5. Can bacteria become resistant to Triclosan?
- 5.1 What are the mechanisms of resistance to triclosan?
- 5.2 Could exposure to triclosan lead to cross-resistance to antibiotics?
- 5.3 Has bacterial resistance to triclosan been observed in the environment?
- 5.4 How can bacterial resistance be determined?
5.1 What are the mechanisms of resistance to triclosan?
The SCCS opinion states:
Unlike antibiotic resistance, the issues relating to biocide resistance in the healthcare environment are considered to have a very low profile and priority (Cookson 2005). Despite the widespread use of disinfectants and antiseptics in healthcare settings, acquired resistance to biocides in bacteria isolated from clinical specimens or the environment is not routinely characterised. Emerging bacterial resistance to biocides has been well described in vitro, but evidence in practice is still lacking (Russell 2002b, Cookson 2005, Maillard and Denyer 2009).
It is widely accepted that biocides have multiple target sites against bacteria (Denyer and Maillard 2002, Lambert 2002, Maillard 2002, Maillard 2007, Poole 2004, Stickler 2004, Gilbert and Moore 2005, Maillard 2005b)with their efficacy depending on a range of intrinsic and extrinsic factors, (Reuter 1984, 1989, 1994, EFSA 2008, SCENIHR 2009). Thus, the emergence of general bacterial resistance is likely to arise from a mechanism/process causing the decrease of the intracellular concentration of a biocide, under the threshold that is harmful to the bacterium (Tattawasart et al. 2000a, Tattawasart et al. 2000b; Braoudaki and Hilton 2005; Maillard 2005a, Maillard and Denyer 2009). Several mechanisms based on this principle (mode of action) have been described including change in cell envelope, change in permeability, efflux and degradation (Table 5). Bacteria in biofilms are also less susceptible to biocides because of a number of factors. It is likely that some of these mechanisms operate synergistically although very few studies investigating multiple bacterial mechanisms of resistance following exposure to a biocide have been performed.
Bacterial resistance to biocides is not a new phenomenon and it has been reported since the 1950’s (Adair et al. 1971; Russell 2002b; Chapman 2003). To date, bacterial resistance has been described for all the biocides that have been investigated. Resistance often occurs following an improper usage of the formulated biocide, leading notably to a decrease in active concentration (Sanford 1970, Prince and Ayliffe 1972, Russell 2002b).
It is worth noting that some mechanisms (e.g. efflux, target protection, degradation) can be horizontally transferred to other bacteria (Poole 2002a, Quinn et al. 2006, Roberts and Mullany 2009, Yazdankhah et al. 2006; Hawkey and Jones 2009, Juhas et al. 2009). In addition, Pearce et al. (1999) showed that some biocides, at a sub-lethal concentration, may increase or decrease the frequency of gene transfer by conjugation and transduction.
6.2. General considerations on the study of triclosan
Triclosan is described as a broad spectrum biocide. However, some bacteria are intrinsically resistant to triclosan, notably P. aeruginosa (Lear et al. 2002) and triclosan is not active against bacterial endospores. This is likely due to the structure of the Gram-negative bacteria and particularly the outer membrane, preventing triclosan to penetrate through the bacterium to reach its target sites.
Bacterial resistance mechanisms to triclosan have been widely studied. However, most studies have considered resistance as an increase in MIC and not necessarily as an increase in MBC. Using MICs to measure bacterial resistance is arguable, since much higher concentrations of biocides have usually been used in practice and, therefore, failing to achieve lethality because of elevated MICs is unlikely. Some studies have shown that bacterial strains showing a significant increase in MICs to some biocides, such as cationics, were nevertheless susceptible to higher (in use) concentrations of the same biocide (Thomas et al. 2005) or triclosan (Lear et al. 2006). MRSA showing a 40-fold increase in MIC to triclosan remained susceptible to 1 mg/L (Suller and Russell 1999). Concentration is central to the definition of resistance in practice (Maillard and Denyer 2009). Hence, bacterial resistance based on the determination of MIC does not reflect accurately the true resistance profile of biocides, including triclosan.
Concentration is one the most important factors that will affect the activity and efficacy of a biocide (Russell and McDonnell 2000, Maillard 2005a, b 2007). Biocides with a high concentration exponent (Russell and McDonnell 2000) such as triclosan are particularly affected by dilution since a small decrease in concentration will profoundly affect efficacy. Hence, it might not be surprising that products with a low concentration of a phenolic biocide or other biocides with a high concentration exponent (e.g. alcohols) are less effective and might be prone to bacterial contamination and growth.
Most laboratory studies have been performed with triclosan dissolved in a solvent such as DMSO, and in some cases alcohol, and did not investigate commercially available formulations. Differences between laboratory (in vitro) investigations and situations in practice have not been addressed to date (Maillard and Denyer 2009). Hence, emerging bacterial resistance to triclosan investigated in vitro conditions might not necessarily reflect such development of resistance in situ. Components of the formulations might have a potentiation effect (or not) on the activity of triclosan, and their role on emerging bacterial resistance to triclosan has not been studied.
Bacterial resistance against triclosan involves both intrinsic and acquired mechanisms (Yazdankhah et al. 2006), and include: mechanical barrier (altering intracellular concentration), change in target site (mutation of the target site) (Heath et al. 1998), efflux, and by-pass of metabolic pathway (Webber et al. 2008b). These mechanisms have been also described to confer antibiotic resistance (Davin-Regli et al. 2008).
Change in enoyl acyl carrier reductase
At sub-lethal concentrations, triclosan has been shown to affect specific bacterial targets. Triclosan interacts specifically with an enoyl-acyl reductase carrier protein (ENR) at a low concentration (Heath et al. 1999; Levy et al. 1999, Roujeinikova et al. 1999, Stewart et al. 1999). Triclosan was found to inhibit fatty acid synthesis by targeting FabI in E. coli (Heath et al. 1998) and S. aureus (Heath et al. 2000), and InhA in M. smegmatis (McMurry et al. 1999) and M. tuberculosis (Parikh et al. 2000).
Triclosan resistant mutations in fabI decrease the interaction of triclosan with the ENR-NAD+ binding. Mutation in fabI in E. coli was shown to confer a 60-fold decreased susceptibility to triclosan (Heath et al. 1998). Mutation in fabI has led to an increase in triclosan MIC in a number of bacterial genera (McMurry et al. 1998a, Parikh et al. 2000, Health et al. 2000, Slater-Radosti et al. 2001, Massengo-Tiassé and Cronan 2008, Webber et al. 2008b). In Acinetobacter baumannii high-level triclosan resistance could be explained by a Gly95Ser mutation of FabI, whilst wild-type fabI was observed to be overexpressed in low-level resistant isolates (Chen et al. 2009). Likewise in Ps. aeruginosa, high-level resistance to triclosan has been shown to be associated with FabV (Zhu et al. 2010).
McMurry et al. (1998b) postulated that mutations at mar and sox in E. coli only conferred a 2-fold increase in resistance presumably by a modest overexpression of AcrAB. This expression is unlikely to decrease the efficacy of triclosan. However such a mutation, together with mutations at other loci such as fabI (increasing resistance to 90-140-fold) could be more significant
Efflux of antimicrobials
Triclosan is a substrate of AcrAB efflux pump in E. coli, of MexAB-OprM and MexCD-OprJ, MexEF-OprN, MexJK-OprH multidrug efflux pumps in P. aeruginosa, of AcrB in S. enterica serovar Typhimurium, and CmeB in Campylobacter spp. (Piddock 1996; McMurry et al. 1998; Chuanchuen et al. 2001, 2002, 2003; Schweizer 1998). These efflux pumps are similar to other efflux pumps in other Gram-negative pathogens (Piddock 2006) and as such, it is likely that triclosan is a substrate of such pumps in other Gram-negative bacteria.
In S. enterica serovar Typhimurium, active efflux via AcrAB-TolC conferred decreased susceptibility to triclosan. The triclosan resistant mutants (MIC ≥32 mg/L) did not lose any fitness when compared to wild-type strains (Webber et al. 2008a). The pump efflux system of P. aeruginosa has been shown to confer a high level of intrinsic triclosan resistance (Mima et al. 2007). In addition, mutants of E. coli, and S. enterica which overexpress the AcrAB– TolC efflux system, have decreased susceptibility to various agents, including triclosan, demonstrating that triclosan is a substrate for efflux pumps (Webber et al. 2008a).
As previously reported for antibiotics, the presence of active efflux pumps is required for the acquisition of target mutations, which in turn increase the level of resistance (Webber et al. 2008b). In Acinetobacter baumannii, although active efflux did not appear to be a major reason for triclosan resistance, the acquisition of resistance appeared to be dependent on a background of intrinsic triclosan efflux (Chen et al. 2009).
By-pass of metabolic blockage
The proteomic analysis of S. enterica serovar Typhimurium triclosan-resistant mutants showed a set of proteins with commonly altered expression in all resistant strains. This “triclosan resistance network” included 9 proteins involved in production of pyruvate or fatty acid and represents a mechanism to increase fatty acid synthesis by an alternative pathway (Webber et al. 2008b). In addition to the expression of this “network”, these mutants showed specific patterns of protein expression leading to the conclusion that triclosan resistance was multifactorial and potentially involved a number of mechanisms acting synergistically to attain high-level resistance (≥32 mg/L) (Webber et al. 2008b). In S. aureus, a modification of the membrane lipid composition associated with the alteration of the expression of various genes involved in the fatty acid metabolism were observed in triclosan resistant strains (Tkachenko et al. 2007).
Seaman et al. (2007) studied the appearance of small colony variants in MRSA following exposure to triclosan in vitro. The small colony variants displayed reduced susceptibility (23-60 fold; 1.5-4 mg/L from 0.063 mg/L) to triclosan and resistance to penicillin and gentamicin. Bayston et al. (2009) noted that prolonged exposure (i.e. 72 h) to triclosan-impregnated silicone resulted in the induction of small colony variants and a 67-fold increase in triclosan MIC.
Recent evidence highlighted that bacterial swarming motility might confer some resistance to triclosan (5 mg/mL in B. subtilis and 0.1 mg/mL in E.coli) when compared to no swarming bacteria. The mechanism(s) by which swarming might confer some resistance is unknown, but is unlikely to be caused by efflux (Lai et al. 2009).
Involvement of multiple mechanisms
At bactericidal concentrations, triclosan seems to act against multiple and various targets, causing disruption of the bacterial control of cell wall permeability (Villalain et al. 2001; Guillén et al. 2004). One study in particular, investigated the role of both the permeability barrier and efflux in increase resistance to triclosan in E. coli. The MIC of triclosan-resistant E. coli mutants (MIC >1000 mg/L) was reduced to 10-25 mg/L when treated with both ethylene diamine tetra-acetic acid (EDTA; a chelating agent enhancing outer membrane permeability) and carbonyl cyanide m-chlorophenylhydrazone (CCCP; a proton motive force uncoupler), as compared to a MIC of 0.1 mg/L in sensitive E. coli strain, indicating that potentially both permeability and efflux worked together to provide the high level resistance to triclosan. However, neither CCCP nor EDTA reduced the susceptibility of P. aeruginosa to triclosan (Gomez Escalada 2003). In Acinetobacter baumannii, triclosan-resistant isolates were characterized by antibiotic susceptibility, clonal relatedness, fabI mutation, fabI expression, and efflux pump expression (Chen et al. 2009). Yu et al. (2010) described a multiple mechanism response in E. coli following exposure to triclosan. The involvement of a number of mechanisms was shown to confer triclosan resistance up to 80 mg/L.
Generally, bacteria are attached to surfaces and associated in a community (termed biofilm) and are rarely present as single cells (planktonic). Bacterial biofilms have been shown to be highly resistant to antimicrobials compared to planktonic cultures. A biofilm-associated phenotype has been described (Brown and Gilbert 1993, Ashby et al. 1994, Das et al. 1998; Gilbert et al. 2003). The mechanisms of resistance involved in a bacterial biofilm include decreased metabolism, quiescence, reduced penetration due to the extracellular polymeric matrix (Pan et al. 2006), enzymatic inactivation of biocides (Sondossi et al. 1985) Giwercman et al. 1991, Huang et al. 1995), and the induction of multi-drug resistant operons and efflux pumps (Maira-Litran et al. 2000). Bacterial biofilm resistance to triclosan has been poorly studied.
One study reported that the tolerance to triclosan of Salmonella in biofilm was attributed to low diffusion through the extracellular matrix, while changes of gene expression might provide further resistance both to triclosan and to other antimicrobials (Tabak et al. 2007). McBain et al. (2003) investigated the fate of a complex bacterial biofilm exposed to sub- lethal concentrations of triclosan (2–4 g/L) over a 3 month period. The authors identified a change in the composition of the biofilm and an increase in resistance of the complex population as measured by MIC. Interestingly, the composition of the biofilm changed, with a decrease of species diversity. The triclosan tolerant species such as Pseudomonads and Stenotrophomonads were still present, but other triclosan tolerant species (Achromobacter xylosoxidans) demonstrated a clonal expansion. Most importantly, the authors noted that the antibiotic susceptibility profile was not affected.
A study investigating the effect of triclosan in the development of bacterial biofilms on urinary catheters highlighted the selectivity of triclosan. While triclosan inhibited P. mirabilis, it had little effect on other common bacterial pathogens (Jones et al. 2006). In addition, the control of P. mirabilis by triclosan resulted in emerging triclosan-resistant strains in vitro. While most of these strains were still susceptible to the triclosan concentration used in the urinary catheter, one strain (MIC = 40 mg/L) was not (Stickler and Jones 2008). Smith and Hunter (2008) showed that recommended concentrations of three biocidal products used in healthcare (one containing benzalkonium chloride 10 g/L, one containing chlorhexidine gluconate 40 g/L and one containing triclosan 10 g/L), were ineffective in eliminating hospital-acquired MRSA or P. aeruginosa biofilms, highlighting differences in susceptibility between planktonic and biofilm bacteria.
It is however interesting to note that Tabak et al. (2009) observed a synergistic action of sequential treatment of triclosan (500 mg/L) followed by ciprofloxacin (500 mg/L) against biofilm of S. enterica serovar Typhimurium. There is little information in the literature about the potentiation of activity between a biocide and an antibiotic and such a study is important and describes an interesting application/effect of triclosan.
6.4. Mutation rates and transfer of resistance
The development of bacterial resistance through acquired mechanisms such as mutation and the acquisition of resistant determinants are of concern since a bacterium that was previously susceptible can become insusceptible to a compound or a group of compounds (Russell 2002a). In S. enterica serovar Typhimurium, mutation frequency following exposure to triclosan was low (5 x 10-9), lower than mutation frequency observed following antibiotic exposure (Birošová and Mikulášová 2009).
Cookson et al. (1991) isolated MRSA strains exhibiting triclosan resistance (2-4 mg/L) from patients using mupirocin and triclosan baths. Although in this study the resistance was shown to be transferable in association with the plasmid encoding for mupirocin resistance, this could not be confirmed subsequently by other studies. The transfer of a plasmid encoding for mupirocin resistance to a triclosan sensitive S. aureus strain failed to increase resistance to triclosan (Suller and Russell 2000). Other studies investigating clinical S. aureus isolates resistant to mupirocin also failed to observe this association (Bamber and Neal 1999). Although various genetic mobile elements have been described to be involved in the dissemination of cross-resistance towards biocides-antibiotics (Roberts and Mullany 2009, Schlüter et al. 2007) no specific genetic mobile element has been associated with triclosan resistance.
6.5. Induction of resistance
There are two types of induction. The first corresponds to the trigger of genes governing the genetic cascade (global regulation) which promotes the expression of efflux pumps and/or down regulates membrane permeability (porin synthesis). The second corresponds to the direct activation of the promoter region (local regulation) for example controlling efflux genes (Davin-Regli et al. 2008).
The induction of bacterial resistance mechanisms following exposure to a low concentration of a biocide has been reported in a number of studies for a number of biocides (SCENIHR 2009). In some occasions, a specific mechanism has not been established and a phenotypic change leading to the emergence of resistance to several unrelated compounds in vitro has been reported following exposure to a low concentration of a biocide (Moken et al. 1997).
It is possible that triclosan induces a stress response followed by, or in addition to, the expression of mechanisms that reduce the deleterious effect of the biocide (McMurry et al. 1998b; Gilbert et al. 2002). A decrease in growth rates in E. coli and P. aeruginosa has been described following exposure to sub-lethal concentrations of triclosan, which indicates the generation of a stress to the organism (Gomez Escalada et al. 2005).
Triclosan induces bacterial resistance through the over-expression of efflux pumps via activation of mar and ram (Randall et al. 2007; Webber et al. 2008a; Bailey et al. 2009), over-expression and mutagenesis of fab1, expression of regulatory genes involved in the control of antibiotic resistance cascades (activator of drug efflux, decrease of membrane permeability) and fatty acid metabolism in a number of bacterial genera (Jang et al. 2008, Webber et al. 2008b, Bailey et al. 2009). These genes are involved in resistance to triclosan, but also in possible cross-resistance and multi-resistance to different antibiotic and biocide classes. In Stenotrophomonas maltophilia, the overexpression of an efflux pump (SmeDEF), involved in antibiotic resistance, was demonstrated in several triclosan-selected mutants (Sánchez et al. 2005). In E. coli, overexpression of acrAB or marA or soxS (positive regulator of acrAB) decreased susceptibility to triclosan 2-fold. Deletion of the acrAB locus increased susceptibility to triclosan approximately 10-fold. It was observed that clinical isolates overexpressing acrAB showed enhanced resistance to triclosan. A clinical strain overexpressing marA had a triclosan MIC of 0.27mg/L as compared to susceptible strain with an MIC of 0.090 mg/L. In S enterica serovar Typhimurium overexpressing AcrAB and C. jejuni overexpressing CmeB, triclosan MIC increased to 32 mg/L (Pumbwe et al. 2005; Buckley et al. 2006). Moken et al. (1997) described the induction of the MDR phenotype in E. coli and its relevance to cross-resistance between pine oil, triclosan and multiple antibiotics. Jang et al. (2008) reported that, in S. aureus, exposure to triclosan (0.015 mg/L) resulted in down-regulation of the clpB chaperone-related genes, which might trigger the expression of resistant determinants. A recent study demonstrated that triclosan activates the expression of several groups of genes in E. coli and S. enterica (Bailey et al 2009). Transcriptome analyses (including microarray and RT-PCR experimental approaches) of bacteria exposed to triclosan (0.12 mg/L for 30 minutes) indicated an induction of the expression of various genes involved in drug efflux (e.g. acrB), in the genetic activation of resistance genes (e.g. marA), in the control of oxidative and drug response (e.g. soxS), and in the control of membrane permeability (e.g. ompR). Despite some differences in the response level observed between the two bacterial species, triclosan was shown to induce a rapid and adaptative response including the activation of several regulatory and structural genes involved in antibiotic resistance (Bailey et al. 2009).
McBain et al. (2004), however, failed to demonstrate a biologically significant induction of drug resistance in a number of bacterial species exposed to sub-lethal concentrations of triclosan, suggesting that triclosan-induced drug resistance is not generally readily inducible nor is it transferred across bacterial species.
5.2 Could exposure to triclosan lead to cross-resistance to antibiotics?
The SCCS opinion states:
6.6.1. General considerations
The possibility that the mechanisms involved in triclosan resistance may contribute to reduced susceptibility to clinically important and structurally unrelated antimicrobials is of major concern. It is important to note that antibacterial actions from antibiotics and biocides show some similarities in their mechanisms of action, behaviour and clinical aspects (Poole 2007).
Among the similarities, we can mention (i) the penetration/uptake through bacterial envelope by diffusion, (ii) the effect on the membrane integrity and morphology, (iii) the effect on diverse key steps of bacterial metabolism (replication, transcription, translation, transport, various enzymes). Faced with this chemical aggression and stress, bacteria mobilise similar defence mechanisms conferring resistance against structurally non-related molecules (Walsh and Fanning 2008).
A number of (but not all) laboratory studies have demonstrated an association between triclosan resistance and resistance to other antimicrobials. However, this link has not been confirmed in the limited number of in situ studies that have been performed to date. A number of bacterial mechanisms potentially conferring cross-resistance has been identified in laboratory investigations (see Table 6).
Studies on S. enterica and Stenotrophomonas maltophilia described the effect of triclosan on emerging bacterial cross-resistance. In S. enterica, Karatzas et al. (2007) reported that a triclosan-resistant strain overexpressing an efflux pump was less susceptible to antibiotics than the wild type original strain. Another study described the survival of S. enterica serovar Typhimurium following exposure to various disinfectants at a low concentration on the resulting changes in antibiotic profile (Randall et al. 2007). The authors concluded that growth of Salmonella with sub-inhibitory concentrations of biocides favours the emergence of strains resistant to different classes of antibiotics. In Stenotrophomonas, Sanchez et al. (2005) analysed the effect of triclosan on the selection of mutants overexpressing the efflux pump SmeDEF involved in both intrinsic and acquired resistance to antibiotics. The authors demonstrated that triclosan was able to select 5 mutants overexpressing this pump, out of a total of 12 mutants. This overexpression conferred resistance to a number of antibiotics such as tetracycline, chloramphenicol and ciprofloxacin.
Similar results have been reported with S. enterica and E. coli (Braoudaki and Hilton 2004). E. coli O157 strains, involved in the "hamburger disease", acquired high- levels of resistance to triclosan after only two sublethal exposures and when adapted, repeatedly demonstrated decreased susceptibilities to various antibiotics, including chloramphenicol, erythromycin, imipenem, tetracycline, and trimethoprim, as well as to a number of biocides. Bailey et al. (2009) showed that triclosan triggered the expression of a number of genes (e.g. encoding for efflux pumps, porins) directly involved in antibiotic resistance, and regulatory genes involved in the control of the antibiotic resistance gene cascade (activator of drug efflux, decrease of membrane permeability). Alteration in InhA in M. smegmatis following exposure to triclosan resulted in resistance to isoniazid (McMurry et al. 1999). Likewise, exposure of M. tuberculosis to triclosan led to mutation in inhA causing cross-resistance to isoniazid. However, isoniazid-resistant mutants were still susceptible to triclosan (Parikh et al. 2000).
Pycke et al. (2010) observed that triclosan exposure of the environmental α- proteobacterium Rhodospirillum rubrum led to an increase in triclosan MIC. The extent of this increase as well as the generation of different antibiotic susceptibility profiles was triclosan-concentration dependent, indicating the expression of distinct resistance mechanisms.
However, direct linkage between triclosan usage and bacterial resistance to other biocides and antibiotics might not be universal. Cottell et al. (2009) investigated the antibiotic susceptibility of triclosan tolerant S. aureus, E. coli and Acinetobacter johnsonii and reported that these strains remain susceptible to antibiotics used in clinical settings. In addition, triclosan-tolerant E. coli were found to be significantly more susceptible to aminoglycosides (Cottell et al. 2009). Likewise, triclosan resistant mutants in S. aureus did not show an altered antibiotic susceptibility profile compared to their parent strains (Suller and Russell 2000). Lear et al. (2006) demonstrated that environmental isolates with an increased MIC to triclosan remained susceptible to other biocides and antibiotics. Birošová and Mikulášová (2009) reported that continuous exposure of sub-inhibitory concentrations of triclosan did not increase emerging antibiotic resistance in S. enterica serovar Typhimurium but helped to maintain antibiotic-resistant bacteria in the population, notably those showing a mar phenotype. A short-term exposure to triclosan (30 min at 0.5 MIC, i.e. 0.098 mg/L) did not result in the selection of antibiotic resistant mutants.
5.3 Has bacterial resistance to triclosan been observed in the environment?
The SCCS opinion states:
Triclosan has been the most studied biocide with respect to its anti-bacterial activity. However, investigations concerned mainly laboratory experiments and only very few studies are available to date on bacterial resistance to triclosan in situ. Furthermore, in most in vitro studies, resistance to triclosan has been measured as an increase in MIC. As mentioned in section 6.2 above, the measurement of resistance based on MIC only, might have little bearing on bacterial survival to concentrations found in situ.
Ledder et al. (2006) investigated acquired high-level triclosan resistance in a number of distinct environmental isolates and reported that a relatively small number of strains showed a decrease in triclosan susceptibility (E. coli, Klebsiella oxytoca, Aranicola proteolyticus and S. maltophilia) while the susceptibility of the remaining 35 species remained unchanged. They concluded that repeated exposures to triclosan did not systematically produce high-level triclosan resistance in all bacteria. Furthermore, among the strains with decreased susceptibility, there was no change in antibiotic susceptibility or susceptibility to other biocides. Similarly, another study by the same group on repeated exposure of dental bacteria to triclosan resulted in the same conclusions (McBain et al. 2004).
Cole et al. (2003) collected 1238 isolates from the homes of users and non-users of antibacterial product and were unable to demonstrate any cross-resistance to antibiotic and antibacterial agents in target bacteria. In addition, this study showed an increased prevalence of potential pathogens in the homes of non-users of antibacterial products. However, in this study, the isolates were selected based on their antibiotic resistance and were then tested for their insusceptibility to biocides. With our current state of knowledge, it is generally accepted that antibiotic resistance in clinical isolates is not necessarily associated with resistance to biocides. Sullivan et al. (2003) studied the effect of triclosan in toothpaste on some bacterial species from the oral flora of 9 human volunteers over a 14- day period. Triclosan usage contributed to a decrease in lactobacilli although this decrease had no clinical significance. Furthermore, the antibiotic susceptibility profile of the oral streptococci investigated did not change following the use of triclosan containing toothpaste. Aiello et al. (2004) did not find any statistical significance between elevated triclosan MICs and antibiotic susceptibility in bacterial isolates taken from the hands of individuals using antibacterial cleaning and hygiene products for a 1-year period. Earlier studies reported no change in the ecology of the oral flora or resistance to triclosan following the use of triclosan-containing toothpaste. Jones et al. (1987) reported no change in the predominant plaque flora in 13 volunteers following the use of triclosan (2 g/L) for seven months. The authors did not observe any increase in triclosan MIC in these bacteria. Similar conclusions were reported by Walker et al. (1994) who reported no changes in the microbial flora in 144 patients following the use of 3 g/L triclosan-containing toothpaste. A meta-analysis of 16 clinical studies of the long-term effect (at least 6 month) of using triclosan toothpaste showed reduction in dental plaques and gingivitis (Davies et al. 2004).
7. TRICLOSAN BIOAVAILABILITY AND FORMULATION EFFECTS
The concentration of triclosan that comes in contact with a micro-organism governs the subsequent effect on that micro-organism (e.g. inhibitory, lethal, adaptation, selection). Hence the bioavailability of triclosan is paramount.
As described in Chapter 5, triclosan present in various environmental media is susceptible to degradation by oxidation by ozone, chlorine and sunlight, and to biodegradation by micro- organisms. The main route of exposure to soil is expected to be via the application of sewage sludge to agricultural soil. The bioavailability will depend on the sorption, mobility and degradation in soil under various physical conditions. Triclosan is released into surface waters via effluents from WWTP, and the bioavailability of the triclosan to micro-organisms in these media will depend upon sedimentation by binding with the particulate matter and stability of the compound during the exposure period.
The US EPA (2008) states on stability of triclosan in the environment that: "Triclosan is hydrolytically stable under abiotic and buffered conditions over the pH 4-9 range based on data from a preliminary test at 50°C.
Photolytically, triclosan degrades rapidly under continuous irradiation from artificial light at 25°C in a pH 7 aqueous solution, with a calculated aqueous photolytic half-life of 41 minutes.
Triclosan degrades rapidly in aerobic soils maintained in darkness at 20 ± 2°C, with calculated half-lives of 2.9-3.8 days.
In aerobic water-sediment systems maintained in darkness at 20 ±2°C, triclosan degraded with calculated nonlinear half-lives of 1.3-1.4 days in the water, 53.7-60.3 days in the sediment, and 39.8-55.9 days in the total system.
In soil, triclosan is expected to be immobile based on an estimated Koc of 9,200.
Triclosan is not expected to volatilize from soil (moist or dry) or water surfaces based on an estimated Henry’s Law constant of 1.5 x 10-7 atm-m3/mole.
Triclosan partially exists in the dissociated form in the environment based on a pKa of 7.9, and anions do not generally adsorb more strongly to organic carbon and clay than their neutral counterparts.
Hydrolysis is not expected to be an important environmental fate process due to the stability of triclosan in the presence of strong acids and bases. However, triclosan is susceptible to degradation via aqueous photolysis, with a half-life of <1 hour under abiotic conditions, and up to 10 days in lake water. An atmospheric half-life of 8 hours has also been estimated based on the reaction of triclosan with photochemically produced hydroxyl radicals.
In the laboratory, triclosan degraded via aerobic soil metabolism and aerobic aquatic metabolism, with half-lives of <4 days in soils and half-lives of <1.5 days (water layer) and up to 60 days (sediment and total system) in water-sediment systems."
Samsøe-Petersen et al. (2004) have described that half-life of triclosan for three experimental soils was calculated to be in the range of 17.4 to 35.2 days
Some observed concentrations of triclosan in the environment (e.g. Kumar et al. 2010) are high enough to induce changes in the microbial population. However, the bioavailability of triclosan in these environments (WWTP effluents, sludges, sediments, etc.) has not been determined. It is therefore important that the concentration effects of bioavailable triclosan are measured during the exposure period under study.
The presence of other chemicals (e.g. antibiotics, other biocides, surfactants...) in the environment may also affect the microbial population. Therefore it may be difficult to assess the effect of triclosan alone against microbial populations in the environment.
Triclosan-containing products are complex formulations since triclosan is poorly soluble in water. The role of the formulations is important to ensure the bioavailability of triclosan. Formulations might also enhance biocidal activity and/or reduce microbial aggregation, improving the biocidal activity of the product. The bioavailability of triclosan in surfaces or textiles, etc., is product dependent. Some manufacturers claim that triclosan does not leach out of their product.
5.4 How can bacterial resistance be determined?
The SCCS opinion states:
8. MEASUREMENT OF RESISTANCE AND CROSS-RESISTANCE
Concentration is central to the definition of bacterial resistance in practice (McDonnell and Russell 1999, Maillard and Denyer 2009). Therefore, the measurement of bacterial lethality rather than the measurement of bacterial growth inhibition is paramount. The determination of the lethality of the in-use concentration of a biocide will indicate, by comparison to a reference strain, whether a bacterial strain is insusceptible (i.e. intrinsically resistant) or has acquired resistance to a biocide or not.
The determination of minimum bactericidal concentrations (MBCs) is also an appropriate methodology that allows the comparison of lethality between a reference strain and “resistant” clinical/environmental isolates. Here, the reference strains represent the population of bacteria which is normally susceptible to the biocide. In addition the determination of the lethality of a biocide must involve the use of a neutralising agent or the removal of the biocide. Failure to do so will provide an over-estimation of the lethality of the biocide.
The determination of bacterial growth kinetics in the presence of a low concentration of a biocide can also provide indications to a change in bacterial phenotype (Thomas et al. 2004; Gomez Escalada et al. 2005a; Maillard 2007), but it does not indicate whether bacteria will become resistant to the biocide and cross-resistant to unrelated compounds or not.
Likewise, a number of protocols have been used to measure antibiotic susceptibility in bacterial isolates showing resistance, tolerance or increased insusceptibility to biocides or vice versa. The variety of protocols used contributes to the variability of the results reported on antibiotic “resistance”. For example, some studies based a change in antibiotic susceptibility profile on measurement of zone of inhibition (Tattawasart et al. 1999; Thomas et al. 2005). More meaningfully studies used standardised antibiotic susceptibility methodologies such as those given by the British Society for Antimicrobial Chemotherapy (BSAC) or Clinical and Laboratory Standards Institute (CLSI) to measure a change in antibiotic susceptibility profile. However a limited number of studies have looked at a decrease in antibiotic susceptibility that would be associated with treatment failure (Lear et al. 2006; Cottell et al. 2009).The effect of biocides on antibiotic susceptibility in bacteria has been measured indirectly, whereby a bacterial population is treated first with a biocide and the surviving bacteria then investigated for their susceptibility to antibiotics. However, there are currently no well-referenced criteria or standard protocols for the evaluation of the capability of a biocide to induce or select for resistance to antibiotics. Therefore, tools need to be developed to define for example the "minimal selecting concentration": the minimal concentration of a biocide which is able to select or trigger the emergence/expression of a resistance mechanism that will confer clinical resistance to an antibiotic class in a defined bacterium (SCENIHR 2009).
Since cross-resistance can be conferred by a number of distinct mechanisms, it is important to evaluate the propensity of a bacterium to express these mechanisms. Advances in modern genetic methods (e.g. PCR, -omics) and the development of an assay using specific chemosensitizers or markers (e.g. efflux pumps inhibitors) might allow the development of routine tests to identify resistance mechanisms.