Comparative susceptibility of resident and transient hand bacteria to para-chloro-meta-xylenol and triclosan

Journal of Applied Microbiology 2002, 93, 336–344

Comparative susceptibility of resident and transient hand bacteria to para-chloro-meta-xylenol and triclosan S.A. Johnson1, P.A. Goddard2, C. Iliffe2, B. Timmins2, A.H. Rickard1, G. Robson1 and P.S. Handley1 1

University of Manchester, Manchester and 2Reckitt Benckiser Healthcare, Hull, UK

2001 ⁄ 362: received 16 November 2001, revised 15 April 2002 and accepted 24 April 2002

S.A. JOHNSON, P.A. GODDARD, C. ILIFFE, B. TIMMINS, A.H. RICKARD, G. ROBSON AND P . S . H A N D L E Y . 2002.

Aims: To determine the susceptibility of planktonic and biofilm-grown strains of resident and transient skin bacteria to the liquid hand soap biocides para-chloro-meta-xylenol (PCMX) and triclosan. Methods and Results: Freshly isolated hand bacteria were identified by partial 16S rRNA gene sequencing. Two resident and three transient strains, as well as four exogenous potential transient strains, were selected for biocide susceptibility testing. The minimum inhibitory concentrations (MIC) and minimum bactericidal concentrations (MBC) of planktonic cells were determined. Resident and transient strains showed a range of susceptibilities to both biocides (PCMX, MIC 12Æ5–200 mg l)1, MBC 100–400 mg l)1; triclosan, MIC 0Æ6– > 40 mg l)1, MBC 1Æ3– > 40 mg l)1). Strains were attached to polystyrene plates for 65 h in 96-well microtitre plates and challenged with biocide to determine the biofilm inhibitory concentration and biofilm eradicating concentration. For all strains tested, biofilms were twoto eightfold less susceptible than planktonic cells to PCMX. Conclusions: Very few transients were detected on the hand. Transients were not more sensitive than residents to the biocides and susceptibility to PCMX and triclosan was strain dependent. Biofilm-grown strains were less susceptible to PCMX than planktonic cells. Significance and Impact of the Study: The study provides increased knowledge about the susceptibility of skin bacteria to biocides present in typical liquid antibacterial hand soaps and suggests that the concentration of biocide employed in such products is in excess of that required to kill the low numbers of transient bacteria typically found on skin. INTRODUCTION Despite the presence of rigorous hand wash regimes within hospitals and food production environments, food-poisoning outbreaks and nosocomial infections still occur with significant cost to health services (Webster et al. 1994; Pittet et al. 2000). Hand washing with antimicrobial soaps gives a greater reduction in bacterial numbers on the hand compared with soaps without added antimicrobial agents (Ayliffe et al. 1988; Namura et al. 1993). Therefore, antimicrobial agents are included in liquid hand soap formulations to improve hand hygiene in critical areas where Correspondence to: Dr P.S. Handley, Department of Biological Sciences, 1.800 Stopford Building, Manchester University, Oxford Road, Manchester M13 9PT, UK.

infections are easily spread, e.g. hospitals (Boyce 2001), food production areas (Cogan et al. 1999) and domestic settings (Barker et al. 2001). However, despite the ever-increasing use of antimicrobial soaps there are currently no studies available comparing the susceptibilities of resident and transient skin bacteria with the antimicrobial agents used in liquid soaps. Two distinct bacterial populations exist on the skin of healthy human hands. These are resident bacteria, described by Price (1938), which multiply and persist on the skin of healthy individuals, and transient bacteria, which are not believed to colonize healthy human skin and are present only in low numbers. Marples (1994) suggested that residents grow on the skin as small microcolonies encased in exopolymer. It is thought that these microcolonies are found either amongst the most superficial layers of the ª 2002 The Society for Applied Microbiology

BIOCIDE SUSCEPTIBILITY OF BACTERIA

stratum corneum (Noble 1968) or within the lumen of pilosebaceous follicles, as is the case for propionibacteria (Leeming et al. 1984). The predominant resident skin bacterium is the Gram-positive Staphylococcus epidermidis, although residents include other staphylococcal species, aerobic coryneforms, Micrococcus spp., propionibacteria and the yeast Malasseszia furfur (Kloos and Musselwhite 1975; Leeming et al. 1984). Typically, the transient microflora consists of Gram-negative rods (Lowbury 1969). It is assumed that transient bacteria are deposited on the skin surface from environmental sources or from other reservoirs within the human body, e.g. gastrointestinal tract, and it is generally presumed they are unable to become residents and multiply as microcolonies on healthy skin. The most welldocumented transients are those that cause skin infections, such as Pseudomonas aeruginosa (Fluit et al. 2001; Ahlen et al. 2001) and Staph. aureus (Higaki et al. 2000). Staphylococcus aureus is predominantly recovered from the anterior nares, although it has been recovered from other areas of the skin of healthy individuals without causing skin infection (Kloos and Musselwhite 1975). It has been shown that Staph. aureus is persistently carried by certain individuals (Williams 1969); however, its absence on other individuals has led to an unclear classification amongst researchers as to its resident or transient status. As a result of its ability to cause skin infections, Staph. aureus will be considered a transient for the purpose of this report. Other important potentially pathogenic transient bacteria on hands could include food-poisoning organisms, such as pathogenic strains of Escherichia coli (Todd and Dundas 2001), Salmonella spp. (Gallay et al. 2000) and Listeria monocytogenes (Hof 2001). It is not known whether transient bacteria are more susceptible to biocides than resident bacteria so, to investigate this, two active ingredients in liquid antimicrobial hand soap, para-chloro-meta-xylenol (PCMX) and triclosan, were used in this study. Both are phenolic biocides with a broad spectrum of activity against Gram-positive and -negative bacteria as well as against some yeasts and fungi (Brunch 1996; Rego¨s and Hitz 1974). To investigate the comparative susceptibilities of resident and transient skin bacteria to biocides, bacteria were isolated from the hand and identified by 16S rRNA gene sequencing. Biocide susceptibilities are usually determined on bacteria growing in suspension (planktonic cells). Therefore, the minimum inhibitory (MIC) and minimum bactericidal (MBC) concentrations of planktonically-grown residents and transients to PCMX and triclosan were determined. Resident bacteria are reported to grow as microcolonies on the skin and, therefore, they may exhibit the biofilm property of decreased biocide susceptibility in situ. Therefore, to determine whether skin bacteria grown as biofilms have decreased susceptibility to biocides, the biofilm inhibitory (BIC) and biofilm-eradicating (BEC)

337

concentrations of bacteria grown attached to polystyrene were determined.

MATERIALS AND METHODS Isolation of bacteria Resident and transient bacterial strains were isolated from the hands of volunteers using a modification of the glove wash technique used by Adams and Marrie (1982). The hand was placed in a non-powdered latex glove containing 50 ml tryptone soya broth (TSB; Oxoid, Basingstoke, UK) and agitated for 45 s. The resulting suspension was serially diluted, plated out onto the surface of tryptone soya agar (TSA) and incubated aerobically at 37C for 48 h. Although the temperature of the skin varies from 30 to 40C, 37C is used extensively for isolating skin bacteria (Adams and Marrie 1982; Leeming et al. 1984). This non-selective method isolates both obligate aerobes and facultative anaerobes which live on the exposed skin surfaces and which would be susceptible to hand wash biocides. Resident strains were identified as colony types that were found repeatedly in high numbers over 10 weeks. Transient bacteria were typically present on the hand wash dilution plates as single colonies, each with a distinct colony morphology differentiating them from resident bacteria. Bacterial strains Two resident Gram-positive strains and three transient Gram-negative rods were selected for biocide susceptibility testing and further identification (Table 1). The resident RAR1 was chosen as its colony morphology was typical of the predominant colony type found on all isolation plates over a 4-week period. The resident RNF1 was chosen as its colony morphology was typical of a colony type found persistently in lower numbers than RAR1 over the same 4-week period. Colonies of transients were present on isolation plates in very low numbers and only eight Gram-negative transient strains were found over the 4-week period of sampling. The three Gram-negative transients used in this study were selected as they had the most distinct colony types of the Gram-negative rods. Four other strains were also tested. Escherichia coli NCTC 10418 is a laboratory strain currently used in biocide testing (Reckitt Benckiser Healthcare, Hull, UK). Staphylococcus aureus SA8, isolated from a catheter, represents a potential hand transient and human opportunistic pathogen involved in nosocomial infection. A clinical strain of L. monocytogenes strain C52 represents an opportunistic pathogen important in food-poisoning outbreaks, which could potentially be transmitted via hands. Finally, Ps. veronii BL146, isolated from a swimming pool, was included as a potential transient from an environmental source.

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

338 S . A . J O H N S O N ET AL.

Table 1 Characteristics of the transient and resident skin strains together with their proposed genus and species names

Strain

Skin status

RAR1

Resident

RNF1

Resident

TSJ5

Transient

TSJ14

Transient

TSJ3

Transient

Colony morphology White circular entire edge Yellow circular entire edge Pink irregular edge rugose Mucoid circular entire edge Mucoid circular entire edge

Gram stain

Catalase

Oxidase

Highest % sequence similarity*

+ ve coccus

+

)

100

+ ve coccus

+

)

98Æ87

) ve rod

+

+

99Æ83

) ve rod

+

+

98Æ36

) ve rod

+

+

99Æ09

Proposed identity

EMBL accession nos

Staphylococcus epidermidis Kocuria rhizophila Pseudomonas stutzeri Brevundimonas diminuta Agrobacterium tumefaciens

AJ344227 AJ344225 AJ344226 AJ344224 AJ417902

*Highest percentage sequence similarity of two resident and three transient strains with named organisms in the EMBL database.

All strains were grown on TSA plates at 37C for 48 h. Liquid cultures were prepared in TSB and incubated with shaking at 200 rev min)1 at 37C for 20 h. Pseudomonas veronii BL146 was grown at 25C with shaking. Strains were stored at )70C in 50% sterile glycerol and TSB.

Sequencing was performed in a sequencer (ABI 377; PerkinElmer; PE Applied Biosystems, Foster City, CA, USA) and the sequences for each isolate then compiled using InheritTM (Perkin-Elmer). Compiled sequences of approx. 650 bases in length were obtained from each strain and compared with known sequences in the EMBL database.

Identification of isolated bacteria

Preparation of biocides

Strains were identified by polymerase chain reaction (PCR) amplification and partial sequencing of the 16S rRNA gene fragment (Rickard et al. 2000). Amplification of the 16S rRNA was performed by taking one bacterial colony of each organism grown on TSA, boiling it in 100 ll sterile nanopure water for 5 min and using 10 ll of the suspension as template DNA for PCR. Degenerate primers, 806R (Wilson et al. 1990) and 8FPL (Weisberg et al. 1991), were used to amplify a fragment of 16S rDNA, which corresponds to nucleotides 8–806 in the E. coli 16S rRNA gene sequence. The PCR reactions were carried out in PCR reaction buffer (Boehringer Mannheim, Indianapolis, IN, USA) containing 3Æ2 lmol l)1 of each primer, 0Æ5 mmol l)1 dNTPs and 2 U Taq DNA polymerase (Boehringer Mannheim) per 100 ll. The PCR cycles consisted of 35 cycles of 94C (1 min), 53C (1 min) and 72C (1 min), plus a final cycle with a 15-min chain elongation step at 72C. Amplified products were purified using the QIA quick PCR purification kits (Qiagen, Warrington, UK) according to the manufacturer’s instructions. The PCR products were subsequently sequenced using the primers 806R and 8FPL. Sequencing reactions consisted of 30–100 ng PCR product, 10 ng primer and 4 ll Big DyeTM (PE Applied Biosystems, Foster City, CA, USA) in a total volume of 20 ll. The samples were incubated at 94C (4 min) followed by 25 cycles of 96C (30 s), 50C (15 s) and 60C (4 min).

Stock solutions of biocides (obtained from Reckitt Benckiser Healthcare, Hull, UK) were prepared by dissolving 0Æ1 g of the biocide in the solvent dimethylsulphoxide (DMSO) to give a concentration of 10 000 mg l)1. The stock solution of triclosan was diluted 10-fold with DMSO to give a concentration of 1000 mg l)1. Biocide solutions were then diluted to the required concentration with sterile water and filter sterilized before use. Biocides were prepared to give a maximum of 4% (v ⁄ v) DMSO in the final solution. Previous experimentation had shown 4% (v ⁄ v) to be the maximum concentration of DMSO not to adversely effect cell viability (data not shown).

Media and strain maintenance

Susceptibility of planktonic cells to biocides The MIC and MBC of planktonic bacteria were determined in flat-bottomed 96-well polystyrene microtitre plates (Costar; Corning Inc., Corning, NY, USA). An 18-h bacterial culture was diluted to 1 · 105 cells ml)1 in double strength TSB and aliquots of inoculum (100 ll) added to 100 ll doubling dilutions of biocide in sterile double strength phosphate-buffered saline (PBS; Sigma St Louis, MO, USA). Plates were incubated for 24 h and the optical density (O.D.630 nm) used to quantify bacterial growth. The MIC was determined to be the lowest concentration of biocide where no growth was detected.

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

BIOCIDE SUSCEPTIBILITY OF BACTERIA

After incubation with biocide, the MBC was detected by using a metal loop to spread a small volume of solution over the surface of TSA plates, diluting the solution to neutralize any active biocide. Plates were then incubated for 24 h at 37C. No quenching agent (neutralizer) was used as phenolics have a high concentration exponent of 6 and, therefore, dilution is sufficient to inactivate the biocide (Denyer and Stewart 1998). The MBC was determined as the lowest concentration of biocide where no bacterial colonies appeared on the TSA plates. Biocide susceptibility of bacteria grown as 65-h biofilms Strains were initially compared for their ability to attach and form biofilms on the bottom of polystyrene microtitre wells. To produce biofilms, an inoculum of 1 · 105 cells ml)1 in double strength TSB was added to 96-well microtitre plates (Costar) and incubated for 65 h to allow development of biofilms on the polystyrene surface. After incubation at 37C (25C for Ps. veronii BL146) for 65 h the planktonic phase cells were gently removed and wells washed three times with 200 ll PBS to remove unattached cells. Double strength TSB (100 ll) and double strength biocide of varying concentrations (100 ll) was then added to the wells. The O.D.630 nm of the wells was measured to give the O.D. due to the biofilm and the plate then incubated for 24 h at 37 or 25C. After incubation, the O.D.630 nm was re-measured and the BIC determined as the lowest concentration where no

339

increase in O.D. occurred compared with the O.D.630 nm at t ¼ 0. Where no growth occurred in the supernatant fluid, a metal loop was used to take samples of biofilms from the bottom of wells, which were then spread over the surface of TSA plates. These plates were incubated for 24 h at 37 or 25C. As for the MBC, dilution was considered sufficient to inactivate the biocide, allowing the BEC to be determined as the lowest concentration at which no bacterial growth occurred on the TSA plates.

RESULTS Identification of isolated resident and transient bacteria The proposed identity and accession numbers for each strain in the EMBL database are shown in Table 1. Colony morphology, cell shape, catalase and oxidase tests were also carried out to support identification. The two resident strains were identified as Staph. epidermidis RAR1 and Kocuria rhizophila RNF1. The three transient Gramnegative strains were identified as Agrobacterium tumefaciens TSJ3, Ps. stutzeri TSJ5 and Brevundimonas diminuta TSJ14. Para-chloro-meta-xylenol susceptibility of strains grown in the planktonic phase A broad range of MIC and MBC values was obtained, MIC values ranged from 12Æ5 to 200 mg l)1 and MBC values

Table 2 Susceptibility to biocides of strains grown in the planktonic phase

PCMX Strain Agrobacterium tumefaciens TSJ3 Transient Brevundimonas diminuta TSJ14 Transient Escherichia coli NCTC 10418 Laboratory strain Kocuria rhizophila RNF1 Resident Listeria monocytogenes C52 Representative food pathogen Pseudomonas stutzeri TSJ5 Transient Pseudomonas veronii BL146 Representative transient Staphylococcus aureus SA8 Clinical isolate Staphylococcus epidermidis RAR1 Resident

Triclosan )1

)1

MIC (mg l ) MBC (mg l ) MIC (mg l)1) MBC (mg l)1) 12Æ5

120

6Æ5

> 40

30

200

12Æ5

> 40

200

400

0Æ6

1Æ3

100

200

> 40

> 40

75

100

> 20

> 20

150

400

> 40

> 40

75

100

> 30

> 30

75

400

3Æ1

7Æ5

100

400

5

7Æ5

PCMX, Para-chloro-meta-xylenol; MIC, minimum inhibitory concentration; MBC, minimum bactericidal concentration. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

340 S . A . J O H N S O N ET AL.

from 100 to 400 mg l)1 (Table 2). The strains most susceptible to the bactericidal effects of PCMX were the transients L. monocytogenes C52 and Ps. veronii BL146 (MBC 100 mg l)1). The least susceptible strains were Staph. aureus SA8, Staph. epidermidis RAR1, E. coli NCTC 10418 and Ps. stutzeri TSJ5 (MBC 400 mg l)1), which included both residents and transients. The group of transient bacteria tested, therefore, contained both the most and the least susceptible strains to PCMX, with a fourfold difference between them. There was no correlation between the resident and transient status of the strains and their susceptibility to PCMX. Triclosan susceptibility of strains grown in the planktonic phase A broad range of MIC and MBC values was obtained, MIC values ranged from 0Æ6 to > 40 mg l)1 and MBC values from 1Æ3 to > 40 mg l)1 (Table 2). Due to the low aqueous solubility of triclosan and the toxicity of DMSO at concentrations higher than 4% (v ⁄ v), it was not possible to precisely characterize the MIC and MBC values for less susceptible organisms. The strain most susceptible to the bactericidal effect of triclosan was E. coli NCTC 10418 (MBC 1Æ3 mg l)1) with a greater than 30-fold difference in susceptibility between the most and least susceptible strains (MBC > 40 mg l)1). The least susceptible strains were mostly transients, although the resident K. rhizophila was in this group. There was no correlation between the resident and transient status of the strains and their susceptibility to triclosan.

Strain Agrobacterium tumefaciens TSJ3 Transient Brevundimonas diminuta TSJ14 Transient Kocuria rhizophila RNF1 Resident Pseudomonas veronii BL146 Representative transient Staphylococcus aureus SA8 Clinical isolate Staphylococcus epidermidis RAR1 Resident

Biocide susceptibility of biofilm-grown strains Biofilm-grown strains were not tested against triclosan, as the MIC for the majority of strains already exceeded the triclosan concentration achievable in aqueous 4% (v ⁄ v) DMSO. Single species biofilms for the six strains were challenged with a range of PCMX concentrations from 25 to 700 mg l)1. The BICs and BECs were compared with the values obtained for the planktonic cells (Table 3) to show apparent changes in susceptibility due to biofilm growth. For all strains tested, attachment to polystyrene as a biofilm conferred decreased susceptibility to PCMX. The BIC was between two- and eightfold greater than the concentration required to inhibit growth in suspension (MIC), with BIC values ranging between 100 and 300 mg l)1 (Table 3). The biofilms of the three Gram-negative transient strains were the most sensitive to PCMX and the three Gram-positive cocci the least sensitive. Figure 1 shows the concentration of biocide at which inhibition of the biofilm occurred and that further incubation of the biofilm in biocide for up to 72 h did not alter the BIC value obtained after 24 h. The BEC was between 1Æ25- and twofold greater than the MBC of cells in suspension with BEC values ranging from 150 to 600 mg l)1. Biofilm cells of Ps. veronii BL146 were the most susceptible to killing by PCMX with a BEC of 150 mg l)1. The least susceptible strains were the two Gram-positive cocci, Staph. epidermidis RAR1 (BEC 600 mg l)1) and Staph. aureus SA8 (BEC 500 mg l)1).

Decrease in susceptibility compared BIC (mg l)1) with MIC

Decrease in susceptibility compared BEC (mg l)1) with MBC

100

8-fold

200

1Æ6-fold

100

3Æ3-fold

300

1Æ5-fold

250

2Æ5-fold

400

2-fold

150

2-fold

150

1Æ5-fold

300

4-fold

500

1Æ25-fold

200

2-fold

600

1Æ5-fold

Table 3 Susceptibility to para-chloro-metaxylenol of strains grown as biofilms for 65 h

BIC, Biofilm inhibitory concentration; BEC, biofilm-eradicating concentration; MIC, minimum inhibitory concentration; MBC, minimum bactericidal concentration. ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

BIOCIDE SUSCEPTIBILITY OF BACTERIA

opportunistic human pathogen. The proposed identity of TSJ5 was Ps. stutzeri, a common water and soil inhabitant (Sarnaik and Kanekar 1995) which has also been implicated as a human pathogen causing eye infections (Ben-Tovim et al. 1974). TSJ14 was identified as a Br. diminuta (previously classified as Pseudomonas), which has also been found in aquatic environments (Hernandez Duquino and Rosenberg 1987; Mouget et al. 1995).

1·6 1·4 1·2 O. D. 630 nm

341

1·0 0·8

Biocide susceptibility of strains in planktonic phase

0·6 0·4 0·2 0·0 0

25

50

75

100 150 200 300 400 600 700

Biocide concentration (mg l–1)

Fig. 1 Growth of 65-h Pseudomonas veronii BL146 biofilms in 96-well microtitre plates after 24 h (s) and 72 h (h) in the presence of increasing concentrations of para-chloro-meta-xylenol. Bacterial growth was measured in a 96-well microtitre plate reader. Positive growth was determined as an increase in optical density at a wavelength of 630 nm compared with the 65-h biofilm present at t ¼ 0 h (n) before re-incubation with the biocide

DISCUSSION Classification of skin strains as resident or transient The two resident strains RAR1 and RNF1 were identified as Staph. epidermidis and K. rhizophila, respectively. Staphylococcus epidermidis is the predominant resident on skin (Kloos and Schleifer 1975). It is also a major nosocomial pathogen associated with implanted medical devices (O’Gara and Humphreys 2001). Kocuria rhizophila has not previously been identified as a resident on the skin, although other members of the genus Kocuria (previously classified as Micrococcus), such as K. rosea, K. kristinae and K. varians, have been identified as residents (Kloos and Musselwhite 1975). Therefore, for this reason and the fact that it was repeatedly isolated, it was considered a resident in this study. The three transient strains isolated were all common water and soil inhabitants, which could easily be picked up as hand transients. TSJ3 was identified as Agrobacterium tumefaciens, which is often isolated from soil and plants (Ogawa et al. 2000; Bien et al. 1990). Although primarily a plant pathogen, Ag. tumefaciens has been isolated from cases of bacteraemia in immuno-compromised patients (Southern 1996; Salavert et al. 1997) illustrating its potential as a

Biocide susceptibility varied according to species, with resident and transient groups both showing a range of susceptibilities. For both PCMX and triclosan, the transient strains showed both the highest and lowest susceptibility indicating no correlation between biocide susceptibility and resident or transient status with the limited number of bacteria tested within this study. Planktonic biocide sensitivity did not correlate with cell wall structure. Gram-negative and -positive strains showed both high and low susceptibility for both PCMX and triclosan. The reasons for this are not understood as, typically, bacteria vary in their susceptibility to biocides, with Gram-negative bacteria reportedly less susceptible to biocides than Gram-positive bacteria (Russell 1999). The difference in susceptibility to biocides is thought to be due to differences in the permeability or binding capacity of the cells (Russell and Day 1996). In particular, Gram-negative bacteria can have a high cell impermeability, so that access of biocide to its target sites can be prevented or reduced, causing cells to be intrinsically less susceptible to biocides (Russell 1993, 1999). However, in this study, the Gramnegative strains as a group were not, as might be expected, less susceptible than the Gram-positive strains. The nine strains challenged with PCMX had MBC values ranging from 100 to 400 mg l)1, confirming that PCMX is a broad spectrum biocide acting on both Gram-positive and -negative strains. The MBC values for PCMX are well below the concentration in commercial liquid hand soaps (typically 5000 mg l)1). This indicates that the bacteria would be susceptible to liquid hand soaps in practice, even though the exposure to PCMX during a hand wash is shorter than in this study and liquid hand soaps are considerably diluted during use. Published data on bacterial susceptibility to PCMX are limited and test methods vary. This study showed that E. coli NCTC 10418 and Staph. aureus SA8 had MICs higher than those previously published for these species (Brunch 1996). However, Staph. epidermidis RAR1 had an MIC lower than the MICs previously published for a range of Staphylococcus spp. (Brunch 1996). The variation in test methods used to determine MIC and MBC values makes

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

342 S . A . J O H N S O N ET AL.

direct comparison between studies difficult; however, it would appear that susceptibility to PCMX varies between strains as well as species. Six of the nine strains challenged with triclosan had MBC values greater than 30 mg l)1. However, the concentration of triclosan in commercial liquid hand soaps varies but is typically 1000–5000 mg l)1, suggesting that liquid hand soap would be effective at killing bacteria. The study showed that the MICs of the strains tested for triclosan susceptibility were higher than those previously reported. The MICs for a range of strains have been determined and ranged between 0Æ1 and 30 mg l)1 (Rego¨s and Hitz 1974; Vischer and Rego¨s 1974), although agar incorporation tests were used. Staphylococcus aureus SA8 had an MIC of 3Æ1 mg l)1 although Staph. aureus strains have been isolated with higher MICs of 32 mg l)1 (Bamber and Neal 1999). Clinical isolates of Staph. aureus have demonstrated lower MIC values than Staph. aureus SA8; however, direct comparison cannot be made with the present work as an agar incorporation method was used (Suller and Russell 2000). In this study, Ps. veronii BL 146 had a high MIC (30 mg l)1) and this agrees with previous work showing that Pseudomonas species have high MICs for triclosan. For example, Ps. fluorescens had an MIC > 100 mg l)1 and Ps. aeruginosa an MIC > 1000 mg l)1 (Rego¨s and Hitz 1974). Pseudomonas stutzeri TSJ5 also had a high MIC although this species has previously been reported as sensitive to triclosan (Tattawasart et al. 1999). As with PCMX, it would appear that sensitivity to triclosan varies between species and strains. It has been suggested that triclosan in the environment may cause reduced triclosan susceptibility, which could result in strains being less susceptible to antibiotics (Levy 2001; McMurry et al. 1998). A link between triclosan and antibiotic resistance has been shown in certain strains (Schweizer 2001) yet other studies have failed to develop stable enhanced resistance to triclosan in Staph. aureus on exposure to low levels of the biocide (Cox 1987; Suller and Russell 2000), suggesting that further research is required. Biocide susceptibility of strains grown as biofilms All strains grown as biofilms on polystyrene showed decreased sensitivity to PCMX and triclosan although the reduction was only 1Æ25–2-fold. It is well established that biofilms show decreased susceptibility to antimicrobial agents, although the factor increases range from twofold (Das et al. 1998) up to 1000-fold (Nickel et al. 1985; Gristina et al. 1987; Prosser et al. 1987). The present results are in agreement with more recent studies, using microtitre wells, where thin Ps. aeruginosa biofilms exhibited a two- to threefold reduction in susceptibility to biocides (Cochran et al. 2000) and 24 h Staph. aureus biofilms were 10–20-fold

less susceptible to antibiotics compared with planktonic cells (Williams et al. 1997). The reasons for biofilm recalcitrance to antimicrobial agents are complex and involve a number of factors. For example, the slow growth rate of biofilm cells and attachment-specific resistant phenotypes, as well as reaction diffusion limitation, have been implicated in biofilm resistance to antimicrobial agents (Allison et al. 2000; Gilbert and Allison 1999). The present data clearly show that all transient and resident strains can develop a biofilm phenotype with a reduced susceptibility to biocides. Hygienic hand washes in hospital, food production and domestic environments are formulated with the aim of killing the transient bacteria on the hands without eliminating the resident bacterial population. In this study, it has been shown that the resident strains are not intrinsically more susceptible to the biocides than transients, so the differential killing effect must be achieved in other ways. Firstly, the lower numbers of transients will be killed quickly compared with the much higher numbers of resident bacteria. Secondly, since resident skin bacteria are proposed to grow as microcolonies on the skin and they also exhibit the biofilm property of reduced biocide susceptibility, they would have an enhanced survival advantage over the transients which are not thought to exhibit a biofilm phenotype on the skin. The adherence of microcolonies growing in the superficial layers of the stratum corneum would also be expected to be far stronger than superficial transient bacteria. The data presented here were necessarily obtained using standard sensitivity assays for comparative purposes. However, the in vitro biofilms tested in this study do not accurately reflect the types of biofilm found in situ on the skin. The size of the biofilms on skin is dependent on their location. Microcolonies of less than 10 bacteria occur on the exposed skin surface (Malcolm and Hughes 1980) compared with microcolonies of up to 100 bacteria associated with hair follicles (Montes and Wilborn 1970) and sebaceous ducts (Ho¨lze and Kligman 1978; Williams 1963). At present, there are no experimental methods to test the biocide susceptibility of skin bacteria that suitably reflect the attached status of bacteria on the hand. Therefore, work is currently underway to develop a method where bacteria attached to skin cells as small microcolonies can be challenged and their biocide susceptibility compared with planktonic cells. REFERENCES Adams, B.G. and Marrie, T.J. (1982) Hand carriage of aerobic Gramnegative rods by health care personnel. Journal of Hygiene 89, 23–29. Ahlen, C., Mandal, L.H. and Iversen, O. (2001) The impact of environmental Pseudomonas aeruginosa genotypes on skin infections

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

BIOCIDE SUSCEPTIBILITY OF BACTERIA

in occupational saturation diving systems. Scandinavian Journal of Infectious Disease 33(6), 413–419. Allison, D.G., McBain, A.J. and Gilbert, P. (2000) Microbial biofilms: Problems of control. In Community Structure and Cooperation in Biofilms ed. Lappin-Scott, H., Gilbert, P., Wilson, M. and Roberts, D. pp. 309–327. Cambridge: Cambridge University Press. Ayliffe, G.A.J., Babb, J.R., Davies, J.G. and Lilly, H.A. (1988) Hand disinfection: a comparison of various agents in laboratory and ward studies. Journal of Hospital Infection 11, 226–243. Bamber, A.I. and Neal, T.J. (1999) An assessment of triclosan susceptibility in methicillin-resistant and methicillin-sensitive Staphylococcus aureus. Journal of Hospital Infection 41, 107–109. Barker, J., Stevens, D. and Bloomfield, S.F. (2001) Spread and prevention of some common viral infections in community facilities and domestic homes. Journal of Applied Microbiology 91, 7–21. Ben-Tovim, T., Eylan, E., Romano, A. and Stein, R. (1974) Gramnegative bacteria isolated from external eye infections. Infection 2, 162–165. Bien, E., Lorenz, D., Eichhorn, K. and Plapp, P. (1990) Isolation and characterization of Agrobacterium tumefaciens from German vine region Rheinpfalz. Zeitschrift fu¨r Pflanzenkrankheiten und Pflanzenschutz 97, 313–322. Boyce, J.M. (2001) MRSA patients: proven methods to treat colonization and infection. Journal of Hospital Infection 48 (Suppl. A), s9– s14. Brunch, M.K. (1996) Chloroxylenol: an old–new antimicrobial. In Handbook of Disinfectants and Antiseptics ed. Asceuzi, J.M. pp. 265– 294. New York: Marcel Dekker. Cochran, W.L., McFeters, G.A. and Stewart, P.S. (2000) Reduced susceptibility of thin Pseudomonas aeruginosa biofilms to hydrogen peroxide and monochloramine. Journal of Applied Microbiology 88, 22–30. Cogan, T.A., Bloomfield, S.F. and Humphrey, T.J. (1999) The effectiveness of hygiene procedures for prevention of cross-contamination from chicken carcasses in the domestic kitchen. Letters in Applied Microbiology 29, 354–358. Cox, A.R. (1987) Efficacy of the antimicrobial agent triclosan in topical deodorant products: Recent developments in vivo. Journal of the Society of Cosmetic Chemists 38, 223–231. Das, J.R., Bhakoo, M., Jones, M.V. and Gilbert, P. (1998) Changes in the biocide susceptibility of Staphylococcus epidermidis and Escherichia coli cells associated with rapid attachment to plastic surfaces. Journal of Applied Microbiology 84, 852–858. Denyer, S.P. and Stewart, G.S.A.B. (1998) Mechanisms of action of disinfectants. International Journal of Biodeterioration and Biodegradation 41, 261–268. Fluit, A.C., Schmitz, F.J. and Verhoef, J. (2001) Frequency and isolation of pathogens from bloodstream, nosocomial pneumonia, skin and soft tissue, and urinary tract infections occurring in European patients. European Journal of Clinical Microbiology and Infections 20(3), 188–191. Gallay, A., Vaillant, V., Bouvet, P., Grimont, P. and Desenclos, J. (2000) How many food borne outbreaks of Salmonella infection occurred in France in 1995? Application of the capture-recapture method to three surveillance systems. American Journal of Epidemiology 152(2), 171–177.

343

Gilbert, P. and Allison, D. (1999) Biofilms and their resistance towards antimicrobial agents. In Dental Plaque Revisited ed. Newman, H. and Wilson, M. pp 125–143. Cardiff: Bioline. Gristina, A.G., Hobgood, C.B., Webb, L.X. and Myrvik, Q.N. (1987) Adhesive colonisation of biomaterials and antibiotic resistance. Biomaterials 8, 423–426. Hernandez Duquino, H. and Rosenberg, F.A. (1987) Antibioticresistant Pseudomonas in bottled drinking water. Canadian Journal of Microbiology 33, 286–289. Higaki, S., Kitagawa, T., Kagoura, M., Morohashi, M. and Yamagishi, T. (2000) Predominant Staphylococcus aureus isolated from various skin diseases. Journal of International Medical Research 28(4), 187– 190. Hof, H. (2001) Listeria monocytogenes: A causative agent of gastroenteritis. European Journal of Clinical Microbiology and Infections 20(6), 369–373. Ho¨lze, E. and Kligman, A.M. (1978) The pathogenesis of miliaria rubra: role of the resident microflora. British Journal of Dermatology 86, 117. Kloos, W.E. and Musselwhite, M.S. (1975) Distribution and persistence of Staphylococcus and Micrococcus species and other aerobic bacteria on human skin. Applied Microbiology 30, 381–395. Kloos, W.E. and Schleifer, K.H. (1975) Isolation and characterization of staphylococci from human skin. International Journal of Systemic Bacteriology 25, 62–79. Leeming, J.P., Holland, K.T. and Cunliffe, W.J. (1984) The microbial ecology of pilosebaceous units from human skin. Journal of General Microbiology 130, 803–807. Levy, S.B. (2001) Antibacterial household products: cause for concern. Emerging Infectious Diseases 7(3) (Suppl.), S512–S515. Lowbury, E.J.L. (1969) Gram-negative bacilli on the skin. British Journal of Dermatology 81 (Suppl. 1), 55–61. Malcolm, S.A. and Hughes, T.C. (1980) The demonstration of bacteria on and within the stratum corneum using scanning electron microscopy. British Journal of Dermatology 102, 267–275. Marples, R.R. (1994) The normal flora of the skin is a biofilm. In Bacterial Biofilms and Their Control in Medicine and Industry ed. Wimpenny, J., Nichols, W., Stickler, D. and Lappin-Scott, H. pp. 127–129. Cardiff: Bioline. McMurry, L.A., Oethinger, M. and Levy, S.B. (1998) Over expression of marA, soxS, or acrAB produces resistance to triclosan in laboratory and clinical strains of Escherichia coli. FEMS Microbiology Letters 166, 305–309. Montes, L.F. and Wilborn, W.H. (1970) Anatomical location of normal skin flora. Archives of Dermatology 101, 145–159. Mouget, J.L., Dakhama, A., Lavoie, M.C. and De La Noue, J. (1995) Algal growth enhancement by bacteria: is consumption of photosynthetic oxygen involved. FEMS Microbiology Ecology 18(1), 35– 44. Namura, S., Nishijima, S., McGinley, K.J. and Leyden, J. (1993) A study of the efficacy of antimicrobial detergents for hand washing using the full-hand touch plates method. Journal of Dermatology (Tokyo) 20(2), 88–93. Nickel, J.C., Ruseska, I., Wright, J.B. and Costerton, J.W. (1985) Tobramycin resistance of Pseudomonas aeruginosa cells growing as a biofilm on urinary tract catheter material. Antimicrobial Agents and Chemotherapy 27, 619–624.

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344

344 S . A . J O H N S O N ET AL.

Noble, W.C. (1968) Observations on the surface flora of the skin and the skin pH. British Journal of Dermatology 80, 279. O’Gara, J.P. and Humphreys, H. (2001) Review: Staphylococcus epidermidis biofilms: importance and implications. Journal of Medical Microbiology 50, 582–587. Ogawa, Y., Ishikawa, K. and Mii, M. (2000) Highly tumorigenic Agrobacterium tumefaciens strain from crown gall tumors in chrysanthemum. Archive Microbiologica 173, 311–315. Pittet, D., Hugonnet, S., Harbarth, S., Mourouga, P. and Sauvan, V. (2000) Effectiveness of a hospital-wide programme to improve compliance with hand hygiene. Lancet 356, 1307–1312. Price, P.B. (1938) The bacteriology of normal skin; a new quantitative test applied to a study of the bacterial flora and the disinfectant action of mechanical cleansing. Journal of Infectious Disease 63, 301. Prosser, B.T., Taylor, D., Dix, B.A. and Cleeland, R. (1987) Methods of evaluating effects of antibiotics upon bacterial biofilms. Antimicrobial Agents and Chemotherapy 31, 1502–1506. Rego¨s, J. and Hitz, H.R. (1974) Investigations on the mode of action of triclosan, a broad spectrum antimicrobial agent. Zentralblatt fu¨r Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene-Erste Abt. Orig.-Reihe A (Medizinsche, Mikrobiologie und Parasitologie) 226, 390–401. Rickard, A.H., Leach, S.A., Buswell, C.M., High, N.J. and Handley, P.S. (2000) Coaggregation between aquatic bacteria is mediated by specific-growth-phase-dependent lectin–saccharide interactions. Applied and Environmental Microbiology 66(1), 431–434. Russell, A.D. (1993) Microbial cell walls and resistance of bacteria and fungi to antibiotics and biocides. Journal of Infectious Diseases 168(5), 1339–1340. Russell, A.D. (1999) Bacterial resistance to disinfectants: Present knowledge and future problems. Journal of Hospital Infection 43 (Suppl.), S57–S68. Russell, A.D. and Day, M.J. (1996) Antibiotic and biocide resistance. Microbios 85(342), 45–65. Salavert, M., Breton, J.R., Perez-Tamarit, D., Perez-Belles, C. and Gobernado, M. (1997) Persistent bacteraemia and infection of an intravascular device by Agrobacterium radiobacter (tumefaciens) in a boy with AIDS. Enfermedades Infecciosas y Microbiologia Clinical 15, 496–497. Sarnaik, S. and Kanekar, P. (1995) Bioremediation of colour of methyl violet and phenol from a dye-industry waste effluent using

Pseudomonas spp. isolated from factory soil. Journal of Applied Bacteriology 79, 459–469. Schweizer, H.P. (2001) Triclosan: a widely used biocide and its link to antibiotics. FEMS Microbiology Letters 202, 1–7. Southern, P.M. (1996) Bacteremia due to Agrobacterium tumefaciens (radiobacter). Report of infection in a pregnant woman and her stillborn foetus. Diagnosis of Microbial Infectious Diseases 24, 43–45. Suller, M.T.E. and Russell, A.D. (2000) Triclosan and antibiotic resistance in Staphylococcus aureus. Journal of Antimicrobial Chemotherapy 46, 11–18. Tattawasart, U., Maillard, J.Y., Furr, J.R. and Russell, A.D. (1999) Comparative responses of Pseudomonas stutzeri and Pseudomonas aeruginosa to antibacterial agents. Journal of Applied Microbiology 87, 323–331. Todd, W.T.A. and Dundas, S. (2001) The management of VTEC 0157 infection. International Journal of Food Microbiology 66(1–2), 103– 110. Vischer, W.A. and Rego¨s, J. (1974) Antimicrobial spectrum of triclosan, a broad-spectrum antimicrobial agent for topical application. Zentralblatt fu¨r Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene-Erste Abt. Orig.-Reihe A (Medizinsche, Mikrobiologie und Parasitologie) 226, 376–389. Webster, J., Faoagali, J.L. and Cartwright, D. (1994) Elimination of methicillin-resistant Staphylococcus aureus from a neonatal intensive care unit after handwashing with triclosan. Journal of Paediatrics and Child Health 30, 59–64. Weisberg, W.G., Barns, S.M., Pelletier, D.A. and Lane, D.J. (1991) 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173, 697–703. Williams, R.E.O. (1963) Healthy carriage of Staphylococcus aureus: its prevalence and importance. Bacteriological Reviews 27, 56. Williams, R.E.O. (1969) Staphylococcus aureus on the skin. British Journal of Dermatology 81 (Suppl. 1), 33–36. Williams, I., Venables, W.A., Lloyd, D., Paul, F. and Critchley, I. (1997) The effects of adherence to silicone surfaces on antibiotic susceptibility in Staphylococcus aureus. Microbiology 143, 2407– 2413. Wilson, K.H., Blitchington, R.B. and Greene, R.C. (1990) Amplification of bacterial 16S ribosomal DNA with polymerase chain reaction. Journal of Clinical Microbiology 28, 1942–1946.

ª 2002 The Society for Applied Microbiology, Journal of Applied Microbiology, 93, 336–344