Probiotics have been defined as “live micro-organisms that, when administered in adequate amounts, confer a health benefit to the host” (Hill et al., 2014). Usually members of the genera lactobacilli and bifidobacteria, probiotics have been reported to have beneficial effects such as prevention of antibiotic-associated diarrhoea (Vanderhoof et al., 1999) and reduction in the incidence of atopic disease (Kalliomäki et al., 2001). The mechanisms used by the bacteria to exert positive effects are varied and include inhibition of pathogens (Coconnier et al., 1993; Mukherjee and Ramesh, 2015), modulation of immune responses (Jespersen et al., 2015; Rossi et al., 2016) and enhancement of epithelial barrier functions (Laval et al., 2015; Simeoli et al., 2015). However, in general, the bacterial molecules underlying probiosis remain poorly characterised.

Due to the success of probiotics in the treatment of gut disorders, many authors have investigated whether lactic acid bacteria (LAB) hold any promise when used topically for the treatment or prevention of skin diseases (reviewed in Knackstedt et al., 2020). However, due to the potential problems associated with maintaining bacterial viability in topical formulations, use of LAB-derived molecules or their lysates have been investigated in a number of studies (Golkar et al., 2021; Nam et al., 2021). Although the term is still evolving, bacterially derived preparations are generally referred to as ‘postbiotics’ (Aguilar-Toalá et al., 2021; Salminen et al., 2021). We have been assessing a lysate of Lacticaseibacillus rhamnosus GG (LGG) as a novel topical treatment for skin in health and disease. Our previous in vitro data suggest that LGG lysate could be developed as a novel anti-infective agent to inhibit Staphylococcus aureus infection of skin. We have demonstrated that LGG lysate can protect epidermal keratinocytes from the toxic effects of S. aureus (Mohammedsaeed et al., 2014). Specifically, LGG lysate inhibits S. aureus adhesion to primary human keratinocytes and can both competitively exclude and displace S. aureus from keratinocyte binding sites. Additionally, LGG lysate can inhibit the growth of S. aureus (Mohammedsaeed et al., 2014). These two activities of LGG lead to an increase in keratinocyte survival in the presence of S. aureus (Mohammedsaeed et al., 2014).

In the current study, we have investigated the anti-adhesive effects of the LGG lysate against S. aureus. The mechanism by which LGG adheres to intestinal mucus has been shown previously to involve pili (Kankainen et al., 2009; Nishiyama et al., 2016) that promote retention of LGG in the gastrointestinal tract (Lebeer et al., 2012; Nishiyama et al., 2016). Pilus production in LGG is encoded by the spaCBA gene cluster, with the SpaC protein having been shown to be primarily responsible for the high mucus binding activity of LGG (Kankainen et al., 2009; Lebeer et al., 2012; Nishiyama et al., 2016). In our previous work, we showed that SpaC was at least part of the mechanism whereby live LGG inhibits binding of S. aureus to keratinocytes (Spacova et al., 2020). In other LAB, several other adhesins have been identified. Often, proteins of central metabolism, these are referred to as “moonlighting” proteins because of their dual functionality. However, the presence of moonlighting adhesins has been underexplored in LGG. We hypothesised that the protective activity of the LGG lysate may involve quenching of staphylococcal keratinocyte binding sites by molecules within the LGG lysate, which would otherwise be available for S. aureus binding. Thus, the aim of the present study was to investigate the molecules underlying LGG-mediated inhibition of S. aureus binding and protection of keratinocyte viability.

We have been studying the potential of LGG cell-free lysates as a topical therapy targeted at prevention/treatment of S. aureus skin infection. For topical applications, a cell-free lysate has many advantages over the use of viable bacteria. Importantly, the safety concerns surrounding use of live bacteria on the skin would be negated as would be the potential problems associated with formulating live bacteria. Such bacterial preparations have recently been referred to as “postbiotics” which are defined as “a preparation of inanimate microorganisms and/or their components that confers a health benefit on the host” (Aguilar-Toalá et al., 2021; Salminen et al., 2021). However, if potential postbiotics like LGG lysate are to fulfil their potential as topical therapeutic agents, an understanding of the effector molecules mediating their effects is a pre-requisite. We have therefore investigated the molecules mediating the anti-adhesive effects of LGG against S. aureus. The efficacious molecules are apparently proteins because heat denaturation or protease treatment completely destroyed the activity of the lysate against S. aureus. However, the possibility that other molecules such as sugars on the surface of LGG may also be important to the anti-adhesive action of the lysate cannot be completely excluded. Indeed, exopolysaccharides have been shown to be important for LGG binding to the host in the intestine (Ruas-Madiedo et al., 2006).

A number of protein adhesins have been previously identified in lactobacilli (Pancholi and Fischetti, 1992; Edelman et al., 2012; Kainulainen and Korhonen, 2014). Of these, the involvement of the pilus protein SpaC as a mucus binding protein has been shown in a number of studies (Kankainen et al., 2009; Lebeer et al., 2012). Our present work shows that SpaC may also be involved in the mechanism by which LGG inhibits S. aureus adhesion. This is suggested by a number of observations: firstly, fractionation of the lysate and analysis of the fractions show the most efficacious fraction to contain SpaC; secondly, recombinant SpaC inhibits S. aureus adhesion in a dose-dependent manner; and lastly, the toxic effects of S. aureus on keratinocyte viability are negated by SpaC, but not a control protein, BSA.

Overall, these data are consistent with a conclusion that SpaC is involved in the mechanism by which LGG lysate inhibits the adhesion of S. aureus to keratinocytes. Indeed, in other work, SpaC was shown to be intimately involved in the mechanism by which live LGG protects keratinocytes from the effects of S. aureus (Spacova et al., 2020). However, the previous study only investigated the role of SpaC when live LGG was added to keratinocytes at the same time as the pathogen (Spacova et al., 2020). In the present study, using an LGG lysate, SpaC is almost certainly not the only protein involved. Of note in this regard is the observation that SpaC excludes but does not displace S. aureus from keratinocyte binding sites, whereas the whole LGG lysate possesses both activities. This suggests that other proteins are also involved in the full anti-adhesive activities of the lysate. Indeed, several other proteins were found in the 50% fraction and this fraction contains all the anti-adhesive activities of the lysate. Crucially, the 50% fraction does not inhibit the growth of S. aureus thus suggesting that adhesive proteins are likely to be the only molecules in this fraction responsible for its observed effects. The proteins contained within the 50% fraction were further concentrated by HPLC into a fraction (F4) which showed the highest efficacy in both adhesion and viability assays. The most abundant proteins in this F4 fraction are likely to be elongation factor Tu (EF-TU), glyceraldehyde-3-phosphate dehydrogenase (G3P), enolase and triosephosphate isomerase (TPI), because the major components of the fraction (as judged by gel electrophoresis) were proteins of the same molecular weights as these. Interestingly, SpaC must be a low abundance protein in both the 50% and F4 fractions because gel-top Tandem Mass Spectrometry identification of proteins did not detect it. Indeed, this technique is known to be reliable only for identification of abundant proteins (D. Knight, University of Manchester protein sequencing unit, personal communication). However, SpaC was shown to be present by Western blotting (which has an amplification effect and so is more sensitive than mass spectrometry) suggesting it is present in the fraction and hence may be part of the anti-adhesive mechanism. SpaC is part of the heteropolymeric SpaCBA pili that are complex structures. It is known that low protein expression can still result in high adhesive ability due to the "zipperlike” mechanism of adhesion’ (Segers and Lebeer, 2014). However, the mechanism of inhibiting S. aureus binding to keratinocytes appears to additionally involve different proteins probably because it is a completely different system.

The proteins EF-TU, G3P, enolase, and TPI have been previously reported to be important for adhesive function in several species of lactobacilli (Pancholi and Fischetti, 1992; Granato et al., 2004; Ramiah et al., 2008; Kainulainen et al., 2012; Dhanani and Bagchi, 2013; Glenting et al., 2013; Spurbeck et al., 2015; Jeffery, 2018). All these proteins have been previously described as so-called “moonlighting proteins”, i.e., proteins with an ability to perform functions unrelated to the canonical function ascribed to the protein (Kainulainen and Korhonen, 2014). For example, G3P is an intracellular enzyme central to glycolysis. However, it is found as a cell surface adhesion protein on several prokaryotes including L. plantarum and L. crispatus (Pancholi and Fischetti, 1992; Kainulainen et al., 2012). TPI, another glycolytic enzyme, has been shown to be involved in competitive exclusion and displacement of Clostridium sporogenes and Enterococcus faecalis from Caco-2 cells by L. plantarum (Ramiah et al., 2008). EF-Tu is involved in protein translation but is found at the cell surface as an adhesin mediating attachment of lactobacilli to mucins (Granato et al., 2004; Dhanani and Bagchi, 2013). Many of these moonlighting proteins have been shown to mediate bacterial adhesion to eukaryotic cells by binding to specific eukaryotic proteins such as fibronectin (Muñoz-Provencio et al., 2011; Salzillo et al., 2015). Evidence has demonstrated that enolase, EF-TU and G3P to be cell surface proteins of LGG (Espino et al., 2015) as well as having their usual cytoplasmic location. Such dual localisation is usually suggestive of moonlighting function (Kainulainen and Korhonen, 2014). In the current study, we show for the first time an anti-staphylococcal adhesive function of the moonlighting proteins TPI and enolase in L. rhamnosus GG. Interestingly, G3P and EF-Tu that did not inhibit S. aureus adhesion also did not protect keratinocytes demonstrating that these effects are due to specific proteins and not a random effect of protein present within the assay. These data suggest that one mechanism by which the lysate protects form the S. aureus-induced cell death of keratinocytes involves the ability of SpaC, TPI and enolase proteins to prevent the adhesion of S. aureus to the keratinocyte surface. Of note is the small but significant effects on adhesion which appear to be sufficient for a large effect on viability of keratinocytes. This may be due to reduced bacterial load. Potentially a small reduction in load may be sufficient to render the keratinocytes more capable of resisting the toxic effects of S. aureus perhaps by processes such as production of antimicrobial peptides (Mohanty et al., 2017; Abu-Humaidan et al., 2018).

Both the concentration and timing of LGG protein addition to the keratinocytes monolayers appear to be critical to its protective effect. When SpaC, TPI and enolase were added before S. aureus, keratinocytes were protected from the toxic effects of the pathogen in a dose-dependent manner. However, only the higher concentration (25 and 50 μg/mL) of TPI and enolase were able to protect and reduce keratinocyte cell death levels when added after the pathogen. Taken together our data suggest that various LGG lysate protein components confer different mechanisms of protection involving exclusion and displacement of S. aureus from keratinocyte binding sites.

In summary, we suggest that the overall anti-adhesive function of LGG against S. aureus may be facilitated by a number of proteins including SpaC, and the moonlighting proteins TPI and enolase, although others may also be involved. We hypothesise that a number of proteins, including the ones identified, act together to inhibit S. aureus adhesion although the exact contribution of each remains to be established.

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

RC and CE-C designed and conducted the experiments, and performed data analysis on SpaC and moonlighting proteins. CE-C drafted the manuscript with support from CO. WM performed some additional viability assays and lysate fractionation. VK produced the SpaC antibody. AM, SL, and RS assisted with theoretical interpretation. CO conceived the project, supervised the work and oversaw the production of the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that this study received funding from SkinBiotherapeutics PLC. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication.

CO and AM have equity in SkinBioTherapeutics. However, the work described in this manuscript is purely of academic interest.

The remaining 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.