and other FSU countries. Methods Bacterial isolates The study sample included consecutively isolated M. tuberculosis isolates (n = 96) recovered from 2 January to 30 December 2014 from 96 newly diagnosed patients with TB who were admitted to the North Estonian Medical Centre (Tallinn, Estonia). The survey area of North Estonia includes the capital city of Tallinn and 8 of 15 counties of the country; it represents 70% of the total population and a major part of the TB burden in Estonia.4 According to the census 2011 data, the total population of Estonia was 1294455 and it was estimated to be 1314000 in 2014 (https://en.wikipedia.org/wiki/Demographics_of_Estonia). All samples used in this study were coded and lacked personal information about the patients, particularly names and addresses, to maintain their anonymity. Patient-related data were obtained in an anonymous way from the North Estonian Medical Centre and Estonian Tuberculosis Registry, National Institute for Health and Development; no individual patient information is disclosed in this article. M. tuberculosis culture, identification and drug susceptibility testing of first- and second-line drugs was performed as recommended.5,6 For all drugs, susceptibility testing was performed using Bactec MGIT 960 system (Becton Dickinson, Sparks, MD, USA) according to the manufacturer’s instructions. Drugs and their critical concentrations were as follows: 1.0 mg/L streptomycin; 0.1 mg/L isoniazid; 1.0 mg/L rifampicin; 5.0 mg/L ethambutol; 100 mg/L pyrazinamide; 2.0 mg/L ofloxacin; 2.5 mg/L prothionamide; 1.0 mg/L amikacin; 5.0 mg/L kanamycin; and 2.5 mg/L capreomycin. The bacteriology laboratory in the North Estonian Medical Centre is externally quality assured by INSTAND (Düsseldorf, Germany; first-line drug susceptibility testing) and the WHO TB Supranational Reference Laboratory for drug-resistant TB at the Public Health Agency of Sweden, Solna, Sweden (first- and second-line drugs susceptibility testing). Detection of drug resistance mutations Rifampicin and isoniazid mutations in genes rpoB, katG and inhA were detected using real-time PCR assay (AmpliTub-MLU-RV; Syntol, Russia). Resistance mutations to ethambutol and second-line drugs in embB, gyrA and rrs were detected using GenoType® MTBDRsl (Hain Lifescience GmbH, Germany). Genotyping DNA was extracted from bacterial cultures using a GenoLyse® kit (Hain Lifescience). After exclusion of isolates with failed PCR, 93 isolates were analysed. Beijing genotype was detected by PCR of the dnaA-dnaN::IS6110 insertion as described previously.7 All non-Beijing isolates were subjected to spoligotyping following the macroarray-based standard protocol;8 the profiles were compared with SITVIT_WEB (http://www.pasteur-guadeloupe.fr:8081/SITVIT_ONLINE/query). The M. tuberculosis LAM family was identified by PCR-RFLP analysis of the specific Rv0129c SNP.9 Genomic deletions RD115, RD174 and RD-Rio and the IS6110 insertion specific to the LAM-RUS branch were detected using a PCR-based approach as described previously.9–11 The Beijing B0/W148-cluster was identified based on the Rv2664–Rv2665::IS6110 insertion.2 The Beijing 94-32 cluster (∼A0 cluster) was detected based on the specific two-locus signature (five copies in MIRU26 and eight copies in QUB26) as reported in Mokrousov et al.12 for A0 cluster isolates. Variable number tandem repeat (VNTR) typing was performed as described by Supply et al.13 Statistics A χ2 test was used to detect any significant difference between the two groups. Yates-corrected χ2 and P values were calculated with 95% CI at http://www.medcalc.org/calc/odds_ratio.php online resource. Results Patients, isolates and resistance This population-based study included patients with newly diagnosed TB in northern Estonia enrolled within a 1 year period in 2014 (n = 96). Sixty-five were male and 31 were female. The mean age was 48 years (range = 24–91 years; SD = ±13.7). The first available isolate from each patient was included in the microbiological and molecular study. Three isolates were excluded owing to repeated PCR failure, thus the further analysis was done on the collection of 93 isolates (Table 1). Among the possible risk factors (HIV coinfection, prison history, alcohol abuse), the latter was most frequently found (42 of 93). Table 1. Comparison of genotypic and phenotypic resistance of the studied M. tuberculosis strains Drug; resistance genes targeted  Drug-susceptible isolates: without resistance mutation/total  Drug-resistant isolates: with resistance mutation/total  Number of isolates with particular drug resistance mutations  INH; katG, inhA  64/64  28/29  0 with inhA mutations; 28 katG315 AGC>ACC (Ser>Thr)  RIF; rpoB  70/70  23/23  22 rpoB531 TCG>TTG (Ser>Leu); 1 rpoB526 CAC>CTG (His>Leu)  EMB; embB  71/72  16/21  17 embB306 ATG>GTG (Met>Val)  OFX; gyrA  89/90  3/3  1 gyrA91 TCG>CCG (Ser/Pro); 2 gyrA94 GAC>GCC (Asp/Ala)a; 1 gyrA94 GAC>AAC (Asp/Asn)  Drug; resistance genes targeted  Drug-susceptible isolates: without resistance mutation/total  Drug-resistant isolates: with resistance mutation/total  Number of isolates with particular drug resistance mutations  INH; katG, inhA  64/64  28/29  0 with inhA mutations; 28 katG315 AGC>ACC (Ser>Thr)  RIF; rpoB  70/70  23/23  22 rpoB531 TCG>TTG (Ser>Leu); 1 rpoB526 CAC>CTG (His>Leu)  EMB; embB  71/72  16/21  17 embB306 ATG>GTG (Met>Val)  OFX; gyrA  89/90  3/3  1 gyrA91 TCG>CCG (Ser/Pro); 2 gyrA94 GAC>GCC (Asp/Ala)a; 1 gyrA94 GAC>AAC (Asp/Asn)  INH, isoniazid; RIF, rifampicin; EMB, ethambutol; OFX, ofloxacin. a One isolate was susceptible. By country of birth, most patients were born in Estonia (n = 73), nine in Russia, four in Ukraine, two in Belarus, two in Kyrgyzstan, one in Gambia and one in Lithuania. In other words, all but one represented the FSU. One patient from Gambia was infected with Mycobacterium africanum, which is known to be endemic in West Africa. Stratification of results by country of birth would not be informative because only the Gambia-born patient had lived in Estonia for only 1–2 years. Other foreign-born patients had lived for >5 years in Estonia. As they were newly diagnosed it is more likely that they were infected with a local Estonian strain rather than had a reactivation of the latent isolate brought from abroad. Phenotypically, 52 of 93 isolates were pan-susceptible while 21 were MDR. No XDR isolates were identified, but eight isolates were pre-XDR. Three of them were MDR and ofloxacin-resistant, and five were MDR and resistant to kanamycin, capreomycin and/or amikacin. Analysis of the major gene targets associated with drug resistance revealed a strong correlation between presence of mutation and resistant phenotype for the two key first-line drugs, rifampicin and isoniazid, and second-line ofloxacin (Table 1). In particular, all rifampicin- and ofloxacin-resistant isolates harboured a mutation in the targeted genes. One isoniazid-resistant isolate had no mutation in inhA and katG, and resistance was apparently due to the presence of other mutations not included in the molecular assay. The isoniazid resistance mechanism is known to be quite complex and multiple genes may be involved. In two cases, a phenotypically susceptible isolate harboured a drug resistance mutation: one was ofloxacin susceptible and one was ethambutol susceptible. The latter kind of discrepancy is not unusual and the controversial role of the embB306 mutations with regard to ethambutol resistance and MDR has been described previously.14,15 Along with high sensitivity of the molecular methods to predict drug resistance, we also note a high homogeneity of the detected mutations in rpoB and katG. All katG mutations were in codon 315 (AGC > ACC) and all but one rpoB mutation were in codon 531 (TCG > TTG). M. tuberculosis genotypes To outline the population structure at the M. tuberculosis family/subfamily level, the isolates were analysed by several genotyping methods. As a first step, they were subdivided into the Beijing genotype and non-Beijing groups, and the latter were subjected to spoligotyping. In addition, Beijing and LAM isolates were tested for major clonal clusters and phylogenetic sublineages known to be of epidemiological and/or clinical relevance. Spoligotyping profiles of the non-Beijing isolates were assigned to the particular families/clades based on comparison with the SITVIT_WEB database with correction for certain families based on expert knowledge of robust molecular signatures. The genotypes and clusters were compared for drug resistance pattern and patient-related features. Table 2 shows summarized information on genotypes and subtypes (the T family is not shown separately because it is a polyphyletic group). Table 2. Molecular characteristics of M. tuberculosis strains stratified by strain genotype, drug resistance and patient data   All, n = 93  Beijing, all, n = 37  Beijing B0/W148, n = 21  Beijing 94-32, n = 7  Non-Beijing, alla, n = 55  LAM, n = 12  Ural, n = 10  Haarlem, n = 9  Pan-susceptible  52  4  –  4  47  11  9  7  Drug resistant (not MDR)  20 (12 mono, 8 poly)  12  4 (poly)  –  8  1 (mono)  1 (poly)  2 (mono)  MDR  21  21  17 (6 pre-XDR)  3 (2 pre-XDR)  –  –  –  –  Estonia-born  73  28  16  4  45  11  9  8  FSU-bornb  18  8  4  3  10  1  1 (no data)  1  Estonia-born in 1980 and before  60  20  12  3  40  8  9  8  Estonia-born in 1981 and after  13  8  4  1  5  3  –  –  HIV coinfection  12  7  2  1  5  2  2  –  Alcoholic  42  19  12  2  23  5  5  5  Prison history  18  9  6  –  9  3  1  –    All, n = 93  Beijing, all, n = 37  Beijing B0/W148, n = 21  Beijing 94-32, n = 7  Non-Beijing, alla, n = 55  LAM, n = 12  Ural, n = 10  Haarlem, n = 9  Pan-susceptible  52  4  –  4  47  11  9  7  Drug resistant (not MDR)  20 (12 mono, 8 poly)  12  4 (poly)  –  8  1 (mono)  1 (poly)  2 (mono)  MDR  21  21  17 (6 pre-XDR)  3 (2 pre-XDR)  –  –  –  –  Estonia-born  73  28  16  4  45  11  9  8  FSU-bornb  18  8  4  3  10  1  1 (no data)  1  Estonia-born in 1980 and before  60  20  12  3  40  8  9  8  Estonia-born in 1981 and after  13  8  4  1  5  3  –  –  HIV coinfection  12  7  2  1  5  2  2  –  Alcoholic  42  19  12  2  23  5  5  5  Prison history  18  9  6  –  9  3  1  –  poly, polyresistant (resistant to more than one drug, but not MDR); mono, monoresistant. a Non-Beijing: without M. africanum. b FSU-born: without one patient born in Lithuania (infected with Beijing B0/W148 isolate). The 39.8% (37 of 93) of M. tuberculosis isolates were assigned to the Beijing genotype; the Beijing B0/W148 cluster was identified in 59.5% (22 of 37) of Beijing isolates. Russian/Central Asian Beijing type 94-32 were found in 7 of 37 Beijing isolates, and ancient Beijing isolates in 2 of 37 Beijing. The pre-XDR phenotype was found in eight isolates, of which six were B0 of W148 and two were the 94-32 type. MDR was found in 22.8% (21 of 92) of isolates and only in Beijing genotype isolates. MDR was more prevalent among Beijing B0/W148-cluster isolates (81.8%; 18 of 22) compared with other Beijing types (20.0%; 3 of 15; P = 0.0007). An increased prevalence of the B0/W148 isolates in high-risk population groups might indicate their increased transmission capacity. However, we did not find any marked bias in prevalence of B0/W148 in any risk group (Table 2) although indeed B0/W148 isolates compared with all other genotypes had higher rates among former prison inmates (6 of 22 versus 12 of 71) and alcoholics (12 of 22 versus 30 of 71) although non-significantly (P = 0.28 and P = 0.31, respectively). This study enrolled newly diagnosed patients, and a very high rate of MDR among Beijing B0/W148 isolates (81.8%) could in principle be due to the active transmission of a single clone. However, it should be noted that the B0/W148 cluster includes very closely related isolates that are virtually indistinguishable by 24-MIRU-VNTR typing (as we previously showed for these isolates circulating in the neighbouring Russian regions2,12). WGS would be required to adequately differentiate these isolates, but it was beyond the scope of this study. Among 55 non-Beijing isolates, 30 spoligotypes (SIT) of different genetic lineages were identified: T (16.3% of the whole collection), LAM (12%), Ural (11%), Haarlem (10%), X (2%) and unclassified (3%). Four isolates had new spoligotypes not found in SITVIT_WEB. One isolate (SIT326) was defined as M. africanum. The 14.5% (8 of 55) of non-Beijing isolates were resistant to streptomycin (n = 5), isoniazid (n = 1), rifampicin (n = 1) and streptomycin + isoniazid (n = 1); none was MDR. The number of non-Estonia-born patients was rather small (20 of 93). Furthermore, they all lived in Estonia for more than 5 years and a comparison of the FSU-born versus Estonia-born would not be justified. However, we considered it meaningful to compare within the Estonia-born group depending on age: (i) those born in 1980 and before, i.e. infected during active human exchange with FSU versus (ii) those born in 1981 and after, i.e. infected already in independent Estonia and under reduced exchange with/influx from FSU (Table 3). Some of such comparisons showed different, opposite trends, but owing to the small sample size of those born in Estonia after 1980, this difference was not significant. In particular, the prevalence was higher for non-Beijing genotypes in the ‘older’ (40 of 60) versus ‘younger’ (5 of 13) group. Although this difference was non-significant (P = 0.06), this may reflect the onset of active dissemination of the Beijing isolates in Estonia only in the last two decades. Table 3. Comparison within the Estonia-born group of patients   Pan-susceptible  MDR  HIV coinfection  Alcoholic  Prison history  Estonia-born in 1980 and before (n = 60)  37  10  4  29  10  Estonia-born in 1981 and after (n = 13)  5  3  5  3  4    Pan-susceptible  MDR  HIV coinfection  Alcoholic  Prison history  Estonia-born in 1980 and before (n = 60)  37  10  4  29  10  Estonia-born in 1981 and after (n = 13)  5  3  5  3  4  Interestingly, in the group of patients born before 1950, only one (born in Kyrgyzstan) was infected with a Beijing genotype isolate (B0/W148 type) while other patients were born in Estonia (seven), Russia (one) or Ukraine (one) and represented genotypes T (three), Haarlem (three), LAM (two) and Ural (one). However, Beijing B0/W148 is found in central Asia, Kyrgyzstan included, at 2%–3% at most (reviewed by Mokrousov2 and Skiba et al.16). Thus, it is more likely that this patient was infected with a B0/W148-cluster isolate already in Estonia; hypothetically, he/she may have been infected during a likely stay in Russia although this kind of information was absent in the epidemiological record. Discussion Drug resistance of M. tuberculosis genotypes in Estonia Drug resistance was unequally distributed across the families and clonal groups in the studied collection. All three major non-Beijing genotypes (LAM, Haarlem and Ural) were mainly susceptible, with mainly single monoresistant isolates (Table 2). In contrast, the Beijing genotype was dominated by MDR isolates (21 of 37). An even stronger prevalence of MDR was noted in the Beijing B0/W148 cluster (17 of 21) whereas no monoresistant or susceptible isolates were found. The situation with MDR prevalence is similar to 20 years ago when all Beijing B0/W148 isolates were drug resistant, 60% of other Beijing isolates were drug resistant (now it is 75%) and the non-Beijing genotypes were mainly susceptible (14% drug resistance rate in both 1994 and 2014). In the Russian spondylitis study the B0/W148 cluster included 29% of the Beijing isolates and 87% of them were MDR and none was susceptible.17 Altogether, even compared with other Beijing isolates in the FSU, the Beijing B0/W148 cluster is remarkable in its degree of multidrug resistance as shown by meta-analysis of studies in the FSU countries,2 and its increasing trend in Estonia is alarming. The high prevalence of certain drug resistance mutations known to be of low fitness cost should be noted (Table 1). In practical terms, this demonstrated their utility to serve as specific and sensitive molecular markers of isoniazid and rifampicin resistance in Estonia. This situation resembles that in the neighbouring Russian province of Pskov where the rpoB531 mutation was found in 19 of 21 rifampicin-resistant Beijing genotype isolates in a recent study.18 At the same time, the earlier Russian study in neighbouring St Petersburg in 1997–2002 revealed a somewhat lower prevalence of this mutation (77%).19 Overall, population structures in neighbouring Estonia and Pskov are quite similar (Figure 1), but in Pskov, rifampicin-resistant and MDR isolates were found in two major genotype families, Beijing and LAM (although in the latter, rpoB516 mutations were prevalent). On the other hand, the high rate of this mutation in rpoB531 in Beijing isolates is not a unique feature of this genotype. For example, a study in Bulgaria found a high rate of this mutation in the rifampicin-resistant isolates of different (non-Beijing) genotypes and it was hypothesized that this mutation rpoB S531L might correlate with some specific features of the national TB control programmes, e.g. the quality of the drugs used.20 Figure 1. View largeDownload slide Distribution of the major genotype families of M. tuberculosis strains in Estonia and neighbouring regions assessed in this and previous studies. Information on Estonia in the 1990s was based on re-estimation of data from Krüüner et al.3 and SITVIT_WEB. In particular, from Krüüner et al.,3 it was possible to assess only Beijing (29%), LAM (14%) and Ural (10%) genotypes. Other genotypes (T, Haarlem and other) were assessed from the SITVIT_WEB data. Circle size is not to scale. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Distribution of the major genotype families of M. tuberculosis strains in Estonia and neighbouring regions assessed in this and previous studies. Information on Estonia in the 1990s was based on re-estimation of data from Krüüner et al.3 and SITVIT_WEB. In particular, from Krüüner et al.,3 it was possible to assess only Beijing (29%), LAM (14%) and Ural (10%) genotypes. Other genotypes (T, Haarlem and other) were assessed from the SITVIT_WEB data. Circle size is not to scale. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. History and geography of M. tuberculosis genotypes in Estonia Reanalysis of the previous studies published 20 years ago in light of current knowledge, along with methodological approach used in the present study, permitted us to compare our data with previous results3 and the SITVIT_WEB database. In particular, the genotype families were assigned based on current knowledge and only partly using SITVIT_WEB rules, although some limitations were inevitable. For example, Krüüner et al.3 showed spoligotypes only for ‘representative’ isolates from different IS6110-RFLP clusters and it was not possible to reliably assess the T and Haarlem families. Analysis of the geographic distribution of the major genotype families found in this study, and in the neighbouring countries/regions in this part of Europe18,21–25 revealed a certain trend although a different study design requires caution in interpretation (Figure 1). In particular, high prevalence of the Beijing genotype in most settings should be noted, although these isolates in the foreign-born group in Finland may have different and more diverse origins compared with the Baltic countries and Russia. In addition, the LAM family appears to present a stable and somewhat lower rate in Estonia compared with Latvia and neighbouring Russian regions. Regarding dynamic changes within Estonia, we compared our data with the situation 20 years ago.3 This showed an increase of Beijing genotype from 29.2% [61 of 209 in 1994 to 39.7% (37 of 93) in 2014, but non-significantly (P = 0.07)]. However, particular Beijing subtypes showed different trajectories as detailed below. In the present study, the major Beijing subtype B0/W148 (∼Mlva15-9 type 100-32) previously designated as a ‘successful Russian clone’ was found in 22.6% of the entire collection, which is similar to the recent nationwide study (19.5%).4 In 1994 in Estonia, the B0/W148 prevalence rate was 7.6%. Thus there is a clear increase in its prevalence in Estonia [16 of 209 versus 21 of 93; P = 0.0005; OR = 0.284 (95% CI = 0.140–0.575)]. In the Russian regions bordering Estonia, the prevalence of B0/W148 in recent years was reported as 8% in Pskov and 13% in St Petersburg and 19% in the more distant Karelia, mainly in patients with pulmonary TB. In a spinal TB study in European Russia, B0/W148 was identified in 22%.17 It may be noted that the rate of B0/W148 was higher in Estonia than in Pskov and St Petersburg, but the absolute prevalence values were higher in Russia, owing to the higher incidence (see the WHO estimations). Similarly, the other important Beijing cluster 94-32 prevalent in both Russia and FSU Central Asia and named the Central Asian/Russian strain16,17,26,27 remains at a low prevalence in Estonia, at the rate of 7.5% in this study and in 3.3% in the study of Toit et al.4 An active spread of 94-32 in Russia could have only started after 1991 and thus this strain had less opportunity to be imported to Estonia. Finally, isolates of the ancient sublineage of the Beijing genotype remain sporadic in this part of Europe. Two ancient Beijing isolates (5.4%, of 37 Beijing strains) were found in this study. This is similar to 5% in north-western Russia in the early 2000s28 and 6.2% in a recent study in the European part of Russia.17 In spite of the prevalence of the notorious and hazardous Beijing isolates, the emerging clones in eastern Europe and the FSU expand far beyond the Beijing family and the non-Beijing isolates should not be underestimated. These are the LAM and Ural families and their intriguing subtypes. The Ural genotype isolates (11% in this study) were represented by two spoligotypes, SIT35 (n = 2) and SIT262 (n = 8). Compared with the situation 20 years ago (Koivula/SITVIT_WEB), the Ural group was present at the similarly low prevalence of 11%, but showed more diversity, while SIT35 and SIT2622 included two and three isolates, respectively. Thus, in the Ural group, we note both a decreasing diversity and increasing prevalence of SIT262 (8 of 93 versus 3 of 119) although at a non-significant level [P = 0.06; OR = 3.6392 (95% CI = 0.9377–14.1241)]. However, all Ural isolates, except for one polyresistant SIT262 isolate, were susceptible. Accordingly, SIT262 appears transmissible rather than an MDR-prone clone, at least in Estonia. Indirectly, this highlights the exceptional capacity of the Beijing B0/W148 strain circulating in the same country, under the same TB control programme and in the same human population, to rapidly develop drug resistance. In contrast, more recently emerged MDR-associated LAM types SIT252 and SIT26629,30 were absent in the studied collection, but evolutionary older and more widespread types SIT254 and SIT264 were present. As hypothesized regarding Beijing 94-32, it may be possible that SIT252 and SIT266 emerging in central Russia and Belarus, respectively, have started their dissemination in the last two decades and had less time/chance to be brought to Estonia. Conclusion Our data suggest that the major pool of the Beijing isolates was imported to Estonia before 1990 during the Soviet period of its history. However, an active circulation of the most hazardous MDR-associated Beijing B0/W148 cluster (‘successful Russian strain’2) started only in the last 20 years in independent Estonia. Absence or low prevalence of some recently emerging Russian subtypes (Beijing type 94-32, LAM SIT252) in the studied collection may be due to the significantly reduced human influx from Russia during the last 25 years. Both the direction and volume of human migration appear to be crucial factors in the global and regional spread of M. tuberculosis and should be taken into consideration by national TB control programmes. Whereas no dramatic shift for Beijing and non-Beijing genotypes occurred in the M. tuberculosis population in Estonia in the last 20 years, the changes at the subtype level are alarming. First, a significantly increased circulation of the MDR-associated Beijing B0/W148 cluster isolates and the increasing proportion of drug resistance among other Beijing isolates present the main threat to the TB control programme in Estonia. Similarly, growing circulation of the Ural family isolates of spoligotype SIT262, albeit susceptible, requires caution. The situation of overwhelming prevalence of the low fitness cost mutations rpoB531 and katG315 in the MDR-associated isolates of the Beijing genotype requires special attention, in particular with regard to early case identification and improved epidemiological surveillance. Funding This work was supported by the Russian Science Foundation (grant no. 14-14-00292). Transparency declarations None to declare. References 1 De Beer JL, Kodmon C, van der Werf MJ et al.  ; ECDC MDR-TB Molecular Surveillance Project Participants. Molecular surveillance of multi- and extensively drug-resistant tuberculosis transmission in the European Union from 2003 to 2011. Euro Surveill  2014; 19: pii: 20742. 2 Mokrousov I. Insights into the origin, emergence, and current spread of a successful Russian clone of Mycobacterium tuberculosis. Clin Microbiol Rev  2013; 26: 342– 60. Google Scholar CrossRef Search ADS PubMed  3 Krüüner A, Hoffner SE, Sillastu H et al.   Spread of drug-resistant pulmonary tuberculosis in Estonia. J Clin Microbiol  2001; 39: 3339– 45. Google Scholar CrossRef Search ADS PubMed  4 Toit K, Altraja A, Acosta CD et al.   A four-year nationwide molecular epidemiological study in Estonia: risk factors for tuberculosis transmission. Public Health Action  2014; 4 Suppl 2: S34– 40. Google Scholar CrossRef Search ADS PubMed  5 Rüsch-Gerdes S, Pfyffer GE, Casal M 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: 688– 92. Google Scholar CrossRef Search ADS PubMed  6 WHO. Policy Guidance on Drug-Susceptibility Testing (DST) of Second-Line Antituberculosis Drugs. 2008. http://www.who.int/tb/publications/2008/whohtmtb_2008_392/en/. 7 Mokrousov I, Vyazovaya A, Zhuravlev V et al.   Real-time PCR assay for rapid detection of epidemiologically and clinically significant Mycobacterium tuberculosis Beijing genotype. J Clin Microbiol  2014; 52: 1691– 3. Google Scholar CrossRef Search ADS PubMed  8 Kamerbeek J, Schouls L, Kolk A et al.   Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol  1997; 35: 907– 14. Google Scholar PubMed  9 Gibson AL, Huard RC, Gey van Pittius NC et al.   Application of sensitive and specific molecular methods to uncover global dissemination of the major RDRio sublineage of the Latin American-Mediterranean Mycobacterium tuberculosis spoligotype family. J Clin Microbiol  2008; 46: 1259– 67. Google Scholar CrossRef Search ADS PubMed  10 Dubiley S, Kirillov E, Ignatova A et al.   Molecular characteristics of the Mycobacterium tuberculosis LAM-RUS family prevalent in Central Russia. J Clin Microbiol  2007; 45: 4036– 8. Google Scholar CrossRef Search ADS PubMed  11 Mokrousov I, Vyazovaya A, Narvskaya O. Mycobacterium tuberculosis Latin-American Mediterranean family and its sublineages: in the light of evolutionary robust markers. J Bacteriol  2014; 196: 1833– 41. Google Scholar CrossRef Search ADS PubMed  12 Mokrousov I, Narvskaya O, Vyazovaya A et al.   Mycobacterium tuberculosis Beijing genotype in Russia: in search of informative variable-number tandem-repeat loci. J Clin Microbiol  2008; 46: 3576– 84. Google Scholar CrossRef Search ADS PubMed  13 Supply P, Allix C, Lesjean S et al.   Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis. J Clin Microbiol  2006; 44: 4498– 510. Google Scholar CrossRef Search ADS PubMed  14 Mokrousov I, Otten T, Vyshnevskiy B et al.   Detection of embB306 mutations in ethambutol-susceptible Mycobacterium tuberculosis clinical isolates from northwestern Russia: implications for genotypic resistance testing. J Clin Microbiol  2002; 40: 3810– 3. Google Scholar CrossRef Search ADS PubMed  15 Hazbón MH, Bobadilla del Valle M, Guerrero MI et al.   Role of embB codon 306 mutations in Mycobacterium tuberculosis revisited: a novel association with broad drug resistance and IS6110 clustering rather than ethambutol resistance. Antimicrob Agents Chemother  2005; 49: 3794– 802. Google Scholar CrossRef Search ADS PubMed  16 Skiba Y, Mokrousov I, Ismagulova G et al.   Molecular snapshot of Mycobacterium tuberculosis population in Kazakhstan: a country-wide study. Tuberculosis  2015; 95: 538– 46. Google Scholar CrossRef Search ADS PubMed  17 Vyazovaya A, Mokrousov I, Solovieva N et al.   Tuberculous spondylitis in Russia and prominent role of multidrug-resistant clone Mycobacterium tuberculosis Beijing B0/W148. Antimicrob Agents Chemother  2015; 59: 2349– 57. Google Scholar CrossRef Search ADS PubMed  18 Mokrousov I, Vyazovaya A, Otten T et al.   Mycobacterium tuberculosis population in northwestern Russia: an update from Russian-EU/Latvian border region. PLoS One  2012; 7: e41318. Google Scholar CrossRef Search ADS PubMed  19 Mokrousov I, Otten T, Vyazovaya A et al.   PCR-based methodology for detecting multidrug-resistant strains of Mycobacterium tuberculosis Beijing family circulating in Russia. Eur J Clin Microbiol Infect Dis  2003; 22: 342– 8. Google Scholar CrossRef Search ADS PubMed  20 Valcheva V, Mokrousov I, Narvskaya O et al.   Molecular snapshot of drug-resistant and drug-susceptible Mycobacterium tuberculosis strains circulating in Bulgaria. Infect Genet Evol  2008; 8: 657– 63. Google Scholar CrossRef Search ADS PubMed  21 Tracevska T, Jansone I, Baumanis V et al.   Prevalence of Beijing genotype in Latvian multidrug-resistant Mycobacterium tuberculosis isolates. Int J Tuberc Lung Dis  2003; 7: 1097– 103. Google Scholar PubMed  22 Narvskaya O, Mokrousov I, Otten T et al.   Molecular markers: application for studies of Mycobacterium tuberculosis population in Russia. In: Read MM, ed. Trends in DNA Fingerprinting Research . New York: Nova Science Publishers, 2005; 111– 125. 23 Mäkinen J, Marjamäki M, Haanperä-Heikkinen M et al.   Extremely high prevalence of multidrug resistant tuberculosis in Murmansk, Russia: a population-based study. Eur J Clin Microbiol Infect Dis  2011; 30: 1119– 26. Google Scholar CrossRef Search ADS PubMed  24 Mokrousov I, Vyazovaya A, Solovieva N et al.   Trends in molecular epidemiology of drug-resistant tuberculosis in Republic of Karelia, Russian Federation. BMC Microbiol  2015; 15: 279. Google Scholar CrossRef Search ADS PubMed  25 Smit PW, Haanperä M, Rantala P et al.   Molecular epidemiology of tuberculosis in Finland, 2008-2011. PLoS One  2013; 8: e85027. Google Scholar CrossRef Search ADS PubMed  26 Merker M, Blin C, Mona S et al.   Evolutionary history and global spread of the Mycobacterium tuberculosis Beijing lineage. Nat Genet  2015; 47: 242– 9. Google Scholar CrossRef Search ADS PubMed  27 Luo T, Comas I, Luo D et al.   Southern East Asian origin and coexpansion of Mycobacterium tuberculosis Beijing family with Han Chinese. Proc Natl Acad Sci USA  2015; 112: 8136– 41. Google Scholar CrossRef Search ADS PubMed  28 Mokrousov I, Narvskaya O, Otten T et al.   Phylogenetic reconstruction within Mycobacterium tuberculosis Beijing genotype in northwest Russia. Res Microbiol  2002; 153: 629– 37. Google Scholar CrossRef Search ADS PubMed  29 Zalutskaya A, Wijkander M, Jureen P et al.   Multidrug-resistant Myobacterium tuberculosis caused by the Beijing genotype and a specific T1 genotype clone (SIT no. 266) is widely transmitted in Minsk. Int J Mycobacteriol  2013; 2: 194– 8. Google Scholar CrossRef Search ADS PubMed  30 Vyazovaya A, Mokrousov I, Pole I et al.   Multidrug-resistant clone M. tuberculosis LAM SIT252 emerging in some regions of European Russia and Eastern Europe. In: Abstracts of the Thirty-seventh Annual Congress of the European Society of Mycobacteriology, Catania, Italy, 2016. Abstract P35, p. 81–82. Agency KONSENS Ltd, Werne, Germany. © The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Emerging resistant clones of Mycobacterium tuberculosis in a spatiotemporal context

Loading next page...
 
/lp/ou_press/emerging-resistant-clones-of-mycobacterium-tuberculosis-in-a-jSLc8YQcGs
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dkx372
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives We assessed the genetic structure of the Mycobacterium tuberculosis population in Estonia with a special focus on major epidemic/endemic clones and drug resistance determinants. We investigated the hypothesis of the decisive impact of massive human influx on the locally circulating genotypes. Estonia received a mass immigration from Russia during 1945–90 followed by enhanced interaction with the EU since 1991. Methods The study sample included M. tuberculosis isolates from patients newly diagnosed with TB in 2014 in North Estonia (including the capital Tallinn). The isolates were subjected to first- and second-line drug susceptibility testing, detection of mutations in rpoB, katG, inhA, rrs, embB and gyrA and lineage/clone-specific genotyping. Results Of the M. tuberculosis isolates, 39.8% were assigned to the Beijing genotype; 56.8% of them were MDR. In contrast, all three major non-Beijing genotypes (LAM, Haarlem and Ural) were mainly drug susceptible. MDR was more prevalent among Beijing B0/W148-cluster isolates (81.8%) compared with other Beijing isolates (20.0%; P = 0.0007). The pre-XDR phenotype was found in eight isolates, of which six belonged to Beijing B0/W148. All rifampicin-resistant and ofloxacin-resistant and 97% of isoniazid-resistant isolates harboured resistance mutations in rpoB, gyrA and katG. The rpoB S531L, katG S315T and embB M306V mutations were the most prevalent. Conclusions The major pool of the Beijing isolates was brought to Estonia before 1990. However, an active circulation of the most hazardous MDR-associated Beijing B0/W148-cluster started only in the last 20 years and its significantly increased circulation presents the major threat to TB control in Estonia. The overwhelming prevalence of the rpoB531 and katG315 mutations in the MDR-associated Beijing isolates requires attention. Introduction Human immigration is an important factor in transmission of MDR-TB in Western European countries of the EU where such isolates are mainly imported from high-burden regions. In contrast, transmission of local isolates is a major concern for those EU members that used to be a part of the Former Soviet Union (FSU).1 Estonia is a prime example. According to the WHO estimation for 2015, the median (range) TB incidence (per 100000 persons) in Estonia was 18 (15–21), compared with 80 (69–92) in Russia and 41 (35–47) in Latvia. The incidence of MDR-TB (per 100000) in Estonia was 5.4 (3.8–7), which is similar to Latvia [5.0 (3.8–6.1)], much lower than in Russia [42 (34–49)], but much higher than in Poland [0.18 (0.13–0.24)] or Finland [0.31 (0.11–0.51)] (http://www.who.int/tb/country/data/profiles/en/). A high prevalence of the Mycobacterium tuberculosis isolates of the Beijing genotype (and consequently, a high MDR-TB rate) in Baltic countries was previously linked to massive immigration from Russia during 1945–90.2 Estonia restored its independence in 1991, following which the human influx from Russia and other post-Soviet countries was reduced, whereas human exchange with the EU increased. This situation may have led to a certain shift in the population structure of major human pathogens, including M. tuberculosis. Previous molecular M. tuberculosis studies in Estonia are separated by almost 20 years.3,4 Toit et al.4 concluded that recent TB transmission in Estonia is particularly associated with MDR- and XDR-TB and the Beijing lineage. However, 24-loci typing used in that study may not be discriminatory enough for Beijing types 100-32 and 94-32 to provide unambiguous proof of recent transmission, and some caution in interpretation is required. In this study, we assessed the genetic structure of the M. tuberculosis population in Estonia, with a special focus on the major epidemic/endemic clones and drug resistance determinants. The results were compared with the reanalysed earlier Estonian studies and were placed within the broader context of emerging epidemic clusters of M. tuberculosis in this part of northern Europe and northern Eurasia, i.e. Russia and other FSU countries. Methods Bacterial isolates The study sample included consecutively isolated M. tuberculosis isolates (n = 96) recovered from 2 January to 30 December 2014 from 96 newly diagnosed patients with TB who were admitted to the North Estonian Medical Centre (Tallinn, Estonia). The survey area of North Estonia includes the capital city of Tallinn and 8 of 15 counties of the country; it represents 70% of the total population and a major part of the TB burden in Estonia.4 According to the census 2011 data, the total population of Estonia was 1294455 and it was estimated to be 1314000 in 2014 (https://en.wikipedia.org/wiki/Demographics_of_Estonia). All samples used in this study were coded and lacked personal information about the patients, particularly names and addresses, to maintain their anonymity. Patient-related data were obtained in an anonymous way from the North Estonian Medical Centre and Estonian Tuberculosis Registry, National Institute for Health and Development; no individual patient information is disclosed in this article. M. tuberculosis culture, identification and drug susceptibility testing of first- and second-line drugs was performed as recommended.5,6 For all drugs, susceptibility testing was performed using Bactec MGIT 960 system (Becton Dickinson, Sparks, MD, USA) according to the manufacturer’s instructions. Drugs and their critical concentrations were as follows: 1.0 mg/L streptomycin; 0.1 mg/L isoniazid; 1.0 mg/L rifampicin; 5.0 mg/L ethambutol; 100 mg/L pyrazinamide; 2.0 mg/L ofloxacin; 2.5 mg/L prothionamide; 1.0 mg/L amikacin; 5.0 mg/L kanamycin; and 2.5 mg/L capreomycin. The bacteriology laboratory in the North Estonian Medical Centre is externally quality assured by INSTAND (Düsseldorf, Germany; first-line drug susceptibility testing) and the WHO TB Supranational Reference Laboratory for drug-resistant TB at the Public Health Agency of Sweden, Solna, Sweden (first- and second-line drugs susceptibility testing). Detection of drug resistance mutations Rifampicin and isoniazid mutations in genes rpoB, katG and inhA were detected using real-time PCR assay (AmpliTub-MLU-RV; Syntol, Russia). Resistance mutations to ethambutol and second-line drugs in embB, gyrA and rrs were detected using GenoType® MTBDRsl (Hain Lifescience GmbH, Germany). Genotyping DNA was extracted from bacterial cultures using a GenoLyse® kit (Hain Lifescience). After exclusion of isolates with failed PCR, 93 isolates were analysed. Beijing genotype was detected by PCR of the dnaA-dnaN::IS6110 insertion as described previously.7 All non-Beijing isolates were subjected to spoligotyping following the macroarray-based standard protocol;8 the profiles were compared with SITVIT_WEB (http://www.pasteur-guadeloupe.fr:8081/SITVIT_ONLINE/query). The M. tuberculosis LAM family was identified by PCR-RFLP analysis of the specific Rv0129c SNP.9 Genomic deletions RD115, RD174 and RD-Rio and the IS6110 insertion specific to the LAM-RUS branch were detected using a PCR-based approach as described previously.9–11 The Beijing B0/W148-cluster was identified based on the Rv2664–Rv2665::IS6110 insertion.2 The Beijing 94-32 cluster (∼A0 cluster) was detected based on the specific two-locus signature (five copies in MIRU26 and eight copies in QUB26) as reported in Mokrousov et al.12 for A0 cluster isolates. Variable number tandem repeat (VNTR) typing was performed as described by Supply et al.13 Statistics A χ2 test was used to detect any significant difference between the two groups. Yates-corrected χ2 and P values were calculated with 95% CI at http://www.medcalc.org/calc/odds_ratio.php online resource. Results Patients, isolates and resistance This population-based study included patients with newly diagnosed TB in northern Estonia enrolled within a 1 year period in 2014 (n = 96). Sixty-five were male and 31 were female. The mean age was 48 years (range = 24–91 years; SD = ±13.7). The first available isolate from each patient was included in the microbiological and molecular study. Three isolates were excluded owing to repeated PCR failure, thus the further analysis was done on the collection of 93 isolates (Table 1). Among the possible risk factors (HIV coinfection, prison history, alcohol abuse), the latter was most frequently found (42 of 93). Table 1. Comparison of genotypic and phenotypic resistance of the studied M. tuberculosis strains Drug; resistance genes targeted  Drug-susceptible isolates: without resistance mutation/total  Drug-resistant isolates: with resistance mutation/total  Number of isolates with particular drug resistance mutations  INH; katG, inhA  64/64  28/29  0 with inhA mutations; 28 katG315 AGC>ACC (Ser>Thr)  RIF; rpoB  70/70  23/23  22 rpoB531 TCG>TTG (Ser>Leu); 1 rpoB526 CAC>CTG (His>Leu)  EMB; embB  71/72  16/21  17 embB306 ATG>GTG (Met>Val)  OFX; gyrA  89/90  3/3  1 gyrA91 TCG>CCG (Ser/Pro); 2 gyrA94 GAC>GCC (Asp/Ala)a; 1 gyrA94 GAC>AAC (Asp/Asn)  Drug; resistance genes targeted  Drug-susceptible isolates: without resistance mutation/total  Drug-resistant isolates: with resistance mutation/total  Number of isolates with particular drug resistance mutations  INH; katG, inhA  64/64  28/29  0 with inhA mutations; 28 katG315 AGC>ACC (Ser>Thr)  RIF; rpoB  70/70  23/23  22 rpoB531 TCG>TTG (Ser>Leu); 1 rpoB526 CAC>CTG (His>Leu)  EMB; embB  71/72  16/21  17 embB306 ATG>GTG (Met>Val)  OFX; gyrA  89/90  3/3  1 gyrA91 TCG>CCG (Ser/Pro); 2 gyrA94 GAC>GCC (Asp/Ala)a; 1 gyrA94 GAC>AAC (Asp/Asn)  INH, isoniazid; RIF, rifampicin; EMB, ethambutol; OFX, ofloxacin. a One isolate was susceptible. By country of birth, most patients were born in Estonia (n = 73), nine in Russia, four in Ukraine, two in Belarus, two in Kyrgyzstan, one in Gambia and one in Lithuania. In other words, all but one represented the FSU. One patient from Gambia was infected with Mycobacterium africanum, which is known to be endemic in West Africa. Stratification of results by country of birth would not be informative because only the Gambia-born patient had lived in Estonia for only 1–2 years. Other foreign-born patients had lived for >5 years in Estonia. As they were newly diagnosed it is more likely that they were infected with a local Estonian strain rather than had a reactivation of the latent isolate brought from abroad. Phenotypically, 52 of 93 isolates were pan-susceptible while 21 were MDR. No XDR isolates were identified, but eight isolates were pre-XDR. Three of them were MDR and ofloxacin-resistant, and five were MDR and resistant to kanamycin, capreomycin and/or amikacin. Analysis of the major gene targets associated with drug resistance revealed a strong correlation between presence of mutation and resistant phenotype for the two key first-line drugs, rifampicin and isoniazid, and second-line ofloxacin (Table 1). In particular, all rifampicin- and ofloxacin-resistant isolates harboured a mutation in the targeted genes. One isoniazid-resistant isolate had no mutation in inhA and katG, and resistance was apparently due to the presence of other mutations not included in the molecular assay. The isoniazid resistance mechanism is known to be quite complex and multiple genes may be involved. In two cases, a phenotypically susceptible isolate harboured a drug resistance mutation: one was ofloxacin susceptible and one was ethambutol susceptible. The latter kind of discrepancy is not unusual and the controversial role of the embB306 mutations with regard to ethambutol resistance and MDR has been described previously.14,15 Along with high sensitivity of the molecular methods to predict drug resistance, we also note a high homogeneity of the detected mutations in rpoB and katG. All katG mutations were in codon 315 (AGC > ACC) and all but one rpoB mutation were in codon 531 (TCG > TTG). M. tuberculosis genotypes To outline the population structure at the M. tuberculosis family/subfamily level, the isolates were analysed by several genotyping methods. As a first step, they were subdivided into the Beijing genotype and non-Beijing groups, and the latter were subjected to spoligotyping. In addition, Beijing and LAM isolates were tested for major clonal clusters and phylogenetic sublineages known to be of epidemiological and/or clinical relevance. Spoligotyping profiles of the non-Beijing isolates were assigned to the particular families/clades based on comparison with the SITVIT_WEB database with correction for certain families based on expert knowledge of robust molecular signatures. The genotypes and clusters were compared for drug resistance pattern and patient-related features. Table 2 shows summarized information on genotypes and subtypes (the T family is not shown separately because it is a polyphyletic group). Table 2. Molecular characteristics of M. tuberculosis strains stratified by strain genotype, drug resistance and patient data   All, n = 93  Beijing, all, n = 37  Beijing B0/W148, n = 21  Beijing 94-32, n = 7  Non-Beijing, alla, n = 55  LAM, n = 12  Ural, n = 10  Haarlem, n = 9  Pan-susceptible  52  4  –  4  47  11  9  7  Drug resistant (not MDR)  20 (12 mono, 8 poly)  12  4 (poly)  –  8  1 (mono)  1 (poly)  2 (mono)  MDR  21  21  17 (6 pre-XDR)  3 (2 pre-XDR)  –  –  –  –  Estonia-born  73  28  16  4  45  11  9  8  FSU-bornb  18  8  4  3  10  1  1 (no data)  1  Estonia-born in 1980 and before  60  20  12  3  40  8  9  8  Estonia-born in 1981 and after  13  8  4  1  5  3  –  –  HIV coinfection  12  7  2  1  5  2  2  –  Alcoholic  42  19  12  2  23  5  5  5  Prison history  18  9  6  –  9  3  1  –    All, n = 93  Beijing, all, n = 37  Beijing B0/W148, n = 21  Beijing 94-32, n = 7  Non-Beijing, alla, n = 55  LAM, n = 12  Ural, n = 10  Haarlem, n = 9  Pan-susceptible  52  4  –  4  47  11  9  7  Drug resistant (not MDR)  20 (12 mono, 8 poly)  12  4 (poly)  –  8  1 (mono)  1 (poly)  2 (mono)  MDR  21  21  17 (6 pre-XDR)  3 (2 pre-XDR)  –  –  –  –  Estonia-born  73  28  16  4  45  11  9  8  FSU-bornb  18  8  4  3  10  1  1 (no data)  1  Estonia-born in 1980 and before  60  20  12  3  40  8  9  8  Estonia-born in 1981 and after  13  8  4  1  5  3  –  –  HIV coinfection  12  7  2  1  5  2  2  –  Alcoholic  42  19  12  2  23  5  5  5  Prison history  18  9  6  –  9  3  1  –  poly, polyresistant (resistant to more than one drug, but not MDR); mono, monoresistant. a Non-Beijing: without M. africanum. b FSU-born: without one patient born in Lithuania (infected with Beijing B0/W148 isolate). The 39.8% (37 of 93) of M. tuberculosis isolates were assigned to the Beijing genotype; the Beijing B0/W148 cluster was identified in 59.5% (22 of 37) of Beijing isolates. Russian/Central Asian Beijing type 94-32 were found in 7 of 37 Beijing isolates, and ancient Beijing isolates in 2 of 37 Beijing. The pre-XDR phenotype was found in eight isolates, of which six were B0 of W148 and two were the 94-32 type. MDR was found in 22.8% (21 of 92) of isolates and only in Beijing genotype isolates. MDR was more prevalent among Beijing B0/W148-cluster isolates (81.8%; 18 of 22) compared with other Beijing types (20.0%; 3 of 15; P = 0.0007). An increased prevalence of the B0/W148 isolates in high-risk population groups might indicate their increased transmission capacity. However, we did not find any marked bias in prevalence of B0/W148 in any risk group (Table 2) although indeed B0/W148 isolates compared with all other genotypes had higher rates among former prison inmates (6 of 22 versus 12 of 71) and alcoholics (12 of 22 versus 30 of 71) although non-significantly (P = 0.28 and P = 0.31, respectively). This study enrolled newly diagnosed patients, and a very high rate of MDR among Beijing B0/W148 isolates (81.8%) could in principle be due to the active transmission of a single clone. However, it should be noted that the B0/W148 cluster includes very closely related isolates that are virtually indistinguishable by 24-MIRU-VNTR typing (as we previously showed for these isolates circulating in the neighbouring Russian regions2,12). WGS would be required to adequately differentiate these isolates, but it was beyond the scope of this study. Among 55 non-Beijing isolates, 30 spoligotypes (SIT) of different genetic lineages were identified: T (16.3% of the whole collection), LAM (12%), Ural (11%), Haarlem (10%), X (2%) and unclassified (3%). Four isolates had new spoligotypes not found in SITVIT_WEB. One isolate (SIT326) was defined as M. africanum. The 14.5% (8 of 55) of non-Beijing isolates were resistant to streptomycin (n = 5), isoniazid (n = 1), rifampicin (n = 1) and streptomycin + isoniazid (n = 1); none was MDR. The number of non-Estonia-born patients was rather small (20 of 93). Furthermore, they all lived in Estonia for more than 5 years and a comparison of the FSU-born versus Estonia-born would not be justified. However, we considered it meaningful to compare within the Estonia-born group depending on age: (i) those born in 1980 and before, i.e. infected during active human exchange with FSU versus (ii) those born in 1981 and after, i.e. infected already in independent Estonia and under reduced exchange with/influx from FSU (Table 3). Some of such comparisons showed different, opposite trends, but owing to the small sample size of those born in Estonia after 1980, this difference was not significant. In particular, the prevalence was higher for non-Beijing genotypes in the ‘older’ (40 of 60) versus ‘younger’ (5 of 13) group. Although this difference was non-significant (P = 0.06), this may reflect the onset of active dissemination of the Beijing isolates in Estonia only in the last two decades. Table 3. Comparison within the Estonia-born group of patients   Pan-susceptible  MDR  HIV coinfection  Alcoholic  Prison history  Estonia-born in 1980 and before (n = 60)  37  10  4  29  10  Estonia-born in 1981 and after (n = 13)  5  3  5  3  4    Pan-susceptible  MDR  HIV coinfection  Alcoholic  Prison history  Estonia-born in 1980 and before (n = 60)  37  10  4  29  10  Estonia-born in 1981 and after (n = 13)  5  3  5  3  4  Interestingly, in the group of patients born before 1950, only one (born in Kyrgyzstan) was infected with a Beijing genotype isolate (B0/W148 type) while other patients were born in Estonia (seven), Russia (one) or Ukraine (one) and represented genotypes T (three), Haarlem (three), LAM (two) and Ural (one). However, Beijing B0/W148 is found in central Asia, Kyrgyzstan included, at 2%–3% at most (reviewed by Mokrousov2 and Skiba et al.16). Thus, it is more likely that this patient was infected with a B0/W148-cluster isolate already in Estonia; hypothetically, he/she may have been infected during a likely stay in Russia although this kind of information was absent in the epidemiological record. Discussion Drug resistance of M. tuberculosis genotypes in Estonia Drug resistance was unequally distributed across the families and clonal groups in the studied collection. All three major non-Beijing genotypes (LAM, Haarlem and Ural) were mainly susceptible, with mainly single monoresistant isolates (Table 2). In contrast, the Beijing genotype was dominated by MDR isolates (21 of 37). An even stronger prevalence of MDR was noted in the Beijing B0/W148 cluster (17 of 21) whereas no monoresistant or susceptible isolates were found. The situation with MDR prevalence is similar to 20 years ago when all Beijing B0/W148 isolates were drug resistant, 60% of other Beijing isolates were drug resistant (now it is 75%) and the non-Beijing genotypes were mainly susceptible (14% drug resistance rate in both 1994 and 2014). In the Russian spondylitis study the B0/W148 cluster included 29% of the Beijing isolates and 87% of them were MDR and none was susceptible.17 Altogether, even compared with other Beijing isolates in the FSU, the Beijing B0/W148 cluster is remarkable in its degree of multidrug resistance as shown by meta-analysis of studies in the FSU countries,2 and its increasing trend in Estonia is alarming. The high prevalence of certain drug resistance mutations known to be of low fitness cost should be noted (Table 1). In practical terms, this demonstrated their utility to serve as specific and sensitive molecular markers of isoniazid and rifampicin resistance in Estonia. This situation resembles that in the neighbouring Russian province of Pskov where the rpoB531 mutation was found in 19 of 21 rifampicin-resistant Beijing genotype isolates in a recent study.18 At the same time, the earlier Russian study in neighbouring St Petersburg in 1997–2002 revealed a somewhat lower prevalence of this mutation (77%).19 Overall, population structures in neighbouring Estonia and Pskov are quite similar (Figure 1), but in Pskov, rifampicin-resistant and MDR isolates were found in two major genotype families, Beijing and LAM (although in the latter, rpoB516 mutations were prevalent). On the other hand, the high rate of this mutation in rpoB531 in Beijing isolates is not a unique feature of this genotype. For example, a study in Bulgaria found a high rate of this mutation in the rifampicin-resistant isolates of different (non-Beijing) genotypes and it was hypothesized that this mutation rpoB S531L might correlate with some specific features of the national TB control programmes, e.g. the quality of the drugs used.20 Figure 1. View largeDownload slide Distribution of the major genotype families of M. tuberculosis strains in Estonia and neighbouring regions assessed in this and previous studies. Information on Estonia in the 1990s was based on re-estimation of data from Krüüner et al.3 and SITVIT_WEB. In particular, from Krüüner et al.,3 it was possible to assess only Beijing (29%), LAM (14%) and Ural (10%) genotypes. Other genotypes (T, Haarlem and other) were assessed from the SITVIT_WEB data. Circle size is not to scale. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Distribution of the major genotype families of M. tuberculosis strains in Estonia and neighbouring regions assessed in this and previous studies. Information on Estonia in the 1990s was based on re-estimation of data from Krüüner et al.3 and SITVIT_WEB. In particular, from Krüüner et al.,3 it was possible to assess only Beijing (29%), LAM (14%) and Ural (10%) genotypes. Other genotypes (T, Haarlem and other) were assessed from the SITVIT_WEB data. Circle size is not to scale. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. History and geography of M. tuberculosis genotypes in Estonia Reanalysis of the previous studies published 20 years ago in light of current knowledge, along with methodological approach used in the present study, permitted us to compare our data with previous results3 and the SITVIT_WEB database. In particular, the genotype families were assigned based on current knowledge and only partly using SITVIT_WEB rules, although some limitations were inevitable. For example, Krüüner et al.3 showed spoligotypes only for ‘representative’ isolates from different IS6110-RFLP clusters and it was not possible to reliably assess the T and Haarlem families. Analysis of the geographic distribution of the major genotype families found in this study, and in the neighbouring countries/regions in this part of Europe18,21–25 revealed a certain trend although a different study design requires caution in interpretation (Figure 1). In particular, high prevalence of the Beijing genotype in most settings should be noted, although these isolates in the foreign-born group in Finland may have different and more diverse origins compared with the Baltic countries and Russia. In addition, the LAM family appears to present a stable and somewhat lower rate in Estonia compared with Latvia and neighbouring Russian regions. Regarding dynamic changes within Estonia, we compared our data with the situation 20 years ago.3 This showed an increase of Beijing genotype from 29.2% [61 of 209 in 1994 to 39.7% (37 of 93) in 2014, but non-significantly (P = 0.07)]. However, particular Beijing subtypes showed different trajectories as detailed below. In the present study, the major Beijing subtype B0/W148 (∼Mlva15-9 type 100-32) previously designated as a ‘successful Russian clone’ was found in 22.6% of the entire collection, which is similar to the recent nationwide study (19.5%).4 In 1994 in Estonia, the B0/W148 prevalence rate was 7.6%. Thus there is a clear increase in its prevalence in Estonia [16 of 209 versus 21 of 93; P = 0.0005; OR = 0.284 (95% CI = 0.140–0.575)]. In the Russian regions bordering Estonia, the prevalence of B0/W148 in recent years was reported as 8% in Pskov and 13% in St Petersburg and 19% in the more distant Karelia, mainly in patients with pulmonary TB. In a spinal TB study in European Russia, B0/W148 was identified in 22%.17 It may be noted that the rate of B0/W148 was higher in Estonia than in Pskov and St Petersburg, but the absolute prevalence values were higher in Russia, owing to the higher incidence (see the WHO estimations). Similarly, the other important Beijing cluster 94-32 prevalent in both Russia and FSU Central Asia and named the Central Asian/Russian strain16,17,26,27 remains at a low prevalence in Estonia, at the rate of 7.5% in this study and in 3.3% in the study of Toit et al.4 An active spread of 94-32 in Russia could have only started after 1991 and thus this strain had less opportunity to be imported to Estonia. Finally, isolates of the ancient sublineage of the Beijing genotype remain sporadic in this part of Europe. Two ancient Beijing isolates (5.4%, of 37 Beijing strains) were found in this study. This is similar to 5% in north-western Russia in the early 2000s28 and 6.2% in a recent study in the European part of Russia.17 In spite of the prevalence of the notorious and hazardous Beijing isolates, the emerging clones in eastern Europe and the FSU expand far beyond the Beijing family and the non-Beijing isolates should not be underestimated. These are the LAM and Ural families and their intriguing subtypes. The Ural genotype isolates (11% in this study) were represented by two spoligotypes, SIT35 (n = 2) and SIT262 (n = 8). Compared with the situation 20 years ago (Koivula/SITVIT_WEB), the Ural group was present at the similarly low prevalence of 11%, but showed more diversity, while SIT35 and SIT2622 included two and three isolates, respectively. Thus, in the Ural group, we note both a decreasing diversity and increasing prevalence of SIT262 (8 of 93 versus 3 of 119) although at a non-significant level [P = 0.06; OR = 3.6392 (95% CI = 0.9377–14.1241)]. However, all Ural isolates, except for one polyresistant SIT262 isolate, were susceptible. Accordingly, SIT262 appears transmissible rather than an MDR-prone clone, at least in Estonia. Indirectly, this highlights the exceptional capacity of the Beijing B0/W148 strain circulating in the same country, under the same TB control programme and in the same human population, to rapidly develop drug resistance. In contrast, more recently emerged MDR-associated LAM types SIT252 and SIT26629,30 were absent in the studied collection, but evolutionary older and more widespread types SIT254 and SIT264 were present. As hypothesized regarding Beijing 94-32, it may be possible that SIT252 and SIT266 emerging in central Russia and Belarus, respectively, have started their dissemination in the last two decades and had less time/chance to be brought to Estonia. Conclusion Our data suggest that the major pool of the Beijing isolates was imported to Estonia before 1990 during the Soviet period of its history. However, an active circulation of the most hazardous MDR-associated Beijing B0/W148 cluster (‘successful Russian strain’2) started only in the last 20 years in independent Estonia. Absence or low prevalence of some recently emerging Russian subtypes (Beijing type 94-32, LAM SIT252) in the studied collection may be due to the significantly reduced human influx from Russia during the last 25 years. Both the direction and volume of human migration appear to be crucial factors in the global and regional spread of M. tuberculosis and should be taken into consideration by national TB control programmes. Whereas no dramatic shift for Beijing and non-Beijing genotypes occurred in the M. tuberculosis population in Estonia in the last 20 years, the changes at the subtype level are alarming. First, a significantly increased circulation of the MDR-associated Beijing B0/W148 cluster isolates and the increasing proportion of drug resistance among other Beijing isolates present the main threat to the TB control programme in Estonia. Similarly, growing circulation of the Ural family isolates of spoligotype SIT262, albeit susceptible, requires caution. The situation of overwhelming prevalence of the low fitness cost mutations rpoB531 and katG315 in the MDR-associated isolates of the Beijing genotype requires special attention, in particular with regard to early case identification and improved epidemiological surveillance. Funding This work was supported by the Russian Science Foundation (grant no. 14-14-00292). Transparency declarations None to declare. References 1 De Beer JL, Kodmon C, van der Werf MJ et al.  ; ECDC MDR-TB Molecular Surveillance Project Participants. Molecular surveillance of multi- and extensively drug-resistant tuberculosis transmission in the European Union from 2003 to 2011. Euro Surveill  2014; 19: pii: 20742. 2 Mokrousov I. Insights into the origin, emergence, and current spread of a successful Russian clone of Mycobacterium tuberculosis. Clin Microbiol Rev  2013; 26: 342– 60. Google Scholar CrossRef Search ADS PubMed  3 Krüüner A, Hoffner SE, Sillastu H et al.   Spread of drug-resistant pulmonary tuberculosis in Estonia. J Clin Microbiol  2001; 39: 3339– 45. Google Scholar CrossRef Search ADS PubMed  4 Toit K, Altraja A, Acosta CD et al.   A four-year nationwide molecular epidemiological study in Estonia: risk factors for tuberculosis transmission. Public Health Action  2014; 4 Suppl 2: S34– 40. Google Scholar CrossRef Search ADS PubMed  5 Rüsch-Gerdes S, Pfyffer GE, Casal M 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: 688– 92. Google Scholar CrossRef Search ADS PubMed  6 WHO. Policy Guidance on Drug-Susceptibility Testing (DST) of Second-Line Antituberculosis Drugs. 2008. http://www.who.int/tb/publications/2008/whohtmtb_2008_392/en/. 7 Mokrousov I, Vyazovaya A, Zhuravlev V et al.   Real-time PCR assay for rapid detection of epidemiologically and clinically significant Mycobacterium tuberculosis Beijing genotype. J Clin Microbiol  2014; 52: 1691– 3. Google Scholar CrossRef Search ADS PubMed  8 Kamerbeek J, Schouls L, Kolk A et al.   Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol  1997; 35: 907– 14. Google Scholar PubMed  9 Gibson AL, Huard RC, Gey van Pittius NC et al.   Application of sensitive and specific molecular methods to uncover global dissemination of the major RDRio sublineage of the Latin American-Mediterranean Mycobacterium tuberculosis spoligotype family. J Clin Microbiol  2008; 46: 1259– 67. Google Scholar CrossRef Search ADS PubMed  10 Dubiley S, Kirillov E, Ignatova A et al.   Molecular characteristics of the Mycobacterium tuberculosis LAM-RUS family prevalent in Central Russia. J Clin Microbiol  2007; 45: 4036– 8. Google Scholar CrossRef Search ADS PubMed  11 Mokrousov I, Vyazovaya A, Narvskaya O. Mycobacterium tuberculosis Latin-American Mediterranean family and its sublineages: in the light of evolutionary robust markers. J Bacteriol  2014; 196: 1833– 41. Google Scholar CrossRef Search ADS PubMed  12 Mokrousov I, Narvskaya O, Vyazovaya A et al.   Mycobacterium tuberculosis Beijing genotype in Russia: in search of informative variable-number tandem-repeat loci. J Clin Microbiol  2008; 46: 3576– 84. Google Scholar CrossRef Search ADS PubMed  13 Supply P, Allix C, Lesjean S et al.   Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis. J Clin Microbiol  2006; 44: 4498– 510. Google Scholar CrossRef Search ADS PubMed  14 Mokrousov I, Otten T, Vyshnevskiy B et al.   Detection of embB306 mutations in ethambutol-susceptible Mycobacterium tuberculosis clinical isolates from northwestern Russia: implications for genotypic resistance testing. J Clin Microbiol  2002; 40: 3810– 3. Google Scholar CrossRef Search ADS PubMed  15 Hazbón MH, Bobadilla del Valle M, Guerrero MI et al.   Role of embB codon 306 mutations in Mycobacterium tuberculosis revisited: a novel association with broad drug resistance and IS6110 clustering rather than ethambutol resistance. Antimicrob Agents Chemother  2005; 49: 3794– 802. Google Scholar CrossRef Search ADS PubMed  16 Skiba Y, Mokrousov I, Ismagulova G et al.   Molecular snapshot of Mycobacterium tuberculosis population in Kazakhstan: a country-wide study. Tuberculosis  2015; 95: 538– 46. Google Scholar CrossRef Search ADS PubMed  17 Vyazovaya A, Mokrousov I, Solovieva N et al.   Tuberculous spondylitis in Russia and prominent role of multidrug-resistant clone Mycobacterium tuberculosis Beijing B0/W148. Antimicrob Agents Chemother  2015; 59: 2349– 57. Google Scholar CrossRef Search ADS PubMed  18 Mokrousov I, Vyazovaya A, Otten T et al.   Mycobacterium tuberculosis population in northwestern Russia: an update from Russian-EU/Latvian border region. PLoS One  2012; 7: e41318. Google Scholar CrossRef Search ADS PubMed  19 Mokrousov I, Otten T, Vyazovaya A et al.   PCR-based methodology for detecting multidrug-resistant strains of Mycobacterium tuberculosis Beijing family circulating in Russia. Eur J Clin Microbiol Infect Dis  2003; 22: 342– 8. Google Scholar CrossRef Search ADS PubMed  20 Valcheva V, Mokrousov I, Narvskaya O et al.   Molecular snapshot of drug-resistant and drug-susceptible Mycobacterium tuberculosis strains circulating in Bulgaria. Infect Genet Evol  2008; 8: 657– 63. Google Scholar CrossRef Search ADS PubMed  21 Tracevska T, Jansone I, Baumanis V et al.   Prevalence of Beijing genotype in Latvian multidrug-resistant Mycobacterium tuberculosis isolates. Int J Tuberc Lung Dis  2003; 7: 1097– 103. Google Scholar PubMed  22 Narvskaya O, Mokrousov I, Otten T et al.   Molecular markers: application for studies of Mycobacterium tuberculosis population in Russia. In: Read MM, ed. Trends in DNA Fingerprinting Research . New York: Nova Science Publishers, 2005; 111– 125. 23 Mäkinen J, Marjamäki M, Haanperä-Heikkinen M et al.   Extremely high prevalence of multidrug resistant tuberculosis in Murmansk, Russia: a population-based study. Eur J Clin Microbiol Infect Dis  2011; 30: 1119– 26. Google Scholar CrossRef Search ADS PubMed  24 Mokrousov I, Vyazovaya A, Solovieva N et al.   Trends in molecular epidemiology of drug-resistant tuberculosis in Republic of Karelia, Russian Federation. BMC Microbiol  2015; 15: 279. Google Scholar CrossRef Search ADS PubMed  25 Smit PW, Haanperä M, Rantala P et al.   Molecular epidemiology of tuberculosis in Finland, 2008-2011. PLoS One  2013; 8: e85027. Google Scholar CrossRef Search ADS PubMed  26 Merker M, Blin C, Mona S et al.   Evolutionary history and global spread of the Mycobacterium tuberculosis Beijing lineage. Nat Genet  2015; 47: 242– 9. Google Scholar CrossRef Search ADS PubMed  27 Luo T, Comas I, Luo D et al.   Southern East Asian origin and coexpansion of Mycobacterium tuberculosis Beijing family with Han Chinese. Proc Natl Acad Sci USA  2015; 112: 8136– 41. Google Scholar CrossRef Search ADS PubMed  28 Mokrousov I, Narvskaya O, Otten T et al.   Phylogenetic reconstruction within Mycobacterium tuberculosis Beijing genotype in northwest Russia. Res Microbiol  2002; 153: 629– 37. Google Scholar CrossRef Search ADS PubMed  29 Zalutskaya A, Wijkander M, Jureen P et al.   Multidrug-resistant Myobacterium tuberculosis caused by the Beijing genotype and a specific T1 genotype clone (SIT no. 266) is widely transmitted in Minsk. Int J Mycobacteriol  2013; 2: 194– 8. Google Scholar CrossRef Search ADS PubMed  30 Vyazovaya A, Mokrousov I, Pole I et al.   Multidrug-resistant clone M. tuberculosis LAM SIT252 emerging in some regions of European Russia and Eastern Europe. In: Abstracts of the Thirty-seventh Annual Congress of the European Society of Mycobacteriology, Catania, Italy, 2016. Abstract P35, p. 81–82. Agency KONSENS Ltd, Werne, Germany. © The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from Google Scholar, PubMed
Create lists to organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off