|Home | About | Journals | Submit | Contact Us | Français|
Biofilms are microbial communities that form on surfaces and are embedded in an extracellular matrix. C. albicans forms pathogenic mucosal biofilms that are evoked by changes in host immunity or mucosal ecology. Mucosal surfaces are inhabited by many microbial species; hence these biofilms are polymicrobial. Several recent studies have applied paradigms of biofilm analysis to study mucosal C. albicans infections. These studies reveal that the Bcr1 transcription factor is a master regulator of C. albicans biofilm formation under diverse conditions, though the most relevant Bcr1 target genes can vary with the biofilm niche. An important determinant of mucosal biofilm formation is the interaction with host defenses. Finally, studies of interactions between bacterial species and C. albicans provide insight into the communication mechanisms that endow polymicrobial biofilms with unique properties.
Biofilms are complex three-dimensional communities of microorganisms embedded in an extracellular matrix (ECM) layer that display unique phenotypes compared to free-living (planktonic) cells. They are found adhering to living tissue such as mucosal surfaces, or to a non-living surface such as an implanted medical device [1,2]. Most infections result from pathogenic biofilms, so the biomedical significance of biofilms is substantial [3,4]. Our focus is Candida albicans, a commensal fungus found on mucosal surfaces of the oral cavity, gastrointestinal tract, and genitourinary tract. It causes severe and recurrent mucosal infections as well as fatal invasive infections in both immunocompromised and immunocompetent individuals [5,6]. Until recently, most knowledge of C. albicans biofilms derived from the studies with abiotic substrates, such as intravascular catheters. The question that arises is how abiotic-surface biofilms are related to mucosal biofilms, and to what extent the findings with one model can be extrapolated to other models. Are functional determinants of mucosal biofilms distinct from those of abiotic-surface biofilms? Recent studies of mucosal biofilms found in a vaginal candidiasis model, an oropharyngeal candidiasis (OPC) model, and a denture biofilm model indicate that some of the major genetic determinants are similar. In addition, gene expression analysis of an intestinal colonization model resembles that of biofilms. Overall, these studies argue that diverse mucosal biofilms have substantial similarity to abiotic-surface biofilms.
C. albicans abiotic-surface biofilms consist of two general types of cells: yeast cells and filamentous cells [1,7]. In vitro, the basal biofilm layer is composed of yeast cells from which filamentous cells emanate. These yeast and filamentous cells are embedded in a dense layer of ECM [1,7]. The primary component of biofilm matrix is beta-glucan [1,8]. In vivo, biofilm architecture is more chaotic, with interspersed yeast and filamentous cells, and an ECM layer that contains host immune cells such as neutrophils [9,10].
Advances in genetic manipulation and expression profiling have helped define the regulatory pathways and mechanisms that govern C. albicans biofilm development [1,11] (Table 1). Several classes of gene products control biofilm development of in vitro abiotic-surface biofilms. These genes specify transcription factors (e.g. EFG1, BCR1, TYE7), protein kinases (e.g. IRE1, CBK1), known or predicted cell wall proteins (e.g. HWP1, ALS3), and an alcohol dehydrogenase (ADH1) [1,12,13]. BCR1, HWP1, and ALS3 lie in a regulatory pathway that is also required for in vivo abiotic-surface biofilm formation in a rat catheter model . Thus this regulatory system may be operative in many niches that support biofilm formation. Like bacterial biofilms, C. albicans abiotic-surface biofilms are associated with increased drug resistance compared to planktonic cells [1,7]. Decreased toxicity of clinically used antifungals, such as amphotericin B and flucanozole, to biofilm cells is due to ECM adsorbtion of drugs , formation of persister cells , and perhaps other mechanisms . Indeed, a recent expression-profiling study detected no transcriptional response of a biofilm to fluconazole, and a muted response to amphotericin B , in keeping with the idea that the drugs are denied access to biofilm cells. The extent and mechanism of antimicrobial resistance of mucosal biofilms is clearly an important issue from multiple standpoints.
About 75% of all women are affected by vulvovaginal candidiasis at least once in their lifetime, with a subset (5–8%) experiencing recurrent infections . Vaginitis infection models have been used extensively for some time . Recently, Noverr and colleagues looked specifically at vaginal infection as a biofilms (Table 1), using both in vivo and ex vivo approaches . Their study determined the kinetics of biofilm development in vivo by harvesting vaginae at 8, 24, 48 and 72 hours post-inoculation. This analysis revealed that the initial 8 hour time point involved yeast cell proliferation on vaginal epithelium with little or no ECM. By 24 hours, there was extensive filamentous cell differentiation and some ECM production. A similar sequence of events occurs in abiotic-surface biofilm formation in vitro [1,11]. ECM accumulation extended through 48 hours, and by 72 hours the ECM had entirely covered the biofilm. There was a substantial increase in fungal burden from 8 to 48 hours, followed by little increase between 48 and 72 hours. Interestingly, the ex vivo biofilm, with a substrate of processed vaginal tissue, included yeast form cells at the apical surface of the biofilms. These yeast cells may be engaged in biofilm dispersal [11,20]. It would be interesting to know if the many analogies between vaginal and abiotic-surface biofilms may extend to drug resistance.
Two key regulators of abiotic-surface biofilms, Efg1 and Bcr1, are required for formation of vaginal biofilms as well. Both of these transcription factors are required for expression of many filamentation genes, and functional evidence from abiotic-surface biofilms suggests that their target genes are required primarily for filamentous cell adherence [14,21]. Both efg1 and bcr1 mutants colonized the vaginal models extensively, but failed to produce ECM. Perhaps Efg1- and Bcr1-dependent adhesins contribute to ECM in the vaginal model. The fact that colonization and biofilm formation are readily distinguished in these models offers the opportunity to gain insight into features of biofilm formation that are not accessible in abiotic-surface models.
Oral mucosal Candida infections are prevalent among individuals with weakened, suppressed, or underdeveloped immune systems . Approximately 5% of newborns, 10% of elderly patients and almost 90% of HIV-infected individuals develop oral pseudomembraneous candidiasis . Oral mucosal infections have been studied for some time (Table 1) with both an in vivo oropharyngeal candidasis (OPC) model that utilizes immunosuppressed mice , as well as a reconstituted human epithelium (RHE) infection model in vitro [23,24]. Recent efforts by Dongari-Bagtzoglou and colleagues have asked how oral mucosal infection is related to biofilm formation . Microscopic analysis of both OPC and RHE infected surfaces revealed structural similarities with abiotic-surface biofilm architecture. As expected, the OPC biofilm is complex, in that the ECM layer contains commensal bacterial flora and host components such as neutrophils and keratin from desquamating epithelial cells. The ECM layer was abundant on cells at the basal end of biofilm close to the mucosal tissue and on cells invading the submucosal compartment .
As found with abiotic-surface and vaginal biofilms, the transcription factor Bcr1 is required for OPC biofilm formation as well . In the case of OPC biofilms, the adhesin Hwp1 is a pivotal Bcr1 target, because overexpression of Hwp1 in a strain lacking Bcr1 restored OPC biofilm formation. This result is consistent with pioneering work of Sundstrom and colleagues , who showed that Hwp1 functions as an epithelial cell adhesin through its covalent crosslinking to host cells, mediated by mammalian host transglutaminase. A second Bcr1 target, the surface protein Hyr1, is required for the transition from a mucosal surface biofilm to epithelial tissue invasion . Hyr1 protects C. albicans against neutrophil killing [26,28], and it is reasonable that this protective role would facilitate tissue invasion. It does not have an obvious role in abiotic-surface biofilm formation in vitro , which is consistent with its main role being to combat host defenses. Clearly, though, Hyr1 function should be tested in other pathogenic biofilm models. Finally, the adhesin Als3, a Bcr1 target with a pivotal role in abiotic-surface biofilms , has a minor role if any in OPC biofilms . This conclusion is qualified by the caveat that Als3 is part of a large adhesin family , and the oral environment may promote expression of other Als family members that substitute for Als3. In any event, the case of Als3 illustrates that biofilm formation requirements may differ significantly, depending upon the surface or niche. The case of Hyr1 illustrates that biofilm formation can be separated from tissue invasion and pathogenesis. Finally, the case of Hwp1 illustrates that a single protein may have distinct roles in diverse niches – as a substrate for host crosslinking activity in a mucosal biofilm; as a host enzyme-independent adhesin in abiotic-surface biofilms in vitro. It is unknown whether these OPC biofilm models exhibit altered antifungal drug sensitivity. The oral environment does support biofilm drug resistance, as illustrated by the denture model, discussed below. We believe that this is an important area for future investigation.
Denture stomatitis is the most common form of oral Candida infection, affecting 65% of denture wearers . Andes and colleagues have developed an in vivo biofilm denture model to mimic denture-associated stomatitis . In this model, the device and mucosal surfaces are both colonized due to contact between the host mucosal surface and the denture device. The model involves placement of denture material in immunosuppressed rats followed by inoculation of C. albicans cells, which are allowed to grow up to 72 hours. Histopathological analysis of denture samples showed adherence of yeast cells 6 hours and a confluent layer of yeast cells surrounded by ECM at 24 hours post inoculation. At 48 hours, a mature biofilm consisting of yeast and filamentous cells encased in matrix was visualized. This matrix component also contained host cells and bacteria.
The biofilm drug resistance phenotype is preserved in this model . Topical or systemic treatments with several drugs had little impact on fungal burden. Even suspended denture biofilm cells had much higher MICs than planktonic cells to such drugs as amphotericin B, fluconazole and micafungin. Resistance to amphotericin B and fluconazole is also evident in an in vitro denture biofilm model . However, resistance to an echinocandin (micafungin) has not been observed in other C. albicans biofilms, and may indicate that a unique resistance mechanism is operative in this niche.
One notable feature of this in vivo model is that ECM appears prior to accumulation of filamentous cells. It is possible that the ECM is primarily of host or bacterial origin; there is a substantial bacterial component of these biofilms [30,32]. A different perspective is that filamentation may be delayed (compared to overall growth) in this model. That notion is consistent with the inhibition of C. albicans filamentation by bacterial quorum sensing molecules , which are presumably present in the mixed-species biofilms that form here.
Two observations indicate that some abiotic-surface biofilm regulatory mechanisms are operative in this model. First, biofilm formation is Bcr1-dependent . Second, seven genes tested are up-regulated in these biofilm cells (in comparison to planktonic cells) as well as in rat venous catheter biofilm cells [30,34]. Hence the unique features of denture biofilms, such as echinocandin resistance, may derive from novel environmental responses in the context of established biofilm regulatory pathways.
C. albicans can colonize the gastrointestinal tract, where cells are in close proximity to mucosal surfaces surfaces. The transcription factor Efh1 has a pivotal role in control of colonization efficiency . Efh1 is a paralog of Efg1, which regulates abiotic-surface biofilms . No functional relationship between Efh1 and Bcr1 has been reported, though several Bcr1-responsive genes are induced during cecum colonization, as is Tec1, which stimulates Bcr1 expression [35,37]. The analogies between cecum colonization and abiotic-surface biofilms are fragmentary, but future study of this question is clearly warranted.
Mucosal biofilms are multi-species in nature. Indeed, the C. albicans denture and oral mucosal biofilms discussed above had significant numbers of bacterial and fungal cells [25,30]. We are beginning to understand the full extent of the microbial diversity of the oral cavity and in oral biofilms [38,39]. Interactions among C. albicans, bacteria, and the host can affect C. albicans survival, competitiveness, and morphogenesis . These interactions may promote or prevent disease in niches co-inhabited by C. albicans and bacteria . Several studies show interactions between C. albicans and oral bacteria like Streptococcus mutans, Streptococcus gordonii and Staphylococcus aureus [41–45]. S. mutans inhibits filamentous cell formation in C. albicans through a quorum-sensing molecule (competence-stimulating peptide) . Conversely, S. gordonii promotes C. albicans filamentation: it interacts with C. albicans Als3 through its adhesin SspB . Physical adherence facilitates C. albicans filamentation and biofilm development through response to the bacterial signal, autoinducer 2 . S. gordonii also binds to the adhesins Eap1 and Hwp1 . Mixed-species biofilms indeed have unique properties, a point illustrated by proteomic comparison of biofilms comprising S. aureus, C. albicans, or both, revealing differential regulation of virulence factor genes . Thus bacteria can attenuate or enhance fungal invasion and virulence in C. albicans mucosal biofilms. Finally, in light of a recent study that revealed fungal diversity in the oral microbiome , perhaps fungal-fungal interactions are significant factors in infection as well. The vaginal mucosa is colonized by bacteria, especially Lactobacilli species . Suppression of fungal growth by Lactobacilli has been documented in several studies [40,48–50]. The significance of these observations is reflected in the augmented risk of vaginal Candida infections from systemic antibacterial antibiotic therapy . These interactions argue that mucosal ecology plays a key role in promoting or preventing disease .
Mucosal epithelial cells play an important role in development of an innate immunity against Candida species [51–53]. Host cytokines particularly, interleukin-17 (IL-17, produced by Th17 cells) and IL-23 play a protective role against oral mucosal C. albicans infections [54,55]. The capacity of C. albicans to colonize mucosal surfaces and their subsequent recognition as a potential pathogen by host cells are important steps in development of mucosal infections . Oral epithelial cells can indeed differentiate between low commensal fungal burdens on mucosal surfaces and high damage-inducing pathogenic states, highlighted by studies showing a NF-κB and MAPK mediated biphasic response against C. albicans . C. albicans cells also undergo pathogenicity-associated changes, including upregulation of genes that facilitate host cell damage, confer resistance to host defense responses, or are associated with filamentation . The multi-species nature of mucosal biofilms is also acknowledged by host cells. For example, bacterial antigens determine the type of host immune response to mixed infections of bacteria and C. albicans . Thus highly significant interactions occur between the host and all resident microbes.
It has long been appreciated that mucosal infections are in essence biofilms. Recent studies have used the knowledge from abiotic-surface biofilm systems to guide functional analysis. Common themes have emerged, such as the requirement for Bcr1 for biofilms in most niches. Unique features have emerged as well, such as the influence of host interactions on mucosal biofilms, the complex interactions with bacterial flora, and biofilm-associated echinocandin resistance. These analogies point toward important questions for future study in both mucosal and abiotic-surface biofilms.
This work was supported by NIH grant R01 AI067703 to APM. We thank Drs. Jill Blankenship, Saranna Fanning, and Ron Stamper for their comments.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.