Abstract
The interactions between beneficial bacteria and pathogens are understudied. Here we investigate the interactions between the probiotic strain Bacillus subtilis PS-216 and the pathogen Salmonella Typhimurium SL1344. We show here that the sporulation of B. subtilis is impaired when it competes with S. Typhimurium in a nutrient-depleted medium. The sporulation impairment in B. subtilis is mediated by the sigma factor B (SigB)-dependent general stress response, as the ΔsigB mutant remains blind to manipulative cues from S. Typhimurium. Furthermore, we show that decreased sporulation frequency in B. subtilis depends on cell–cell contact between the two species involving the S. Typhimurium Type VI Secretion System, whereas B. subtilis uses the SigB-dependent response to trade spore quantity for higher spore quality.
Introduction
Microbial interactions are driving forces in microbial ecology [1] and play a crucial role in the development and management of biotechnological and biocontrol applications [2]. Although the number of studies addressing microbial social interactions is increasing, the role of interspecies interactions in shaping specific adaptive responses remains poorly understood.
Bacterial resistance to antibiotics emerged before their widespread use in medicine and agriculture. The increased use of antibiotics has driven the evolution of resistance mechanisms, which is especially problematic in pathogenic bacteria. As pathogens circulate between hosts and the environment, the environment has become a key reservoir of resistance genes, facilitating their acquisition by bacteria upon environmental exposure [3].
Salmonella enterica serovar Typhimurium (S. Typhimurium), a well-known pathogen, that causes diarrheal disease, is among the top most problematic carriers of antibiotic resistance [4]. Although S. Typhimurium is primarily recognized as a foodborne pathogen, it is also widespread in the environment, including soil [5], surface water [6], and plants [7]. Additionally, it can survive on abiotic surfaces causing issues in both industrial and domestic setting [8]. Survival outside a host requires rapid adaptation to various stress factors, such as low nutrient availability, enabling this pathogen to persist for a longer period and increasing its chances of re-entry into the food chain [5]. Studies have shown that S. Typhimurium can spread from soil to plants, including a variety of crops, posing the risk for consumers. Due to its wide host and habitat range Salmonella represents a major public health concern [9].
The use of probiotics offers an alternative strategy to combat the growing problem of antimicrobial resistant pathogens. Bacillus subtilis is a Gram-positive bacterium beneficial to plants and animals, known for its ability to produce various antimicrobial compounds [10]. It is mostly associated with soil and rhizosphere, but is also a member of the human and animal microbiota [11, 12]. B. subtilis is capable of inhibiting common pathogens such as Staphylococcus aureus [13], Campylobacter jejuni [14–16], and Salmonella Typhimurium [17]. In addition, the ability of B. subtilis to produce a variety of biologically active molecules, such as extracellular enzymes, vitamins, and antibiotics, expands the range of its applications [18], and makes it attractive for industrial use [19]. Moreover, as a spore forming bacterium, B. subtilis easily survives harsh environmental conditions through endospore formation. This resilience also contributes to its growing popularity as a probiotic for both humans and animals [20, 21].
Endospores are differentiated dormant cells that are highly resistant to a variety of environmental stressors. Spore development is induced when nutrients are lacking and growth is not favoured [22]. This highly regulated process is initiated via phosphorylation of the Spo0A transcription factor through a complex phosphorelay system [23, 24]. When Spo0A-P reaches a critical concentration, it binds to sporulation promoters and triggers the transcription of the associated sporulation genes [25]. After this stage, the cell is committed to spore development for several hours and is unable to grow until the sporulation process is completed. Therefore, B. subtilis has evolved numerous checkpoints and mechanisms to control premature commitment to sporulation [26]. For example, upon nutrient and energy stress, B. subtilis initially induces the general stress response (GSR) controlled by the sigma B transcription factor (SigB) (reviewed in [27]). SigB initiates a rapid response to environmental stressors [28–32] and temporarily inhibits sporulation by interfering with the Spo0A activity [33, 34]. However, starvation signals further evoke the Spo0A activity which in turn triggers sporulation in B. subtilis [35].
Recent work shows that interspecies competition can influence the sporulation of B. subtilis. For example, Pseudomonas chlororaphis PCL1606 stimulates sporulation in a contact-dependent manner by inducing specific histidine kinases, KinA and KinB, which are required for sporulation [36]. Siderophores pirated by B. subtilis from other species, such as Escherichia coli, also induce the sporulation process [37]. In general, studies have shown that B. subtilis upregulates sporulation in the presence of other bacteria, mainly for its protection against competitors [38–40].
Here we explore the influence of the common enteropathogen Salmonella Typhimurium [41] on the sporulation of B. subtilis. In contrast to prior studies utilizing other competitors, we show that Salmonella impairs the sporulation of B. subtilis and that in co-culture with S. Typhimurium, B. subtilis sporulation is affected through the SigB-dependent general stress response. Our results show that the impairment in sporulation is contingent on a sufficiently high Salmonella density and the presence of an active Type 6 Secretion System (T6SS) in Salmonella. These findings contribute to the understanding of the molecular links underlying social and adaptive strategies that shape the dynamics of microbial communities.
Materials and methods
Bacterial strains and strain construction
The strains used in this study are listed in Supplementary Table 1. Specific reporter fusions and gene deletions were inserted into the recipient strains by transformation. B. subtilis recipient strains were obtained using a standard transformation protocol in which the B. subtilis PS-216 [42] recipient strain was transformed with added DNA (plasmid or PCR product) [43]. Transformants were plated on LB agar plates with the appropriate antibiotics: erythromycin (Ery) 10 μg/ml, kanamycin (Kn) 50 μg/ml, or spectinomycin (Sp) 100 μg/ml.
B. subtilis PS-216 strains labelled with the red fluorescent protein mKate2, whose gene is linked to a constitutive Physpank or P43 promoter integrated into two different loci, amyE::Physpank-mKate2 [44–46] and sacA::P43-mKate2, were used.
To assay sporulation initiation, we generated the plasmid pEM1121 (Supplementary Table 2) carrying a Pspo0A-yfp reporter fusion. Briefly, the Pspo0A promoter region was PCR-amplified from B. subtilis PS-216 WT using the Pspo0A-F (EcoRI) and Pspo0A-R (HindIII) primer pair (Supplementary Table 3). The PCR product was then digested with EcoRI and HindIII restriction enzymes and ligated into the previously digested plasmid pKM3 [47] to obtain the plasmid pEM1121 (Supplementary Table 2). The plasmid pKM3 was digested with EcoRI and HindIII restriction enzymes to remove the PspoIIQ promoter region from the original vector. Plasmid pEM1121 was further transformed into B. subtilis PS-216 sacA::P43-mKate2 (BM1629 [48]), producing the B. subtilis BM1996 strain.
The B. subtilis PS-216 ΔsigB mutant strain was generated by amplifying the erythromycin inactivated sigB gene from the genomic DNA of B. subtilis BKE24610 using the 5pL-sigB and 3pR-sigB primers [49] (listed in Supplementary Table 3). The PCR product of the erythromycin inactivated sigB gene was then transformed into the recipient strains B. subtilis PS-216 WT and B. subtilis PS-216 amyE::Physpank-mKate2 (BM1097 [44]) to obtain the B. subtilis BM1930 and BM1931 strains, respectively. The plasmid pMS17 [48] was transformed into B. subtilis PS-216 ΔsigB to generate B. subtilis PS-216 ΔsigB sacA::P43-mKate2 (BM1992) reporter stain.
The plasmid pKM3 (Supplementary Table 2) carrying the PspoIIQ-yfp reporter fusion was transformed into B. subtilis PS-216 ΔsigB sacA::P43-mKate2 (BM1956) strain to obtain the BM2017 strain.
S. Typhimurium SL1344 (WT) [50] and S. Typhimurium ATCC 14028 [51] were fluorescently labelled via the gfpmut3 gene expressed from a plasmid [52, 53].
The S. Typhimurium SL1344 ΔclpV, Δhcp, ΔcpxA, ΔcpxP, ΔcpxR, and ΔrpoE mutant strains were generated by phage P22 transduction. Donor strains were obtained from the S. Typhimurium ATCC 14028 background [54]. Briefly, 500 μl of donor strains were mixed with different amounts of phage (5 μl and 20 μl) and ~ 6 ml of TopAgar containing 8 g/L LB agar, 5 mM CaCl2, and 10 mM MgSO4, and poured on LB agar. Plates were checked for the presence of plaques the next day and the TopAgar was scraped off and pushed through a syringe and needle (22 G). The samples were centrifuged for 10 min at 13 000 g and the supernatants were transferred to a new tube to which a few drops of chloroform were added. After precipitation, the upper, clear supernatant was transferred to a new tube and mixed with 500 μl of the recipient strain, S. Typhimurium SL1344 WT. After centrifugation for 5 min at 5000 g, the pellet was dissolved in 100 μl of LB supplemented with 5 mM CaCl2, 10 mM MgSO4, and different amounts of phage (5 μl and 20 μl). The mixture was plated on LB supplemented with 20 mM EGTA and kanamycin for S. Typhimurium SL1344 ΔclpV and Δhcp mutant strains or ampicillin for S. Typhimurium SL1344 ΔcpxA, ΔcpxP, ΔcpxR, and ΔrpoE mutant strains.
To determine the T6SS activity of Salmonella, we constructed a reporter fusion plasmid to measure the expression of clpV. Hereto, we inserted the promoter region of clpV upstream of the promoter-less gfpmut3 gene in the pFPV25 backbone [52] using primers described in [55]. The construct was verified by PCR and sequence analysis.
Growth conditions
To prepare overnight cultures, bacterial strains were grown in tryptic soy broth (TSB, Conda, Spain) supplemented with appropriate antibiotics at 37°C and shaken at 200 rpm for 16 h. The antibiotic concentrations in the medium were as follows: Cm 10 μg/ml, Kn 50 μg /ml, Ery 10 μg /ml, Sp 100 μg /ml, and Amp 100 μg /ml.
As described previously [17], overnight cultures were centrifuged for 10 min at 10 000 g, the supernatant discarded and the pellet re-suspended in 20-fold diluted (1/20) TSB medium. Suspensions of different S. Typhimurium strains were then diluted to OD650 ~ 0.1 absorbance units (a.u.), and suspensions of different B. subtilis strains were diluted to OD650 ~ 0.2 a.u. to obtain ~107 cells/ml. S. Typhimurium was mixed with B. subtilis in a 1:1 ratio (V:V) and 100 μl of each co-culture sample was transferred to the wells of a 96-well F-bottom microtiter plate and incubated further for 24 h (or other time points if indicated) at 37°C under static conditions. In experiments in which different cell densities of S. Typhimurium were tested against B. subtilis, the initial inoculum containing 107 cells/ml was diluted 10, 100, and 1000 times to obtain 106, 105, 104 cells/ml, respectively. Monocultures were prepared in a 1:1 ratio with 1/20 TSB medium as control. The prepared monoculture and co-culture samples had an initial density before dilution ~107 cells/ml for both bacteria, with the number of cells per species kept the same in the inoculum for monoculture and co-culture.
To estimate the cell numbers at the beginning and end of the experiment in monocultures and co-cultures, whole samples were disrupted by vigorous pipetting and vortexing. Disintegration of cell clumps into separate cells has been verified under the microscope.
Estimation of the sporulation frequency in B. subtilis
At different times of incubation, the total cell counts and the spore fraction of B. subtilis cells were determined by CFU/ml. To determine the number of spores, the samples were incubated for 30 min at 80°C and then plated on LB agar plates supplemented with the appropriate antibiotics. The sporulation frequency (SF) was then calculated by dividing the number of spores by the total cell count.
Isolation of spores and estimation of the relative spore survival in B. subtilis
B. subtilis WT and the ΔsigB mutant monoculture and co-culture samples were prepared as described above and incubated for 24 h at 37°C. To isolate the spores, the samples were incubated for 30 min at 80°C and then centrifuged three times for 10 min at 10 000 g. The supernatant was discarded and the pellet re-suspended in saline solution. Glycerol was then added in final concentration of 11% to store the spores at −20°C for further experiments.
The isolated spores were exposed to high temperature of 100°C for 30 min and then plated on LB plates. The relative spore survival was then calculated as the number of spores that survived the treatment divided by the total number of isolated spores before the treatment.
Determination of the type of competition between B. subtilis and S. Typhimurium
Determination of the effect of conditioned medium on the sporulation
S. Typhimurium SL1344 monoculture and co-culture of S. Typhimurium SL1344 with B. subtilis PS-216 were prepared as described above and incubated for 24 h at 37°C. After incubation, the samples were centrifuged at 10 000 g for 10 min. To prepare conditioned medium, the supernatants were filtered through a filter with a pore diameter of 0.2 μm. Finally, the conditioned medium was mixed at a 1:1 (V/V) ratio with the B. subtilis strain prepared as described previously. The SF was calculated after 24 h of incubation.
Determination of the effect of lysed S. Typhimurium cells on the sporulation
Overnight culture of S. Typhimurium SL1344 was prepared in 1/20 TSB medium at 37°C and 200 rpm and then autoclaved at 121°C. The autoclaved S. Typhimurium culture was co-cultured with B. subtilis strain as described above. The SF was calculated after 24 h of incubation.
Transwell assay
The cultures were prepared as described above. An insert with a translucent PET membrane with a pore diameter of 0.4 μm was inserted into the tested wells (6-well plates, Greiner). First, 2.5 ml of S. Typhimurium culture was transferred into the well and then an insert containing 2.5 ml of B. subtilis culture was immersed into the well. The SF was calculated after 24 h of incubation.
Visualization of B. subtilis sporulation
After 16 and 24 h of incubation, the samples were first disrupted by pipetting and 15 μl of each sample was transferred to glass slides (10 wells/ 6 mm). The slides were previously prepared by adding 10 μl of 0.05% poly-L-lysine to the well and air-dried at room temperature for 1.5 h as previously described [56]. After 15 min of samples incubation at room temperature, the excess liquid was aspirated. The samples were then rinsed with 15 μl of PBS buffer to remove unattached cells and dried for 10 min at room temperature. Finally, 2 μl of SlowFade Gold reagent was added to prevent fluorescence fading and the slides were covered with coverslips.
Visualization of B. subtilis PS-216 cells constitutively labelled with P43-mKate2, carrying PspoIIQ-yfp was performed. DIC microscopy allowed visualization of unlabelled S. Typhimurium cells in co-culture samples (not shown). Excitation of YFP was performed at 488 nm. The variable dichroics were set to reflect the emitted light in the range from 400–580 nm, which allowed to effectively capture fluorescence of YFP in the range of 500 to 580 nm by GaAsP PMT detector. Excitation of the red fluorescence protein (mKate2) was performed at 561 nm and the emitted fluorescence was captured in the range of 580 to 700 nm by GaAsP PMT detector. The laser intensities and GaAsP PMT detector gain were 4.5% and 650 V and 4% and 800 V for YFP and mKate2, respectively. The pinhole size was 55 μm. The wells on the slide were scanned with a 100 X / 0.4 N.A. objective and the captured images were 1024 × 1024 pixels with 16-bit colour depth.
Zen 2.3 Software (Carl Zeiss) was used for image processing and image noise was reduced by a single pixel filter (threshold = 1.5).
Expression of different reporter strains
Bulk fluorescence measurements
The strains were prepared as described above in and then inspected as described previously [17]. Briefly, 100 μl of monocultures and co-cultures were allocated to at least three technical replicates in the wells of a 96-well black transparent bottom microtiter plate and incubated for 24 h at 37°C statically in a Cytation 3 imaging reader (BioTek, USA). To monitor gene expression, the fluorescence intensity of YFP was measured by excitation at 500 nm and emission at 530 nm, whereas the fluorescence intensity of mKate2 with excitation at 570 nm and emission at 620 nm. The gains of both fluorescence intensities were set to 100 and were measured every half hour. In unmarked samples, we measured both fluorescence intensities to obtain background values. Both values were then subtracted from the corresponding fluorescence intensities of the tagged samples. The mKate2 fluorescence intensity of the monoculture that represented the maximum value was set to a value of 1 and the remaining mKate2 fluorescence intensities were divided by the highest mKate2 fluorescence intensity. The YFP fluorescence intensity was then divided by the normalized mKate2 fluorescence intensity to obtain the fraction of activated cells.
Single cell measurements
To measure the activity of the PspoIIQ at the single-cell level, static monocultures and co-cultures in 1/20 TSB medium were prepared as described above. Three biological replicates were prepared for each promoter and each biological replicate consisted of two duplicates. The PspoIIQ activity was measured after 16 h and after 24 h. Prior to this, the samples were disrupted by vigorous pipetting and vortexing. Afterwards, 10 000 cells were measured with FACS Melody Cell Sorter (BD). Gene expression was analysed with FlowJo software. First, singlets were identified based on the forward and side scattering. Second, B. subtilis cells were identified based on the red fluorescence of constitutively expressed
P43-mKate2. Subsequently, the activity of the PspoIIQ was quantified by measuring the fluorescence intensity of the yellow fluorescent protein [17].
Data presentation and statistical analysis
All experiments were performed in at least three independent biological replicates. The datasets obtained by CLSM are presented as one of the most representative images using Zen imaging software. Furthermore, statistical analysis and data presentation were performed in Origin program. Mean values and standard deviation (or standard error) were calculated from the obtained datasets and presented in the figures. Student’s t-test or one-way analyses of variance (ANOVA) with accompanying Tukey's post hoc test were used to compare means, and compared samples that showed a P value <0.05 were considered statistically different.
Results
B. subtilis sporulation is impaired in co-culture with S. Typhimurium
Nutrient-depleted conditions lead to the formation of metabolically inactive B. subtilis spores [57] (Fig. 1A). This adaptive process has also been reported to serve as a defence mechanism during competition with other bacteria [36–40]. Here we focus on the sporulation frequency (SF) of B. subtilis in a previously characterized interaction with S. Typhimurium SL1344 [17]. We performed the experiments in nutrient-depleted medium (1/20 TSB), which promotes B. subtilis sporulation. Our results showed that B. subtilis PS-216 co-cultured with S. Typhimurium SL1344 exhibited a lower SF compared to monoculture conditions. This phenomenon was also observed in other B. subtilis strains, namely the laboratory wild-type strain B. subtilis NCIB 3610 and two soil isolates, B. subtilis PS-218 and B. subtilis PS-196 (Fig. 1B). Moreover, in addition to S. Typhimurium SL1344, S. Typhimurium ATCC 14028 also decreased SF of B. subtilis PS-216 (Fig. 1B). Therefore, impaired sporulation is a widespread and not a strain specific phenomenon, occurring across different B. subtilis and S. Typhimurium strains. Cell number may also play an important role in interspecies interactions as some effects may be cell density dependent. S. Typhimurium reaches similar densities in co-cultures with different B. subtilis strains in 1/20 medium (Fig. 1C). However, SF was not decreased if the S. Typhimurium inoculum was diluted 100- or 1000-fold (Fig. 1D), suggesting that impaired sporulation by S. Typhimurium is cell density dependent and does not occur at low pathogen densities.
S. Typhimurium impairs B. subtilis sporulation in a density-dependent manner. (A) B. subtilis shifts between two lifestyles: It grows when nutrients are available and during starvation it differentiates into dormant spores. (B) Sporulation frequency (SF) of different B. subtilis strains in monoculture and co-culture with S. Typhimurium SL1344 and ATCC 14028. (C) Cell counts of S. Typhimurium SL1344 in monoculture and co-culture with different B. subtilis strains. (D) SF of B. subtilis PS-216 in monoculture and in co-culture with undiluted (107 cells/ml), 10x diluted (106 cells/ml), 100x diluted (105 cells/ml) and 1000x diluted (104 cells/ml) initial inocula of S. Typhimurium SL1344. The SF for B. subtilis and cell counts for S. Typhimurium were determined after 24 h of static incubation at 37°C in 1/20 TSB medium. Data for (B), (C), and (D) are presented as mean values and error bars represent standard deviation of the mean values of three biologically independent experiments. Student’s t-test was performed to compare the means of SF and cell counts between monoculture and co-culture for (B) and (C) where statistically significant differences were determined (*P < 0.05). One-way ANOVA with Tukey's post hoc test (P < 0.05) was performed to compare the means of SF between monoculture and different co-cultures for (D), where different letters above the columns indicate a significant difference between samples.
We also determined the sporulation dynamics (Fig. 2A, Supplementary Fig. 1). Sporulation of PS-216 started after 10 h of incubation in both monoculture and co-culture, but consistently reached higher SFs in monoculture conditions, reaching up to 60% at the end of the experiment. In contrast, in the presence of Salmonella, sporulation stagnated around 20%. Total cell counts and spore counts are shown in Supplementary Fig. 1. To visualize and measure sporulation dynamics, the B. subtilis PS-216 PspoIIQ-yfp P43-mKate2 reporter strain [48] was grown with S. Typhimurium or alone, using the expression of the spoIIQ gene (expressed in the forespore [58]) as a sporulation marker. This fluorescent reporter strain showed similar monoculture and co-culture sporulation dynamics at the population level as our previous assay using traditional plating methods, confirming the utility of this reporter strain for determining sporulation dynamics (Fig. 2B). Subsequently, we measured and visualized the PspoIIQ-yfp activity at the single cell level after 16 h and 24 h of incubation using flow cytometry and confocal laser scanning microscopy (Fig. 2C, Fig. 2D). These assays confirm that sporulation genes were induced in smaller subpopulation in co-culture compared to monoculture.
Influence of S. Typhimurium on B. subtilis PS-216 sporulation dynamics. (A) B. subtilis SF was determined after 10, 16, 18, 20 and 24 h of static growth at 37°C in 1/20 TSB medium for monoculture and co-culture. (B) Bulk measurements of the PspoIIQ-yfp promoter activity in B. subtilis PS-216 monoculture and co-culture with S. Typhimurium SL1344. Only one out of three biologically independent experiments is shown on the panel (performed in six technical replicates). Measurements were performed every half hour for 24 h in 1/20 TSB medium. Data are presented as mean values and error bars represent standard error of the mean values. (C) Flow measurements and (D) microscopic images of single-cell transcriptional activity of the PspoIIQ-yfp of B. subtilis PS-216 in monoculture and co-culture with S. Typhimurium SL1344 were performed after 16 h and 24 h of static growth at 37°C in 1/20 TSB medium. Data are presented as mean values and error bars represent standard deviation of the mean values of three biologically independent experiments. Student’s t-test was performed to compare the means of SF and the proportion of PspoIIQ-yfp expressing cells between monoculture and co-culture, where statistically significant differences were determined (*P < 0.05). One representative repeat of at least three biologically independent experiments (performed in at least two technical replicates) is shown for both (C) and (D). In the microscopic images, the scale bar represents 5 μm.
To test whether S. Typhimurium negatively affects sporulation before stage II, we constructed a reporter strain carrying the promoter of the master regulator spo0A fused to YFP. The Pspo0A-yfp transcriptional activity was initially higher in the monoculture than in the co-culture, confirming that the sporulation impairment occurs early during spore development. As expected, at later stages, from 17 h onwards, when Pspo0A-yfp activity in monoculture dropped rapidly, the Pspo0A-yfp activity in the co-culture also dropped, but not as intensely as in monoculture (Supplementary Fig. 2).
S. Typhimurium affects sporulation through the transcriptional factor SigB leading to higher spore quality
In B. subtilis, the alternative SigB is responsible for sporulation inhibition under various stress conditions, including nutrient-limitation and energy stress. SigB acts at an early stage of spore development by interfering with Spo0A-P-dependent effects [33, 34]. SigB controls the GSR, which is a rapid and reversible response to different stressors including low nutrients [30, 59, 60]. In contrast, sporulation is an irreversible response following the induction of sporulation stage II genes [61]. Therefore, we tested whether competition with Salmonella affects sporulation via a SigB-dependent pathway. We compared the SF of the B. subtilis PS-216 WT and ΔsigB mutant in monoculture and co-culture. S. Typhimurium did not impair sporulation in the PS-216 ΔsigB mutant (Fig. 3A), indicating that pathogen-mediated disruption of sporulation is SigB-dependent. Furthermore, the ΔsigB mutant did not reduce the growth of the pathogen in co-culture, confirming that the lack of negative impact on sporulation is not due to a lower cell counts of the pathogen (Fig. 3B). The spore count and the total number of ΔsigB cells of B. subtilis is shown in Supplementary Fig. 3.
Impairment of B. subtilis PS-216 sporulation is dependent on the SigB transcription factor. (A) SF of the B. subtilis PS-216 ΔsigB mutant in monoculture and co-culture with S. Typhimurium SL1344. SF was determined after 24 h of static incubation at 37°C in 1/20 TSB medium (B) cell counts of S. Typhimurium SL1344 in monoculture and co-culture with B. subtilis PS-216 ΔsigB mutant after 24 h of static incubation at 37°C in 1/20 TSB medium. Data for (A) and (B) are presented as mean values and error bars represent standard deviation of the mean values of three biologically independent experiments. Student’s t-test was performed to compare the means of SF and cell counts between monoculture and co-culture, where statistically significant differences were determined (*P < 0.05). (C) Bulk measurements of the PspoIIQ-yfp promoter transcriptional activity in B. subtilis PS-216 ΔsigB mutant monoculture and co-culture with S. Typhimurium SL1344. Only one out of three biologically independent experiments is shown on the panel (performed in six technical replicates). Measurements were performed every half hour for 24 h in 1/20 TSB medium. Data are presented as mean values and error bars represent standard error of the mean values. (D) B. subtilis PS-216 WT and the ΔsigB mutant relative spore survival in monoculture and co-culture with S. Typhimurium SL1344. Spore survival was determined after 30 min of incubation at 100°C and was divided by total spores. Data are presented as mean values and error bars represent standard deviation of the mean values of four biologically independent experiments. Student’s t-test was performed to compare the means between monoculture and co-culture, where statistically significant differences were determined (*P < 0.05).
To test whether SigB affects the expression of sporulation genes, we introduced the PspoIIQ-yfp reporter into the PS-216 ΔsigB background and tested promoter activity in monoculture and co-culture with S. Typhimurium. Again, we did not observe any repression of the PspoIIQ promoter activity in the mutant. In contrast, the promoter activity in the PspoIIQ-yfp reporter in the co-cultured mutant was initially comparable to that in monoculture, and then even started to increase during the last stages of incubation (Fig. 3C). These results support the conclusion that S. Typhimurium impairs sporulation via a SigB-dependent pathway.
Recent reports show that B. subtilis sporulation can follow a quality-quantity trade-off [62, 63]. We test here whether the competition between B. subtilis and S. Typhimurium can also affect spore quality. We defined quality as the durability of spores at high temperatures (100°C). Therefore, we isolated spores after 24 h from monoculture and co-culture samples of B. subtilis WT (sporulation impairment in co-culture) and the ΔsigB mutant (no sporulation impairment). We then exposed the spores to 100°C for 30 min. The results showed that the WT spores produced in co-culture had a higher survival frequency at high temperatures than the spores from the monoculture. In addition, the increase in spore quality was dependent on SigB, as the spores of the co-cultured ΔsigB mutant do not show an increased tolerance to heat (Fig. 3D).
Interactions between B. subtilis and S. Typhimurium are contact-dependent
We aimed to understand how S. Typhimurium triggers the SigB-dependent effect on sporulation frequency. We hypothesized that S. Typhimurium might act through either a) secretion of diffusible factors, b) shedding of its own cell components that act as a signal, or c) contact-mediated mechanisms. To test these predictions, we cultured B. subtilis PS-216 with conditioned medium from either S. Typhimurium SL1344 monoculture or co-culture, as the presence of B. subtilis may alter the compounds secreted by Salmonella. We also tested autoclaved (heat killed) S. Typhimurium SL1344 cells to evaluate the impact of the cell wall components on sporulation. Finally, we tested the effects of cell–cell contact by co-culturing both species in a transwell setup, where both bacteria were separated by a semi-permeable membrane. The results show that only live Salmonella decreased SF of B. subtilis, whereas conditioned medium or heat killed S. Typhimurium did not lower SF. Moreover, preventing cell-to-cell contact between the pathogen and B. subtilis also abolished the decrease in SF (Fig. 4A). Therefore, we conclude that direct cell-to-cell contact between viable bacteria is required for the sporulation impairment phenotype. The total cell counts and spore counts for the different co-cultures are shown in Supplementary Fig. 4.
Direct cell-to-cell contact and S. Typhimurium T6SS is needed for sporulation impairment. (A) SF of B. subtilis PS-216 in monoculture and co-cultures with live and heat-killed S. Typhimurium SL1344, conditioned medium and in transwell assay. SF was determined after 24 h of static incubation at 37°C in 1/20 TSB medium. (B) SF of the B. subtilis PS-216 in monoculture and co-culture with S. Typhimurium SL1344 WT strain and mutants in T6SS and envelope stress response and (C) cell counts of S. Typhimurium SL1344 WT strain and mutants in T6SS and envelope stress response. SF and cell counts were determined after 24 h of static incubation at 37°C in 1/20 TSB medium. Data for (A), (B), and (C) are presented as mean values and error bars represent standard deviation of the mean values of three biologically independent experiments. One-way ANOVA with Tukey's post hoc test (P < 0.05) was performed to compare the means of SF between monocultures and co-cultures for (A) and (B), where different letters above the columns indicate a significant difference between samples. Student’s t-test was performed to compare the means of SF and cell counts between monoculture and co-culture for (C), where statistically significant differences were determined (*P < 0.05). (D) Bulk measurements of the PspoIIQ-yfp promoter transcriptional activity in B. subtilis PS-216 monoculture and co-culture with the S. Typhimurium SL1344 ΔclpV and Δhcp mutants. Measurements were performed every half hour for 24 h in 1/20 TSB medium. Only one out of three biologically independent experiments is shown on the panel (performed in six technical replicates). Measurements were performed every half hour for 24 h in 1/20 TSB medium. Data are presented as mean values and error bars represent standard error of the mean values.
We speculated that the contact-dependent Type 6 Secretion System (T6SS) might contribute to sporulation impairment. The T6SS, a complex contractile needle-like system, is used by Gram negative bacteria for the injection of toxins into competitors [64, 65]. To test this hypothesis, we co-cultured B. subtilis PS-216 with T6SS-defective Salmonella mutants and measured the SF of PS-216 after 24 h of incubation. These mutants with a deletion in clpV or hcp, respectively encoding the ATPase and the inner tube of the T6SS [64], did not decrease SF (Fig. 4B).
It has been shown that T6SS expression is mediated by the envelope stress response in E. coli and Citrobacter rodentium [66, 67]. S. Typhimurium has various systems that respond to changes to the membrane [68], with the best studied being the σE and the two component Cpx system. σE responds to stress at the outer membrane, whereas Cpx recognizes stress at the inner membrane [69]. In the Cpx system, CpxA is the sensor histidine kinase, whereas CpxR is the response regulator [70]. The third component of the Cpx system, CpxP, inhibits the activity of CpxA [71]. S. Typhimurium mutants defective in the envelope stress response (i.e. ΔcpxA, ΔcpxP, ΔcpxR, ΔrpoE) were also unable to reduce B. subtilis SF (Fig. 4B). The deletion in genes for T6SS and envelope stress response in S. Typhimurium result in a similar number of B. subtilis cells in co-culture (Supplementary Fig. 5).
Because the results can also be explained by reduced growth of the Salmonella mutants in co-cultures, we determined their cell density in co-cultures. Mutants ΔclpV and ΔcpxA were found to have reduced growth in co-cultures (Fig. 4C). However, other mutants, including T6SS-related mutants, showed comparable growth in co-culture to wild-type S. Typhimurium (Fig. 4C). This suggests that a complex, contact-dependent interplay between B. subtilis and S. Typhimurium, involving the pathogen’s T6SS and some envelope stress response genes, may be responsible in the sporulation impairment phenotype.
We then monitored the PspoIIQ-yfp expression of B. subtilis in co-culture with S. Typhimurium ΔclpV and Δhcp mutants. The results showed that the PspoIIQ-yfp expression was comparable to that in the monoculture during the first 16 h. However, thereafter we detected an increase in the PspoIIQ-yfp expression in the co-culture with the two S. Typhimurium mutants (Fig. 4D). These results further confirm our findings that the T6SS-defective S. Typhimurium mutants do not impair B. subtilis sporulation.
Because B. subtilis requires iron for sporulation [72], we hypothesized that iron limitation may contribute to the T6SS-mediated sporulation impairment. To test this, we added FeCl₃ (0.05 and 0.1 mM) to 1/20 TSB medium and measured sporulation frequency. Under iron-rich conditions, sporulation of B. subtilis in co-culture with S. Typhimurium was no longer impaired (Supplementary Fig. 6).
Discussion
S. Typhimurium is a good biofilm former, which allows it to survive in different environments [8] including non-host environments [5]. Its widespread environmental persistence poses a major challenge to pathogen control and contributes to the spread of antimicrobial resistance [41]. The persistence of Salmonella in soil has been found to largely depend on soil microbiota diversity, where greater diversity inhibits Salmonella invasion and abundance, whereas lower diversity promotes it [73]. Understanding the ecological perspective of S. Typhimurium and its interactions with other soil microorganisms, including B. subtilis, may contribute to understanding S. Typhimurium lifestyle outside the host and thus reduce the associated public health risks.
The endospore development of B. subtilis has been intensively studied for decades as spores are among the most resilient life forms on the planet. Endospores confer B. subtilis with increased tolerance to a range of stressors, including low and high temperatures, UV and γ-radiation, desiccation, harmful chemicals, predation, and nutrient depletion [22, 74, 75]. Therefore, sporulation is mainly induced in stress environments, including nutrient-depleted conditions. However, once B. subtilis enters stage II of sporulation (asymmetric septum formation), it cannot resume vegetative growth before sporulation is completed [61]. Consequently, B. subtilis has evolved several mechanisms to prevent premature entry into sporulation, which include activation of spo0A transcription and Spo0A phosphorylation [76, 77]. In addition, rapid entry into sporulation upon nutrient limitation is prevented via SigB. This stress response system senses nutrient and energy stress and postpones sporulation by acting as a transcriptional inhibitor of spore development [34].
Despite the importance of sporulation, knowledge of how interspecies social interactions influence spore development is still limited. Most published work highlights the induction of sporulation as a defence mechanism of B. subtilis against competitors such as Myxococcus xanthus, E. coli, and P. chlororaphis [36–40]. In contrast, our work shows that competition with the enteropathogen S. Typhimurium leads to an impairment in sporulation. The sporulation impairment requires pathogen’s T6SS system and close cell–cell interactions. Although the experiments are performed in liquid medium, under static conditions, both species form submerged biofilm enable mixing and thus facing the close contact between competitors [17]. This is consistent with our observation that T6SS of S. Typhimurium is involved in sporulation impairment and that the phenotype depends on cell–cell contact. T6SSs are found in 25% of Gram-negative species capable of releasing diverse effectors into bacterial and eukaryotic cells [78]. However, the knowledge of the T6SS attack mechanisms against Bacillus is limited [40, 79]. Recent study shows that the plant beneficial bacterium P. chlororaphis delivers the Tse1 toxin via T6SS to B. subtilis cells. Tse1 is a hydrolase that degrades Bacillus peptidoglycan and indirectly damages Bacillus membrane functionality and activates the SigW dependent pathway, which promotes sporulation. It was proposed that this response permits B. subtilis to defend against the toxicity of T6SS-mobilized Tse1 effector [40]. In contrast, the T6SS of S. Typhimurium induces the SigB-dependent general stress response in B. subtilis, impairing the commitment to sporulation with concomitant increase in spore quality. S. Typhimurium SL1344 strain we used here has only one known effector, Tae4, which is an amidase that cleaves the γ-d-glutamyl-l-meso-diaminopimelic acid amide bond of peptidoglycan [64]. Because the deletion of the clpV and hcp genes, which are essential for a functional T6SS, did not alter the fitness of either S. Typhimurium or B. subtilis in co-culture, we cannot claim at this point that T6SS mediates the attack of B. subtilis via toxin release. In Fig. 5, we summarized the current knowledge on interspecies interactions between P. chlororaphis or S. Typhimurium with B. subtilis and their impact on B. subtilis sporulation.
Model of the B. subtilis survival strategies of in competition with P. chlororaphis and S. Typhimurium. During interaction with P. chlororaphis on LB agar, B. subtilis responds to Tse1-T6SS-mediated envelope disturbance by upregulating sporulation via SigW, KinA, and KinB. Under nutrient-depleted conditions in 1/20 TSB, B. subtilis sporulates in response to starvation. However, co-cultivation with S. Typhimurium impairs sporulation. This effect is mediated by the T6SS of S. Typhimurium, which in an unknown manner activates the SigB-dependent general stress response, resulting in formation of fewer but higher quality spores. Created in BioRender. Podnar, E. (2025) https://BioRender.com/x41m214.
From an evolutionary perspective, sporulation is a conserved process in bacteria [80]. However, according to our results, prompt sporulation does not seem to be beneficial in the competition with S. Typhimurium. The impairment in sporulation gives B. subtilis an ecological advantage as the higher quality of the spores is ensured by affecting the SigB-dependent pathway. Although the deletion of the sigB gene does not negatively affect the growth of either species, it seems that low quantity and high-quality trade-off in spore development is an adaptive strategy of B. subtilis and less likely a manipulative strategy of S. Typhimurium to gain a competitive advantage. Although, we cannot completely rule out that S. Typhimurium rips benefits form inhibiting sporulation during prolonged co-incubations spanning beyond 24 h. The trade-off between spore quality and quantity has recently been hypothesized to promote the survival of B. subtilis under harsh conditions [62, 63]. As a result, the trade-off may result in B. subtilis having a better chance of reproducing when conditions improve. The quality-quantity trade-off is also consistent with our observation that SigB plays an important role in competition between the two species. For example, the ΔsigB mutants of Bacillus cereus have been shown to produce low-quality spores [81]. Thus, by impairing sporulation, SigB-dependent GSR also mediates a trade-off between quality and quantity during competition with S. Typhimurium. Also, B. subtilis interactions with fungus Fusarium verticillioides induce SigB-dependent stress response, resulting in the surfactin mediated antagonism against the fungus [82]. This result suggests that SigB-dependent regulatory response to competition may be more-wide spread, but more research is needed to clarify whether it is linked to quality-quantity trade-off, we observed in interactions with S. Typhimurium.
External conditions, such as medium composition, significantly influence spore properties. Divalent cations (Mn2+, Ca2+, Zn2+, Mg2+) enhance spore stability and yield [83]. It is likely that interspecies competition changes the availability of essential nutrients, including ions. This could be sensed by B. subtilis SigB activating pathway and consequently affects the spore quantity-quality trade off. In line with these findings, T6SS also plays a role in ion sequestration in some bacterial species [84]. Although the T6SS role in ion sequestration by S. Typhimurium needs to be tested in the future, it is intriguing that addition of iron to the nutrient limited medium bypasses the sporulation impairment triggered by S. Typhimurium. This suggests that competition for iron between S. Typhimurium and B. subtilis could alter the competition dynamics and consequently affect adaptive responses of B. subtilis (for discussion see Supplementary Fig. 6 and associated text).
At this point exact environmental factors and molecular cues triggering the S. Typhimurium T6SS dependent activation of the SigB general stress response in B. subtilis remain unclear. It will be important to test the role of specific T6SS effectors and/or nutrient signals to better define the molecular cues exchanged during interspecies competition.
Overall, our results reveal a response of B. subtilis to interspecific competition and nutrient limitation, in which it favours growth over sporulation at an earlier stage. Although sporulation has been considered as the primary defence mechanism of B. subtilis, here we show that co-culture with S. Typhimurium stimulates B. subtilis to adopt a strategy to impair sporulation. This work reveals new insights into the intricate dynamics of bacterial interactions dictated by SigB and the GSR of B. subtilis, leading to a change in the commitment of B. subtilis to sporulation and the production of higher quality spores during interspecies competition. This could be of great importance for understanding the ecology of this bacterium within microbial communities under nutrient-limited conditions.
Acknowledgements
We acknowledge the support of the University’s infrastructural centre for Microscopy of biological samples located at the Biotechnical Faculty, University of Ljubljana. We also thank the National Bioresource Project (NIG, Japan): B. subtilis for providing us with the BKE library. We acknowledge Miona Kovachevikj and Maruša Požar for their technical assistance.
Author contributions
EP, TD, IMM, and HS designed research. EP, KD, and BL performed the experiments. EP, EK, TD, and BL contributed to data analysis. All authors discussed the results. EP drafted the manuscript. TD, IMM, BL, and HS revised the manuscript.
Conflicts of interest
All authors have declared that there is no conflict of interest.
Funding
We acknowledge the Slovenian Research and Innovation Agency (ARIS) for funding of the ARIS project J4–4550; the research program grant P4–0116; the young research program grant ARRS plus, awarded to IMM. We also acknowledge the KU Leuven research fund (CELSA/18/031 and C24/18/046) and Research Foundation Flanders (FWO) (grant S004824N).
Data availability
All the data are provided in the published article and its supplementary information files. Raw data of this study have been deposited in Figshare: https://doi.org/10.6084/m9.figshare.26526805
