Differential Modulation of Listeria monocytogenes Fitness, In Vitro Virulence, and Transcription of Virulence-Associated Genes in Response to the Presence of Different Microorganisms

Listeria monocytogenes is the etiological agent of the severe foodborne disease listeriosis. Important insight regarding the physiology and the infection biology of this microorganism has been acquired in the past 20 years. However, despite the fact that L. monocytogenes coexists with various microorganisms throughout its life cycle and during transmission from the environment to foods and then to the host, there is still limited knowledge related to the impact of surrounding microorganisms on L. monocytogenes' biological functions. In this study, we showed that L. monocytogenes modulates specific biological activities (i.e., growth and virulence potential) as a response to coexisting microorganisms and differentially alters the expression of virulence-associated genes when confronted with different bacterial genera and species. Our work suggests that the interaction with different bacteria plays a key role in the survival strategies of L. monocytogenes and supports the need to incorporate biotic factors into the research conducted to identify mechanisms deployed by this organism for establishment in different environments.

ABSTRACT

Interactions between Listeria monocytogenes and food-associated or environmental bacteria are critical not only for the growth but also for a number of key biological processes of the microorganism. In this regard, limited information exists on the impact of other microorganisms on the virulence of L. monocytogenes. In this study, the growth of L. monocytogenes was evaluated in a single culture or in coculture with L. innocua, Bacillus subtilis, Lactobacillus plantarum, or Pseudomonas aeruginosa in tryptic soy broth (10°C/10 days and 37°C/24 h). Transcriptional levels of 9 key virulence genes (inlA, inlB, inlC, inlJ, sigB, prfA, hly, plcA, and plcB) and invasion efficiency and intracellular growth in Caco-2 cells were determined for L. monocytogenes following growth in mono- or coculture for 3 days at 10°C or 9 h at 37°C. The growth of L. monocytogenes was negatively affected by the presence of L. innocua and B. subtilis, while the effect of cell-to-cell contact on L. monocytogenes growth was dependent on the competing microorganism. Cocultivation affected the in vitro virulence properties of L. monocytogenes in a microorganism-specific manner, with L. innocua mainly enhancing and B. subtilis reducing the invasion of the pathogen in Caco-2 cells. Assessment of the mRNA levels of L. monocytogenes virulence genes in the presence of the four tested bacteria revealed a complex pattern in which the observed up- or downregulation was only partially correlated with growth or in vitro virulence and mainly suggested that L. monocytogenes may display a microorganism-specific transcriptional response.

IMPORTANCE Listeria monocytogenes is the etiological agent of the severe foodborne disease listeriosis. Important insight regarding the physiology and the infection biology of this microorganism has been acquired in the past 20 years. However, despite the fact that L. monocytogenes coexists with various microorganisms throughout its life cycle and during transmission from the environment to foods and then to the host, there is still limited knowledge related to the impact of surrounding microorganisms on L. monocytogenes' biological functions. In this study, we showed that L. monocytogenes modulates specific biological activities (i.e., growth and virulence potential) as a response to coexisting microorganisms and differentially alters the expression of virulence-associated genes when confronted with different bacterial genera and species. Our work suggests that the interaction with different bacteria plays a key role in the survival strategies of L. monocytogenes and supports the need to incorporate biotic factors into the research conducted to identify mechanisms deployed by this organism for establishment in different environments.

INTRODUCTION

Microbial communities in foods constitute pools of biological diversity where various taxonomic units may coexist in variable microniches of food matrices, such as structure, nutrients, aeration, etc. Microbial interactions may define the dominance of certain spoilage groups or the persistence of species that can cause foodborne illness (1–8). As a result, the study of these interactions is crucial for the development of accurate and realistic predictions of microbial growth and thus the design of control strategies to ensure food quality and safety. Listeria monocytogenes is a foodborne pathogen responsible for the severe invasive disease termed listeriosis, which primarily occurs in immunocompromised individuals, but it can also cause gastroenteritis in healthy persons (9). The genome of L. monocytogenes is typical of ubiquitous bacteria (i.e., high abundance of genes encoding transport proteins and regulators) (10, 11). Thus, a number of studies have reported on the remarkable ability of L. monocytogenes to adapt, survive, and/or grow in diverse niches and often under extreme environmental conditions (12). It is apparent that, depending on environmental factors, L. monocytogenes may regulate its responses (e.g., growth, biofilm formation, ability to invade a host) accordingly (13–16) to optimize its survival potential. The same strategy may be applicable to the effort of L. monocytogenes to become established in habitats where different bacteria may induce different responses to this microorganism. Previous studies examining the transcriptional profile of Pseudomonas fluorescens have suggested that a bacterium may alter the expression of genes related to various processes (e.g., antibiotic production) in the presence of phylogenetically different microorganisms (17, 18).

With reference to the above, the behavior of L. monocytogenes when confronted with other bacterial species has been studied mostly in relation to its growth capacity (8). Other characteristics, such as its virulence potential and the underpinning mechanisms which drive the responses of L. monocytogenes in the presence of other bacteria, have not been extensively investigated. To date, there is evidence for the differential regulation of metabolic and antibiotic resistance-related genes of L. monocytogenes in mixed cultures with Bacillus subtilis (19, 20). Likewise, it has been shown that L. monocytogenes reshapes its entire transcriptome in the presence of two Lactobacillus species (21), while in the presence of Bifidobacterium longum a number of virulence-related L. monocytogenes genes may be downregulated (22). At the same time, a number of studies, though limited, indicate changes in the pathogenicity of L. monocytogenes when surrounded by specific groups of microorganisms colonizing the intestinal environment (23–25).

Nevertheless, information on the contribution of different microorganisms in the modulation of L. monocytogenes' biological functions is scarce. To better understand the dynamics of L. monocytogenes evolution, transmission, and pathogenesis, we need to gain deeper insight into the interactions between this microorganism and other bacterial species, both at the phenotypic level and at the molecular level. Our hypothesis was that growth and virulence of L. monocytogenes are differently regulated in response to the presence of specific microorganisms. Therefore, the objective of this study was to assess the growth, in vitro virulence, and transcriptional profile of L. monocytogenes virulence-associated genes following cocultivation with four different bacterial genera or species.

RESULTS

Growth potential of bacterial strains in monocultures.

The growth potential of L. monocytogenes (ScottA), L. innocua (7510), B. subtilis (LMBF B109), L. plantarum, and P. aeruginosa (LMBF B26) was assessed in monocultures. When grown in TSB-Y, L. monocytogenes and L. innocua attained approximately 10⁹ CFU/ml at both 10°C and 37°C (Fig. 1). Incubation at 37°C for 24 h resulted in an 8 to 8.5 log CFU/ml final population for P. aeruginosa, while B. subtilis and L. plantarum reached approximately 7.5 to 8 log CFU/ml (Fig. 1A). B. subtilis did not grow at 10°C, while P. aeruginosa and L. plantarum reached ca. 6.5 log CFU/ml within 10 days (Fig. 1B).

FIG 1.

Growth kinetics of L. monocytogenes, L. innocua, B. subtilis, P. aeruginosa, and L. plantarum in TSB-Y at 37°C for 24 h (A) and at 10°C for 10 days (B). Data represented as log (CFU/ml) are mean values ± standard deviations (SD) of three biological replicates performed in duplicate.

Growth potential of L. monocytogenes in the presence of competing microorganisms.

We compared the growth of L. monocytogenes as a monoculture with its growth in coculture with L. innocua, B. subtilis, L. plantarum, or P. aeruginosa. When L. monocytogenes was incubated with B. subtilis at 37°C, growth was significantly (P < 0.05) suppressed after 9 h of incubation compared to the corresponding growth in monoculture (Fig. 2A). More specifically, similarly to the monoculture, L. monocytogenes attained almost 6 log CFU/ml within 9 h of incubation, followed by population decline. After 24 h of incubation, the levels of L. monocytogenes decreased to the initial inoculation levels (3 log CFU/ml). Contrary to the case at 37°C, coincubation of L. monocytogenes with B. subtilis at 10°C did not influence the growth of L. monocytogenes (Fig. 2B).

FIG 2.

Growth kinetics of L. monocytogenes in the presence of L. innocua, B. subtilis, P. aeruginosa, or L. plantarum in TSB-Y at 37°C for 24 h (A) and at 10°C for 10 days (B). Data represented as log (CFU/ml) are mean values ± SD of three biological replicates performed in duplicate.

Cocultivation with L. innocua affected the growth of L. monocytogenes at 37°C (Fig. 2A) and at 10°C (Fig. 2B), resulting in approximately 1 log CFU/ml lower maximum population density compared to the monoculture. The presence of P. aeruginosa and L. plantarum did not influence the growth potential of L. monocytogenes regardless of the incubation temperature (Fig. 2A and B).

Considering the above, among the tested microorganisms we identified L. innocua and B. subtilis as capable of inhibiting the growth of the L. monocytogenes population and possibly even reducing the population levels.

Cell contact.

L. monocytogenes was cocultivated in direct cell contact or in the absence of cell contact with L. innocua and B. subtilis to assess whether the inhibitory effect of the latter microorganisms on the growth of L. monocytogenes is cell contact dependent. It was observed that the population density of L. monocytogenes was 1 log CFU/ml lower than the monoculture at 21, 22, and 24 h either in direct contact or without cell contact with L. innocua (Fig. 3A). In contrast, the restriction of L. monocytogenes growth in the presence of B. subtilis significantly abated when the two microorganisms were not allowed to intermix (Fig. 3B). In the absence of cell contact, B. subtilis could only reduce the final population of L. monocytogenes by 1 log CFU/ml. The levels of growth of both L. innocua and B. subtilis in mixed cultures with L. monocytogenes (with or without cell contact) were not different from the growth levels in their monocultures (data not shown). These results showed that growth inhibition of L. monocytogenes is contact mediated for B. subtilis but not for L. innocua.

FIG 3.

Growth kinetics of L. monocytogenes in monoculture (single) or in coculture with (contact) or without (no contact [separated by a 0.4-μm membrane]) physical cell contact with L. innocua (A) or B. subtilis (B) in TSB-Y at 37°C for 24 h. Data represented as log (CFU/ml) are mean values ± SD of three biological replicates performed in duplicate.

Coaggregation of L. monocytogenes with B. subtilis.

The coaggregation of L. monocytogenes with B. subtilis was investigated as a potential factor affecting the observed substantial growth inhibition of L. monocytogenes by B. subtilis. Therefore, coaggregation was assessed at 37°C within 24 h. Coaggregation occurred up to 18% (i.e., 18% of L. monocytogenes cells coaggregated with B. subtilis) (Fig. 4). Initially, at 6 h the coaggregating ability of L. monocytogenes with B. subtilis was approximately 8%. This ability increased with incubation time, reaching the highest percentage of 18% after 24 h.

FIG 4.

Coaggregation of L. monocytogenes with B. subtilis. Data represented as % coaggregation are mean values ± SD of two biological replicates performed in duplicate.

In vitro virulence of L. monocytogenes in the presence of competing microorganisms.

The potential of the four competing organisms (i.e., L. innocua, B. subtilis, L. plantarum, and P. aeruginosa) to affect the virulence of L. monocytogenes was also investigated. An in vitro Caco-2 cell assay was employed to determine L. monocytogenes CFU that could invade and proliferate intracellularly in the presence of the aforementioned microorganisms. Under the conditions tested in this experiment, none of the competing microorganisms could adhere to or invade epithelial cells.

The presence of L. innocua during infection of Caco-2 cells significantly enhanced the invasion efficiency of L. monocytogenes (black bar) regardless of the previous growth temperature, i.e., 37°C (Fig. 5A, light gray bars) or 10°C (Fig. 5C, light gray bars). Interestingly, the invasiveness of L. monocytogenes also increased if cocultivation of the two microorganisms preceded the infection of epithelial cells (Fig. 5A and C, dark gray bars). On the other hand, coincubation with B. subtilis at 37°C significantly attenuated the potential of L. monocytogenes to invade Caco-2 cells (Fig. 5A, dark gray bars). The same result was also observed when B. subtilis and L. monocytogenes were grown independently at 37°C and then combined to infect the cells (Fig. 5A, light gray bars). The presence of L. plantarum and P. aeruginosa during growth and/or during infection of Caco-2 cells did not change the invasion efficiency of L. monocytogenes (Fig. 5A and C).

FIG 5.

Invasion efficiency (%) (A and C) and intracellular growth (IGC) (B and D) of L. monocytogenes grown in TSB-Y at 37°C for 9 h (A and B) or at 10°C for 3 days (C and D) singly (black bars) cocultivated with L. innocua, B. subtilis, P. aeruginosa, and L. plantarum (dark gray bars) or grown alone and combined with the latter microorganisms prior to the virulence assay (IG, light gray bars). Data represent mean values ± standard error of the mean (SEM) of three biological replicates performed in triplicate. *, significant differences compared to single culture (P < 0.05).

All microorganisms reduced the intracellular growth of L. monocytogenes in Caco-2 cells at 10°C (Fig. 5D); this was more pronounced if these microorganisms had been previously cocultivated with L. monocytogenes before the virulence assay. At 37°C (Fig. 5B), the same effect was evident only for the cocultures of L. monocytogenes with P. aeruginosa and B. subtilis. Especially at 10°C, the presence of P. aeruginosa and B. subtilis during infection of Caco-2 cells decreased the intracellular growth coefficient (IGC) of L. monocytogenes compared to the monoculture, even if previous cocultivation with the two organisms had not taken place (Fig. 5D, light gray bars). At 37°C, the intracellular proliferation of L. monocytogenes was significantly decreased if P. aeruginosa was present during infection of Caco-2 cells, with or without previous cocultivation of the two microorganisms (Fig. 5B). The highest reduction (35-fold) in IGC of L. monocytogenes in Caco-2 compared to monoculture (black bar) was observed at 37°C when L. monocytogenes was cultivated with B. subtilis before the infection of epithelial cells (Fig. 5B, dark gray bars). As for the impact of L. plantarum, the IGC of L. monocytogenes at 37°C was lower than for the single culture only when L. plantarum was added in the assay just before the infection of Caco-2 cells (Fig. 5B, light gray bars).

Taken together, our data show that individual microorganisms have different effects on the virulence potential of L. monocytogenes. These results indicate a strong positive influence of L. innocua and a significant negative influence of B. subtilis on the invasion of L. monocytogenes, suggesting that invasion attenuation does not necessarily correlate with growth inhibition. Overall, a competing microorganism may distinctly and differently affect growth, invasion, and intracellular growth of L. monocytogenes.

Transcription of L. monocytogenes virulence genes in the presence of competing microorganisms.

The ability of the four different tested microorganisms to influence the in vitro virulence properties of L. monocytogenes led us to investigate the transcriptional profiles of certain L. monocytogenes key virulence genes that are associated with invasion and intracellular proliferation into host cells. The expression levels of inlA, inlB, inlC, inlJ, sigB, prfA, hly, plcA, and plcB of L. monocytogenes were determined following growth in monoculture for 9 h or 3 days at 37°C or 10°C, respectively. This expression profile was compared with the expression of the same genes when L. monocytogenes was grown in coculture with L. innocua, B. subtilis, L. plantarum, or P. aeruginosa. A complex pattern was revealed, with no quite clear or evident consistent trends regarding the combined role of competing organisms and growth conditions in the expression of the virulence genes. The mRNA levels of inlB, inlJ, sigB, and plcA were higher in the presence of L. innocua compared to the monoculture in TSB-Y at 37°C (Fig. 6A); notably, the transcription of prfA was 7-fold (log₂) higher in comparison to the monoculture. On the contrary, inlA and inlB were downregulated following incubation with B. subtilis at 37°C (Fig. 6C). When L. monocytogenes was cocultured with P. aeruginosa at 37°C, expression levels of inlJ were significantly higher (at least 2-fold) compared to the monoculture (Fig. 6D). In the presence of L. plantarum, expression of the genes under study was not affected (Fig. 6B). At 10°C, prfA was upregulated in the presence of B. subtilis (Fig. 7C) but downregulated in the presence of L. innocua (Fig. 7A) or L. plantarum (Fig. 7B). Cocultivation with P. aeruginosa caused a significant increase in the transcription of four virulence genes compared to the monoculture, namely, inlB, prfA, plcA, and plcB (Fig. 7D). Our data suggest that L. monocytogenes may respond differently when interacting with different microorganisms and amend its virulence-related responses at the transcriptional level depending on the competitor. In addition, the effect of the same microorganism might change depending on the growth temperature during cocultivation.

FIG 6.

Relative changes in mRNA levels of L. monocytogenes virulence-related genes in the presence of L. innocua (A), L. plantarum (B), B. subtilis (C), and P. aeruginosa (D) in TSB-Y at 37°C for 9 h. Values were normalized to rpoB expression levels and are given as log₂ fold change compared to the values of single culture. Data represent mean values ± SEM of two biological replicates performed in duplicate. The asterisk indicates that the values are significantly (P < 0.05) above 2 or below −2.

FIG 7.

Relative changes in mRNA levels of L. monocytogenes virulence-related genes in the presence of L. innocua (A), L. plantarum (B), B. subtilis (C), and P. aeruginosa (D) in TSB-Y at 10°C for 3 days. Values were normalized to rpoB expression levels and are given as log₂ fold change compared to the values of single culture. Data represent mean values ± SEM of two biological replicates performed in duplicate. The asterisk indicates that the values are significantly (P < 0.05) above 2 or below −2.

DISCUSSION

The existing knowledge about the relationships of L. monocytogenes with other microorganisms and how different bacteria may influence the evolution and responses of the pathogen to environmental or host-produced stimuli is relatively limited. This knowledge is relevant to food safety since certain microorganisms could inhibit or promote the growth of L. monocytogenes on foods or enrichment media, thereby possibly introducing bias in detection. At the same time, these microorganisms may trigger virulence or generally have a divergent impact on the behavior of the pathogen within the host and affect the severity of a listerial infection. Our observations cannot be extrapolated to real food systems, where additional characteristics such as food structure may interfere with bacterial interactions (26, 27). In addition, during passage through the gastrointestinal tract, L. monocytogenes would encounter an environment drastically different from laboratory media, and interactions with the gut microbiome would likely overwhelm any observed interactions between L. monocytogenes and the tested organisms. However, such data may set the ground for the design of more realistic studies.

In this study, we used four ubiquitous bacteria of different genera and phylogenetic proximity to L. monocytogenes that are found in habitats where L. monocytogenes may also be found. During time as a saprophyte, L. monocytogenes encounters such other bacteria in the environment and on foods which could be ingested and accompany the microorganism throughout the route to the host. Often, these bacteria may colonize the intestinal tract and/or become members of the transient gut microbiota. Both B. subtilis and L. plantarum are considered important inhabitants of the human intestinal microbiota (28), while P. aeruginosa may also colonize the intestinal epithelium of susceptible individuals (29, 30).

We showed that L. innocua and B. subtilis have the potential to restrain the growth of L. monocytogenes, in contrast to P. aeruginosa and L. plantarum, which did not seem to have an effect on the fitness of L. monocytogenes under the conditions tested. In contrast to our results, previous studies have stressed the inhibiting effect of lactic acid bacteria on the growth of L. monocytogenes and suggested that they would make suitable candidates for biocontrol of the pathogen (31). Interestingly, Pseudomonas species may act either as competitors or as promoters of L. monocytogenes growth depending on the surrounding environment (8). Growth competition between L. monocytogenes and L. innocua has long been a subject of research, mainly due to the reported underdetection of L. monocytogenes in the presence of L. innocua (32). This has previously been attributed to the higher growth rates or the more efficient nutrient uptake by L. innocua compared to L. monocytogenes (33, 34), which does not seem to agree with our results since both microorganisms have equal growth potential as monocultures. L. innocua may also outcompete L. monocytogenes in coculture through the production of toxic metabolites (35, 36). Since cell contact was not involved in growth inhibition of L. monocytogenes by L. innocua, we could speculate that the decrease of the L. monocytogenes population observed in the current study might be ascribed to the release of allelochemical compounds.

Regarding the effect of B. subtilis on the growth of L. monocytogenes, our results suggest that possible secretion of toxic compounds, as well as cell contact, may have contributed to the lethal effect of B. subtilis on L. monocytogenes. Previous studies have attributed the antibacterial properties of B. subtilis to the production of bioactive lipopeptides such as surfactin (32). This compound has been shown to reduce adhesion and biofilm formation by L. monocytogenes on abiotic surfaces (37, 38) and to exert a significant antimicrobial effect against L. monocytogenes and other Gram-positive microorganisms (39, 40). On the other hand, as mentioned above, physical interactions between B. subtilis and L. monocytogenes that relate to a cell-contact-mediated inhibitory effect of B. subtilis may have taken place. Proteins of the YD-repeat family have been shown to be involved in interstrain growth inhibition of B. subtilis (41), while Holberger et al. identified the role of B. subtilis LXG toxins (proteins widespread among Firmicutes) as RNases potentially involved in interbacterial competition (42). In addition to the above, our results suggest that cell aggregation of L. monocytogenes in the presence of B. subtilis might also have partially resulted in contact-mediated inhibition. Available evidence from previous studies has illustrated that B. subtilis is highly coaggregative with L. monocytogenes (43). This difference may be related not only to the microorganism species but also to the particular strains that are tested for coaggregation. Coaggregation is thought to play a significant role in gut colonization by pathogens, as noncoaggregating microorganisms may be more easily removed from the intestinal tract (44, 45). The above suggests that underestimation of the actual numbers of L. monocytogenes bacteria could potentially occur due to cell aggregation in the presence of B. subtilis.

Over the course of investigating the effect of cocultivation on the virulence properties of L. monocytogenes, we found that L. innocua boosted the in vitro invasion efficiency of L. monocytogenes as opposed to its negative effect on the growth of the pathogen. Interestingly, the presence of L. innocua during infection of Caco-2 cells could still increase the invasion efficiency of L. monocytogenes, though to a much lesser extent. This suggests an important role of exposure to L. innocua prior to the infection of host cells in the invasiveness of L. monocytogenes. L. innocua is a common cohabitant of L. monocytogenes, often isolated from the same niches (46). Due to the potential of L. innocua to impair L. monocytogenes growth and hinder its isolation from foods during selective enrichment, a possible effect of this microorganism on the ability of L. monocytogenes to invade a host may dictate the need for a more critical evaluation of L. monocytogenes-negative but L. innocua-positive samples.

We also observed that the attenuation of L. monocytogenes invasion due to cocultivation with B. subtilis was only manifested after incubation at a temperature permissive for the growth of B. subtilis (37°C). It has been reported that the peptidoglycan fragments and muropeptides released by actively growing Gram-positive cells may serve as signals whereby bacteria may monitor their surroundings for the presence of microorganisms and respond accordingly. For instance, they may alter the expression of their curli and fimbriae or induce the production of antimicrobial compounds (47–49). The above findings underline the importance of cocultivation conditions (i.e., favorable [or not] also for the competing bacterium) to the subsequent effect of a microorganism on L. monocytogenes invasiveness, e.g., when coexisting in a food matrix or the host environment.

Our data on the intracellular growth of L. monocytogenes indicated that, during infection of Caco-2 cells, the preceding cocultivation with the four competing bacteria did not promote but rather decreased the potential of the pathogen to proliferate intracellularly. In fact, P. aeruginosa and L. plantarum, though unable to affect the growth or the invasion of L. monocytogenes, could reduce the intracellular growth of the microorganism. Similar to environmental factors, which do not always distinctly affect every biological activity of L. monocytogenes, it is likely that a neighboring microorganism may only influence specific responses of the pathogen. Previously, Moroni et al. (50) have shown that three tested bifidobacterial strains could influence in vitro invasion but not the adhesion of L. monocytogenes to Caco-2 cells, which could also be attributed to strain-specific factors or the relative ratio of the population density of the competing microorganisms. It is likely that L. monocytogenes' response to the presence of another microorganism depends on the time and the manner by which this microorganism is likely to interfere with a particular biological process. According to Basler et al. (51), under competition stress and when they sense adjacent microorganisms potentially threatening their survival, bacteria may employ mechanisms—often fundamentally associated with virulence in mammalian cells—to eliminate heterologous microorganisms. Recently, Quereda et al. (24) demonstrated that listeriolysin S (LLS), a virulence-associated bacteriocin of L. monocytogenes, is expressed against specific microorganisms of the host intestinal microbiota that produce substances harmful for the pathogen.

Our data on the transcriptional profile of L. monocytogenes did not reveal a systematic correlation between the regulation of virulence genes and the behavior of L. monocytogenes during growth or during infection of Caco-2 cells in the presence of the other bacteria. At 37°C, increased inlB and inlJ mRNA levels observed in L. monocytogenes during simultaneous growth with L. innocua could account for the increased invasion of Caco-2 cells by L. monocytogenes. Furthermore, prfA upregulation may be related to reduced L. monocytogenes fitness. Serving as the major regulator of L. monocytogenes virulence genes, prfA has a central role in balancing the survival and transition of L. monocytogenes between the host and the environment. Overexpression of prfA has been linked to attenuated growth and subsidence of processes associated with improved fitness outside the host (52–54). It could be hypothesized that the presence of specific bacteria in foods or the environment prepares L. monocytogenes for invasion and replication inside mammalian cells. However, this was not the case at 10°C, where despite the downregulation of inlA, increased L. monocytogenes invasion was observed. This discrepancy could be attributed to other internal factors which were not included in the present study and which might have a putative role in L. monocytogenes virulence that has not yet been determined. In addition, a temperature-dependent pattern in the expression levels of internalins has been described before (55). In that study, inlA, inlB, inlC, and inlJ had the lowest mRNA levels at 16°C (the lowest temperature examined in that particular study) and the highest mRNA levels at 37°C. In our study, this pattern was altered depending on the bacterial species cocultured with L. monocytogenes. In line with this, Archambaud et al. (21) showed that Lactobacillus casei and Lactobacillus paracasei, although closely related taxonomically, could induce different transcriptional responses in L. monocytogenes during infection of mice in vivo.

L. innocua also triggered upregulation of sigB in L. monocytogenes at 37°C. sigB codes for the alternative factor σ^B, regulating a series of activities related to the infectious cycle of L. monocytogenes, such as the transcription of prfA (56, 57), inlA, inlB (58), and other σ^B-dependent virulence genes (59), as well as its survival under stressful conditions (60, 61). Thus, increased levels of transcription during coexistence of L. monocytogenes and L. innocua may be critical for the survival of the former within polymicrobial communities inside or outside the host. On the other hand, the trend toward downregulation of σ^B-mediated inlA and inlB in the presence of B. subtilis at 37°C may be reflected in the decreased ability for L. monocytogenes to invade Caco-2 cells following simultaneous growth with this bacterium. These observations, combined with the decline of the L. monocytogenes population in cocultivation with B. subtilis, may indicate a shutdown of defensive responses and mechanisms important for the survival and replication of the organism, both in the environment and in host cells.

In conclusion, we showed that L. monocytogenes adjusts its behavior (i.e., growth and virulence potential) as a response to coexisting microorganisms in its microenvironment. We showed that the regulation of L. monocytogenes virulence factors by other microorganisms may take place both during cocultivation outside the host and during in vitro infection of mammalian cells. This plasticity in L. monocytogenes responses depending on the surrounding microorganisms underscores the importance of interspecific interactions for the survival and pathogenicity of the bacterium. Further investigation is needed to assess potential common transcriptional alterations induced by different bacteria and to elucidate the complex pathways via which specific microorganisms provoke the manifestation of specific responses (common or different) in L. monocytogenes. Experimentation on real food systems, where physicochemical parameters and microstructure play a definitive role in microbial interactions (e.g., affecting, among other properties, cell motility and proximity of colonies), is also necessary. In addition, in vivo studies and/or studies including bacteria from the gut microbiome are necessary. The microbiome within the gut interacts not only with L. monocytogenes but also with the other incoming bacteria. These multidirectional interactions define the entrance into the host cells or the inhibition of the pathogen from entering them. A study including all these parameters would approach a more realistic scenario and would test the biological relevance of our in vitro data. Besides, in natural settings (e.g., foods, host gut, soil), L. monocytogenes is more commonly found in bacterial communities where interactions with other community members define the survival and replication of the pathogen. The study of these interactions may contribute to accurate predictions regarding L. monocytogenes' behavior in more realistic environments.

MATERIALS AND METHODS

Bacterial strains, culture, and growth conditions.

In this study, the strain L. monocytogenes ScottA was used as a reference virulent strain. Four microorganisms with different phylogenetic proximities to L. monocytogenes were used in the coculture studies: L. innocua (LQC 7510), Bacillus subtilis (FMCC B109), Lactobacillus plantarum (LQC 6193), and Pseudomonas aeruginosa (FMCC B26). All strains were stored at −80°C in tryptic soy broth (TSB; LabM, Lancashire, UK) or TSB with 0.6% yeast extract (TSB-Y, pH 7.2 [for listeriae]), both containing 20% glycerol. Accordingly, strains were maintained on tryptic soy agar (TSA; LabM, Lancashire, UK) or TSA supplemented with 0.6% yeast extract (TSA-Y [for listeriae]).

One single colony from each TSA or TSA-Y stock culture of L. monocytogenes, L. innocua, B. subtilis, L. plantarum, and P. aeruginosa was inoculated into 10 ml of TSB-Y for listeriae (30°C), TSB for B. subtilis (37°C) and P. aeruginosa (37°C), or de Man, Rogosa & Sharpe (MRS) broth for L. plantarum (30°C) and incubated for 24 h. Subsequently, 100 μl of each bacterial culture was transferred to 10 ml of fresh medium and incubated for 18 h at the optimal incubation temperature for each microorganism.

Determination of bacterial growth.

The 18-h cultures of all bacterial strains, corresponding to approximately 10⁹ CFU/ml for listeriae and 10⁸ CFU/ml for B. subtilis, P. aeruginosa, and L. plantarum, were washed twice with Ringer solution, resuspended in 10 ml TSB-Y, and serially diluted in TSB-Y to obtain a final inoculum of ca. 10³ CFU/ml. All microorganisms were grown at 10°C for 10 days or at 37°C for 24 h, in 6-well culture plates (Greiner Bio-One) as monocultures. In addition, we assessed the growth of L. monocytogenes in dual cocultures by mixing L. monocytogenes with each of the tested microorganisms at a 1:1 ratio. Sampling was performed on days 0, 1, 3, 5, 7, and 10 at 10°C and at 0, 2, 6, 9, 21, 22, and 24 h at 37°C. Bacterial CFU were determined by plating appropriate serial dilutions onto TSA-Y and, additionally for L. monocytogenes, onto Agar Listeria Ottaviani Agosti (ALOA; Biolife). The experiment was performed three independent times with duplicate samples.

Evaluation of cell contact-dependent growth inhibition of L. monocytogenes.

L. monocytogenes was cocultivated with B. subtilis or L. innocua. The following experiments were carried out in TSB-Y similarly to the aforementioned growth assays. Polyethylene tetraphthalate (PET) track-etched membrane inserts (Greiner Bio-One) of 0.4 μm pore size were placed in 6-well culture plates to create two compartments in the same well separated by the porous membrane. The multiwell plates and the membrane inserts used in the study are designed to allow the effective mixing of the growth medium and metabolic products during the spontaneous growth of two cultures, but not the passage of cells. Aliquots (2 ml) of L. monocytogenes culture were added to the upper chamber of the wells, while the cultures of L. innocua or B. subtilis were added to the lower chamber of the wells at the same volume. According to our previous screening experiments and published work (62), L. monocytogenes cells do not pass through the membrane pores of the inserts within the time frame of incubation. To confirm this, at the end of each incubation, 1 ml of the content of the lower chamber was spread on ALOA and no L. monocytogenes cells were enumerated. Therefore, we chose this as the best arrangement to safely ensure that no bacterial intermixing would occur and gravity would not be an issue. Furthermore, before each sampling, the cultures were shaken and well mixed, thus aiding in a better flow of metabolites. In addition, growth of L. monocytogenes in monoculture and coculture in direct contact (i.e., not separated by the membrane) with L. innocua or B. subtilis was monitored in separate wells. The colonies on ALOA that did not have a halo were enumerated as L. innocua; likewise, visual inspection and enumeration of B. subtilis colonies on TSA was possible due to the characteristic size and shape of the microorganism’s colonies. Cells were incubated at 37°C for 24 h and sampling was performed at 0, 2, 6, 9, 21, 22, and 24 h.

Coaggregation assay.

The ability of L. monocytogenes to coaggregate with B. subtilis was investigated following the protocol described by Collado et al. (44), with some modifications. Bacterial cells from activated cultures of each microorganism were washed twice with phosphate-buffered saline (PBS) by centrifugation (10 min, 4°C, 10,000 × g) and resuspended in the same buffer. The optical density at 600 nm (OD₆₀₀) of the bacterial suspensions was adjusted to 0.25 ± 0.05 in order to attain cell densities of 10⁷ to 10⁸ CFU/ml. Subsequently, 100 μl of L. monocytogenes and 100 μl of B. subtilis suspension were transferred and mixed in the wells of a 96-well microtiter plate. In addition, aliquots of 200 μl of each microorganism monoculture were transferred to separate wells and used as controls. The plate was incubated at 37°C for 24 h and the OD₆₀₀ of the bacterial suspensions was recorded every 6 h. Coaggregation was calculated according to equation 1:

$$\%\hspace{0.17em}\text{Coaggregation}=\frac{\frac{O{D}_{600}(A_{\text{Lm}}+A_{\text{Bs}})}{2}-O{D}_{600}(A_{\text{mix}})}{\frac{O{D}_{600}(A_{\text{Lm}}+A_{\text{Bs}})}{2}}\hspace{0.17em}\cdot \hspace{0.17em}100$$ (1)

where A_Lm and A_Bs represent the OD₆₀₀s of L. monocytogenes and B. subtilis suspensions, respectively, and A_mix represents the OD₆₀₀ of their mixture at the recorded time points. The experiment was repeated twice in duplicate.

In vitro virulence assay.

Human intestinal epithelial Caco-2 cells (American Type Culture Collection, ATCC) were cultured in Eagle’s minimum essential medium (MEM), supplemented with 15% (vol/vol) fetal bovine serum (FBS), 2 mM l-glutamine, 100 units/ml penicillin/streptomycin, and 1% (vol/vol) nonessential amino acids (all from Biochrom), at 37°C and 5% CO₂.

Invasion efficiency and intracellular proliferation of L. monocytogenes in Caco-2 cell monolayers were assessed as previously described by Zilelidou et al. (62). Briefly, Caco-2 cells were plated into 24-well tissue culture plates (Greiner Bio-One) in MEM supplemented with 15% (vol/vol) FBS and incubated until they reached confluence. Twenty-four hours prior to the experiment, the culture medium was replaced by antibiotic-free MEM, supplemented with 0.1% (vol/vol) FBS.

Microorganisms were cultivated similarly to the growth experiments in TSB-Y at 10°C and 37°C for 3 days and 9 h, respectively, in 50 ml TSB-Y. At the aforementioned time points, we did not observe any significant differences between the population of L. monocytogenes in monoculture and the population in the cocultures. In addition, the cell densities of L. monocytogenes and of the competing microorganisms in the cocultures were at the same levels, thus excluding the possibility that different population levels would interfere with in vitro virulence and transcriptional analysis results.

Bacterial cultures were centrifuged (3,000 × g, 2 min) and resuspended in MEM, prewarmed at 37°C, to obtain a multiplicity of infection of ∼25. An aliquot of the L. monocytogenes inoculum was plated onto TSA-Y to ensure the bacterial population used for the infection was similar to that estimated during growth experiments. Caco-2 cell monolayers were infected with the cultures for 1 h at 37°C. Sixty minutes after infection, Caco-2 cells were washed twice with Dulbecco’s phosphate-buffered saline (DPBS) and incubated in MEM containing 0.1% FBS and 200 μg/ml gentamicin (Biochrom). At 45 min (invasion assay) or 4 h (intracellular proliferation assay) posttreatment with gentamicin, Caco-2 cells were washed twice with DPBS and lysed with 1 ml of cold 0.1% Triton X-100 (Applichem). Numbers of viable L. monocytogenes cells after 45 min or 4 h were determined by plating appropriate dilutions on both TSA-Y and ALOA. Plates were incubated at 37°C for 24 or 48 h, respectively.

Invasion efficiency was calculated as the number of invading bacteria divided by the number of bacteria in the initial inoculum, multiplied by 100. The intracellular growth index designated intracellular growth coefficient (IGC) was determined by equation 2:

$$\text{IGC}=\frac{\text{no. of bacteria after proliferation assay}-no\mathrm{.}\text{of bacteria after invasion assay}}{\text{no. of bacteria after invasion assay}}$$ (2)

The invasion efficiency and intracellular growth of L. monocytogenes were determined for monocultures, cocultures, or cultures grown individually but combined before the virulence assay.

To examine whether the other microorganisms used in the study had the potential to adhere to and invade the host cells, their monocultures were also tested for in vitro adherence and invasiveness. Caco-2 cell monolayers were infected with the monocultures of the microorganisms at 37°C for 1 h. Following infection, Caco-2 cells were washed twice with DPBS and either lysed with 1 ml of cold 0.1% Triton X-100 to determine adherence potential or incubated in MEM containing 0.1% FBS and 200 μg/ml gentamicin for 45 min and then lysed to determine the invasion efficiency of the tested microorganisms. Numbers of viable bacterial cells were determined by plating appropriate dilutions on TSA. In addition, the minimal bactericidal concentration (MBC) of gentamicin for the microorganisms was determined (<200 μg/ml) to ensure that detection of cells from the competing microorganisms would not be due to their resistance to gentamicin at the concentration used. Experiments were performed at least three independent times in triplicate.

In vitro gene expression assay.

RNA samples were collected at the time points described in the in vitro virulence assay section (i.e., following growth of bacteria at 10°C and 37°C for 3 days and 9 h, respectively). Bacterial cultures were mixed (1:3) with RNAlater solution (Ambion, Waltham, MA, USA) and centrifuged (12,000 × g; 1 min). The supernatant was discarded and the biomass was resuspended in 200 μl of RNAlater solution (Ambion, Waltham, MA, USA) and stored at –20°C for 2 weeks until further use. The experiment was performed two independent times in duplicate.

RNA extraction was performed using the PureLink RNA minikit (Ambion, Waltham, MA, USA) and cDNA synthesis using the SuperScript first-strand synthesis system for reverse transcriptase PCR (RT-PCR) (Invitrogen, Waltham, MA, USA) according to the instructions of the manufacturer. Real-time PCR was performed using KAPA SYBR qPCR kit Master Mix (2×) for ABI Prism (KapaBiosystems, Boston, MA, USA). Primers and PCR conditions are presented in Table 1. The intergenic spacer (IGS), rpoB, and tpi were evaluated as housekeeping genes, and values were normalized using rpoB as the reference gene; prfA, sigB, hly, plcA, plcB, inlA, inlB, inlC, and inlJ were selected due to the significance of these genes in virulence of L. monocytogenes. Two RT reactions were performed for each sample, containing ca. 0.1 μg RNA each and the resulting cDNA was used for the assessment of the expression of the genes under study. The threshold cycle (C_T) values of the target genes obtained during the real-time quantitative PCR (RT-qPCR) experiments were converted to relative expression and log₂ values for further analysis according to the procedure of Hadjilouka et al. (63).

TABLE 1.

Primer sequences, amplicon sizes, and PCR conditions used for the in vitro gene expression assay

Gene	Primer name	Primer sequence	Amplicon size (bp)	PCR efficiencya	Reference
Reference genes
tpi	tpi_f	AACACGGCATGACACCAATC	93	2.03	(64)
	tpi_r	CACGGATTTGACCACGTACC
rpoB	rpob_f	CCGCGATGCGAAAACAAT	69	2.04	(65)
	rpob_r	CCWACAGAGATACGGTTATCRAATGC
IGS	IGS_f	GGCCTATAGCTCAGCTGGTTA	135	2.03	(66)
	IGS_r	GCTGAGCTAAGGCCCCGTAAA
Virulence-associated genes
prfA	prfA_f	CTATTTGCGGTCAACTTTTAATCCT	100	2.09	(65)
	prfA_r	CCTAACTCCTGCATTGTTAAATTATCC
sigB	sigB_f	CCAAGAAAATGGCGATCAAGAC	166	2.13	(66)
	sigB_r	CGTTGCATCATATCTTCTAATAGCT
hly	hly_f	TACATTAGTGGAAAGATGG	153	1.98	(66)
	hly_r	ACATTCAAGCTATTATTTACA
plcA	plcA_f	CTAGAAGCAGGAATACGGTACA	115	1.94	(66)
	plcA_r	ATTGAGTAATCGTTTCTAAT
plcB	plcB_f	CAGGCTACCACTGTGCATATGAA	72	2.00	(65)
	plcB_r	CCATGTCTTCYGTTGCTTGATAATTG
inlA	inlA_f	AATGCTCAGGCAGCTACAMTTACA	114	2.12	(65)
	inlA_r	CGTGTCTGTTACRTTCGTTTTTCC
inlB	inlB_f	AAGCAMGATTTCATGGGAGAGT	78	2.04	(65)
	inlB_r	TTACCGTTCCATCAACATCATAACTT
inlC	inlC_f	ACTGGTCAGAAATGTGTGAATGA	80	2.06	(63)
	inlC_r	CCATCTGGGTCTTTGACAGT
inlJ	inlJ_f	TGCGTAAATGCTCACATCCAAG	81	2.03	(63)
	inlJ_r	TTGCCCTTCAGCATCCAAGT

^aThermocycling conditions: initial denaturation at 95°C for 20 s and then 40 cycles of 95°C for 10 s, 60°C for 30 s, and 72°C for 30 s. Melting curve analysis: 95°C for 15 s, then 60°C for 1 min and raise to 95°C at 0.3°C/s. Primer concentration: 1 μM.

Statistical analysis.

Data analysis was performed using Microsoft Excel 2013 and SPSS 22.0 for Mac (SPSS Inc., Chicago, IL, USA). For all pairwise comparisons, the Student’s t test was used. Differences were considered to be significant for P values of <0.05. Gene transcription data analysis was performed with the Excel Analysis ToolPak using a one-sample t test. Changes in gene expression were considered over- or underexpression when the log₂ of the fold change was significantly (P < 0.05) above 2 or below −2, respectively.