International Wound Infection Institute Project
The significance of Biofilms in Wound Infections
Three sub-committees for the International Wound Infection Institute (IWII) have been established with a focus on Education, Evidence and Research. One of the areas that the Research sub-committee is investigating is the possibility of providing a new biofilm testing protocol to improve detection and subsequent treatment of wounds colonised with biofilms.
Thanks to Val Edwards-Jones, Greg Schultz and Jude Douglass of the International Wound Infection Institute for the development of this paper
Non-healing wounds and biofilms
The widely-accepted definition of an infected wound, as being one which contains >105 colony forming units per gram of tissue (CFU/g), is a useful rule of thumb, but is not wholly accurate. This was established in studies on burns and acute wounds, which were usually infected with a single microorganism (for example: [1]), and is the level at which most microorganisms produce virulence factors and begin the infectious process. However, many chronic wounds are polymicrobial and can have more or fewer than 105 organisms without displaying any obvious clinical signs of infection, yet they still fail to heal. These clinical observations led to the adoption of the term “critical colonisation” to describe wounds that exist in a state between normal colonisation and frank infection. It is now becoming clear that this term is in fact a euphemism for colonisation with a polymicrobial community or a biofilm, rather than a true description of a condition lying somewhere on a linear spectrum between colonisation and infection.[2] The term “critical colonisation” was in fact only adopted as a functional term to try to describe a situation that was not fully understood at the time.
Wounds that fail to heal may be impaired due to the presence of a biofilm which slows down the wound-healing process without necessarily inducing any signs of invasive growth.[2]
Research over the last 20 years has suggested that chronic wound infections fail to heal because of the presence of biofilms [3-7] but there has been little direct evidence to support this theory. Biofilm bacteria exist as structured polymicrobial communities and are not distributed evenly in the wound. Therefore standard biopsy or swab techniques may not accurately sample the bacterial community in the wound, making it difficult to interpret laboratory results. Even if biofilm colonies are captured by the sampling technique, ‘negative’ results may be obtained as traditional laboratory culturing techniques cannot determine which of the many microorganisms are responsible for the condition of the wound (a biofilm can be formed by a single species of bacteria, such as Staphylocossus epidermis on catheters, but when in chronic wounds, biofilms are typically formed of multiple species.)
If there are no obvious ‘typical’ wound pathogens, a report of ‘no significant growth’ or ‘no pathogens isolated’ may be issued by the laboratory.
A recently published study by James et al (2008) now provides some of the missing evidence for the presence of biofilm in chronic wounds. Using light and scanning electron microscopy, they found that about 60% of chronic wounds were colonised with biofilm.[8] Fifty chronic wounds of varying aetiology and of at least 30 days’ duration, and 16 acute wounds were investigated by microscopy, while 27 chronic wounds were also subject to population analysis using molecular techniques. Microscopic analysis revealed the presence of densely aggregated colonies of bacteria, often surrounded by extracellular matrix. This was observed in only one acute wound while 30 of the chronic wounds had clear biofilm formation (p<0.001). Population studies using molecular techniques revealed diverse polymicrobial communities and in some specimens, strict anaerobes that were not revealed by standard culture techniques.
Of course, an alternative interpretation of the results is that the material observed in microscopy may not have been a true biofilm but may simply be slough. However, it is worth asking what is the nature of “slough”? It consists predominantly of fibrin and necrotic tissue within which is embedded numerous bacteria. This accumulates on the surface of the chronic wound and is rarely seen in an acute wound. If the biofilm were not present it is likely there would also not be slough.
Further evidence for the presence of biofilm comes from Bjarnsholt et al (2008) who analysed sections from chronic wounds using fluorescent in situ hybridisation (FISH), and found distinct microcolonies, the basic structures of biofilms.[2]
The development of biofilms
Bacterial biofilms are structured communities of bacterial cells enclosed in a self-produced polymeric matrix adhering to inert or living surfaces.[9] The cells attached to the surface are called ‘sessile’ and those free living at the surface of the biofilms are called planktonic bacteria. Biofilms in natural environments are resistant to bacteriophages (bacterial viruses), amoebae and to the powerful chemical biocides used in the industrial environment. In man, sessile bacterial cells can resist innate and acquired host immune responses and are up to 1000 times more resistant to antibiotics than planktonic cells. Whether this is due to a property of the sessile cells or to a protective effect from the other bacteria in the community is dependent upon the nature of the biofilm.
It is important to understand the mechanisms by which biofilms form as these may provide opportunities for the development of medication that could prevent or disrupt their formation. It is generally accepted that there are a number of phases in the development of a biofilm: initial attachment and formation of microcolonies; maturation of attached bacteria into a differentiated biofilm; and dispersal of planktonic bacteria from mature biofilm.
Attachment and formation of microcolonies. When a bacterium encounters a suitable surface it explores the surface in a process called “twitching”.[10, 11] Specific cell surface components are required for adhesion to a surface and the very process of attachment initiates synthesis of extracellular matrix.[12] This was also demonstrated by Davies et al (1995) who showed that, within minutes of attachment of P aeruginosa to a solid surface, genes that synthesise extracellular polysaccharide (alginate) were up-regulated.[13]
During the first hour of attachment, over 800 new proteins can be expressed,[14] radically changing the phenotype of the bacterium.[15] The attached cells continue to grow and divide while recruiting additional planktonic cells to the colony. At a certain stage of the biofilm formation, the microcolonies differentiate into mature biofilms with water-filled channels through which nutrients can flow in and waste products can be removed. This is an essential stage for the survival of the biofilm which continues to mature.
Maturation and differentiation. Research on the behaviour of bacteria when in groups, has shown that individual bacterial cells communicate via cell-to-cell signalling systems. In Gram-negative bacteria, specific genes direct the production of extracellular acylhomoserine lactone molecules (acyl-HLSs), that function in a similar way to hormones, diffusing out of the cell and into other bacterial cells. The acyl-HLSs accumulate in proportion to the total number of cells present and therefore provide an index of the density of the population, in effect, a quorum sensor. Gram-positive bacteria use small peptides, around 5 to 17 amino acids in length, as signalling molecules.[16]
When these “quorum sensing” signals accumulate to a certain critical level, genes are activated that trigger differentiation of the biofilm into loosely-packed pillar and mushroom like colonies, with the bacterial cells embedded in thick polymeric walls resistant to biocides.[17] The polymeric wall may contain polysaccharides, alginate, extracellular DNA and components such as proteins and lipids.[2] Living, fully hydrated biofilms are composed of around 15% by volume of cells and 85% of matrix material.[3]
Rhoads suggests that the cells at the base of the biofilm are “rooted” to the surface, and that the biofilm will continue to regenerate so long as the root is still attached to the source of nutrients.[18] Modern methods (FISH) have demonstrated the architecture and special organisation of the relative species in a multi-soecies biofilm. In addition, expression studies have shown that there is a variation of activity between sessile and planktonic forms of the same species [19,20]
Seeding of planktonic bacteria. A mechanism of dispersion is required if bacteria within a biofilm are to colonise new areas. Pieces of biofilm may be physically detached and carried to new surfaces and it has also been suggested that sessile bacteria may produce an enzyme that digests the alginate matrix allowing dispersion of individual bacteria.[21]
The impact of biofilms on wounds
Biofilm formation confers great advantage to the bacteria: the exopolysaccharide shield provides a physical barrier; bacteria within the biofilm can reduce their metabolic activity and thereby increase their tolerance to antibiotics; and sub-communities of facultative anaerobes or strict anaerobes can develop within niches of the biofilm community where nutrients and oxygen may be limited. It has also been suggested that biofilms modulate their virulence in order to remain attached to their source of nutrients.[22, 23]
Because of this, biofilms often remain in stasis and infections can be slow to produce overt symptoms, if at all. Biofilms do release antigens that stimulate the production of antibodies, but these are incapable of killing the protected sessile bacteria and instead cause damage to surrounding tissues.[2] Antibiotics can resolve the symptoms caused by planktonic cells but have little or no effect on the biofilm itself.[2, 24]
Bacteria in biofilms are protected from systemic and topical agents, from endogenous antibodies and phagocytic cells[25, 26], and thus are relatively free to cause extensive tissue damage and delay healing. Biofilms are highly inflammatory, constantly shedding bacteria onto the surface of the wound, exciting an immunological response which causes tissue damage and maintains chronic inflammation. Biofilms therefore appear to “recur” despite repeated attempts at antibiotic therapy.
As a result of wound biofilms, keratinocytes are unable to migrate across the wound bed, there is low oxygen at the wound surface, and host defences are unable to penetrate or remove the biofilm. Bjarnsholt et al (2008) suggest that this is due to the fact that biofilms disable host defences. In studies on biofilm microcolonies of P aeruginosa the quorum sensing communication system was shown to regulate production of a rhamnolipid that can eliminate polymorphonuclear leukocytes (PMNs).[27, 28] This effectively builds a “PMN shield” around the colony that suppresses the cellular defence system. Killing and lysis of PMNs result in the release of large quantities of cytotoxic enzymes, free oxygen radicals and inflammatory mediators that damage host tissue and maintain the chronicity of the wound, without affecting the biofilm.
Treatment of biofilms
Effect of antibiotics on biofilms
Microorganisms in a biofilm have an inherent lack of susceptibility to antibiotics, compared to planktonic cultures of the same organism.[29] In 1999, Ceri formed biofilms from NCCLS reference strains of E coli, P aeruginosa and S aureus and tested the minimal inhibitory concentration (MIC) of a range of antibiotics indicated for each type of bacteria.[30] While the MIC values for the planktonic form of each strain were as expected, the minimal biofilm eradication concentrations (MBECs) were generally 100 to 1000 times higher. Hoyle et al (1992) found that planktonic bacterial cells were 15 times more susceptible to tobramycin than cells in intact biofilms.[31] These findings have been confirmed by many other studies; some MBECs have even been found to exceed the maximum allowable prescription level.[27, 30-33]
Mathematical modelling of the processes of diffusion suggests that exopolysaccharide matrices can be highly effective barriers if the antimicrobial agent can be deactivated faster than it can diffuse through the matrix. The negatively charged polymeric substances that make up the walls of a biofilm are known to retard the diffusion of antibiotics. Suci et al (1994) showed that there was delayed penetration of ciprofloxacin into P aeruginosa biofilms: the normal penetration time of 40 seconds was increased to 21 minutes on a biofilm-containing surface.[36] A 2% suspension of alginate isolated from P aeruginosa was able to inhibit diffusion of gentamicin and tobramycin,[33] and Allison and Matthews (1992) observed reduced diffusion and antimicrobial activity of tobramycin and to a lesser extent of ciprofloxacin through P aeruginosa biofilm.[38] Using four different antibiotic groups, Koseoglu et al (2006) found that development of E coli biofilms on urethral catheters could only be delayed, not prevented.[35]
The polymeric barrier is also effective against reactive species, such as hypochlorite and hydrogen peroxide which are produced by phagocytic cells. Alginate is a scavenger of free oxygen radicals, prevents phagocytosis, and can bind cationic antibiotics such as the aminoglycosides.[36, 37,38, 39]
It has also been shown that biofilm bacteria, like planktonic bacteria, contain efflux pumps that are able to pump out antimicrobial agents. However, using strains of P aeruginosa with either defective, or over-productive, efflux pumps, Brooun et al (2000) were not able to detect any pumping mechanism and suggested that this resistance mechanism may be masked by other mechanisms that are also operating in the biofilm.[25]
Another possibility is that at least some of the cells in a biofilm exist in a slow-growing or semi-starved state.[29] Evans et al (1990) found that the slowest growing E coli cells in biofilms were the most resistant to cetrimide.[40] As many antibiotics work by targeting the metabolic processes of bacteria, cells in a state of reduced metabolism will not be highly susceptible to antimicrobial agents, and a proportion will therefore always be more likely to survive an attack.
A third proposed mechanism is that some of the bacteria in a biofilm develop into a distinct phenotype which is inherently more resistant to traditional antibiotics. Dagostino et al (1991) suggested that the initial adhesion to a surface might result in certain genetic changes which could affect the physiological responses of the bacteria.[41]
The fact that free-floating planktonic bacteria can be susceptible to standard doses of antibiotics can cause confusion and difficulty in managing wounds that are colonised with biofilms. Planktonic bacteria are shed from the biofilm and can be eliminated with antibiotics, thereby providing symptomatic relief. However, because the biofilm is not eliminated, once the antibiotic is stopped, it appears to the clinician that re-infection has occurred.
Antimicrobials and biofilms
Percival et al (2008) demonstrated that a silver hydrogel dressing (ionic silver) was able to kill 90% of all sessile bacteria within three common biofilms in 24 hours, and total bacterial kill was achieved within 48 hours. The biofilms studied were P aeruginosa, Enterobacter cloacae, S aureus and mixed communities of more than one type of bacterium.[42] In an earlier paper, the authors demonstrated that sessile bacteria were far less susceptible than planktonic bacteria to silver dressings.[43] Using an in vitro model they found that a silver-containing hydrofibre dressing was more effective against strains of P aeruginosa, Candida albicans and S aureus; while a nanocrystalline silver dressing was more effective against strains of Klebsiella pneumoniae, Enterococcus faecalis and E coli.
However, these biofilms were immature, only 24 hours old, and further research has shown that silver hydrogel dressings are less effective on three-day old, mature biofilms.(unpublished thesis, Q Yang, Institute for Wound Research, University of Florida) This and other research demonstrates that immature biofilms behave differently to mature biofilms in terms of susceptibility to antimicrobial agents and dressings.
However, it does appear that mature biofilms can be fully killed by iodine releasing dressings.(as above, Q Yang) Akiyama et al (2006) were able to clear S aureus biofilm in vitro and in vivo using a mouse model.[44] They suggest that the cadexomer iodine beads adsorb biofilm-protected S aureus cells and that the protective glycocalyx collapses during dehydration. This is followed by liberation of iodine which is capable of directly killing the biofilm S aureus cells.
New approaches
Clearly, there are many points in the process of biofilm development which are potential therapeutic targets: disruption of the cell surface components that are required for adhesion; disruption of the quorum sensing process; and stimulation of the enzymes that digest alginate.
A number of agents have been developed in industrial and dental sectors to counter biofilm development. In the food industry, bovine lactoferrin is used as an anti-biofilm agent against P aeruginosa. This substance prevents the bacteria from attaching to the target surface, rendering them unable to produce biofilm structures. [10, 45]
A number of studies have proven the validity of the concept of quorum-sensing inhibitors (QSIs) as antimicrobial treatment. Differentiation and maturation of the biofilm can be disrupted if the cell-to-cell signalling process is blocked.[25] In P aeruginosa biofilms it has been shown that a synthetic halogenated furanone compound is capable of interfering with acyl-HLS signalling.[46] The compound can penetrate microcolonies and block cell signalling and quorum sensing. It does not affect initial attachment to the substrate but it does affect the architecture of the biofilm, enhancing the process of detachment. Application of the drug to P aeruginosa biofilms in mice inhibited quorum sensing of bacteria and promoted their clearance by the immune response of the mouse.[47]
Other quorum sensing inhibitors have been identified such as RNA III inhibitory peptide which inhibits the agr system of Gram-positive bacteria.[48-52]
Debridement of biofilms
Until some of these newer methods reach the stage of commercialisation, the only sure method of treating biofilm is physical removal through thorough debridement, followed by the application of effective antimicrobials, antibiotics or antiseptics to prevent re-formation. Disrupting the community through debridement will not directly kill the bacteria, but the disturbance does allow other concomitant therapies and natural host defences to work more effectively.[53] The aim of a debride-and-cover strategy is therefore to remove the nidus of infection that perpetuates a state of chronic inflammation and to give the host defences the best possible change to recover.[54]