Interaction of Candida albicans with human gut epithelium in the presence of Live Biotherapeutic Products (LBPs)

The gut microbiome is arguably one of the most mysterious ‘organs’ of the human body and is vital to all aspects of human health and our sense of wellbeing. Over the past few decades significant research has been conducted to understand the dynamics involved in microbe-to-microbe and microbe-to-gut interactions and how these impact on overall human health. One microbe that has been the subject of ongoing investigation is Candida albicans, an opportunistic pathogenic yeast found in about 70% of people¹. As a polymorphic fungus, it is generally considered to inhabit the body as a commensal, kept under control by the host’s beneficial microbiota. However, in certain circumstances, such as in immunocompromised individuals², and during prolonged antibiotic therapy³, C. albicans is able to overgrow within its local environment, that is, within the gastrointestinal (GI) tract or translocate across the gut epithelium leading to systemic Candidiasis. This disease has a high morbidity and mortality ranging from 20% to 49%^4–6. Studies suggest that C. albicans adversely affects inflammatory bowel disease (IBD) exacerbating inflammation in the gut and delaying healing of ulcerative colitis, for example in humans and mice^7–9. An increased abundance of C. albicans has been observed in patients with IBD compared with healthy subjects suggesting that fungi may play a role in its pathogenesis^10,11.

Scientific investigations have identified various fungal genes known to play a role in C. albicans pathogenicity. However, there are still gaps in current knowledge of atypical virulence mechanisms, particularly in understanding the ability of C. albicans to invade gut epithelial cells. The majority of studies have focused on the interaction of C. albicans with oral/vaginal epithelial cells as opposed to gastrointestinal (GI) epithelial cells¹². While antibiotics are still the drug of choice to treat bacterial and fungal infections in clinic, Live Biotherapeutic Products (LBPs) have been suggested as an alternative for treating infections. LBPs are defined by the Food and Drug Administration (FDA) Centre for Biologic Evaluation and Research (CBER) as ‘a live biological product that: (1) contains live organisms, such as bacteria; (2) is applicable to the prevention, treatment or cure of a disease or condition of human beings; and (3) is not a vaccine’¹³. They are further described as ‘medicinal products containing live micro-organisms (bacteria or yeasts) for human use’ by the European Pharmacopoeia (Ph. Eur.) (which excludes faecal microbiota transplants and gene therapy agents from this category)¹⁴. The investigation of LBPs in treating inflammatory, autoimmune and even malignant conditions is accelerating at an astonishing rate, being recognised as novel drug candidates that aim to change the medical paradigm in treating human illness¹⁵. In this study we investigated the competitive inhibitory effects of six LBPs on adhesion and invasion of C. albicans using a co-culture of Caco-2:HT29-MTX cells as a model of human gut epithelium to provide insight into the potential of these LBPs for managing Candidiasis.

Current methods to investigate pathogenic interactions of microbes on gut epithelium rely on using cell lines that resemble biomimetic synthetic intestines, mainly Caco-2 or HT-29 cell lines^16–18. Caco-2 cells are differentiated in culture medium to form a polarized cell monolayer with tight junctions and microvilli to resemble important characteristics of human intestinal mature enterocytes. The main drawback of this cell line is that it does not produce a sufficient mucus layer. HT29, with methotrexate (MTX) adaptation, differentiates in culture media to secret mucin¹⁹, an essential component of the gut epithelium. We used a co-culture of Caco-2 and HT29-MTX cells to investigate the interaction of C. albicans ATCC 10231 with the gut epithelium. Six LBP candidates were selected and provided by Servatus Biopharmaceuticals: SVT 01D1, SVT 04P1, SVT 05P2, SVT 06B1, SVT 07R1 and SVT 08Z1.

Interaction of C. albicans ATCC 10231 at a final concentration of 10⁶ CFU/mL with the co-culture Caco-2:HT29-MTX (9:1) alone and in the presence of each of the LBPs (10⁶ CFU/mL) was assessed by measuring reduction in C. albicans colonisation when co-inoculated with LBP strains, and when pre-inoculated for 60 min with LBP strains prior to inoculation of C. albicans. The number of adhering C. albicans per cell was recorded to identify the competitive ability of LBP strains to inhibit adherence of C. albicans per cell. The results indicated that the LBP strains (except SVT 04P1) reduced the colonisation of C. albicans on the co-culture cells by 21–43% (Figure 1a) and adhesion per cell by 21–64% (Figure 1b) both in co-inoculation and pre-inoculation assays.

Figure 1.  Percent colonisation on co-culture of Caco-2:HT29-MTX by C. albicans ATCC 10231 (a) and number of adhering C. albicans per cell (b) when co-inoculated with LBPs (blue) and following pre-inoculation with LBPs (orange). E. coli strain 46-4 was used as a negative control. Error bars represent SEM.

All LBP strains (except SVT 04P1) demonstrated a significant reduction in colonisation and adhesion per cell of C. albicans (P < 0.01 in both co- and pre-inoculation). Overall, reduction in number of adhering C. albicans ATCC 10231 was seen for both co-inoculation and pre-inoculation, with SVT 07R1 showing the highest reduction (P = 0.0005).

In scanning electron microscopy (Figure 2) of the co-culture assay, C. albicans ATCC 10231 was shown to be highly invasive in that the C. albicans hypha was seen penetrating the epithelial cell monolayer. A similar procedure to the adhesion assay was used for co-inoculation and pre-inoculation in the invasion assay. A suspension of C. albicans was inoculated into 96-well plates at a final concentration of 10⁷ CFU/well. After 90 min the wells were inoculated with nystatin (24 μg/mL) for 60 min to kill any extracellular C. albicans, followed by incubation with 0.1% Triton-X-10 (Sigma-Aldrich) for 15 min to lyse the monolayer releasing invading pathogens and enumerating them. The results showed a reduction in invasion of C. albicans in the presence of LBP strains that demonstrated variable efficacy (Figure 3) with SVT 01D1 showing the highest reduction overall.

Figure 2.  Scanning electron micrograph showing invasion of C. albicans strain ATCC 10231 into Caco-2/HT29-MTX cells after a 20-min incubation. Scale bar = 5 µm.

Figure 3.  The number of invading C. albicans ATCC 10231 cells in a co-culture of Caco-2:HT29-MTX cells when co-inoculated with the LBPs (blue) and following pre-inoculation with LBPs (orange). Error bars represent SEM.

The escalating need to develop alternative approaches for managing C. albicans infection has highlighted the potential for the use of LBPs as anti-fungal therapeutics. We showed that most LBPs used in this study showed a significant reduction in the adhesion and invasion of C. albicans in our human gut epithelium cell culture model, although these effects varied among the LBPs. The potential use of these LBPs as a therapeutic or as a prophylactic measure was also tested using co-inoculation and pre-inoculation models of the LBPs against the C. albicans. While there was a significant reduction in colonisation and invasion of the cells by C. albicans in the presence of LBPs, we did not observe a significant difference between co-inoculation and pre-inoculation of LBPs one hour before the addition of C. albicans. Poupet et al. studied the curative effect of LBP L. rhamnosus Lcr35^® on Caenorhabditis elegans survival after C. albicans exposure, and found that the 2-h and the 4-h pre-inoculation periods were most protective against C. albicans infection²⁰. Future studies to investigate the effect of longer incubation periods of LBPs used in our study can provide a better understanding of the impact of pre-inoculation time in clinical studies aiming to assess the prophylactic effect of LBPs against C. albicans. Furthermore, it could prove insightful to explore the efficacy of various other LBP strains against C. albicans. Similarly, the use of further Candida strains in future studies for comparison would strengthen our findings.

Although the most reliable model to establish the impact of LBPs against C. albicans and other enteric pathogens is clinical trials in humans, in vitro studies utilising a co-culture of Caco-2 and HT29-MTX cells lines as used in this study, provide a suitable model to mimic the human gut epithelium. Caco-2 cells can be differentiated in the culture medium to form a polarized cell monolayer with tight junctions and microvilli that resemble important characteristics of human intestinal mature enterocytes. The other cell line, HT29, with methotrexate (MTX) adaptation, differentiates in culture media to secret mucin. In this study we used this co-culture model to investigate the efficacy of the LBPs against C. albicans, however, to achieve a far more reliable and robust gut epithelium model which resembles biomimetic molecular mechanisms in the intestinal niche, further improvements of this model such as the use of secretory IgAs and/or various other crucial antibodies/cytokines necessary for managing gut microbiome homeostasis are necessary²¹.

This fascinating field of research has significant potential for determining the link in a chain of events involving interactions between C. albicans and the gut epithelium where LBPs are used to treat the invading pathogens.

We are currently investigating the cellular response of the gut epithelial cells to C. albicans colonisation by comparing global gene expression (using RNA sequencing) with and without co-inoculation of LBPs. The RNAs will represent a snapshot of interaction/non-interaction of C. albicans with gut epithelium cells following the competitive adhesion of C. albicans with and without LBPs and identify genes that play a major role during these interactions. This will further our understanding of the mechanisms associated with using LBPs to treat invading pathogens.

The authors declare no conflicts of interest.

This research was funded by a Research Training Program (RTP) scholarship from University of the Sunshine Coast and a top-up scholarship from Servatus Biopharmaceuticals.