Time survivor study of Escherichia coli with polyhexamethylene biguanide on cotton

Abstract

Time survivor or time kill studies are commonly used to investigate the efficacy of antimicrobial agents in homogeneous solutions. Such a study was attempted via a textile treated with an antimicrobial agent. For this study, a finished undyed cotton fabric and a commercially available antimicrobial agent, polyhexamethylene biguanide, were used. The release of the antimicrobial agent from the cotton fabric when submerged in water with a liquor-to-cloth ratio of 20:1 was evaluated. The antibacterial agent-treated cotton fabric was also tested according to the JIS L 1902 absorption antibacterial testing method at various agent concentrations applied to the fabric and incubation times. The treated textile showed a quick release of agent when submerged in water and the results of the antibacterial tests showed increasing antibacterial activity with increases in concentration, as has been found in homogeneous solutions. Fabrics treated with lower concentrations of the agent show bacteriostatic action. A regrowth of microorganisms was additionally noted at certain incubation times.

Keywords

antibacterial textiles cotton eosin staining JIS L 1902 release

Antibacterial textiles are of increasing interest in textile applications. Textiles can be made antibacterial using substances such as quaternary ammonium compounds, metal-based agents such as silver or other metallic salts, copper or titanium oxide, natural polymers such as chitosan and it’s quaternized derivatives, triclosan, regenerable halamines, dyes, and poly biguanides, etc.¹ Polyhexamethylene biguanide is an important polybiguanide that is known to be a disinfectant and is also used as an antibacterial agent for cotton or cotton–polyester textiles. It is applied in wound dressings as well.² The biguanide consists of water-soluble polycationic polymers with a hydrophobic backbone having multiple cationic groupings separated by hexamethylene chains, as shown in Figure 1. The terminating end groups can be amine, guanide, or cyanoguanide groups.

Figure 1.

Structure of polyhexamethylene biguanide. Source: reproduced with permission from Elsevier, 2011.¹⁹

The cationic polyhexamethylene biguanide bonds to cotton through anionic carboxylate groups on the cellulose. The adsorption isotherm of this agent on cotton is said to be of a Langmuir type at lower concentrations, when the weaker electrostatic interactions prevail, and of a Freundlich type at higher concentrations, when hydrogen bonding is predominant.³

Polyhexamethylene biguanide is considered as a membrane-active agent similar to quaternary ammonium compounds. The first step in its attack on the bacterium cell wall is the immediate attraction to the bacterium’s divalent cations, such as Mg²⁺ and Ca²⁺. The agent then proceeds to interact with the liposaccharides and peptidoglycan layer of the cell wall and bind with the cell membrane layer. The polyhexamethylene biguanide molecules tend to aggregate over the lipid bilayer with protein sites of the cellular membrane where the charge density is high. The proteins lose their function, while the lipid membrane slowly starts to solubilize, leading to cellular leakage and loss of the membrane permeability barrier, as schematically shown in Figure 2.⁴ Figure 2(c) and (d) show the aggregation of the agent and the subsequent dissolution of the membrane.

Figure 2.

Mechanism of antibacterial activity of PHMB. Source: reproduced with permission from John Wiley & Sons, 2011.⁴

While the hydrophobic inflexible hexamethylene can enter the core of the cellular membrane, unlike other membrane active agents the entire molecule cannot. This then renders it less susceptible to resistance mechanisms through efflux pumps. There is currently no evidence to suggest that it would render certain strains of bacteria resistant.⁴ This was one of the primary reasons for the interest of the authors in this antibacterial agent. The other reason being that this agent is a commercially available agent for finishing on cotton and cotton–polyester substrates and requires no synthesis; therefore, it is ideal to study its mechanism for future use on a commercial scale in the laundry industry.

Time survivor or time kill studies are common in the field of disinfectants where time of action and concentration are considered to be the primary variables.^5,6 The aim of this paper is, therefore, to perform a similar study via a textile interface. In the area of textiles, studies exist showing the efficacy of textiles treated with polyhexamethylene biguanide at certain concentrations in killing bacteria, along with laundering fastness studies.^7–9 The study of the efficacy of a polyhexamethylene biguanide treated textile for various incubation times for a particular concentration has been attempted as well.¹⁰ This study was done in order to have some insight into the antibacterial agent activity on textiles through the combination of time and concentration of the agent.

Experimental details

Materials and methods

A 20% aqueous polyhexamethylene biguanide hydrochloride stock solution (called Reputex 20) was procured from Lonza Group Limited, UK. This procured stock solution contains 0.073 mol/l of polyhexamethylene biguanide hydrochloride, considering its molar mass of 2750 g/mol (C Chadwick, personal communication, 2011). This stock solution is referred to as PHMB in this paper. The textile used for testing was bleached white plain woven cotton with a fabric density of 180 g/m². The E. coli strain for the tests was ATCC 11229 obtained from the LGC standard company, UK.¹¹ The material for the Luria Broth medium and microbiological agar, Triton X100, and Eosin Y (C.I. Acid Red 87, 2′, 4′, 5′, 7′-tetrabromofluorescein) were procured from Sigma Aldrich.

PHMB incorporation

Textile samples of 30 cm by 30 cm were treated with PHMB solutions using an exhaustion method at 40℃ and at pH 7 for 30 min in a beaker glass under stirring. The pH was adjusted by the addition of sodium hydroxide or acetic acid drops to the exhaustion beaker glass.

The liquor-to-cloth ratio (LCR) was 20:1. The samples were treated with PHMB to obtain 0.4%, 0.8%, 1.2%, 1.6%, and 2.0% o.w.f. The concentrations in the liquor were 0.02 v/v%, 0.04 v/v%, 0.06 v/v%, 0.08 v/v%, and 0.1 v/v%. It has to be noted that the concentrations calculated on the basis of the weight of fabric do not refer to the actual amount of PHMB fixed to the fabric. The amount of PHMB added to the liquor was calculated from the dry weight of the sample and the desired percentage o.w.f. After treatment, the samples were rinsed in cold water once and then air dried.

Study of the release of PHMB from the textile

For the release studies, 0.25 g fabric treated with 2% PHMB o.w.f. was placed in a test tube containing 5 ml of water (LCR 20:1) at room temperature. 1 ml from the test tube was taken at time intervals (initially, every minute for the first 5 minutes and then every second minute from the 5th to the 15th minute and then every five minutes till the 40th minute of the release, and finally every 15 minutes till the 70th minute). The last measurement was at 100 minutes. The absorbance was found using a Cary 100 UV spectrophotometer (Agilent Technologies, US) and before each measurement the test tube was shaken up and down once, and the water inside was pipetted in and out twice. The measurements were done at a maximum absorbance of 236 nm.¹² Three samples were tested and then averaged for each measured point of the release study. 2% PHMB o.w.f. was chosen for the release study because the PHMB provider noted that at least 2% o.w.f. is required for a durable finish of up to 50 washes and therefore it was likely that such a concentration on the weight of fabric is interesting for users.¹³ Moreover, the antibacterial tests also showed that this concentration of the agent on the weight of the fabric was optimal for a good antibacterial activity.

Antibacterial activity tests

For the antibacterial activity tests, 0.25 g of textile material was taken for the inoculation with 200 µl of 1–3 × 10⁶ CFU (colony forming unit)/ml of E. coli bacterium. This concentration was obtained by measuring the optical density of the inoculum with a WPACO 8000 Biowave personal cell density meter (Biochrom, UK) at 600 nm. The required optical density was obtained after diluting the bacterium grown to 10⁸ CFU/ml with Luria Broth medium. This procedure was done according to the protocol described in the JIS L 1902 absorption testing method.¹⁴ This testing method requires 24 hours of incubation; we, however, modified this by also doing tests at various other incubation times.

Three PHMB-treated samples were inoculated for each measurement. For the control, an untreated cotton fabric of similar dimensions and weight was taken and inoculated in the same way. After the inoculation, the samples were placed on the inside of a petri dish lid and then covered with an inverted agar-filled petri dish bottom, and finally sealed with paraffin tape to prevent drying of the sample during the incubation. The samples were then placed in an incubator, operating at 37℃ for the required duration (1, 3, 6, 12, and 24 hours).

The amount of bacteria eluted from untreated and treated samples at time 0 (C₀ and T₀) was 1–3 × 10⁶ CFU/ml. This was determined by eluting the bacteria immediately from the samples after inoculation using the above-described procedure, but without incubation step.

After the incubation period, the samples were taken out and shaken out in 20 ml of physiological saline containing 2 g/l Triton X100. 1 ml of this shaken out solution was then serially diluted in 9 ml of physiological saline for the required number of dilutions and then 0.1 ml was plated. The number of colonies grown on the plate was manually counted the next day to calculate the bacteria eluted from each specimen (through a plate count method). The resolution of the plate count method was 6 × 10³ CFU per specimen, below which the CFUs obtained were considered to be 0 CFUs, the reason being that only plates with 30–300 colonies were considered.

Eosin staining

The goal of this study was to correlate the amount of applied PHMB o.w.f. (%) on the textile to the K/S value of the stained textile. The staining of PHMB-treated fabrics was done with the anionic dye Eosin Y. Eosin and PHMB are known to bind 1:1 stoichiometrically and, in fact, Eosin Y can be used to predict the antibacterial activity of PHMB in homogeneous media.¹⁵ 50 ml of 0.6% Eosin Y dye was dissolved in 350 ml demi-water, and 50 g sodium citrate was then added and the volume of the dye solution was made up to 500 ml with demi-water again. Each 0.25 g textile piece was treated with PHMB separately, put into 25 ml of the dye solution, and stirred for 10 min at room temperature. The sample was then rinsed in cold water and air dried. The same was done for the untreated textile. The samples were measured for reflectance with an X-Rite color spectrophotometer (X-Rite, Germany) with D65 illuminant and a 10° observer. The reflectance values at 530 nm were taken for the Kubelka Munk calculations. From the reflectance values at 530 nm, an absorption-scattering coefficient (K/S) was used with the Kubelka Munk equation to determine color strength, as shown in equation (1). We correlated this to the amount of PHMB concentration o.w.f. (%).¹³

\frac{K}{S} = \frac{(1 - R)^{2}}{2 R}

(1)

Where K is the absorption coefficient, S is the scattering coefficient, and R is the reflectance at a particular wavelength.

Results and discussion

Release of PHMB from the treated fabric

An earlier study relating to the release of PHMB from cellulose wound dressing at a LCR ratio of 20:1 indicated that more than 75% of PHMB (of unknown concentration o.w.f.) was released from the wound dressing within 24 hours at static conditions (i.e., without stirring or mixing) in saline solution.¹⁶ Here, the goal of this experiment was to observe the release of PHMB in water.

A calibration curve was obtained with each point being an average of three measurements for the release kinetics study, as shown in Figure 3(a). The R² value of coefficient of correlation obtained was 0.97. In addition to the calibration curve, measurements were made to check the linear range of absorbance values and the UV spectrophotometer limit to the absorbance measurements, as shown in Figure 3(b). For these measurements, absorbances were noted for the stock solution (solution as obtained from the source) and then subsequently stepwise serial log dilutions were made and absorbances were noted for each. These measurements were repeated three times. It can be seen in Figure 3(b) that a linear range exists between 1 and 1000 ppm. The UV spectrophotometer limit appears to be 1 ppm. It should be noted that the commercial PHMB solution obtained from the provider is considered to be 10⁶ppm. This is calculated by considering that 0.01% of the procured PHMB solution in liquor is said to be 100 ppm by the provider.¹³

Figure 3.

(a) Calibration of various PHMB concentrations against absorbance at 236 nm. (b) Absorbance measurements of different dilutions of PHMB at 236 nm with a UV Vis spectrophotometer. The stock solution is considered to be 10⁶ppm or 6 (log ppm). The solution is serially diluted at each step (by 10 times). The values are an average of three readings.

The results of the release kinetics study showed that at an LCR of 20:1, in 100 min 23.5 ppm was released into the water from a fabric treated with 2% PHMB o.w.f., as shown in Figure 4. The amount of PHMB released, as shown in this figure, was calculated from the absorbance measurements obtained and using the equation obtained from the calibration curve in Figure 3(a).

Figure 4.

Release with time of PHMB from the PHMB treated textile of 2% o.w.f. in the water.

The graph shows that at close to static conditions the commercial formulation of PHMB releases quickly into the water in which the textile is placed. The release of about 10 ppm occurs within 5 min and this concentration is known to be much above the minimum inhibitory concentration (MIC) of this commercial formulation of PHMB against E. coli, which is 1 ppm.¹³ (A minimum concentration of the agent required to inhibit the growth or to give a bacteriostatic effect is known as the minimum inhibitory concentration (MIC) and the minimum concentration of the agent required to be bactericidal is known as the minimum biocidal concentration (MBC).¹⁷) The release study of PHMB indicates that in the case of laundered items of PHMB-treated textiles the washed liquor may contain the agent at concentrations above the MIC of certain microorganisms. However, this leaching phenomenon or disassociation of the agent from the carboxylate groups in the textile goes hand in hand with the agent’s efficiency as an antibacterial agent. It was found that when a reactive dye was used to bind the polyhexamethylene biguanide covalently to the textile, the release was impeded and its antibacterial efficiency reduced. This was due to its strong ionic linkages with the reactive dyes.¹⁸ In order to inhibit this leaching, certain other methods to enhance binding to the textile have also been mentioned, such as the modification of polyhexamethylene biguanide with glycidyl methacrylate, where the antibacterial activity is said not to have been compromised.¹⁹ However, details of the antibacterial tests are not given in the later paper. Washing trials were not done in this study, but this agent is known to show consistent antibacterial activity even after several washing cycles.^7–9 This could be because the leaching rate of the agent is low enough to allow the functionality to last the required washing cycles at the concentrations applied in these tests. Further investigations are needed to explore methods to bind polyhexamethylene biguanide covalently to cotton without having to compromise the antibacterial activity.

Antibacterial activity tests

Figure 5(a) shows the antibacterial activity of a PHMB -treated textile. The graph shows the number of bacteria eluted for particular concentrations of the agent and incubation times. Each measuring point is an average of three samples tested, that is the CFU eluted from each sample was converted to log scale and then the results of the three such samples were averaged.

Figure 5.

(a) Amount of bacteria recovered from samples (log CFU) against different concentrations of applied PHMB (% o.w.f.). (b) Amount of bacteria recovered from samples (log CFU) against time. Data presented here are the same as those presented in Figure 5(a).

The results show that at higher concentrations (1.6% and 2%) of the agent on the textile, no viable bacteria were recovered from the sample (therefore showing bactericidal action). At lower concentrations, it can be noted that the trend of the amount of bacteria eluted appears to be mixed. On the untreated or control sample, the bacteria appear to grow by a factor of 100 over 6 hours and then stabilize. Since the inoculum contained Luria Broth medium and the samples were placed inside agar plates that were sealed, this stabilizing of the growth could not be due to lack of nutrition or moisture, but rather due to the characteristic growth curve of E. coli. In general, bacteria are known to show a steady stationary growth phase after an initial lag phase and growth curve.

Figure 5(b) shows the same data more clearly in relation to time. It is noted that at lower concentrations (0.4%, 0.8%, and 1.2%), a bacteriostatic effect occurs and a regrowth of bacteria is seen. This regrowth is known to occur when the agent concentration falls below the lethal limit (MIC).²⁰ It is known that cells treated with bacteriostatic levels of polyhexamethylene biguanide can recover despite having lost up to 40% of the K⁺ in their cells. Above certain concentrations of the agent, recovery is not seen even when the agent has been removed from the vicinity and this is irrespective of the time of recovery.²¹ At higher antibacterial agent concentrations, nucleotide leakage occurs, and with further increases in concentration the precipitation of cell contents occurs; in other words, there is permanent damage.¹³

Eosin staining

The reflectance curves for eosin-stained PHMB samples showed that there was a significant difference in the reflectance curve between untreated and treated samples, as seen in Figure 6. The values of the K/S measurements of stained untreated and unstained untreated samples are shown in Table 1 (each value is an average of three readings). From Table 1, it is clear that the K/S measurement could be easily used to distinguish between treated and untreated fabrics. Figure 7 shows the K/S values of the stained PHMB-treated samples after the K/S values of the stained untreated samples have been deducted from it. The difference between the K/S values of the treated samples used in this study was minimal. However, the K/S values of the PHMB-treated samples could be used to note if any agent was left on the fabric after washing.

Figure 6.

Reflectance curves obtained from X-Rite spectrophotometry of eosin stained PHMB treated (% o.w.f.) and untreated cotton.

Figure 7.

Difference in the K/S values between the Eosin Y stained control and PHMB treated fabrics.

Table 1.

K/S values for the Eosin Y stained control and PHMB treated fabrics. The values are an average of three readings

Concentration o.w.f. (%) applied	K/S	Standard deviation
0	7.62	1.89
0.4	14.30	0.23
0.8	14.93	0.42
1.2	15.14	1.74
1.6	14.35	1.35
2.0	14.34	2.08

Conclusion

Reputex 20, a commercially available polyhexamethylene biguanide antimicrobial agent for textile finishing, was applied to cotton textiles at various concentrations. The release of this agent from the textile submerged in water was studied. The release of the commercial polyhexamethylene biguanide from a fabric at close to static conditions showed that within some minutes a concentration above the MIC of certain microorganisms such as E. coli is released into the liquor. A time survivor study was conducted with E. coli bacterium. The agent behaved as a bacteriostat at lower concentrations of application. At higher concentrations (1.6–2% o.w.f.), a biocidal or bactericidal action was noted where no viable bacteria were recovered from the textile.

Footnotes

Acknowledgments

For the fruition of this study, we are greatly indebted to Audrey Tourrette (University of Toulouse), Miha Lavric, Wendy Dankers, and Mark Poels (Biomaterial Science and Technology Group), Anouk Leusink and Hugo Alves (Tissue Regeneration Group, University of Twente), Kirsten Lijenhorst-Groener (NanoBioPhysics Group, University of Twente), Christopher Niel Chadwick (Lonza group Ltd (previously Arch chemicals)), Brigita Tomsic (University of Ljubljana), Henk Gooijer (TKT), and the EFSM foundation.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Simoncic

Tomsic

. Structures of novel antimicrobial agents for textiles – a review. Text Res J 2010; 80(16): 1721–1737.

Block

. Disinfection, sterilization, and preservation, Philadelphia: Lippincott Williams & Wilkins, 2001.

Blackburn

Harvey

Kettle

. Sorption of poly (hexamethylenebiguanide) on cellulose: Mechanism of binding and molecular recognition. Langmuir 2006; 22(13): 5636–5644.

Gilbert

Moore

. Cationic antiseptics: diversity of action under a common epithet. J Appl Microbiol 2005; 99(4): 703–715.

Hugo

Denyer

. The concentration exponent of disinfectants and preservatives (biocides). Soc Appl Bacteriol Tech Ser 1987; 22: 281–291.

Russell

McDonnell

. Concentration: a major factor in studying biocidal action. J. Hosp Infect 2000; 44(1): 1–3.

Wallace

. Testing the efficacy of polyhexamethylene biguanide as antimicrobial treatment for cotton fabric. AATCC Rev 2001; 1(11): 18–20.

Chen-Yu

Eberhardt

Kincade

. Antibacterial and laundering properties of AMS and PHMB as finishing agents on fabric for health care workers' uniforms. Clothing Text Res J 2007; 25(3): 258–272.

Payne

Kudner

. A new durable antimicrobial finish for cotton textiles. Am Dyest Rep 1996; 85(6): 26–30.

10.

Chadeau

Brunon

Degraeve

. Evaluation of antimicrobial activity of a polyhexamethylene biguanide-coated textile monitoring both bacterial growth (ISO 20743/2005 standard) and viability (live/dead backlight kit). J Food Safety 2012; 32(2): 141–151.

11.

LGC standards. Available at: http://www.lgcstandards-atcc.org (accessed 1 November 2012).

12.

De Paula

Netto

Mattoso

LHC

. Physical and chemical characterization of poly (hexamethylene biguanide) hydrochloride. Polymers 2011; 3(2): 928–941.

13.

Technical brief. Lonza Group Ltd (Arch chemicals), 2011.

14.

JIS (Japanese Industrial Standard) L1902 (2002) Absorption method.

15.

Huth

Powell

. Methods, compositions and instruments to predict antimicrobial or preservative activity (2003). Google Patents.

16.

Wilhelms T. et al., Release of polyhexamethylene biguanidine hydrochloride (PHMB) from a hydrobalanced cellulose wound dressing with PHMB. In European Wound Management Association Conference, 2007; 7: 113.

17.

Pankey

Sabath

. Clinical relevance of bacteriostatic versus bactericidal mechanisms of action in the treatment of Gram-positive bacterial infections. Clin Infect Dis 2004; 38(6): 864–870.

18.

Kawabata

Taylor

. The effect of reactive dyes upon the uptake and antibacterial efficacy of poly(hexamethylene biguanide) on cotton. Part 3: Reduction in the antibacterial efficacy of poly (hexamethylene biguanide) on cotton, dyed with bis (monochlorotriazinyl) reactive dyes. Carbohyd Polym 2007; 67(3): 375–389.

19.

Wei

Zhou

Guan

. Investigation on the reaction between polyhexamethylene guanidine hydrochloride oligomer and glycidyl methacrylate. J Appl Polym Sci 2012; 127(5): 666–674.

20.

. A review of microbial injury and recovery methods in food. Food microbiol 2008; 25(6): 735–744.

21.

Broxton

Woodcock

. and Gilbert P. Injury and recovery of Escherichia coli ATCC 8739 from treatment with some polyhexamethylene biguanides. Microbios 1984; 40(161–162): 187–193.