A toxin/antitoxin system targeting the replication sliding-clamp induces competence in Streptococcus pneumoniae

1 A toxin/antitoxin system targeting the replication sliding-clamp 2 induces competence in Streptococcus pneumoniae. 3 Maziero Mickaël 1, 2 , Juillot Dimitri 3, Mortier-Barrière Isabelle 1, 2 , Carballido-Lopez Rut 3, 4 Campo Nathalie 1, 2 , Genevaux Pierre 1, 2 , Bordes Patricia 1, 2 , Patrice Polard 1, 2 * and Mathieu 5 Bergé1, 2*. 6 1 Laboratoire de Microbiologie et Génétique Moléculaires (LMGM), Centre de Biologie 7 Intégrative (CBI), CNRS, Toulouse, France. 8 2 Université de Toulouse, Toulouse, France. 9 3 Université Paris-Saclay, INRAE, AgroParisTech, Micalis Institute, 78350, Jouy-en-Josas, France 10 *Corresponding authors 11 Abstract: 12 Streptococcus pneumoniae is a pathogenic bacterium capable of entering a cellular 13 differentiation state, called competence, which enables it to acquire new genetic functions by 14 natural transformation, as well as physiological functions such as tolerance to a number of 15 antibiotics. The transition to this state is regulated by various environmental or intracellular 16 signals that converge on the comCDE operon, which groups together the competence 17 initiation genes. A fraction of activated cells is sufficient to propagate competence to the 18 whole population via the product of the comC gene, the competence stimulating peptide 19 (CSP). Remarkably, depletion of the essential ClpX/ ClpP AAA+ protease has been shown to 20 induce the comCDE operon. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 21 Here we demonstrate that the ClpX-dependent induction of competence relies on the Spr1630 22 toxin (RipA), part of a Rosmer toxin-antitoxin system. We show that this toxin generates 23 replicative stress by acting on the sliding clamp of replication, inducing transcription of the 24 comCDE operon. Bacteria that produce RipA appear to lose their viability but remain 25 metabolically active and able to produce CSP, thereby transferring competence to viable 26 neighbouring cells. 27 Authors’ summary: 28 The environment in which bacteria live puts them under a great deal of stress, forcing them 29 to adapt constantly, either temporarily or permanently. Streptococcus pneumoniae , a 30 pathogenic bacterium implicated in various pathologies such as otitis, meningitis and 31 pneumonia, is also subject to stress, whether from its host, antibiotic treatments or the 32 microbiota in which it lives. In response to this, S. pneumoniae is able to switch to a 33 differentiated state called competence. This allows it to acquire new genetic characteristics 34 through natural transformation, but also to better tolerate stresses such as antibiotics 35 pressure. 36 The underlying signals and signaling pathways of this phenotypic switch remain poorly 37 characterized. In this study, we identified a novel toxin–antitoxin system that, when activated, 38 causes a subset of the population to self-sacrifice by disrupting its own DNA replication. This 39 self-induced arrest serves as a signal that promotes the transition to competence in 40 neighboring cells, thereby improving the capacity for adaptation at the populational level. 41 42 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 43 Introduction: 44 Streptococcus pneumoniae is a commensal bacterium of the human nasopharynx, but can 45 become an invasive or non-invasive pathogen leading to a wide range of diseases such as 46 otitis, pneumonia, meningitis [1,2]. The emergence of antibiotic resistance in S. pneumoniae 47 is making treatment more complex, and led to the deaths of around 600,000 people in 2019 48 [3]. This is one of the reasons why S. pneumoniae was ranked among the 12 priority pathogens 49 by the World Health Organization in 2017. Several studies have strongly suggested that the 50 emergence and rapid spread of antibiotic resistance is due to the ability of S. pneumoniae to 51 undergo natural transformation [4]. 52 Natural transformation is a horizontal gene transfer mechanism that enables bacteria to 53 capture exogenous DNA and integrate it into their chromosomes through homologous 54 recombination. First identified in S. pneumoniae [5,6], then in a wide range of bacteria, this 55 mechanism has been relatively well characterized at the molecular level in a variety of bacteria 56 [7–9]. In S. pneumoniae, most of the proteins required for natural transformation are only 57 expressed during a period when the bacterium enters a particular physiological state called 58 competence. The induction of competence is regulated by the level of transcription of two 59 operons, comCDE and comAB. The comC gene encodes for a peptide [10] which is exported 60 and matured by the ABC-transporter ComAB; the externalized product is called Competence 61 Stimulating Peptide (CSP). CSP is able to activate the ComDE two component system [11]. In 62 turn, phosphorylated ComE activates transcription of the comAB and comCDE operons as part 63 of the early competence regulon, inducing an autocatalytic loop [11,12]. Two waves of genes 64 are then successively induced: late and delayed competence genes [13,14], resulting in ~14% 65 of the transcriptome being modified during competence. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 66 The transition to competence is initially an individual behavior and when a sufficient fraction 67 of the population has reached competence, the entire population is converted to competence 68 following transmission of CSP via cell contact [15] by the self-inducing fraction [16]. The time 69 required for this inductive fraction to appear is highly dependent on the stimuli received 70 individually by the cells, which will affect the transcription of the early comCDE operon. 71 Several stresses are known to promote competence, such as drugs that alter replication or 72 genome integrity [17–19], temperature [20,21] and ribosome decoding [22]. At least two 73 transduction pathways from these signals to comCDE have been described [20], but it is clear 74 that not all pathways have been identified, and those that have remain poorly characterized. 75 Early on in the molecular characterization of competence, it was noted that ClpP appeared to 76 repress its development [23]. The clpP gene encodes the proteolytic subunit that forms the 77 Clp ATP-dependent protease complex, together with an ATPase subunit ClpC, ClpE, ClpL, or 78 ClpX. Systematic CRISPRi (clustered regularly interspaced short palindromic repeats 79 interference) of clpP or any of the genes encoding the ATPase subunits demonstrated that 80 only depletion of ClpP or ClpX induced competence development [24]. It was logically 81 proposed that ClpX is the main ATPase implicated with ClpP to repress competence. 82 Interestingly, ClpX is the only ATPase subunit of the ClpP complex that is essential for S. 83 pneumoniae viability [25]. The essentiality of clpX was shown to be due to the presence of the 84 gene spr1630 which is proposed to encode the toxin of a so far uncharacterized toxin-antitoxin 85 (TA) system [26]. 86 The aim of the present work was to characterize the molecular pathway inducing competence 87 under the control of ClpXP. Using genetic approaches, we demonstrated that competence 88 induction by depletion of ClpX but not ClpP is dependent on spr1630. We demonstrated that 89 Spr1630 is able to induce competence alone. We experimentally demonstrated that Spr1629 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 90 antagonizes the effects of Spr1630 leading us to propose that Spr1630 and Spr1629 form a 91 toxin-antitoxin system (TA) that belongs to the large Rosmer TA family. Furthermore, our data 92 strongly suggest that Spr1630 targets DnaN, the sliding clamp in DNA replication, and as such 93 we propose to rename this TA system RipAB (Replication Interfering Protein). Finally, we place 94 the impact of this TA system in the population context of the onset of competence and show 95 that production of the RipA toxin gives rise to an inducing fraction that acts as a group of 96 sentinels that warn the rest of the population of a stress by propagating competence 97 development. 98 Results: 99 Competence induction by clpX depletion is RipA-dependent. 100 We first sought to reproduce the induction of competence by transcriptional depletion of clpX 101 or clpP as described previously [23,24]. To monitor the development of competence, we used 102 a transcriptional fusion of the comCDE early operon promoter with the luciferase gene as a 103 reporter [17,27]. We then constructed strains carrying constructions that allow Clustered 104 Regularly Interspaced Short Palindromic Repeats interference (CRISPRi) in S. pneumoniae [24]. 105 Briefly, the defective Streptococcus pyogenes Cas9 (dCas9) controlled by an IPTG-dependent 106 promoter was integrated into the S. pneumoniae chromosome together with the lacI gene and 107 a single-guide RNA (sgRNA), targeting clpX or clpP, under the control of constitutive promoters 108 [24]. When these cells were cultured in an unfavourable environment for the spontaneous 109 development of competence (pH 7.0), no changes in the transcription of the comCDE operon 110 were observed (Fig 1A, black lines). The addition of IPTG induced a strong increase in luciferase 111 production reflecting the induction of transcription of the comCDE operon in either ClpX and .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 112 ClpP depletion (Fig 1A, red lines), suggesting that the depletion of ClpX or ClpP is a competence 113 trigger, as previously published [24]. Notably, the increase of comCDE transcription was 114 concomitant with a reduction in growth, which reports on the depletion of the Clp proteins. 115 Since the biological essentiality of ClpX has previously been linked to the ripA gene [26], we 116 explored the possibility that RipA is also involved in triggering competence. As expected, 117 deletion of ripA significantly improved growth of ClpX depleted cultures (Fig 1B). Importantly, 118 competence induction was completely abolished in the ClpX depleted culture lacking ripA (Fig 119 1B, top left panel), indicating that competence depends solely on RipA when ClpX is depleted. 120 In an interesting way, and in contrast to the ClpX depletion background, ClpP depletion was 121 still able to trigger competence in the ripA null mutant strain (Fig 1B, top right panel ), 122 suggesting that ClpP dependent induction of competence is not functionally related to RipA. 123 However, as the absence of ripA consistently enhanced growth of the ClpP-depleted strain 124 (Fig 1B, bottom right panel), we cannot rule out the involvement of ClpP in RipA toxicity 125 management. Altogether, these results confirmed the link between ClpX and RipA, and 126 revealed a ClpX-dependent role of RipA in competence induction. 127 The RipA toxin induces competence and is antagonized by the RipB antitoxin. 128 To test the ability of RipA to induce competence, we produced the protein under the control 129 of an IPTG inducible promoter [28]. To ensure tight repression of the IPTG-sensitive promoter, 130 a second copy of the lacI gene driven by the constitutive PF6 promoter was integrated into 131 the S. pneumoniae chromosome [24]. As ripA is part of an operon together with the essential 132 ripB gene (spr1629) (Fig 2A), we also constructed strains containing the whole operon or ripB 133 alone under IPTG induction. Fig 2B shows that expression of ripA induces competence 134 development in an IPTG-dependent manner whereas ripB expression does not. Interestingly, .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 135 expression of the ripAB operon does not lead to competence development either, suggesting 136 that ripB expression inhibits the action of RipA. We also note that ripA expression alone causes 137 growth retardation in correlation with IPTG concentration and again co-expression of ripB 138 impairs this growth defect (Fig 2B). We confirmed this by assessing the growth capacity of 139 these strains on agarose medium in presence of increasing IPTG concentrations (Fig 3AB). 140 From 10 µM IPTG, the strain expressing only ripA began to lose viability. Growth ability was 141 almost completely lost from 20 µM IPTG. Production of ripB in operon with ripA completely 142 cancelled out this toxicity. To test whether RipB could exert its protective effect in trans, we 143 expressed ripA and ripB under IPTG inducible promoters, but at two different chromosomal 144 loci. In this condition, RipB again antagonised the effect of RipA (Fig 3C). These results are 145 consistent with the idea that RipA and RipB form a TA system as previously proposed [26]. We 146 next tested whether RipA could be responsible for the previously observed essential nature 147 of ripB [29] by attempting to delete ripB in different genetic backgrounds. Transformation 148 results shown in Fig 3D demonstrate that ripB deletion could be obtained only when ripA was 149 absent, further supporting that RipAB is a bona fide TA system in S. pneumoniae. Since RipA is 150 responsible for the essentiality of both clpX [26] and ripB in S. pneumoniae, these data suggest 151 that both proteins are controlling RipA activity. 152 Suppressor mutations in dnaN antagonize RipA toxicity 153 In search for RipA potential target(s), we took advantage of a S. pneumoniae interactome 154 study based on yeast two-hybrid (Y2H) approach, which suggests that RipA interacts with 155 DnaN [30]. As DnaN encodes the sliding clamp, an essential element in DNA replication, we 156 considered that DnaN could be the target of RipA. To test this hypothesis, we searched for 157 genetic suppressors of RipA toxicity in the dnaN locus. To this end, the strain producing RipA .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 158 under the control of an IPTG-inducible promoter was transformed by error-prone ~ 4kb PCR 159 fragments centred on the dnaN locus and plated on agar medium supplemented with 20 µM 160 IPTG. Interestingly, transformation with error-prone dnaN PCR fragments resulted in a higher 161 number of colonies (about tenfold) compared to non-transformed control cells (Fig 4A). 162 Sequencing the 4kb dnaN region of these transformants revealed that all mutations were 163 located in the dnaN gene (Fig 4B). Seven strains with single mutations were obtained, leading 164 to seven different amino acid substitutions affecting 5 different residues of DnaN (H183, L185, 165 Y249, Y333, L371). Remarkably, these five residues are clustered in the same region of the 166 DnaN structure (Fig 4B), a hydrophobic pocket implicated in DnaN protein-protein interactions 167 [31]. We then characterized the ability of these mutations to suppress the toxicity linked to 168 increased production of RipA (Fig 4C). We found that, the majority of the mutations allow 169 significant growth at IPTG concentrations similar to the chosen selection threshold (20 µM 170 IPTG). Beyond this concentration, their effect diminishes drastically as ripA transcription 171 increases. However, two mutations (Y249N) and (Y333H) enable growth at the highest IPTG 172 concentration tested (200µM IPTG). This suggests that the various mutations do not exhibit 173 the same ability to prevent toxicity of RipA. We have therefore categorised them as strong 174 (Y249N, Y333H), intermediate (L185R) or weak (H183Q, H183L, H183P, L371F) suppressors. 175 DnaN substitutions impair DnaN-RipA interaction in yeast two-hybrid assay. 176 Since RipA interacts with DnaN in yeast two-hybrid assays (Y2H) [30], we investigated whether 177 the single nucleotide polymorphism suppressive mutations in dnaN alter this interaction. We 178 decided to test this interaction with one of the strongest suppressive mutations (Y249N), as 179 well as a representative of the weaker mutations (H183P), and a mutation that displays an 180 intermediate suppressive phenotype (L185R). First, we detected an interaction between DnaN .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 181 with itself as expected, but also for RipA with itself, suggesting a potential dimerization of the 182 protein RipA (Fig 5A). We also reproduced the Y2H interaction between RipA and DnaN in both 183 directions of interactions. However the RipA:DnaN interaction was completely abolished 184 whatever the suppressive mutation analysed, (Fig 5B) suggesting a loss or reduction of 185 physical interaction between RipA and DnaN proteins that express suppressive mutations. 186 Together, these data suggest that RipA targets DnaN through a physical interaction. 187 DnaN substitutions impair competence development mediated by RipA. 188 We next investigated whether the interaction between RipA and DnaN was responsible for 189 the induction of competence in cells overexpressing RipA. To answer this, we tested the 190 induction of competence in strains carrying the three categories of suppressive mutations 191 described above. All strains carrying a suppressive mutation showed better growth in liquid 192 medium compared to the reference strain in IPTG+ conditions (Fig 6A). Mutant strains 193 displayed a gradation in the suppressive effect of the toxicity, as observed in spot test assays 194 (Fig 4C). Thus, the strain carrying the most effective suppressive mutation (DnaNY249R) did not 195 show any growth problems whatever the IPTG concentration. Conversely, the weakest 196 suppressor mutant (DnaN H183P) remained quite strongly impacted at the highest RipA 197 production level. The DnaNL185R mutant exhibited an intermediate growth behaviour (Fig 6A). 198 Overall, competence induction of the mutant strains was strikingly impaired compared to the 199 WT reference strain (Fig 6A). However, enlarging the comCDE transcription observation scale 200 by a logarithm (Fig 6B) revealed a gradation in behaviour during the development of 201 competence in the three representative mutant strains. Indeed, DnaN Y249R never induced 202 competence whatever the concentration of IPTG whereas, DnaN H183P and DnaN L185R strains 203 only displayed a residual competence burst at high ITPG concentration (Fig 6B). It is interesting .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 204 to note that the ability to trigger competence appears inversely proportional to the 205 categorisation proposed to describe the ability to withstand the toxicity of RipA. 206 “Doomed cells” are able to promote competence of naive cells. 207 All the above elements support a link between the toxicity of RipA and its ability to induce 208 competence. To test this, we compared the toxicity and competence induction kinetics. When 209 RipA-producing cells were observed by microscopy, it was clear that bacterial multiplication 210 was affected after 90 minutes of IPTG exposition, but that cell integrity was maintained for 30 211 additional minutes (Fig 7A). To complete this observation, we evaluated the ability of the same 212 culture to generate colonies on IPTG-free agarose medium after different RipA exposure 213 duration. Fig 7B shows that most cells were unable to generate colonies after 45 minutes of 214 exposure. However, as competence development occurred after about 120 minutes of RipA 215 toxicity exposure (Fig 2B), we wondered about the physical state of these cells. To test 216 whether these cells are physiologically active, at least regarding competence regulation, we 217 incubated RipA-producing cells under the control of IPTG for more than 60 minutes, with 218 synthetic CSP to trigger rapid competence induction in the culture. These cells carrying the 219 comC::luc reporter gene were still able to respond to CSP at a higher level than unstressed 220 cells that did not produce RipA (Fig 7C). We then tested if RipA-exposed cells were also able 221 to produce enough CSP to propagate competence to naive cells. For this, we grew a strain 222 expressing ripA under IPTG promoter but lacking the comC::luc reporter for 120 minutes in 223 medium containing IPTG to allow induction of RipA-mediated competence. As a negative 224 control, we also used a strain with the same background, but unable to produce CSP (comC0 225 strain). After the allotted time, these cells were mixed (1:1 ratio) with naive wild-type cells 226 carrying the comC::luc reporter (strain R825). Almost instantaneously, wild-type cells switched .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 227 to competence when mixed with RipA stressed cells capable of producing CSP, whereas no 228 change was observed for wild-type cells alone or mixed with comC0 cells (Fig 7D). These 229 observations were complemented by measuring the ability of strain R825 to transform. Under 230 these conditions, transformation can only occur in the presence of cells that produce ripA and 231 are capable of producing CSP (Fig 7D, bottom panel). Altogether these results show that the 232 RipA-producing cells, although rapidly losing their viability, remain metabolically active and 233 produce sufficient CSP to propagate the competence to naive cells and promote their ability 234 to transform. 235 Discussion: 236 ClpX and ClpP use independent pathways to repress competence. 237 We found that the RipA toxin is responsible for the induction of competence generated by 238 ClpX depletion but not ClpP depletion. Since deletion of ripA slightly improves the growth of a 239 clpP strain, it is not possible to exclude the possibility of a partial link between RipA and ClpP, 240 but ClpP-mediated induction of competence appears to be almost independent of RipA. It is 241 very likely that ClpP, in association with other accessory proteins like ClpE and ClpC, ensures 242 the homeostasis of several competence-regulating proteins, such as ComX or ComW 243 respectively as previously demonstrated [32] or hypothesized, such as ComE [23]. Liu et al., 244 demonstrated that among the ClpP accessory proteins, only depletion of ClpX by CRISPRi was 245 capable of inducing competence [24]. However, it should be kept in mind that the intensity of 246 protein depletion generated by CRISPRi can be extremely variable from one gene to another 247 or from one guide to another [33]. In addition, the quantity of accessory proteins required to .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 248 ensure repression of competence can also be extremely variable depending on the protein. 249 Understanding the role of ClpP in competence repression will require further investigation. 250 How are RipA, RipB and ClpX related? 251 As ripA is responsible for clpX essentiality, it has already been hypothesised that RipA is a toxin 252 that is impaired in its activity by ClpX [26]. We have furthermore shown here that the essential 253 nature of RipB is also linked to the presence of RipA. The C-terminus of RipB protein shares 254 homology with ImmA, a Zn metalloprotease involved in regulation of excision and transfer of 255 the mobile genetic element ICEBs1 [34,35]. The N-terminus contains a transcriptional 256 regulator domain of the XRE-family, sharing homology with the HipB antitoxin. As ectopic 257 expression of ripA was impaired by trans-expression of ripB, it is reasonable to suggest that 258 RipB acts as an antitoxin and that RipAB belongs to the large Rosmer TA family [36]. However, 259 the RipAB system is not autonomous, since the action of RipB is strictly dependant on clpX 260 integrity. This could be explained by different mechanisms. Firstly, we could assume that ClpX 261 somehow modulates the toxin/antitoxin activation cycle. The involvement of ClpX in 262 managing the function of a toxin-antitoxin system has already been reported, but until now 263 its role has been described as activating the degradation of the antitoxin and therefore 264 releasing the toxin [37–39]. In this work, ClpX plays a negative role in the action of the toxin. 265 Other more indirect relationships might also be involved, for example, in a ClpX-deficient 266 genetic context, DnaN appeared to be more sensitive to the action of RipA. 267 The DnaN sliding clamp is a target of RipA 268 Given that RipA toxicity can be counteracted by mutations in the DnaN gene and that the same 269 mutations abolish the Y2H interaction between RipA and DnaN, it is reasonable to propose .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 270 that the sliding clamp is the target of RipA. The most obvious molecular action would be that 271 RipA affects the main action of the sliding clamp, by disrupting DNA replication. Replicative 272 stresses are clearly known to be factors that induce competence [16,17,19,40], which is in line 273 with our results showing that RipA production induces competence and that suppressive 274 mutations in dnaN strongly antagonise it. However, we cannot rule out the hypothesis that 275 DnaN is an intermediate target of RipA and that its final target would be a partner of the sliding 276 clamp. DnaN is known to act as a hub for interactions between different proteins such as MutL 277 and MutS, DNA polymerases or ligases [41]. The majority of these interactions occur via a 278 hydrophobic pocket localized on the anti-Pol face of the sliding Clamp [31]. All the suppressive 279 mutations uncovered in this work are localized in the hydrophobic pocket close to the 280 interaction zone. It is therefore conceivable that RipA disrupts the dynamics of the sliding 281 clamp interactome, thereby inducing perturbations in the replicative machinery. 282 Biological role of RipA in S. pneumoniae 283 Several TA systems affecting replication have been discovered mostly affecting 284 topoisomerases [42,43], but very few directly affect the sliding-clamp [44]. For example, SocB, 285 the toxin of the SocAB type VI TA, was the first toxin described to target the sliding-clamp in 286 Caulobacter crescentus [45]. The discovery of its target was also achieved using genetic 287 approaches. The mutations suppressing the action of SocB were located in the same 288 hydrophobic pocket of the sliding clamp as the one supressing RipA toxicity. In addition, it is 289 interesting to note that SocB induces the SOS response system by altering replication in C. 290 crescentus; while in S. pneumoniae, a bacterium lacking an SOS response system [45], ripA 291 expression induces the development of competence, a physiological state frequently 292 proposed as a replacement for the SOS system [19,40,46,47]. However, consultation of .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 293 alignment tools did not reveal any sequence or structural homology between SocB and RipA, 294 nor between their respective antitoxins (https://www.rcsb.org/alignment). Taken together, 295 these elements suggest a functional convergence of these TA systems towards a central target 296 of the cell cycle. 297 However, one notable difference needs to be highlighted: while the induction of the SOS 298 system remains an individual behaviour, the induction of competence leads to a population 299 behaviour. In fact, it is well established that certain replicative stresses induce competence in 300 S. pneumoniae and in this way, a fraction of the population subjected to RipA stress could 301 transmit this signal through the production of CSP to the whole of the unstressed population, 302 through a potential sacrificial behaviour (Fig 8). In the end, the switch to competence allows 303 the generation of a heterogeneous population, both physiologically and genetically, allowing 304 the emergence of potentially better adapted individuals [16]. The fate of the inducer cells is 305 of particular interest, as it will provide clues as to the biological signals that activate RipA. 306 Indeed, the transition of these cells to a stasis state could be consistent with the ability of 307 antibiotics to induce competence [17]. In such a model, stressed cells that activate RipA trigger 308 two mechanisms that allow them to transition to a potentially persistent state while enabling 309 other cells to become competent. This could be relevant, given that competence enables the 310 population to tolerate certain antibiotics more effectively [16]. Alternatively, sacrificial 311 behaviour could limit the spread of danger while warning neighbouring cells of potential 312 stress. This behaviour is reminiscent of phage resistance mechanisms, in which TA systems are 313 involved [44,48]. This suggests that we should explore the possibility of a previously unknown 314 link between competence and phages in S. pneumoniae. 315 Material and methods: .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 316 Strains and growth media. 317 All S. pneumoniae strains used were derived from R800 strain [49] and are listed in the 318 supporting tables file. 319 Standard procedures for transformation and growth media were used [50]. Briefly, pre- 320 competent cells were treated at 37°C for 10 min with synthetic CSP1 (100ng mL −1) to induce 321 competence, then exposed to transforming DNA for 20 min at 30°C. Transformants were then 322 plated on CAT agar supplemented with 4% horse blood and incubated for 120min at 37°C. 323 Transformants were then selected by addition of a second layer of agar medium containing 324 the appropriate antibiotic and incubated overnight at 37°C. Antibiotic concentrations (µg mL– 325 1) used for the selection were: chloramphenicol (Cm), 4.5; kanamycin (Kan), 250; 326 spectinomycin (Spc), 100; streptomycin (Sm), 200, gentamycin (G), 40 and erythromycin (E), 327 0.1. Unless otherwise described, pre-competent cultures were prepared by growing cells to 328 an OD 550nm of 0.1 in C+Y medium (pH 6.8). Then cells were 10-fold concentrated in C+Y 329 medium supplemented with 15% glycerol and storage at –80°C. 330 Deletion and invalidation mutagenesis. 331 Deletion or invalidation mutagenesis was based on strand overlap extension (SOE) [51]. 332 Briefly, primers MP170 and MP173 were used to generate PCR fragments carrying 333 spectinomycin or kanamycin resistance gene from plasmids pr412 (SpcR) or pr413 (KanR) 334 previously described [52]. The two PCR fragments that flank the integration site of the 335 resistance gene were amplified with the specific primer pairs described in the supporting 336 tables file. These pairs are composed of a primer defining the integration site and carrying the 337 sequence complementary to MP170 or MP173 in its 5' region and a distal primer at around 1 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 338 to 2 kb. The three PCR amplified fragments were purified and used as template to produce a 339 unique PCR fragment using the two distal primers. The resulting fragment was used to 340 transform a recipient strain as described above. The transformed strains were selected by 341 appropriate antibiotic selection and sequenced (Eurofins genomics). 342 IPTG-dependent expression platform construction. 343 SOE was also used to generate expression platforms CEPlac- spr1630, CEPlac spr1629 or 344 CEPlac-spr1630-1629. Primers pairs MB293-MB294 and MB295-MB291 were used to amplify 345 an expression platform from strain R3833 previously described [28]. Internal PCR fragments 346 carrying genes of interest were amplified from R800 strain using the following primers pairs 347 MB296-MB297 (spr1629-1630), MB296-MB298 (spr1630), and MB299-MB297 (spr1629). Each 348 of these fragments was used with MB293-MB294 and MB295-MB291 primer pairs as 349 templates to generate a unique PCR fragment with primers MB293 and MB291. This final 350 fragment was used to transform strains of interest as described above (kanamycin selection). 351 The PcepII-lac platform was generated by transferring the Pcep-lac platform to the cpsN locus. 352 Briefly, MM56 and MM57 were used to amplify the platform from strain R5139. Flanking cpsN 353 loci were amplified using YA09-MM58 et YA14-MM59 primer pairs with R4631 strain as a 354 template [53]. These three PCR fragments were used as templates to amplify a single PCR 355 fragment using primers YA09-YA14. This fragment was used to transform strain R5198. 356 CRISPRi depletion 357 Strains carrying the CRISPRi system were constructed as described previously [24]. Plasmids 358 carrying lacI repressor (pPEPY-PF6-lacI ), Cas9 enzyme (pJWV102-PL-dCas9) and expressing 359 interfering sgRNA (pPEPX-P3-sgRNAluc) were purchased from addgene. Plasmids pPEPX-P3- .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 360 sgRNA clpX and pPEPX-P3-clpP were constructed using the In-Fusion HD cloning kit (Takara) 361 as previously described [24]. Primers OIM134 and OIM135 were used to amplify pPEPX-P3- 362 sgRNA backbone. Primers pairs used to generate the RNA guide, MB551-MB552 (clpX), MM54- 363 MM55 (clpP), were published previously [24]. Generated plasmids were sequentially 364 integrated into the genome of strain R800 by transformation and homologous recombinaison. 365 Monitoring of growth and luciferase expression. 366 For monitoring operon comCDE expression, a transcriptional fusion was used, inserting a 367 S. pneumoniae comCDE promoter fragment (HindIII - BamHI) upstream of the luc gene and 368 carried by pR414 plasmid. Homology-dependent integration of the non-replicative 369 recombinant plasmid into the pneumococcal chromosome was selected using erythromycin 370 resistance [52]. 371 For the monitoring of growth and luciferase expression, precultures were gently thawed and 372 aliquots were inoculated (unless otherwise described) at OD 550nm of 0.005 in luciferin- 373 containing C+Y medium (pH 7) as previously described [50] and distributed into a 96-wells 374 (300 µl per well) white microplate with clear bottom (Corning). Relative luminescence units 375 (RLU) and OD492nm values were recorded at defined time points throughout incubation at 37°C 376 in a Varioskan luminometer (ThermoFisher). 377 DnaN error prone PCR and screening. 378 DnaN locus was amplified from R800 genomic DNA using primers MB313 and MB314 and 379 DreamTaq DNA Polymerase (thermoFisher scientific) as recommended by the supplier with or 380 without adding 25µM of MnCl 2 to reduce the fidelity of the enzyme. The resulting fragments 381 were used to transform strain R5086 as described above. After an incubation of 120min in CAT .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 382 medium at 37°C, transformant cells were plated and grown 24 hours on CAT agar 383 supplemented with 4% horse blood and 20µM IPTG. DnaN locus was amplified on potential 384 suppressor clones using primers MB313 and MB314 and a high-fidelity DNA polymerase, 385 Primestar max DNA polymerase (Takara). PCR fragments were used to transform R5086 as 386 previously described to confirm that suppressive mutations were linked to DnaN locus. 387 Relevant PCR fragments were sequenced (Eurofins genomics). 388 3D mapping of suppressive mutations on DnaN. 389 Suppressive mutations were positioned on 3D DnaN structure from Protein Data Bank (PDB), 390 www.pdb.org (PDB ID code 2AWA) using Pymol software V0.99. 391 Yeast Two hybrid. 392 Gene-coding sequences of S. pneumoniae Spr1630 and DnaN proteins were PCR amplified 393 using R1501 DNA as template, and inserted in Gal4-based plasmids, PGAD-C1 and PGBD-C1 394 [54], using the In-Fusion HD cloning kit (Takara). Primers used were MM30 and MM31 395 (Spr1630), MM28 and MM29 (DnaN) and OCN424 and OCN425 (PGAD-C1 and PGBD-C1). 396 Resulting plasmids were then transformed independently in the yeast strains PJ69-4a and 397 PJ69-4α. Saccharomyces cerevisiae cells expressing S. pneumoniae proteins as GAL4 Binding 398 Domain (BD) fusions were mated with cells expressing some of these proteins as GAL4 399 Activating Domain (AD) fusions. Binary interactions were identified by growth of diploid cells 400 after 8 or 20 days at 30°C on synthetic complete medium [55] lacking leucine, uracil and 401 histidine (to select expression of the HIS3 interaction reporter). Controls with empty vector 402 plasmids (i.e., carrying only the BD or AD domain) were systematically included. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 403 Mixed culture Assay 404 Strains R5204 and R5086 were inoculated at 0.04 OD 550nm in C+Y medium and grown in the 405 presence of 80 µM of IPTG for 120 minutes. In parallel R825 strain was inoculated at 0.04 406 OD550nm in C+Y medium (pH 7) and grown for 60 minutes. Cells R5204 or R5086 were mixed 407 with R825 (volume to volume) to an expected ratio of 1 to 1. R825 luciferase activity was then 408 monitored. PCR amplified DNA carrying rpsL41 allele (Streptomycin resistance) was added at 409 the time of mixing. 100 minutes after mixing a culture sample was collected and plated on 410 agar medium supplemented with erythromycin and streptomycin to determine the 411 percentage of transformation of strain R825. 412 Statistical tests: 413 Pairwise comparisons were done with a nonparametric MannWhitney test. P values were 414 displayed as follows: ***, 0.0001 < P < 0.001; **, 0.001< P < 0.01; *, 0.01 < P  0.05. 416 Acknowledgements. 417 We would like to extend our special thanks to Calum Johnston and Hélène Cordier for their 418 critical review of the manuscript. 419 References. 420 1. Prina E, Ranzani OT, Torres A. Community-acquired pneumonia. The Lancet. 2015;386: 1097– 421 1108. doi:10.1016/S0140-6736(15)60733-4 422 2. Simell B, Auranen K, Käyhty H, Goldblatt D, Dagan R, O’Brien KL. The fundamental link between 423 pneumococcal carriage and disease. Expert Rev Vaccines. 2012;11: 841–855. 424 doi:10.1586/erv.12.53 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 425 3. Murray CJL, Ikuta KS, Sharara F, Swetschinski L, Robles Aguilar G, Gray A, et al. Global burden of 426 bacterial antimicrobial resistance in 2019: a systematic analysis. The Lancet. 2022;399: 629– 427 655. doi:10.1016/S0140-6736(21)02724-0 428 4. Straume D, Stamsås GA, Håvarstein LS. Natural transformation and genome evolution in 429 Streptococcus pneumoniae. Infect Genet Evol. 2015;33: 371–380. 430 doi:10.1016/j.meegid.2014.10.020 431 5. Griffith F. The Significance of Pneumococcal Types. Epidemiol Infect. 1928;27: 113–159. 432 doi:10.1017/S0022172400031879 433 6. Avery OT, MacLeod CM, McCarty M. STUDIES ON THE CHEMICAL NATURE OF THE SUBSTANCE 434 INDUCING TRANSFORMATION OF PNEUMOCOCCAL TYPES : INDUCTION OF TRANSFORMATION 435 BY A DESOXYRIBONUCLEIC ACID FRACTION ISOLATED FROM PNEUMOCOCCUS TYPE III. J Exp 436 Med. 1944;79: 137–158. doi:10.1084/jem.79.2.137 437 7. Johnston C, Campo N, Bergé MJ, Polard P, Claverys J-P. Streptococcus pneumoniae, le 438 transformiste. Trends Microbiol. 2014;22: 113–119. doi:10.1016/j.tim.2014.01.002 439 8. Claverys J-P, Martin B, Polard P. The genetic transformation machinery: composition, 440 localization, and mechanism. FEMS Microbiol Rev. 2009;33: 643–656. doi:10.1111/j.1574- 441 6976.2009.00164.x 442 9. Dubnau D, Blokesch M. Mechanisms of DNA Uptake by Naturally Competent Bacteria. Annu Rev 443 Genet. 2019;53: 217–237. doi:10.1146/annurev-genet-112618-043641 444 10. Håvarstein LS, Coomaraswamy G, Morrison DA. An unmodified heptadecapeptide pheromone 445 induces competence for genetic transformation in Streptococcus pneumoniae. Proc Natl Acad 446 Sci U S A. 1995;92: 11140–11144. 447 11. Pestova EV, Håvarstein LS, Morrison DA. Regulation of competence for genetic transformation 448 in Streptococcus pneumoniae by an auto-induced peptide pheromone and a two-component 449 regulatory system. Mol Microbiol. 1996;21: 853–862. 450 12. Ween O, Gaustad P, Håvarstein LS. Identification of DNA binding sites for ComE, a key regulator 451 of natural competence in Streptococcus pneumoniae. Mol Microbiol. 1999;33: 817–827. 452 13. Dagkessamanskaia A, Moscoso M, Hénard V, Guiral S, Overweg K, Reuter M, et al. 453 Interconnection of competence, stress and CiaR regulons in Streptococcus pneumoniae: 454 competence triggers stationary phase autolysis of ciaR mutant cells. Mol Microbiol. 2004;51: 455 1071–1086. 456 14. Peterson SN, Sung CK, Cline R, Desai BV, Snesrud EC, Luo P, et al. Identification of competence 457 pheromone responsive genes in Streptococcus pneumoniae by use of DNA microarrays. Mol 458 Microbiol. 2004;51: 1051–1070. 459 15. Prudhomme M, Berge M, Martin B, Polard P. Pneumococcal Competence Coordination Relies 460 on a Cell-Contact Sensing Mechanism. PLoS Genet. 2016;12. doi:10.1371/journal.pgen.1006113 461 16. Prudhomme M, Johnston CHG, Soulet A-L, Boyeldieu A, De Lemos D, Campo N, et al. 462 Pneumococcal competence is a populational health sensor driving multilevel heterogeneity in 463 response to antibiotics. Nat Commun. 2024;15: 5625. doi:10.1038/s41467-024-49853-2 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 464 17. Prudhomme M, Attaiech L, Sanchez G, Martin B, Claverys J-P. Antibiotic stress induces genetic 465 transformability in the human pathogen Streptococcus pneumoniae. Science. 2006;313: 89–92. 466 doi:10.1126/science.1127912 467 18. Domenech A, Slager J, Veening J-W. Antibiotic-Induced Cell Chaining Triggers Pneumococcal 468 Competence by Reshaping Quorum Sensing to Autocrine-Like Signaling. Cell Rep. 2018;25: 469 2390-2400.e3. doi:10.1016/j.celrep.2018.11.007 470 19. Slager J, Kjos M, Attaiech L, Veening J-W. Antibiotic-induced replication stress triggers bacterial 471 competence by increasing gene dosage near the origin. Cell. 2014;157: 395–406. 472 doi:10.1016/j.cell.2014.01.068 473 20. Maziero M, Lane D, Polard P, Bergé M. Fever-like temperature bursts promote competence 474 development via an HtrA-dependent pathway in Streptococcus pneumoniae. PLoS Genet. 475 2023;19: e1010946. doi:10.1371/journal.pgen.1010946 476 21. Luo P, Li H, Morrison DA. ComX is a unique link between multiple quorum sensing outputs and 477 competence in Streptococcus pneumoniae. Mol Microbiol. 2003;50: 623–633. 478 22. Stevens KE, Chang D, Zwack EE, Sebert ME. Competence in Streptococcus pneumoniae is 479 regulated by the rate of ribosomal decoding errors. mBio. 2011;2. doi:10.1128/mBio.00071-11 480 23. Chastanet A, Prudhomme M, Claverys JP, Msadek T. Regulation of Streptococcus pneumoniae 481 clp genes and their role in competence development and stress survival. J Bacteriol. 2001;183: 482 7295–7307. doi:10.1128/JB.183.24.7295-7307.2001 483 24. Liu X, Gallay C, Kjos M, Domenech A, Slager J, van Kessel SP, et al. High-throughput CRISPRi 484 phenotyping identifies new essential genes in Streptococcus pneumoniae. Mol Syst Biol. 485 2017;13. doi:10.15252/msb.20167449 486 25. Robertson GT, Ng W-L, Gilmour R, Winkler ME. Essentiality of clpX, but not clpP, clpL, clpC, or 487 clpE, in Streptococcus pneumoniae R6. J Bacteriol. 2003;185: 2961–2966. 488 doi:10.1128/JB.185.9.2961-2966.2003 489 26. Piotrowski A, Burghout P, Morrison DA. spr1630 is responsible for the lethality of clpX 490 mutations in Streptococcus pneumoniae. J Bacteriol. 2009;191: 4888–4895. 491 doi:10.1128/JB.00285-09 492 27. Martin B, Granadel C, Campo N, Hénard V, Prudhomme M, Claverys J-P. Expression and 493 maintenance of ComD-ComE, the two-component signal-transduction system that controls 494 competence of Streptococcus pneumoniae. Mol Microbiol. 2010;75: 1513–1528. 495 doi:10.1111/j.1365-2958.2010.07071.x 496 28. Johnston C, Mortier-Barriere I, Khemici V, Polard P. Fine-tuning cellular levels of DprA ensures 497 transformant fitness in the human pathogen Streptococcus pneumoniae. Mol Microbiol. 498 2018;109: 663–675. doi:10.1111/mmi.14068 499 29. van Opijnen T, Bodi KL, Camilli A. Tn-seq: high-throughput parallel sequencing for fitness and 500 genetic interaction studies in microorganisms. Nat Methods. 2009;6: 767–772. 501 doi:10.1038/nmeth.1377 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 502 30. Wuchty S, Rajagopala SV, Blazie SM, Parrish JR, Khuri S, Finley RL, et al. The Protein Interactome 503 of Streptococcus pneumoniae and Bacterial Meta-interactomes Improve Function Predictions. 504 mSystems. 2017;2. doi:10.1128/mSystems.00019-17 505 31. Simonsen S, Søgaard CK, Olsen JG, Otterlei M, Kragelund BB. The bacterial DNA sliding clamp, β- 506 clamp: structure, interactions, dynamics and drug discovery. Cell Mol Life Sci CMLS. 2024;81: 507 245. doi:10.1007/s00018-024-05252-w 508 32. Piotrowski A, Luo P, Morrison DA. Competence for genetic transformation in Streptococcus 509 pneumoniae: termination of activity of the alternative sigma factor ComX is independent of 510 proteolysis of ComX and ComW. J Bacteriol. 2009;191: 3359–3366. doi:10.1128/JB.01750-08 511 33. Calvo-Villamañán A, Ng JW, Planel R, Ménager H, Chen A, Cui L, et al. On-target activity 512 predictions enable improved CRISPR–dCas9 screens in bacteria. Nucleic Acids Res. 2020;48: 513 e64. doi:10.1093/nar/gkaa294 514 34. Bose B, Auchtung JM, Lee CA, Grossman AD. A conserved anti-repressor controls horizontal 515 gene transfer by proteolysis. Mol Microbiol. 2008;70: 570–582. doi:10.1111/j.1365- 516 2958.2008.06414.x 517 35. Bose B, Grossman AD. Regulation of horizontal gene transfer in Bacillus subtilis by activation of 518 a conserved site-specific protease. J Bacteriol. 2011;193: 22–29. doi:10.1128/JB.01143-10 519 36. Millman A, Melamed S, Leavitt A, Doron S, Bernheim A, Hör J, et al. An expanded arsenal of 520 immune systems that protect bacteria from phages. Cell Host Microbe. 2022;30: 1556-1569.e5. 521 doi:10.1016/j.chom.2022.09.017 522 37. Texier P, Bordes P, Nagpal J, Sala AJ, Mansour M, Cirinesi A-M, et al. ClpXP-mediated 523 Degradation of the TAC Antitoxin is Neutralized by the SecB-like Chaperone in Mycobacterium 524 tuberculosis. J Mol Biol. 2021;433: 166815. doi:10.1016/j.jmb.2021.166815 525 38. Lehnherr H, Yarmolinsky MB. Addiction protein Phd of plasmid prophage P1 is a substrate of 526 the ClpXP serine protease of Escherichia coli. Proc Natl Acad Sci U S A. 1995;92: 3274–3277. 527 doi:10.1073/pnas.92.8.3274 528 39. Prysak MH, Mozdzierz CJ, Cook AM, Zhu L, Zhang Y, Inouye M, et al. Bacterial toxin YafQ is an 529 endoribonuclease that associates with the ribosome and blocks translation elongation through 530 sequence-specific and frame-dependent mRNA cleavage. Mol Microbiol. 2009;71: 1071–1087. 531 doi:10.1111/j.1365-2958.2008.06572.x 532 40. Khemici V, Prudhomme M, Polard P. Tight Interplay between Replication Stress and 533 Competence Induction in Streptococcus pneumoniae. Cells. 2021;10: 1938. 534 doi:10.3390/cells10081938 535 41. López de Saro FJ, O’Donnell M. Interaction of the beta sliding clamp with MutS, ligase, and DNA 536 polymerase I. Proc Natl Acad Sci U S A. 2001;98: 8376–8380. doi:10.1073/pnas.121009498 537 42. Bernard P, Couturier M. Cell killing by the F plasmid CcdB protein involves poisoning of DNA- 538 topoisomerase II complexes. J Mol Biol. 1992;226: 735–745. doi:10.1016/0022-2836(92)90629- 539 x .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 540 43. Jiang Y, Pogliano J, Helinski DR, Konieczny I. ParE toxin encoded by the broad-host-range 541 plasmid RK2 is an inhibitor of Escherichia coli gyrase. Mol Microbiol. 2002;44: 971–979. 542 doi:10.1046/j.1365-2958.2002.02921.x 543 44. LeRoux M, Laub MT. Toxin-Antitoxin Systems as Phage Defense Elements. Annu Rev Microbiol. 544 2022;76: 21–43. doi:10.1146/annurev-micro-020722-013730 545 45. Aakre CD, Phung TN, Huang D, Laub MT. A Bacterial Toxin Inhibits DNA Replication Elongation 546 Through a Direct Interaction with the β Sliding Clamp. Mol Cell. 2013;52: 617–628. 547 doi:10.1016/j.molcel.2013.10.014 548 46. Charpentier X, Polard P, Claverys J-P. Induction of competence for genetic transformation by 549 antibiotics: convergent evolution of stress responses in distant bacterial species lacking SOS? 550 Curr Opin Microbiol. 2012;15: 570–576. doi:10.1016/j.mib.2012.08.001 551 47. Gasc AM, Sicard N, Claverys JP, Sicard AM. Lack of SOS repair in Streptococcus pneumoniae. 552 Mutat Res. 1980;70: 157–165. 553 48. Ernits K, Saha CK, Brodiazhenko T, Chouhan B, Shenoy A, Buttress JA, et al. The structural basis 554 of hyperpromiscuity in a core combinatorial network of type II toxin-antitoxin and related 555 phage defense systems. Proc Natl Acad Sci U S A. 2023;120: e2305393120. 556 doi:10.1073/pnas.2305393120 557 49. Lefevre JC, Claverys JP, Sicard AM. Donor deoxyribonucleic acid length and marker effect in 558 pneumococcal transformation. J Bacteriol. 1979;138: 80–86. 559 50. Martin B, Prudhomme M, Alloing G, Granadel C, Claverys JP. Cross-regulation of competence 560 pheromone production and export in the early control of transformation in Streptococcus 561 pneumoniae. Mol Microbiol. 2000;38: 867–878. 562 51. Horton RM. PCR-mediated recombination and mutagenesis. SOEing together tailor-made 563 genes. Mol Biotechnol. 1995;3: 93–99. doi:10.1007/BF02789105 564 52. Bergé M, Moscoso M, Prudhomme M, Martin B, Claverys J-P. Uptake of transforming DNA in 565 Gram-positive bacteria: a view from Streptococcus pneumoniae. Mol Microbiol. 2002;45: 411– 566 421. 567 53. Johnston CHG, Hope R, Soulet A-L, Dewailly M, De Lemos D, Polard P. The RecA-directed 568 recombination pathway of natural transformation initiates at chromosomal replication forks in 569 the pneumococcus. Proc Natl Acad Sci U S A. 2023;120: e2213867120. 570 doi:10.1073/pnas.2213867120 571 54. James P, Halladay J, Craig EA. Genomic libraries and a host strain designed for highly efficient 572 two-hybrid selection in yeast. Genetics. 1996;144: 1425–1436. 573 doi:10.1093/genetics/144.4.1425 574 55. Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. 575 (Academic Pr, 1998). 1998. 576 577 Figure Legends. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 578 Fig. 1: Depletion of clpX or clpP induces competence. 579 A. comCDE expression was monitored in strains carrying CRISPRi guide RNA targeting clpX 580 R4993 (left) or clpP R5203 (right). Cells were grown in C+Y medium at 37°C with or without 581 IPTG, as shown in the colour key. Culture was initiated 100 minutes prior to the first 582 measurement at time 0. comCDE expression values are expressed in relative light units per 583 OD (RLU/OD) in the top panel and the corresponding growth curves are reported as OD492nm 584 in the bottom panel. 585 B. Identical to A in ripA strains carrying CRISPRi guide RNA targeting respectively 4995 (clpX 586 depletion, left) and R5250 (clpP, right). 587 For the sake of clarity, only a single data set, representative of at least three independent 588 experiments carried out on different days, is presented. 589 590 Fig 2: ripA expression induces competence development 591 A. Schematic representation of ripA-ripB locus and its chromosomal surroundings. 592 B. comCDE expression was monitored in strains expressing ripA (R5139), ripB (R5138) or ripA- 593 ripB (R5140) under the control of an IPTG inducible promotor. Cells were grown in C+Y 594 medium at 37°C with increasing concentration of IPTG from the first measurement at time 0. 595 Top panels: luciferase activity expressed in relative light units per OD (RLU/OD). Bottom 596 panels: corresponding growth curves. 597 598 Fig 3: Trans-dependant suppression of RipA toxicity by RipB. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 599 A. Strains expressing ripA (R5139), ripB (R5138) or ripA-ripB (R5140) under the control of an 600 IPTG inducible promotor were serially diluted and spotted on CAT agar supplemented with 4% 601 horse blood containing different concentrations of IPTG. Plates were incubated at 37 °C 602 overnight. 603 For clarity, only a single spot test is presented and also represented as a histogram form in B 604 to reflect the efficiency of plating. Three independent spot tests were carried out on different 605 days displaying same results. 606 C. Same as B. for strains expressing ripA under the control of CepIIlac IPTG inducible promotor 607 at the cps locus (see material and methods), (R5239 and R5259). In strain R5259, ripB is 608 expressed under the control of CEPlac IPTG inducible promotor at the“ami locus”. 609 D. Percentage of transformation of strain wild type (wt) (R800) or invalidated for ripA (R4796) 610 with DNA carrying ripB invalidation (spectinomycin insertion). In the control, these two strains 611 were transformed with DNA carrying a point mutation conferring resistance to 612 streptomycin(rpsl41). 613 Fig 4: Mutations in DnaN confer different levels of resistance to RipA toxicity. 614 A. Strain R5086, expressing ripA under the control of an IPTG inducible promotor was transformed 615 with different error prone PCR fragments amplified from dnaN locus (see material and methods). 616 Transformant cells were plated on CAT agar supplemented with 4% horse blood containing 20µM 617 IPTG. CFU were numerated after overnight incubation at 37 °C. 618 B. Top panel. Schematic representation of the dnaN locus and its environment amplified by 619 primers MB313 and MB314. Red vertical lines indicate the position of mutated residues that .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 620 supress RipA toxicity. Positioning of the different suppressive mutations on the primary 621 structure (middle panel) and on the tertiary 3D structure of DnaN (bottom panel). 622 C. Efficiency of plating of strains expressing ripA under the control of an IPTG inducible 623 promotor in a dnaN wild type genetic background (R5139) or in a DnaN suppressor mutations 624 genetic background (R5165 to R5171). The strain expressing ripA-ripB under the control of 625 IPTG was used as a growth control (R5140). Cells were plated on CAT agar supplemented with 626 4% horse blood containing different concentrations of IPTG. For clarity, only a single spot test 627 is presented as a histogram. Three independent determinations were made on different days 628 displaying same results. 629 630 Fig 5: Suppressor mutations in DnaN reduce interactions with RipA in Yeast- 631 two-hybrid system. 632 A. Yeast-two-hybrid matrices produced to test interactions between RipA and DnaN. 633 B. Yeast-two-hybrid matrices performed to test interactions between RipA and DnaN 634 suppressive alleles. 635 636 Fig 6: Suppressor mutations in DnaN impair RipA dependant competence 637 development. 638 A. comCDE expression was monitored in strains expressing ripA under the control of an IPTG inducible 639 promotor in a wild type dnaN genetic background (R5138) or in strains carrying suppressive alleles of 640 dnaN, H183P (R5167), L185R (R5168), Y249N (R5169). Cells were grown in C+Y medium at 37°C with .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint 641 increasing amounts of IPTG from the first measurement at time 0. Top panels: luciferase activity 642 expressed in relative light units per OD (RLU/OD). Bottom panels: corresponding growth curves. 643 B. zoom in on transcription induction curves (A) in strains carrying suppressor mutations. 644 Fig 7: RipA toxicity allows propagation of competence to naive cells. 645 A. Phase contrast time-lapses of the strain expressing ripA under the control of an IPTG 646 inducible promotor (R5139) without or with 80µM of IPTG (respectively top and bottom 647 panels). 648 B. Ability of R5139 strain to generate colony after IPTG exposure. R5204 strain was exposed 649 or not to 80µM of IPTG for different times (X-axis), washed with fresh medium and plated on 650 Agar medium without IPTG. 651 C. Responsiveness of R5139 strain to CSP after one hour exposition or not to IPTG. 652 D. Mixed culture of strains expressing ripA under the control of an IPTG inducible promotor, 653 R5204 (comC+) or R5086 (comC0), with R825 strain used as competence reporter cells 654 through its PcomCDE::luc construct. Top panels: luciferase activity of R825 strain alone (grey) 655 or mixed with R5204 (red) or R5086 (blue). Middle panels: corresponding growth curves. For 656 clarity, only a single data set, representative of at least three independent determinations 657 made on different days, is presented. 658 Bottom panel. R825 transformation frequency that occurs during mixed cultures. 659 Fig 8: Working model of RipA action on competence development at single cell 660 and population level in S. pneumoniae. 661 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 2, 2025. ; https://doi.org/10.1101/2025.09.02.673660doi: bioRxiv preprint