Type VI secretion system killing by commensal Neisseria is influenced by expression of type four pili.

Type VI Secretion Systems (T6SSs) are widespread in bacteria and can dictate the development and organisation of polymicrobial ecosystems by mediating contact dependent killing. In Neisseria species, including Neisseria cinerea a commensal of the human respiratory tract, interbacterial contacts are mediated by Type four pili (Tfp) which promote formation of aggregates and govern the spatial dynamics of growing Neisseria microcolonies. Here, we show that N. cinerea expresses a plasmid-encoded T6SS that is active and can limit growth of related pathogens. We explored the impact of Tfp on N. cinerea T6SS-dependent killing within a colony and show that pilus expression by a prey strain enhances susceptibility to T6SS compared to a non-piliated prey, by preventing segregation from a T6SS-wielding attacker. Our findings have important implications for understanding how spatial constraints during contact-dependent antagonism can shape the evolution of microbial communities.

Introduction

The human microbiota is critical for the development of a healthy gastrointestinal immune system (Round and Mazmanian, 2009; Sommer and Bäckhed, 2013) and can also protect the host from invasion by pathogenic bacteria (Kamada et al., 2013). The microbes that carry out these important functions live as part of complex communities shaped by their fitness and ability to adapt to their environment, and which can be remodeled through mutualistic and antagonistic interactions (García-Bayona and Comstock, 2018; Little et al., 2008; Nadell et al., 2016). Competition for niche and host-derived resources has therefore driven the evolution in bacteria of an array of mechanisms to suppress growth of or kill neighbouring microbes. One mechanism, the Type VI Secretion System (T6SS), provides an effective strategy to eliminate competitors in a contact-dependent manner and is widespread in Gram negative bacteria from many different environments (Coulthurst, 2019). The T6SS is a contractile, bacteriophage-like nanomachine that delivers toxins into target organisms (Cianfanelli et al., 2016; Ho et al., 2014). T6SS-associated effectors possess a broad range of activities, including nucleases (Koskiniemi et al., 2013; Ma et al., 2014; Pissaridou et al., 2018), phospholipases (Flaugnatti et al., 2016; Russell et al., 2013), peptidoglycan hydrolases (Whitney et al., 2013), and pore-forming proteins (Mariano et al., 2019); each effector is associated with a cognate immunity protein to prevent self-intoxication and to protect against kin cells (Alcoforado Diniz et al., 2015; Unterweger et al., 2014). In pathogens such as Pseudomonas, Vibrio, Salmonella, and Shigella, the impact of the T6SS in pathogenesis and bacterial competition has been established in vitro and in some cases in vivo (Anderson et al., 2017; Sana et al., 2016). Commensal bacteria also harbour T6SSs, although how these systems combat pathogens has only been elucidated for Bacteroidetes in the intestinal tract (Russell et al., 2014); further studies are needed to gain a greater appreciation of how T6SSs in commensals influence microbial communities and pathogens in other niches.

The human nasopharynx hosts a polymicrobial community (Kumpitsch et al., 2019; Cleary and Clarke, 2017; Ramos-Sevillano et al., 2019), which can include the obligate human pathogen Neisseria meningitidis, as well as related but generally non-pathogenic, commensal Neisseria species (Diallo et al., 2016; Dorey et al., 2019; Gold et al., 1978; Knapp and Hook, 1988; Sheikhi et al., 2015). In vivo studies have demonstrated an inverse relationship between carriage of commensal Neisseria lactamica and N. meningitidis (Deasy et al., 2015), whereas in vitro studies have revealed that some commensal Neisseria demonstrate potentially antagonistic effects against their pathogenic relatives (Custodio et al., 2020; Kim et al., 2019). Commensal and pathogenic Neisseria species have also been shown to interact closely in mixed populations (Custodio et al., 2020; Higashi et al., 2011). Social interactions among Neisseria are mediated by surface structures known as Type IV pili (Tfp). These filamentous organelles enable pathogenic Neisseria to adhere to host cells (Nassif et al., 1993; Virji et al., 1991) and are crucial for microbe-microbe interactions and the formation of bacterial aggregates and microcolonies (Helaine et al., 2007; Higashi et al., 2007). In addition, Tfp interactions can dictate bacterial positioning within a community; non-piliated strains have been shown to be excluded to the expanding edge of colonies growing on solid media (Oldewurtel et al., 2015; Zöllner et al., 2017) while heterogeneity in pili, for example through post translational modifications, can alter how cells integrate into microcolonies (Zöllner et al., 2017).

Neisseria cinerea is one of the commensal Neisseria species that has been previously isolated from the upper respiratory tracts of adults and children (Knapp and Hook, 1988; Sheikhi et al., 2015). This species expresses Tfp that promote microcolony formation, can closely interact with N. meningitidis in a Tfp-dependent manner and impairs meningococcal association with human epithelial cells (Custodio et al., 2020). Here, whole genome sequence analysis revealed that the N. cinerea isolate used in our studies encodes a T6SS. Similarly, T6SS genes have been recently identified in other Neisseria spp. isolated from human throat swab cultures (Calder et al., 2020). Here, we provide the first description of a functional T6SS in Neisseria spp. We show that the N. cinerea T6SS is encoded on a plasmid and antagonises pathogenic relatives, N. meningitidis and Neisseria gonorrhoeae. Moreover, we examined whether Tfp influence the competitiveness of microbes in response to T6SS-mediated antagonism and demonstrate that T6SS–mediated competition is facilitated by Tfp in bacterial communities.

Results

N. cinerea 346T encodes a functional T6SS on a plasmid

We identified a single locus in N. cinerea isolate CCUG346T (346T) (https://www.ccug.se/strain?id=346) that encodes homologues of all 13 components that are necessary for a functional T6SS (Cascales and Cambillau, 2012), including genes predicted to encode canonical T6SS components Hcp and VgrG (Figure 1A and Supplementary file 1). We used T6SS-effector prediction software tools (Li et al., 2015) to search for putative effectors. In total we identified six putative effector and immunity genes, termed nte and nti for Neisseria T6SS effector/immunity, respectively.

Figure 1.

N. cinerea expresses a functional T6SS.

(A) Schematic representation of T6SS genes in N. cinerea 346T. Canonical tss nomenclature was used for genes in the T6SS cluster. (B) Map of the T6SS-associated genes encoded by the N. cinerea 346T plasmid. See also Figure 1—figure supplement 1. (C) Expression and secretion of Hcp by wild-type N. cinerea 346T (Wt) and the tssB mutant (ΔtssB). Hcp protein was detected in the whole cell lysates (W) and supernatants (S) by western blot analysis. For strain ΔtssB::tssBsfGFP, bacteria were grown in the presence (+) or absence (-) of 1 mM IPTG; molecular weight marker shown in kDa. RecA is only detected in whole cell lysates. (D) Survival of the prey, N. cinerea 27178A, after 4 and 24 h co-incubation with wild-type N. cinerea 346T or the T6SS mutant (ΔT6SS) at approximately 10:1 ratio, attacker:prey. The mean ± SD of three independent experiments is shown: ***p < 0.0001 using unpaired two-tailed Student’s t-test.

Overlapping PCR and sequencing confirms extra-chromosomally closed circular DNA fragment. A total of 25 PCR fragments (pink bars) were amplified from N. cinerea 346T gDNA to confirm the plasmid predicted by PacBio whole-genome sequencing. Yellow shows regions which were sequenced, and green indicates the overlapping amplified regions.

Figure 1—figure supplement 1.

The N. cinerea 346T T6SS is encoded on a plasmid.

Overlapping PCR and sequencing confirms extra-chromosomally closed circular DNA fragment. A total of 25 PCR fragments (pink bars) were amplified from N. cinerea 346T gDNA to confirm the plasmid predicted by PacBio whole-genome sequencing. Yellow shows regions which were sequenced, and green indicates the overlapping amplified regions.

Of note, all predicted T6SS-related orfs and Nte/Ntis in N. cinerea 346T were found to be encoded on a 108,141 bp plasmid, revealed by PacBio sequencing, and confirmed by PCR and sequencing. Nte/Nti 1–5 are encoded adjacent to the structural gene cluster, with Nte6/Nti6 encoded elsewhere in the plasmid (Figure 1B and Figure 1—figure supplement 1). No other PAAR, Rhs, or VgrG homologues were found outside the plasmid. Thus, our analysis reveals that the human commensal N. cinerea 346T harbours a plasmid-borne T6SS together with six putative effector-immunity pairs.

Contraction of the T6SS leads to Hcp secretion, a hallmark of a functional T6SS (Cascales and Cambillau, 2012). Therefore, to establish whether the N. cinerea T6SS is functional, we assessed Hcp levels in whole cell lysates and supernatants from wild-type N. cinerea and a ΔtssB mutant, based on previous work demonstrating that TssB is a component of the T6SS-tail-sheath required for contraction (Brackmann et al., 2018). A ΔT6SS mutant which lacks 10 core genes including hcp was analysed as a negative control. As expected, Hcp was detected in both fractions from the wild-type strain but not in the negative control strain (ΔT6SS mutant) (Figure 1C). Importantly, Hcp was present in cell lysates from the ΔtssB mutant, but not detected in cell supernatants, while Hcp secretion was restored by complementation of the ΔtssB mutant by chromosomal expression of TssB with a C-terminal sfGFP fusion (ΔtssB::tssB-sfGFP) (Figure 1C).

Next, we performed competition assays between N. cinerea 346T or the ΔT6SS mutant against N. cinerea 27178A which lacks a T6SS and Nte/Nti pairs identified in N. cinerea 346T. The survival of N. cinerea 27178A was reduced by around an order of magnitude following incubation with N. cinerea 346T compared with the ΔT6SS mutant (Figure 1D), confirming that the N. cinerea 346T T6SS is active during inter-bacterial competition.

Dynamic behaviour of the Neisseria T6SS in the presence of prey cells

We further analysed the activity of the T6SS by visualising assembly and contraction in N. cinerea 346TΔtssB::tssB-sfGFP; this strain exhibits comparable T6SS killing as wild-type N. cinerea 346T (Figure 2—figure supplement 1). Time-lapse microscopy revealed dynamic T6SS foci inside bacteria, with structures extending/contracting over seconds (Figure 2A and Figure 2—video 1) consistent with T6SS activity (Gerc et al., 2015; Ringel et al., 2017). To further confirm T6SS activity, we deleted the gene encoding the TssM homologue in strain 346TΔtssB::tssB-sfGFP, abolishing T6SS activity (Figure 2B and Figure 2—figure supplement 1) and confirmed that in the absence of TssM, fluorescent structures were rarely seen (< 5% of cells in the ΔtssM background, compared with > 60% in the strain expressing TssM; Figure 2C and Figure 2—video 2).

Figure 2—figure supplement 1.

N. cinerea T6SS with a TssB C-terminal sfGFP fusion is functional and activity is lost upon deletion of tssM.

Competition assay measuring the recovery of prey (N. meningitidis 8013) after 4 hr co-incubation with wild-type N. cinerea 346T (Wt) and specified mutants at ratio of 100:1 (attacker:prey). Data shown are the mean ± SD of two independent experiments: NS, not significant, **p < 0.005 using one-way ANOVA test for multiple comparisons.

Figure 2.

Visualisation of T6SS activity in N. cinerea.

(A) Assembly and contraction of the T6SS in N. cinerea; white arrows indicate contracting T6SSs. Time-lapse images of N. cinerea 346TΔtssB::tssBsfGFP (green) and prey N. cinerea 27178A_ sfCherry (red); the arrowhead shows a non-dynamic focus, scale bar, 1 µm. See also Figure 2—video 1. (B) Representative images of N. cinerea strains with the TssB::sfGFP fusion with (upper panels) or without (lower panels) TssM. Loss of fluorescent foci upon deletion of tssM indicates that foci correspond to active T6SSs. The scale bar represents 2 µm. (C) Quantification of TssB-sfGFP foci in different strains. T6SS foci were quantified using ‘analyse particle’ (Fiji) followed by manual inspection. For each strain, at least two images from gel pads were obtained on two independent occasions. Percentage of cells with 0, 1, or 2+ foci are shown and n = number of cells analysed. Data shown are mean ± SD of two independent experiments: ***p<0.0001 using two-way ANOVA test for multiple comparison. See also Figure 2—video 2.

Competition assay measuring the recovery of prey (N. meningitidis 8013) after 4 hr co-incubation with wild-type N. cinerea 346T (Wt) and specified mutants at ratio of 100:1 (attacker:prey). Data shown are the mean ± SD of two independent experiments: NS, not significant, **p < 0.005 using one-way ANOVA test for multiple comparisons.

N. cinerea 346TΔtssB::tssBsfGFP (green) and prey cells N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted on a 1% agarose PBS pad supplemented with 0.1 mM IPTG. The cells were imaged for 12 s with a rate of 1 image per second.

N. cinerea 346TΔtssB::tssBsfGFP (green) or N. cinerea 346TΔtssBΔtssM::tssBsfGFP (green) and prey cells N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted on a PBS 1% agarose pad supplemented with 0.1 mM IPTG. The cells were imaged for 200 s with a rate of one image per 10 s.

Figure 2—video 1.

Visualisation of N. cinerea T6SS contraction.

N. cinerea 346TΔtssB::tssBsfGFP (green) and prey cells N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted on a 1% agarose PBS pad supplemented with 0.1 mM IPTG. The cells were imaged for 12 s with a rate of 1 image per second.

Figure 2—video 2.

Visualisation of N. cinerea T6SS foci.

N. cinerea 346TΔtssB::tssBsfGFP (green) or N. cinerea 346TΔtssBΔtssM::tssBsfGFP (green) and prey cells N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted on a PBS 1% agarose pad supplemented with 0.1 mM IPTG. The cells were imaged for 200 s with a rate of one image per 10 s.

Finally, we examined whether T6SS assembly induces lysis of prey cells. We imaged N. cinerea 346TΔtssB::tssB-sfGFP with N. cinerea 27178 expressing sfCherry on gel pads with SYTOX Blue as an indicator of target cell permeability (Ringel et al., 2017). Interestingly, we detected increased SYTOX staining of prey cells immediately adjacent to predator bacteria displaying T6SS activity (Figure 3 and Figure 3—video 1), indicating that the N. cinerea T6SS induces cell damage and lysis of its prey.

Figure 3.

N. cinerea T6SS induces lysis in prey bacteria.

(A) Assembly of T6SSs and prey lysis. Time-lapse series of merged images with phase contrast, N. cinerea 346T ΔtssB+tssBsfGFP (green), and N. cinerea 27178A sfCherry (red); scale bar, 1 µm. (B) Top row shows merged images of GFP (green, indicating T6SS assembly/contraction), mCherry (red, prey strain), and SYTOX Blue (cyan, showing membrane permeabilisation) channels. The bottom row arrows highlight a prey cell losing membrane integrity (increase in SYTOX Blue staining inside cells) arrows. Representative image from two biological repeats. Scale bars represent 1 µm. See also Figure 3—video 1.

N. cinerea 346TΔtssB::tssBsfGFP (green) and N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted onto agarose pads supplemented with 0.1 mM IPTG and the cell-impermeable DNA stain SYTOX Blue (0.5 μM) as an indicator for loss of membrane integrity. The cells were imaged for 10 min with an image taken every 10 s.

Figure 3—video 1.

N. cinerea T6SS elicits prey lysis.

N. cinerea 346TΔtssB::tssBsfGFP (green) and N. cinerea 27178A_sfCherry (red) were mixed at a ratio of 1:1 and spotted onto agarose pads supplemented with 0.1 mM IPTG and the cell-impermeable DNA stain SYTOX Blue (0.5 μM) as an indicator for loss of membrane integrity. The cells were imaged for 10 min with an image taken every 10 s.

N. cinerea T6SS effectors are functional toxin/immunity pairs

To characterise the six putative T6SS effectors identified, we first used sequence analysis to determine their predicted domain structure. As shown in Figure 4, all Ntes contain a conserved Rhs domain, frequently associated with polymorphic toxins (Busby et al., 2013), and a C-terminal region with predicted activities previously described in T6SS effectors (Alcoforado Diniz et al., 2015). Nte1 contains an N-terminal PAAR motif, which can associate with the VgrG tip of the T6SS (Shneider et al., 2013) and C-terminal phospholipase domain (cd00618). Nte2 also contains an N-terminal PAAR domain and has a predicted RNase domain (pfam15606) in its C-terminal region. Using BLASTp analysis and the PAAR-like domain sequence from Nte1 as the query sequence, we did not identify any other PAAR encoding genes in the WGS of 346T. Nte3 is a putative endonuclease of the HNH/Endo VII family with conserved LHH (pfam14411). Nte4 contains a GIY-YIG nuclease domain (cd00719) and Nte5 is predicted to be an HNH/endo VII nuclease with conserved AHH (pfam14412), with Nte6 predicted to contain an HNHc endonuclease active site (cd00085).

Figure 4.

Predicted domain organisation of N. cinerea 346T T6SS effectors.

Schematic representation of bioinformatically identified effectors in N. cinerea 346T. The domain organisation of the putative effectors is shown, with PAAR motifs indicated in orange, Rhs domains in blue, endonuclease motifs (Tox-LHH pfam14411; Tox-GIY/YIG cd00719; Tox-AHH pfam14412; and Tox-HNHc cd00085) in green, RNase (Ntox34, pfam15606) motif in yellow and the phospholipase (PLA2_like, cd00618) domain in red. The conserved domains annotation was retrieved from the NCBI database.

To further characterise the possible effector/immunity pairs, we expressed each Nte alone or with its corresponding Nti using an inducible expression plasmid in E. coli (Figure 5). We were only able to clone wild-type Nte6 in presence of its immunity protein, so Nte6R1300S was used to analyse toxicity of this protein. In addition, as Nte1 encodes a predicted phospholipase that should be active against cell membranes (Flaugnatti et al., 2016), we targeted the putative phospholipase domain of Nte1 to the periplasm by fusing it to the PelB signal sequence (Singh et al., 2013); cytoplasmic expression of the Nte1 phospholipase domain does not inhibit bacterial growth (Figure 5—figure supplement 1). All Ntes are toxic, with their expression leading to decreased viability and reduced optical density (OD) of E. coli cultures compared to empty vector controls; toxicity was counteracted by co-expression of the corresponding Nti.

Figure 5.

Putative N. cinerea T6SS effectors are toxic in E. coli.

(A) Arabinose (Ara)-induced expression of T6SS effector Nte1 in periplasm of E. coli leads to reduction in CFU and OD at 600 nm (OD600). Co-expression of putative immunity Nti1 restores growth to levels of strain with empty vector (pBAD33). See also Figure 5—figure supplement 1. (B-E) Cytoplasmic expression of putative effectors Nte2-5 without cognate immunity reduces growth and survival of E. coli. (F) Expression of Nte6R1300S reduces viability and growth when expressed in E. coli. Expression of Nti6 with Nte6 does not impact growth. In (A-F) number of CFU at 120 min post-induction are shown. Data shown are the mean ± SD of three independent experiments: NS, not significant, ***p<0.0001, *p<0.05 using two-way ANOVA test for multiple comparison. Images of colonies for Nte1 and Nte6 are composite as strains were spotted to different areas of the same plates.

(A) Toxicity of Nte1 with or without the PelB signal sequence (ssPelB) following expression in E. coli. Expression was induced with L-arabinose at 60 min and bacterial growth was monitored by measuring the OD600 of cultures. A reduction on OD600 was only observed when Nte1 was expressed with ssPelB. (B) Samples collected after addition of arabinose were plated to media containing glucose to repress toxin expression and enumerated. Data shown are mean ± SD of three independent experiments: NS, not significant, ***p<0.0001, *p<0.05 using two-way ANOVA test for multiple comparison.

Figure 5—figure supplement 1.

N cinerea putative T6SS effector Nte1 requires a PelB signal sequence for toxicity in E. coli.

(A) Toxicity of Nte1 with or without the PelB signal sequence (ssPelB) following expression in E. coli. Expression was induced with L-arabinose at 60 min and bacterial growth was monitored by measuring the OD600 of cultures. A reduction on OD600 was only observed when Nte1 was expressed with ssPelB. (B) Samples collected after addition of arabinose were plated to media containing glucose to repress toxin expression and enumerated. Data shown are mean ± SD of three independent experiments: NS, not significant, ***p<0.0001, *p<0.05 using two-way ANOVA test for multiple comparison.

Commensal Neisseria T6SS kills human pathogens

We next investigated whether N. cinerea can deploy the T6SS to antagonise the related pathogenic species, N. meningitidis and N. gonorrhoeae. We performed competition assays with three N. meningitidis strains (belonging to different lineages and expressing different polysaccharide capsules i.e., serogroup B or C), and a strain of N. gonorrhoeae. N. cinerea 346T caused between a 50- to 100-fold decrease in survival of the meningococcus compared with the ΔT6SS strain, irrespective of lineage or serogroup (Figure 6A) and an approximately fivefold reduction in survival of the gonococcus (Figure 6B). We also investigated whether the meningococcal capsule protects against T6SS assault. Using a capsule-null strain (ΔsiaD) in competition assays with wild-type N. cinerea 346T or the T6SS mutant, we found reduced survival of the unencapsulated mutant compared to the wild-type (Figure 6C). Therefore, the meningococcal capsule protects bacteria against T6SS attack.

Figure 6.

N. cinerea T6SS is active against pathogenic N. meningitidis and N. gonorrhoeae.

(A) Recovery of wild-type N. meningitidis (Nm8013, NmMC58, NmS3) after 4 hr co-incubation with N. cinerea 346T wild-type (Wt) or the T6SS mutant (ΔT6SS) at approx. 100:1 attacker:prey ratio. (B) Recovery of wild-type N. gonorrhoeae (FA1090) after 4 hr co-incubation with N. cinerea 346T wild-type (Wt) or the T6SS mutant (ΔT6SS) at approximately 10:1 attacker:prey ratio. (C) Unencapsulated N. meningitidis (NmMC58ΔsiaD) is more susceptible to T6SS-mediated killing than wild-type N. meningitidis. Recovery of NmMC58 or the capsule-null mutant (NmMC58ΔsiaD) after 4 hr co-culture with N. cinerea 346T (Wt) or a T6SS-deficient mutant (ΔT6SS) at ratio of approximately 100:1, attacker:prey. Data shown are the mean ± SD of three independent experiments: NS, not significant, ***p < 0.0001, **p < 0.001 using unpaired two-tailed Student’s t-test for pairwise comparison (B and C) or one-way ANOVA test for multiple comparison (A).

Spatial segregation driven by type IV pili dictates prey survival against T6SS assault

Despite the potency of the T6SS in Neisseria warfare, this nanomachine operates when bacteria are in close proximity, so we hypothesised that Tfp, which are critical for the formation of Neisseria microcolonies and organisation of bacterial communities (Higashi et al., 2007; Mairey et al., 2006; Oldewurtel et al., 2015; Zöllner et al., 2017), could influence T6SS-mediated antagonism. To test this, we constructed fluorophore expressing ‘prey’ strains (i.e. sfCherry-expressing 346TΔnte/i3-5; Figure 7—figure supplement 1) with and without Tfp. Prey strains were mixed with piliated attacker strain N. cinerea 346T expressing sfGFP at a 1:1 ratio on solid media, and the spatiotemporal dynamics of bacterial growth examined by time-lapse stereo microscopy over 24 hr, while the relative proportion of each strain was analysed by flow cytometry at 24 hr (Figure 7—figure supplement 2). As expected based on previous observations of Tfp-mediated cell sorting in Neisseria (Oldewurtel et al., 2015; Zöllner et al., 2017), the non-piliated prey strain (346TΔnte/i3-5ΔpilE1/2_sfCherry; red) segregates to the periphery of the colony, in this location the prey strain escapes T6SS-mediated assault and dominates the expanding colony (Figure 7A and Figure 7—video 1). In contrast, when the prey is piliated, pilus-mediated cell interactions prevent displacement of cells to the expanding front (Oldewurtel et al., 2015; Pönisch et al., 2018; Zöllner et al., 2017), so the susceptible strain (Tfp-expressing 346TΔnte/i3-5_sfCherry Tfp+, red) is outcompeted by the T6SS+ strain (Tfp-expressing 346T_sfGfp Tfp+, green) (Figure 7B and Figure 7—video 2). When both strains are piliated and immune to T6SS attack, there is no dominance of either strain (Figure 7C and Figure 7—video 3). Assessment of the relative recovery of piliated and non-piliated prey in competition assays also supported the observation that the piliation status of the prey impacts survival against T6SS (Figure 7D and Figure 7—figure supplement 3). These results highlight that Tfp influence the outcome of T6SS-mediated antagonism through structuring and partitioning bacteria in mixed microcolonies.

Figure 7—figure supplement 1.

N. cinerea 346TΔnte/i3-5 prey is susceptible to T6SS-killing by wild-type N. cinerea 346T and fluorophore expressing mutants have comparable growth.

(A) Competition assay measuring the recovery of the indicated N. cinerea 346TΔNte/i3-5 mutant strain after 4 hr of co-incubation wild-type N. cinerea 346T (Wt) or a tssM-deficient mutant (ΔtssM) at ratio of 10:1 (attacker: prey). Data shown are the mean ± SD of three independent experiments performed in triplicate: *p<0.05 using unpaired two-tailed Student’s t-test. (B) N. cinerea 346T WT and mutant strains expressing fluorophores were grown in liquid BHI media for 8 hr at 37°C with 5% CO2. Growth was monitored by measuring the OD600 of cultures. Data are representative of two independent experiments: NS, not significant, using two-way ANOVA test for multiple comparison.

Figure 7—figure supplement 2.

Flow cytometry analysis of relative proportion of attacker and prey strains.

(A) Representative FACS plots corresponding to analysis of fixed input suspensions of indicated attacker and prey strains mixed at a 1:1 ratio. Plots revealed a population of dual labelled (sfCherry+sfGFP+) particles which likely correspond to associated sfGFP+ and sfCherry+ cells, although additional experiments are required to confirm this. Data from two experiments indicated no significant difference in the % of total population between sfGFP+ and sfCherry+ strains. (B) Total bacteria in mixed colonies grown on agar plates for 24 hr (Figure 7A-C) were collected, fixed, and analysed by Flow cytometry. Plots are representative of two independent experiments and show relative proportion of sfCherry+ (prey), sfGFP+ (attacker), non-fluorescent (NF), and dual-labelled populations (sfCherry+sfGFP+) populations. To determine whether the dual labelled population could affect interpretation of the relative overall proportions of attacker (sfGFP+) and prey (sfCherry+) cells, the fluorescence of this population was analysed further. The Median fluorescence Intensity (MFI) for sfGFP and sfCherry of this population was determined. MFI values were normalised relative to the MFI of the sfGFP or sfCherry single populations respectively, to allow a comparison of the red and green fluorescence of this dual labelled population. This revealed no significant difference for all samples except for the Wt attacker vs Tfp- prey, in which we observed a slight increase towards the dominant population, that is, MFIsfCherry>MFIsfGFP of the sfGFP+sfCherry+ population. Therefore, dual fluorescence populations were excluded from analysis and graphs indicate the percentage of total population for sfGFP+ and sfCherry+ populations as mean ± SD of two independent experiments with duplicate samples. When WT:WT are mixed (both strains are T6SS+ and Tfp+) both were detected in approximately equal proportions (red 346T_sfCherry 29 ± 1%, green 346T_gfp 31 ± 1%, p=0.046). In the context of piliated attacker and prey strains, the attacker cells constitute the majority of the population (red prey 346TΔNte/i3-5 9 ± 2%, green attacker 346T_gfp 65 ± 5%, p<0.0001); however, when non-piliated, the prey strain (red) dominates, despite being susceptible to T6SS-mediated attack (red prey 346TΔNte/i3-5ΔpilE1/2_sfCherry 56 ± 4%, green attacker 346T_gfp 13 ± 3%; p<0.0001). NS, not significant, ***p < 0.0001 using unpaired two-tailed Student’s t-test.

Figure 7.

Attacker and prey piliation promotes T6SS killing.

(A) Fluorescence microscopy images taken at specific times after inoculation of mixed (1:1 ratio) bacterial colonies. A T6SS-susceptible, non-piliated prey strain (346TΔnte/i3-5ΔpilE1/2_sfCherry, red) migrates to the expanding edge of the colony over time, segregating from the T6SS+ attacker strain (N. cinerea 346T_gfp, green) and dominating the expanding population. See also Figure 7—video 1, (B) The same susceptible prey strain but expressing pili does not segregate, and after 24 hr is outcompeted by the piliated T6SS+ attacker. See also Figure 7—video 2. (C) The non-T6SS-susceptible, piliated prey strain (346T_sfCherry, red) and piliated attacker strain (346T_sfGfp, green) do not segregate, but due to immunity against T6SS attack, no dominance is observed. Images of colonies are representative of three independent experiments. See also Figure 7—video 3. Scale bar, 500 µm. Expanding colony edge images are stills at indicated times from time-lapse imaging performed on one occasion. Scale bar 100 µm. Flow cytometry data are presented in Figure 7—figure supplement 2. (D) The influence of piliation on T6SS killing. Recovery of non-piliated and piliated prey strains after 24 hr co-culture with N. cinerea 346T (Wt) and a tssM-deficient mutant (ΔtssM) at ratio of approx. 10:1, attacker:prey. Relative survival is defined as the fold change in recovery of prey following incubation with wild-type attacker N. cinerea compared to N. cinerea ΔtssM. Data shown are the mean ± SD of three independent experiments: **p < 0.01 using unpaired two-tailed Student’s t-test for pairwise comparison. See also Figure 7—figure supplement 3.

(A) Competition assay measuring the recovery of the indicated N. cinerea 346TΔNte/i3-5 mutant strain after 4 hr of co-incubation wild-type N. cinerea 346T (Wt) or a tssM-deficient mutant (ΔtssM) at ratio of 10:1 (attacker: prey). Data shown are the mean ± SD of three independent experiments performed in triplicate: *p<0.05 using unpaired two-tailed Student’s t-test. (B) N. cinerea 346T WT and mutant strains expressing fluorophores were grown in liquid BHI media for 8 hr at 37°C with 5% CO2. Growth was monitored by measuring the OD600 of cultures. Data are representative of two independent experiments: NS, not significant, using two-way ANOVA test for multiple comparison.

(A) Representative FACS plots corresponding to analysis of fixed input suspensions of indicated attacker and prey strains mixed at a 1:1 ratio. Plots revealed a population of dual labelled (sfCherry+sfGFP+) particles which likely correspond to associated sfGFP+ and sfCherry+ cells, although additional experiments are required to confirm this. Data from two experiments indicated no significant difference in the % of total population between sfGFP+ and sfCherry+ strains. (B) Total bacteria in mixed colonies grown on agar plates for 24 hr (Figure 7A-C) were collected, fixed, and analysed by Flow cytometry. Plots are representative of two independent experiments and show relative proportion of sfCherry+ (prey), sfGFP+ (attacker), non-fluorescent (NF), and dual-labelled populations (sfCherry+sfGFP+) populations. To determine whether the dual labelled population could affect interpretation of the relative overall proportions of attacker (sfGFP+) and prey (sfCherry+) cells, the fluorescence of this population was analysed further. The Median fluorescence Intensity (MFI) for sfGFP and sfCherry of this population was determined. MFI values were normalised relative to the MFI of the sfGFP or sfCherry single populations respectively, to allow a comparison of the red and green fluorescence of this dual labelled population. This revealed no significant difference for all samples except for the Wt attacker vs Tfp- prey, in which we observed a slight increase towards the dominant population, that is, MFIsfCherry>MFIsfGFP of the sfGFP+sfCherry+ population. Therefore, dual fluorescence populations were excluded from analysis and graphs indicate the percentage of total population for sfGFP+ and sfCherry+ populations as mean ± SD of two independent experiments with duplicate samples. When WT:WT are mixed (both strains are T6SS+ and Tfp+) both were detected in approximately equal proportions (red 346T_sfCherry 29 ± 1%, green 346T_gfp 31 ± 1%, p=0.046). In the context of piliated attacker and prey strains, the attacker cells constitute the majority of the population (red prey 346TΔNte/i3-5 9 ± 2%, green attacker 346T_gfp 65 ± 5%, p<0.0001); however, when non-piliated, the prey strain (red) dominates, despite being susceptible to T6SS-mediated attack (red prey 346TΔNte/i3-5ΔpilE1/2_sfCherry 56 ± 4%, green attacker 346T_gfp 13 ± 3%; p<0.0001). NS, not significant, ***p < 0.0001 using unpaired two-tailed Student’s t-test.

(A) Data shown in Figure 7D presented as total recovery in CFU/ml of non-piliated (346TΔnte/i3-5ΔpilE1/2 as Tfp-prey) and piliated prey (346TΔnte/i3-5 as Tfp+prey) strains after 24 hr co-culture with N. cinerea 346T (Wt) and a tssM-deficient mutant (ΔtssM) at ratio of 10:1, attacker:prey. (B) Data shown in Figure 7D presented as percentage survival of prey (Tfp-prey and Tfp+prey) relative to input levels of each prey strain following incubation with wild-type attacker N. cinerea. Data shown are the mean ± SD of three independent experiments: NS, not significant *p<0.05 and **p<0.01 using or two-way ANOVA test for multiple comparison.

Strains were spotted at a ratio of 1:1. Colonies were imaged every 2 hr between 4 and 16 hr post inoculation. Over time a population of non-piliated prey segregates to the edge, escape T6SS assault and dominates the growing colony.

Strains were spotted at a ratio of 1:1. Colonies were imaged every 2 hr between 4 and 16 hr post inoculation. Over time the T6SS expressing strain dominates the colony.

N. cinerea 346T_gfp (green) and N. cinerea 346T_sfCherry (red). Strains were spotted at a ratio of 1:1. Colonies were imaged between 4 and 16 hr post inoculation. Images obtained every 2 hr. No dominance of either strain is observed.

Figure 7—video 1.

Growing edge of colonies with a piliated attacker N. cinerea 346T_gfp, (green) and non-piliated prey 346TΔnte/i3-5ΔpilE1/2_sfCherry (red).

Strains were spotted at a ratio of 1:1. Colonies were imaged every 2 hr between 4 and 16 hr post inoculation. Over time a population of non-piliated prey segregates to the edge, escape T6SS assault and dominates the growing colony.

Figure 7—video 2.

Growing edge of colonies with a piliated attacker N. cinerea 346T_gfp, (green) and piliated prey 346TΔnte/i3-5_sfCherry (red).

Strains were spotted at a ratio of 1:1. Colonies were imaged every 2 hr between 4 and 16 hr post inoculation. Over time the T6SS expressing strain dominates the colony.

Figure 7—video 3.

Growing edge of colonies with two wild-type strains.

N. cinerea 346T_gfp (green) and N. cinerea 346T_sfCherry (red). Strains were spotted at a ratio of 1:1. Colonies were imaged between 4 and 16 hr post inoculation. Images obtained every 2 hr. No dominance of either strain is observed.

Figure 7—figure supplement 3.

CFU and % survival data for Tfp+/- strains.

(A) Data shown in Figure 7D presented as total recovery in CFU/ml of non-piliated (346TΔnte/i3-5ΔpilE1/2 as Tfp-prey) and piliated prey (346TΔnte/i3-5 as Tfp+prey) strains after 24 hr co-culture with N. cinerea 346T (Wt) and a tssM-deficient mutant (ΔtssM) at ratio of 10:1, attacker:prey. (B) Data shown in Figure 7D presented as percentage survival of prey (Tfp-prey and Tfp+prey) relative to input levels of each prey strain following incubation with wild-type attacker N. cinerea. Data shown are the mean ± SD of three independent experiments: NS, not significant *p<0.05 and **p<0.01 using or two-way ANOVA test for multiple comparison.

We also considered that Tfp might contribute to increased prey survival through mechanisms other than the spatial organisation of strains within bacterial colonies. For example, Tfp-Tfp interactions are known to contribute to kin recognition (Adams et al., 2019) and promote aggregation (Hélaine et al., 2005), which could impact T6SS activity by anchoring neighbouring cells in closer proximity. Alternatively, Tfp may have a role in provoking T6SS activity, similar to the T6SS response to exogenous T6SS (Basler et al., 2013) or cell envelope perturbations in Pseudomonas (Ho et al., 2013; Stolle et al., 2021). To address this, we compared the survival of piliated and non-piliated prey, in presence of Tfp+ or Tfp- attacker strain. Where pili are expressed on both or neither strain, segregation should not occur, enabling comparative analysis of the impact of Tfp on prey survival, independent of segregation. Competition assays to assess prey survival revealed that increased prey survival is only observed when the prey is non-piliated, but not when the attacker is non-piliated. Moreover, prey survival was equivalent when attacker and prey either both have, or both lack Tfp (Figure 8A and Figure 8—figure supplement 1). These data support the idea that the enhanced prey survival is due to segregation of the non-piliated prey from the piliated attacker, allowing the prey to achieve a favourable position for outgrowth at the edge of the colony.

Figure 8.

Tfp loss influences prey survival.

(A) Role of Tfp in the attacker and prey population during competition. Recovery of non-piliated (346TΔnte/i3-5ΔpilE1/2_sfCherry) and piliated prey (346TΔnte/i3-5_sfCherry) strains after 24 hr co-culture with piliated N. cinerea 346T (346T_sfGfp) and non-piliated attacker (346TΔpilE1/2_sfGfp) strains at ratio of approx. 10:1, attacker:prey. Data shown are the mean ± SD of three independent experiments: NS, not significant, **p<0.001 using one-way ANOVA test for multiple comparison. See also Figure 8—figure supplement 1. (B) Fluorescence microscopy images taken at 24 hr after inoculation of mixed (1:1 ratio) bacterial colonies. A T6SS-susceptible, piliated prey strain (Tfp+ prey, 346TΔnte/i3-5_sfCherry, red) does not segregate, and after 24 hr is outcompeted by the piliated T6SS+ attacker (Tfp+ attacker, 346T_sfGfp, green). The same prey, but non-piliated (Tfp- prey, 346TΔnte/i3-5ΔpilE1/2_sfCherry, red), segregates from the piliated T6SS+attacker strain (Tfp+ attacker, 346T_sfGfp, green) and dominates the edge of the colony. When the prey is piliated (Tfp+ prey, 346TΔnte/i3-5_sfCherry, red) and attacker is non-piliated (Tfp- attacker, 346TΔpilE1/2_sfGfp, green), the non-piliated attacker population segregates to the edge and dominates the outer region of the colony. In a mixed population with a non-piliated prey (Tfp- prey, 346TΔnte/i3-5ΔpilE1/2_sfCherry, red) and a non-piliated attacker (Tfp- attacker, 346TΔpilE1/2_sfGfp, green), the prey does not segregate from attacker and attacker and prey form expanding sectors in the region of outgrowth at the colony edge. Images of colonies are representative of three independent experiments. Scale bar, 500 µm. See also Figure 8—figure supplement 2.

Data shown in Figure 8A presented as percentage of prey (Tfp-prey and Tfp+prey) relative to input, following incubation with attacker N. cinerea (Tfp- attacker and Tfp+ attacker). Data shown are the mean ± SD of three independent experiments: NS, not significant ***p < 0.0001 using one-way ANOVA test for multiple comparison.

Additional examples of fluorescence microscopy images taken at 24 hr after inoculation of mixed (1:1 ratio) bacterial colonies. Gfp-expressing attacker, with Tfp (Tfp+, 346T_sfGfp, green) or without Tfp (Tfp-, 346T_sfGfpΔpilE1/2, green) and sfCherry-expressing T6SS susceptible prey strains with Tfp (346TΔnte/i3-5_sfCherry, red) or without Tfp (346TΔnte/i3-5ΔpilE1/2_sfCherry, red) were mixed as indicated.

Figure 8—figure supplement 1.

% survival data for prey strains +/-Tfp.

Data shown in Figure 8A presented as percentage of prey (Tfp-prey and Tfp+prey) relative to input, following incubation with attacker N. cinerea (Tfp- attacker and Tfp+ attacker). Data shown are the mean ± SD of three independent experiments: NS, not significant ***p < 0.0001 using one-way ANOVA test for multiple comparison.

Figure 8—figure supplement 2.

Fluorescence microscopy images of colonies of piliated (Tfp+/Tfp+) or non-piliated (Tfp-/Tfp-) attacker and prey strains.

Additional examples of fluorescence microscopy images taken at 24 hr after inoculation of mixed (1:1 ratio) bacterial colonies. Gfp-expressing attacker, with Tfp (Tfp+, 346T_sfGfp, green) or without Tfp (Tfp-, 346T_sfGfpΔpilE1/2, green) and sfCherry-expressing T6SS susceptible prey strains with Tfp (346TΔnte/i3-5_sfCherry, red) or without Tfp (346TΔnte/i3-5ΔpilE1/2_sfCherry, red) were mixed as indicated.

We also used fluorescently labelled piliated and non-piliated attacker and prey to observe the same strains in mixed colonies as previously. Given that Tfp heterogeneity impacts segregation within a colony (Oldewurtel et al., 2015; Pönisch et al., 2018; Zöllner et al., 2017), when one of the two strains lacks pili, the non-piliated strain segregates and outgrowth is clearly visible at the edge of the colony (Figure 8B and Figure 7A). In the case where the non-piliated attacker segregates and dominates the colony edge, this appears to prevent any expansion of the prey, consistent with the lack of enhanced prey survival observed in competition assays. Interestingly, comparison of colonies where both attacker and prey are piliated with colonies where neither strain expresses Tfp revealed differences in prey expansion at the edge of the colony. In colonies with both strains lacking Tfp, we observe higher abundance of expanding sectors of the emerging prey population compared to colonies where attacker and prey cells are piliated (Figure 8B and Figure 8—figure supplement 2). One possible explanation is that Tfp interactions at the expanding edge bring adjacent attacker and prey cells into close proximity and thus result in a more effective reduction in the prey compared to when neither is piliated. Although competition assays did not reveal any difference in levels of prey survival when neither or both attacker and prey are piliated, this could be due to methodological limitations which mean that this very local effect is not detected at the population level. Further work is therefore necessary to explore the contribution of Tfp to T6SS-mediated attack beyond the impact on spatial reorganisation within a colony. Overall, data presented here confirm that Tfp influence the outcome of T6SS-mediated antagonism.

Discussion

Here, we identified a T6SS in a commensal Neisseria spp. which can kill T6SS-deficient N. cinerea isolates and the related pathogens, N. meningitidis, with which it shares an ecological niche (Knapp and Hook, 1988), and N. gonorrhoeae. Of note, the N. cinerea T6SS is encoded on a large plasmid, with structural genes for the single T6SS apparatus clustered in one locus, similar to other T6SSs (Anderson et al., 2017; Liaw et al., 2019; Sana et al., 2016). To date, plasmid encoded T6SSs have only been described in Campylobacter species (Marasini and Fakhr, 2016), with this plasmid T6SS mobilised via conjugation (Marasini et al., 2020). Although other small plasmids have been reported in N. cinerea (Knapp et al., 1984; Roberts, 1989) and N. cinerea can be a recipient of N. gonorrhoeae plasmids (Genco et al., 1984), it is not yet known whether T6SS plasmids are widespread among Neisseria, or whether the plasmid can be mobilised by conjugation or transformation. Interestingly, in Acinetobacter baylyi, T6SS-induced prey cell lysis contributes to acquisition of plasmids from target cells (Ringel et al., 2017). Therefore, it will be interesting to see whether other Neisseria species with T6SS genes (Marri et al., 2010) harbour T6SS-expressing plasmids.

In total six genes encoding putative effectors were identified based on their proximity to the T6SS locus, their pairwise arrangement with genes encoding proteins with homology to immunity proteins and the presence of conserved domains such as PAAR and Rhs domains in the predicted proteins (Alcoforado Diniz et al., 2015). Our bioinformatic predictions suggest Nte3, 4, 5, and 6 may be cargo effectors while Nte1 and Nte2 are more typical of specialised or evolved effectors with integral PAAR domains at their N-termini (Durand et al., 2014). Of note nte5 encodes a protein with a shorter Rhs domain compared to the other five putative effectors (319AA compared to 695-735AA), raising the possibility that this may represent an orphan Rhs-CT, as described in other genomes (Kirchberger et al., 2017). Based on previous work, Hcp, or VgrG could be responsible for delivery of effectors encoded nearby in the T6SS locus (Hachani et al., 2014), with or without the involvement of the adjacent DUF2169 family protein (Figure 1). As nte/nti6 are not part of the T6SS locus and are encoded elsewhere on the plasmid, Nte/Nti6 may not be associated with the T6SS. Thus, although our bioinformatic analysis and protein expression in E. coli suggests Nte1-6 as effectors, additional experimentation will be needed to further confirm their contributions to T6SS activity and killing, and to demonstrate direct secretion via the T6SS.

Examination of N. cinerea T6SS activity revealed several interesting features. Microscopy demonstrated that T6SS attack (tit-for-tat) is not required to provoke firing of the system. Instead, the T6SS appears to be constitutively active in N. cinerea (Figure 2). Furthermore, the system is capable of inducing lysis of prey bacteria (Figure 3). The consequences of T6SS attack are determined by the repertoire and activities of effectors, and their site of delivery. Many different effector activities have been proposed including lipases, peptidoglycan hydrolases, metalloproteases, and nucleases (Lewis et al., 2019). Effector activities can result in target cell lysis to varying degrees (Ringel et al., 2017; Smith et al., 2020). Of the six Ntes we identified, lysis could be mediated by Nte1 which harbours a putative phospholipase domain in the C-terminus. Alternatively, a combination of effectors might be needed to elicit prey lysis.

Polysaccharide capsules are largely thought to provide bacteria with a strategy for evading host immune killing (Lewis and Ram, 2014). Here, we found that the meningococcal capsule has an alternative role in defence against other bacteria. Meningococcal strains lacking a capsule were at a significant disadvantage in the face of a T6SS-expressing competitor implicating this surface polysaccharide in protection against T6SS assault. Similar findings have been reported for other bacteria; for example, the extracellular polysaccharide of V. cholerae and the colanic acid capsule of E. coli confer defence against T6SS attack (Hersch et al., 2020; Toska et al., 2018). One potential mechanism is that the capsule sterically impairs the ability of the T6SS to penetrate the target cell membrane, and/or inhibits access of T6SS effectors to their cellular targets. Interestingly, recent genetic evidence indicates that some commensal Neisseria species also have capacity to produce polysaccharide capsules (Clemence et al., 2018), which might also confer a survival advantage in mixed populations that include strains expressing T6SS.

Most bacteria exist within complex polymicrobial communities in which the spatial and temporal dynamics of proliferation and death have a major effect on their fitness and survival (Nadell et al., 2016). While structured complex microbial societies can benefit all their members (Gabrilska and Rumbaugh, 2015; Wolcott et al., 2013), antagonistic neighbours, especially those deploying contact-dependent killing mechanisms, can disrupt communities. Although T6SS-mediated killing can be advantageous to a producing strain during bacterial competition, this requires intimate association with its prey (MacIntyre et al., 2010; Russell et al., 2014). Thus, one way for susceptible bacteria to evade T6SS killing is to avoid direct contact with attacking cells (Borenstein et al., 2015; Smith et al., 2020). In Neisseria, the Tfp is a key mediator of interbacterial and interspecies interactions (Custodio et al., 2020; Higashi et al., 2011) and pilus-mediated interactions influence the spatial structure of a growing community (Oldewurtel et al., 2015; Zöllner et al., 2017). In N. gonorrhoeae, non-piliated bacteria segregate to the expanding front of the colony and Tfp-mediated spatial reorganisation can allow bacteria to avoid external stresses or strains competing for resources (Oldewurtel et al., 2015; Zöllner et al., 2017). We predicted that this would be especially relevant in the context of T6SS-mediated antagonism. For example, physical exclusion driven by Tfp-loss or modification could be an effective strategy to evade and survive an antagonistic interaction, while pilus-mediated interactions might be less favourable for a susceptible prey. Importantly, Tfp loss may occur naturally in a polymicrobial environment and is an established phenomenon in pathogenic Neisseria (Hagblom et al., 1985; Helm and Seifert, 2010). Our results demonstrate that within a bacterial community of attacker and prey cells with or without pili, the sorting of the non-piliated prey to the colony edge results in enhanced survival, likely through segregation of the prey from the attacker. Reduction in prey survival was equivalent whether both or neither the attacker and prey express Tfp, demonstrating that Tfp are not required for T6SS activity. However, observation of the prey in mixed colonies where both strains are piliated compared to when neither strain expresses Tfp suggests a possible localised contribution of Tfp. It will be interesting to further explore the contribution of Tfp to T6SS activity at the single-cell level, to ascertain their localised impact and explore for example how this affected by pilus retraction or pilin sequence variation. It is noteworthy that many bacteria (e.g. Pseudomonas aeruginosa, Vibrio cholerae, Acinetobacter baumannii, enteropathogenic E. coli) that employ T6SSs for inter-bacterial competition also express Tfp. Therefore, our findings are of broad relevance for the impact of contact-dependent killing, and further emphasise how precise spatial relationships can have profound effects on how antagonistic and mutualistic factors combine to influence the development of microbial communities.

Materials and methods

Bacterial strains and growth

Bacterial strains used in this study are shown in Key Resources Table (Appendix). Neisseria spp. were grown on Brain Heart Infusion (BHI, Oxoid) agar with 5% defibrinated horse blood or in BHI broth at 37°C with 5% CO2 or GC-medium supplemented with 1.5% base agar (w/v) and 1% Vitox (v/v; Oxoid). GW-medium (Wade and Graver, 2007) was used for N. cinerea microscopy experiments. E. coli was grown on LB (Lennox Broth base, Invitrogen) agar or in liquid LB at 37°C with shaking. Antibiotics were added at the following concentrations: for E. coli, carbenicillin (carb) 100 µg/ml, kanamycin (kan) 50 µg/ml, and chloramphenicol (cm) 20 µg/ml; for Neisseria spp. kan 75 µg/ml, spectinomycin (spec) 65 µg/ml, erythromycin (ery) 15 µg/ml, and polymyxin B (pmB) 10 µg/ml.

DNA isolation and whole-genome sequencing (WGS)

Genomic DNA was extracted using the Wizard Genomic Kit (Promega), and sequenced by PacBio (Earlham Institute, Norwich) using single-molecule real-time (SMRT) technology; reads were assembled de novo with HGAP3 (Chin et al., 2013).

Bioinformatic analysis of putative T6SS genes

All ORFs on the N. cinerea 346T plasmid were analysed manually using NCBI BLASTp against non-redundant protein databases at NCBI using default search parameters to confirm the presence of T6SS-associated conserved domains. The PAAR-like domain and Rhs domain from N. cinerea 346T Nte1 plus VgrG amino acid sequences from N. cinerea 346T T6SS locus were used as query sequences in BLASTp analysis using the PubMLST BLAST tool. The default parameter of word size (length of the initial identical match that is required before extending a hit) of 11 was used for all searches. Output of 10 hits per isolate was selected to enable identification of multiple homologues within a genome. BLAST results were then subjected to further manual refinement by filtering the hits obtained using a cut-off of at least 20% homology to the query sequence and 20% coverage. The FASTA nucleotide sequence of the hits (including 100 bp flanking sequence) were extracted from the PubMLST database and mapped onto the 346T PacBio genome using SnapGene. Each of the ORFs mapped on the genome were further analysed by NCBI BLASTx against non-redundant protein databases at NCBI using default search parameters to confirm the presence of T6SS-associated conserved domains. T6SS-effector prediction software tools (Li et al., 2015) were also used to identify putative effectors.

Construction of N. cinerea mutants

Primers used in this study are listed in key resources table (Appendix). Target genes were replaced with antibiotic cassettes as previously (Wörmann et al., 2016). Constructs were assembled into pUC19 by Gibson Assembly (New England Biolabs), and hosted in Escherichia coli DH5α. Plasmids were linearised with ScaI, and gel extracted, relevant linearised fragments used to transform N. cinerea; transformants were checked by PCR and sequencing. Complementation or chromosomal insertion of genes encoding fluorophores was achieved using pNCC1-Spec, a spectinomycin-resistant derivative of pNCC1 (Wörmann et al., 2016). For visualisation of T6SS-sheaths, sfgfp was cloned in-frame with tssB and a short linker (encoding 3×Ala 3×Gly) by Gibson Assembly (New England Biolabs) into pNCC1-Spec to allow IPTG-inducible expression of TssB-sfGFP. PCR was performed using Herculase II (Agilent) or Q5 High-fidelity DNA Polymerase (New England Biolabs).

Analysis of effector/immunity activity in E. coli

Putative effector coding sequences with or without cognate immunity gene were amplified by PCR from N. cinerea 346T gDNA and either assembled by Gibson Assembly (NEB) into pBAD33 or, for Nte1 with or without addition of the PelB signal sequence, cloned in to pBAD33 using XbaI / SphI restriction enzyme sites. All forward primers also included the E. coli ribosomal binding site (RBS: AAGAAGG) upstream of the start codon. Plasmids were transformed into E. coli DH5α and verified by sequencing (Source Bioscience). For assessment of toxicity, strains with recombinant or empty pBAD33 plasmids were grown overnight in LB supplemented with 0.8% glucose (w/v), then diluted to an OD600 of 0.1 and incubated for 1 hr at 180 rpm and 37°C; bacteria were pelleted and resuspended in LB with arabinose (0.8% w/v) to induce expression and incubated at 37°C, 180 rpm for a further 4 hr. The OD600 and CFU/ml of cultures were determined; aliquots were diluted and plated to media containing 0.8% glucose at relevant time points up to 5 hr.

Hcp protein expression, purification, and antibody generation

Codon optimised hcp was synthesised with a sequence encoding an N-terminal 6x His Tag and a 3C protease cleavage site, and flanked by NcoI and XhoI restriction sites (ThermoFisher). The fragment was ligated into NcoI and XhoI sites in pET28a (Novagen) using QuickStick T4 DNA Ligase (Bioline) and transformed into E. coli B834. Bacteria were grown at 37°C, 150 rpm to an OD600 of 1.0, and expression of 6xHis-3C-Hcp was induced with 1 mM IPTG for 24 hr at 16°C. Cells were resuspended in Buffer A (50 mM Tris-HCl buffer pH 7.5, 10 mM Imidazole, 500 mM NaCl, 1 mM DTT) containing protease inhibitors, 1 mg/mL lysozyme and 100 μg/mL DNase then subsequently homogenised with an EmulsiFlex-C5 (Avestin). Lysed cells were ultracentrifuged, and the cleared supernatant loaded onto a Ni Sepharose 6 Fast Flow His Trap column (GE Healthcare) equilibrated with Buffer A. The column was washed with Buffer A, then Buffer B (50 mM Tris-HCl buffer pH 7.5, 35 mM Imidazole, 500 mM NaCl, 1 mM DTT) before elution with 10 mL of Buffer C (50 mM Tris-HCl buffer pH 7.5, 300 mM Imidazole, 150 mM NaCl, 1 mM DTT). The eluate was incubated with the HRV-3C protease (Sigma) then applied to a Ni Sepharose column. The eluate containing protease and cleaved protein was concentrated using Amicon Ultra 10,000 MWCO (Millipore), then passed through a Superdex-200 column (GE Healthcare, Buckinghamshire, UK). Fractions were analysed by SDS-PAGE and Coomassie blue staining, and those with Hcp pooled and used to generate polyclonal antibodies (EuroGentec).

Hcp secretion assay

Bacteria were grown in BHI broth for 4–5 hr then harvested and lysed in an equal volume of SDS-PAGE lysis buffer (500 mM Tris-HCl [pH 6.8], 5% SDS, 15% glycerol, 0.5% bromophenol blue containing 100 mM β-mercaptoethanol); supernatants were filtered (0.22 µm pore, Millipore) and proteins precipitated with 20% (v/v) trichloroacetic acid. Hcp was detected by Western blot with anti-Hcp (1:10,000 dilution) and goat anti-rabbit IgG–HRP (1:5000, sc-2004; Santa Cruz). Anti-RecA (1:5000 dilution, ab63797; Abcam) followed by goat anti-rabbit IgG–HRP and detection with ECL detection Reagent (GE Healthcare) or Coomassie blue staining were used as loading controls.

Live cell imaging of T6SS activity

Bacteria were grown overnight on BHI agar, resuspended in PBS and 20 µl spotted onto fresh BHI agar containing 1 mM IPTG and incubated for 4 hr at 37°C. After incubation, 500 µl of 109 CFU/mL bacterial suspension of attacker was mixed with the prey strain at a 1:1 ratio. Cells were harvested by centrifugation for 3 min at 6000 rpm, resuspended in 100 μL of PBS or GW media and 2 µl spotted on 1% agarose pads (for T6SS dynamics) or GW media with 0.1 mM IPTG and 0.5 µM SYTOXBlue (Thermo Fisher Scientific) for assessment of prey permeability. Fluorescence microscopy image sequences were acquired within 20–30 min of sample preparation with an inverted Zeiss 880 Airyscan microscope equipped with Plan-Apochromat 63×/1.4-NA oil lens and fitted with a climate chamber mounted around the objective to perform the imaging at 37°C with 5% CO2. Automated images were collected at 1 s, 10 s or 1 min intervals and processed with Fiji (Schindelin et al., 2012). Background noise was reduced using the ‘Despeckle’ filter. The XY drift was corrected using StackReg with ‘Rigid Body’ transformation (Thévenaz et al., 1998). Experiments and imaging were performed on at least two independent occasions.

Quantitative competition assays

Strains grown overnight on BHI agar were resuspended in PBS and diluted to 109 CFU/mL based on OD quantification, mixed at an approximate ratio of ~10:1 for N. cinerea/ N. cinerea and N. cinerea/N. gonorrhoeae, or ~100:1 for N. cinerea/ N. meningitidis (actual CFU are indicated in source data files where available), then 20 µl spotted onto BHI agar in triplicate and incubated at 37°C with 5% CO2. At specific time-points, entire spots were harvested and resuspended in 1 mL of PBS. The cellular suspension was then serially diluted in PBS and aliquots spotted onto selective media. Colonies were counted after ~16 hr incubation at 37°C with 5% CO2. Experiments were performed on at least three independent occasions. For different prey analysis, relative survival was defined as the fold change in recovery of prey following incubation with wild-type attacker N. cinerea compared to a T6SS-deficient N. cinerea.

Competition assays assessed by fluorescence microscopy and flow cytometry

Bacteria were grown overnight on BHI, resuspended in PBS and diluted to 109 CFU/mL. 100 µl of each suspension (attacker/prey) were mixed thoroughly (i.e., a 1:1 ratio) and 1 µl spotted in duplicate onto GC-medium supplemented with 0.5% base agar (w/v) and 1% Vitox (v/v; Oxoid). Plates were incubated for 24 hr at 37°C, 5% CO2. For flow cytometry analysis, the remaining input suspension was then centrifuged for 3 mins at 6000 rpm then pellets resuspended in 500 µL of 4% paraformaldehyde and fixed for 20 min at room temperature. Following centrifugation, the fixed bacteria were then resuspended in 250 µL PBS and stored at 4°C for 24 hr prior to analysis. At various time points, expanding colonies were imaged using a M125C stereo microscope equipped with a DFC310FX digital camera (Leica Microsystems), and images processed with Fiji. Images were imported using ‘Image Sequence’ and corrected with StackReg as described above. At 24 hr, colonies were harvested, fixed with 4% PFA for 20 min then washed with PBS. Samples were analysed using a Cytoflex LX (Beckman Coulter), and at least 104 events recorded. Fluorescence, forward and side scatter data were collected to distinguish between debris and bacteria. Results were analysed by calculating the number of events positive for either GFP or Cherry signal in FlowJo v10 software (Becton Dickinson Company). The negative population (non-fluorescent cells) was established using 346T Wt, the GFP+ population was determined using N. cinerea 346T Wt_sfGFP, and the Cherry+ population using N. cinerea 346T Wt_sfCherry. Quadrants were set to delineate the GFP+, Cherry+, GFP+Cherry+ and the percentage of cells representing each population within the different competition spots was recorded. Flow cytometry analysis was performed on two independent occasions. Stereo microscopy analysis was performed on three independent occasions with technical duplicates each time.

Statistical analyses

Graphpad Prism7 software (San Diego, CA) was used for statistical analysis. We used One-way/two-way ANOVA with Tukey post hoc testing for multiple comparisons and unpaired two-tailed Student’s t-test for pairwise comparisons. In all cases, p < 0.05 was considered statistically significant.

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The type VI secretion system (T6SS) is an antibacterial system that has been characterized in a large number of Gram-negative bacteria either pathogenic or environmental. This study represents the first functional description of T6SS in the genus Neisseria, which, unusually, is plasmid encoded. The work demonstrates that this T6SS may be important in allowing commensal Neisseria cinerea to resist pathogenic Neisseria, and has implications for understanding how spatial considerations impact T6SS-mediated inter-bacterial interactions.

Decision letter after peer review:

Thank you for submitting your article "Type VI secretion system killing by commensal Neisseria is influenced by the spatial dynamics of bacteria" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Alain Filloux as the Reviewing Editor and Reviewer #1, and the evaluation has been overseen by Gisela Storz as the Senior Editor.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our policy on revisions we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, as they do with your paper, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Our expectation is that the authors will eventually carry out the additional experiments and report on how they affect the relevant conclusions either in a preprint on bioRxiv or medRxiv, or if appropriate, as a Research Advance in eLife, either of which would be linked to the original paper.

Summary:

The type VI secretion system (T6SS) is an antibacterial system that has been characterized in a large number of Gram-negative bacteria either pathogenic or environmental.

This works represents the first functional description of T6SS in the genus Neisseria, which, unusually, is plasmid encoded, demonstrates that this T6SS may be important in allowing commensal Neisseria cinerea to resist pathogenic Neisseria, assess the role of T6SS effectors and how the capsule on the target cell reduces the efficacy of T6SS killing.

It brings important new aspects to the emerging theme in the field that spatial considerations are critical to understanding T6SS-mediated inter-bacterial interactions. Notably, it demonstrates how the type four pilus (Tfp) on the target cell contributes to T6SS-dependent killing. In a mixed population of bacteria possessing or not the Tfp, bacteria without Tfp are excluded at the edge of the microcolony where they can outgrow suggesting that both T6SS and Tfp contribute to the spatial organisation of the microcolonies.

Overall, the findings will be of great interest to the T6SS field and also those interested in microbial interactions more broadly.

Essential revisions:

1. The focus of the paper is on the spatial dynamics/spatial segregration in mixed microcolonies. Yet, the authors do not exclude a very localized role of the Tfp at the interface between the attacker and the prey. It is not clear for me whether the impact of the Tfp is due to organisation of the bacterial community (with Tfp minus strains segregating at the periphery of the colony) or whether the Tfp has a very localized impact either by bringing the attacker and the prey in close contact or triggering a response of the attacker as described for vibrio T4SS/mating pair formation system. Indeed, in figure 6, it seems that there are also major differences in the center of the inoculation zone between Tfp- and Tfp+ strains.

Assessing the role of Tfp in the attacker cells (in the experiment presented in Figure 6) would be very helpful for the understanding of the role of Tfp.

2. There are 6 toxin/immunity pairs described. While Nte1-5 are encoded in the T6SS cluster, nte6/nti6 is found as a remote pair. What suggests that Nte6 is a T6SS-dependent toxin? In this respect there are no demonstration of a direct T6SS-dependent secretion of any of the 6 toxins. This is not essential but if tools available and at least one toxin could be tested it will add value to this publication.

3. There does not seem to be any paar gene in the cluster. Is there any elsewhere on the plasmid? If not which of the Ntes have a PAAR-like at the N terminus. Only Nte1 and Nte2? If this is true, then deletion of these two should likely prevent the other toxins to be delivered since it is accepted that at least one PAAR is required for the T6SS to be effective. This may be worth assessing but at least some more information on the domain organisation would be helpful. Along this line, the Nte5 (and to a lesser extent Nte4) effector is much smaller than a typical full-length T6SS-associated Rhs protein would be. Is it (or both) actually an 'orphan' Rhs CT- immunity pair (presumed to be displaced from the upstream full-length Rhs by homologous recombination, leaving only a part of the Rhs domain)? Perhaps this could be discussed?

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Type VI secretion system killing by commensal Neisseria is influenced by the spatial dynamics of bacteria" for further consideration by eLife. Your revised article has been evaluated by Gisela Storz as the Senior Editor, and a Reviewing Editor.

All previous reviewers have had the opportunity to look carefully at the changes you made and we are happy to say that there is a consensus that most comments have been well addressed. However, there is still the pending issue that the contribution of type IV pili (Tfp) loss to increased prey survival may not simply be due to the gross spatial reorganisation in the colony. One way to look a bit further into this would be for example to have attacker and prey both lacking pili so that the role of Tfp in T6SS-dependent combat independently of segregation could be further tested. One reviewer suggests for example an experiment where you make one more strain (GFP Tfp- strain to be used as T6SS+/Tfp- attacker) and then use this with the Tfp+ and Tfp- prey cells (Tfp- attacker with Tfp- prey would give a non-segregated population but might retain other aspects of loss of Tfp, so would be interesting to see if prey still survive better/T6SS still less efficient).

The eLife policy is not to ask for a second round of revision. However, in this case we thought we could give you the opportunity to address this point, especially since we felt that it should not be a too challenging experiment.

Will you decide not to perform the experiment then we would ask you to modify your text accordingly so that the discussion about the role of Tfp takes all possible aspects into consideration. We would also suggest to modify the title to highlight the role of the Tfp without any emphasis on the spatial dynamics."

Essential revisions:

1. The focus of the paper is on the spatial dynamics/spatial segregration in mixed microcolonies. Yet, the authors do not exclude a very localized role of the Tfp at the interface between the attacker and the prey. It is not clear for me whether the impact of the Tfp is due to organisation of the bacterial community (with Tfp minus strains segregating at the periphery of the colony) or whether the Tfp has a very localized impact either by bringing the attacker and the prey in close contact or triggering a response of the attacker as described for vibrio T4SS/mating pair formation system. Indeed, in figure 6, it seems that there are also major differences in the center of the inoculation zone between Tfp- and Tfp+ strains.

Assessing the role of Tfp in the attacker cells (in the experiment presented in Figure 6) would be very helpful for the understanding of the role of Tfp.

We agree with the reviewer that Tfp may have a localised impact through pilus-pilus interactions bringing individual cells into close proximity for T6SS dependent killing. However, the assays presented in this paper do not enable us to either ascertain or exclude this effect. Further experiments (e.g. single cell analysis of piliated and non-piliated attacker/prey using sytox blue as an indicator of prey lysis) are required to address this but we feel are beyond the scope of this work. We have modified our discussion to clarify that our current work is limited to understanding the impact of prey piliation of a susceptible prey on survival against T6SS attack within a bacterial community, and that the contribution of pili on the attacker and at the cell-cell level merits future investigation

In addition, analysis of the impact of pili on attacker and the impact of pilin subunit variation, Tfp dynamics and the contribution of effectors should be addressed in future work to provide a comprehensive view of the interplay between Tfp expression and T6SS attack in mixed microcolonies.

2. There are 6 toxin/immunity pairs described. While Nte1-5 are encoded in the T6SS cluster, nte6/nti6 is found as a remote pair. What suggests that Nte6 is a T6SS-dependent toxin? In this respect there are no demonstration of a direct T6SS-dependent secretion of any of the 6 toxins. This is not essential but if tools available and at least one toxin could be tested it will add value to this publication.

The reviewer raises a valid point. Our reasons for proposing Nte6 as a T6SS-dependent toxin were based on bioinformatic analysis which reveals that the protein encoded by Nte6 has features commonly found in T6SS effectors (namely Rhs domains and a C-terminal region encoding a toxin belonging to the HNH superfamily, see revised manuscript Figure 4). In addition, some preliminary data we have obtained indicates this effector is linked with the T6SS. As shown in Author response image 1, deletion of nte6 in combination with the three other genes encoding putative nuclease effectors (generating 346TΔnte/nti3,4,5,6) led to survival of prey (N. meningitidis) at levels equivalent to survival in presence of a ΔT6SS mutant, and abolished Hcp secretion. This observation suggests that Nte3, Nte4, Nte5 and Nte6 are linked to the T6SS in N. cinerea and may be required for T6SS activity, reminiscent of findings from Agrobacterium tumefaciens in which effective T6SS activity requires loading of cargo effectors (Wu et al., 2020). However, these findings remain preliminary and experimental data demonstrating T6SS-dependent secretion of Nte1-6 in N. cinerea 346T are still required. Therefore, we have been careful to describe these as putative effectors in our current manuscript, and to clarify that further experiments are required to demonstrate direct T6SS-dependent secretion for all the toxins .

Author response image 1.

The deletion of all four putative endonuclease effectors impairs T6SS function and Hcp secretion.

(A) Recovery of wild-type N. meningitidis 8013 after 4 h co-incubation with N. cinerea 346T wild-type (Wt) and the T6SS mutant (ΔT6SS) or strains lacking specific effectors at a 100:1 attacker:prey ratio. Data shown are the mean ± SD of three independent experiments: NS, not significant, ***p < 0.0001, one-way ANOVA test for multiple comparison. (B) Western blot detection of Hcp in whole cell lysates (W) and supernatants (S) from bacteria grown in liquid BHI media for 4-5 h. WT (wild-type N. cinerea 346T), ΔTssB (strain with deletion of tssB), Δ4Nte (strain lacking nte/nti 3,4,5,6).

3. There does not seem to be any paar gene in the cluster. Is there any elsewhere on the plasmid? If not which of the Ntes have a PAAR-like at the N terminus. Only Nte1 and Nte2? If this is true, then deletion of these two should likely prevent the other toxins to be delivered since it is accepted that at least one PAAR is required for the T6SS to be effective. This may be worth assessing but at least some more information on the domain organisation would be helpful.

The reviewer is correct, only Nte1 and Nte2 have PAAR domains at the N terminus (indicated in new Figure 4 ). We analysed the PacBio WGS of N. cinerea 346T for additional PAAR domains using BLASTp analysis and the PAAR-like domain sequence from Nte1 as query sequence. This analysis did not identify any other PAAR genes either on the plasmid or the chromosome. We have now included this information in the revised manuscript .

The reviewer is correct that based on previous work, deletion of both PAAR containing effectors (Nte1 and Nte2) may prevent secretion, and this is something we intend to address in future work, along with demonstration of the T6SS-dependent secretion of the toxins as indicated in response to point 2. As suggested, our revised manuscript includes more information about the predicted effectors (new Figure 4- schematic of domain structure, results , discussion and Materials and methods ).

Along this line, the Nte5 (and to a lesser extent Nte4) effector is much smaller than a typical full-length T6SS-associated Rhs protein would be. Is it (or both) actually an 'orphan' Rhs CT- immunity pair (presumed to be displaced from the upstream full-length Rhs by homologous recombination, leaving only a part of the Rhs domain)? Perhaps this could be discussed?

The reviewer raises and interesting point. Orphan effector-immunity gene pairs have been reported, for example in Vibrio species, where they are found in conserved gene neighbourhoods, not adjacent to the T6SS gene cluster and sometimes accompanied by transposable elements (Salomon et al., 2015). In addition, orphan immunity genes have been identified in several species (Ma et al., 2017; Ross et al., 2019) and can be found as islets, or located downstream of effector-immunity modules (Kirchberger et al., 2017). Nte5 encodes a putative nuclease effector with a shorter Rhs domain compared to the other five putative effectors (319AA compared to 695-735AA) and therefore may represent an orphan RHS-CT gene. We have now included this possibility in the revised manuscript .

References:

Custodio, R., Johnson, E., Liu, G., Tang, C.M., and Exley, R.M. (2020). Commensal Neisseria cinerea impairs Neisseria meningitidis microcolony development and reduces pathogen colonisation of epithelial cells. PLoS pathogens 16, e1008372.

Hagblom, P., Segal, E., Billyard, E., and So, M. (1985). Intragenic recombination leads to pilus antigenic variation in Neisseria gonorrhoeae. Nature 315, 156-158.

Helm, R.A., and Seifert, H.S. (2010). Frequency and rate of pilin antigenic variation of Neisseria meningitidis. J Bacteriol 192, 3822-3823.

Kirchberger, P.C., Unterweger, D., Provenzano, D., Pukatzki, S., and Boucher, Y. (2017). Sequential displacement of Type VI Secretion System effector genes leads to evolution of diverse immunity gene arrays in Vibrio cholerae. Scientific reports 7, 45133.

Knapp, J.S., and Hook, E.W., 3rd (1988). Prevalence and persistence of Neisseria cinerea and other Neisseria spp. in adults. J Clin Microbiol 26, 896-900.

Ma, J., Pan, Z., Huang, J., Sun, M., Lu, C., and Yao, H. (2017). The Hcp proteins fused with diverse extended-toxin domains represent a novel pattern of antibacterial effectors in type VI secretion systems. Virulence 8, 1189-1202.

Marri, P.R., Paniscus, M., Weyand, N.J., Rendon, M.A., Calton, C.M., Hernandez, D.R., Higashi, D.L., Sodergren, E., Weinstock, G.M., Rounsley, S.D., et al. (2010). Genome sequencing reveals widespread virulence gene exchange among human Neisseria species. PLoS One 5, e11835.

Matthey, N., Stutzmann, S., Stoudmann, C., Guex, N., Iseli, C., and Blokesch, M. (2019). Neighbor predation linked to natural competence fosters the transfer of large genomic regions in Vibrio cholerae. eLife 8.

Ross, B.D., Verster, A.J., Radey, M.C., Schmidtke, D.T., Pope, C.E., Hoffman, L.R., Hajjar, A.M., Peterson, S.B., Borenstein, E., and Mougous, J.D. (2019). Human gut bacteria contain acquired interbacterial defence systems. Nature 575, 224-228.

Salomon, D., Klimko, J.A., Trudgian, D.C., Kinch, L.N., Grishin, N.V., Mirzaei, H., and Orth, K. (2015). Type VI Secretion System Toxins Horizontally Shared between Marine Bacteria. PLoS pathogens 11, e1005128.

Sheikhi, R., Amin, M., Rostami, S., Shoja, S., and Ebrahimi, N. (2015). Oropharyngeal Colonization With Neisseria lactamica, Other Nonpathogenic Neisseria Species and Moraxella catarrhalis Among Young Healthy Children in Ahvaz, Iran. Jundishapur J Microb 8.

Wu, C.F., Lien, Y.W., Bondage, D., Lin, J.S., Pilhofer, M., Shih, Y.L., Chang, J.H., and Lai, E.M. (2020). Effector loading onto the VgrG carrier activates type VI secretion system assembly. EMBO reports 21, e47961.

[Editors' note: further revisions were suggested prior to acceptance, as described below.]

All previous reviewers have had the opportunity to look carefully at the changes you made and we are happy to say that there is a consensus that most comments have been well addressed. However, there is still the pending issue that the contribution of type IV pili (Tfp) loss to increased prey survival may not simply be due to the gross spatial reorganisation in the colony. One way to look a bit further into this would be for example to have attacker and prey both lacking pili so that the role of Tfp in T6SS-dependent combat independently of segregation could be further tested. One reviewer suggests for example an experiment where you make one more strain (GFP Tfp- strain to be used as T6SS+/Tfp- attacker) and then use this with the Tfp+ and Tfp- prey cells (Tfp- attacker with Tfp- prey would give a non-segregated population but might retain other aspects of loss of Tfp, so would be interesting to see if prey still survive better/T6SS still less efficient).

The eLife policy is not to ask for a second round of revision. However, in this case we thought we could give you the opportunity to address this point, especially since we felt that it should not be a too challenging experiment.

We are very grateful to the reviewer for this suggestion. We had available the non-piliated, T6SS+ strain expressing GFP (included in revised Key resources table) and have now performed this experiment using (i) competition assays to quantitatively assess prey survival and (ii) fluorescence microscopy to qualitatively assess the distribution of strains in mixed colonies. The competition assays revealed that increased prey survival is observed only when the prey is non-piliated, thus, these data support our conclusions that prey survival is enhanced upon loss of pili due to segregation. However, although competition assays demonstrated that prey survival was equivalent when attacker and prey either both have, or both lack Tfp, the microscopy analysis of mixed colonies indicate qualitative differences in prey distribution between colonies composed of Tfp+/Tfp+ attacker and prey and Tfp-/Tfp- attacker and prey. These observations suggests that Tfp may have an effect beyond gross spatial reorganisation, possibly a local effect that is not detected at the population level. We have included the new data and discussion of these findings in our revised manuscript (Figure 8, Figure 8—figure supplement 1 and 2 and corresponding legends, results , discussion).

Will you decide not to perform the experiment then we would ask you to modify your text accordingly so that the discussion about the role of Tfp takes all possible aspects into consideration. We would also suggest to modify the title to highlight the role of the Tfp without any emphasis on the spatial dynamics."

We have performed the experiment and have also modified the text accordingly. As our findings do not rule out contributions of Tfp other than Tfp loss enhancing prey survival through segregation, we have included other possible roles of Tfp in the results and discussion and still include that further work is necessary to explore the contribution of Tfp to T6SS mediated attack beyond the impact on spatial reorganisation within a colony .

Regarding the title of our paper, we felt that the reviewers’ suggestion was helpful even though we have included further data to support the observation that T6SS killing in N. cinerea is influenced by spatial dynamics. We have therefore modified the title to be a broader description of the work presented. The new title is “Type VI secretion system killing by commensal Neisseria is influenced by expression of Type four pili” and highlights the role of Tfp without any emphasis on the spatial dynamics.

Appendix 1—key resources table

Reagent type (species) or resource	Designation	Source or reference	Identifiers	Additional information
Antibody	Goat polyclonal anti-rabbit IgG–HRP	Santa Cruz	sc-2004	Target rabbit IgG antibodies WB (1:5000)
Antibody	Rabbit polyclonal anti-RecA	Abcam	ab63797	Target bacterial RecA protein WB (1:5000)
Antibody	Rabbit polyclonal anti-Hcp sera	This paper		Antibody raised to target full-length N. cinerea 346T Hcp protein WB (1:10000)
Chemical compound, drug	SYTOX Blue	Thermo Fisher Scientific	S34857	SYTOX Blue is a high-affinity nucleic acid stain that does not penetrate uncompromised cell membranes
Software, algorithm	FiJi	Schindelin et al., 2012 DOI:10.1038/nmeth.2019	https://fiji.sc RRID:SCR_002285
Software, algorithm	Graphpad Prism7	San Diego, CA	https://www.graphpad.com/ RRID:SCR_002798
Software, algorithm	FlowJo v10	Becton Dickinson Company	https://www.flowjo.com/ RRID:SCR_008520
Strain, strain background (Neisseria cinerea)	CCUG346T (346T)	Bennett et al., 2012 DOI:10.1099/mic.0.056077-0		wild-type N. cinerea
Strain, strain background (Neisseria cinerea)	CCUG27178A (27178A)	Bennett et al., 2012 DOI:10.1099/mic.0.056077-0		wild-type N. cinerea
Strain, strain background (Neisseria cinerea)	346T_sfGFP	Wörmann et al., 2016 DOI: 10.1099/mic.0.000248		346T with chromosomally integrated sfGfp; EryR
Strain, strain background (Neisseria cinerea)	346T_sfGFPΔpilE1/2	Wörmann et al., 2016 DOI: 10.1099/mic.0.000248		346T with pilE1 and pilE2 deleted by insertion mutagenesis, and chromosomally integrated sfGfp; EryR and KanR
Strain, strain background (Neisseria cinerea)	346T_sfCherry	This paper		346T with chromosomally integrated sfCherry; EryR
Strain, strain background (Neisseria cinerea)	27178A_sfCherry	This paper		27178 with chromosomally integrated sfCherry; SpecR
Strain, strain background (Neisseria cinerea)	346TΔT6SS	This paper		346T with insertion-deletion of tssC – vgrG region; EryR
Strain, strain background (Neisseria cinerea)	346TΔtssB	This paper		346T with insertion-deletion of tssB; EryR
Strain, strain background (Neisseria cinerea)	346TΔtssB::tssBsfGFP	This paper		346T with insertion-deletion of native tssB and ectopic chromosomal insertion of tssB-sfGFP fusion; SpecR EryR
Strain, strain background (Neisseria cinerea)	346TΔtssM	This paper		346T with insertion-deletion of tssM; TetR
Strain, strain background (Neisseria cinerea)	346TΔtssBΔtssM::tssB-sfGFP	This paper		346T with insertion-deletion of native tssB and tssM and ectopic chromosomal insertion of tssB-sfGFP fusion; SpecR EryR TetR
Strain, strain background (Neisseria cinerea)	346TΔnte3Δnte4Δnte5	This paper	immunity genes; EryR	deletion mutagenesis, nte3-nte5 locus deletion including respective immunity genes; EryR
Strain, strain background (Neisseria cinerea)	346TΔnte/i3-5_sfCherry	This paper		346T with insertion-deletion of nte/i3-5 region and ectopic chromosomal insertion of sfCherry; SpecR EryR
Strain, strain background (Neisseria cinerea)	346TΔnte6	This paper		deletion mutagenesis, nte6 deficient; SpecR
Strain, strain background (Neisseria cinerea)	346TΔnte3Δnte4Δnte5Δnte6	This paper		deletion mutagenesis, nte3-nte5 locus deletion including respective immunity genes plus nte6 deletion; EryR SpecR
Strain, strain background (Neisseria cinerea)	346TΔnte/i3-5ΔpilE1/2_sfCherry	This paper		346T with insertion-deletion of nte/i3-5 region; ectopic chromosomal insertion of sfCherry; insertion-deletion of pilE1 and pilE2; kanR,SpecR EryR
Strain, strain background (Neisseria meningitidis)	8013	Rusniok et al., 2009 DOI: 10.1186/gb-2009-10-10-r110		N. meningitidis wild-type
Strain, strain background (Neisseria meningitidis)	MC58	Tettelin et al., 2000 DOI: 10.1126/science.287.5459.1809.		N. meningitidis wild-type
Strain, strain background (Neisseria meningitidis)	S3	Uria et al., 2008 DOI: 10.1084/jem.20072577		N. meningitidis wild-type
Strain, strain background (Neisseria meningitidis)	MC58ΔsiaD	Virji et al., 1995 DOI:10.1111/j.1365-2958.1995.mmi_18040741.x		deletion mutagenesis, NEIS0051; KanR
Strain, strain background (Neisseria gonorrhoeae)	FA1090 pGCC4	Mehr and Seifert, 1997 DOI:10.1046/j.1365-2958.1997.2971660.x		FA1090 with chromosomally integrated plasmid pGCC4; EryR
Strain, strain background (Escherichia coli)	Dh5α	Lab collection		DH5α is an E. coli strain used for general cloning applications.
Strain, strain background (Escherichia coli)	Dh5α pNCC1-Spec	This paper		Dh5α with pNCC1SpecR plasmid
Strain, strain background (Escherichia coli)	Dh5α pNCC1-Spec-sfGFP	This paper		Dh5α with pNCC1-Spec with sfGFP insert;
Strain, strain background (Escherichia coli)	Dh5α pNCC101-Spec-sfCherry	Lab collection		DH5α with plasmid pNCC101+sfCherry insert. SpecR
Strain, strain background (Escherichia coli)	Dh5α pUC19	Lab collection	pUC19 vector RRID:Addgene_50005	E. coli DH5α strain harbouring pUC19 for general cloning applications.
Strain, strain background (Escherichia coli)	Dh5α pUC19::ΔtssB	This paper		DH5α with pUC19::ΔtssB deletion construct; CarbR EryR
Strain, strain background (Escherichia coli)	Dh5α pUC19::ΔtssM	This paper		DH5α with pUC19::ΔtssM deletion construct; CarbR TetR
Strain, strain background (Escherichia coli)	Dh5α pUC19::ΔT6SS	This paper		DH5α with pUC19::ΔtssC-vgrG locus deletion construct; CarbR EryR
Strain, strain background (Escherichia coli)	Dh5αpUC19:: Δnte3Δnte4Δnte5	This paper		DH5α with pUC19::Δnte3Δnte4Δnte5 region including respective immunity genes deletion construct; CarbR EryR
Strain, strain background (Escherichia coli)	Dh5α pUC19:: Δnte6	This paper		nte6 deletion construct; CarbR SpecR
Strain, strain background (Escherichia coli)	B834 pET28a	Lab collection	pET28a Novagen Cat. No. 69864–3	B834 with pET28a IPTG-inducible expression vector, KanR
Strain, strain background (Escherichia coli)	Dh5α pET28a-His-3C-Hcp	This paper		Dh5α with pET28a vector for IPTG inducible expression of Nc 346T Hcp with N-terminal cleavable HIS tag. KanR
Strain, strain background (Escherichia coli)	B834 pET28a-His-3C-Hcp	This paper		B834 expression strain, with pET28a vector for IPTG inducible expression of Nc 346T Hcp with N-terminal cleavable HIS tag. KanR
Strain, strain background (Escherichia coli)	Dh5α pBAD33	Lab collection	pBAD33 RRID:Addgene_36267	Dh5α with pBAD33 vector for Arabinose-inducible expression, CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::(ssPelB)Nte1-His	This paper		Dh5α with pBAD33 encoding Nte1 with N-terminal PelB leader peptide and C-terminal his-tag under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33:: (ssPelB)Nte1+Nti1	This paper		Dh5α with pBAD33 encoding Nte1 with N-terminal PelB leader peptide and C-terminal his-tag plus Nti, under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte1-His	This paper		Dh5α with pBAD33 encoding Nte1 with N-terminal his-tag under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte2	This paper		Dh5α with pBAD33 encoding Nte2 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte2+Nti2	This paper		Dh5α with pBAD33 encoding Nte2+Nti2 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte3	This paper		Dh5α with pBAD33 encoding Nte3 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte3+Nti3	This paper		Dh5α with pBAD33 encoding Nte3+Nti3 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte4	This paper		Dh5α with pBAD33 encoding Nte4 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte4+Nti4	This paper		Dh5α with pBAD33 encoding Nte4+Nti4 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte5	This paper		Dh5α with pBAD33 encoding Nte5 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte5+Nti5	This paper		Dh5α with pBAD33 encoding Nte5+Nti5 under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte6R1300S	This paper		Dh5α with pBAD33 encoding Nte6R1300S under arabinose-inducible promoter control; CmR
Strain, strain background (Escherichia coli)	Dh5α pBAD33::Nte6+Nti6	This paper		Dh5α with pBAD33 encoding Nte6+Nit6 under arabinose-inducible promoter control; CmR
Sequence-based reagent	T6SSdel-1	This paper	5’-CGAAAAGTG CCACCTGACGTATGACTGAAAAGCAATTAGATATC	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-2	This paper	5’-GTTAAATTTAAGGATAAGAAACGTGGCAG	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-3	This paper	5’-TTTCTTATCC TTAAATTTAACGATCACTCATCATG	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-4	This paper	5’-ACTCAAACATTTACTTATTAAATAATTTATAGCTATTGAAAAG	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-5	This paper	5’-TTAATAAGTAAATGTTTGAGTTGCAGAACTTTAC	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-6	This paper	5’-GATAATAATGGTTTCTTAGACGTGCCGTTCCAATAGGCCATAG	Deletion of tssC-vgrG locus
Sequence-based reagent	T6SSdel-conf-F	This paper	5’-CCTAAAGCG GCTTCCAAAGACG	Confirmation of tssC-vgrG locus deletion
Sequence-based reagent	T6SSdel-conf-R	This paper	5’-CCATGCCGG TAAAGGTCAGT	Confirmation of tssC-vgrG locus deletion
Sequence-based reagent	TssBdel-1	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCACTTACCCTGATCCACAAAGCC	Deletion of tssB
Sequence-based reagent	TssBdel-2	This paper	5’-ATTCAATGACCTTTAAATGATAAAAGTTGT	Deletion of tssB
Sequence-based reagent	TssBdel-3	This paper	5’-ACAACTTTTATCATTTAAAGGTCATTGAATATGAACGAGAAAAATATAAAACACAGTC	Deletion of tssB
Sequence-based reagent	TssBdel-4	This paper	5’-TTACTTATTA AATAATTTATAGCTATTGAAAAGAGATAAGAATTG	Deletion of tssB
Sequence-based reagent	TssBdel-5	This paper	5’-TATAAATTATTTAATAAGTAAGCTTCCAAAGACGAGCAGTAA	Deletion of tssB
Sequence-based reagent	TssBdel-6	This paper	5’-CAGGAAACA GCTATGACCATGATTACGCCTAAGTTGCGGGCAACTTCTT	Deletion of tssB
Sequence-based reagent	TssBdel-conf-F	This paper	5’-ATAGAAACCTACTTTTTCGGAAAGC	Confirmation of tssB deletion
Sequence-based reagent	TssBdel-conf-R	This paper	5’-TTACTTATTA AATAATTTATAGCTATTGAAAAGAGATAAGAATTG	Confirmation of tssB deletion
Sequence-based reagent	TssMdel-1	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAACCCTGTCTTGGCTAGAGTC	Deletion of tssM
Sequence-based reagent	TssMdel-2	This paper	5’-ATTTGTTTTT CCGTATCAATCCAATTTCA	Deletion of tssM
Sequence-based reagent	TssMdel-3	This paper	5’-ATTGGATTGATACGGAAAAACAAATATGAAAATTATTAATATTGGAGTTTTAGCTCATGTT	Deletion of tssM
Sequence-based reagent	TssMdel-4	This paper	5’-CTAAGTTATTTTATTGAACATATATCGTACTTTATCTATCCG	Deletion of tssM
Sequence-based reagent	TssMdel-5	This paper	5’-AAGTACGATATATGTTCAATAAAATAACTTAGAATAAATTAAGGAATTTTCAGTGCATTTGAAG	Deletion of tssM
Sequence-based reagent	TssMdel-6	This paper	5’-CAGGAAACA GCTATGACCATGATTACGCCGGCAATATCTAGAACGGATTTATCG	Deletion of tssM
Sequence-based reagent	TssMdel-Conf-F	This paper	5’-AGGACTTCC AAGATAGAAGTACGG	Confirmation of tssM deletion
Sequence-based reagent	TssMdel-Conf-R	This paper	5’-AAAGCCCCT TGTACGATAGC	Confirmation of tssM deletion
Sequence-based reagent	Nte345del-1	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAGACCTTCATGCTGACTAGTGAT	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-2	This paper	5’-GAAGTGTTG GATGAACTTTTTCTATG	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-3	This paper	5’-CATAGAAAAAGTTCATCCAACACTTCTTAAATTTAACGATCACTCATCATGT	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-4	This paper	5’-TTACTTATTA AATAATTTATAGCTATTG	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-5	This paper	5’-CAATAGCTAT AAATTATTTAATAAGTAAAATAAGAAACTGTAAACACAGTGTG	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-6	This paper	5’-CAGGAAACA GCTATGACCATGATTACGCCAGTTTAACTGTTCGGAAAGGGTGT	Deletion of Nte3-Nte5 locus
Sequence-based reagent	Nte345del-conf-F	This paper	5’-GTTTTCGTTGGTGAGGACGG	Confirmation of Nte3-Nte5 locus deletion
Sequence-based reagent	Nte345del-conf-R	This paper	5’-CTACTTATAATCCAAATATTTTATTGAACAGAGAAC	Confirmation of Nte3-Nte5 locus deletion
Sequence-based reagent	TssBsfGFP1	This paper	5’-CATGATTACGAATTCCCGGATTAATTAAAATGTCACGAAACAAATCATCCGG	tssB amplification to fuse with sfGFP and clone into pNCC1-spec
Sequence-based reagent	TssBsfGFP2	This paper	5’-CTGCTCGTCTTTGGAAGC	tssB amplification to fuse with sfGFP and clone into pNCC1-spec
Sequence-based reagent	TssBsfGFP3	This paper	5’-GCTTCCAAA GACGAGCAGGCAGCAGCAGGTGGTGGTAGCAAAGGAGAAGAACTTTTCAC	sfGFP amplification and addition of DNA linker to fuse with tssB and clone into pNCC1-spec
Sequence-based reagent	TssBsfGFP4	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCTCATTTGTAGAGCTCATCCATGC	sfGFP amplification and addition of DNA linker to fuse with tssB and clone into pNCC1-spec
Sequence-based reagent	sfGFP-Prom-F	This paper	5’-TGACCCGGG TCATTTGTAGAGCTCATCCATGCC	sfGFP amplification from pNCC1-sfGFP to clone into pNCC1-spec
Sequence-based reagent	sfGFP-Prom-R	This paper	5’-TGAAAGCTTTTGACAGCTAGCTCAGTCCTAGGTATAATGCTAGCCCAACATGTTACACAATAATGGAGTAATGAACATATGAGCAAAGGAGAAGAACT	sfGFP amplification from pNCC1-sfGFP to clone into pNCC1-spec
Sequence-based reagent	pGib-RBS-Nte2-F	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAAAGAAGGAGATATACCATGGCATTCAATAAAATCGCCC	Nte2 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte2-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATCATTTTTTCCTATTGTTACATTTATCCT	Nte2 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nti2-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATTATTCAAATTTCTTTAGCAGTATTTTTCT	Nte2 and Nti2 amplification plus addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte3-F	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAAAGAAGGAGATATACCATGGCCTCTTTCGGTAAC	Nte3 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte3-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATCATTTAATACCTCTTCTTGATAATTCTTT	Nte3 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nti3-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCACTATTCACCCAACAATGTTTCT	Nte3 and Nti3 amplification plus addition of RBS to clone into pBAD33
Sequence-based reagent	PGIB-RBS-NTE4-F	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAAAGAAGGAGATATACCATGGTCGAACACAACCAG	Nte4 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte4-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATTAAATTATTGGAAGATTTTTACAACCA	Nte4 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nti4-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATTACGCTTTTAAATTCCGGTG	Nte4 and Nti4 amplification plus addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte5-F	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAAAGAAGGAGATATACCATGGGTCGTCTGAAAAGC	Nte5 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte5-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCACTAATCTAATCGTTTGGGCG	Nte5 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nti5-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATTAATCCCAATAACTGTCTAAATTGT	Nte5 and Nti5 amplification plus addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte6-F	This paper	5’-GATCCTCTA GAGTCGACCTGCAGGCATGCAAAGAAGGAGATATACCATGGCCTCTTTCGGTAAC	Nte6 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nte6-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCACTATTATCTAGGAACAATCTGATTAATTATTCC	Nte6 amplification and addition of RBS to clone into pBAD33
Sequence-based reagent	pGib-RBS-Nti6-R	This paper	5’-AAAATCTTCTCTCATCCGCCAAAACAGCCATTAAATTTCCTCTAGTTTTTCTTTCATC	Nte6 and Nti6 amplification plus addition of RBS to clone into pBAD33
Sequence-based reagent	CE043-F	This paper	5’-GGCCGGTCT AGAAAGAAGGAGATATACCATGAAATACCTGCTGCCGACCGCTGCTGCTGGTCTGCTGCTCCTCGC	Addition of 5′ PelB leader peptide and 3′ 6xHIS-tag to PLA2 domain to clone into pBAD33
Sequence-based reagent	CE044-F	This paper	5’-GGTCTGCTG CTCCTCGCTGCCCAGCCGGCGATGGCCATGGGGGGAAGTAATTTTTATGCGTTTGCA	PLA2 domain amplification and addition of 5′ PelB leader peptide
Sequence-based reagent	CE046-R	This paper	5’-CCGGCCGCA TGCCTAGTGATGGTGATGGTGATGCCTATGATTTTTAGAC	Addition of 3′ 6xHIS-tag to PLA2 domain with or without 5′ PelB leader peptide to clone into pBAD33
Sequence-based reagent	CE047-R	This paper	5’-GATGCCTAT GATTTTTAGACGTTTTTTTAATTGTTTTATCG	PLA2 domain amplification with or without addition of 5′ PelB leader peptide
Sequence-based reagent	CE048-R	This paper	5’-CCGGCCGC ATGCCTAGTGATGGTGATGGTGATGATTAAGTTTGGATAGTTTGAAAATTTTTTTAAGCTTATATATAAG	PLA2 domain amplification with or without a 5′ PelB leader peptide and amplification of Nti1 adding a 3′ 6xHIS-tag to clone into pBAD33
Sequence-based reagent	CE083-F	This paper	5’-GGCCGGTC TAGAAAGAAGGAGATATACCATGGGGGGAAGTAATTTTTATGCGTTTGCA	PLA2 domain amplification and addition of 3′ 6xHIS-tag to clone into pBAD33
Sequence-based reagent	MW312	This paper	5’- TATAAGGAG GAACATATGGAATACATGTTATAATAACTATAAC	Spectinomycin cassette amplification from pDG1728 to clone into pNCC1
Sequence-based reagent	MW313	This paper	5’- GTATTCCATATGTTCCTCCTTATAAAATTAGTATAATTATAG	pNCC1 plasmid backbone amplification
Sequence-based reagent	MW314	This paper	5’- GCATCCCTTAACGACGTCAATTGAAAAAAGTGTTTCCACC	Spectinomycin cassette amplification from pDG1728 to clone into pNCC1
Sequence-based reagent	MW315	This paper	5’-TCAATTGACGTCGTTAAGGGATGCATAAACTGCATCCCTTAAC	pNCC1 plasmid backbone amplification