Large-scale comparison of toxin and antitoxins in Listeria monocytogenes

(1)

toxins

Article

Large-Scale Comparison of Toxin and Antitoxins in

Listeria monocytogenes

José Antonio Agüero1,2, Hatice Akarsu2,3,4 , Lisandra Aguilar-Bultet4,5, Anna Oevermann4

and Laurent Falquet2,3,*

1 _{CENSA National Center for Animal and Plant Health, San José de las Lajas Municipality 32700, Mayabeque,}

Cuba; jaaguero@censa.edu.cu

2 _{SIB Swiss Institute of Bioinformatics, 1015 Lausanne, Switzerland; hatice.akarsuegger@vetsuisse.unibe.ch} 3 _{Department of Biology, UniFr University of Fribourg, 1700 Fribourg, Switzerland}

4 _{Vetsuisse Faculty, University of Bern, 3012 Bern, Switzerland; lisandra.aguilarbultet@usb.ch (L.A.-B.);}

anna.oevermann@vetsuisse.unibe.ch (A.O.)

5 _{USB University Hospital Basel, 4031 Basel, Switzerland}

* Correspondence: Laurent.Falquet@unifr.ch

Received: 27 September 2019; Accepted: 20 December 2019; Published: 2 January 2020 

Abstract: Toxin–antitoxin systems (TASs) are widely distributed in prokaryotes and encode pairs

of genes involved in many bacterial biological processes and mechanisms, including pathogenesis. The TASs have not been extensively studied in Listeria monocytogenes (Lm), a pathogenic bacterium of the Firmicutes phylum causing infections in animals and humans. Using our recently published TASmania database, we focused on the known and new putative TASs in 352 Listeria monocytogenes genomes and identified the putative core gene TASs (cgTASs) with the Pasteur BIGSdb-Lm database and, by complementarity, the putative accessory gene TAS (acTASs). We combined the cgTASs with those of an additional 227 L. monocytogenes isolates from our previous studies containing metadata information. We discovered that the differences in 14 cgTAS alleles are sufficient to separate the four main lineages of Listeria monocytogenes. Analyzing these differences in more details, we uncovered potentially co-evolving residues in some pairs of proteins in cgTASs, probably essential for protein–protein interactions within the TAS complex.

Keywords: toxin–antitoxin systems; Listeria monocytogenes; co-evolution

Key Contribution:Systematic annotation of TASs in a large number of L. monocytogenes assemblies.

1. Introduction

Toxin–antitoxin systems (TASs) were discovered because of their involvement in a biological process called post-segregational killing (PSK), a plasmid maintenance mechanism based on two plasmid-encoded genes: a toxin gene (T) and its antagonistic antitoxin (A) [1–3]. In this context, the toxin and antitoxin are equally distributed in the two daughter cells; however, the instability of the antitoxin will lead to an active toxin killing the cell lacking the plasmid because the antitoxin cannot be replaced. This system also could be important under some stress (e.g., antibiotic in the medium), where the toxin is released from its less stable antitoxin partner, leading to a transient metabolic shutdown and growth arrest or cell death, similar to apoptosis in higher organisms. Of particular interest, TASs have been associated with pathogenic bacterial intracellular infection and with quorum sensing [4]. In addition, pandemic bacterial strains have been shown to carry more TASs compared to non-epidemic related species [5].

TAS toxicity relies on various molecular mechanisms targeting diverse biological processes, e.g., cell membrane integrity, assembly of the translational machinery, tRNA and mRNA stability,

(2)

Toxins 2020, 12, 29 2 of 14

translation initiation and elongation steps, DNA replication, and ATP synthesis (see [6,7] for reviews). For example, the toxins MazF or PemK target and cut both free mRNA and mRNA bound to the translational machinery, while the toxin ParE targets DNA replication by inhibiting the DNA gyrase [6]. In terms of gene structure, TASs are usually found in operons with either AT or TA orientations. Those TAS operons can be acquired from mobile genetic elements such as plasmids or phages, and are also present in bacterial chromosomes [8], leading to complex heterogeneous genomic landscapes of TASs with both horizontal and vertical transmissions.

Listeria monocytogenes (Lm) is a Gram-positive bacterium and opportunistic food-borne pathogen [9–11]. It is the etiological agent of listeriosis in humans and animals causing abortion, septicemia, gastroenteritis, and central nervous system (CNS) infections [12,13]. L. monocytogenes strains are grouped into four distinct phylogenetic lineages called I, II, III, and IV [14–16]. Strains belonging to lineages I and II are the most abundant isolates worldwide and particularly Lineage I is frequent in disease [16,17], while Lineage III and IV strains are very rare and predominantly isolated from (asymptomatic) animals [18]. In silico predictions using TADB2 reveal only two TASs in L. monocytogenes EGD-e: lmo0113-0114 and lmo0887-0888 [19]. There are a few studies investigating TAS systems in L. monocytogenes [20–22] focusing only on a few L. monocytogenes strains and a few TAS pairs. The following TAS pairs were identified mainly using in silico methods in the strain ATCC19117 (with their corresponding gene names in L. monocytogenes EGD-e): lmo0113-0114, lmo0887-0888, and lmo1301-1302, predicting their 3D structure and potential inhibitory peptides. This report also showed by qPCR that lmo0113 is upregulated upon heat stress [21,22]. Another report investigated the TAS lmo0887-888 of L. monocytogenes EGD-e in more detail, ruling out the classical MazF action, but without demonstrating the exact role of this TAS in L. monocytogenes [20]. They showed that lmo0887-888 does not affect the level of persister formation upon antibiotic treatment, but the expression of σB-dependent genes opuCA and lmo0880 under sub-inhibitory norfloxacin treatment [20].

In this article, we first identified putative core gene TASs in a larger set of 579 L. monocytogenes genomes and discovered how some of these gene pairs have potentially co-evolved. In a second step using putative accessory gene TASs, we correlated our phenotypic metadata to the presence/absence of TAS genes, revealing the potential impact of specific TASs in the pathogenicity of L. monocytogenes strains.

2. Results

2.1. Identification of Core Gene TASs

The TASmania database [23] was queried to identify the putative TASs in the 352 Lm genomes that are currently in the database (list of L. monocytogenes genomes in TASmania, Table S1, list of TASs, Table S2). These 352 genomes were typed using the scheme cgMLST1748 at the Pasteur BIGSdb-Lm web site [24] to obtain the core gene alleles for all strains. By comparing with the TASmania hits we identified n= 14 core gene TASs (cgTASs) (Table1) and their respective alleles (list of alleles in the 14 cgTASs, Table S3). The current knowledge on TASs in L. monocytogenes is rather scarce and our list included the two cgTASs already identified by TADB2 (lmo0113-0114 and lmo0887-0888). TASmania extended the number of TAS candidates by one order of magnitude in all L. monocytogenes strains, but only a subset of them consists of core genes according to cgMLST1748.

Out of these 14 cgTASs, two cgTASs appear as orphans (lmo0168 and lmo0887). Their genetic environment was studied in the annotations of the EGD-e strain [25]. In the case of lmo0168, no possible partner was found, as the gene is surrounded by genes located on the other strand, confirming a probable orphan antitoxin. The potential partner of lmo0887 is lmo0888, which likely is a toxin mRNA interferase containing a pemK-like domain and a plasmid_toxin domain (according to TASmania), is a mazEF (according to TADB2); however, it is not identified as a core gene according to the cgMLST1748 scheme and thus must be ignored in our co-evolution analysis below.

(3)

Table 1. List of 14 putative cgTASs identified in TASmania L.monocytogenes. Gene locus names are taken from the reference EGD-e annotation (NCBI accession number NC_003210.1). The term “Guilt-by-association” refers to potential T or A proteins identified only by their neighborhood to a hit

in TASmania [23]. The alternative shading is used to group the pseudo-operon TAS loci.

Core Gene Locus Putative Type HMM Cluster Hit Description TAS Pair ID and Orientation lm0113 Toxin Guilt-by-association Peptidase_M78 domain-containing protein 1 AT lm0114 Antitoxin TASMANIA.A78 Toxin–antitoxin system, antitoxin component, Xre family 1 AT

lmo0168 Antitoxin TASMANIA.A8 Orphan antitoxin

mazE 2 A orphan

lmo0887 Antitoxin TASMANIA.A5

CopG family ribbon-helix-helix

protein

3 AT

Chromosome partitioning protein

ParB

4 TA

lmo1310 Toxin Guilt-by-association

DUF3440 domain-containing

protein

4 TA

lmo1466 Antitoxin Guilt-by-association

Cyclic-di-AMP phosphodiesterase

PgpH

5 TA

lmo1467 Toxin TASMANIA.T1 PhoH family

protein 5 TA

lmo2041 Toxin Guilt-by-association

Ribosomal RNA small subunit methyltransferase

H

6 AT

lmo2042 Antitoxin TASMANIA.A2 Transcriptional

regulator MraZ 6 AT

ParB/RepB/Spo0J family partition

protein

7 TA

lmo2791 Toxin Guilt-by-association Partition protein,

ParA homolog 7 TA

lmo2793 Toxin Guilt-by-association Uncharacterized

protein 8 AT

lmo2794 Antitoxin TASMANIA.A1 Nucleoid occlusion

protein 8 AT

In a second step, we added our own collection of L. monocytogenes strains [26] (n= 227) with their cgMLST alleles (Table S4) and metadata annotation (Table S5). We extracted their non-redundant gene alleles corresponding to the previously identified 14 cgTASs (Table S6). By clustering the non-redundant cgTAS alleles patterns with nominal hierarchical clustering, we obtained the dendrogram (Figure1) showing that the 14 cgTASs are sufficient to separate the L. monocytogenes lineages (I, II, III, and IV). These results are in agreement with previous MLST analysis using seven housekeeping genes [14]. This clustering allows for characterizing the lineage membership of strains that did not have metadata

(4)

Toxins 2020, 12, 29 4 of 14

information available (e.g., Lm_gca_000729665 as Lineage I and Lm_gca_001466115 as Lineage II). However, one cannot infer any causal role for the cgTASs in the lineage separation, as clustering the alleles of some subsets of core genes among the 1748 core genes would potentially show the same

lineage split.Toxins 2019, 11, x FOR PEER REVIEW 4 of 15

Figure 1. Dendrogram of the non-redundant L. monocytogenes strains based on the cgTAS alleles. The

lineage coloration is based on the known lineages thanks to the metadata. Those without metadata are tentatively colored according to the branch. Black = Lineage I; red = Lineage II; blue = Lineage III; green = Lineage IV; light blue = Listeria innocua as outgroup (wrongly annotated L. monocytogenes in ENA).

2.2. Co-Evolution Analysis

We analyzed the six complete cgTAS pairs for co-evolutionary residues and excluded the orphan cgTASs (lmo0168 and lmo0887). We grouped the sequences by gene, translated them to proteins and fused the toxin (T) alleles with the corresponding antitoxin (A) alleles per genome in a multiple sequence alignment (MSA) using our own Perl script and MAFFT [27]. With this MSA, the BIS2Analyzer server [28] was able to identify potential co-evolutionary residues, i.e. a mutated amino acid in a toxin associated to another mutated residue in the cognate antitoxin of the same TAS (e.g., red arrows in Figure 2 and Table 2). Only two cgTAS pairs (lmo1466-1467 and lmo1309-1310) revealed co-evolving residues between the two partners. For instance, the residue at location 435 (within the first partner lmo1466) co-evolves with the residue at location 828 (within the second partner lmo1467) (Figure 2a and Table 2). More than one residue per partner can be co-evolving as shown with the pair lmo1309-1310 (Figure 2b and Table 2). One cgTAS pair (lmo2793-2794) had co-evolving residues visible on the MSA using Jalview (Figure 2c red arrows), but it failed to reach a significant p-value (<0.05) in BIS2Analyzer. Lm 07p f077 6 I L_m 0₈ 5₅ 78 II Lm 1 040 3s II Lm 442 3 _II Lm lm 18 80 II Lm atcc 19117 I Lm c_fs an 00 23 49 Lm egd e II Lm f690_{0 II} L_m fin la_n d 1₉ 9₈ II Lm fs l f6 68 4 Lm fs l j1 19 4 I Lm fsl j1 208 IV Lm fsl j2 071 III Lm fs l n 1 0₁₇ I Lm fsl n3 16 5 I I Lm gc a 0004 38625 JF49 31 II Lm gca 00 0438 665 JF55_{94 I} L m s e roty pe 4b s tr f 2 Lm gca 000 725 94 5 Lm lm 1886 I L m s_h l0 0₇ I Lm g_ca 0 00 72 61 85 Lm gca 000726425 Lm gc a 000 7264 75 Lm gca 000₇₂₆ 685 Lm gca 00072₇₂₈₅ Lm gca 000 7273 85 Lm gc a 000 7281 45 Lm gca₀₀₀ 728₉₃₅ Lm gca 000729665 Lm gc a 0₀₀₇ 540₂₅ Lm gca 00 077 25 95 JF5₈₃₂ II Lm gc a 000 8145 55 N11 .128 5 II JF5592 I JF 57 61 II N 12.253 2 II L m g ca 0₀ 0 9 7 4 3 2 5 JF 58₂₉ I Lm gca 001005 925 Lm gca 001012875 Lm gc a 001 0270 65 N 13.071 4 II Lm g_ca 0 01 04 77 15 JF5 171 II JF51 72 II JF 5336 II Lm gca 001399895 L m g ca 0 01 4 32 3 65 LM N C 35_{1 II} Lm gca₀₀₁₄₆ 6115 Lm lm 18 40 II Lm gc a 001466 215 Lm gca 001 466235 Lm gca 001751825 Lm gca 001752305 Lm hcc23 III Lm lm 1 82 3 II Lm q oc 1 II Lm qo c2 II L_m r4₇ 9_a II Lm se roty pe 4 b s_{tr h} 785 8 I e ro ty_p e 4 bv s tr _ls 5₄ 2 I Lm se_ro typ e 4_bv str_ls 64₂ Lm s er oty pe 4 bv_s tr _ls 64 3 I Lm sh_l0 08 I L_m s h l0 0 9 II Lm shl0 11 II Lm sh l01 3 II Lm slc c237 2 II Lm slc c237 8 I Lm s_lc c2 54 0 I N 12. 0588 II N 1₁ .2₁ 3₄ II N 12.1608 I LM NC 331 I L_M N C 2₈ 4 I JF57_{88 II} JF 4976 II JF₅ 3₃ 7 II LM N_C 08 3 II LM N C 318 III LM N C 244 II JF55 88 I LMN C 306 I JF49 36 I JF55 19 I LM NC 542 I LM N C 283 II JF57 94 II JF4975 I L_M N C 1₄ 2 I JF₅ 1₈ 6 I LMN C 121 II LM NC27 3 II JF₅ 8₂ 3 II L M N C 3 20 I LM NC 37 8 I LM NC 06 2 I LM N C4 2 7 I LM N C 11₉ I LM N C 06 0 I LM N_C 27 7 I LM N C415 II JF 4914 II LM N C431 II JF49 74 I JF49 87 I JF 4930 II LM N_{C 27} 8 I L M N C 41 2 I I JF55 77 II JF 55 76 I I JF5 0 52 II LM N C 359 II JF49 20 II LMN C 160 I LM N C 287 II N 12 .110 7 II LM NC 55 1 I N 1₃ .1 0₇ 9 I LM N C 5₄ 9 I N 12.092 2 II N 12.1 387 I N 13.0581 I N 13. 14 07 II N 11 .1 25 5 II N 11.1_{415 II} N₁₁ .25 53_II N 1 2.0₆₆ 7 II

Figure 1. Dendrogram of the non-redundant L. monocytogenes strains based on the cgTAS alleles. The lineage coloration is based on the known lineages thanks to the metadata. Those without metadata are tentatively colored according to the branch. Black= Lineage I; red = Lineage II; blue = Lineage III; green= Lineage IV; light blue = Listeria innocua as outgroup (wrongly annotated L. monocytogenes in ENA).

2.2. Co-Evolution Analysis

We analyzed the six complete cgTAS pairs for co-evolutionary residues and excluded the orphan cgTASs (lmo0168 and lmo0887). We grouped the sequences by gene, translated them to proteins and fused the toxin (T) alleles with the corresponding antitoxin (A) alleles per genome in a multiple sequence alignment (MSA) using our own Perl script and MAFFT [27]. With this MSA, the BIS2Analyzer server [28] was able to identify potential co-evolutionary residues, i.e. a mutated amino acid in a toxin associated to another mutated residue in the cognate antitoxin of the same TAS (e.g., red arrows in Figure2and Table2). Only two cgTAS pairs (lmo1466-1467 and lmo1309-1310) revealed co-evolving residues between the two partners. For instance, the residue at location 435 (within the first partner lmo1466) co-evolves with the residue at location 828 (within the second partner lmo1467) (Figure2a and Table2). More than one residue per partner can be co-evolving as shown with the pair lmo1309-1310 (Figure2b and Table2). One cgTAS pair (lmo2793-2794) had co-evolving residues visible on the MSA using Jalview (Figure2c red arrows), but it failed to reach a significant p-value (<0.05) in BIS2Analyzer.

(5)

Toxins 2019, 11, x FOR PEER REVIEW 5 of 15

(a) Lmo1466-1467

(b) Lmo1309-1310

(c) Lmo2793-2794

Figure 2. The MSA of the fusion pseudo-protein (first partner in orange, second partner in green) with

the co-evolving amino acid residues highlighted (inter-protein = red arrows; intra-protein = blue arrows) within Jalview [29]. In that view, all common residues letters are hidden, only varying Figure 2. The MSA of the fusion pseudo-protein (first partner in orange, second partner in green) with the co-evolving amino acid residues highlighted (inter-protein= red arrows; intra-protein = blue arrows) within Jalview [29]. In that view, all common residues letters are hidden, only varying residues are displayed with their letters. The numbering at the top corresponds to the amino acid positions along the artificially fused protein sequence (same numbers and colors as Table2). For clarity, only the red arrows (inter-protein) are labeled with residues highlighted in Table2.

(6)

Toxins 2020, 12, 29 6 of 14

Table 2. List of residues that are inter-protein co-evolving for each pair of cgTASs according to BIS2Analyzer. Positions numbered for the fusion pseudo-protein; orange= residues in the first partner; green= residues in the second partner. See Materials and Methods for more details. (Intra-protein co-evolving residues are shown in TableA1). Add fusion protein info.

cgTAS p-Value Positions from BIS2Analyzer Jalview Confirmed

Lmo1466 - 1467 1.126763e-14

residues are displayed with their letters. The numbering at the top corresponds to the amino acid positions along the artificially fused protein sequence (same numbers and colors as Table 2). For clarity, only the red arrows (inter-protein) are labeled with residues highlighted in Table 2.

Table 2. List of residues that are inter-protein co-evolving for each pair of cgTASs according to

BIS2Analyzer. Positions numbered for the fusion pseudo-protein; orange = residues in the first partner; green = residues in the second partner. See Materials and Methods for more details. (Intra-protein co-evolving residues are shown in Table A1). Add fusion (Intra-protein info.

cgTAS p-Value Positions from BIS2Analyzer Jalview Confirmed

Lmo1466-1467 1.126763e-14 Positions: 435 828 28 sequences: I T 22 sequences: V M Yes inter-protein Lmo1309-1310 3.039477e-06 Positions: 31 56 85 360 387 35 sequences: T D F T V 3 sequences: L G Y K I 1 sequence: P N M N T Yes inter-protein 1.215791e-05 Positions: 166 221 248 284 422 35 sequences: T R T N A 4 sequences: A Q S D S Yes inter-protein Lmo2793-2794 Not significant Positions: 68 302 25 sequences: V L 9 sequences: I I

Yes inter-protein but not significant with

BIS2Analyzer.

2.3. Accessory Gene TAS Analysis

By subtracting the cgTAS hits from the L. monocytogenes strains in TASmania, all other TASs are classified as accessory TASs (acTASs) (Table S7, Figure A1). In order to obtain a better understanding of the role of accessory TASs, we performed a gene analysis with a set of 227 L. monocytogenes strains [26] for which we have the corresponding metadata (Table S5). These isolates are not part of the TASmania database, so we processed them by first predicting protein genes with Prodigal [30] and annotating them with the TASmania HMMs [23]. Finally, we built a heatmap of those acTASs (Table S8) for antitoxins (Figure 3a) and toxins (Figure 3b) after removing the core genes and the redundancy as described above.

Yes inter-protein

Lmo1309 - 1310 3.039477e-06

BIS2Analyzer.

Yes inter-protein

1.215791e-05

BIS2Analyzer.

Yes inter-protein

Lmo2793 - 2794 Not significant

BIS2Analyzer.

2.3. Accessory Gene TAS Analysis

By subtracting the cgTAS hits from the L. monocytogenes strains in TASmania, all other TASs are classified as accessory TASs (acTASs) (Table S7, FigureA1). In order to obtain a better understanding of the role of accessory TASs, we performed a gene analysis with a set of 227 L. monocytogenes strains [26] for which we have the corresponding metadata (Table S5). These isolates are not part of the TASmania database, so we processed them by first predicting protein genes with Prodigal [30] and annotating them with the TASmania HMMs [23]. Finally, we built a heatmap of those acTASs (Table S8) for antitoxins (Figure3a) and toxins (Figure3b) after removing the core genes and the redundancy as described above.

Figure3a shows on the rightmost column that the antitoxin cluster A32 (HTH_3) is the most abundant and is more frequently found in animal cases of Lineage II. In TASmania, this A32 antitoxin cluster is observed as being paired with at least six different toxin clusters (T2, T13, T14, T38, T48, and T85, whose nearest Pfam identifiers are HipA_C, Zeta_toxin, HipA_C, RelE, Gp49 and Gp49, respectively). All of these pairs were analyzed for their co-occurrence in the L. monocytogenes strains above, but no candidate pair seems to exist in Lineage I, while only A32.T2 (nearest Pfam HTH_3.HipA_C) can be found rarely in Lineage II (TableA2).

In Figure3b, a small, interesting group of isolates carry one or two genes having a hit to the T138 (nearest Pfam AbiEii) toxin cluster. It is mainly found in Lineage I isolates causing rhombencephalitis, 12 out of 15 (9 in cattle, 2 in goat, and 1 in sheep). The three remaining isolates are not rhombencephalitis, but could be related by their proximity to animals: 1 sheep abortion and 2 environmental isolates. In addition, one case carrying T138 in Lineage II, a cattle rhombencephalitis, is reported. This renders this toxin quite interesting regarding rhombencephalitis. However, in the L. monocytogenes strains of TASmania, no antitoxin partner is known for this toxin that seems to be often encoded in a prophage region.

(7)

Toxins 2019, 11, x FOR PEER REVIEW 7 of 15 (a) Antitoxins TA SM AN IA.A1 11 TA SM AN IA.A9 3 TA SM AN IA.A8 3 TA SM AN IA.A7 8 TA SM AN IA.A1 24 TA SM AN IA.A1 5 TA SM AN IA.A6 6 TA SM AN IA.A2 TA SM AN IA.A3 6 TA SM AN IA.A3 1 TA SM AN IA.A8 8 TA SM AN IA.A1 15 TA SM AN IA.A1 2 TA SM AN IA.A2 7 TA SM AN IA.A3 7 TA SM AN IA.A2 4 TA SM AN IA.A2 9 TA SM AN IA.A5 7 TA SM AN IA.A7 3 TA SM AN IA.A5 5 TA SM AN IA.A1 0 TA SM AN IA.A7 7 TA SM AN IA.A7 6 TA SM AN IA.A8 1 TA SM AN IA.A3 0 TA SM AN IA.A7 2 TA SM AN IA.A1 01 TA SM AN IA.A8 TA SM AN IA.A8 7 TA SM AN IA.A4 7 TA SM AN IA.A5 0 TA SM AN IA.A2 3 TA SM AN IA.A9 TA SM AN IA.A8 2 TA SM AN IA.A4 2 TA SM AN IA.A1 19 TA SM AN IA.A2 2 TA SM AN IA.A5 4 TA SM AN IA.A7 TA SM AN IA.A4 0 TA SM AN IA.A1 13 TA SM AN IA.A8 5 TA SM AN IA.A1 TA SM AN IA.A3 4 TA SM AN IA.A3 2 LM_LMNC159 LM_LMNC332 LM_LMNC288 LM_LMNC287 LM_JF5809 LM_JF5828 LM_LMNC379 LM_LMNC359 LM_LMNC410 LM_JF5577 LM_JF5171 LM_LMNC325 LM_JF5794 LM_JF4941 LM_LMNC307 LM_LMNC309 LM_LMNC253 LM_LMNC247 LM_LMNC347 LM_JF5164 LM_LMNC351 LM_JF4976 LM_LMNC065 LM_LMNC368 LM_JF5593 LM_LMNC317 LM_JF5788 LM_LMNC354 LM_LMNC422 LM_LMNC374 LM_LMNC297 LM_JF5576 LM_LMNC356 LM_LMNC255 LM_JF5761 LM_LMNC266 LM_JF4991 LM_JF5541 LM_JF5172 LM_LMNC148 LM_JF5337 LM_LMNC327 LM_LMNC083 LM_LMNC322 LM_JF4939 LM_JF5199 LM_JF4906 LM_JF5208 LM_LMNC079 LM_JF5822 LM_LMNC278 LM_JF5594 LM_LMNC047 LM_LMNC424 LM_LMNC062 LM_JF4902 LM_LMNC409 LM_LMNC119 LM_LMNC060 LM_JF5201 LM_LMNC426 LM_JF5829 LM_LMNC277 LM_LMNC329 LM_JF4929 LM_JF4987 LM_JF4908 LM_JF4934 LM_LMNC320 LM_LMNC064 LM_LMNC542 LM_JF5186 LM_LMNC155 LM_LMNC358 LM_LMNC284 LM_JF4909 LM_LMNC420 LM_LMNC445 LM_LMNC411 LM_JF5207 LM_LMNC146 LM_JF4975 LM_LMNC318 LM_LMNC112 LM_LMNC108 LM_JF5592 LM_JF4913 LM_JF4935 LM_LMNC283 LM_JF5832 LM_JF4839 LM_LMNC269 LM_LMNC273 LM_JF4931 LM_LMNC314 LM_JF5336 LM_JF4982 LM_LMNC078 LM_LMNC150 LM_LMNC323 LM_LMNC080 LM_JF5825 LM_LMNC431 LM_LMNC121 LM_JF5830 LM_LMNC156 LM_LMNC154 LM_JF4981 LM_LMNC301 LM_JF5827 LM_JF5831 LM_LMNC415 LM_JF5826 LM_LMNC476 LM_LMNC412 is o la tio n _yea r s our c e _ ty pe so u rce_ d e sc rip tio n ph y loge ne tic _ lin e ag e ph ylog enetic_lineag e I II III so urce_descriptio n Cattle rhombencephalitis Goat rhombencephalitis Sheep rhombencephalitis CNS infection Cattle abortion Goat abortion Sheep abortion Cattle mastitis Sheep enteritis Cattle environment Goat environment Sheep environment Environment Food so urce_type Animal Food Human Natural environment isolation _y ear 2015 2000 0 2 4 6 8

(b) Toxins

Figure 3. Heatmaps of the counts of accessory genes for (a) Antitoxins and (b) Toxins in 227 L.

monocytogenes isolates. Metadata is given on the left side of the heatmap: lineage classification, source description, source type, and isolation year.

Figure 3a shows on the rightmost column that the antitoxin cluster A32 (HTH_3) is the most abundant and is more frequently found in animal cases of Lineage II. In TASmania, this A32 antitoxin cluster is observed as being paired with at least six different toxin clusters (T2, T13, T14, T38, T48, and T85, whose nearest Pfam identifiers are HipA_C, Zeta_toxin, HipA_C, RelE, Gp49 and Gp49, respectively). All of these pairs were analyzed for their co-occurrence in the L. monocytogenes strains above, but no candidate pair seems to exist in Lineage I, while only A32.T2 (nearest Pfam HTH_3.HipA_C) can be found rarely in Lineage II (Table A2).

In Figure 3b, a small, interesting group of isolates carry one or two genes having a hit to the T138 (nearest Pfam AbiEii) toxin cluster. It is mainly found in Lineage I isolates causing rhombencephalitis, 12 out of 15 (9 in cattle, 2 in goat, and 1 in sheep). The three remaining isolates are not rhombencephalitis, but could be related by their proximity to animals: 1 sheep abortion and 2 environmental isolates. In addition, one case carrying T138 in Lineage II, a cattle rhombencephalitis, is reported. This renders this toxin quite interesting regarding rhombencephalitis. However, in the L. monocytogenes strains of TASmania, no antitoxin partner is known for this toxin that seems to be often encoded in a prophage region.

For the 352 L. monocytogenes strains described in TASmania unfortunately, no metadata is available, but the list of acTASs and their abundance is similar to our annotated dataset, with A32, A1, A2, and A34 (nearest Pfam HTH_3, ParBc, MraZ, and Omega_Repress/Rep_trans, respectively) being the most prevalent antitoxin clusters, and T9, T1, T57, T7, and T44 (nearest Pfam ParE, PhoH, PemK_toxin, Gp49, and HicA_toxin, respectively) being the main toxin clusters, as shown in Figure A1.

Using TASmania, a closer look at the Pfam annotation of the toxin clusters highlights the diversity of the TA systems uncovered in L.monocytogenes. Indeed, ParE targets the DNA gyrase and,

TA SM AN IA.T 4 9 TA SM AN IA.T 5 5 TA SM AN IA.T 7 0 TA SM AN IA.T 2 9 TA SM AN IA.T 1 1 TA SM AN IA.T 1 7 TA SM AN IA.T 4 4 TA SM AN IA.T 1 3 TA SM AN IA.T 1 TA SM AN IA.T 7 TA SM AN IA.T 8 5 TA SM AN IA.T 2 6 TA SM AN IA.T 6 7 TA SM AN IA.T 4 8 TA SM AN IA.T 2 0 TA SM AN IA.T 1 27 TA SM AN IA.T 1 6 TA SM AN IA.T 1 0 TA SM AN IA.T 3 9 TA SM AN IA.T 1 18 TA SM AN IA.T 3 3 TA SM AN IA.T 1 4 TA SM AN IA.T 1 05 TA SM AN IA.T 3 8 TA SM AN IA.T 1 20 TA SM AN IA.T 9 9 TA SM AN IA.T 2 5 TA SM AN IA.T 3 1 TA SM AN IA.T 6 8 TA SM AN IA.T 6 TA SM AN IA.T 1 8 TA SM AN IA.T 7 1 TA SM AN IA.T 2 7 TA SM AN IA.T 5 0 TA SM AN IA.T 2 TA SM AN IA.T 3 6 TA SM AN IA.T 6 1 TA SM AN IA.T 1 43 TA SM AN IA.T 1 38 TA SM AN IA.T 5 TA SM AN IA.T 1 14 TA SM AN IA.T 7 4 TA SM AN IA.T 9 TA SM AN IA.T 5 7 LM_LMNC250 LM_JF5831 LM_JF5576 LM_LMNC051 LM_JF4935 LM_JF4930 LM_LMNC344 LM_JF4976 LM_LMNC325 LM_LMNC087 LM_JF5827 LM_LMNC356 LM_LMNC379 LM_LMNC080 LM_JF5832 LM_LMNC266 LM_LMNC410 LM_JF5541 LM_LMNC374 LM_JF4991 LM_LMNC273 LM_LMNC159 LM_LMNC283 LM_LMNC307 LM_LMNC156 LM_JF5823 LM_JF5336 LM_LMNC412 LM_JF5577 LM_LMNC415 LM_LMNC148 LM_LMNC319 LM_LMNC078 LM_LMNC150 LM_JF4982 LM_LMNC359 LM_LMNC332 LM_LMNC288 LM_JF4920 LM_JF5171 LM_JF5809 LM_JF5828 LM_JF5794 LM_JF4839 LM_JF5761 LM_LMNC121 LM_JF5337 LM_LMNC297 LM_LMNC476 LM_JF5164 LM_JF4941 LM_JF5825 LM_LMNC331 LM_LMNC306 LM_JF5861 LM_LMNC284 LM_LMNC358 LM_JF5594 LM_LMNC420 LM_JF5829 LM_LMNC112 LM_LMNC060 LM_LMNC109 LM_JF4975 LM_LMNC542 LM_JF4909 LM_JF5822 LM_JF5592 LM_LMNC160 LM_LMNC414 LM_LMNC448 LM_LMNC427 LM_JF4990 LM_JF5833 LM_JF4987 LM_LMNC277 LM_JF4902 LM_LMNC409 LM_LMNC146 LM_JF5201 LM_LMNC062 LM_JF5181 LM_LMNC357 LM_JF5191 LM_JF4927 LM_LMNC047 LM_LMNC411 LM_LMNC424 LM_JF5207 LM_LMNC445 LM_LMNC326 is o la tio n _yea r s our c e _ ty pe so u rce_ d e sc rip tio n ph y loge ne tic _ lin e a g e phylog enetic_lineag e I II III sou rce_descrip tion

Cattle rhombencephalitis Goat rhombencephalitis Sheep rhombencephalitis CNS infection Cattle abortion Goat abortion Sheep abortion Cattle mastitis Sheep enteritis Cattle environment Goat environment Sheep environment Environment Food sou rce_type Animal Food Human Natural environment isolation_y ear 2015 2000 0 1 2 3

Figure 3. Heatmaps of the counts of accessory genes for (a) Antitoxins and (b) Toxins in 227 L. monocytogenes isolates. Metadata is given on the left side of the heatmap: lineage classification, source description, source type, and isolation year.

(8)

Toxins 2020, 12, 29 8 of 14

For the 352 L. monocytogenes strains described in TASmania unfortunately, no metadata is available, but the list of acTASs and their abundance is similar to our annotated dataset, with A32, A1, A2, and A34 (nearest Pfam HTH_3, ParBc, MraZ, and Omega_Repress/Rep_trans, respectively) being the most prevalent antitoxin clusters, and T9, T1, T57, T7, and T44 (nearest Pfam ParE, PhoH, PemK_toxin, Gp49, and HicA_toxin, respectively) being the main toxin clusters, as shown in FigureA1.

Using TASmania, a closer look at the Pfam annotation of the toxin clusters highlights the diversity of the TA systems uncovered in L. monocytogenes. Indeed, ParE targets the DNA gyrase and, along with Gp49, belongs to the Pfam clan called “plasmid_antitox CL0136”, whose members are originally described as plasmid-encoded TASs involved in plasmid maintenance. The PhOH domain is found in cytoplasmic proteins predicted as ATPase and which are induced by phosphate starvation [31]. PemK toxins belong to the Pfam superfamily of CcdB/PemK (CL0624) known as growth inhibitors that can bind to their own promoter and act also as endonucleases. HicA is an mRNA interferase that binds to target mRNA potentially in a translation-independent manner. The Firmicutes phylum has a prevalence of ParE and PemK like toxins in chromosomal TASs, which correspond to the prevalence in L. monocytogenes (FigureA2). More experimental investigation is required to confirm these putative TASs and to understand the conditions that regulate their expression in L. monocytogenes of various pathogenicity.

3. Discussion

Toxins and antitoxins have many roles in the bacterial cells by targeting a broad range of biological processes. Their role in pathogenic bacteria involves plasmid and pathogenicity island maintenance or biofilm formation [4]. We previously stated [23] that the TASs identified in the TASmania database are putative TASs predicted purely in silico by computational means and would benefit from in vivo validation. This is also the case in this report; all TASs that we identified in L. monocytogenes are candidates to be confirmed experimentally.

We obtained those candidates by combining large-scale database searches leveraging on the “guilt-by-association” neighborhood criteria (“guilt-by-association” refers to potential T or A proteins

identified only by their neighborhood to a hit in TASmania) with the available metadata.

By looking at L. monocytogenes core genes, we identified 14 putative cgTASs that were analyzed for potential co-evolving residues. In addition, clustering their alleles confirmed the L. monocytogenes lineage separation. However, this lineage split is not necessarily due to the presence of these cgTASs. One of the cgTASs (lmo0168) appears as a putative orphan antitoxin (nearest Pfam “MazE_antitoxin”). Whether this putative antitoxin is expressed and functional remains unclear. If expressed, one could speculate about an interaction in trans with the toxin of another TAS.

Since we unraveled the possibility of co-evolving residues, further work is needed to demonstrate these hypothetical protein–protein contacts between those residues. Interestingly, co-evolution was reported for a Type III toxin between amino acid residues of the toxin CptIN and nucleic acids of its cognate antitoxin ncRNA [32]. An intriguing gene is lmo0888, which is recognized as part of a TAS together with lmo0887, but it is not defined as a core gene contrary to its partner lmo0887. It is not clear to us why this happens, as, intuitively, the whole pair should be part of the core genes. One possible explanation could be that this gene is found in multiple copies in part of the L. monocytogenes genomes and thus was removed from the core genes [24].

Additional knowledge or hypotheses on TASs could be extracted, such as TAS association with diseases. For instance, when looking at the phenotypic link to acTASs, the toxin cluster AbiEii T138 is of particular interest given its frequent association with rhombencephalitis in ruminants. Even though not all strains isolated from rhombencephalitis harbored this toxin, the association of toxin T138 with ruminant rhombencephalitis was significant (p-value= 0.0413, X2_{test). Looking at the TASs,} no obvious partner antitoxin is found for this toxin probably because it is part of a prophage region that is usually less well annotated. Further work remains to be done to demonstrate which pathway or

(9)

target is activated. The presence of this prophage should be validated for its suitability as a diagnostic tool in the surveillance of animal farms.

By using large-scale in silico analysis of toxin–antitoxin systems in Listeria monocytogenes, we demonstrated that knowledge could be extracted from combined genome sequences and associated metadata. This includes potential co-evolutionary residues, the detection of putative new toxin or antitoxin partners, as well as the suspected role for a specific prophage TAS in rhombencephalitis in ruminants.

4. Materials and Methods

4.1. Data Used

Sequences were obtained from ENSEMBL.Bacteria [33] (352 L. monocytogenes strains chosen because they are included in our TASmania database) and from our previous work PRJEB15123, PRJEB15195 [34,35], and PRJEB22706 [26] (227 L. monocytogenes strains collected by our group with their associated metadata).

4.2. Identification of Putative TAS Genes using the TASmania Database

TAS candidates were extracted from our TASmania databasehttps://bugfri.unibe.ch/tasmania[23] in all 352 L. monocytogenes strains included in the TASmania database with an R script directly querying the database. This list of candidates (Table S2) allows for the discovery of new TAS pairs.

4.3. Identification of Core Gene TASs

In a first step, typing of all the 352 L. monocytogenes strains was performed at the BIGSdb-Lm

https://bigsdb.pasteur.fr/listeria/database[24] with a full genome sequence query on the cgMLST1748 scheme, in order to identify the alleles of each of the 1748 core genes. In the second step, the candidate TAS genes of TASmania were crossed with the core genes, leading to a list of 14 core gene TASs (cgTASs) (Table1). Core genes (cgMLST1748) are 1748 genes that were identified among all L. monocytogenes comparing thousands of L. monocytogenes genomes [24]. Accessory genes are only found in a subset of L. monocytogenes strains.

Another set, including 227 L. monocytogenes field isolates not belonging to TASmania, was added to the study. For these isolates, we knew the cgMLST profiles from a preceding study (Aguilar-Bultet, manuscript in preparation). The cgTAS genes identified above were identified in these 227 genomes by BLASTn and combined with the alleles of the 352 L. monocytogenes strains. From this, a matrix with allele information of all isolates, but containing only the 14 cgTAS loci previously identified from the combination TASmania-BIGSdb, was obtained.

4.4. Removing Redundancy and Clustering

First, the strains containing more than 50% of alleles with no hits in the 14 core gene set were removed. Second, only one genome was kept as representative for all the strains with 100% identical cgTAS allele patterns (in the 14 cgTASs).

The representative cgTAS patterns (Table S6) were clustered according to the cgMSLT allele identifiers using the Nominal Clustering nomclust R package [36]. The Nominal Clustering performs hierarchical cluster analysis (HCA) with objects characterized by nominal (categorical) variables with a distance measured by the Goodall 1 dissimilarity measure. The dendrogram obtained with hclust was then plotted with plot as a fan with labels colored according to the lineages (Figure1).

4.5. Identification of Accessory Gene TASs

The accessory gene TASs (acTASs) were deduced by subtracting the cgTASs from the list of TASs extracted from TASmania. Redundancy among the 100% identical acTAS pattern was removed by

(10)

Toxins 2020, 12, 29 10 of 14

keeping a single representative genome. The heatmaps were calculated with R package pheatmap, allowing for the addition of metadata (Tables S5 and S8).

4.6. Identification of Potential Co-Evolving Residues

The BIS2Analyzer web site [28] was used to identify potential co-evolving residues. The input MSA for each of the 6 cgTAS pairs was built with an in-house Perl script and computed with MAFFT (File F1). The Dimension parameter was set to 2, allowing up to 2 exceptions on a column, and all other parameters were kept as defaults (Table2). The residue positions and p-values (Fischer test) were evaluated by the CLAG score [37] and the server, respectively.

Supplementary Materials: The following are available online athttp://www.mdpi.com/2072-6651/12/1/29/s1, Table S1: genomes, Table S2: all TAShits, Table S3, 14cg TAS Allele, Table S4: All Sample Allele, Table S5: metadata, Table S6: NonRedundant 14cgTAS TASmania BIGSdb Allele, Table S7: acTAS TASMANIA, Table S8: acTAS_additionalstrains.

Author Contributions:Methodology, software, validation, formal analysis, funding acquisition, writing—review and editing: J.A.A.; resources, writing—review and editing: A.O.; formal analysis, data curation, writing—review and editing: H.A.; formal analysis, data curation, and writing—review and editing: L.A.-B.; formal analysis, supervision, project administration, funding acquisition, and writing—original draft preparation: L.F. All authors have read and agreed to the published version of the manuscript.

Funding:This research was funded by the Swiss National Science Foundation (CRSII3_160703 & CRSII3_147692 http://www.snf.ch), and J.A.A. was funded by the Swiss Excellence Fellowships for Foreign Scholars (ESKAS). Acknowledgments:We thank Sylvain Brisse, Marc Lecuit, and Alexandra Moura for helping us with the strain typing via their cgMLST1748 platform at Pasteur. We are grateful to Joachim Frey for the initial funding of the Listeria monocytogenes project.

Conflicts of Interest:The authors declare that there is no conflict of interest.

Appendix A

NonRedundant 14cgTAS TASmania BIGSdb Allele, Table S7: acTAS TASMANIA, Table S8: acTAS_additionalstrains.

Author Contributions: Methodology, software, validation, formal analysis, funding acquisition, writing— review and editing: J.A.A.; resources, writing—review and editing: A.O.; formal analysis, data curation, writing—review and editing: H.A.; formal analysis, data curation, and writing—review and editing: L.A-B.; formal analysis, supervision, project administration, funding acquisition, and writing—original draft preparation: L.F. All authors have read and agreed to the published version of the manuscript.

Funding: This research was funded by the Swiss National Science Foundation (CRSII3_160703 & CRSII3_147692 http://www.snf.ch), and J.A.A. was funded by the Swiss Excellence Fellowships for Foreign Scholars (ESKAS). Acknowledgments: We thank Sylvain Brisse, Marc Lecuit, and Alexandra Moura for helping us with the strain typing via their cgMLST1748 platform at Pasteur. We are grateful to Joachim Frey for the initial funding of the Listeria monocytogenes project.

Conflicts of Interest: The authors declare that there is no conflict of interest.

Appendix A

(a) antitoxins

A 32.HT H_3 A1 .P a rBc A2 .M ra Z A 34.Rep_tr ans A7 .P a rBc A 11 3. H icB _lk_antito x A 15.B rnA _antito xin A 124.B rnA _antito xin A 78.HT H_3 A 83.P ar B c_2 A 8.MazE _antito xin A5 .N ik R_ C A3 .P a rBc A93.RH H_1 A 111.P hdY eFM_antito x A 85.P arD A 40.HT H_3 A29.RH H_1 A 9.P hdY eFM_antito x A 55.DUF772 A 66.P hdY eFM_antito x A 30.MazE _antito xin A2 2 .Y o kU A 1 2. H icB _lk_antito x A 10.AbiE i_4 A42.RH H_1 A 23.HT H_3 A5 0. A rc A 54.RelB A 49.P arD_lik e A82.RH H_5 A 119.P hdY eFM_antito x A 33 .P rlF_antito xin A53.RH H_3 A 101.Re pr ess or_Mnt A 91.S p oV T_C A 77.HT H_3 A 73.P arD_lik e A 9 5. H icB _lk_antito x A 94.P hdY eFM_antito x A 47.MazE _antito xin A 67.P a rB c A 116.RelB A 27.P hdY eFM_antito x A 37.HicB A 24.P hdY eFM_antito x A6 .P a rBc A62.RH H_1 A 108.MazE _antito xin A 81.P hdY eFM_antito x A 14.S eqA A7 2. A rc A 25.P a rB

c lm_gca_001466295lm_4423lm_gca_000970525lm_gca_000814565lm_n53_1_gca_000626575lm_gca_000814575lm_gca_000814545lm_hpb2262lm_gca_000987705lm_gca_000973005lm_gca_001466135lm_gca_001302385lm_gca_001466035lm_f6900lm_j0161lm_slcc2372lm_gca_001465135lm_gca_000729565lm_gca_000438605lm_6179lm_gca_001466165lm_gca_000727385lm_gca_000800335lm_gca_001257635lm_wslc1042lm_gca_000950775lm_gca_000600015lm_gca_000681515lm_serotype_4b_str_f2365lm_gca_000754025lm_07pf0776lm_gca_000772595lm_gca_000726425lm_slcc2378lm_gca_000438665lm_gca_001463985lm_gca_000960155lm_gca_001447225lm_serotype_4bv_str_ls542lm_gca_001466255lm_gca_000438705lm_serotype_4bv_str_ls642lm_gca_001027085lm_slcc2479lm_cfsan002349lm_fsl_n3_165lm_gca_001432365lm_gca_000972995lm_gca_001027125lm_gca_000960145lm_gca_000726185lm_gca_001466215lm_gca_000808125lm_slcc7179lm_shl013lm_gca_001466155lm_fsl_f6_684lm_finland_1998lm_qoc1lm_wslc1001lm_shl009lm_qoc2lm_egdlm_10403slm_gca_001302395lmlm_shl006lm_gca_000746625lm_fsl_r2_561lm_gca_000978685lm_gca_000814555lm_gca_000951975lm_gca_000729645lm_gca_000725945lm_r479alm_08_5578lm_gca_000712425lm_n53_1lm_08_5923lm_gca_001280245lm_gca_001399895lm_serotype_1_2a_str_f6854lm_gca_001466115lm_gca_001466095lm_gca_001466075lm_gca_001262395lm_shl001lm_gca_001047715lm_gca_000726265lm_gca_001466235lm_gca_000726225lm_gca_000438685lm_fsl_n1_017lm_shl002lm_gca_000726475lm_gca_000727285lm_gca_000728365lm_gca_000726075lm_gca_000726545lm_gca_000729325lm_gca_000726305lm_gca_000726925lm_gca_000728465lm_gca_000727095lm_gca_000729345lm_gca_000729125lm_gca_000726685lm_gca_000727165lm_lm_1880lm_gca_000726765lm_gca_000729075lm_slcc2376lm_gca_000728055lm_gca_001400985lm_gca_001483225lm_lm_1889lm_lm_1840lm_fsl_j1_208lm_lm_1886lm_atcc_19117lm_gca_000728135lm_gca_000727905lm_gca_000729545lm_gca_000729185lm_gca_000764145lm_gca_000727085lm_l99lm_hcc23lm_m7lm_gca_000725925lm_gca_000727395lm_gca_000729135lm_lm_1823lm_fsl_j2_071lm_gca_000438625lm_serotype_7_str_slcc2482lm_lm_1824lm_gca_000729665lm_shl007lm_shl012lm_fsl_j1_194lm_shl008lm_fsl_r2_503lm_gca_000438585lm_gca_001466265lm_gca_000974345lm_gca_001399905lm_slcc2755lm_gca_001468565lm_serotype_4b_str_h7858lm_gca_001027165lm_slcc2540lm_gca_000726155lm_gca_000727825lm_gca_000974325lm_gca_000726105lm_shl010lm_gca_000729045lm_gca_001466025lm_gca_000955695lm_gca_000960445lm_gca_000955685lm_gca_000727755lm_gca_000726905lm_gca_000727465lm_gca_000726065lm_gca_000727065lm_gca_000725955lm_gca_000726385lm_gca_001005925lm_gca_000726235lm_gca_000726145lm_gca_000726865lm_gca_000727605lm_gca_000729685lm_gca_000726045lm_gca_000728045lm_gca_001027205 0 1 2 3 4 5 6 7 8 9 10 Value 02468 C o lo r K ey an d D ensity P lot D ens ity

(11)

Toxins 2020, 12, 29 11 of 14

(b) toxins

Figure A1. Heatmaps of the counts of accessory TAS genes (non-redundant for the strains) for a) Antitoxins and b) Toxins in 322 L. monocytogenes of TASMANIA. The closest Pfam name is given to the TASMANIA cluster.

Figure A2. Comparison of the Firmicutes toxin closest to the Pfam family distribution between plasmid and chromosome. For simplicity, only TASs from the non-ambiguous two-gene AT/TA pseudo-operons were analyzed. The xT/Tx or Ax/xA pairs were ignored. L. monocytogenes strains, as Firmicutes in general, present a high prevalence of PemK and ParE-like chromosomal toxins, while YoeB and HicA toxins are more prevalent in plasmids.

T9.P ar E _to xin T1.P hoH T 57.P emK _to xin T7.Gp49 T44.HicA _to xin T11.HOK _G E F T 13.Zeta_to xin T 70.Zeta_to xin T55.S ymE _to xin T 29.P IN_5 T17.HipA _C T 49.P emK _to xin T74.P IN T 92.Y oeB _to xin T5.Cpta_to xin T114.P IN T85.Gp49 T 36.S poIIS A _to xin T 105.P emK _to xin T 71.P emK _to xin T27.F ic _N T 143.P emK _to xin T26.P IN T61.P arE _to xin T1 3 8.Abi Ei i T2.H ip A_C T98.Y afQ_to xin T4.Y afQ_to xin T50.HicA _to xin T 6.Y oeB _to xin T78.P arE _to xin T31.P IN T25.HigB _to xin T40.P arE _to xin T118.P IN T33.Fic T37.P arE _to xin T 3 9. A biE ii T12.HOK _G E F T14.HipA _C T64.P arE _to xin T 35.S poIIS A _to xin T22.HipA _C T86.P arE _to xin T 77.P emK _to xin T38.RelE T20.P IN T67.HigB −lik e_to xin T152.HigB −lik e_to xin T52.S ymE _to xin T48.Gp49 T8.Ldr_to xin T19.P IN T59.P arE _to xin lm_lm_1840 lm_gca_001400985 lm_lm_1886 lm_gca_001483225 lm_lm_1889 lm_lm_1823 lm_lm_1824 lm_gca_000438685 lm_fsl_j2_071 lm_slcc2376 lm_l99 lm_hcc23 lm_m7 lm_lm_1880 lm_cfsan002349 lm_gca_000726185 lm lm_fsl_n3_165 lm_gca_001466155 lm_gca_001466135 lm_gca_001466165 lm_gca_001012875 lm_gca_001465135 lm_r479a lm_08_5578 lm_gca_000814555 lm_6179 lm_n53_1_gca_000626575 lm_4423 lm_gca_000729565 lm_gca_000725945 lm_gca_000729645 lm_gca_001399895 lm_gca_001466035 lm_f6900 lm_j0161 lm_gca_001027065 lm_gca_000746625 lm_08_5923 lm_gca_000712385 lm_gca_001466215 lm_slcc7179 lm_gca_001466075 lm_gca_001280245 lm_gca_001466115 lm_shl013 lm_gca_000808125 lm_gca_000973005 lm_shl001 lm_gca_001027125 lm_gca_000438605 lm_finland_1998 lm_gca_000951975 lm_gca_000727385 lm_slcc2479 lm_shl006 lm_gca_001432365 lm_gca_000438645 lm_fsl_r2_561 lm_gca_001463995 lm_slcc5850 lm_egd lm_wslc1001 lm_fsl_f6_684 lm_10403s lm_gca_001466015 lm_gca_000974345 lm_gca_000974325 lm_gca_000726925 lm_gca_000726685 lm_gca_000727095 lm_fsl_j1_194 lm_gca_001005925 lm_gca_000726145 lm_gca_000726865 lm_gca_000726045 lm_gca_000725955 lm_serotype_7_str_slcc2482 lm_slcc2755 lm_fsl_r2_503 lm_gca_001466265 lm_gca_000438585 lm_gca_001399905 lm_fsl_n1_017 lm_shl008 lm_shl002 lm_gca_001468565 lm_gca_001466235 lm_gca_000728305 lm_gca_000726235 lm_serotype_4b_str_h7858 lm_gca_000987705 lm_gca_001463975 lm_gca_001463985 lm_gca_001466025 lm_gca_000729135 lm_gca_000726305 lm_gca_000728055 lm_gca_000727285 lm_atcc_19117 lm_gca_000727395 lm_hpb2262 lm_gca_000729185 lm_gca_000727905 lm_gca_000725925 lm_gca_000729665 lm_gca_000726075 lm_gca_000438705 lm_serotype_4bv_str_ls542 lm_serotype_4bv_str_ls642 lm_serotype_4b_str_f2365 lm_wslc1042 lm_serotype_4b_str_ll195 lm_gca_000600015 lm_gca_001447225 lm_shl010 lm_gca_001466295 lm_gca_000438665 lm_slcc2378 lm_gca_000772595 lm_gca_000726065 lm_shl007 lm_gca_000955685 lm_gca_000726475 lm_gca_000726265 lm_slcc2540 lm_gca_001047715 lm_gca_000681515 lm_gca_000438625 lm_07pf0776 0 1 2 3 4 Value 02 46 8 1 0 12 C o lo r K ey an d D ensity P lot Dens ity 0 25 50 75 100 chromosome plasmid % Pfam AbiEii Fic GidB Gp49 HicA HigB ParE PemK PIN SpoIISA YafQ YoeB Zeta other

Figure A1. Heatmaps of the counts of accessory TAS genes (non-redundant for the strains) for (a) Antitoxins and (b) Toxins in 322 L. monocytogenes of TASMANIA. The closest Pfam name is given to the TASMANIA cluster.

(b) toxins

Figure A1. Heatmaps of the counts of accessory TAS genes (non-redundant for the strains) for a) Antitoxins and b) Toxins in 322 L. monocytogenes of TASMANIA. The closest Pfam name is given to the TASMANIA cluster.

Figure A2. Comparison of the Firmicutes toxin closest to the Pfam family distribution between plasmid and chromosome. For simplicity, only TASs from the non-ambiguous two-gene AT/TA pseudo-operons were analyzed. The xT/Tx or Ax/xA pairs were ignored. L. monocytogenes strains, as Firmicutes in general, present a high prevalence of PemK and ParE-like chromosomal toxins, while YoeB and HicA toxins are more prevalent in plasmids.

T9.P ar E _to xin T1.P hoH T 57.P emK _to xin T7.Gp49 T44.HicA _to xin T11.HOK _G E F T 13.Zeta_to xin T 70.Zeta_to xin T55.S ymE _to xin T 29.P IN_5 T17.HipA _C T 49.P emK _to xin T74.P IN T 92.Y oeB _to xin T5.Cpta_to xin T114.P IN T85.Gp49 T 36.S poIIS A _to xin T 105.P emK _to xin T 71.P emK _to xin T27.F ic _N T 143.P emK _to xin T26.P IN T61.P arE _to xin T1 3 8.Abi Ei i T2.H ip A_C T98.Y afQ_to xin T4.Y afQ_to xin T50.HicA _to xin T 6.Y oeB _to xin T78.P arE _to xin T31.P IN T25.HigB _to xin T40.P arE _to xin T118.P IN T33.Fic T37.P arE _to xin T 3 9. A biE ii T12.HOK _G E F T14.HipA _C T64.P arE _to xin T 35.S poIIS A _to xin T22.HipA _C T86.P arE _to xin T 77.P emK _to xin T38.RelE T20.P IN T67.HigB −lik e_to xin T152.HigB −lik e_to xin T52.S ymE _to xin T48.Gp49 T8.Ldr_to xin T19.P IN T59.P arE _to xin lm_lm_1840 lm_gca_001400985 lm_lm_1886 lm_gca_001483225 lm_lm_1889 lm_lm_1823 lm_lm_1824 lm_gca_000438685 lm_fsl_j2_071 lm_slcc2376 lm_l99 lm_hcc23 lm_m7 lm_lm_1880 lm_cfsan002349 lm_gca_000726185 lm lm_fsl_n3_165 lm_gca_001466155 lm_gca_001466135 lm_gca_001466165 lm_gca_001012875 lm_gca_001465135 lm_r479a lm_08_5578 lm_gca_000814555 lm_6179 lm_n53_1_gca_000626575 lm_4423 lm_gca_000729565 lm_gca_000725945 lm_gca_000729645 lm_gca_001399895 lm_gca_001466035 lm_f6900 lm_j0161 lm_gca_001027065 lm_gca_000746625 lm_08_5923 lm_gca_000712385 lm_gca_001466215 lm_slcc7179 lm_gca_001466075 lm_gca_001280245 lm_gca_001466115 lm_shl013 lm_gca_000808125 lm_gca_000973005 lm_shl001 lm_gca_001027125 lm_gca_000438605 lm_finland_1998 lm_gca_000951975 lm_gca_000727385 lm_slcc2479 lm_shl006 lm_gca_001432365 lm_gca_000438645 lm_fsl_r2_561 lm_gca_001463995 lm_slcc5850 lm_egd lm_wslc1001 lm_fsl_f6_684 lm_10403s lm_gca_001466015 lm_gca_000974345 lm_gca_000974325 lm_gca_000726925 lm_gca_000726685 lm_gca_000727095 lm_fsl_j1_194 lm_gca_001005925 lm_gca_000726145 lm_gca_000726865 lm_gca_000726045 lm_gca_000725955 lm_serotype_7_str_slcc2482 lm_slcc2755 lm_fsl_r2_503 lm_gca_001466265 lm_gca_000438585 lm_gca_001399905 lm_fsl_n1_017 lm_shl008 lm_shl002 lm_gca_001468565 lm_gca_001466235 lm_gca_000728305 lm_gca_000726235 lm_serotype_4b_str_h7858 lm_gca_000987705 lm_gca_001463975 lm_gca_001463985 lm_gca_001466025 lm_gca_000729135 lm_gca_000726305 lm_gca_000728055 lm_gca_000727285 lm_atcc_19117 lm_gca_000727395 lm_hpb2262 lm_gca_000729185 lm_gca_000727905 lm_gca_000725925 lm_gca_000729665 lm_gca_000726075 lm_gca_000438705 lm_serotype_4bv_str_ls542 lm_serotype_4bv_str_ls642 lm_serotype_4b_str_f2365 lm_wslc1042 lm_serotype_4b_str_ll195 lm_gca_000600015 lm_gca_001447225 lm_shl010 lm_gca_001466295 lm_gca_000438665 lm_slcc2378 lm_gca_000772595 lm_gca_000726065 lm_shl007 lm_gca_000955685 lm_gca_000726475 lm_gca_000726265 lm_slcc2540 lm_gca_001047715 lm_gca_000681515 lm_gca_000438625 lm_07pf0776 0 1 2 3 4 Value 02 46 8 1 0 12 C o lo r K ey an d D ensity P lot Dens ity 0 25 50 75 100 chromosome plasmid % Pfam AbiEii Fic GidB Gp49 HicA HigB ParE PemK PIN SpoIISA YafQ YoeB Zeta other

Figure A2.Comparison of the Firmicutes toxin closest to the Pfam family distribution between plasmid and chromosome. For simplicity, only TASs from the non-ambiguous two-gene AT/TA pseudo-operons were analyzed. The xT/Tx or Ax/xA pairs were ignored. L. monocytogenes strains, as Firmicutes in general, present a high prevalence of PemK and ParE-like chromosomal toxins, while YoeB and HicA toxins are more prevalent in plasmids.

(12)

Toxins 2020, 12, 29 12 of 14

Table A1.List of residues that are significantly co-evolving intra-protein for each pair of cgTASs.

Lmo1466-1467 4.024153e-16

Table A1. List of residues that are significantly co-evolving intra-protein for each pair of cgTASs.

Lmo1466-1467 4.024153e-16 Positions: 340 412 27 sequences: F M 22 sequences: L V 1 sequence: M A intra-protein Lmo1466 4.719742e-07 Positions: 114 134 186 235 351 424 45 sequences: K S E V A L 5 sequences: N T A I S F intra-protein Lmo1466 4.342162e-06 Positions: 141 315 46 sequences: N T 4 sequences: S S intra-protein Lmo1466 Lmo1309-1310 3.977592e-11 Positions: 186 281 412 24 sequences: V S D 15 sequences: A T E intra-protein Lmo1310

Table A2. List of possible A32 TAS pairs in TASMANIA (all bacteria) vs. Lm strains.

First in

Pair Second in Pair

Pair Present in

Lm Remark

A32 T13 (Zeta

toxin) No

A32 T14 (HipA_C) No

A32 T2 (HipA_C) Yes

Only found rarely in Lineage II, in animals with mastitis (cattle) or rhomboencephalitis

(goat and sheep)

T85 (Gp49) A32 No

T48 (Gp49) A32 No

T38 (relE) A32 No

References

1. Jaffé, A.; Ogura, T.; Hiraga, S. Effects of the ccd function of the F plasmid on bacterial growth. J. Bacteriol.

1985, 163, 841–849.

2. Gerdes, K.; Rasmussen, P.B.; Molin, S. Unique type of plasmid maintenance function: Postsegregational killing of plasmid-free cells. Proc. Natl. Acad. Sci. USA 1986, 83, 3116–3120.

3. Gerdes, K.; Larsen, J.E.; Molin, S. Stable inheritance of plasmid R1 requires two different loci. J. Bacteriol.

1985, 161, 292–298.

4. Lobato-Márquez, D.; Díaz-Orejas, R.; García-del Portillo, F. Toxin-antitoxins and bacterial virulence. FEMS Microbiol. Rev. 2016, 40, 592–609.

intra-protein Lmo1466

4.719742e-07

First in

Pair Present in

Lm Remark

A32 T13 (Zeta

toxin) No

A32 T14 (HipA_C) No

A32 T2 (HipA_C) Yes

(goat and sheep)

T85 (Gp49) A32 No

T48 (Gp49) A32 No

T38 (relE) A32 No

References

1985, 163, 841–849.

1985, 161, 292–298.

4.342162e-06

First in

Pair Present in

Lm Remark

A32 T13 (Zeta

toxin) No

A32 T14 (HipA_C) No

A32 T2 (HipA_C) Yes

(goat and sheep)

T85 (Gp49) A32 No

T48 (Gp49) A32 No

T38 (relE) A32 No

References

1985, 163, 841–849.

1985, 161, 292–298.

Lmo1309-1310 3.977592e-11

First in

Pair Present in

Lm Remark

A32 T13 (Zeta

toxin) No

A32 T14 (HipA_C) No

A32 T2 (HipA_C) Yes

(goat and sheep)

T85 (Gp49) A32 No

T48 (Gp49) A32 No

T38 (relE) A32 No

References

1985, 163, 841–849.

1985, 161, 292–298.

Table A2.List of possible A32 TAS pairs in TASMANIA (all bacteria) vs. Lm strains. First in Pair Second in Pair Pair Present in Lm Remark

A32 T13 (Zeta toxin) No

A32 T14 (HipA_C) No

A32 T2 (HipA_C) Yes

Only found rarely in Lineage II, in animals with mastitis (cattle) or

rhomboencephalitis (goat and sheep)

T85 (Gp49) A32 No

T48 (Gp49) A32 No

T38 (relE) A32 No

References

1. Jaffé, A.; Ogura, T.; Hiraga, S. Effects of the ccd function of the F plasmid on bacterial growth. J. Bacteriol. 1985, 163, 841–849. [PubMed]

2. Gerdes, K.; Rasmussen, P.B.; Molin, S. Unique type of plasmid maintenance function: Postsegregational killing of plasmid-free cells. Proc. Natl. Acad. Sci. USA 1986, 83, 3116–3120. [CrossRef] [PubMed]

3. Gerdes, K.; Larsen, J.E.; Molin, S. Stable inheritance of plasmid R1 requires two different loci. J. Bacteriol. 1985, 161, 292–298. [PubMed]

4. Lobato-Márquez, D.; Díaz-Orejas, R.; García-del Portillo, F. Toxin-antitoxins and bacterial virulence. FEMS Microbiol. Rev. 2016, 40, 592–609. [CrossRef] [PubMed]

5. Georgiades, K.; Raoult, D. Genomes of the Most Dangerous Epidemic Bacteria Have a Virulence Repertoire Characterized by Fewer Genes but More Toxin-Antitoxin Modules. PLoS ONE 2011, 6, e17962. [CrossRef] [PubMed]

6. Harms, A.; Brodersen, D.E.; Mitarai, N.; Gerdes, K. Toxins, Targets, and Triggers: An Overview of Toxin-Antitoxin Biology. Mol. Cell 2018, 70, 768–784. [CrossRef]

7. Yamaguchi, Y.; Park, J.-H.; Inouye, M. Toxin-antitoxin systems in bacteria and archaea. Annu. Rev. Genet. 2011, 45, 61–79. [CrossRef]

(13)

8. Hayes, F.; Van Melderen, L. Toxins-antitoxins: Diversity, evolution and function. Crit. Rev. Biochem. Mol. Biol. 2011, 46, 386–408. [CrossRef]

9. Conly, J.; Johnston, B. Listeria: A persistent food-borne pathogen. Can. J. Infect. Dis Med. Microbiol. 2008, 19, 327–328. [CrossRef]

10. Vázquez-Boland, J.A.; Kuhn, M.; Berche, P.; Chakraborty, T.; Domínguez-Bernal, G.; Goebel, W.; González-Zorn, B.; Wehland, J.; Kreft, J. Listeria pathogenesis and molecular virulence determinants. Clin. Microbiol. Rev. 2001, 14, 584–640. [CrossRef]

11. Farber, J.M.; Peterkin, P.I. Listeria monocytogenes, a food-borne pathogen. Microbiol. Rev. 1991, 55, 476–511. [PubMed]

12. Low, J.C.; Donachie, W. A review of Listeria monocytogenes and listeriosis. Vet. J. 1997, 153, 9–29. [CrossRef] 13. Oevermann, A.; Zurbriggen, A.; Vandevelde, M. Rhombencephalitis Caused by Listeria monocytogenes in Humans and Ruminants: A Zoonosis on the Rise? Interdiscip. Perspect. Infect. Dis. 2010, 2010, 632513. [CrossRef] [PubMed]

14. Ward, T.J.; Ducey, T.F.; Usgaard, T.; Dunn, K.A.; Bielawski, J.P. Multilocus genotyping assays for single nucleotide polymorphism-based subtyping of Listeria monocytogenes isolates. Appl. Environ. Microbiol. 2008, 74, 7629–7642. [CrossRef] [PubMed]

15. Den Bakker, H.C.; Bundrant, B.N.; Fortes, E.D.; Orsi, R.H.; Wiedmann, M. A population genetics-based and phylogenetic approach to understanding the evolution of virulence in the genus Listeria. Appl. Environ. Microbiol. 2010, 76, 6085–6100. [CrossRef]

16. Orsi, R.H.; den Bakker, H.C.; Wiedmann, M. Listeria monocytogenes lineages: Genomics, evolution, ecology, and phenotypic characteristics. Int. J. Med. Microbiol. 2011, 301, 79–96. [CrossRef]

17. Chenal-Francisque, V.; Lopez, J.; Cantinelli, T.; Caro, V.; Tran, C.; Leclercq, A.; Lecuit, M.; Brisse, S. Worldwide distribution of major clones of Listeria monocytogenes. Emerg. Infect. Dis. 2011, 17, 1110–1112. [CrossRef] 18. Kuenne, C.; Billion, A.; Mraheil, M.A.; Strittmatter, A.; Daniel, R.; Goesmann, A.; Barbuddhe, S.; Hain, T.;

Chakraborty, T. Reassessment of the Listeria monocytogenes pan-genome reveals dynamic integration hotspots and mobile genetic elements as major components of the accessory genome. BMC Genomics 2013, 14, 47. [CrossRef]

19. Xie, Y.; Wei, Y.; Shen, Y.; Li, X.; Zhou, H.; Tai, C.; Deng, Z.; Ou, H.-Y. TADB 2.0: An updated database of bacterial type II toxin-antitoxin loci. Nucleic Acids Res. 2018, 46, D749–D753. [CrossRef]

20. Curtis, T.; Takeuchi, I.; Gram, L.; Knudsen, G. The Influence of the Toxin/Antitoxin mazEF on Growth and Survival of Listeria monocytogenes under Stress. Toxins 2017, 9, 31. [CrossRef]

21. Kalani, B.S.; Irajian, G.; Lotfollahi, L.; Abdollahzadeh, E.; Razavi, S. Putative type II toxin-antitoxin systems in Listeria monocytogenes isolated from clinical, food, and animal samples in Iran. Microb. Pathog. 2018, 122, 19–24. [CrossRef] [PubMed]

22. Mohammadzadeh, R.; Shivaee, A.; Ohadi, E.; Kalani, B.S. In Silico Insight into the Dominant Type II Toxin–Antitoxin Systems and Clp Proteases in Listeria monocytogenes and Designation of Derived Peptides as a Novel Approach to Interfere with this System. Int. J. Pept. Res. Ther. 2019, 1–11. [CrossRef]

23. Akarsu, H.; Bordes, P.; Mansour, M.; Bigot, D.-J.; Genevaux, P.; Falquet, L. TASmania: A bacterial Toxin-Antitoxin Systems database. PLoS Comput. Biol. 2019, 15, e1006946. [CrossRef] [PubMed]

24. Moura, A.; Criscuolo, A.; Pouseele, H.; Maury, M.M.; Leclercq, A.; Tarr, C.; Björkman, J.T.; Dallman, T.; Reimer, A.; Enouf, V.; et al. Whole genome-based population biology and epidemiological surveillance of Listeria monocytogenes. Nat. Microbiol. 2016, 2, 16185. [CrossRef] [PubMed]

25. Bierne, H.; Sabet, C.; Personnic, N.; Cossart, P. Internalins: A complex family of leucine-rich repeat-containing proteins in Listeria monocytogenes. Microbes Infect. 2007, 9, 1156–1166. [CrossRef]

26. Aguilar-Bultet, L.; Nicholson, P.; Rychener, L.; Dreyer, M.; Gözel, B.; Origgi, F.C.; Oevermann, A.; Frey, J.; Falquet, L. Genetic Separation of Listeria monocytogenes Causing Central Nervous System Infections in Animals. Front. Cell. Infect. Microbiol. 2018, 8, 20. [CrossRef]

27. Nakamura, T.; Yamada, K.D.; Tomii, K.; Katoh, K. Parallelization of MAFFT for large-scale multiple sequence alignments. Bioinformatics 2018, 34, 2490–2492. [CrossRef]

28. Oteri, F.; Nadalin, F.; Champeimont, R.; Carbone, A. BIS2Analyzer: A server for co-evolution analysis of conserved protein families. Nucleic Acids Res. 2017, 45, W307–W314. [CrossRef]

29. Waterhouse, A.M.; Procter, J.B.; Martin, D.M.A.; Clamp, M.; Barton, G.J. Jalview Version 2—A multiple sequence alignment editor and analysis workbench. Bioinformatics 2009, 25, 1189–1191. [CrossRef]