(C) PLOS One [1]. This unaltered content originally appeared in journals.plosone.org.
Licensed under Creative Commons Attribution (CC BY) license.
url:https://journals.plos.org/plosone/s/licenses-and-copyright

------------



Minimalistic mycoplasmas harbor different functional toxin-antitoxin systems

['Virginia Hill', 'Institute Of Veterinary Bacteriology', 'University Of Bern', 'Bern', 'Graduate School For Biomedical Science', 'Hatice Akarsu', 'Rubén Sánchez Barbarroja', 'Valentina L. Cippà', 'Peter Kuhnert', 'Martin Heller']

Date: 2021-12

Mycoplasmas are minute bacteria controlled by very small genomes ranging from 0.6 to 1.4 Mbp. They encompass several important medical and veterinary pathogens that are often associated with a wide range of chronic diseases. The long persistence of mycoplasma cells in their hosts can exacerbate the spread of antimicrobial resistance observed for many species. However, the nature of the virulence factors driving this phenomenon in mycoplasmas is still unclear. Toxin-antitoxin systems (TA systems) are genetic elements widespread in many bacteria that were historically associated with bacterial persistence. Their presence on mycoplasma genomes has never been carefully assessed, especially for pathogenic species. Here we investigated three candidate TA systems in M. mycoides subsp. capri encoding a (i) novel AAA-ATPase/subtilisin-like serine protease module, (ii) a putative AbiEii/AbiEi pair and (iii) a putative Fic/RelB pair. We sequence analyzed fourteen genomes of M. mycoides subsp. capri and confirmed the presence of at least one TA module in each of them. Interestingly, horizontal gene transfer signatures were also found in several genomic loci containing TA systems for several mycoplasma species. Transcriptomic and proteomic data confirmed differential expression profiles of these TA systems during mycoplasma growth in vitro. While the use of heterologous expression systems based on E. coli and B. subtilis showed clear limitations, the functionality and neutralization capacities of all three candidate TA systems were successfully confirmed using M. capricolum subsp. capricolum as a host. Additionally, M. capricolum subsp. capricolum was used to confirm the presence of functional TA system homologs in mycoplasmas of the Hominis and Pneumoniae phylogenetic groups. Finally, we showed that several of these M. mycoides subsp. capri toxins tested in this study, and particularly the subtilisin-like serine protease, could be used to establish a kill switch in mycoplasmas for industrial applications.

Mycoplasmas belong to a class of cell-wall deficient bacteria characterized by minimal genomes acquired through regressive evolution. Historically, they were thought to lack many of the common bacterial virulence traits including classical exotoxins and toxin-antitoxin (TA) systems. In this work, we confirmed the presence of different functional TA systems in several isolates of the caprine pathogen Mycoplasma mycoides subsp. capri. Our data also indicate that TA systems are widespread in other mycoplasma species of veterinary importance. This work paves the way for the investigation of the biological role of TA systems during mycoplasma chronic infections as they are likely to contribute to the parasitic lifestyle of mycoplasmas, persistence in their hosts as well as the buildup of antimicrobial resistance, as recently observed. The availability of synthetic genomics tools to modify a range of Mycoplasma pathogens and well-established challenge models will foster future research and shed the light on the importance of TA systems in mycoplasmas.

Funding: This study was supported by the National Science Foundation, USA [to JJ (grant number IOS-1110151, www.nsf.gov )] The study was also supported by the Swiss National Science Foundation [to JJ (grant number 310030_201152, www.snf.ch/en )]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

In this study, we investigated further the presence, distribution, and functionality of TA systems in mycoplasmas. First, we selected three candidate TA systems in Mmc GM12, including two TA systems identified in silico in addition to one candidate TA system reported recently in the Syn2.0 cell and described above. The overall distribution of these TA systems in Mmc was first assessed by performing whole genome sequencing on fourteen strains and subsequent mapping of TA system homologues. Next, we performed proteomic and transcriptomic analyses in Mmc GM12 to assess their expression profiles in different conditions, including heat stress. We subsequently used toxicity neutralization assays to assess the functionality of each TA module using different heterologous expression systems, including E. coli, B. subtilis and Mycoplasma capricolum (Mcap). Thereafter, we extended the search for homologous TA systems to all Mollicutes and successfully confirmed and tested the functionality of one system in M. feriruminatoris, M. bovis and M. gallisepticum. Finally, we used the previously identified toxins in a programmed cell death using an inducible expression system in mycoplasmas.

In a recent work, the presence of a candidate TA system in the strain Syn2.0, a minimized mycoplasma cell, controlled by a synthetic genome derived of strain GM12, has been identified using global transposon mutagenesis [ 23 ]. This system consists of a gene pair annotated as an AAA-ATPase (JCVISYN2_132) and a hypothetical protein carrying a subtilisin-like serine protease domain (JCVISYN2_133), considered as the putative antitoxin and toxin, respectively. Mycoplasma genomes carrying deletions of both the gene pair as well as the putative toxin JCVISYN2_133 resulted in viable cells, while the sole deletion of the antitoxin JCVISYN2_132 was shown to be lethal [ 23 ]. So far, functional TA systems have not been proposed in other mycoplasmas, especially not in pathogenic field strains, but in silico searches using the TASmania database identified candidate TA systems in different mycoplasmas based on sequence homology searches [ 24 ].

Toxin-antitoxin (TA) systems are involved in cell growth arrest and bacterial persistence in several bacteria including pathogens of the gastrointestinal, urogenital and respiratory tract such as Clostridia, Enterobacteriaceae and Mycobacteria, respectively [ 15 ]. Additionally, they were often associated with a reduction of the cellular metabolism and/or the build-up of biofilms [ 19 ]. Originally discovered as addiction modules involved in plasmid maintenance [ 20 ], TA systems were since linked to many additional cellular processes such as DNA replication, translation, or cell division among others. Currently, TA systems are grouped into six different types based on the on the molecular mechanism resulting in toxin inhibition [ 21 ]. All types have in common that the toxin consists of a stable protein usually encoded in an operon together with a less stable cognate antitoxin, which counteracts the toxin activity. Type I and type III systems have a noncoding small RNA-type antitoxin that neutralizes the toxin via base pairing with the mRNA of the toxin or via direct binding to the toxin, respectively. Type II systems, the most common TA systems reported so far, have a proteinaceous antitoxin that binds and thereby inactivates the toxin. In type IV systems the antitoxin and toxin compete for the same binding partner instead of interacting with each other, while in type V systems the antitoxin is an RNase that degrades the mRNA of the toxin. Type VI systems have a proteolytic adaptor antitoxin that promotes the degradation of the toxin. Most recently, a type VII has been proposed, which is characterized by an antitoxin that renders the toxin inactive via post-translational modification [ 22 ].

The genus Mycoplasma encompasses important human pathogens such as M. genitalium [ 6 ] and M. pneumoniae [ 7 ] as well as important veterinary pathogens such as M. mycoides subsp. mycoides [ 8 ], M. capricolum subsp. capripneumoniae (Mccp) [ 9 ], M. gallisepticum [ 10 ], M. bovis [ 11 ] and M. hyopneumoniae [ 12 ]. Although few mycoplasmas such as Mccp can cause acute diseases with a relatively short incubation time and high lethality [ 13 ], many mycoplasma infections result in a rather chronic form of disease associated with a long persistence of the causative agent [ 2 , 14 ]. Bacterial persistence is generally associated with a state of growth arrest resulting in non-dividing cells that can survive environmental stresses such as temperature, pH, nutrient starvation [ 15 ]. This phenomenon can also be triggered in response to external factors such as phage infections or antibiotic exposure [ 16 ]. In the absence of efficient vaccines for many mycoplasma infections, antibiotic treatment is the only viable option to treat the latter. The continuous rise of antibiotic resistance observed for several mycoplasma species [ 17 ] could be the result of an increased selection of persister cells as it was recently shown for several natural and laboratory strains of E. coli [ 18 ].

Mycoplasmas are minute bacteria that evolved through regressive evolution from a Gram-positive ancestor with low G+C content related to the genus Clostridia [ 1 ]. This extensive loss of genes affected many metabolic pathways including the biosynthesis of the peptidoglycan cell wall and even their capacity to synthesize essential cellular building blocks and nutrients, which make them strictly dependent on their hosts [ 2 ]. As a consequence, mycoplasma genomes were historically considered as degenerated and streamlined genomes, mainly composed of essential genes. The recent construction of the minimal synthetic mycoplasma cell (Syn3.0) [ 3 ], based on the genome of the caprine pathogen M. mycoides subsp. capri (Mmc) GM12 [ 4 ], contradicted this paradigm. Indeed, the Syn3.0 genome is composed of only 473 genes but 80 of them (17% of its genome) still encode proteins of unknown function [ 5 ], confirming that mycoplasma genomes still carry many genetic features yet to be unraveled.

Results

In silico identification and selection of candidate TA systems TASmania [24], which is a discovery-oriented database, was used to identify candidate TA systems in the genome of Mmc GM12. Our in silico search revealed 38 genes encoding candidate toxins or antitoxins but, despite high confidence scores, many of them were found as orphans (S1 Table). Two of them, namely TA 160/1 and TA 752/3 were identified as putative TA systems. Gene descriptions in the TASmania database referred to the TA 160/1 partners as abortive infection proteins, AbiEii (T 160 ) and AbiEi (A 161 ), which belong to a well-described protein family (Abi) reported to act as type IV toxin-antitoxin systems in e.g. Streptococcus agalactiae [25] (Table 1). The T 752 -encoding gene was denominated as cell filamentation protein (Fic) whereas the one coding for the A 753 protein was described as DNA-damage-inducible protein J presenting similarities with the E. coli RelB family [26]. In addition, we also selected a third candidate TA system, namely TA 132/3 , previously suggested to encode a type II AAA-ATPase/subtilisin-like serine protease TA module in the semi-minimal mycoplasma cell JCVI-syn2.0 [23] to further investigate its functionality. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 1. Candidate toxins and antitoxins of M. mycoides subsp. capri GM12 investigated in this study. The top two candidate TA systems were identified in silico by using the TASmania database. https://doi.org/10.1371/journal.pgen.1009365.t001

Genomic organization of TA systems The close vicinity of the genes encoding each partner of the three TA elements suggested that they can be structured in operons, which is a common feature of many TA systems. The presence of antitoxin-encoding genes upstream of their toxin counterparts, correlated with the absence of intergenic regions in between the two partners, confirmed that hypothesis for both TA 132/3 and TA 160/1 . In contrast, the TA 752/3 system presented a different genomic organization with the gene coding for the toxin upstream of its antitoxin partner, spaced by an intergenic region of 207 bp. An analysis of this intergenic region using BPROM identified one possible promotor, including putative -10 and -35 boxes, indicating that the antitoxin may have its own promotor. To confirm these preliminary findings, we first tested the presence of transcripts covering the genes encoding the three candidate TA systems using Mmc GM12 RNA preparations (S1 Fig). Transcripts of all three TA systems were successfully detected via reverse-transcription PCR performed on complementary DNA (cDNA) using primers targeting the genes encoding the candidate toxins and antitoxins (S1A Fig) but also using primers spanning an overlapping region between the two partners (S1B Fig). Fainter amplifications were observed for the T 752 and TA 752/3 transcripts . Primer specificity was confirmed when used to amplify the corresponding regions on Mmc GM12 genomic DNA (S1C and S1D Fig). To conclude on this, RNA-seq data were successfully generated and deposited in the public domain (NCBI project number PRJNA765891). Based on these data, we used two independent softwares, namely RockHopper and ANNOgesic, to predict operon structures. Both tools had very similar results and predicted 188 operons and 186 operons, respectively. Of the three candidate TA systems described here, only the TA 132/3 pair was confirmed as an operon.

PacBio sequencing, genome assembly, annotation and genome organization of Mmc strains To further investigate the distribution of the candidate TA systems in the subspecies Mmc, we sequenced and analyzed fourteen additional strains showing a wide diversity of origins (Table 2). We obtained between 67,246 and 203,646 PacBio long reads per genome sequenced, with at least 231x coverage (S2 Table). Each of the genomes was assembled into one circularized chromosome and the key features associated to each of them are displayed in the Table 2. The size of the 14 genomes ranged from 1,019,889 to 1,172,610 bp, with a G+C content of 23.7%. All genomes contained 30 tRNAs, two ribosomal RNA operons and 1 tmRNA and encoded between 789 and 915 proteins coding sequences. The functional annotation, based on a BLASTP analysis of the coding sequences against UniProtKB, revealed an average of 122 hypothetical proteins (~15% of the total annotated CDS) for each genome. We also observed a high level of synteny between all the genomes and only the strains 152/93 and PG3 carried a large inversion of more than 100 kbp, which was confirmed by PCR amplifications (S2 Fig). Interestingly, the strain 152/93 contained a plasmid (GenBank accession number CP068011) with a size of 1,875 bp, which was highly similar to the previously described pKMK1 [27] and pMmc-95010 plasmids [28]. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Table 2. M. mycoides subsp. capri strains used in this study. https://doi.org/10.1371/journal.pgen.1009365.t002

Mapping of the three candidate TA systems in the Mmc These newly sequenced Mmc strains, as well as the genomes of the two well-characterized strains GM12 and 95010 available in GenBank, were used to map the locations of all the homologs of the three putative TA systems (Fig 1). We constructed a phylogenetic tree of all 16 Mmc strains using whole genome data and carried out a TBLASTN analysis to identify the TA homologs (Fig 1 and S1 File). At first, the candidate TA modules were found unevenly distributed throughout the genomes but, strikingly, they appeared to cluster into four genomic regions, arbitrarily named “TA regions I to IV” (Fig 1). The phylogenetic tree split in two main branches, one containing the GM12 strain (referred as ‘GM12 group’) and one harboring the PG3 strain (referred as the ´PG3 group’) and the distribution of each of the three candidate TA systems was found extremely divergent between the two groups. For instance, both the TA 132/3 and TA 160/1 were found in multiple copies in the ‘GM12 group’ but, besides the presence of few orphans, both were completely missing in the ‘PG3 group’. Three copies of the TA 132/3 module were identified in the TA regions I, II and IV of the My-325, M-18, My-1 and 95010 while the other strains of the ‘GM12 group’ harbored only 1 or 2 copies. Except for 95010, most of the strains of the ‘GM12 group’ harbored 1 or 2 copies of the TA 160/1 system located in the TA regions I and IV. Lastly, the TA 752/3 pair was found as a single copy in all strains of the ‘GM12 group’, as well as more than half of the strains included in the ‘PG3 group’, always located in the TA region III. Only the strains 152/93, Wi8079 and PG3 did not harbor the full TA 752/3 module but orphans (T 752 or A 753 ) were detected. Finally, the presence of the TA 132/3 and TA 160/1 systems in the phylogenetically closely-related M. mycoides subsp. mycoides strain Gladysdale, used as an outgroup, gave some indications that other mycoplasma species might also harbor TA modules in their genomes. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Cartoon displaying the genomic localization of candidate TA systems of M. mycoides subsp. capri GM12 in different M. mycoides subsp. capri strains and in one M. mycoides subsp. mycoides strain. Tree displaying the phylogenetic relationship of the strains based on whole genome sequence data is presented (left). M. mycoides subsp. mycoides strain Gladysdale was used as an outgroup. The tree was constructed in Bionumerics 8.0 using the comparative genomics tool applying standard parameters. The genomic locations of the different copies of the three candidate TA systems investigated in this study are depicted with vertical lines of different colors as detailed in the figure legend (right). The four main genomic regions harboring candidates TA modules are indicated as TA regions I to IV. https://doi.org/10.1371/journal.pgen.1009365.g001

Detection of proteins encoded by candidate TA systems using a proteomic approach We investigated the expression levels of the candidate toxins and antitoxins in Mmc GM12 using a shot gun proteomic approach. First, a culture of Mmc GM12, grown at 37°C, was harvested at mid-logarithmic phase as served as a reference. Two additional Mmc GM12 cultures were grown until early stationary phase at either 37°C and 41.5°C, the latter mimicking a heat stress similar to the fever episodes encountered in their natural hosts during infections. The expressions of the TA proteins in the three different conditions were analyzed and plotted using the mean distributed normalized spectral abundance factor (dNSAF) (Fig 3). In order to compare between all three conditions, the expression of the house-keeping enzyme DNA gyrase subunit B (GyrB) and the cold shock protein (CspG) was used. When grown at 37°C until mid-log phase, they presented a log 10 (mean dNSAF) value of -3.0 and -2.0, respectively. In comparison, the A 132 , A 753 and A 161 as well as T 133 proteins were detectable in low abundance with log 10 (mean dNSAF) values below -3.7. The other candidate toxins and antitoxins were below the detection limit of the mass spectrometry analysis (Fig 3 and S3 File). During the heat shock treatment, the house keeping control protein GyrB abundance remained constant, while the CspG protein’s abundance strongly decreased, as expected. The latter was also observed for the antitoxins A 161 and A 753 with the A 161 even going below detection limits. The changes in the expression levels of the TA 132/3 proteins was less obvious as the heat shock treatment resulted in a small shift in the toxin-antitoxin expression ratio with the toxin becoming slightly more abundant than its cognate antitoxin. PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 3. Detection of proteomic signatures of three TA systems investigated in in vitro grown Mmc GM12. The histogram shows relative protein abundance in strain GM12 [frequency over distributed normalized spectral abundance factor (dNSAF)], with candidate toxins (T) and antitoxins marked with different symbols. The analysis is based on three biological replicates. The protein abundance was tested in early log phase, early stationary phase and early stationary phase during heat stress using SP5 medium. Candidate toxins and antitoxins not displayed were below detection limits or absent. We included the house keeping proteins CspG and GyrB for comparison. The figure illustrates the dynamic changes of TA proteomic signatures over time, the TA 132/3 had the strongest signatures and its antitoxin signatures was getting smaller compared to its toxin pair under heat stress. https://doi.org/10.1371/journal.pgen.1009365.g003

Functionality testing of candidate TA systems using phylogenetically distant heterologous expression systems First, codon-optimized synthetic genes encoding candidate mycoplasma toxins and antitoxins were individually cloned into E. coli. The codon optimization was necessary since the mycoplasmas use a genetic code that differs from the universal genetic code resulting in the introduction of premature opal codons (UGA) once introduced in E. coli [32]. Candidate encoding genes were inserted under the control of the arabinose-inducible araBAD promoter of the pBAD/His expression system. The E. coli endoribonuclease toxin MazF [33] and the empty pBAD/His vector were used as positive and negative controls, respectively. Expression of the recombinant proteins was either repressed using glucose or induced using L-arabinose and cultures were incubated at 37°C for 5 hours followed by spot dilutions onto plates using ten-fold dilutions up to 10−6 (Fig 4A). No growth difference was observed between the different clones when the araBAD promoter was repressed. Bacteria concentrations were all ranging between 1.5 x 108 and 3 x 108 cells/mL (Fig 4A, left panels). In contrast, a 5-log reduction in bacteria concentration was observed upon induction of the recombinant E. coli MazF toxin and the candidate Mmc T 133 (Fig 4A, right panels). No toxic activity was recorded for the other E. coli clones upon induction of the remaining TA toxins and antitoxins. The expression of each candidate His6-tagged mycoplasma protein was confirmed by immunoblotting (S4 Fig). Variable expression levels were observed but all proteins were expressed at least three hours after induction. We further investigated the capacity of the antitoxin A 133 to neutralize the toxic activity of the candidate T 132 . To do so, we built an E. coli dual expression system based on the pET28a expression system (Fig 4B). This system allows the individual or concomitant expression of the two proteins using different inducers (arabinose and IPTG). The functionality of this system was tested for the individual expression of each protein. As expected, we observed a 3-log reduction in bacteria concentration when the expression of the T 133 protein was induced whereas no growth difference was observed upon induction of the A 132 counterpart (Fig 4B). More importantly, a complete neutralization of the T 133 toxic activity was obtained when the two proteins were expressed together (Fig 4B). Expression of recombinant proteins was confirmed. We then phenotypically characterized the toxic activity observed for the T 133 . Therefore, we followed the growth of the E. coli clone carrying the pBAD-T 133 for seven hours under repressed (Fig 4C) or induced conditions (Fig 4C). Under repressed conditions, no growth defect was observed for this clone when compared to the other E. coli clones carrying the pBAD-T 160 or the pBAD-T 752 . When arabinose was added, all three clones grew similarly until late exponential phase (4 hours post-induction). However, four hours post-induction, the OD 600 of E. coli clone harboring the pBAD-T 133 construct started to decrease drastically compared to the various controls (Fig 4C). Interestingly, the growth defect observed for this clone was also associated with a clear clumping of cells in the liquid culture. To better characterize this phenotypic observation, we visualized these E. coli cells at 7 hours post-induction using scanning electron microscopy (SEM) (Fig 4D). Micrographs revealed that the cells expressing the T 133 were presenting several morphological alterations when compared to the cells expressing the A 132 or those where the T 133 expression was repressed. Cells were abnormally elongated and showed atypical z-shape morphologies (Fig 4D, indicated by an asterisk). Obvious blebbing was also observed at one cell pole indicating early signs of cell death (Fig 4D). Lastly, burst E. coli cells were present as well as cellular debris (Fig 4D). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 4. Toxicity of recombinant candidate TA systems in E. coli. (A) Spot assays of E. coli LMG194 in response to induction or repression of heterologous expression of the different toxins, antitoxins and entire TA systems. The empty vector pBAD/His was used as negative control and the toxin MazF was used as positive control. Each data point represents the mean of three biological replicates, bars indicate standard deviation. (B) Neutralization experiments employing a dual expression system based on the pET28a vector enabling the expression of the two TA partners individually or in combination. Schematic representation of the construct is displayed (left) while the spot assay results are presented (right). A toxic effect is only observed after induction of the toxin but neutralization of the toxin activity was obtained via concomitant expression of the antitoxin. (C) Growth curves of E. coli LMG194 in response to induction or repression of heterologous expression of T 133 , T 160 , T 752 and empty vector pBAD/His. Each data point represents the mean of three biological replicates, bars indicate standard deviation. The p-values are displayed (* p ≤ 0.05, ** p ≤ 0.01).). (D) Scanning electron micrograph (magnification 10,000x) displaying morphological changes of E. coli LMG194 observed 7 hours after the induction of T 133 and A 132 recombinant proteins. Blebbing at the pole is indicated by an arrow, z-shaped cells by asterisk and cell debris by arrowheads. https://doi.org/10.1371/journal.pgen.1009365.g004 Altogether, these results confirmed that one out of the three candidate Mmc TA systems, namely the TA 132/3 system, was functional in E. coli. Despite lacking a cell wall, mycoplasmas are closely related to the gram-positive bacteria. Therefore, we wanted to use the closely related Bacillus subtilis to assess the functionality of the three candidate Mmc TA modules, in particular for those that could not be previously characterized. The expression of the candidate TA partners, either individually or in combination, was attempted using the commercially available pHT01-based expression system transformed in B. subtilis strain 168. The monitoring of the optical density of individual TA elements showed only a lag in growth for clones harboring pHT01-T 752 and pHT01-A 132 (S4A Fig). This phenotype was confirmed by the spot dilution assay (S4B Fig). Next, we tested the expression levels in B. subtilis, to confirm expression and to correlate expression with a growth phenotype. Despite several attempts, we failed to obtain detectable levels of expression for all candidates except for recombinant A 132 and T 133 proteins (S4C Fig). Even though expression of the T 752 protein was not visible in a Coomassie-stained gel, SEM micrographs showed an altered cell morphology compared to cells harboring the empty vector (S4D Fig). A large fraction of cells harboring T 752 burst, shrunk or presented holes at their surface. In contrast we did not observe a phenotype related to cell death in the clones harboring A 132 .

Functional characterization of candidate Mmc TA systems using M. capricolum In light of the issues encountered in using phylogenetically distant heterologous systems such as E. coli or B. subtilis, we opted to ultimately characterize all three candidate Mmc TA systems directly in mycoplasmas. We took advantage of the availability of the mycoplasma-based replicating plasmid pMYCO1 [34], as well as previously developed transformation protocols [35], to study the function of each TA partner in Mcap. We first constructed a set of four plasmids for each of the three candidate TA systems. These constructions included the gene(s) encoding (i) the entire TA module, (ii) the antitoxin only, (iii) the toxin only, all under the control of their natural promoter regions, as well as (iv) the toxin only under the control of the strong spiralin promotor [36]. Sequence-verified plasmids (S1 File) were transformed into Mcap and transformation efficacies were monitored to assess toxicity (Fig 5). Transformations with plasmids carrying the antitoxins yielded the same amount of transformants (~105−106 transformants per μg of plasmid) than those performed with the empty vector (Fig 5A). The transformation rates significantly dropped (3–3.5 logs difference) when plasmids harboring the toxins only were used. The number of transformants harboring the toxin with the natural promotor compared to the ones containing the spiralin promoter did not differ significantly, except when pMYC01-pSpi-T 133 was transformed as no transformants were isolated. Interestingly, transformation rates obtained with pMYCO1 plasmids harboring both TA partners were identical to the empty control, conclusively showing the neutralization capacity of all the antitoxins (Fig 5A). A maximum of five transformants per construct and per experiment were selected, passaged and sequenced. No mutations were found in the sequences of the genes coding the antitoxins and entire TA systems. Mcap cells transformed with the toxin pMYCO1-pNat-T 133 were found to not contain the toxin gene. About half of the transformants harboring the pMYC1-pNat-T 752 were intact whereas none of the pMYCO1-pSpi-T 752 ones contained the toxin genes anymore. No mutations were found in the nucleotide sequences of the constructs containing the toxin T 160 (both versions). PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 5. Effect of cloned toxins, antitoxins and TA systems on the transformation rate into M. capricolum subsp. capricolum. (A) M. capricolum subsp. capricolum ATCC 27343T was transformed with different plasmid constructs harboring either entire TA operons (TA 132/3 , TA 752/3 and TA 160/1 ), individual toxins or antitoxins. Empty pMYCO1 plasmid was used as a positive control. Each column represents the mean of three independent biological replicates and bars indicate standard deviations. Significance is indicated (** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001). (B) M. capricolum subsp. capricolum ATCC 27343T was transformed with different plasmid constructs harboring the toxins (T 133 , T 752 and T 160 ) under the control of an inducible promotor. Empty pMYCO1-ChloR plasmid was used as control. Graphs represents the mean of three independent biological replicates and bars indicate standard deviations. Significance is indicated (** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001). https://doi.org/10.1371/journal.pgen.1009365.g005

Induction of cell death in mycoplasmas The identification of three functional Mmc toxins prompted us to test if such toxic proteins could be used to trigger cell death in mycoplasmas. We built additional plasmid constructs based on the only available inducible promoter developed for mycoplasmas, namely the tetracycline-inducible promoter TetR-Pxyl/tetO 2 [37,38]. A chemically-synthetized sequence of this inducible promoter was introduced in the pMYCO1-ChloR where a chloramphenicol counterpart replaced the original tetracycline resistance cassette, thus enabling the selection of the transformants. Each of the three toxin-encoding genes was inserted under the control of the tetracycline-inducible promoter and plasmids were transformed in Mcap as previously described. Resulting clones were grown until late logarithmic phase then tetracycline (1 μg/mL) was added to induce the expression of the toxins. In parallel, the same Mcap clones were also grown in absence of tetracycline and were used as negative controls. Samples were taken at different time-points; ten-fold serially diluted up to 10−9 in non-selective SP5 medium and incubated at 37°C in 96-well plates. Color changing units (CCUs) were calculated as the last dilution where an acidification (from red to yellow) of the medium due to Mcap growth was recorded. The results of such experiments are displayed on Fig 5B. The induction of all three toxins resulted in a significant decrease of bacteria concentrations as early as 6 hours post-induction. The maximal toxic activity was recorded 12 hours post-inoculation. The toxin T 133 was particularly effective in triggering cell death as Mcap concentrations dropped from 109 CCU/mL at T0h to as low as 101 CCU/mL at T12h (Fig 5B). The toxin T 752 was also effective and 5–6 logs difference in bacterial concentrations were consistently observed at T12h. In the absence of tetracycline, all cultures were able to survive at least for 24 hours in stationary phase before entering the death phase.

[END]

[1] Url: https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1009365

(C) Plos One. "Accelerating the publication of peer-reviewed science."
Licensed under Creative Commons Attribution (CC BY 4.0)
URL: https://creativecommons.org/licenses/by/4.0/


via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/

You are viewing proxied material from magical.fish. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.