|Home | About | Journals | Submit | Contact Us | Français|
Biofilms are a principal form of microbial growth and are critical to development of clinical infection. They are responsible for a broad spectrum of microbial infections in the human host. Many medically important fungi produce biofilms, including Candida , Aspergillus , Cryptococcus , Trichosporon , Coccidioides , and Pneumocystis . In this review we emphasize common features among fungal biofilms, and point toward genes and pathways that may have conserved roles.
Biofilm cell communities are more resistant to antifungal drugs than planktonic cells. Contributing factors include biofilm structural complexity, presence of extracellular matrix (ECM), metabolic heterogeneity intrinsic to biofilms, and biofilm-associated up-regulation of efflux pump genes. The actual fold increase in resistance varies with both the drug and species. Candida albicans and Candida parapsilosis biofilms are relatively resistant to fluconazole, amphotericin B, nystatin, voriconazole, and others. Aspergillus fumigatus biofilms are relatively resistant to itraconazole and, to some extent, to caspofungin. Cryptococcal biofilms are unaffected by fluconazole and voriconazole, and biofilms of Trichosporon asahii display elevated resistance to amphotericin B, caspofungin, voriconazole, and fluconazole. Azole and amphotericin B therapies are ineffective against Pneumocystis carinii biofilms. Biofilm-associated resistance mechanisms have been characterized in C. albicans and A. fumigatus and include drug binding by ECM and production of persister cells ,  (see supplementary references for this section in Text S1). Persister cells represent only a fraction of the population, and probably reflect its metabolic heterogeneity. These mechanisms may pertain to other fungi as well.
Biofilms are complex surface-associated cell populations embedded in an ECM that possess distinct phenotypes compared to their planktonic cell counterparts. Nutrients, quorum-sensing molecules, and surface contact are contributory factors. C. albicans biofilms are comprised primarily of yeast-form and hyphal cells, both of which are required for biofilm formation . Formation is a sequential process involving adherence to a substrate (either abiotic or mucosal surface), proliferation of yeast cells over the surface, and induction of hyphal formation . ECM accumulates as the biofilm matures, and seems to contribute to cohesion . C. albicans biofilms form on numerous abiotic  and biotic surfaces –. In denture stomatitis, a combination of biotic mucosal (the host) and abiotic surface (the denture) biofilm formation exists . Other Candida spp. including C. tropicalis, C. parapsilosis, and C. glabrata form ECM-containing biofilms but do not produce true hyphae .
Aspergillus biofilms can form both on abiotic and biotic surfaces . The initial colonizing cells that adhere to the substrate are conidia. Mycelia (the hyphal form) develop as the biofilm matures . ECM that binds the biofilm together  has been observed in vitro  and in vivo . Hyphal organisation is different in the two forms of A. fumigatus biofilm infection: aspergilloma infections present an intertwined ball of hyphae; aspergillosis infections present individual separated hyphae . Hyphae of C. albicans and of A. fumigatus can form pores or channels through biotic surfaces , .
The emerging fungal pathogen T. asahii forms biofilms comprised of yeast and hyphal cells embedded in matrix , as do those of Coccidioides immitis . C. neoformans forms biofilms consisting of yeast cells on many abiotic substrates , and shed capsular polysaccharide forms the ECM. Although C. neoformans forms hyphae in the course of mating, no hyphae have been observed in C. neoformans biofilms to date. Similarly, Pneumocystis species do not produce hyphal structures as part of their biofilms . Thus, hyphal formation is not a uniform feature of fungal biofilms.
Transcription factors play fundamental roles in both positive and negative regulation of biofilm formation through regulation of hyphal formation and cell surface proteins responsible for adherence . Bcr1, a C2H2 zinc finger transcription factor, is a critical determinant of C. albicans biofilm formation in all environments studied to date , , , . Bcr1 seems to be a conserved regulator of biofilm formation, because the Bcr1 ortholog of C. parapsilosis is required for biofilm formation as well . Ace2, another C2H2 zinc finger transcription factor, also contributes to C. albicans biofilm formation, probably through its role in adherence as well as hypha formation . The C. albicans transcription factor Efg1, a global regulator of cell surface protein genes and hyphal formation , is required for biofilm formation as well. The orthologs or best hits of Bcr1, Ace2, and Efg1, including C. glabrata CAGL0E06116g, CAGL0M04323g, and CAGL0L01771g, and C. parapsilosis CPAG00564, CPAG00148, and CPAG00178, are good candidates for biofilm regulators in those species. Transcription factors with analogous roles to Bcr1, Ace2, and Efg1 of C. albicans in A. fumigatus may be identified amongst the 124 uniquely expressed or upregulated transcription factors identified in biofilm culture by Gibbons et al., 2011 . The A. fumigatus transcription factor LAEA, a regulator of cell type and secondary metabolism gene clusters in A. fumigatus, is highly upregulated in biofilms , and it remains to be seen whether it may influence biofilm phenotypes.
Cell wall proteins are of particular interest in biofilm formation. Besides its expected role in adherence, the cell wall may have a sensory role that promotes adherence-induced responses . Numerous cell wall protein genes that may function in these capacities are upregulated early in C. albicans and A. fumigatus biofilm formation , , . C. albicans cell surface proteins have been reviewed authoritatively (see Text S1). Among the upregulated A. fumigatus surface proteins are the hydrophobins RodA, RodB, RodD, and RodE. RODB is thought to play the most crucial role with its expression increased over 4,000-fold in biofilm versus planktonic growth conditions . Ten other putative adhesins have been identified by Gibbons et al., 2011 . It is possible these Aspergillus proteins have functions analogous to known adhesins in C. albicans.
Biofilm cells have phenotypes distinct from planktonic cells, and this difference is reflected in greatest detail at the gene expression level. Detailed gene expression profiling comparisons, conducted in both C. albicans and A. fumigatus, have revealed substantial changes in gene expression between biofilm and planktonic cells , . Changes in transcription factor expression is characteristic of C. albicans biofilm formation in vitro and in vivo –, suggesting biofilm formation to be a highly regulated process. Similarly, almost 50% of the predicted transcription factors of A. fumigatus, including many with roles in asexual and sexual development, are upregulated in biofilms compared to planktonic cells.
Although biofilms are thought to include dormant cells, biofilms of C. albicans and A. fumigatus have increased expression of genes involved in protein synthesis. These genes encode ribosomal proteins, protein turnover, and translation factors as well as ribosomal proteins, indicating increased protein translation and ribosome production in biofilms to be a feature of biofilms , , . If indeed biofilm cells are nutrient limited, these particular gene expression features may optimize recycling of cellular constituents.
Upregulation of multi-drug resistance transporter genes is common to A. fumigatus (MDR1, MDR2, MDR4) and C. albicans (MDR1, CDR1, CDR2) biofilms in vitro . C. albicans MDR1 and CDR2 are upregulated in in vivo biofilms, as is PDR16, which is increased in fluconazole-resistant cells that overexpress CDR1 and CDR2 . Phase dependency of these transporters exists in vivo for C. albicans CDR1 and in A. fumigatus for MDR4 , . Additionally, ergosterol gene expression may account for increased drug resistance of biofilms. Genes involved in sterol biosynthesis are upregulated in A. fumigatus and C. albicans biofilms , , . Increases in ERG gene expression as well as multi-drug resistance transporters has been correlated with increased azole resistance in C. albicans patient isolate samples, though their contribution to biofilm-specific azole resistance has not been detected in mature biofilms (see Text S1).
Increased expression of adherence genes is also a property of biofilm cells. ALS1 is the most upregulated of the known adherence genes of C. albicans under biofilm conditions. Garcia-Sanchez et al. (2004)  highlight that the ALS genes are differentially expressed in biofilms and have autonomous contributions in the biofilm transcriptome. Nett et al. (2009)  observed differential expression of ALS genes at different stages of biofilm formation and potential for overlap of function in vivo. A similar pattern of differential adhesin expression is seen in vitro in the A. fumigatus biofilm environment . The inducing signal for biofilm adherence genes is clearly an area of interest as a basic biological question as well as a direction for prospective therapeutic development.
A significant number of primary metabolism genes, including those for amino acid synthesis, in particular sulfur amino acid biosynthesis, and nucleotide synthesis, are upregulated in C. albicans biofilms in vitro ,  and in vivo , relative to in planktonic cells in vitro. Many are regulated by GCN4, a transcriptional activator required for biofilm formation . Genes involved in amino acid metabolism are also upregulated in A. fumigatus biofilms including amino acid permeases, transporters, and amino peptidases. Secondary metabolism gene upregulation is significant in A. fumigatus biofilms, possibly due to upregulation of LAEA, a secondary metabolism regulator . Altered metabolic gene expression may reflect nutrient limitation, but the rapid kinetics of induction (in C. albicans at least ) may reflect a different regulatory signal.
Many cell wall biogenesis genes are induced in the biofilm environment. Altered expression of genes for β-1,3 glucan synthesis and modification are features of in vivo C. albicans biofilms including FKS1, BGL2, and XOG1 . Given the connection between the β-glucan pathway and biofilm matrix production, these may also contribute to ECM production. Nett et al. (2009)  highlight downregulation of β-1,3 glucan degrading enzymes in 24-hour biofilms and suggest this functions in glucan conservation for matrix production. In contrast, altered expression of α- and β-1,3 glucan synthesis genes is not observed in A. fumigatus biofilms. Although it is not directly reflected by the expression of polysaccharide synthase genes, the presence of α-1,3 glucan, galactosaminogalactan, and galactomannan in the mycelial extacellular matrix is correlated to the aerial growth of the mycelium of A. fumigatus . Expression of more than 50% of cell wall genes investigated in A. fumigatus is, however, altered in the biofilm habitat, including upregulation of the ROD genes. Thus, these two organisms both restructure their cell surfaces in biofilms, though they may use different mechanisms to achieve that outcome.
Genetic exchange is a feature of bacterial biofilms, mediated in part by extracellular DNA. Although extracellular DNA has been detected in C. albicans biofilms , the main mechanism of biofilm-associated genetic exchange involves mating and cell fusion. Most biofilm studies have been conducted with nonmating a/α cells, but biofilm formation of the mating-capable cell types, a/a and α/α, has revealed a unique regulatory pathway intimately tied to pheromone signalling. In order to mate, C. albicans must go through a switch from the white to opaque cell type. Upon switching, α/α opaque cells release a mating pheromone that induces a mating response in a/a opaque cells and vice versa. Pheromone release also induces an adhesive phenotype among the mating-incompetent a/a white cells , leading to mixed biofilm formation and ultimately mating .
Notably, genes upregulated specifically in white cells in response to pheromone exposure specify primarily cell wall and surface proteins . Several of these genes contribute to a/a-α/α biofilm formation . The configuration of the mating type locus also seems to affect global biofilm properties , which may result from distinct signalling pathways . If C. albicans has distinct ways to make a biofilm, it seems likely that other fungi will as well.
Fungal biofilms reflect a range of architectures. Regulators of biofilm formation may be conserved even among disparate biofilm architectures. From detailed analysis in C. albicans and A. fumigatus, there are numerous candidate genes that could be investigated in other biofilm-forming fungi. In addition to hyphal gene expression, characteristic biofilm gene expression patterns include increased expression of transcription factors and protein synthesis genes. Differential adhesin expression, upregulation of cell wall genes, and increased primary metabolism are features of the biofilm environment. Studies of mating pheromone effects on adherence highlight how a small portion of biofilm constituents can have a significant impact on biofilm formation. The presence of highly drug tolerant persister cells in biofilms (discussed above) is another illustration of the contribution of cell heterogeneity to overall biofilm properties. How other heterogeneous properties among biofilm cells may contribute to the overall development and integrity of pathogenic fungal biofilms will be an interesting question for future research.
We are grateful to past and present lab members for many helpful discussions about biofilms, and to Joe Suhan for insight into biofilm with electron microscopy. We thank Stephanie Guadagnini, Anne Beauvais, and J.P. Latge for providing the A. fumigatus biofilm image in Figure 1, and for critical reading of the manuscript. We apologize to our colleagues whose papers we could not cite due to space limitations.
The authors have declared that no competing interests exist.
Our work on biofilms was funded by NIH research grant R01 AI067703, and support from the Richard King Mellon Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.