Interdiscip Sci Comput Life Sci DOI 10.1007/s12539-015-0135-6
ORIGINAL RESEARCH ARTICLE
Prediction of Type II Toxin-Antitoxin Loci in Klebsiella pneumoniae Genome Sequences Yi-Qing Wei1 • De-Xi Bi1 • Dong-Qing Wei1 • Hong-Yu Ou1
Received: 6 April 2015 / Revised: 10 May 2015 / Accepted: 26 May 2015 Ó International Association of Scientists in the Interdisciplinary Areas and Springer-Verlag Berlin Heidelberg 2015
Abstract Klebsiella pneumoniae is an increasingly important bacterial pathogen to human. This Gram-negative bacterium species has become a serious concern due to its dramatic increase in the levels of multiple antibiotic resistances, particularly to carbapenems. The toxin-antitoxin (TA) system has recently been reported to be involved in the formation of drug-tolerant persister cells. The type II TA system is composed of a stable toxin protein and a relatively unstable antitoxin protein that is able to inhibit the toxin. Here, we examine the type II TA locus distribution and compare the TA diversity throughout ten completely sequenced K. pneumoniae genomes by using bioinformatics approaches. Two hundred and twelve putative type II TA loci were identified in 30 replicons of these K. pneumoniae strains. The amino acid sequence similarity-based grouping shows that these loci distribute differently not only among different K. pneumoniae strains isolated from diverse sources, but also between their chromosomes and plasmids. Keywords Type II toxin-antitoxin loci Klebsiella pneumoniae In silico detection
Electronic supplementary material The online version of this article (doi:10.1007/s12539-015-0135-6) contains supplementary material, which is available to authorized users. & Dong-Qing Wei [email protected] & Hong-Yu Ou [email protected] 1
State Key Laboratory for Microbial Metabolism and School of Life Sciences and Biotechnology, Shanghai Jiaotong University, Shanghai 200030, China
1 Introduction Klebsiella pneumoniae is an important opportunistic pathogen to human. It is also a common environmental organism that is frequently found in soil, water and the surface of plants. With an increasing level of multiple antibiotic resistance, it causes urinary tract infections, nosocomial pneumonia and intraabdominal infections. Carbapenem-resistant K. pneumoniae strains, often associated with severe infections with few treatment options, have become an urgent threat to public health [1]. The comparative analysis for available complete genome sequences of K. pneumoniae strains has revealed that this species possesses an extremely plastic genome [2, 3]. Bacterial toxin-antitoxin (TA) system is composed of a stable toxin and a relatively unstable antitoxin which inhibits the toxin. TA systems are not essential for normal cell growth, but they may inhibit cell growth and lead to a programmed cell death-like process when exposed to disadvantageous conditions such as antibiotic, starvation, improper temperature [4]. In normally growing cells, toxins are neutralized by their cognate antitoxins, which are more labile than toxins and degraded under stress conditions. The cellular targets of toxins are diverse, including DNA replication, mRNA stability, biofilm formation, protein synthesis and ATP synthesis. Bacterial TA systems have been classified into five types according to the molecular nature of the antitoxin and its mode of interaction with the toxin [5]. Among all types of TA systems, the type II TA system is the most extensively studied. Both the antitoxin and toxin are proteins, and antitoxin neutralizes the toxicity of toxin by forming a toxinantitoxin complex. To date, type II TA system has been shown to be involved in plasmid maintenance, formation of persister cells, programmed cell death and stress resistance;
123
Interdiscip Sci Comput Life Sci
these systems are being exploited in synthetic biology for their wide applications in genetic manipulation [6, 7]. In type I systems, antitoxins are antisense RNA transcribed from the toxin genes. In type III, non-coding RNA antitoxins neutralize toxins by binding to toxin proteins. In YeeUV (YeeUCbtA) TA system, which is classified as a type IV TA system, the antitoxin YeeU functions as an antagonist for CbtA toxicity [8]. In type V TA system, the antitoxin cleaves the mRNA of toxin to inhibit its translation [9]. The type II TA systems are widely distributed in bacteria and archaea [10, 11]. For example, 79 TA loci are found in Mycobacterium tuberculosis [12], and no TA locus has been found in Buchnera aphidicola and Mycoplasma hyopneumoniae yet [11]. But the detailed distribution of TA systems among K. pneumoniae is unknown. In this study, we identified the type II TA loci in ten completely sequenced K. pneumoniae genomes, and compare the TA loci found in these chromosomes and plasmids. This in silico investigation of the content of TA loci will be useful in revealing the wider DNA blueprints of K. pneumoniae species and mapping out the extent and nature of chromosome-borne, functional-associated TA systems.
2 Materials and Methods 2.1 Klebsiella pneumoniae Genome Sequences and Annotations Nucleotide sequences and annotations of 23 plasmids and ten chromosomes of ten completely sequenced K. pneumoniae genomes were collected from the NCBI Refseq project. The included K. pneumoniae strains were 1084, 342, CG43, HS11286, JM45, KCTC 2242, MGH 78,578, NTUH K2044, subsp. rhinoscleromatis SB3432 and KPNIH27 (supplementary Table S1). 2.2 Prediction of TA Loci 2.2.1 Short ORF Identification Toxins and antitoxins are usually small proteins, containing 30–200 amino acid residues (a.a.) [13]. The small proteincoding genes (\300 bp) are sometimes missed in the original NCBI Refseq annotation. Therefore, Prodigal [14] was first used to identify small ORFs as a supplementary protein-coding regions due to its advantage in detection of short genes. 2.2.2 Protein Sequence Searches NCBI BLASTP was used to find candidates of type II TA proteins homologous to the toxin and antitoxin proteins archived by the type II toxin-antitoxin locus database,
123
TADB (http://bioinfo-mml.sjtu.edu.cn/TADB/) [15]. TADB currently contains 11,964 type II TA proteins, including 5985 toxins, 5975 antitoxins and four regulators, in which 78 TA pairs were experimentally validated. Considering low sequence similarities have been shown among toxins and even less among antitoxins, BLASTP hits were obtained when their corresponding identities were [40 % for toxin proteins, and [30 % for antitoxin proteins. 2.2.3 Conserved Domain Detections The hidden Markov model (HMM) profile approach was used to detect conserved domains of toxin or antitoxin protein candidates. We had collected and built 343 HMM profiles to feature type II TA protein-related conserved domains, which were then grouped into 47 superfamilies. The HMM profiles were searched against the toxin or antitoxin proteins by using HMMer 3.0 with the e-value of 0.01 [16]. 2.2.4 Identification of Putative Type II TA Loci Toxin and antitoxin protein candidates were obtained by BLASTP-based sequence searches and HMMer-based conserved domain searches as described above. Typical features of type II TA locus are considered to define whether a pair of two short genes is organized as an operon coding for both toxin and antitoxin proteins. Both toxins and antitoxins usually are small proteins with the size of 30–200 a.a. However, exceptions existed, for example, MosT toxin is 316 a.a. and its cognate antitoxin MosA is 277 a.a. in length [17], and HipA toxin and cognate antitoxin HipB is 440 and 88 a.a. in length [18]. Therefore, the length cutoff for toxin and antitoxin protein was set as 30–500 a.a. In addition, a type II TA locus is encoded by an operon consisting of a toxin gene and an antitoxin gene on the same strand, and the two TA genes usually overlap by a few bases or apart only by a few bases, with a distance of -20 to 30 bp. However, exception was also seen in some TA families such as the YeeUV family. After a comprehensive statistics analysis of the TADB data, the distance range was set as -20 to 150 bp. Thus, a putative K. pneumoniae TA pair was obtained based on these cutoffs mentioned above (Fig. 1). 2.3 TA Locus Group The obtained putative K. pneumoniae TA loci were then grouped based on significant sequence similarities of their encoding proteins. The TA pairs coding for toxins and antitoxins sharing a BLASTP identities [90 % and a mismatch \10 a.a. were considered as members of the same TA locus group.
Interdiscip Sci Comput Life Sci
Fig. 1 Flowchart of prediction approach of type II TA loci in ten completely sequenced K. pneumoniae genomes
Table S2, which were missed by NCBI annotations. After manual curations, these small ORFs show high sequence similarities among other K. pneumoniae genomes and were as such regarded as the potential protein-coding regions. Among ten K. pneumoniae chromosomes under analysis, three K. pneumoniae strains (342, CG43 and KCTC 2242) harbor maximally 18 TA loci while the K. pneumoniae strain 1084 carries the minimum TA pair numbers (13 TA loci) (supplementary Table S1). The chromosomes
3 Results and Discussion 3.1 Putative TA Loci Found in Ten Completely Sequenced K. pneumoniae Strains A total of 212 putative type II TA loci were detected in 33 replicons of ten K. pneumoniae strains (Table 1, supplementary Table S1 and S2). Forty Prodigal-predicted new small ORFs (37–393 bp in size) were available in supplementary
Table 1 Number of predicted TA loci found in 10 completely sequenced Klebsiella pneumoniae genomes No.
Organism
Host, source
Chr. size (Mb)
Number of plasmid
Number of TA in Chr.
Number of TA in plasmid
Number of TA found
1
K. pneumoniae 1084
Human, liver
5.39
0
13
–
13
2
K. pneumoniae 342
Maize, root
5.64
2
18
10
28
3
K. pneumoniae CG43
Human, liver
5.17
0
18
–
18
4
K. pneumoniae HS11286
Human, sputum
5.33
6
16
3
19
5
K. pneumoniae JM45
Human, colon
5.27
2
15
7
22
6
K. pneumoniae KCTC 2242
Human, blood
5.26
1
18
4
22
7
K. pneumoniae MGH 78578
Human, sputum
5.32
5
16
7
23
8
K. pneumoniae NTUH K2044
Human, blood
5.25
1
15
4
19
9
K. pneumoniae subsp. rhinoscleromatis SB3432
Human, nasal
5.27
1
17
2
19
10
K. pneumoniae KPNIH27
Human, groin
5.24
5
16
13
29
123
Interdiscip Sci Comput Life Sci Table 2 Twenty-nine TA groups classified based on significant amino acid sequence similarities of both toxin and antitoxin proteins Group Id
Group namea
Representative TA locus (Replicon|TA no.)
TAG_1c
PIN_AbrB_1
NC_009650|TA_2
2
d
relE_RHH_1
CP007731|TA_12
8
TAG_3b
PIN_AbrB_2
NC_012731|TA_7
6
TAG_4c
relE_1
NC_009649|TA_2
2
TAG_5d
mosAT_1
NC_017540|TA_1
6
TAG_6d
relE_2
NC_018522|TA_1
11
TAG_7d TAG_8d
relE_3 Fic_PHD
NC_011283|TA_10 CP007731|TA_4
9 6
TAG_9d
GNAT_HTH_1
NC_022566|TA_5
10
TAG_10d
GNAT_HTH_2
NC_022566|TA_11
10
TAG_11d
hipA_HTH
NC_018522|TA_7
9
TAG_12
relE_4
NC_018522|TA_3
2
TAG_13d
GNAT_HTH_3
NC_022082|TA_1
9
TAG_14c
GNAT_RHH_1
NC_011282|TA_3
5
TAG_15c
relBE_1
NC_016846|TA_2
5
TAG_16d
GNAT_RHH_2
CP007731|TA_11
TAG_17d
Fic_YhfG
NC_016845|TA_11
10
TAG_18d
GNAT_HTH_4
NC_009648|TA_9
5
TAG_19d
relBE_2
NC_021232|TA_11
3
TAG_20d
GNAT_HTH_5
NC_012731|TA_9
4
TAG_21b
COG5654_COG5642
NC_017541|TA_4
4
TAG_22c TAG_23c
relE_5 relBE_3
NC_016839|TA_1 NC_022078|TA_6
2 2
TAG_24d
relBE_4
NC_022566|TA_1
3
TAG_25d
PIN_RHH
NC_022082|TA_11
10
TAG_26c
MazF_RHH
CP007734|TA_1
3
TAG_27d
MazF
NC_022566|TA_4
2
TAG_28
GNAT_HTH_6
NC_017540|TA_12
10
TAG_48d
relE_Xre_1
CP007731|TA_5
TAG_2
d
d
a
Number of TA members
8
4
Each TA group is named after its classification of toxin and antitoxin based on the amino acid sequence alignments
b
Members of this TA group are either chromosome- or plasmid-borne
c
All members of this TA group are chromosome-borne
d
All members of this TA group are plasmid-borne
of the ten strains share similar size, ranging from 5.17 to 5.64 Mb. However, the number of the TA loci does not appear to be correlated with the chromosome size, since the largest and smallest chromosomes (strain 342 and CG43, respectively) both have 18 identified TA loci. Although sample size of this study was small, this phenomenon is consistent with the previous conclusion that number of TA loci has nothing to do with genome size [10, 11]. In eight K. pneumoniae strains that carry 23 native plasmids, 50 plasmid-borne TA loci were identified (supplementary Table S1). The plasmid pKPN-262 of the strain
123
KPNIH27 and plasmid p1 of the JM45 both carry six predicted TA loci, which is the most among all these plasmids under study. However, not all plasmids were predicted to carry TA loci. Seven of the 23 plasmids, ranging 1308–122,799 bp in length, have no TA locus predicted. The distribution of predicted TA loci on plasmid also revealed that replicon size has no direct correlation with number of TA loci. Considering the fact that the TA loci were frequently detected on plasmid, the plasmid-borne TA locus not only plays vital role in plasmid maintenance, but also is likely to be transferred between strains.
Interdiscip Sci Comput Life Sci Fig. 2 Distribution of 72 TA groups in 33 replicons of ten completely sequenced K. pneumonia strains. Heat map was generated with R (http:// www.r-project.org) from the BLASTp identities of toxin or antitoxin proteins reflecting TA group distribution in the replicons of K. pneumonia (details available in the supplementary Table S2)
3.2 TA Groups Found in K. pneumoniae Chromosome and Plasmid In general, family classification of the known type II TA loci is based on sequence similarity of toxin protein, associated with a specific cognate antitoxin protein [10, 19]. Not only on basis of the toxin protein sequence similarity but also on toxin tertiary structure, TADB has defined 11 two-component TA families (ccd, hicBA, hipBA, mazEF/chpA, mosAT, parD/PemKI, parDE, phddoc, relBE, vapBC and yeeUV), and 3 three-component TA families (x-e-f, pasABC, paaR-paaA-parE) [15]. In this study, we employ a stricter pair-wise alignment-based method to group type II TA loci found in the same bacterial species. The 212 predicted K. pneumoniae type II TA pairs were divided into 72 groups based on significant amino acid
sequence similarities (supplementary Table S3), in which 29 TA groups consist of multiple TA locus members (Table 2 and supplementary Table S4). Interestingly, those TA groups conserved among K. pneumonia chromosomes are found to be usually absent on K. pneumonia plasmids, whereas those TA groups frequently carried on plasmids are less conserved on K. pneumonia chromosomes (Fig. 2). RelBE is one of the best studied TA systems [20, 21]. The toxin, RelE, is capable of cleaving mRNA in the ribosomal A site and inhibits cell growth by reducing protein translation level to decrease amino acid consumption under amino acid starvation. Some toxins of other TA families show homology to RelE toxin, such as HigB, or RelE-like toxins, while the cognate antitoxin may not be RelB-like antitoxins. Interestingly, the relBE-like TA groups in our study show distinct distribution. The relBE_1 and relBE_2 TA loci distribute in the same K. pneumoniae
123
Interdiscip Sci Comput Life Sci
replicons, but with different similarities among the chromosomes and plasmids, which is consistent with the source of TA loci members in each group. All members belonged to the relBE_1 group are from K. pneumoniae plasmids, and those in relBE_2 are from K. pneumonia chromosomes. Of all the 72 group of K. pneumonia TA loci, there are 24 TA groups containing GNAT (Gcn5-related acetyl transferase) domain-related toxin proteins. In the corresponding antitoxins, the DNA binding-associated domains HTH, RHH and DUF1778 are usually found. Although some operons with toxin proteins coding for GNAT domains had been predicted as TA systems [22, 23], the genetic and molecular verification experiments have not been reported. Among the 24 GNAT-related TA groups, most of the GNAT_HTH TA groups were found to be present in all chromosomes of the ten K. pneumonia strains. In contrast, the GNAT_RHH and GNAT_DUF1778 TA groups were present on both chromosomes and plasmids, but those carried by chromosomes display lower amino acid sequence similarities. Fic_PHD TA group is present among the chromosomes and absent among the plasmids of the K. pneumonia strains analyzed. Fic domain-related proteins play critical roles in multiple cellular processes through AMPylation of (transfer of AMP to) target proteins. In addition, the Doc domain, a member of Fic domain family, was often encoded by the toxin genes. The Doc protein is a kinase that inhibits bacterial translation by phosphorylating the conserved threonine (Thr382) of the translation elongation factor EFTu [24]. MosAT TA group family is present among the chromosomes of seven K. pneumoniae strains (342, CG43, HS11286, JM45, KCTC 2242, MGH 78,578 and KPNIH27). It is the first TA system found to be harbored within an integrative and conjugative element, SXT in Vibrio cholerae MJ-1236 [17], and related to genetic maintenance of SXT. Interestingly, alignments of the genomic contexts of the mosAT locus show that three K. pneumoniae strains (1084, NTUH K2044 and subsp. rhinoscleromatis SB3432) lose a 2-kb bp DNA fragment coding for the MosAT system (supplementary Figure S1).
4 Summary In this study, 212 putative type II TA loci were identified in ten completely sequenced K. pneumoniae strains by using BLASTP- and HMMer-based search approaches. The TA group classification based on significant amino acid sequence similarities shows that these type II TA loci distribute differently not only among different K. pneumoniae strains isolated from diverse sources, but also
123
between chromosomes and plasmids. The function of plasmid-borne type II TA systems can be explained for their post-segregational killing effect; however, the physiological functions of diverse TA systems on chromosomes remain to be explored, especially considering its prevalence. Whether the TA systems promote the niche fitness and facilitate the genome evolution of K. pneumoniae awaits further investigation within the population. Acknowledgments This study was supported by partial funding from the 973 program, Ministry of Science and Technology, China (2015CB554202, 2012CB721000); the National Natural Science Foundation of China (31170082); the Specialized Research Fund for the Doctoral Program of Higher Education, China (20130073110062).
References 1. Nordmann P, Cuzon G, Naas T (2009) The real threat of Klebsiella pneumoniae carbapenemase-producing bacteria. Lancet Infect Dis 9(4):228–236 2. Liu P, Li P, Jiang X, Bi D, Xie Y, Tai C, Deng Z, Rajakumar K, Ou HY (2012) Complete genome sequence of Klebsiella pneumoniae subsp. pneumoniae HS11286, a multidrug-resistant strain isolated from human sputum. J Bacteriol 194(7):1841–1842 3. Zhang J, van Aartsen JJ, Jiang X, Shao Y, Tai C, He X, Tan Z, Deng Z, Jia S, Rajakumar K, Ou HY (2011) Expansion of the known Klebsiella pneumoniae species gene pool by characterization of novel alien DNA islands integrated into tmRNA gene sites. J Microbiol Methods 84(2):283–289 4. Yamaguchi Y, Park JH, Inouye M (2011) Toxin-antitoxin systems in bacteria and archaea. Annu Rev Genet 45:61–79 5. Ghafourian S, Raftari M, Nourkhoda S, Sekawi Z. (2013) Toxinantitoxin systems: classification, biological function and application in biotechnology. Horizon Scientific Press, New York, vol 16, no Biol, pp 9–14 6. Wang X, Wood TK (2011) Toxin-antitoxin systems influence biofilm and persister cell formation and the general stress response. Appl Environ Microbiol 77(16):5577–5583 7. Maisonneuve E, Gerdes K (2014) Molecular mechanisms underlying bacterial persisters. Cell 157(3):539–548 8. Masuda H, Tan Q, Awano N, Wu KP, Inouye M (2012) YeeU enhances the bundling of cytoskeletal polymers of MreB and FtsZ, antagonizing the CbtA (YeeV) toxicity in Escherichia coli. Mol Microbiol 84(5):979–989 9. Wang X, Lord DM, Cheng HY, Osbourne DO, Hong SH, Sanchez-Torres V, Quiroga C, Zheng K, Herrmann T, Peti W, Benedik MJ, Page R, Wood TK (2012) A new type V toxinantitoxin system where mRNA for toxin GhoT is cleaved by antitoxin GhoS. Nat Chem Biol 8(10):855–861 10. Pandey DP, Gerdes K (2005) Toxin-antitoxin loci are highly abundant in free-living but lost from host-associated prokaryotes. Nucleic Acids Res 33(3):966–976 11. Ou HY, Wei Y, Bi D (2013) Type II toxin-antitoxin loci: phylogeny. In: Prokaryotic toxin-antitoxins. Springer, Heidelberg, pp 239–247 12. Sala A, Bordes P, Genevaux P (2014) Multiple toxin-antitoxin systems in Mycobacterium tuberculosis. Toxins (Basel) 6(3):1002–1020 13. Sevin EW, Barloy-Hubler F (2007) RASTA-Bacteria: a webbased tool for identifying toxin-antitoxin loci in prokaryotes. Genome Biol 8(8):R155
Interdiscip Sci Comput Life Sci 14. Hyatt D, Chen GL, Locascio PF, Land ML, Larimer FW, Hauser LJ (2010) Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinform 11:119 15. Shao Y, Harrison EM, Bi D, Tai C, He X, Ou HY, Rajakumar K, Deng Z (2011) TADB: a web-based resource for type 2 toxinantitoxin loci in bacteria and archaea, Nucleic Acids Res 39(Database issue):D606–D611 16. Finn RD, Clements J, Eddy SR (2011) HMMER web server: interactive sequence similarity searching. Nucleic Acids Res 39(Web Server issue):W29–W37 17. Wozniak RA, Waldor MK (2009) A toxin-antitoxin system promotes the maintenance of an integrative conjugative element. PLoS Genet 5(3):e1000439 18. Black DS, Kelly AJ, Mardis MJ, Moyed HS (1991) Structure and organization of hip, an operon that affects lethality due to inhibition of peptidoglycan or DNA synthesis. J Bacteriol 173(18):5732–5739 19. Van Melderen L, Saavedra De Bast M (2009) Bacterial toxinantitoxin systems: more than selfish entities? PLoS Genet 5(3):e1000437
20. Boggild A, Sofos N, Andersen KR, Feddersen A, Easter AD, Passmore LA, Brodersen DE (2012) The crystal structure of the intact E. coli RelBE toxin-antitoxin complex provides the structural basis for conditional cooperativity. Structure 20(10):1641–1648 21. Tashiro Y, Kawata K, Taniuchi A, Kakinuma K, May T, Okabe S (2012) RelE-mediated dormancy is enhanced at high cell density in Escherichia coli. J Bacteriol 194(5):1169–1176 22. Makarova KS, Wolf YI, Koonin EV (2009) Comprehensive comparative-genomic analysis of type 2 toxin-antitoxin systems and related mobile stress response systems in prokaryotes. Biol Direct 4:19 23. Jurenaite M, Markuckas A, Suziedeliene E (2013) Identification and characterization of type II toxin-antitoxin systems in the opportunistic pathogen Acinetobacter baumannii. J Bacteriol 195(14):3165–3172 24. Castro-Roa D, Garcia-Pino A, De Gieter S, van Nuland NA, Loris R, Zenkin N (2013) The Fic protein Doc uses an inverted substrate to phosphorylate and inactivate EF-Tu. Nat Chem Biol 9(12):811–817
123
ORIGINAL RESEARCH ARTICLE
Prediction of Type II Toxin-Antitoxin Loci in Klebsiella pneumoniae Genome Sequences Yi-Qing Wei1 • De-Xi Bi1 • Dong-Qing Wei1 • Hong-Yu Ou1
Received: 6 April 2015 / Revised: 10 May 2015 / Accepted: 26 May 2015 Ó International Association of Scientists in the Interdisciplinary Areas and Springer-Verlag Berlin Heidelberg 2015
Abstract Klebsiella pneumoniae is an increasingly important bacterial pathogen to human. This Gram-negative bacterium species has become a serious concern due to its dramatic increase in the levels of multiple antibiotic resistances, particularly to carbapenems. The toxin-antitoxin (TA) system has recently been reported to be involved in the formation of drug-tolerant persister cells. The type II TA system is composed of a stable toxin protein and a relatively unstable antitoxin protein that is able to inhibit the toxin. Here, we examine the type II TA locus distribution and compare the TA diversity throughout ten completely sequenced K. pneumoniae genomes by using bioinformatics approaches. Two hundred and twelve putative type II TA loci were identified in 30 replicons of these K. pneumoniae strains. The amino acid sequence similarity-based grouping shows that these loci distribute differently not only among different K. pneumoniae strains isolated from diverse sources, but also between their chromosomes and plasmids. Keywords Type II toxin-antitoxin loci Klebsiella pneumoniae In silico detection
Electronic supplementary material The online version of this article (doi:10.1007/s12539-015-0135-6) contains supplementary material, which is available to authorized users. & Dong-Qing Wei [email protected] & Hong-Yu Ou [email protected] 1
State Key Laboratory for Microbial Metabolism and School of Life Sciences and Biotechnology, Shanghai Jiaotong University, Shanghai 200030, China
1 Introduction Klebsiella pneumoniae is an important opportunistic pathogen to human. It is also a common environmental organism that is frequently found in soil, water and the surface of plants. With an increasing level of multiple antibiotic resistance, it causes urinary tract infections, nosocomial pneumonia and intraabdominal infections. Carbapenem-resistant K. pneumoniae strains, often associated with severe infections with few treatment options, have become an urgent threat to public health [1]. The comparative analysis for available complete genome sequences of K. pneumoniae strains has revealed that this species possesses an extremely plastic genome [2, 3]. Bacterial toxin-antitoxin (TA) system is composed of a stable toxin and a relatively unstable antitoxin which inhibits the toxin. TA systems are not essential for normal cell growth, but they may inhibit cell growth and lead to a programmed cell death-like process when exposed to disadvantageous conditions such as antibiotic, starvation, improper temperature [4]. In normally growing cells, toxins are neutralized by their cognate antitoxins, which are more labile than toxins and degraded under stress conditions. The cellular targets of toxins are diverse, including DNA replication, mRNA stability, biofilm formation, protein synthesis and ATP synthesis. Bacterial TA systems have been classified into five types according to the molecular nature of the antitoxin and its mode of interaction with the toxin [5]. Among all types of TA systems, the type II TA system is the most extensively studied. Both the antitoxin and toxin are proteins, and antitoxin neutralizes the toxicity of toxin by forming a toxinantitoxin complex. To date, type II TA system has been shown to be involved in plasmid maintenance, formation of persister cells, programmed cell death and stress resistance;
123
Interdiscip Sci Comput Life Sci
these systems are being exploited in synthetic biology for their wide applications in genetic manipulation [6, 7]. In type I systems, antitoxins are antisense RNA transcribed from the toxin genes. In type III, non-coding RNA antitoxins neutralize toxins by binding to toxin proteins. In YeeUV (YeeUCbtA) TA system, which is classified as a type IV TA system, the antitoxin YeeU functions as an antagonist for CbtA toxicity [8]. In type V TA system, the antitoxin cleaves the mRNA of toxin to inhibit its translation [9]. The type II TA systems are widely distributed in bacteria and archaea [10, 11]. For example, 79 TA loci are found in Mycobacterium tuberculosis [12], and no TA locus has been found in Buchnera aphidicola and Mycoplasma hyopneumoniae yet [11]. But the detailed distribution of TA systems among K. pneumoniae is unknown. In this study, we identified the type II TA loci in ten completely sequenced K. pneumoniae genomes, and compare the TA loci found in these chromosomes and plasmids. This in silico investigation of the content of TA loci will be useful in revealing the wider DNA blueprints of K. pneumoniae species and mapping out the extent and nature of chromosome-borne, functional-associated TA systems.
2 Materials and Methods 2.1 Klebsiella pneumoniae Genome Sequences and Annotations Nucleotide sequences and annotations of 23 plasmids and ten chromosomes of ten completely sequenced K. pneumoniae genomes were collected from the NCBI Refseq project. The included K. pneumoniae strains were 1084, 342, CG43, HS11286, JM45, KCTC 2242, MGH 78,578, NTUH K2044, subsp. rhinoscleromatis SB3432 and KPNIH27 (supplementary Table S1). 2.2 Prediction of TA Loci 2.2.1 Short ORF Identification Toxins and antitoxins are usually small proteins, containing 30–200 amino acid residues (a.a.) [13]. The small proteincoding genes (\300 bp) are sometimes missed in the original NCBI Refseq annotation. Therefore, Prodigal [14] was first used to identify small ORFs as a supplementary protein-coding regions due to its advantage in detection of short genes. 2.2.2 Protein Sequence Searches NCBI BLASTP was used to find candidates of type II TA proteins homologous to the toxin and antitoxin proteins archived by the type II toxin-antitoxin locus database,
123
TADB (http://bioinfo-mml.sjtu.edu.cn/TADB/) [15]. TADB currently contains 11,964 type II TA proteins, including 5985 toxins, 5975 antitoxins and four regulators, in which 78 TA pairs were experimentally validated. Considering low sequence similarities have been shown among toxins and even less among antitoxins, BLASTP hits were obtained when their corresponding identities were [40 % for toxin proteins, and [30 % for antitoxin proteins. 2.2.3 Conserved Domain Detections The hidden Markov model (HMM) profile approach was used to detect conserved domains of toxin or antitoxin protein candidates. We had collected and built 343 HMM profiles to feature type II TA protein-related conserved domains, which were then grouped into 47 superfamilies. The HMM profiles were searched against the toxin or antitoxin proteins by using HMMer 3.0 with the e-value of 0.01 [16]. 2.2.4 Identification of Putative Type II TA Loci Toxin and antitoxin protein candidates were obtained by BLASTP-based sequence searches and HMMer-based conserved domain searches as described above. Typical features of type II TA locus are considered to define whether a pair of two short genes is organized as an operon coding for both toxin and antitoxin proteins. Both toxins and antitoxins usually are small proteins with the size of 30–200 a.a. However, exceptions existed, for example, MosT toxin is 316 a.a. and its cognate antitoxin MosA is 277 a.a. in length [17], and HipA toxin and cognate antitoxin HipB is 440 and 88 a.a. in length [18]. Therefore, the length cutoff for toxin and antitoxin protein was set as 30–500 a.a. In addition, a type II TA locus is encoded by an operon consisting of a toxin gene and an antitoxin gene on the same strand, and the two TA genes usually overlap by a few bases or apart only by a few bases, with a distance of -20 to 30 bp. However, exception was also seen in some TA families such as the YeeUV family. After a comprehensive statistics analysis of the TADB data, the distance range was set as -20 to 150 bp. Thus, a putative K. pneumoniae TA pair was obtained based on these cutoffs mentioned above (Fig. 1). 2.3 TA Locus Group The obtained putative K. pneumoniae TA loci were then grouped based on significant sequence similarities of their encoding proteins. The TA pairs coding for toxins and antitoxins sharing a BLASTP identities [90 % and a mismatch \10 a.a. were considered as members of the same TA locus group.
Interdiscip Sci Comput Life Sci
Fig. 1 Flowchart of prediction approach of type II TA loci in ten completely sequenced K. pneumoniae genomes
Table S2, which were missed by NCBI annotations. After manual curations, these small ORFs show high sequence similarities among other K. pneumoniae genomes and were as such regarded as the potential protein-coding regions. Among ten K. pneumoniae chromosomes under analysis, three K. pneumoniae strains (342, CG43 and KCTC 2242) harbor maximally 18 TA loci while the K. pneumoniae strain 1084 carries the minimum TA pair numbers (13 TA loci) (supplementary Table S1). The chromosomes
3 Results and Discussion 3.1 Putative TA Loci Found in Ten Completely Sequenced K. pneumoniae Strains A total of 212 putative type II TA loci were detected in 33 replicons of ten K. pneumoniae strains (Table 1, supplementary Table S1 and S2). Forty Prodigal-predicted new small ORFs (37–393 bp in size) were available in supplementary
Table 1 Number of predicted TA loci found in 10 completely sequenced Klebsiella pneumoniae genomes No.
Organism
Host, source
Chr. size (Mb)
Number of plasmid
Number of TA in Chr.
Number of TA in plasmid
Number of TA found
1
K. pneumoniae 1084
Human, liver
5.39
0
13
–
13
2
K. pneumoniae 342
Maize, root
5.64
2
18
10
28
3
K. pneumoniae CG43
Human, liver
5.17
0
18
–
18
4
K. pneumoniae HS11286
Human, sputum
5.33
6
16
3
19
5
K. pneumoniae JM45
Human, colon
5.27
2
15
7
22
6
K. pneumoniae KCTC 2242
Human, blood
5.26
1
18
4
22
7
K. pneumoniae MGH 78578
Human, sputum
5.32
5
16
7
23
8
K. pneumoniae NTUH K2044
Human, blood
5.25
1
15
4
19
9
K. pneumoniae subsp. rhinoscleromatis SB3432
Human, nasal
5.27
1
17
2
19
10
K. pneumoniae KPNIH27
Human, groin
5.24
5
16
13
29
123
Interdiscip Sci Comput Life Sci Table 2 Twenty-nine TA groups classified based on significant amino acid sequence similarities of both toxin and antitoxin proteins Group Id
Group namea
Representative TA locus (Replicon|TA no.)
TAG_1c
PIN_AbrB_1
NC_009650|TA_2
2
d
relE_RHH_1
CP007731|TA_12
8
TAG_3b
PIN_AbrB_2
NC_012731|TA_7
6
TAG_4c
relE_1
NC_009649|TA_2
2
TAG_5d
mosAT_1
NC_017540|TA_1
6
TAG_6d
relE_2
NC_018522|TA_1
11
TAG_7d TAG_8d
relE_3 Fic_PHD
NC_011283|TA_10 CP007731|TA_4
9 6
TAG_9d
GNAT_HTH_1
NC_022566|TA_5
10
TAG_10d
GNAT_HTH_2
NC_022566|TA_11
10
TAG_11d
hipA_HTH
NC_018522|TA_7
9
TAG_12
relE_4
NC_018522|TA_3
2
TAG_13d
GNAT_HTH_3
NC_022082|TA_1
9
TAG_14c
GNAT_RHH_1
NC_011282|TA_3
5
TAG_15c
relBE_1
NC_016846|TA_2
5
TAG_16d
GNAT_RHH_2
CP007731|TA_11
TAG_17d
Fic_YhfG
NC_016845|TA_11
10
TAG_18d
GNAT_HTH_4
NC_009648|TA_9
5
TAG_19d
relBE_2
NC_021232|TA_11
3
TAG_20d
GNAT_HTH_5
NC_012731|TA_9
4
TAG_21b
COG5654_COG5642
NC_017541|TA_4
4
TAG_22c TAG_23c
relE_5 relBE_3
NC_016839|TA_1 NC_022078|TA_6
2 2
TAG_24d
relBE_4
NC_022566|TA_1
3
TAG_25d
PIN_RHH
NC_022082|TA_11
10
TAG_26c
MazF_RHH
CP007734|TA_1
3
TAG_27d
MazF
NC_022566|TA_4
2
TAG_28
GNAT_HTH_6
NC_017540|TA_12
10
TAG_48d
relE_Xre_1
CP007731|TA_5
TAG_2
d
d
a
Number of TA members
8
4
Each TA group is named after its classification of toxin and antitoxin based on the amino acid sequence alignments
b
Members of this TA group are either chromosome- or plasmid-borne
c
All members of this TA group are chromosome-borne
d
All members of this TA group are plasmid-borne
of the ten strains share similar size, ranging from 5.17 to 5.64 Mb. However, the number of the TA loci does not appear to be correlated with the chromosome size, since the largest and smallest chromosomes (strain 342 and CG43, respectively) both have 18 identified TA loci. Although sample size of this study was small, this phenomenon is consistent with the previous conclusion that number of TA loci has nothing to do with genome size [10, 11]. In eight K. pneumoniae strains that carry 23 native plasmids, 50 plasmid-borne TA loci were identified (supplementary Table S1). The plasmid pKPN-262 of the strain
123
KPNIH27 and plasmid p1 of the JM45 both carry six predicted TA loci, which is the most among all these plasmids under study. However, not all plasmids were predicted to carry TA loci. Seven of the 23 plasmids, ranging 1308–122,799 bp in length, have no TA locus predicted. The distribution of predicted TA loci on plasmid also revealed that replicon size has no direct correlation with number of TA loci. Considering the fact that the TA loci were frequently detected on plasmid, the plasmid-borne TA locus not only plays vital role in plasmid maintenance, but also is likely to be transferred between strains.
Interdiscip Sci Comput Life Sci Fig. 2 Distribution of 72 TA groups in 33 replicons of ten completely sequenced K. pneumonia strains. Heat map was generated with R (http:// www.r-project.org) from the BLASTp identities of toxin or antitoxin proteins reflecting TA group distribution in the replicons of K. pneumonia (details available in the supplementary Table S2)
3.2 TA Groups Found in K. pneumoniae Chromosome and Plasmid In general, family classification of the known type II TA loci is based on sequence similarity of toxin protein, associated with a specific cognate antitoxin protein [10, 19]. Not only on basis of the toxin protein sequence similarity but also on toxin tertiary structure, TADB has defined 11 two-component TA families (ccd, hicBA, hipBA, mazEF/chpA, mosAT, parD/PemKI, parDE, phddoc, relBE, vapBC and yeeUV), and 3 three-component TA families (x-e-f, pasABC, paaR-paaA-parE) [15]. In this study, we employ a stricter pair-wise alignment-based method to group type II TA loci found in the same bacterial species. The 212 predicted K. pneumoniae type II TA pairs were divided into 72 groups based on significant amino acid
sequence similarities (supplementary Table S3), in which 29 TA groups consist of multiple TA locus members (Table 2 and supplementary Table S4). Interestingly, those TA groups conserved among K. pneumonia chromosomes are found to be usually absent on K. pneumonia plasmids, whereas those TA groups frequently carried on plasmids are less conserved on K. pneumonia chromosomes (Fig. 2). RelBE is one of the best studied TA systems [20, 21]. The toxin, RelE, is capable of cleaving mRNA in the ribosomal A site and inhibits cell growth by reducing protein translation level to decrease amino acid consumption under amino acid starvation. Some toxins of other TA families show homology to RelE toxin, such as HigB, or RelE-like toxins, while the cognate antitoxin may not be RelB-like antitoxins. Interestingly, the relBE-like TA groups in our study show distinct distribution. The relBE_1 and relBE_2 TA loci distribute in the same K. pneumoniae
123
Interdiscip Sci Comput Life Sci
replicons, but with different similarities among the chromosomes and plasmids, which is consistent with the source of TA loci members in each group. All members belonged to the relBE_1 group are from K. pneumoniae plasmids, and those in relBE_2 are from K. pneumonia chromosomes. Of all the 72 group of K. pneumonia TA loci, there are 24 TA groups containing GNAT (Gcn5-related acetyl transferase) domain-related toxin proteins. In the corresponding antitoxins, the DNA binding-associated domains HTH, RHH and DUF1778 are usually found. Although some operons with toxin proteins coding for GNAT domains had been predicted as TA systems [22, 23], the genetic and molecular verification experiments have not been reported. Among the 24 GNAT-related TA groups, most of the GNAT_HTH TA groups were found to be present in all chromosomes of the ten K. pneumonia strains. In contrast, the GNAT_RHH and GNAT_DUF1778 TA groups were present on both chromosomes and plasmids, but those carried by chromosomes display lower amino acid sequence similarities. Fic_PHD TA group is present among the chromosomes and absent among the plasmids of the K. pneumonia strains analyzed. Fic domain-related proteins play critical roles in multiple cellular processes through AMPylation of (transfer of AMP to) target proteins. In addition, the Doc domain, a member of Fic domain family, was often encoded by the toxin genes. The Doc protein is a kinase that inhibits bacterial translation by phosphorylating the conserved threonine (Thr382) of the translation elongation factor EFTu [24]. MosAT TA group family is present among the chromosomes of seven K. pneumoniae strains (342, CG43, HS11286, JM45, KCTC 2242, MGH 78,578 and KPNIH27). It is the first TA system found to be harbored within an integrative and conjugative element, SXT in Vibrio cholerae MJ-1236 [17], and related to genetic maintenance of SXT. Interestingly, alignments of the genomic contexts of the mosAT locus show that three K. pneumoniae strains (1084, NTUH K2044 and subsp. rhinoscleromatis SB3432) lose a 2-kb bp DNA fragment coding for the MosAT system (supplementary Figure S1).
4 Summary In this study, 212 putative type II TA loci were identified in ten completely sequenced K. pneumoniae strains by using BLASTP- and HMMer-based search approaches. The TA group classification based on significant amino acid sequence similarities shows that these type II TA loci distribute differently not only among different K. pneumoniae strains isolated from diverse sources, but also
123
between chromosomes and plasmids. The function of plasmid-borne type II TA systems can be explained for their post-segregational killing effect; however, the physiological functions of diverse TA systems on chromosomes remain to be explored, especially considering its prevalence. Whether the TA systems promote the niche fitness and facilitate the genome evolution of K. pneumoniae awaits further investigation within the population. Acknowledgments This study was supported by partial funding from the 973 program, Ministry of Science and Technology, China (2015CB554202, 2012CB721000); the National Natural Science Foundation of China (31170082); the Specialized Research Fund for the Doctoral Program of Higher Education, China (20130073110062).
References 1. Nordmann P, Cuzon G, Naas T (2009) The real threat of Klebsiella pneumoniae carbapenemase-producing bacteria. Lancet Infect Dis 9(4):228–236 2. Liu P, Li P, Jiang X, Bi D, Xie Y, Tai C, Deng Z, Rajakumar K, Ou HY (2012) Complete genome sequence of Klebsiella pneumoniae subsp. pneumoniae HS11286, a multidrug-resistant strain isolated from human sputum. J Bacteriol 194(7):1841–1842 3. Zhang J, van Aartsen JJ, Jiang X, Shao Y, Tai C, He X, Tan Z, Deng Z, Jia S, Rajakumar K, Ou HY (2011) Expansion of the known Klebsiella pneumoniae species gene pool by characterization of novel alien DNA islands integrated into tmRNA gene sites. J Microbiol Methods 84(2):283–289 4. Yamaguchi Y, Park JH, Inouye M (2011) Toxin-antitoxin systems in bacteria and archaea. Annu Rev Genet 45:61–79 5. Ghafourian S, Raftari M, Nourkhoda S, Sekawi Z. (2013) Toxinantitoxin systems: classification, biological function and application in biotechnology. Horizon Scientific Press, New York, vol 16, no Biol, pp 9–14 6. Wang X, Wood TK (2011) Toxin-antitoxin systems influence biofilm and persister cell formation and the general stress response. Appl Environ Microbiol 77(16):5577–5583 7. Maisonneuve E, Gerdes K (2014) Molecular mechanisms underlying bacterial persisters. Cell 157(3):539–548 8. Masuda H, Tan Q, Awano N, Wu KP, Inouye M (2012) YeeU enhances the bundling of cytoskeletal polymers of MreB and FtsZ, antagonizing the CbtA (YeeV) toxicity in Escherichia coli. Mol Microbiol 84(5):979–989 9. Wang X, Lord DM, Cheng HY, Osbourne DO, Hong SH, Sanchez-Torres V, Quiroga C, Zheng K, Herrmann T, Peti W, Benedik MJ, Page R, Wood TK (2012) A new type V toxinantitoxin system where mRNA for toxin GhoT is cleaved by antitoxin GhoS. Nat Chem Biol 8(10):855–861 10. Pandey DP, Gerdes K (2005) Toxin-antitoxin loci are highly abundant in free-living but lost from host-associated prokaryotes. Nucleic Acids Res 33(3):966–976 11. Ou HY, Wei Y, Bi D (2013) Type II toxin-antitoxin loci: phylogeny. In: Prokaryotic toxin-antitoxins. Springer, Heidelberg, pp 239–247 12. Sala A, Bordes P, Genevaux P (2014) Multiple toxin-antitoxin systems in Mycobacterium tuberculosis. Toxins (Basel) 6(3):1002–1020 13. Sevin EW, Barloy-Hubler F (2007) RASTA-Bacteria: a webbased tool for identifying toxin-antitoxin loci in prokaryotes. Genome Biol 8(8):R155
Interdiscip Sci Comput Life Sci 14. Hyatt D, Chen GL, Locascio PF, Land ML, Larimer FW, Hauser LJ (2010) Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinform 11:119 15. Shao Y, Harrison EM, Bi D, Tai C, He X, Ou HY, Rajakumar K, Deng Z (2011) TADB: a web-based resource for type 2 toxinantitoxin loci in bacteria and archaea, Nucleic Acids Res 39(Database issue):D606–D611 16. Finn RD, Clements J, Eddy SR (2011) HMMER web server: interactive sequence similarity searching. Nucleic Acids Res 39(Web Server issue):W29–W37 17. Wozniak RA, Waldor MK (2009) A toxin-antitoxin system promotes the maintenance of an integrative conjugative element. PLoS Genet 5(3):e1000439 18. Black DS, Kelly AJ, Mardis MJ, Moyed HS (1991) Structure and organization of hip, an operon that affects lethality due to inhibition of peptidoglycan or DNA synthesis. J Bacteriol 173(18):5732–5739 19. Van Melderen L, Saavedra De Bast M (2009) Bacterial toxinantitoxin systems: more than selfish entities? PLoS Genet 5(3):e1000437
20. Boggild A, Sofos N, Andersen KR, Feddersen A, Easter AD, Passmore LA, Brodersen DE (2012) The crystal structure of the intact E. coli RelBE toxin-antitoxin complex provides the structural basis for conditional cooperativity. Structure 20(10):1641–1648 21. Tashiro Y, Kawata K, Taniuchi A, Kakinuma K, May T, Okabe S (2012) RelE-mediated dormancy is enhanced at high cell density in Escherichia coli. J Bacteriol 194(5):1169–1176 22. Makarova KS, Wolf YI, Koonin EV (2009) Comprehensive comparative-genomic analysis of type 2 toxin-antitoxin systems and related mobile stress response systems in prokaryotes. Biol Direct 4:19 23. Jurenaite M, Markuckas A, Suziedeliene E (2013) Identification and characterization of type II toxin-antitoxin systems in the opportunistic pathogen Acinetobacter baumannii. J Bacteriol 195(14):3165–3172 24. Castro-Roa D, Garcia-Pino A, De Gieter S, van Nuland NA, Loris R, Zenkin N (2013) The Fic protein Doc uses an inverted substrate to phosphorylate and inactivate EF-Tu. Nat Chem Biol 9(12):811–817
123
