Bringing light into the dark site of infection.

Abstract

Quantification of pathogens in infected tissues of experimentally infected animals is a difficult task, but essential to determine virulence attenuation of fungal mutants and to monitor progression or clearance of infection under antifungal therapy. Especially for fungal infections there is still a debate on the best technique to assess fungal burden and current methods based on the detection of cell wall components, the determination of colony forming units (CFU), or quantitative PCR 1, 2. Galactomannan or β-d-glucan ELISA is frequently used to determine soluble fungal cell-wall components from blood samples 3, but these analyses only provide a qualitative assessment of infection. Although quantification of insoluble cell-wall chitin has been improved and provides excellent data on cell-wall chitin contents 4, application of this method for fungal burden determination from infected tissues would require the isolation of fungal cells for flow cytometry analyses, which appears not suitable for rapid diagnostic purposes. Thus, quantitative data are generally obtained from CFU and PCR analyses. However, even these two techniques may provide significantly varying results, since PCR quantification bases on nucleus equivalents, whereby the number of nuclei per cell is strongly species and morphology dependent. On the contrary, filamentous growing fungi frequently produce a dense mycelium composed of a large number of individual cells within the filaments, which may grow as a single colony on plates leading to a severe under determination of the actual load. However, the latter two techniques are generally accepted as a gold standard for quantification of the fungal burden. This is especially true for research on Candida albicans, a dimorphic fungus causing superficial oral, intestinal, and vaginal mucosal infections 5, 6, as well as life-threatening invasive and disseminated disease 7. Interestingly, despite the debate on the most appropriate technique for quantification of fungal burden in a specific tissue, it is frequently completely unknown which body sites are indeed infected. To address this question different in vivo imaging techniques have been established, whereby one of the most sensitive methods is bioluminescence imaging (BLI) 8. BLI has a much lower resolution than conventional fluorescence-based microscopic techniques that are perfectly suited to study cell–cell interactions as shown in a recent automated approach to classify cellular morphology changes of C. albicans during epithelial cell invasion 9. However, BLI can be used as a non-invasive technique to localize pathogens in a living host and may be used to provide localization data for subsequent microscopic analyses. In this respect, the power of BLI for detection of cryptic host sites of infection has recently been shown under antifungal therapy in a disseminated candidiasis model by using a firefly luciferase based BLI system. In this model system, C. albicans is injected via the tail-vein leading to bloodstream infections that allow dissemination to all organs. In the initial phase of infection, C. albicans is detected at all body sites including the liver, but fungal burden declines and the main infection establishes in the kidneys 10. However, by using a BLI approach it was possible to detect C. albicans under antifungal therapy in a central spotted liver area, which turned out to comprise the gall bladder. Within this organ, C. albicans appears to be protected from antifungals and forms a reservoir for re-colonization of the intestinal tract 11. Without an in vivo imaging technique this cryptic reservoir would have been completely overlooked. Additionally, when CFU counting and BLI were used to assess the efficacy of fluconazole in the treatment of OPC, only the BLI method produced significant data at early time points of therapy, although seven out of ten mice revealed negative CFU counts. This indicates that the BLI method produces more stable data that more reliably track differences in the fungal load. In conclusion, the manuscript by Gabrielli et al. 12 provides an excellent example for the application of BLI in determination and quantification of fungal burden in OPC mice. Especially when the oral cavity is monitored at early time points, this technique provides more reliable results compared to standard CFU techniques. However, it should be mentioned that the Gaussia luciferase based system used here is not suitable for monitoring deep-seated infections, which requires ex vivo imaging for assessment of OPC from internal organs such as oesophagus or stomach. Therefore, it would be interesting to compare in future experiments the performance of a firefly luciferase based system in OPC assessment with Gaussia luciferase BLI.