Jiayao Guan1, Chunmei Bao2, Peng Wang3, Ying Jing3, Lingling Wang3, Xinyue Li3, Xiaofei Mu3, Boan Li2, Dongsheng Zhou3, Xuejun Guo1, Zhe Yin3. 1. Key Laboratory of Jilin Province for Zoonosis Prevention and Control, Changchun Veterinary Research Institute, Chinese Academy of Agricultural Sciences, Changchun, Jilin, 130122, People's Republic of China. 2. Fifth Medical Center of Chinese PLA General Hospital, Beijing, 100039, People's Republic of China. 3. State Key Laboratory of Pathogen and Biosecurity, Beijing Institute of Microbiology and Epidemiology, Beijing, 100071, People's Republic of China.
Abstract
Purpose: The aim of this study was to gain a deeper genomics and bioinformatics understanding of diversification of accessory genetic elements (AGEs) in Providencia. Methods: Herein, the complete genome sequences of five Providencia isolates from China were determined, and seven AGEs were identified from the chromosomes. Detailed genetic dissection and sequence comparison were applied to these seven AGEs, together with additional 10 chromosomal ones from GenBank (nine of them came from Providencia). Results: These 17 AGEs were divided into four groups: Tn6512 and its six derivatives, Tn6872 and its two derivatives, Tn6875 and its one derivative, and Tn7 and its four derivatives. These AGEs display high-level diversification in modular structures that had complex mosaic natures, and particularly different multidrug resistance (MDR) regions were presented in these AGEs. At least 52 drug resistance genes, involved in resistance to 15 different categories of antimicrobials and heavy metal, were found in 15 of these 17 AGEs. Conclusion: Integration of these AGEs into the Providencia chromosomes would contribute to the accumulation and distribution of drug resistance genes and enhance the ability of Providencia isolates to survive under drug selection pressure.
Purpose: The aim of this study was to gain a deeper genomics and bioinformatics understanding of diversification of accessory genetic elements (AGEs) in Providencia. Methods: Herein, the complete genome sequences of five Providencia isolates from China were determined, and seven AGEs were identified from the chromosomes. Detailed genetic dissection and sequence comparison were applied to these seven AGEs, together with additional 10 chromosomal ones from GenBank (nine of them came from Providencia). Results: These 17 AGEs were divided into four groups: Tn6512 and its six derivatives, Tn6872 and its two derivatives, Tn6875 and its one derivative, and Tn7 and its four derivatives. These AGEs display high-level diversification in modular structures that had complex mosaic natures, and particularly different multidrug resistance (MDR) regions were presented in these AGEs. At least 52 drug resistance genes, involved in resistance to 15 different categories of antimicrobials and heavy metal, were found in 15 of these 17 AGEs. Conclusion: Integration of these AGEs into the Providencia chromosomes would contribute to the accumulation and distribution of drug resistance genes and enhance the ability of Providencia isolates to survive under drug selection pressure.
Providencia species, mostly frequently identified as Providencia rettgeri and Providencia stuartii, are opportunistic pathogens causing urinary tract infections, diarrhea, and bacteremia in immunocompromised patients.1
Providencia is intrinsically resistant to penicillins and the first-generation cephalosporins due to inducible expression of AmpC β-lactamases,2 aminoglycosides by the reason of inducible expression of AAC(2’)-Ia,3 tetracyclines because of constitutive expression of a multidrug efflux pump AcrAB,4 and polymyxins resulting from the presence of a cell envelop that inhibits colistin to combine with the susceptible lipid target sites or the lipid A modification to reduce binding.5 In recent years, the wide use of aminoglycosides, β-lactams, and fluoroquinolones for antimicrobial therapy aggravates the acquisition and dissemination of diverse antimicrobial resistance genes in Providencia.6 Acquired antimicrobial resistance genes in Providencia are commonly captured and horizontally transferred by accessory genetic elements (AGEs), such as integrative and conjugative elements (ICEs), integrative and mobilizable elements (IMEs), and unit transposons.ICEs have the ability to transfer from one cell to another cell through conjugation and are autonomous in conjugation by mainly utilizing a type IV secretion system (T4SS) and a coupling protein.7 Up to now, three distinct families of ICEs, namely Tn6512 (R391),8 ICEEc2,9 and ICEPm110 have been reported in Providencia. Tn6512 is initially found in Providencia rettgeri in 1972,8 and since then a wealth of Tn6512-related ICEs have been identified in not only Providencia,8,11 but also Enterobacteriaceae,12 Vibrionaceae,13
Shewanella,14
Actinobacillus,15
Alteromonas,16
Pseudoalteromonas.17 Tn6512-related ICEs carry at least five hotspots (located within the intragenic or intergenic sites of their backbone regions) for integration of foreign resistance genes,18 which causes these ICEs being mosaic and diversified with respect to their modular structures.IMEs do not encode the T4SS machinery and the coupling protein and thus is nonautonomous in conjugation.19 The conjugal transfer of an IME from a donor cell into a recipient cell relies on the T4SS gene sets of a helper conjugative element such as an IncA/C conjugative plasmid.20 So far, three distinct families of IMEs, namely Tn6523 (SGI1),21 Tn6588,22 and Tn659122 have been reported in Providencia.ICEs and IMEs can be transferred in an intercellular manner, while the Tn7-family unit transposons use a “cut-and-paste” transposition mechanism to transfer intracellularly, which is typically promoted by the core transposition determents TnsA (endonuclease), TnsB (transposase), TnsC (transposition regulator), and TnsD plus TnsE (target-site selection proteins).23 Tn7-family transposons have widely been found as the vectors of diverse antimicrobial resistance genes in Morganellaceae including Providencia rettgeri,22 Enterobacteriaceae,24 Pseudomonadaceae,25 Burkholderiaceae,26
Acinetobacter,27
Shewanella,24
Hahella,24
Pelobacter,24
Idiomarina,24
Acidithiobacillus,24
Neisseria (eg N. brasiliensis with an accession number CP046027), Nitrosomonas (eg N. stercoris with an accession number AP019755), and Bacillus.24This study presented the complete sequences of seven chromosome-borne AGEs in five sequenced Providencia isolates from China. Detailed genetic dissection and comparison were applied to these seven AGEs, together with additional 10 chromosome-borne ones from GenBank (nine of them came from Providencia). These 17 AGEs could be classified into ICEs, IMEs, and Tn7 derivatives. Data presented here provided a deeper genomics and bioinformatics understanding of diversification of AGEs in Providencia.
Materials and Methods
Providencia rettgeri PROV275, PROV002, and PROV087 () causing nosocomial infections were recovered from three different Chinese public hospitals in 2014, 2013, and 2016, respectively. Providencia alcalifaciens PROV023 and PROV013 () were collected from two different China livestock farms in 2017. Whole-genome sequencing of these five Providencia isolates were conducted with a sheared DNA library with average size of 15 kb (ranged from 10 kb to 20 kb) on a PacBio RSII sequencer (Pacific Biosciences, CA, USA) and further sequence data mining was performed as described previously.28 The sequencing data was conducted using NanoPack29 and FastQC () (). Conjugal transfer of Tn6862 from the wild-type PROV023 isolate into the rifampin-resistant recipient strain Escherichia coli EC600 was performed as described previously.30 Bacterial antimicrobial susceptibility was tested by BioMérieux VITEK 2, and interpreted as per the 2020 Clinical and Laboratory Standards Institute (CLSI) guidelines.31 The antimicrobial drug susceptibility profiles of the five Providencia isolates, EC600, and the transconjugant was shown in . All the wild-type and transconjugant strains were subjected to PCR amplification followed by amplicon sequencing, for determining the sequences of bacterial 16S rDNA, the presence of key markers such as aphA1, strA, strB, int, and cpl, and also the location/boundary of AGEs such as Tn6860, Tn6861, Tn6862, Tn6873, and Tn6876 (data not shown). The complete chromosome sequences of the PROV275, PROV002, PROV023, PROV013, and PROV087 isolates were submitted to GenBank under accession numbers CP059298, CP059345, CP059348, CP059346, and CP059347, respectively. The GenBank accession numbers of all the six plasmids of these five isolates were listed in .
Results
Four Groups of 17 Chromosome-Borne AGEs
In this study, the complete genome sequences from Providencia rettgeri PROV275, PROV002, and PROV087 and Providencia alcalifaciens PROV023 and PROV013 were determined, and then a total of seven chromosome-borne AGEs were identified: i) Tn6860, Tn6861, and a 102.1-kb Tn7-related element T7REPROV087 from strains PROV275, PROV002, and PROV087, respectively; ii) Tn6862 and a 35.9-kb T7REPROV023 from strain PROV023; and iii) Tn6873 and Tn6876 from strain PROV013. Then, a detailed sequence comparison was applied to a collection of four groups of 17 chromosome-borne AGEs (16 of them from Providencia): i) seven related ICEs Tn6512,8 Tn6860, Tn6861, Tn6862, Tn6863,18 Tn6864,11 and Tn6865; ii) three related IMEs Tn6872, Tn6873, and Tn6874; iii) two related IMEs Tn6875 and Tn6876; and iv) Tn7 and its four derivatives T7REPROV087, T7REPROV023, a 58.0-kb T7REMF1, and a 40.7-kb T7REPr-15-2-50 (Table 1). At least 52 drug resistance genes, involved in resistance to 15 different categories of antimicrobials and heavy metal, were identified in 15 (expect for Tn6872 and Tn6875) of these 17 AGEs (Figure 1 and ). All of these T7RE, T1696RE (see below), and T21RE (see below) elements could not be recognized as intact transposons due to truncation of relevant core transposition modules.
Table 1
Major Features of the 17 AGEs Characterized in This Work
Group
Accessory Genetic Element
Accession Number
Chromosomal Nucleotide Position
Length (bp)
Host Bacterium
Reference
Tn6512-related ICEs
Tn6512
AY090559
Not applicable
88,549
Providencia rettgeri 107
8
Tn6860
CP059298
575696.691964
116,269
Providencia rettgeri PROV275
This study
Tn6861
CP059345
727272.869956
142,685
Providencia rettgeri PROV002
This study
Tn6862
CP059348
704853.810761
105,909
Providencia alcalifaciens PROV023
This study
Tn6863
GQ463139
Not applicable
96,586
Providencia alcalifaciens Ban1
18
Tn6864
MT219827
Not applicable
107,207
Providencia rettgeri RF14-2
11
Tn6865
CP031123
751328.820050
68,723
Providencia huaxiensis WCHPr000369
Not applicable
Tn6872-related IMEs
Tn6872
LR134189
025117.3045606
20,490
Providencia rustigianii NCTC6933
Not applicable
Tn6873
CP059346
2973270.3043055
69,786
Providencia alcalifaciens PROV013
This study
Tn6874
AP022371
3358722.3420122
61,401
Providencia rettgeri BML2496
Not applicable
Tn6875-related IMEs
Tn6875
CP031508
1032159.1038038
5880
Providencia stuartii FDAARGOS_87
Not applicable
Tn6876
CP059346
2583916.2605820
21,905
Providencia alcalifaciens PROV013
This study
Tn7-related elements
Tn7
KX117211
Not applicable
14,067
Escherichia coli 3.5-R3
47
102.1-kb T7REPROV087
CP059347
33162.135304
102,143
Providencia rettgeri PROV087
This study
35.9-kb T7REPROV023
CP059348
47906.83831
35,926
Providencia alcalifaciens PROV023
This study
58.0-kb T7REMF1
CP048621
1077700.1135740
58,041
Providencia stuartii MF1
Not applicable
40.7-kb T7REPr-15-2-50
CP039844
31494.72272
40,779
Providencia rettgeri Pr-15-2-50
Not applicable
Figure 1
Heatmap of prevalence of drug resistance genes.
Major Features of the 17 AGEs Characterized in This WorkHeatmap of prevalence of drug resistance genes.
Six Tn6512-Related ICEs Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865
All these seven ICEs including Tn6512 were integrated into the same 17-bp target site located within the 5’ end of the chromosomal gene prfC (peptide chain release factor 3).32 These seven ICEs () had similar conserved backbones,33 containing int (integrase), xis (excisionase), rlx (relaxase), oriT (origin of conjugative replication), cpl (coupling protein), a TivF-type T4SS gene set, and attL/R (attachment sites at the left/right end). A toxin-antitoxin system was encoded by each of the five ICEs (hipBA in Tn6512, and abiEi–ii in Tn6860, Tn6861, Tn6863, and Tn6864) except for Tn6862 and Tn6865. Type I restriction-modification system hsdMSR was carried by five of them except for Tn6512 and Tn6865.The major modular differences were recognized at least 10 regions/sites (Figure 2 and ) across the whole genomes of these seven ICEs. Firstly, a total of 23 events of acquisition of exogenous DNA region occurred at nine regions/sites within these seven ICEs. Secondly, a total of 12 events of backbone region deletion (10 of them were resulted from acquisition of exogenous DNA region) occurred at five regions/sites within these seven ICEs. Finally, totally 21 events of substitution of backbone region occurred at four regions/sites within six ICEs.
Figure 2
Massive gene acquisition and loss across whole-genomes of seven related ICEs Tn6512, Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865.
Massive gene acquisition and loss across whole-genomes of seven related ICEs Tn6512, Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865.Each of these seven ICEs carried two to five accessory modules (), including multidrug resistance (MDR) regions, composite transposons, IME, integron, insertion sequence (IS) elements, and so called “inserted regions”. Firstly, the 33.5-kb, 62.0-kb, 29.8-kb, 14.2-kb, 18.3-kb, and 33.0-kb MDR regions were integrated at the same site within Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865, respectively. Secondly, an IS15DI (a minor variant of IS26)-composite transposon Tn6578 (containing aphA1), an IME Tn6912 (containing blaHMS-1), and an ISPpu12-composite transposon Tn6896 were inserted into the different sites within Tn6512, Tn6860, and Tn6864, respectively. Tn6896 (Figure 3) was bracketed by 8-bp direct repeats (DRs; target site duplication signals for transposition), and carried tetA(C)-carrying ΔTn6309, and a concise class 1 integron In525 with the gene cassette array (GCA) aadA2–ereA–dfrA32. Thirdly, a class 9 integron In9-1 (containing dfrA1) was inserted within Tn6863. Finally, multiple IS elements and inserted regions were found at the different sites within these ICEs.
Figure 3
Organization of Tn6896 from Tn6864, and comparison to related regions.
Organization of Tn6896 from Tn6864, and comparison to related regions.These six MDR regions (Figure 4) shared only a 0.5-kb Tn3-family Tn remnant together with an ISShfr9 element, and they carried dramatically different profiles of resistance loci. The 33.5-kb MDR region carried three resistance loci: a truncated ISCR2–floR unit, a 12.6-kb region (containing blaCTX-M-3-carrying ΔTn6502a) from IncM plasmid R69,34 and ISEc29–mph(E)–IS15DI unit. The 62.0-kb MDR region was composed of five resistance loci: a 23.4-kb Tn1696-related element T1696REPROV002 (see below), tetA(C)-carrying Tn6309, a concise class 1 integron In1793 (GCA: gacG2–aadA6–gacG2), a truncated IS26–mph(A)–IS6100 unit, and mer region. The 29.8-kb, 14.2-kb, and 18.3-kb MDR regions shared a truncated ISCR2–sul2 unit (containing strAB-carrying ΔTn5393c), but each of them integrated one or two additional resistance loci: i) an interrupted ISCR2–floR unit and an IS1006-composite transposon Tn6976 (which was bracketed by 5-bp DRs and contained two resistance loci: aphA1-carrying ΔTn4352 and IS26–aacC4–aph(4)-Ia–ISEc59 unit) in 29.8-kb MDR region; ii) ISCR2–floR unit in 14.2-kb MDR region; and iii) ISCR2–erm42 unit and a truncated ISCR2–tet(X6) unit in 18.3-kb MDR region. The 33.0-kb MDR region had the sole resistance locus In1571, whose 5’-conserved segment (5’-CS: intI–attI1) was interrupted by IS26. In1571 was a complex class 1 integron, which contained two resistance loci: GCA (also called VR1: variable region 1) fosC2–aacA4’-8–ereA1e–blaOXA-21–dfrA1r, and VR2 composed of ISCR1–qnrVC1 unit and a truncated ISCR1–blaPER-4 unit.
Figure 4
Comparison of MDR regions from Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865.
Comparison of MDR regions from Tn6860, Tn6861, Tn6862, Tn6863, Tn6864, and Tn6865.
Three Related IMEs Tn6872, Tn6873, and Tn6874
All these three IMEs were integrated 102-bp upstream the chromosomal gene mutS (DNA mismatch repair protein). The prototype IME Tn6872 was initially found in Providencia rustigianii NCTC6933 (accession number LR134189). These three IMEs (Figure 5) shared the core IME backbone markers int, oriT and attL/R. Tn6872 harbored its three unique backbone regions orf201–to–orf207, orf627–to–orf1068, and orf1338–to–uvrD; correspondingly, Tn6873/Tn6874 carried their unique backbone regions hsdMSR (type I restriction-modification system)–mrr–orf411, orf597–to–orf1107, and orf1311–to–hnhc, instead of the above three Tn6872-unique regions, respectively. No accessory modules were found in Tn6872, but a 47.0-kb MDR region and a 38.5-kb MDR region were integrated at the same site within Tn6873 and Tn6874, respectively.
Figure 5
Comparison of three related IMEs Tn6872, Tn6873, and Tn6874.
Comparison of three related IMEs Tn6872, Tn6873, and Tn6874.The 47.0-kb MDR region (Figure 6) contained six resistance loci: a concise class 1 integron In27 (GCA: dfrA12–gcuF–aadA2), ΔTn4352 containing aphA1, IS26–aph(4)-Ia–aacC4–ISEc59 unit, ISCR2–sul2 unit, a truncated ISCR2–floR unit, and a Tn21-related transposon Tn6974 (see below). The 38.5-kb MDR region carried an IME Tn6977, and a Tn1696-related transposon Tn6910 (see below).
Figure 6
Organization of MDR regions from Tn6873 and Tn6874, and comparison to related regions.
Organization of MDR regions from Tn6873 and Tn6874, and comparison to related regions.
Two Related IMEs Tn6875 and Tn6876
Tn6875 and Tn6876 were integrated 494-bp downstream the chromosomal gene smpB (SsrA-binding protein). The prototype IME Tn6875 was initially found in Providencia stuartii FDAARGOS_87 (accession number CP031508). Tn6875 and Tn6876 (Figure 7) shared the core IME backbone markers int and attL/R, but they displayed two major modular variations: i) Tn6875 contained its unique backbone region rimI–to–orf567, while Tn6876 harbored its unique backbone region containing orf288; and ii) Tn6875 carried no accessory modules, but Tn6876 acquired a 20.0-kb MDR region composed of tetA(B)-containing Tn10 and 10.4-kb T1696REPROV013 (see below).
Figure 7
Comparison of two related IMEs Tn6875 and Tn6876.
Comparison of two related IMEs Tn6875 and Tn6876.
Four Tn7 Derivatives T7REPROV087,T7REPROV023, T7REMF1, and T7REPr-15-2-50
Tn7 and its three derivatives T7REPROV087, T7REPROV023, and T7REPr-15-2-50 were inserted 25-bp downstream of the chromosomal gene glmS (glutamine-fructose-6-phosphate aminotransferase), while the remaining one T7REMF1 was inserted within the chromosomal gene orf1389 (carbohydrate porin). T7REPROV087, T7REPROV023, T7REMF1, and T7REPr-15-2-50 (Figure 8) differed from Tn7 by acquisition of 95.9-kb, 29.7-kb, 51.8-kb, and 33.6-kb MDR regions, respectively, instead of In2-4 (GCA: dfrA1–sat2–aadA1). The above acquisition events resulted in the same truncation of tnsABCDE in these four Tn7 derivatives. Moreover, tnsB of T7REPr-15-2-50 was interrupted by insertion of ISPrst3.
Figure 8
Comparison of Tn7 and its four derivatives.
Comparison of Tn7 and its four derivatives.These four MDR regions (Figure 9) shared only a 1.7-kb Tn1722 remnant together with an IS26 element, and they displayed considerable modular diversification: i) a concise class 2 integron In2-16 with GCA lnu(F)1b–dfrA1–aadA1a in the MDR regions from T7REPROV087, T7REPROV023, and T7REMF1, while a complex In2-16 carrying VR1/GCA as above and additionally two copies of VR2 (ISCR1–bleMBL–blaNDM-1 unit) in the MDR region from T7REPr-15-2-50; ii) a truncated ISCR2–floR unit, ISCR2–sul2 unit, IS26–aacC4–aph(4)-Ia–ISEc59 unit, and a truncated chrA–orf98 unit shared by the MDR regions from T7REPROV087 and T7REMF1; iii) a 71.2-kb Tn21-related element T21REPROV087 (see below) and a 21.2-kb T21REMF1 (see below) in the MDR regions from T7REPROV087 and T7REMF1, respectively; iv) IS26–mph(E)–IS26 unit acquired by the MDR regions from T7REMF1 and T7REPROV023; v) intact or truncated Tn4352 (containing aphA1) in the MDR regions from T7REMF1 and T7REPr-15-2-50; vi) a complex class 1 integron In37 with four resistance loci: VR1/GCA (aacA4cr–blaOXA-1–catB3–arr-3), a disrupted Tn2 (containing blaTEM-1), ΔTn6502a (containing blaCTX-M-3), and VR2 (ISCR1–rmtB unit) in the MDR region from T7REPROV023; and vii) ΔTn1548 carrying ISCR1–armA unit and an interrupted ISEc29–mph(E)–IS15DI unit in the MDR region from T7REPr-15-2-50.
Figure 9
Comparison of MDR regions from four Tn7 derivatives.
Comparison of MDR regions from four Tn7 derivatives.
Three Tn1696 Derivatives Tn6910, T1696RE
Herein, a detailed sequence comparison (Figure 10) was applied to the three Tn1696 derivatives Tn6910, T1696REPROV002, and T1696REPROV013 (identified as the inner components of Tn6874, Tn6861, and Tn6876, respectively; see above), together with Tn1696,35 Tn6338,36 and Tn690922 derived from GenBank. Tn1696 was one of the Tn3-family prototype unit transposons, and it was originally found in plasmid R1003 from Pseudomonas aeruginosa and contained a core backbone structure: IRL (inverted repeat left)–tnpA–tnpR (resolvase)–res (resolution site)–mer–IRR (inverted repeat right).35 These above five Tn1696 derivatives exhibited three major modular differences cross the core backbones: i) an intact version of tnpAR–res in Tn6338, Tn6909, and Tn6910, while a truncated version in T1696REPROV002 and T1696REPROV013; ii) a Tn1696 mer locus in Tn6338 and Tn6910, and a Tn21 mer locus instead in Tn6909; none of mer locus in T1696REPROV002 and T1696REPROV013; and iii) intact IRL/R in Tn6910, and IRL/R interrupted by insertion of IS4321 in Tn6909; only IRL in T1696REPROV013, and IRL interrupted by insertion of IS5075 in Tn6338; only IRR in Tn6338; none of IRL/R in T1696REPROV002.
Figure 10
Comparison of Tn1696 and its three derivatives.
Comparison of Tn1696 and its three derivatives.Each of these five Tn1696 derivatives acquired one or more integrons into res site, instead of In4 (GCA: aacC1–gcuE–aadA2–cmlA1) in Tn1696: a complex class 1 integron In469 in Tn6338, a different complex In469 in T1696REPROV002, a concise class 1 integron In27 (GCA: dfrA12–gcuF–aadA2) in T1696REPROV013, a distinct complex In27 in Tn6909, and the four tandem concise class 1 integrons including In1849 (GCA: dfrA1t–aacA4’–blaOXA-1), two copies of In994 (GCA: blaIMP-1), and In151 [GCA: catB8–lnu(F)3] in Tn6910. These two complex In469 harbored the same VR1/GCA (arr-3–dfrA27–aadA16), but differed from each other by capturing different additional VR modules: VR2 (ISCR1–qnrB4–blaDHA-1 unit) in In469 from Tn6338, and VR2 (ISCR1–blaPER-1 unit) together with VR3 (a truncated ISCR1–aphA6 unit, ΔISSod18, and ISEc29) in In469 from T1696REPROV002. The complex In27 from Tn6909 carried GCA (VR1) as above and still captured four additional resistance modules: a truncated ISCR1–qnrA1 unit (VR2), chrA–orf98 unit, a truncated IS26–mph(A)–IS6100 unit, and a concise class 1 integron In1209 (GCA: aadA1b–dfrA1b–aacA7–blaVIM-1).
Three Tn21 Derivatives Tn6974, T21RE
Herein, a detailed sequence comparison (Figure 11) was applied to the three Tn21 derivatives Tn6974, T21REPROV087, and T21REMF1 (identified as the inner components of Tn6873, T7REPROV087, and T7REMF1, respectively; see above), together with Tn2135 and Tn697537 derived from GenBank. Tn21 was another prototype of Tn3-family unit transposons, and it was initially found in Shigella flexneri plasmid R100 and had a core backbone structure: IRL–tnpA–tnpR–res–mer–IRR.35 The above four Tn21 derivatives shared Tn21 tnpAR–res, but their backbones had at least two major modular differences: i) a Tn21 mer locus in Tn6975, and a Tn1696 mer locus instead in Tn6974; none of mer locus in T21REMF1 and T21REPROV087; and ii) IRL/R in Tn6974 and Tn6975, while only IRL in T21REPROV087 and T21REMF1.
Figure 11
Comparison of Tn21 and its three derivatives.
Comparison of Tn21 and its three derivatives.Each of these four Tn21 derivatives acquired an integron instead of In2 (GCA: aadA1a) in Tn21: a concise class 1 integron In1808 (GCA: aacA4cr–blaOXA-1–ΔcatB3–arr-3–dfrA27–aadA16) in Tn6974, a concise class 1 integron In27 in Tn6975, a distinct complex In27 in T21REMF1, and a complex class 1 integron In263 in T21REPROV087. The concise In27 contained three resistance loci: dfrA12–gcuF–aadA2 (GCA), chrA–orf98 unit, and IS26–mph(A)–IS6100 unit. The complex In27 possessed a VR1/GCA as above and further acquired a VR2 (ISCR1–blaDHA-1 unit). The complex In263 had a VR1/GCA (aacA4cr–arr-3) and additionally nine copies of VR2 (truncated ISCR1–blaDHA-1 unit).
Transferability and Antimicrobial Susceptibility
Tn6862, which was selected to represent Tn6512-related ICEs, could be transferred from the wild-type PROV023 isolate into E. coli EC600 through conjugation, generating the transconjugant EC600/Tn6862. Both PROV023 and EC600/Tn6862 were highly resistant to amikacin with a minimum inhibitory concentration (MIC) value ≥64 μg/mL () owing to production of aminoglycoside 3’-phosphotransferase.
Plasmids in the Five Isolates Sequenced in This Study
Providencia rettgeri PROV087 and Providencia alcalifaciens PROV023 carried no plasmids. Identified were an IncC plasmid pPROV275-1NDM (carrying blaNDM-1, blaCMY-6, aacA4, rmtC, sul1, and qacED1), an IncpGZH766-NDM plasmid pPROV275-2NDM (having blaNDM-1, blaPER-4, blaOXA-10, aadA1, aadB, aphA6, catA1, catB8, sul1, and qacED1) together with a Col3M plasmid pPROV275-qnrD (containing qnrD1) in Providencia rettgeri PROV275, an IncW plasmid pPROV002-IMP (carrying blaIMP-4, blaNDM-1, aacA4, catB3, sul1, and qacED1) in Providencia rettgeri PROV002, and an IncpGZH766-NDM plasmid pPROV013-PER [containing blaPER-4, blaOXA-1, aacA4, mph(E), msr(E), arr-3, catB3, lnu(G), sul1, and qacED1] together with an IncFII plasmid pPROV013-NR (harboring no resistance genes) in Providencia alcalifaciens PROV013.
Discussion
This work presents the complete sequences of seven chromosomal AGEs in Providencia, and they can be further divided into four groups: three Tn6512-related ICEs Tn6860, Tn6861, and Tn6862; one Tn6872-related IME Tn6873; one Tn6875-related IME Tn6876; and two Tn7-related elements T7REPROV087 and T7REPROV023. Newly identified are five (except for T7REPROV087 and T7REPROV023) of the above seven AGEs, together with one unit transposon Tn6974, one composite transposon Tn6976, and two integrons In1793 and In1808, all of which are located within these five AGEs. There are additional 10 newly designated (firstly designated in this study, but with previously determined sequences) AGEs: i) one IME Tn6912 located within Tn6860 sequenced in this study; ii) one ICEs Tn6865, four IMEs Tn6872, Tn6874, Tn6875, and Tn6977, two unit transposons Tn6910 and Tn6975, one composite transposon Tn6896, and one IS element ISPrre11 which come from GenBank and are included in the sequence comparison herein. This is also the first report of identifying Tn6872 and Tn6875 in Providencia. The two previously designated ICEs ICEPalBan1 and ICEPreChnRF14-2 were renamed herein as standard Tn designations Tn6863 and Tn6864, respectively.The transposition of Tn7-related unit transposons, generating bracketed 5-bp DRs, is generally site-specific at 25-bp downstream of glmS, and these target sites with a consensus CCgcGtAAccTgGCaAAatcgGTtACgGTtGAgTaa38 together with Tn7-related elements are presented in a wide range of bacteria as mentioned above. Three of the four Tn7-related elements (all bracketed by 5-bp DRs) characterized herein, as expected, are located downstream of glmS, but T7REMF1 is presented within orf1389. The presence of bracketed 5-bp DRs denotes that T7REMF1 is most likely transposed directly into orf1389 that does not contain the above consensus target site; similarly, additional target sites, not resembling that downstream of glmS, have been proposed previously for integration of Tn7-related unit transposons.25 Tn6512-related ICEs have 17-bp highly conserved attP (attachment site on the ICE) sequences with a consensus ATcATcTCtCACCCtGA. Integration of these ICEs needs the 17-bp relatively non-conserved attB (attachment site on the chromosome) target sites within the 5’ end of prfC,32 and can occur with up to 23.5% sequence mismatch between attB and attP.39 These attB sites together with Tn6512-related ICEs are widely distributed in Gram-negative bacteria as mentioned above. Taken together, Tn7-related unit transposons and Tn6512-related ICEs have a wide host range.Conjugal transfer experiments confirm that Tn6512-related ICEs with the complete gene sets for conjugal transfer have the ability to transfer from the wild-type isolates of Proteus,40
Vibrio,41
Shewanella,42 and Providencia (this study) to the recipient bacteria E. coli, and that a circular extrachromosomal Tn6512 has the ability to transfer into the recipient bacteria Salmonella,39
Enterobacter,39 and Serratia,39 indicating that this group of ICEs have a very broad host range.Integration of Tn6872-related IMEs needs 24-bp highly conserved attB sequences (with a consensus AACCTAcAaTTtAACCTACACTTA) located 102-bp upstream of mutS. These attB sequences together with Tn6872-related elements are mainly found in Morganellaceae including Providencia (this study), Proteus (eg Proteus mirabilis with an accession number CP053894), and Morganella (eg M. morganii with an accession number CP064830). Integration of Tn6875-related IMEs recognizes 11-bp highly conserved attB sequences (with a consensus CcTatTTTATC) located 494-bp downstream of smpB. These attB sites together with Tn6875-related elements are mainly distributed in Morganellaceae including Providencia (this study), Proteus (eg Proteus mirabilis with an accession number CP042857), and Morganella (eg M. morganii with an accession number CP068145), as well as Enterobacteriaceae including Salmonella (eg S. enterica with an accession number CP053319) and Escherichia (eg E. coli with an accession number CP056565). It seems that these two groups of IMEs spread mainly in Morganellaceae and/or Enterobacteriaceae.Five (except for Tn6862 and Tn6865) of the seven Tn6512-related ICEs encode the toxin-antitoxin systems: HipBA in Tn6512, and AbiEi-ii in Tn6860, Tn6861, Tn6863, and Tn6864. These toxin-antitoxin systems have been shown to stabilize Tn6512-related ICEs by preventing their loss when they are extrachromosomal.43,44 The loss of the AbiEi-ii toxin-antitoxin system from Tn6512 and Tn6862 is due to the substitution of ICE backbone region 7 as shown in Figure 2, while that from Tn6865 is caused by the insertion of the MDR region within the ICE backbone gene umuC (site 3 in Figure 2). Five (except for Tn6512 and Tn6865) of the seven Tn6512-related ICEs and two (except for Tn6872) of the three Tn6872-related IMEs carry a type I restriction-modification system HsdMSR. This system can degrade the unmethylated incoming DNA to stabilize AGEs against challenges by competitor elements.45 Tn6512 lose the HsdMSR system because of the substitution of backbones region 5 (Figure 2). The loss of this system from Tn6865 is on account of the insertion of the MDR region. Tn6872 do not harbor this system due the substitution of orf201–to–orf207 region (Figure 5). It is speculated herein that the loss of toxin-antitoxin or restriction-modification systems might lead to the instability of relevant ICEs and IMEs after they are excised from the chromosomes.46Tn7-related unit transposons accomplish transposition using the core transposition module tnsABCDE.23 All the four Tn7-related elements analyzed herein have undergone the truncation of tnsABCDE, which results from the integration of MDR regions. The lesion of tnsABCDE in these four Tn7-related elements would lose their ability of intracellular transfer.Each of 13 of the totally characterized 17 AGEs carries a MDR region, and notably the resulting 13 MDR regions carry at least 49 drug resistance genes, which can be grouped into 15 different categories of antimicrobials and heavy metal, including β-lactam, aminoglycoside, macrolide, tetracycline, trimethoprim, phenicol, sulphonamide, lincosamide, rifampicin, quinolone, fosfomycin, bleomycin, chromate, quaternary ammonium compound, and mercury. Among these 49 drug resistance genes, predominantly found are those resistant to aminoglycosides (n=14) and β-lactams (n=9), which might be due to the wide clinical use of aminoglycosides and β-lactams against Providencia-induced infections. Intact or truncated versions of Tn21- and Tn1696-related transposons, Tn4352, Tn5393c, In27, ISCR2–floR unit, ISCR2–sul2 unit, ISEc29–mph(E)–IS15DI unit, and IS26–aacC4–aph(4)-Ia–ISEc59 unit are frequently found within these 13 MDR regions, indicating the assembly of these MDR regions from various collections of AGEs and associated resistance genes via complex transposition and homologous recombination. In addition, there are so called “concise” accessory modules (Tn6912, Tn6578, Tn6896, In2-4, In9-1, and mer region) that are presented in six AGEs, including four harboring MDR regions and two not. These “concise” accessory modules harbor at least 10 drug resistance genes, which mediate the resistance to six different categories of antimicrobials and heavy metal, including aminoglycoside, β-lactam, tetracycline, trimethoprim, macrolide, and mercury.On the one hand, these four groups of AGEs have a wide range of hosts including Providencia, and they are stable in the host bacteria and further able to transfer across different bacterial species. On the other hand, these four groups of AGEs display high-level diversification in modular structures, which have complex mosaic natures and carry a large number of drug resistance genes, and particularly different MDR regions are presented in these AGEs. Integration of these AGEs into the Providencia chromosomes contributes to the accumulation and distribution of drug resistance genes and enhances the ability of Providencia isolates to survive under drug selection pressure.