Members of the Lactobacillus genus are abundant and predominant in the vaginal niche of healthy women of reproductive age, reaching a concentration of 10⁷ cfu/ml of vaginal sample, and 80% of the whole microbial content (Rönnqvist et al., 2006; Ceccarani et al., 2019). Most healthy women show the dominance of one species of Lactobacillus, and Lactobacillus crispatus, Lactobacillus iners, Lactobacillus jensenii, and Lactobacillus gasseri are the most frequent (Ravel et al., 2011). It is widely demonstrated that vaginal lactobacilli are involved in maintaining the state of vaginal eubiosis, preventing the colonization of genital, and sexually transmitted pathogens. Indeed, it has been reported that various vaginal Lactobacillus strains can interfere with Candida spp. overgrowth and reduce vulvovaginal candidiasis occurrence (Parolin et al., 2015; De Gregorio et al., 2019; Namarta et al., 2020). Anti-Candida activity of lactobacilli can be mediated through multiple mechanisms, including pH lowering, secretion of effective compounds, and impairment of fungal adhesion (Parolin et al., 2015, 2021; Calonghi et al., 2017; Abruzzo et al., 2018, 2021; De Gregorio et al., 2019).

Microorganisms’ growth can occur in two different forms: planktonic and biofilm, and the latter is the predominant form in nature (Rieu et al., 2014). The biofilm mode of growth is characterized by high cell density and secretion of a polymeric matrix, in which microorganisms are immersed (Flemming and Wingender, 2010). The formation of microbial biofilms inside the human body has important implications in health and disease: for example, pathogens’ biofilms generally show increased resistance to antibiotics, making their elimination challenging. On the other hand, the formation of biofilm by members of the human resident microbiota is desirable, as it can improve microbial persistence and favor functionality (Li et al., 2018). The planktonic growth or the alternative establishment of a sessile biofilm by a microbial species implies the remodeling of physiological and metabolic pathways, influencing in turn functional properties of probiotics. For some Lactobacillus species, it has been reported that biofilm mode of life can improve antagonistic effects toward bacterial pathogens and anti-inflammatory and immunomodulatory properties (Rieu et al., 2014; Aoudia et al., 2016; Liu et al., 2021), although few studies have investigated this issue till now.

In the present manuscript, 16 vaginal Lactobacillus strains, belonging to L. crispatus, L. gasseri, Lactobacillus vaginalis, and Lactobacillus plantarum species, were tested for their in vitro capacity to form biofilms. Culture supernatants derived from planktonic and biofilm growth were tested against a panel of Candida clinical isolates (Candida albicans, Candida glabrata, Candida lusitaniae, Candida tropicalis, Candida krusei, and Candida parapsilosis). Anti-Candida activity was then correlated with lactobacilli culture supernatants metabolome, assessed by ¹H-NMR.

The two different forms of microbial growth, i.e., planktonic and biofilm, differ for physiological and metabolic features; indeed, the establishment of a sessile culture by a microbial strain requires the remodeling of several molecular pathways involved in adhesion, secretion, and other cellular functions. Such modifications of the global metabolism can, in turn, influence the functional properties of probiotics, beyond favoring the permanence of probiotics inside the human body (Donlan and Costerton, 2002; Ventolini, 2015).

In this context, we first assessed the ability of Lactobacillus strains of vaginal origin, belonging to L. crispatus, L. gasseri, L. vaginalis, and L. plantarum species, to form biofilm. We found out that L. plantarum species was the best biofilm producer, probably because of its adaptive features to niches (van den Nieuwboer et al., 2016). Indeed, L. plantarum is a versatile species able to adapt to different environmental niches and its genome, considered as one of the largest genome in lactic acid bacteria, encodes a huge repertoire of surface sugar, extracellular proteins, and regulatory peptide involved in the regulation of adherence (Kleerebezem et al., 2003; Sturme et al., 2005). Other authors previously reported that L. plantarum was able to produce a robust biofilm in vitro, resistant to environmental stresses (Kubota et al., 2008). Considering the most peculiar species of the vaginal microbiota, high variability was registered in the level of biofilm formation among L. crispatus and L. gasseri strains. Notably, L. crispatus BC3 biofilm was comparable to L. plantarum strains. Biofilm formation by L. crispatus and L. gasseri has been poorly investigated in the literature, and it has been reported that some L. crispatus strains of vaginal origin were not able to form biofilm and weakly autoaggregated (van der Veer et al., 2018), while Stivala et al. (2021) reported about vaginal L. crispatus and L. gasseri (and L. rhamnosus) strains as weak to moderate biofilm producers, with great variability.

From our results, the shift from a Lactobacillus planktonic culture to an adherent form implied major differences in the metabolism of nitrogen and amino acids, suggesting a key role of this molecule category in the formation of an adherent biomass, as already proposed by Liu et al. for Lactobacillus paraplantarum (Liu et al., 2021). In particular, alanine and lysine were produced/released at higher amounts by lactobacilli in the biofilm form compared to the planktonic one, whereas methionine was highly consumed by Lactobacillus in biofilm growth. These data support the finding that the methionine pathway is involved in the biofilm formation process, as proposed by Liu et al. (2018) in L. paraplantarum.

Although the anti-Candida activity of planktonic-derived Lactobacillus CFS has been reported and explored by several authors (Parolin et al., 2015; Wang et al., 2017; De Gregorio et al., 2019; Jang et al., 2019; Tortelli et al., 2020), no data are available on bf-CFS anti-Candida potential. In the present manuscript, both planktonic- and biofilm-derived CFS were tested against a broad spectrum of Candida clinical isolates, including C. albicans and C. non-albicans species. Overall, CFS did not exert fungistatic activity, with few exceptions. Fungistatic activity was tested in standard conditions following EUCAST protocol and validated in SVF. In accordance with results reported by Parolin et al. (2015), C. krusei resulted in being the most resistant isolate, although acidic conditions enhanced CFS fungistatic effect. C. glabrata isolates were susceptible to Lactobacillus CFS and notably their growth was almost abrogated by L. crispatus bf-CFS, although this species is often reported as resistant to antifungal substances (Rodrigues et al., 2014). Unlike previous results (Parolin et al., 2015) that were obtained with CFS from lactobacilli in exponential growth phase, Lactobacillus CFS from the stationary phase were active toward C. parapsilosis, pointing out the importance of growth phase and resulting metabolites on the anti-Candida activity. In addition, bf-CFS were more effective against C. parapsilosis than the respective pk-CFS.

Lactobacillus CFS that demonstrated high anti-Candida activity all derived from biofilm mode of growth and were produced by L. crispatus and L. plantarum strains; on the contrary, low-activity CFS were derived from L. gasseri (BC11–BC14) biofilm cultures and L. vaginalis BC16. This observation suggested that the anti-Candida activity of L. crispatus strains was enhanced when they were able to establish an adherent biofilm, while L. gasseri strains behaved in the opposite manner. This result, in turn, can offer new hints in the perspective of using L. crispatus or L. gasseri as probiotics to prevent vulvovaginal candidiasis or to restore the vaginal eubiosis status, also in the light of developing fourth-generation probiotics, based on biofilm-forming strains and encapsulation techniques (Salas-Jara et al., 2016). Peculiarly, L. vaginalis BC16, especially when grown in the biofilm mode, slightly stimulated Candida proliferation. This behavior correlated with L. vaginalis BC16 weak attitude to form an adherent biofilm, suggesting that its preferred mode of growth (i.e., planktonic) assured the best conditions to prevent from Candida infection, sustaining in turn the healthy status of the vaginal ecological niche.

In the present manuscript, we also sought for peculiarities in the metabolome of Lactobacillus CFS that demonstrated highly effective toward Candida clinical isolates. Highly active CFS differentiated from low active ones mostly for carbohydrates and amino acid quantification. Among carbohydrates, high anti-Candida activity was accompanied by increased consumption of glucose and fructose (corresponding to lower amounts), two well-known glycolytic fuel molecules. A marked consumption of glucose has already been correlated with Lactobacillus antimicrobial effect (Nardini et al., 2016), highlighting the mechanism of competition for nutrients between human beneficial microbes and pathogens as crucial. On the other hand, galactose concentration was higher in active CFS compared to non-active ones and lactose/galactose metabolism has been identified by overrepresentation analysis as a differential metabolic pathway between active and non-active CFS, denoting the preference of active Lactobacillus strains to utilize other carbon sources than galactose to sustain their growth.

Another recognized mechanism supporting the antimicrobial activity of lactobacilli is lactic acid production by fermentation and acidification (Boskey et al., 2001; O’Hanlon et al., 2013). In this regard, lower amount of lactate was quantified in active CFS compared to non-active ones, suggesting that the observed anti-Candida effect was not directly attributable to this organic acid. CFS endowed with high anti-Candida activity also showed higher amounts of pyruvate with respect to low-activity CFS, suggesting that the antifungal activity did not depend on a burst of the classical lactic fermentation pathway. In addition, active CFS contained high amounts of acetate, although L. crispatus, L. gasseri, and L. plantarum are recognized as homofermentative lactic acid bacteria. Lactobacillus CFS fungistatic activity was assessed in standard and acidic conditions, suggesting that the anti-Candida effect is not merely dependent on the acidic pH. Nevertheless, when CFS acted in an acidic environment, the anti-Candida activity was enhanced, pointing up the existence of a synergic activity among different components of the Lactobacillus supernatants and suggesting that the vaginal physiological environment could improve CFS anti-Candida potential.

To the best of our knowledge, the present study reports the anti-Candida activity of vaginal Lactobacillus grown in a sessile mode for the first time. Biofilm formation by lactobacilli represents a crucial aspect in the vaginal ecology, as this phenomenon can possibly influence endogenous microbiota persistence and functionality. Here, we demonstrated that Lactobacillus adherent growth strongly affects the anti-Candida potential and could in turn modulate Candida colonization of the vaginal niche. In particular, biofilm formation enhanced L. crispatus and L. plantarum fungistatic activity, while L. gasseri became more active when grown in a free form. Lactobacilli metabolism turned different between planktonic and biofilm forms, though the correlation between metabolism and functional properties demands more investigation. Further studies on lactobacilli biofilm formation in the vaginal niche will be approached to better understand its physiological role, also in the perspective to support the development of new antimycotic strategies based on probiotics grown in adherence.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

CP and BV contributed to conception and design of the study. CP supervised the experiments. VC, LL, BG, and PD performed the experiments. CP, VC, LL, and MT analyzed the data. VC and MT produced the figures. CP and VC wrote the first draft of the manuscript. CF provided Candida strains. BV supervised the study. All authors contributed to manuscript revision, read, and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.