Mucosal biofilms: challenges and future directions

Most bacteria and fungi exist, principally, in biofilm communities in nature and, according to a recent public announcement from the US NIH, trigger more than 80% of human soft- and hard-tissue infections [101]. The involvement of mucosal tissue biofilms in human disease has been advocated for several years [1–3], but was only recently demonstrated unequivocally with the identification of extracellular matrix (ECM)-enmeshed microorganisms adhering directly on the mucosal surface of the middle ear of animals and humans with otitis media [4–5]. The combination of resistance to antimicrobials and immune attack, slower propagation rates and variable metabolic activity throughout the biofilm mass is thought to provide a selective advantage to mucosal biofilm microorganisms for long-term survival (for review, see [6,7]). Thus, it is currently hypothesized that the indolent, long-term persistence of a mucosal biofilm provides conditions conducive to recurrent infections, to periods of exacerbations and remissions in chronic infections or to sloughing or release of planktonic organisms and chronic seeding of distant locations [4,7,8].

In recent years, a significant amount of work was conducted to characterize the properties of biofilms forming on inanimate substrates, however, the structure and biological properties of living tissue biofilms have received, in comparison, very little attention. Although abiotic surface biofilms may offer some insights potentially applicable to mucosal biofilms [9–11], for most organisms it is not yet known whether the regulatory pathways that control biofilm development are applicable in both types of biofilm. Mucosal tissues provide a complex nutrient environment to the biofilm organisms, as well as opportunities for unique host receptor–microbial adhesin interactions, which probably affect both biofilm growth and the host response to infection in a manner that is different from abiotic surface biofilms. Host conditioning is the term used to describe a process by which host proteins from serum or saliva passively adsorb to the surface of a medical/dental device as a prelude to biofilm development [12]. The potential for participation of host fluids/molecules and immune and nonimmune cells in this process is even greater in tissue biofilms, owing to the more intimate association of the biofilm with the host. This is exemplified by the composition of Pseudomonas aeruginosa biofilms in cystic fibrosis patients, where the thick pulmonary tissue biofilm consists of bacteria, their alginate matrix and leukocytic cells attempting to clear the infection [13]. This complex tissue environment is also likely to provide unique cues that drive a distinct pattern of gene expression in mucosal biofilms compared with abiotic surface biofilms. In support of this notion, it has been reported that there is approximately 50% discordance in gene expression in tissue and abiotic surface Streptococcus pyogenes biofilms [14].

Although biofilms have been detected in several mucosal locations, their ability to trigger disease in humans is still a matter of active debate and investigation [3,7,8]. The presence of biofilms has been associated with the formation of inflammatory lesions in several mucosal locations in animals and humans [4,5,15], which suggests a direct role in pathogenesis. However, in the case of inflammatory bowel disease or chronic rhinosinusitis, where biofilms have been identified on the respective mucosal surfaces, the possibility exists that biofilm formation is secondary to an aberrant immune response to non-microbial antigens [16,17]. Thus, it is plausible that a chronic mucosal inflammatory response to selfantigens or allergens provides a tissue environment conducive to biofilm growth, since locally released inflammatory cytokines or excess mucus may promote biofilm growth [18]. Regardless of whether biofilms are the primary stimuli of the inflammatory response or a result of it, it is reasonable to expect that, given the appropriate mix of microorganisms and virulence factors, a long-standing biofilm will ultimately contribute to tissue damage. However, neither the mechanisms by which soft-tissue biofilms affect the development of inflammatory events (orvice versa), nor the disease-specific changes in biofilm communities have been determined in most mucosal biofilm infections.

The role of mucosal biofilms in human disease may be defined by one of two host inflammatory response-centered pathogenesis models. In the first model, it is proposed that mucosal biofilms may trigger an excessive proinflammatory response leading to indirect, inflammation-mediated host damage.P.aeruginosa biofilm-triggered pneumonia in cystic fibrosis patients primarily falls under this pathogenesis model [13]. In the second model, biofilm-specific constituents (e.g., ECM molecules) may attenuate the mucosal inflammatory response to infection and dysregulate phagocytic cell function, providing an early growth advantage to biofilm organisms. The relative preponderance of one of the two pathogenetic schemes may depend on mucosal site characteristics, such as the type of epithelium lining the mucosal surface (e.g., stratified vs columnar or keratinized vs nonkeratinized), or even the stage of the infection. In fact, there is evidence to suggest that, during the early stages of biofilm-triggered otitis media, the infection follows the second pathogenesis model as microbial phosphorylcholine components decrease early inflammation, which promotes the establishment of a stable biofilm [19]. By contrast, during the later stages, a chronic exudative inflammatory response to the biofilm microorganisms ensues, which exemplifies the first pathogenesis model [5]. Other factors that may modify mucosal biofilm pathogenesis are the presence and composition of mucus, fluid flow and other site-specific epithelial innate defense mechanisms, such as the ability of the epithelium to synthesize antimicrobial molecules. It is evident that, in order to begin to understand the role of mucosal biofilms in human disease, the host response to these biofilms must be dissected. With the exception of the skin [8,20,21], the host response to biofilms forming on nonkeratinizing epithelial tissues, such as the urinary tract and lower GI tract, is largely unknown.

In addition, virulence characteristics of biofilm organisms may determine which pathogenetic mechanisms will prevail in a certain mucosal site. For example, the ability of a biofilm organism to secrete tissue-damaging hydrolytic enzymes may favor a direct, rather than indirect (i.e., chronic inflammation-mediated) role of biofilm in disease. Since it is possible that the contribution of specific virulence factors to mucosal biofilm-triggered disease differs depending on the stage of tissue destruction or location in the lesion (both of which may affect the availability of nutrients for microbial communities), biofilm stage- and tissue location-dependent analysis of virulence is necessary to understand mucosal biofilm pathogenesis fully. Perhaps, more importantly, the biologic activities of ECM constituents need to be dissected, since they may adversely affect immune and nonimmune cell recognition of microbial cells and antimicrobial host responses and, thus, modify the sequence of events in pathogenesis.

The composition of mucosal biofilm ECM is inherently more complex than abiotic surface biofilms, owing to the fact that host proteins, mucopolysaccharides, nucleic acids and even whole cells may participate in its development [3,21,22]. Perhaps the biggest challenge in recognizing tissue microbial communities as biofilms is the demonstration of ECM presence, since it is easily destroyed by standard histologic processing. Thus, specialized fixation techniques or the use of fresh tissues for ECM analysis are frequently required [5,16]. Lectins, such as concanavalin A (with mannopyranosyl and glucopyranosyl specificity) or wheat germ agglutinin (with sialic acid and N-acetylglucosamine specificity), with affinities to carbohydrate targets that resemble those for antigen–antibody interactions, have been traditionally used to identify the ECM components in abiotic surface biofilms [23,24]. However, since certain components of the ECM can also be found on microbial capsules [25]or cell walls [26], the single use of lectins for ECM labeling,cannot distinguish between cell surface-associated and secreted ECM material. To distinguish between extracellular and cell surface-associated ECM glycoproteins, double-labeling methods, combining lectins with DNA stains (4´-6-diamidino-2-phenylindole or propidium iodide-based) or metabolic stains (e.g., FUN-1), have been traditionally used. However, both metabolic and nuclear stains can be taken up by host cells, which leads to elevated background staining and complicates visualization of tissue microorganisms. In experimental mucosal infections, the use of microorganisms expressing fluorescent proteins under the control of a constitutive promoter is one way to circumvent this problem. Another complicating factor in mucosal tissue biofilm ECM labeling is the fact that host cell-surface glycoproteins may bind to lectins with high affinity. In this case, pathogen-specific anticarbohydrate antibodies may be needed to demonstrate the presence of ECM of microbial origin, but these are not usually readily available.

Mucosal biofilm-triggered infections also present unique therapeutic challenges. For example, persistent oroesophageal thrush, clinically refractory to most antifungals, has been described [27,28], which may be attributed to the fact that certain biofilm-forming cells in these lesions are surviving persisters from a multidrug-tolerant population, a phenomenon unique inCandida spp. biofilms but not planktonic cells [29]. In Staphylococcus aureus dermatitis, where microcolonies avidly attach to surface keratinocytes and expand on them, producing a protective glycocalyx, traditional antimicrobials cannot eradicate bacteria [30]. It has also been suggested that no antimicrobial will be completely effective in the absence of biofilm invasion by neutrophils [8]. Thus, biologic response modifiers that locally augment innate immune and nonimmune cell antimicrobial activities may find appropriate applications in mucosal biofilm infections.

Despite multiple challenges in studying mucosal biofilms, it is critical to continue to recognize many epithelial tissue infections as biofilm infections. Examples of such infections are chronic or recurrent tonsilitis, rhinitis, urethritis, cystitis, otitis and dermatitis, of certain bacterial or fungal origin [5,8,10,11,17,20]. It is also important to experimentally address the unique aspects of biofilm pathogenesis within the epithelial tissue environment. Experimental modeling of these infections is inherently more difficult because it requires either animal models of infection or tissue-engineering approaches. The host is likely to play a more central role in the pathogenesis of these infections compared with abiotic surface biofilms. Finally, the complexity of host–biofilm interactions at mucosal surfaces will probably surpass the complexity of any infection at other sites.

Financial & competing interests disclosure

This author’s work on oral mucosal biofilms is supported by the NIH/National Institute of Dental and Craniofacial Research grant RO1 DE13986. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.