|Home | About | Journals | Submit | Contact Us | Français|
This protocol describes the setup, maintenance and characteristics of models of epithelial Candida infections based on well-established three-dimensional organotypic tissues of human oral and vaginal mucosa. Infection experiments are highly reproducible and can be used for the direct analysis of pathogen–epithelial cell interactions. This allows detailed investigations of Candida albicans wild type or mutant strain interaction with epithelial tissue or the evaluation of the host immune response using histological, biochemical and molecular methods. As such, the models can be utilized as a tool to investigate cellular interactions or protein and gene expression that are not complicated by non-epithelial factors. To study the impact of innate immunity or the antifungal activity of natural and non-natural compounds, the mucosal infection models can be supplemented with immune cells, antimicrobial agents or probiotic bacteria. The model requires at least 3 days to be established and can be maintained thereafter for 2–4 days.
The mucosal epithelium is of central importance in host defense and immune surveillance, as it is the primary cell layer that initially encounters environmental microbes with pathogenic potential. Other commensal microorganisms are regular inhabitants and harmless colonizers of mucosal surfaces with often important roles for the protection and immune regulation of epithelial tissues. Both types of microbes interact with epithelial cells, leading to either a favorable coexistence (commensal) or a breach of the mucosal barrier and subsequent cell injury (pathogen)1. Barrier function alone is usually adequate to restrain commensal microbes, but is often insufficient to protect against microbial pathogens. Accordingly, the oral epithelium is able to secrete a variety of defense effector molecules and to orchestrate an immune inflammatory response to activate myeloid cells in the submucosal layers to clear the invading pathogens2,3.
Epithelial tissue interaction and penetration is central in understanding pathogenic effects of infectious agents. However, the testing protocols have traditionally involved animals or animal cell cultures as an in vivo model of infection. The relevance of these approaches is often limited as human-specific characteristics and end points related to human pathological processes are commonly overlooked. Human cell lines that closely parallel the in vivo situation and allow studies of relevant physiological functions and microbial responses are therefore highly desirable. Recent advances in tissue engineering and molecular and cellular biology have elucidated ways to construct tissues using carefully designed scaffolds on which cells can grow and differentiate. These bioengineered tissues can be manipulated to express phenotypic surface markers and to produce immune modulatory molecules.
The complex interactions between microbial virulence attributes of a pathogen and defense mechanisms of the human host are of particular interest in the case of Candida species, which can either colonize asymptomatically as commensals or cause direct immunopathology as pathogens. Given that Candida is the most common fungal pathogen of man, models of mucosal Candida infections (candidiasis) that closely parallel the in vivo situation and allow the study of relevant physiological functions are particularly beneficial. A number of epithelial tissue models have been utilized to study C. albicans infections4–6; however, in this protocol, we will focus on the use of well-established three-dimensional organotypic epithelial models of human oral and vaginal mucosa provided by SkinEthic Laboratories. SkinEthic is a leading tissue production company specialized in the reconstruction of human epidermal and epithelial tissues for in vitro test applications for the pharmaceutical, chemical, academic and consumer product industry (http://www.skinethic.com).
The epithelial cells are seeded on inert filter substrates that are lifted to the air–liquid interface in a humidified-air incubator. A fully defined nutrient medium feeds the basal cells through the filter substratum. After 5 days, a stratified epithelium is formed that closely resembles human epithelium in vivo. As the oral and vaginal epithelial tissues are reconstituted in a physiologically natural environment and on a chemically defined medium, they express all natural major markers of the epithelial basement membrane and of epithelial differentiation and behave like human in vivo epithelium when treated with pharmacologically active but also irritating products. The models also exhibit tissue repair mechanisms that reflect the natural wound healing processes in vivo.
Moreover, the totally defined and serum-free culture environment allows the detection of very small quantities of inflammatory mediators, cytokines or growth factors secreted by the epithelium in response to topical application of test substances to be assessed in a very reproducible way. Biological controls before use include guaranteed absence of HIV-integrated pro-viral DNA, hepatitis C viral DNA, cytomegalovirus DNA, mycoplasma, hepatitis B antigen HBs and bacteria and fungi (http://www.skinethic.com).
The use of the SkinEthic reconstituted human epithelia (RHE) models was pioneered to study C. albicans infection and specifically the role of the secreted aspartyl proteinases (SAP) gene family in C. albicans pathogenicity7. The SAP family was shown to be differentially regulated during RHE-C. albicans infections. By using a combination of SAP-disrupted mutants and proteinase inhibitors, including the HIV-proteinase inhibitors saquinavir or indinavir, it was demonstrated that certain members of the Sap subfamily were responsible for inducing tissue damage (infection) in the oral or vaginal models8–10.
In recent years, the RHE models have been widely used to analyze the expression patterns of many C. albicans genes and to evaluate the consequence of gene disruption on pathogenicity and environmental sensing11–17. Recently, we have analyzed the genome-wide expression pattern of C. albicans during RHE infection and have compared these data to in vivo transcriptional profiles from patient samples (Zakikhany, Naglik, Schmidt-Westhausen, Holland, Schaller & Hube, unpublished data). These in vivo transcript profiling data confirmed trends observed during experimental oral RHE infections, thus demonstrating the value of the models as surrogates of in vivo infections.
During C. albicans infection of the oral RHE, using real-time RT–PCR, a significant increase in expression of interleukin (IL)-1α, IL-1β, tumor necrosis factor-α, granulocyte macrophage-colony stimulating factor and IL-8 was revealed18. Protein studies confirmed the gene expression data and supported the hypothesis that C. albicans infection induces an epithelial cytokine pattern that may favor a chemotactic and Th-1-type immune response and an environmental switch from an anti- to a pro-inflammatory milieu. Addition of the aspartyl proteinase inhibitor pepstatin A strongly reduced the cytokine response, and mutants lacking SAP genes caused reduced tissue damage and had a significantly reduced potential to stimulate cytokine expression in the vaginal model19. These observations support the hypothesis of an active host–fungus interaction at the epithelial surface, which comprises a dynamic adaptation of both the host and C. albicans during the transition from commensalism to parasitism.
A protective anti-Candida Th1-type epithelial response is likely to contribute to the recruitment of polymorphonuclear cells (PMNs) and lymphocytes to the site of mucosal infection to deal with the fungus. To test this hypothesis in the oral RHE model, a PMN supplementation assay to study the effect of PMNs during experimental oral candidiasis was established20. Infection of RHE with C. albicans alone induced IL-1α, IL-1β and tumor necrosis factor-α, with strong upregulation of granulocyte macrophage-colony stimulating factor and IL-8, which was directly correlated with chemoattraction of PMNs to the site of infection. The addition of PMNs enhanced production of the majority of these cytokines. Notably, C. albicans-induced tissue damage was significantly reduced when PMNs migrated through a modified perforated basal polycarbonate filter or when PMNs were applied to the apical epithelial surface. Interestingly, this protection of the epithelial tissue was also observed when PMNs were placed on the basal side of non-perforated filters, which prevents cell–cell interactions and cell migration but allows free passage of soluble factors, that is, cytokines20. Finally, the addition of saliva to the surface of the oral RHE or the addition of Lactobacilli to the vaginal model in the absence of PMNs also inhibits the C. albicans infection phenotype (unpublished results).
Numerous protocols have been successfully used to observe the pathogenesis of experimental mucosal infection processes. Host–pathogen interactions on the cellular level can be monitored by scanning (Fig. 1) and transmission electron microscopy8,9. Epithelial damage can be visualized by histological analysis of the embedded RHE7–9,20–22 and quantified by the extracellular lactate dehydrogenase (LDH) activity in the culture medium released by damaged epithelial cells9,16,20. Furthermore, immunoelectron microscopy8,9,16,21, confocal laser microscopy (unpublished results), fluorescence-activated cell sorting19 and ELISA18 can be used to measure and localize protein expression, and RT–PCR7–9,18–20,23 and microarray-based transcriptional profiling (unpublished results) for gene expression studies.
The overall RHE infection procedure is outlined in a flowchart form (Fig. 2).
Several further modifications to the RHE model system are possible to analyze certain aspects of mucosal infections, host–pathogen interactions or antimicrobial agents. These include:
Inoculation of the oral or vaginal RHE with reference wild-type strain SC531425 leads to hyphal production caused by contact with the epithelial tissue (Fig. 1) and should induce signs of tissue damage characterized by edema, vacuolization and detachment of keratinocytes by 24 h (Fig. 3a). A temporal progression of epithelial invasion and damage should be clearly evident in time-course experiments by both microscopic analysis of histological sections and increased LDH secretion into the culture medium.
Addition of PMNs to the apical layers of the oral or vaginal epithelium 6 and 12 h after inoculation should significantly reduce C. albicans penetration and epithelial damage (Fig. 3b). Fungal growth and tissue damage should also be significantly reduced when PMNs are applied to the basal side of the intact polycarbonate filter (Fig. 3c) or when PMNs migrate through the filter after perforation with a thin needle (Fig. 3d,e).
We thank all the previous undergraduate and graduate students who have completed a project or a thesis in the laboratory and helped to build the foundations of these protocols. We thank Birgit Fehrenbacher, Renate Nordin, Helga Möller and Hannelore Bischof, University of Tuebingen, for excellent technical assistance and Gudrun Holland and Muhsin Özel, Robert Koch-Institute, for providing Figure 3. M.S. and G.W. were supported by the Deutsche Forschungsgemeinschaft (Sch 897/1; Sch 897/3), B.H. by the Robert Koch-Institute, the Deutsche Forschungsgemeinschaft (Hu 528/8; Hu 528/10) and the European Commission Union (QLK2-2000-00795; “Galar Fungail consortium”) and M.S. and J.N. by NIH grant R01 DE017514-01 and J.N. by a personal Wellcome Trust Value in People (VIP) award.
COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests.
Reprints and permissions information is available online at http://npg.nature.com/reprintsandpermissions