Introduction

Versatile adaptability of Lactobacilli into diverse ecosystems defines the uniqueness of these species1. While these bacteria are frequently associated with dairy products, invertebrates, and craniate, they also form a significant part of human microbiota2. These species are known to stimulate immunomodulatory effects and evince belligerent traits against deleterious bacterial or fungal pathogens3,4,5. Specifically, a probiotic bacterium Lactobacillus plantarum displays diverse morphological phenotypes and a remarkable ability to acclimatize to external settings, qualify it for wide-ranging medicinal and industrial applications5.

Undoubtedly, L. plantarum is a prospective probiotic bacterium in the food and supplements industry6. The capability of multicellular behaviour by this bacterium appears to be an integral factor contributing to its applicability in the food matrices as well as within human host7,8. In addition, L. plantarum successfully colonizes intestinal mucosa and co-exists with other beneficial bacteria in the human gut8. Collectively, probiotics and the indigenous bacterial populations alter the gut microbiome’s overall structure that offers multifarious health benefits to the human host5. For instance, establishment of the commensal bacteria is often thought to either disassemble the preformed biofilms of gut pathogens9 or combat them by triggering host immune response10. Furthermore, probiotics may ameliorate several morbid conditions of metabolic disorders, such as diabetes and obesity, as well as several gut disorders. Some of the probiotics function as psychobiotics as well, yielding cognitive benefits11.

L. plantarum is one of the unique species of Lactobacillus genus known to exhibit multifaceted population heterogeneity12. The cells of L. plantarum manifest a strong ability to auto-aggregate depending on external environments and nutrient availability13. Auto-aggregation or co-aggregation is an exciting adaptation tactic that potentiates the probiotic cells to combat harmful pathogens. A probiotic Lactobacillus gasseri was shown to adhere to the human intestinal mucosa by auto-aggregation, creating a protective blanket, which prevents pathogen colonization. In addition, several strains of L. plantarum are routinely assessed for their auto-aggregation abilities citing its effectiveness for pathogen control14.

Tolerance to acidic conditions is yet another adaptation response manifested by Lactobacilli to curtail pathogen growth in a microbiome. In general, cells of L. plantarum are well-adapted to withstand diverse pH-induced stresses. They grow well at sub-optimal pH conditions (pH 5.0), which may trigger a pre-adaptation response eliciting the cells to swiftly adjust to the stressful environment in a better way15. Other Lactobacilli are reported to tolerate extremely low-pH stresses by arresting growth and favouring colonization. For instance, L. acidophilus cultured in pH 3 revealed enhanced ability to survive and adhere to human intestinal cells16. Other studies have shown that the low pH (3.5–4.5) triggers L. plantarum upsurge in the human vaginal microbiome17. The proliferation, in turn, aids L. plantarum to exhibit complete dominance during competition within the vaginal microbiota. Overall, phenotypes displayed by L. plantarum in response to acidic-pH may account for the remarkable adaptability to survive and combat pathogens thereof, which seems sturdily connected to its origins and distribution.

Recently, the genomic analysis revealed that the genus Lactobacillus relishes three different forms of lifestyles: free-living, nomadic, and host-adapted2,5. The nomadic lifestyle, for instance, enables Lactobacilli to grow in diverse environments2. Thus, understanding the nomadic ancestry of L. plantarum might provide a clearer insight into cellular heterogeneity and dynamic adaptability. This understanding can be realized by examining the animal-derived probiotic model, as it tends to manifest seemingly unique traits like enhanced digestive enzymes and superior antagonistic activities18. Their subtypes are likely dispersed and drifted into different ecosystems (via animal excrement), through paving a way for further evolution and adaptation.

In this study, we characterize morphological traits of the multicellular population of L. plantarum through a unique geometrical arrangement facilitating its survivability under different stress conditions. Furthermore, we provide mechanistic evidence of how the acidic-pH adaptation of this bacterium is tightly linked to its vast antagonistic potential against pathogenic microbial species, for intense Candida albicans, in an in vivo Caenorhabditis elegans model.

Results

Spatiotemporal establishment of conic colonies is governed by an adaptation response of L. plantarum cells to acidic-pH

This investigation was initiated following the observation of unusual cone-shaped colonies formed by L. plantarum on the MRS air–agar interface, which were triggered by acidic pH (Fig. 1). The highly structured small colonies expanded radially and aerially, reaching finite size with a circumferential diameter of 0.416 ± 0.02 cm (Supplementary Fig. S1a); growth ceased despite the availability of nutrients in the vicinity. The most straightforward interpretation of the observed phenomenon was that the L. plantarum cells found a way for aerial expansion within the colony during the growth in acidic pH. On average, a 7d old colony measured an aerial height of 0.3 cm ± 0.16 cm. We also noted the brown-coloured deposits that were apparent on the outer surfaces of these colonies (Fig. 1a and supplementary Fig. S1b). The intensity of the brown coloration increased with incubation time (Supplementary Fig. S2a), which indicated a possible correlation with cellular survivability (Supplementary Fig. S2b). The numbers of dead cells were significantly higher in 15d old colonies than observed in 5 or 7d colonies (Supplementary Fig. S2c). The mechanism of cell death in colonies could be attributed to ROS accumulation and instantaneous cell death (Supplementary Fig. S3). Furthermore, the colonies formed at pH 7 were devoid of brown deposits, resembling a frustum without a conical tip and the reduced colony height (Fig. 1b, c). Despite the phenotypic heterogeneities during growth in altered pH, the viable cell counts remained comparable within the tested colonies (Fig. 1d). These observations indicated that the brown deposits and conic tip formed at pH 5.5 are colony structures developed by the L. plantarum cells during adaptation to acidic-pH.

Fig. 1: Spatiotemporal establishment of conic-shaped colonies in L. plantarum.
figure 1

a Colonies of L. plantarum 3297 were generated onto the MRS hard agar plates (pH 5.5) following 2 and 7d incubation at 37 °C. Scale bar: 0.2 cm. b Estimation of the L. plantarum 3297 colony heights generated on the MRS hard agar plates with different pH conditions. The graph shows the means ± SEMs of 6 individual data points, obtained from two independent measurements. *P < 0.05 vs. the control. Here, the colonies grown at pH 5.5 are taken as control compared to those grown at pH 7. c Effect of ΔpH on conic tip formation and colouration of the L. plantarum colonies. White and black arrowhead show the abolished tip. Scale bar: 0.1 cm. d The CFU quantitation of L. plantarum 3297 cells grown at pH 5.5 and 7 after 7 d of incubation on MRS hard agar. The graph shows the means ± SEMs of individual data points, obtained from two independent measurements. *P > 0.05 vs. the control. Here, the colonies grown at pH 5.5 are taken as control compared to those grown at pH 7. e The CFU quantitation of L. plantarum 3297 cells exposed to desiccation stress. The cells were prior grown on MRS hard agar at pH 5.5 or 7 for 7d. After incubation, the conic colonies were directly transferred to the desiccation unit (at 40% relative humidity and 30 °C for 4 d). The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls. f The CFU quantitation of L. plantarum 3297 cells exposed to desiccation stress. The cells were prior grown on MRS hard agar at pH 5.5 or 7 for 7d. After incubation, the conic colonies were directly transferred to the desiccation unit (at 40% relative humidity and 30 °C for 20 h). The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls.

The cone-shaped colonies accentuate a multi-stress response in cells within

Having linked the cone-shaped structure and brown deposits to the acidic-pH adaptation, we next sought to understand its biological significance. We hypothesized that the observed highly structured spatiotemporal colonization would provide a multi-stress response for the bacterial population. To explore this notion, we exposed the colonies to secondary desiccation stress and assessed the survivability of the cells inside the colonies. The desiccation tolerance was higher in colonies grown at acidic-pH (pH 5.5) than at elevated pH (pH 7) (Fig. 1e, f), signifying that the conic-shaped structure imparts some bio-shielding machinery. The possible interpretation of the observations is that brown-coloured deposits and conic-shaped geometry of the colonies (grown at pH 5.5) might help the cells to improve survivability during subsequent stress.

The cold-shock response triggers consolidated bundle formation

Since cold-shock is regarded as one of the substantial stresses for the bacterial cells, we next tested whether the cone-shaped colonies would provide increased survivability during exposure to the cold stress. The cellular cryotolerance drastically fell in the cold-stressed (−17 °C for 1 h) colonies (Fig. 2a), nonetheless the survivability of cells grown at the acidic-pH stressed colonies was comparatively higher (Fig. 2c). In the course of performing the cold stress experiments, several exciting observations were revealed. The most intriguing finding was the sighting of unique consolidated circular bundles in acidic-pH (pH 5.5) stressed colonies following a freeze-thaw challenge (Fig. 2b). Time-lapse imaging revealed that cells harvested from the conic colony rapidly auto-aggregated to form the circular bundles (Fig. 2d, e). Notably, the bundles consisted of live and dead cells, with the former localized at the centre and the latter occupying the periphery. This observation indicates a possible cellular cryoprotective phenomenon, which apparently needs further investigation.

Fig. 2: Influence of cold stress on survivability of L. plantarum grown in conic colonies.
figure 2

a The CFU quantitation of L. plantarum 3297 conic colonies following exposure to cold stress. The cells were grown on MRS hard agar (pH 5.5) for 7d and transferred to cold stress conditions (−17 °C) for 1 h. After incubation, the colonies were resuspended in phosphate-buffered saline (PBS), vortexed, diluted, and plated on fresh MRS agar plates. Similarly, the colonies were also incubated for 1 week at 4 °C and assessed subsequently. The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls. b Dual staining (SYTO™ and propidium iodide (PI)) of consolidated bundles that was induced following frozen stress (−17 °C, 1 h). Scale bar: 20 µm. The cells stained with PI (true-colour: red) are false-coloured with grey. c The CFU quantitation of L. plantarum 3297 conic colonies following exposure to frozen stress. The cells were grown on MRS hard agar pH 5.5 or 7 for 7d and transferred to cold conditions (−17 °C) for 1 h. After incubation, the colonies were resuspended in phosphate-buffered saline (PBS), vortexed, diluted, and plated on fresh MRS agar plates. The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls. d Time-course analysis of cellular aggregation of L. plantarum 3297 cells from conic colonies to form consolidated bundles (within 1 h). Scale bar: 20 µm. e Schematic experimental setup that led to the sighting of consolidated bundles. The cells stained with PI (true-colour: red) are false-coloured with grey.

A heat shock protein 1 (HSP1) is a keystone factor for consolidated bundling during cold stress

Small heat shock proteins (sHSP) have an influential role during heat or cold acclimations19. Probiotic Lactobacilli have typically have either one or two sHSP genes, with the exemption of L. plantarum that harbors three sHSP genes, namely hsp1, hsp2, and hsp3. Lately, the involvement of hsp1 in cryoprotecion was stated19. We, therefore, explored a possible link between sHSP and consolidated bundle formations by L. plantarum. Using hsp1 knockout mutant (Δhsp1), we show that the bundles’ size was relatively reduced (Fig. 3a and Supplementary Fig. S4). Besides, Δhsp1 mutant also displayed reduced survivability during the freeze-thaw challenge (Fig. 3b), and formed poor biofilms on polystyrene surfaces (Fig. 3c). In contrast, Δhsp2 or Δhsp3 mutants formed either moderate or substantial biofilms, respectively, and did not show a significant difference in the freeze-thaw challenge. Hence, the findings up to now suggest the involvement of HSP1 in governing the consolidated bundle formation and cellular protection during the cold stress (Fig. 3d).

Fig. 3: The heat-shock proteins are involved in cold shock stress response.
figure 3

a Consolidated bundles of wild-type L. plantarum WCFS1 or hsp1 mutant (Δhsp1). Scale bar: 20 µm. The cells stained with PI (true-colour: red) are false-coloured with grey. Yellow arrow indicate consolidated bundles and white arrows indicate compromised bundles. b Colony-forming units (CFUs) of wild-type L. plantarum WCFS1 or hsp1 mutant following exposure to frozen stress. The cells were grown on MRS hard agar pH 5.5 for 7d and transferred to cold stress conditions (−17 °C) for 1 h. After incubation, the colonies were resuspended in phosphate-buffered saline (PBS), vortexed, diluted, and plated on fresh MRS agar plates. The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls. c Crystal violet quantification of biofilms formed by L. plantarum WCFS1, and hsp mutants (Δhsp1, Δhsp2, or Δhsp3) on polystyrene surface. The cells were incubated at 37 °C for 24 h without shaking, following which the cells were stained with 0.4% crystal violet and read at OD595. The graph shows the means ± SEMs 6 individual data points, obtained from two independent measurements. ***P < 0.001 vs. the non-treated controls. Here cells grown at pH 5.5 are taken as a control. d Mechanistic flowchart depicting the involvement of heat-shock protein response in governing the multifaceted traits manifested by L. plantarum in response to acidic-pH.

Formation of the consolidated circular bundles is triggered by signalling molecules derived from the cold-shock colony filtrate (CSCF)

We further hypothesized that the auto-aggregation of the cold-stressed colony cells into the consolidated circular bundles could be triggered by some unidentified, but self-produced signalling molecules. We tested whether the cold-shock colony filtrate (CSCF) could trigger circular bundle formation in unstressed colony cells to address this concept. We noticed a formation of small-to-medium-sized circular bundles within 90 or 120 min (Fig. 4a), confirming our hypothesis. The phenomenon was prevented in cells pre-incubated with lithium chloride (5 M) (Fig. 4a), which removes the S-layer proteins of the bacteria, and renders the cells inefficient to receive any external signals20. The circular bundles were absent also when the colony filtrates (CF) from unstressed colonies (Fig. 4a) or in CSCF exposed to heat treatment (60 °C for 30 min) (Supplementary Fig. S5). Overall, these results reveal that the cold stress potentiates L. plantarum cells to release pre-accumulated signals that coordinate cellular aggregation.

Fig. 4: Effect of CSCF from L. plantarum conic colonies on multicellular behaviour by probiotic Bacilli.
figure 4

a Effect of cold-shock colony filtrates (CSCF) on unstressed L. plantarum 3297 cells from conic colonies. Lithium chloride (5 M) is chemical that prevent cellular aggregation. Scale bar: 20 µm. White arrows indicate small bundles. b Effect of CSCF (10% v/v) on the growth of B. subtilis NCIB 3610 incubated at 37 °C for 24 h at 150 rpm. The graph shows the means ± SEMs of three measurements. *P > 0.05 vs. the control. c Effect of CSCF (10 % v/v) on pellicle formation of B. subtilis NCIB 3610 incubated at 30 °C for 48 h without shaking. d The β-galactosidase activity assay (measured in Miller units) in B. subtilis (YC121) cells harboring PtapA-lacZ transcriptional reporter. The graph shows the means ± SEMs of three measurements. **P < 0.01, ***P < 0.001 vs. the non-treated controls. e Effect of CSCF (10% v/v) on bundle formation in B. subtilis (YC189), harboring a transcriptional reporter (PtapA-cfp) indicating on the matrix gene expressionf. Scale bar: 20 µm.

CSCF activates a heterologous multicellular aggregation in Bacillus subtilis

Lately, B. subtilis was shown to be involved in a symbiotic relationship with L. plantarum21. We hypothesized that the CSCF produced by L. plantarum cells would induce multicellular behaviour in B. subtilis through establishing a cross-talk between these probiotic species. We thus examined supplementation of the CSCF at various doses (1–10% volume per volume (v/v)) on B. subtilis response during growth in Lysogeny broth (LB) medium. A notable induction in pellicle formation by B. subtilis at specific concentrations (5 or 10% v/v) was observed (Fig. 4c), without affecting the bacterial growth (Fig. 4b and Supplementary Fig. S6). The β-galactosidase and fluorescent microscopic assays measuring the tapA (the major operon, required for polymerization of TasA amyloid fibers and their proper anchoring on cell surface toward biofilm formation) expression confirmed the activation of matrix production in response to the CSCF derived from L. plantarum cells (Fig. 4d, e). In addition, CSCF does not possess any suicidal effect on L. plantarum itself, and CF did not induce pellicle formation (Fig. 4c). The most conceivable explanation is that the putative signals are accumulated within the conic colonies and are released following cold stress disruption. The phenomenon could explain the triggering effect of the CSCF toward cellular aggregation (Fig. 4a, e).

Low pH triggers the formation of unique V-shaped geometric cell chains

We further investigated the adaptive morphology of L. plantarum cells during the transition to low pHs. Intriguingly, cells grown at particularly acidic-pH (pH 3.5) displayed unique V-shaped cellular structures (Fig. 5a), though they showed slower growth rates compared to the cells grown at elevated pHs (Fig. 5b). The finding led us to assume that these cells retort to low-pH stress by stalling cell division by locking them in a V-shaped conformation consisting of four undivided cells. Besides, we found a “bottle effect” when the low pH grown cells retained their V-shaped structures while forming robust biofilms on polystyrene and glass surfaces (Fig. 5c, d & Supplementary Fig. S7). Collectively, these results demonstrate a survival mode of growth with increased resistance, which could be linked to biofilm formation.

Fig. 5: Effect of acidic pH on the L. plantarum geometric structuring and subsequent biofilm formation accompanied with antagonistic activity against C. albicans.
figure 5

a Phenotypic appearance of L. plantarum 3297 cells grown at pH 3.5 or 5.5 in liquid MRS. Scale bar: 20 µm. White arrows indicate septation, inset scale bar: 2 µm. b Growth curves of L. plantarum grown at either pH 3.5 or 5.5. The cells were incubated at 37 °C for 24 h. c Crystal violet quantification of L. plantarum biofilms on polystyrene surface. The cells were incubated at 37 °C for 24 h without shaking, following which the cells were stained with 0.4% crystal violet and detected at OD595. The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. ***P < 0.001 vs. the non-treated controls. Here cells grown at pH 5.5 are taken as a control. d Fluorescent microscopic images of L. plantarum biofilms on polystyrene surface stained with SYTO™ 9. Scale bar: 200 µm. e Effect of live probiotic cells (L. plantarum cells grown at either pH 3.5 or 5.5) on mature C. albicans biofilms in in vitro conditions. C. albicans biofilms were grown on polystyrene plates. After 8 h of incubation, the broth containing the planktonic cells were removed, replaced with PBS (controls) or live L. plantarum in PBS, and incubated again for 24 h. Biofilm formation was then assessed microscopically after staining with SYTO™ 9. Scale bar: 200 µm. f Survival rates of nematodes (that was prior fed with L. plantarum cells grown at either pH 3.5 or 5.5 or E. coli OP50) infected with C. albicans. The graph shows the means ± SEMs of 9 individual data points, obtained from three independent measurements. **P < 0.01, and ***P < 0.001 vs. the E. coli OP50 controls, ***P < 0.05 vs. L. plantarum at pH 5.5. g The microscopic image of adult C. elegans exposed to C. albicans and subsequently recused with E. coli OP50 (control) or L. plantarum (pH 3.5 or 5.5) feed. Inset shows C. albicans hyphae that ruptured and killed the nematode in E. coli OP50 fed group. Scale bar: 200 µm. h A scheme depicting the formation of V-shaped cell chains and their pertinence.

The V-shaped L. plantarum cells restrain C. albicans virulence and enhance C. elegans longevity

It has been reported that low pH (3.5–4.5) favours Lactobacilli dominance over some pathogens17. We thus tested if the V-shaped cells could potentially mitigate a medically important yeast pathogen, C. albicans. We tested the effect of V-shaped L. plantarum cells on the C. albicans biofilms in in vitro and C. elegans model. L. plantarum cells (grown in media with adjusted pH 3.5, or 5.5), was added (in potato dextrose broth (PDB)) to an 8 h biofilms of C. albicans and monitored its ability to stall biofilm maturation. We found that the live cells effectually mitigate the biofilms of C. albicans (Fig. 5e). Next, adult C. elegans were fed with L. plantraum (grown in media with either pH 3.5 or 5.5) and then infected with C. albicans. In control (E. coli OP50 fed) groups, C. albicans intensified nematode mortality rates (Fig. 5f) and effectively killed the nematode by piercing the cuticle (Fig. 5g, inset). In contrast, nematodes fed with L. plantarum cells exhibited better survival rates (Fig. 5f and g), particularly the groups fed with V-shaped L. plantarum grown in medium with adjusted pH 3.5 (Fig. 5f). Uninfected nematode survival rates are shown in Supplementary Fig. S8. The best interpretation for the observed phenotypes is that the L. plantarum cells likely colonize C. elegans intestine, which mitigate the C. albicans virulence. It also ties to our previous results that at pH 3.5 L. plantarum forms vigorous biofilms, via exhibiting unique V-shaped conformations (Fig. 5a, Supplementary Fig. S7), and apparently mitigate the colonization of C. albicans within C. elegans, through enhancing C. elegans longevity.

Discussion

The key finding of this study is related to the unusual cone-shaped colonies (Fig. 1a, c) formed by the L. plantarum dairy isolate that appear to be associated with acidic-pH adaptation response. Though a vertical colony growth on hard agars is relatively unusual in Lactobacilli genera, it has been recently investigated in a non-motile and non-EPS producing Escherichia coli using a 3D agent-based model22. The group used computational simulations to show that the mechanical forces, and not the nutrients, induce these aerial structures. Other studies with Pseudomonas fluorescens Pf0-1 have demonstrated that the bacteria use secretions to push and expand vertically23. Lactobacilli is a taxonomically complex genus with some species displaying dynamic adaptability, survivability, and nomadic lifestyles2. Owing to these flexibilities, it is not surprising to witness aerial growth in L. plantarum colonies.

We next questioned the robustness of the cone-shaped colonies by challenging them to multiple stresses like the cold or desiccation stresses. Probiotics often need to endure stressful challenges during processing in food or pharmaceutical industries24. These practices are detrimental as a significant portion of the cells lose viability. Cold stress, for instance, is characterized by combination of multiple stresses that constantly challenge the bacteria, including the formation of ice crystals, osmotic imbalance, and dehydration25. L. plantarum has been known to tolerate cold-stress24. Exposing the conic L. plantarum colonies to extreme cold stress revealed unique consolidated circular bundles (Fig. 2b). Auto-aggregation and co-aggregation of probiotic cells are often regarded as beneficial traits13. The former halts adherence of pathogens to the intestinal mucosa, while the latter facilitates pathogen clearance or attenuation. The formation of consolidated bundles might resemble auto-aggregation, through activating rapid and self-reliant machinery to conserve the surviving bacterial population. That said, the mechanism governing auto-aggregation in probiotic LABs is still unclear and remains largely hypothetical. We suggest that some communication signals are elevated during cold-stress, thus potentiating aggregation of L. plantarum and B. subtilis cells (Fig. 4a, e), which can be viewed as a beneficial trait during cross-talks between the probiotic Bacilli. The finding envisages L. plantarum and B. subtilis as prospective candidates for the combined probiotic therapy.

We further addressed the mechanism of consolidated bundle formations, hypothesizing the involvement of cold-shock protein genes (cspC, cspL, and cspP), DnaK, and GroEL, and the molecular chaperones24. The molecular chaperones or the heat-shock proteins (HSP) are ubiquitous and conserved in all organisms24. In Lactobacilli, a single hsp gene, hsp16, was linked to stress resistance. Interestingly, L. plantarum is the only Lactobacilli that harbours three different hsp genes (hsp1, hsp2, and hsp3) that assist the bacteria in governing diverse biological functions that coordinate adaptation and survival. The involvement of HSP1 in cryoprotection and improved cryotolerance of L. plantarum cells was stated lately26,27. Reduced cyroprotection in a hsp1 knock-out mutant was linked to lower membrane fluidity and reduced biofilm formation19. We here show that the mutant exhibits poor ability to form consolidated bundles (Fig. 3a). Overall, the findings support the involvement of HSP1 in cellular cryoprotection.

The acidic-pH (pH 5.5) stressed conic colonies also exhibited better survivability to desiccation stress when compared to colonies generated at the neutral pH (Fig. 1e, f). The result supports our notion that cells exposed to acidic-pH display enhanced survivability. Subsequently, we were further concerned in studying how the bacteria adjust to extremely low pH environments. In general, L. plantarum cells are well-adapted to low pH settings19,26; hence, the cells undergo rapid physiological reorganization by altering growth or lowering metabolic activities. For instance, Ingham et al. (2008) showed a pH downshift (pH 3) in L. plantarum WCFS1 resulted in the formation of elongated (filamentous) cells lacking septation12. We report distinctive “V-shaped” geometric cell chains in extremely low pH downshifts (pH 3.5) (Fig. 5a). We also detected a decline in the growth in cells displaying the V-shaped geometry (Fig. 5b) accompanied by a mysterious “bottle effect” when these cells exhibited a strong ability to form biofilms on polystyrene and glass surfaces (Fig. 5c, d & Supplementary Fig. S7). A recent study has shown that the exopolysaccharides (EPS) production could be upregulated in L. plantarum cells grown at acidic pH, which supports our observation28. Biofilm formation by probiotic species is a propitious tactic to control pathogenic biofilms, which is relatively based on the competitive-exclusion principle29,30. It was shown that colonizing LAB may notably reduce the biofilms by Listeria monocytogenes, Salmonella typhimurium, and E. coli O157:H731. We experimented the concept with a pathogenic yeast model, C. albicans, and revealed that the “V-shaped” L. plantarum cells influenced C. albicans biofilm formation (Fig. 5e). Lactobacilli were shown to inhibit the early stages of biofilm development in C. albicans32. Likewise, Lactobacillus species are natural competitors of C. albicans in the human vaginal environment and are known to dominate the microbiome when the pH is highly acidic (pH 3.5–4.5)17. The combination of low pH and lactic acid production by LAB may notably control C. albicans infections in vaginal microbiome33. We deduce that the dominance displayed by the LAB in low pH condition has possible connections with the V-shaped cell chaining that trigger biofilm growth.

To explore this concept in an in vivo model, we used a C. elegansC. albicans co-infection experiment34,35,36,37. Infections of C. elegans with C. albicans reduces the life span and longevity of the nematode. On the contrary, feeding C. elegans with LABs enhances its life span37. In our setup, C. elegans fed with V-shaped L. plantarum cells exhibited enhanced longevity and defence against C. albicans (Fig. 5h). L. plantarum also mitigated S. aureus-induced nematode killing (Supplementary Fig. S9), revealing its dominance over bacterial pathogens as well. Overall, the findings shine a light on the dynamic adaptability, dominance, and reign of L. plantarum to compete and thrive in hostile pH conditions34,35,36,37,38.

Conclusion

The probiotic L. plantarum strains have been vastly known as preserving microbial balance in the human gut through mitigating harmful pathogens directly or by triggering the host immune response reactions. However, they lead an unpredictable lifestyle in their ecological niche and exhibit remarkable potentials to undergo a cellular transformation in response to environmental stresses. The current study provided evidence of how L. plantarum expresses numerous cellular phenotypes under acidic-pH conditions. The most intriguing finding appears to be its ability of complex geometrical structuring: cone-shaped colonies, consolidated bundles, and V-shaped chaining. We deduced it as a bio-shielding measure taken by the multicellular bacterial population controlled by a heat-shock response system. Consequently, the unique geometrical arrangement of cells within consolidated bundles could protect the bacterial population from different environmental insults. In addition, the microbial stress recovery concept also applies to the V-shaped structures (Supplementary Fig. S10) because, during stress recovery, a V-shaped cell chain might instantly divide into four cells favouring rapid outgrowth. In terms of growth at acidic pH, L. plantarum adopts active cellular conformations and effectually succeeds in prospering. In total, we present the multifaceted traits manifested by probiotic L. plantarum, which has significant health implications in the food, pharmaceutical, or medical industries.

Materials and methods

Microbial strains and culture media

Specifics of microbial strains used in the study are described in Supplementary Table S1. Microbial strains were cultured and maintained in their respective selective media. For instance, L. plantarum was cultured, maintained and experimented in De Man, Rogosa and Sharpe (MRS) (HI media Pvt. Ltd., India) hard agar and/or liquid medium incubated at 37 °C, non-shaking conditions, B. subtilis in LB (BD Difco, US) (37 °C, 150 rpm for 5 h)39, and C. albicans in potato dextrose broth (PDB) (BD Difco, US), (37 °C, 150 rpm, overnight) or PDB supplemented with agar, and/or Roswell Park Memorial Institute medium-1640 (RPMI) (Gibco, US)40. The colonies were generated by setting the overnight culture to OD600 = 1 and achieving 10−7 dilutions that were subsequently spread on MRS hard agar plates and incubated for either 5, 7, or 15 days at 37 °C. The colony elevation heights and diameters were measured using a standard ruler and a vernier calliper, respectively.

Growth curve analysis

The L. plantarum or B. subtilis cells were grown overnight in MRS or LB, respectively, using incubation conditions described above in “Microbial strains and culture media”. The cultures were diluted 1:100 into fresh MRS (pH 3.5 or 5.5) or LB (with or without CSCF) and incubated for 24 h at 37 °C with shaking at 150 rpm (for B. subtilis) and non-shaking (for L. plantarum). Every 2 h, 1 mL of each sample was collected, and the optical density (OD600) was measured using the Biowave CO8000 cell density meter.

Crystal violet biofilm quantification assay

The crystal violet staining method was used for the biofilm quantification generated in 48-well microtiter plates (Tarsons Products Pvt. Ltd., India)40. Briefly, microbes were inoculated into respective medium and incubated overnight at 37 °C without shaking. Cultures were diluted to OD600 = 0.01 in fresh MRS, re-inoculated into fresh medium, seeded on the microtiter plates, and incubated 37 °C without shaking for 48 h and 72 h. Following incubation, the growth of the cells was measured at optical density (OD) 600 mm using a microtiter plate reader. Then, the cells attached to the surface were stained with 0.1% crystal violet for 20 min, repeatedly washed with sterile distilled water, and suspended in 95% ethanol. Plates were read at 595 nm and OD values were recorded. Graphs were represented as means ± SEMs of four different trials.

Freeze-thaw challenge of L. plantarum conic colonies

An experimental flowchart for L. plantarum freeze-thaw challenge assay is depicted in Fig. 2e. Briefly, overnight culture suspensions of L. plantarum were diluted to OD600 = 0.01 in fresh MRS broth, spread on MRS hard agar, and incubated for 7 or 15 days at 37 °C. After incubation, the plates were sealed and transferred to −17 °C for 1 h, following which the conic colonies were scrapped off the plate with sterile distilled water (DW). The cells in DW were vortexed vigorously for 5 min and sonicated for 2 min (10 s pulse on/off) at 4 °C with 40% amplitude to break any colony clumps. The samples were visualized under a microscope to make sure there were no clumps in the solution. Then, the cells were incubated at 37 °C and periodically scrutinized under the microscope every 15 min till 3 h for cellular aggregation. The cells were stained with filmtracer™ LIVE/DEAD™ biofilm viability kit (Thermofishers Scientific, US) and visualized under Nikon fluorescent microscope (Nikon Eclipse Ti2, Japan). The quantification of the images (measurement of consolidated bundles intensity) were performed using ImageJ V 1.8.0.

Cold-shock colony filtrate (CSCF) preparation and assay with unstressed conic colonies of L. plantarum

An assay was developed to see if the cold-stressed conic colony (−17 °C for 1 h) extracts could potentiate consolidated bundling in unstressed conic colony cells. Briefly, 1 mL of suspension from harvested cold-stressed (treated) and unstressed (control) conic colonies (~5–6 colonies) were resuspended in DW, vortexed, and sonicated. The extracts were filtered using a 0.2 µm membrane filter to remove the cells. The filtrate was named cold-shock colony filtrates (CSCF). Simultaneously, 1 mL cells from the unstressed colonies were vortexed, sonicated, and pelleted. To these pelleted unstressed cells, 1 mL of CSCF was added, aspirated, and incubated at 37 °C for 3 h. At the same time, colony filtrates (CF) from the unstressed colony and heat-treated (60 °C for 30 min) CSCF were prepared and tested. Following incubation, the aggregation pattern was scrutinized under Nikon fluorescent microscope (Nikon Eclipse Ti2, Japan). As a negative control, one vial of unstressed cells was incubated with 5 M lithium chloride (LiCl2) for 30 min before adding 1 mL of CSCF. LiCl2 abolishes the S-layer protein and stalls the ability of the cells to aggregate or form bundles.

pH tolerance assay

L. plantarum tolerance to pH was studied by growing the bacteria in the pH gradient ranging from pH 3.5 to 7. Briefly, MRS liquid media and hard agars were prepared, and the pH was adjusted either with 0.1 M NaOH (for alkaline pH) or 1 M HCl (for acidic pH). MRS liquid media with different pH settings were used for assessing the growth profiles on 50-mL Tarson tubes (Tarsons Products Pvt. Ltd., India). The biofilm formations on polystyrene plates as described in “Growth curve analysis”, while the hard agars were used to scrutinize the formation of conic colonies.

Desiccation stress experiment

The desiccation experiments were performed using the 7-day-old colonies grown on MRS hard agar21. Initially, the L. plantarum cells were inoculated into liquid MRS and incubated overnight at 37 °C for 24 h. The cultures were then diluted to OD600 = 0.01 in fresh MRS, re-inoculated into a fresh medium, spread on MRS hard agar plates, and incubated for 7 days at 37 °C. The plates were kept open in a desiccation cabinet (MRC, Holon, Israel) at 40% relative humidity and 25 °C for 24 h. Following drying, the whole colony was lifted, suspended in sterile PBS or distilled water, aspirated, sonicated to disrupt colony clumps. Then, the cells were serially diluted and plated on MRS hard agars and incubated at 37 °C for 48 h, following which the colony-forming units (CFU) were counted and recorded.

Biofilm formation and the matrix expression assays with Bacillus subtilis

B. subtilis was used as a co-culture probiotic model to examine the biofilm stimulatory effect of cold-shock colony supernatant (CSCF) extracted from L. plantarum. For pellicle formation assays, 5 μl of the bacterial suspensions (5 × 105 CFU/mL) were pipetted into 4 mL of LB in 24-well polystyrene plates, following which plates were incubated at 30 °C for 72 h. Images of the pellicles were captured using a smartphone fitted with Leica Vario-Summilux-H1.6–3.4/16–125 ASPH cameras. B. subtilis strain YC189 (PtapA-cfp harboring strain) was used to assess the biofilm bundles in liquid LB supplemented with or without CSCF (5% v/v and 10% v/v) using CFP filter in a Nikon fluorescent microscope (Nikon Eclipse Ti2, Japan); while the YC121 (PtapA-lacZ expression) strain was used for assessing the β-galactosidase activity41 of the cells collected and resuspended in phosphate-buffered saline (PBS). Typical long bundled chains of cells in the biofilm were disrupted using mild sonication.

C. albicans biofilm inhibition assay with live L. plantarum

Effect of live probiotic cells (L. plantraum grown at pH 3.5 or 5.5) on C. albicans biofilms were tested in C. albicans biofilm inducing conditions42. Briefly, C. albicans biofilms were grown on polystyrene plates for 8 h at 37 °C. After 8 h of incubation, the supernatant was removed, replaced with PBS (controls) or live L. plantarum cells in PBS (grown previously in MRS at pH 3.5, or 5.5), and incubated again at 37 °C for 24 h. Biofilm inhibition was then assessed microscopically after staining with SYTO® 9 dye and crystal violet quantification assay.

C. elegans co-infection assays

Wild-type C. elegans were maintained on nematode growth medium (NGM) with E. coli OP50 as the feed, and synchronization was performed using recently established method34. Briefly, C. elegans were collected by aspiration and bleached with 2% sodium hypochlorite and 0.5 N sodium hydroxide to get the eggs. Eggs were transferred to 48-well microliter plates and were incubated for 24 h at 22 °C for hatching. The hatched juveniles were transferred to fresh E. coli OP50 plates and incubated for 5–7 days to obtain adult nematode. Adults were subsequently used for toxicity, and bacterial or C. albicans colonization assays. For C. albicans infection of C. elegans, the adult nematodes, previously fed on E. coli OP50, were transferred to a C. albicans lawn on NGM agar plate for 4 h. After 4 h, the nematodes were collected in M9 buffer, pipetted into a 96-well, and survival monitored for 7 days. The treatment groups were adult nematodes previously fed with the probiotic L. plantarum lawns prepared from cultures grown at either pH 5.5 or pH 3.5. The live and dead nematodes were counted under bright-field, and DAPi filter and the nematode survival rates were estimated and plotted. The images of C. elegans were acquired using a Nikon fluorescent microscope (Nikon Eclipse Ti2, Japan). The sample size were >50 adult nematodes per NGM or 96-well microtitre plate, on average.

Statistical analysis

All experiments were done in triplicates, and results are expressed as means ± standard deviations with either six or nine individual data points from three biological repeats. The student’s t-test was used to determine the significance of differences between treated and non-treated samples. Statistical significance was accepted for p values <0.05, and significant changes are indicated using asterisks in figures (*p < 0.05, **p < 0.01, and ***p < 0.001).

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.