- Email: [email protected]
Biocidal polymers (II): Determination of biological activity of novel N-halamine biocidal polymers and evaluation for use in water filters
Reactive & Functional Polymers 68 (2008) 1448–1458
Contents lists available at ScienceDirect
Reactive & Functional Polymers journal homepage: www.elsevier.com/locate/react
Biocidal polymers (II): Determination of biological activity of novel N-halamine biocidal polymers and evaluation for use in water filters Abd El-Shafey I. Ahmed a,*, John N. Hay a, Michael E. Bushell b, John N. Wardell b, Gabriel Cavalli a a b
Chemical Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK Microbial Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK
a r t i c l e
i n f o
Article history: Received 11 April 2008 Received in revised form 15 May 2008 Accepted 14 June 2008 Available online 8 July 2008
Key words: Biocidal polymer N-Halamine Halogens Polyurethane Water filters
a b s t r a c t Novel N-halamine biocidal polymers were prepared by co-polymerizing a heterocyclic ring-based monomer with tolylene-2,6-diisocyanate and toluene-2,4-diisocyanate. The resulting polyurethanes were halogenated; chlorinated, brominated or iodinated. The rate of bacterial killing by the halogenated derivatives was determined both with and without halogen quenching and one of them was evaluated for use in water filters. The effect of these polymers on bacterial growth-rates was also determined. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction N-halamine polymers are an important class of biocidal polymers [1–16]. This type of polymer is prepared by introducing a heterocyclic ring containing amino, amide or imide groups into the polymer structure followed by halogenation to the corresponding N-halamines, which confers on the polymer its biological activity [1–16]. Biocidal activity is modulated by halogen stability on the polymer [1–16]; halogenated amines are more stable than amides and imides [8]. In comparison, halogenated imides exhibit the lowest stability but show the most powerful biocidal activity [8,16]. In this work the prepared heterocyclic ring contains imide groups. The N-halogen bond has been stabilized by introducing electron donating groups on the ring; however, a high level of biocidal activity is still apparent [16]. Halogen loading of these poly* Corresponding author. Tel.: +44 (0) 1483 686850; fax: +44 (0) 1483 686851. E-mail address: [email protected] (A.E.-S.I. Ahmed). 1381-5148/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.reactfunctpolym.2008.06.021
mers was increased with respect to other examples in the literature by choosing a heterocyclic ring (uramil) (1), Scheme 2, [16] that can be charged with a maximum of three halogens per unit. After polymerization, the number of the available positions for halogenation was increased to five (Scheme 2) [16] in comparison with the two or three available positions in those similar polymers currently available [1–9,13,14] (examples in Scheme 1d and e). We previously reported polymers prepared using uramil that showed good biological activity and good stability [16]. These polymers were prepared by reacting polyacrylonitrile and polyethylacrylate with uramil (Scheme 1a and b) but the number of available positions for halogens was lower than for those polymers now described. In addition, uramil-derived poly-urea was also prepared (Scheme 1c) [16], the number of positions available for halogens is 7 per repeating unit, and this will be evaluated in future work. In this work we focus on a uramil-derived polyurethane (4) which was halogenated with Cl, Br or I and the most appropriate selected for different applications (e.g., drinking
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
1449
Scheme 1. Different types of N-halamine biocidal polymer.
water filters and sterilization). An analogous uramil-derived polyurethane (8) was chlorinated to compare its biological activity with the chlorinated form of (4). The bacterial killing power of each halogenated derivative, both with and without halogen quenching, was evaluated for cultures of Gram-positive (Staphylococcus aureus) and Gram-negative (Escherichia coli) bacteria. The effect of the non-halogenated polymer (4) on these bacteria was also examined. 2. Experimental 2.1. Materials Barbituric acid, granulated tin, resorcinol, fuming nitric acid, sodium nitrite, toluene-2,4-diisocyanate, tolylene2,6-diisocyanate, bromine and iodine were supplied by Sigma Aldrich Chemicals, UK. Sodium hydroxide, hydrochloric acid, potassium permanganate, sulphuric acid, sodium thiosulfate and dimethylformamide were supplied by Fisher Chemicals, UK. Nutrient broth and Nutrient agar (Oxoid). 2.2. Preparation of polymers The polymers under investigation were prepared according to the methodology reported earlier [16], as follows: 2.2.1. Diazotization of uramil Uramil [17] (1) (5-aminobarbituric acid) (1.40 g, 0.01 mol) was dissolved in 5 ml concentrated sulphuric
acid. The temperature was kept at 0 °C using an external ice bath. A cold solution of NaNO2 [0.69 g of NaNO2 (0.01 mol) + 10 ml water] was added drop-wise to the uramil solution with stirring to form the uramil diazonium salt (2) [16], Scheme 2. 2.2.2. Preparation of 1,3-dihydroxy-4(5-azobarbituric acid)benzene (3) Resorcinol (1.1 g, 0.01 mol) and NaOH (5.5 g, 0.14 mol) were dissolved in 20 ml water and added gradually to cold uramil diazonium salt (2). The dark purple product that precipitated was filtered, washed copiously with cold water, dried and weighed, producing 2.6 g (99% yield), Scheme 2 [16]. Analysis, FTIR (KBr): m (cm 1) 1603, 1705, 1411, 3100, 3432 and 2942. 1H NMR (DMSO, 500 MHz): d 1.3 (s, 1H), 5.4 (s, 1H), 6.2 (s, 2H), 6.9–7.2 (s, 3H) and 10.2 (s, 1H). 13 C NMR (DMSO, 125 MHz): ppm 49, 102.4, 103, 105, 106, 129, 150.3 and 158.3. Elemental analysis, found (%): C, 45.1; H, 2.9; N, 20.9. Calculated for C10H8N4O5 (%): C, 45.5; H, 3; N, 21.2 [16]. 2.2.3. General procedure for the polyurethane polymers (4) and (8) Monomer (3) (2.6 g, 0.01 mol) and a suitable diisocyanate (0.01 mol) were heated in 30 ml dimethylformamide for 5 h at 90 °C. The reaction was cooled and 50 ml of methanol added. The brown product was filtered, washed copiously with methanol, dried and weighed, Scheme 2 [16]. Poly[(1,3-dihydroxy-4(5-azobarbituric acid)-benzene)co-(tolylene-2,6-diisocyanate)] (4) was prepared using
1450
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Scheme 2. Polymers preparation and their halogenation.
tolylene-2,6-diisocyanate while poly[1,3-dihydroxy-4(5azobarbituric acid)benzene)-co-(tolulene-2,4-diisocyanate)] (8) was prepared using tolulene-2,4-diisocyanate, Scheme 2 [16]. Analysis of poly[(1,3-dihydroxy-4(5-azobarbituric acid)-benzene)-co-(tolylene-2,6-diisocyanate)] (4), FTIR (KBr): m (cm 1) 1640, 1700, 1660, 3429, 1135, 1471 and 2920. 1H NMR (DMSO, 500 MHz): d 2.2 (s, 3H), 4.8 (s, 1H), 4.2 (s, 1H), 6.8 (s, 2H), 10.5 (s, 1H) and 7.0–8.4 (s, 6H). 13C NMR (DMSO, 125 MHz): ppm 11.4, 49, 109.9, 111.6, 113.2, 116 117.9, 118.5, 120, 121, 125, 137.3, 137.7, 146, 150, 185, 163.0 and 153.2. Elemental analysis, found (%): C, 51.7; H, 3.2; N, 18.1. Calculated for C19H16N8O7 (%): C, 52.1; H, 3.2; N, 19.2 [16]. Analysis of poly[1,3-dihydroxy-4(5-azobarbituric acid)benzene)-co-(tolulene-2,4-diisocyanate)] (8), FTIR (KBr): m (cm 1) 1617, 1639, 1712, 3417, 3458, 3550, 1110, 1457.
1
H NMR (DMSO, 500 MHz): d 2.0 (s, 3H), 6.4 (s, 1H), 7.1 (s, 1H), 6.9 (s, 2H), 7.4 - 8.5 (m, 6H), and 9.4 (s, 1H). 13C NMR (DMSO, 125 MHz): ppm 12, 49, 106, 110, 113, 114, 115, 116, 119, 119.6, 120, 121, 123, 126, 128, 151, 154 and 160. Elemental analysis, found (%): C, 51.8; H, 3.2; N, 18.5. Calculated for C19H16N8O7 (%): C, 52.1; H, 3.2; N, 19.2 [16].
2.2.4. General procedure for halogenation Polyurethane (4) (0.44 g, 0.001 mol) was suspended in sodium hydroxide (0.20 g, 0.005 mol) solution and the halogen (chlorine, bromine or iodine) was added gradually until neutralization to pH 7. The mixture was stirred for 1 h during which the temperature was kept below 5 °C using an external ice bath. The resulting product was filtered, washed copiously with chlorine-free water, dried and weighed, Scheme 2 [16]. Polyurethane (8) (0.44 g, 0.001 mol) was chlorinated only in the same way, Scheme 2 [16].
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
2.3. Effect of the halogenated polymers on bacterial growth and viability A culture of E. coli was prepared by inoculating one bacterial colony into 20 ml of nutrient broth in a Universal bottle and incubating for 24 h at 37 °C. From the bacterial suspension, 0.1 ml was transferred to a 20 ml Universal bottle containing 10 ml of fresh medium. Six Universals were prepared; three used in testing the effect of the polymer on bacterial growth and the other three to test the effect of the polymer on the viability of the bacteria. To study the effect of the polymer on growth rate of E. coli, 0.5 g of the halogenated polymer was added to the first bottle of broth while 0.5 g of the control polymer (non-halogenated) was added to the second bottle, to act as a polymer control, and the third was left as a bacterial control without polymer. The three bottles were stirred at 37 °C and sampled at timed intervals for viable count, employing the ‘Miles and Misra’ technique [18]. To investigate the effect of the polymer on the viability of E. coli, the other three bottles, inoculated as above, were incubated for 17 h at 37 °C, and the number of bacteria determined by viable count. At this time 0.5 g of the halogenated polymer was added to one bottle; 0.5 g of the control polymer (non-halogenated) was added to the second, to act as a polymer control, and the third vessel was left as a bacterial control. The three bottles were stirred at room temperature, and samples from each culture taken for viable count at regular time intervals, as previously. The procedure was repeated to test the effect of the halogenated polymers on a Gram-positive bacterium (S. aureus). 2.4. Effect of the halogenated polymers on bacterial viability under halogen quenching The previous experiment, investigating the effect of the polymer on the bacterial viability, was repeated. During the viable counts 0.05 ml of 0.5 M sodium thiosulphate was added to each decimal dilution to quench any halogen which may have evolved during the reaction between the polymer and the bacteria. 2.5. Effect of the non-halogenated polymer on the liquid medium Non-halogenated polymer, 0.5 g (non-halogenated 4) was added to each of two Universal bottles each containing 10 ml of sterile liquid medium. One of these was stirred at ambient temperature and the other was stirred at 37 °C for 17 h. The polymer was allowed to settle in each vessel and 5 ml of the overlaying broth removed to a fresh, sterile, Universal bottle. Bacterial suspension, 0.05 ml (either E. coli or S. aureus prepared as described above) was added as inoculum and the growth of the cultures followed by viable count during incubation at 37 °C. Similarly, 5 ml of sterile liquid medium was inoculated as a bacterial control and a viable count performed at the same time intervals.
1451
2.6. Determination of the amount of halogen released from the chlorinated polymer during contact with water, nutrient broth and bacterial medium Chlorinated polymer (5), 0.5 g, was stirred with 10 ml of chlorine-free water. The polymer was filtered, dried and the degree of halogenation of the polymer was determined before and after the experiment using iodometric titration in order to determine the amount of halogen released from the polymer to the water [9,16]. The experiment was repeated using nutrient broth and bacterial suspensions (both E. coli and S. aureus) instead of chlorine-free water and in each case the amount of delivered halogen calculated. 2.7. Evaluation of polymer (5) (chlorinated) in a column water filter Polymer (5), 1.0 g, was packed loosely in a glass tube (15 cm length and 1 cm diameter) to a height of 4 cm. Bacterial suspension [prepared by inoculating one bacterial colony in 20 ml of liquid medium (Nutrient Broth, Oxoid) and incubated for 17 h at 37 °C] was passed through the column and the output from the column recycled through it again. Before recycling, 0.1 ml from the passed liquid was sampled for viable count. Five cycles were performed for each column. Two columns were prepared containing the original polymer (non-chlorinated form of 5) which acted as a control, one for S. aureus and the other for E. coli. A further two columns were prepared for the N-halamine polymer (5), one for S. aureus and the other for E. coli, therefore four columns in total. The number of viable cells in the original bacterial suspensions was determined [3] and the turbidity of the liquid before and after passing through the columns also determined spectrophotometrically at 540 nm [10–12,15].
3. Results and discussion The biological activity and mode of action of a new synthesized N-halamine biocidal polymer was evaluated. The effect of this N-halamine polymer on bacterial growth was examined and one of its halogen derivatives was evaluated for use in water filters. 3.1. Determination of the effect of the halogenated polymers on bacterial viability and growth rate 3.1.1. Growth rate No bacterial growth was recorded in presence of any of the halogenated polymers (5–7). Fig. 1a shows S. aureus does not grow in the presence of the chlorinated polymer (5) or in the presence of the non-halogenated polymer (4) (polymer control); which, as S. aureus is not motile, may be due its adsorption onto the polymer surface. From Fig. 1c, E. coli also did not grow in the presence of the chlorinated polymer. However, unlike S. aureus, it grew in the presence of the control polymer (non-halogenated polymer) but at a different rate and to a lower final population than the E. coli ‘bacterial
1452
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Fig. 1. (a) and (c) Log plot of viable counts (cfu/ml) vs time of S. aureus and E. coli, respectively, grown in the presence of the chlorinated polymer (5), the control polymer (4) and without polymer (bacterial control). (b) and (d). Log plot of viable counts (cfu/ml) vs time of S. aureus and E. coli, respectively, in nutrient broth after stirring with the chlorinated polymer (5), control polymer (4) and without polymer (bacterial control).Where T = Test (bacteria + chlorinated polymer). BC = Bacterial control (no treatment). PC = Polymer control (bacteria + non-halogenated polymer).
control’. This may be due to differences in the motility and surface composition of the two types of bacteria. The same behaviour was recorded for the brominated polymer (6) (Fig. 3a and c) and for the iodinated polymer (7) (Figs. 4a and c). As shown in Figs. 1a and c the non-halogenated polymer (4) inhibited the growth of S. aureus and had some limited effect on the growth of E. coli. A plausible explanation may be the removal of critical nutrient broth components by the polymer through adsorption onto its surface. To investigate this possibility we treated fresh broth with polymer (4) at different temperatures, removing the polymer by allowing it to settle, and then removing and using the overlying liquor. The isolated medium was inoculated with a fresh culture of S. aureus and the bacterial growth followed by viable counts. The results (Figs. 2a and b) show that treating the broth with non-halogenated polymer (4) adversely affects the rate of bacterial growth, even after removal of the polymer (although to a lesser extent than when the polymer is still present). This suggests that there may be a dual mode of action, one by direct contact polymer–bacteria (possibly there is significant bacterial adsorption in the case of S. aureus) and another by affecting the broth composition (through nutrient adsorption onto the polymer).
3.1.2. Viability The chlorinated polymer (5) achieved a 3 log reduction in the bacterial population in 7 min while no bacterial colonies were detected after 15 min (equivalent to a 9 log reduction) in the case of S. aureus, Fig. 1b. A similar behaviour was recorded with E. coli, a 3 log reduction was recorded in the first 7 min while no colonies were detected after 15 min (equivalent to a 10 log reduction), Fig. 1d. The brominated polymer (6) behaves similarly to the chlorinated polymer (5) for E. coli (Fig. 3b), and also for S. aureus but with a 9 log reduction in 15 min (Fig. 3d). The iodinated polymer showed the highest power of sterilization (Fig. 4b) showing a 10 log reduction in the population of E. coli after 7 min following contact between the iodinated polymer (7) and the bacteria. S. aureus behaves similarly; no colonies were detected after 7 min contact between the iodinated polymer (7) and the bacteria (Fig. 4d). To determine the rate of killing by the iodinated polymer (7) the amount of polymer in contact with the bacteria was reduced; the experiment was repeated, but using 0.25 g of polymer in contact with 10 ml of bacterial suspension and the time intervals were reduced to detect any viable bacterial colonies early in the culture. For E. coli, a 5 log reduction in the bacterial population was achieved in 1 min and no viable colonies detected (equivalent to 10
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
log reduction) after 5 min contact time. The results for S. aureus were unequivocal, the rate of killing by the iodinated polymer could not be determined due to the powerful effect of the polymer; no viable colonies were detected after 1 min contact time (equivalent to 9 log reduction) (Figs. 4e and f, respectively).
1453
3.2. Killing rate determination with halogen quenching Most of the research surrounding the N-halamine polymers states that the mode of action of these polymers depends only on the contact between the polymer particles and the bacterial cells [1–15]. However, we suggest that
Fig. 2. (a) and (b) Log plot of viable count (cfu/ml) vs time of S. aureus and E. coli, respectively, grown in liquid medium previously stirred with the nonhalogenated polymer (4). Where Growth of S. aureus (S25) E. coli (E25) in medium pre-treated with the non-halogenated polymer at 25 °C (or, S37 and E37 at 37 °C). Sc and EC are the respective control grown in untreated nutrient broth.
Fig. 3. (a) and (c) Log plot of E. coli and S. aureus viable count (cfu/ml) vs time grown in the presence of the brominated polymer (6), the control polymer (4) and without treatment (bacterial control). (b) and (d) Log no. of colonies (cfu/ml) vs time of E. coli and S. aureus, after exposure to the brominated polymer (6), the non-halogenated polymer (4) and without treatment (bacterial control). Where T = Cells treated with brominated polymer. PC = Cells treated with the control polymer (non-halogenated polymer). BC = Bacterial control (no treatment). BC and PC are superimposed in Fig. 3b.
1454
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Fig. 4. (a) and (c) Log no. of E. coli and S. aureus viable count (cfu/ml) vs time during growth in nutrient broth; in the presence of the iodinated polymer (7), the non-halogenated polymer (4) and without treatment (bacterial control). (b) and (d) Log no. of E. coli and S. aureus, respectively, vs time after treating the bacterial cells with; the iodinated polymer (7) (0.5 g/10 ml bacterial suspension) the non-halogenated polymer and without treatment (bacterial control). (e) and (f) Log no. of E. coli and S. aureus, respectively, at timed intervals after stirring the bacterial cells with; iodinated polymer (7) (0.25 g/10 ml bacterial suspension) and without treatment (bacterial control). Where T = Cells treated with the iodinated polymer. PC = Polymer control, cells treated with nonhalogenated polymer. BC = Bacterial control, untreated cells.
the mechanism of killing, by halogenated polymers can also occur through the release into the medium of soluble halogen species; in addition to contact between the bacteria and polymers. To investigate this, the effect of adding a halogen quencher (sodium thiosulphate) to samples for viable counts immediately prior to counting was examined. Mixing was stopped and the polymer granules allowed to set-
tle before samples of the overlying bacterial suspension were taken for viable count with quenching as described earlier. Any differences in the observed killing rate would reflect a release of active species during mixing of polymer and bacteria. The chlorinated polymer (5) achieved a 2 log reduction in the E. coli population in 40 min and no viable colonies were detected after 1.5 h (9 log reduction). Similarly the
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
chlorinated polymer achieved a 1 log reduction in 40 min for the S. aureus population and no bacterial colonies were detected after 1.5hr (9 log reduction) (Figs. 5a and b, respectively). The brominated polymer (6) showed greater activity, achieving a 4 log reduction in the E. coli population in 40 min and no viable colonies detected after 1.5 h (9 log reduction) while it achieved a 4 log reduction in the S. aur-
1455
eus population in 15 min and no viable colonies detected after 40 min (9 log reduction) (Figs. 5c and d, respectively). For the Iodinated polymer (7), the experiment was performed with a 1 g:40 ml ratio between the polymer weight and the bacterial suspensions due to the high biocidal power of the iodinated polymer. In spite of this reduced quantity of polymer, no E. coli or S. aureus colonies were detected at 7 min (9 log reduction) (Figs. 5e and f, respectively).
Fig. 5. (a) and (b) Effect of the chlorinated polymer (5) on the viability of E. coli and S. aureus, respectively, under chlorine quenching. (c) and (d) Effect of the brominated polymer (6) on the viability of E. coli and S. aureus, respectively, under bromine quenching. (e) and (f) Effect of the iodinated polymer (7) on the viability of E. coli and S. aureus, respectively, under iodine quenching. Where T = Treated cells with the halogenated polymer. BC = Untreated cells (bacterial control).
1456
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
From the previous results (Figs. 1–5) it was clear that the most powerful biocidal effect is exhibited by the iodinated polymer. Without halogen quenching the rate of killing of the chlorinated and brominated polymers are very similar, while with halogen quenching, the brominated polymer shows a more powerful effect on each type of bacteria. This may be related to the stability of the halogen on the polymer; I–N bonds have the lowest stability therefore
the halogen can be easily exchanged between the polymer and bacteria. The Cl–N exhibits the lowest biological power but is the most stable bond; hence its use in water filters. The difference in the killing rate of the polymer with and without halogen quenching (a lower rate with quenching) indicates that halogen species (released from the polymers) are involved in the biocidal action, and suggesting the mechanism of killing is not by contact alone [1–15].
Fig. 6. (a) and (b) Effect of chlorinated polymer (9) on the viability of E. coli and S. aureus, respectively, under chlorine quenching. Where T = Cells treated with the chlorinated polymer. BC = Cells without treatment (bacterial control).
Fig. 7. (a) and (b) Log viable count (cfu/ml) of S. aureus and E. coli recovered from the eluate after each passage through the column containing control polymer. (c) and (d) Log viable counts (cfu/ml) of S. aureus and E. coli, respectively, after each cycle through the column containing chlorinated polymer (5).
1457
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
This dual mechanism of killing (contact + release of halogen species) is currently under investigation and will be reported in due course. As stated earlier there is also a possibility of halogen exchange between the polymer and the protein in the broth medium. This latter possibility can be examined by calculating the amount of released halogen in different liquid media. 3.3. Amount of released chlorine To confirm the presence of halogen species released from the polymer, the amount of chlorine delivered from the polymer to the bacteria was determined by iodometric titration of the halogenated polymer isolated after controlled contact with water, broth medium and the bacterial suspensions [9,16]. The amount of released chlorine was determined as the difference between original chlorine content in the polymer and chlorine content after contact with the different media, the ratio between the weight of polymer and the liquid was 1 g: 20 ml The amounts of delivered halogen ions to water, nutrient broth, E. coli (1.7 109 cfu/ml) in broth medium and S. aureus (3.1 108 cfu/ml) in broth medium were 0.19 ppm ± 0.05, 3.9 ppm ± 0.12, 8.5 ppm ± 0.32 and 5.9 ppm ± 0.24, respectively. The amount of delivered chlorine to water is very low but increased for Nutrient broth, possibly through chlorine exchange between the polymer and protein in the broth medium, which supports the possibility of bacterial death due to changing the nature of the nutrients in the medium, in addition to the release effect and a contact effect. In the presence of bacteria the amount of delivered chlorine increased indicating that more chlorine is released from the polymer when in contact with bacteria. The mechanism or the mode of action of this process is under further investigation but from these results we suggest that the mode of action of this kind of polymer is a combination of factors. 3.4. Determination of the killing rate of the chlorinated polymer (9) under quenching conditions and in comparison with chlorinated polymer (5) This experiment was performed to investigate the suitability of using the alternative commercial, but low purity, toluene-2,4-diisocyante instead of tolylene-2,6-diisocyante. From Fig. 6a, chlorinated polymer (9) achieved a 6 log reduction in the E. coli population in 40 min and no bacterial colonies were detected after 40 min (equivalent to 9 log reduction). For S. aureus, Fig. 6b, the chlorinated polymer (9) achieved a 6 log reduction in 40 min and no colonies detected after 90 min (equivalent to 9 log reduction). From Figs. 6a and b it was clear that using toluene-2,4-diisocyanate (polymer 9) instead of tolylene-2,6-diisocyante (polymer 5) gave very similar results and that the commercial product was equally effective. 3.5. Chlorinated polymer (5) evaluation in water filters This experiment investigated the potential to use one of these halogenated polymers (5) in water filters on a labo-
Table 1 Absorbance (at 540 nm, read against a nutrient broth blank) and viable counts before and after perfusing columns of non-halogenated and halogenated polymers Non-halogenated polymer (control) E. coli Before perfusing the column (viable counts (cfu/ml)) After the fifth cycle (viable counts (cfu/ml)) After incubating the fifth cycle for 24 h (viable counts (cfu/ml))
0.41 9
(2.6 10 ) 0.30 (8.4 108) 1.33
(2.5 109)
Halogenated polymer
S. aureus
E. coli
0.30
0.27 7
(8.5 10 ) 0.04
S. aureus 0.32 9
(1.8 10 ) 0.02
(3.1 108) 0.01
(Nd)
(Nd)
(Nd)
0.46
0.00
0.03
(7.5 107)
(Nd)
(Nd)
Nd = Non-detected.
ratory scale. Figs. 7a and b show the effect of the control polymer (4) contained in a 4 cm length 1 cm diameter column on suspensions of S. aureus and E. coli. After 3 cycles no viable colonies of S. aureus were detected in the eluate (Fig. 7a). Whereas for E. coli, although there was an initial reduction in the population, this recovered and there was no significant reduction overall in the bacterial population over the 5 cycles. This suggests that the control polymer is merely acting as a filter, allowing passage of the motile E. coli but not exhibiting any biocidal action. Results from columns containing the chlorinated polymer (5) are shown in Figs. 7c and d and Table 1. After one passage through the column the population of both types of bacteria was reduced to a non-detectable level (equivalent to an 8 log reduction). This was also reflected in the spectrophotometric measurements (Table 1); the negative values after 5 cycles are due to a bleaching effect on the broth medium. Further incubation of the samples post-passage confirmed the biocidal effect and demonstrates the potential application of the polymer in water filters. 4. Conclusions The rate of bacterial killing of new halamine polymers was examined and the effect of these polymers on bacterial growth determined. No bacterial growth was recorded in the presence of any of the halogenated polymers. The iodinated polymer showed greater biocidal power than chlorinated and brominated polymers. Investigations performed with and without halogen quenching indicate that there is a difference in killing-rate, suggesting that the mode of action of these polymers is dual; proceeding through both release of halogen species into the medium and through bacteria–polymer contact. In addition, a third possibility exists – changing the nature of nutrients around the bacteria through interactions between the halogenated polymer and medium proteins. The non-halogenated polymer also showed an inhibitory effect on growth of S. aureus. One of
1458
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
the prepared N-halamine polymers (5) (chlorinated) was used successfully in a laboratory column for sterilizing a bacterial culture. Acknowledgment The project is sponsored by the Egyptian government. References [1] G. Sun, L.C. Allen, E.P. Luckie, W.B. Wheatly, S.D. Worley, Ind. Eng. Chem. Res. 34 (1994) 4106. [2] G. Sun, W.B. Wheatly, S.D. Worley, Ind. Eng. Chem. Res. 33 (1994) 168. [3] G. Sun, T.Y. Chen, W. Sun, W.B. Wheatley, S.D. Worley, J. Bioact. Compat. Polym. 10 (1995) 135. [4] Y. Chen, L. Wang, H. Yu, Q. Shi, X. Dong, J. Mater. Sci. 42 (2007) 4018. [5] Y. Chen, S.D. Worley, T.S. Huang, J. Weese, J. Kim, C-I. Wei, J.F. Williams, J. Appl. Polym. Sci. 92 (2004) 368.
[6] Y. Chen, S.D. Worley, T.S. Huang, J. Weese, J. Kim, C.-I. Wei, J.F. Williams, J. Appl. Polym. Sci. 92 (2004) 363. [7] Y. Chen, S.D. Worley, J. Kim, C.-I. Wei, T.Y. Chen, Ind. Eng. Chem. Res. 42 (2003) 280. [8] Z. Chen, J. Luo, Y. Sun, Biomaterials 28 (2007) 1597. [9] Z. Chen, Y. Sun, Ind. Eng. Chem. Res. 45 (2006) 2634. [10] A.M. El-Masry, Pigment Resin Technol. 34 (2005) 265. [11] A.M. El-Masry, H.Y. Moustafa, A.I. Ahmed, A.F. Shaaban, Pigment Resin Technol. 33 (2004) 211. [12] A.M. El-Masry, H.Y. Moustafa, A.I. Ahmed, A.F. Shaaban, Pigment Resin Technol. 33 (2004) 75. [13] K. Barnes, J. Liang, R. Wu, S.D. Worley, J. Lee, R.M. Broughton, T.S. Huang, Biomaterials 27 (2006) 4825. [14] J. Liang, J.R. Owens, T.S. Huang, S.D. Worley, J. Appl. Polym. Sci. 101 (2006) 3448. [15] H.Y. Moustafa, Pigment Resin Technol. 35 (2006) 71. [16] A.E.I. Ahmed, J.N. Hay, M.E. Bushell, J.N. Wardell, G. Cavalli, React. Funct. Polym. 68 (2008) 248. [17] S.F. Brian, J.H. Antony, W.G.S. Peter, R.T. Austin, ‘Vogel’ Text Book of Practical Organic Chemistry, fifth ed., Longman Ltd, 1989. 1170. [18] A.A. Miles, S.S. Misra, J. Hygiene (London) 38 (1938) 732.
Contents lists available at ScienceDirect
Reactive & Functional Polymers journal homepage: www.elsevier.com/locate/react
Biocidal polymers (II): Determination of biological activity of novel N-halamine biocidal polymers and evaluation for use in water filters Abd El-Shafey I. Ahmed a,*, John N. Hay a, Michael E. Bushell b, John N. Wardell b, Gabriel Cavalli a a b
Chemical Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK Microbial Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK
a r t i c l e
i n f o
Article history: Received 11 April 2008 Received in revised form 15 May 2008 Accepted 14 June 2008 Available online 8 July 2008
Key words: Biocidal polymer N-Halamine Halogens Polyurethane Water filters
a b s t r a c t Novel N-halamine biocidal polymers were prepared by co-polymerizing a heterocyclic ring-based monomer with tolylene-2,6-diisocyanate and toluene-2,4-diisocyanate. The resulting polyurethanes were halogenated; chlorinated, brominated or iodinated. The rate of bacterial killing by the halogenated derivatives was determined both with and without halogen quenching and one of them was evaluated for use in water filters. The effect of these polymers on bacterial growth-rates was also determined. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction N-halamine polymers are an important class of biocidal polymers [1–16]. This type of polymer is prepared by introducing a heterocyclic ring containing amino, amide or imide groups into the polymer structure followed by halogenation to the corresponding N-halamines, which confers on the polymer its biological activity [1–16]. Biocidal activity is modulated by halogen stability on the polymer [1–16]; halogenated amines are more stable than amides and imides [8]. In comparison, halogenated imides exhibit the lowest stability but show the most powerful biocidal activity [8,16]. In this work the prepared heterocyclic ring contains imide groups. The N-halogen bond has been stabilized by introducing electron donating groups on the ring; however, a high level of biocidal activity is still apparent [16]. Halogen loading of these poly* Corresponding author. Tel.: +44 (0) 1483 686850; fax: +44 (0) 1483 686851. E-mail address: [email protected] (A.E.-S.I. Ahmed). 1381-5148/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.reactfunctpolym.2008.06.021
mers was increased with respect to other examples in the literature by choosing a heterocyclic ring (uramil) (1), Scheme 2, [16] that can be charged with a maximum of three halogens per unit. After polymerization, the number of the available positions for halogenation was increased to five (Scheme 2) [16] in comparison with the two or three available positions in those similar polymers currently available [1–9,13,14] (examples in Scheme 1d and e). We previously reported polymers prepared using uramil that showed good biological activity and good stability [16]. These polymers were prepared by reacting polyacrylonitrile and polyethylacrylate with uramil (Scheme 1a and b) but the number of available positions for halogens was lower than for those polymers now described. In addition, uramil-derived poly-urea was also prepared (Scheme 1c) [16], the number of positions available for halogens is 7 per repeating unit, and this will be evaluated in future work. In this work we focus on a uramil-derived polyurethane (4) which was halogenated with Cl, Br or I and the most appropriate selected for different applications (e.g., drinking
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
1449
Scheme 1. Different types of N-halamine biocidal polymer.
water filters and sterilization). An analogous uramil-derived polyurethane (8) was chlorinated to compare its biological activity with the chlorinated form of (4). The bacterial killing power of each halogenated derivative, both with and without halogen quenching, was evaluated for cultures of Gram-positive (Staphylococcus aureus) and Gram-negative (Escherichia coli) bacteria. The effect of the non-halogenated polymer (4) on these bacteria was also examined. 2. Experimental 2.1. Materials Barbituric acid, granulated tin, resorcinol, fuming nitric acid, sodium nitrite, toluene-2,4-diisocyanate, tolylene2,6-diisocyanate, bromine and iodine were supplied by Sigma Aldrich Chemicals, UK. Sodium hydroxide, hydrochloric acid, potassium permanganate, sulphuric acid, sodium thiosulfate and dimethylformamide were supplied by Fisher Chemicals, UK. Nutrient broth and Nutrient agar (Oxoid). 2.2. Preparation of polymers The polymers under investigation were prepared according to the methodology reported earlier [16], as follows: 2.2.1. Diazotization of uramil Uramil [17] (1) (5-aminobarbituric acid) (1.40 g, 0.01 mol) was dissolved in 5 ml concentrated sulphuric
acid. The temperature was kept at 0 °C using an external ice bath. A cold solution of NaNO2 [0.69 g of NaNO2 (0.01 mol) + 10 ml water] was added drop-wise to the uramil solution with stirring to form the uramil diazonium salt (2) [16], Scheme 2. 2.2.2. Preparation of 1,3-dihydroxy-4(5-azobarbituric acid)benzene (3) Resorcinol (1.1 g, 0.01 mol) and NaOH (5.5 g, 0.14 mol) were dissolved in 20 ml water and added gradually to cold uramil diazonium salt (2). The dark purple product that precipitated was filtered, washed copiously with cold water, dried and weighed, producing 2.6 g (99% yield), Scheme 2 [16]. Analysis, FTIR (KBr): m (cm 1) 1603, 1705, 1411, 3100, 3432 and 2942. 1H NMR (DMSO, 500 MHz): d 1.3 (s, 1H), 5.4 (s, 1H), 6.2 (s, 2H), 6.9–7.2 (s, 3H) and 10.2 (s, 1H). 13 C NMR (DMSO, 125 MHz): ppm 49, 102.4, 103, 105, 106, 129, 150.3 and 158.3. Elemental analysis, found (%): C, 45.1; H, 2.9; N, 20.9. Calculated for C10H8N4O5 (%): C, 45.5; H, 3; N, 21.2 [16]. 2.2.3. General procedure for the polyurethane polymers (4) and (8) Monomer (3) (2.6 g, 0.01 mol) and a suitable diisocyanate (0.01 mol) were heated in 30 ml dimethylformamide for 5 h at 90 °C. The reaction was cooled and 50 ml of methanol added. The brown product was filtered, washed copiously with methanol, dried and weighed, Scheme 2 [16]. Poly[(1,3-dihydroxy-4(5-azobarbituric acid)-benzene)co-(tolylene-2,6-diisocyanate)] (4) was prepared using
1450
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Scheme 2. Polymers preparation and their halogenation.
tolylene-2,6-diisocyanate while poly[1,3-dihydroxy-4(5azobarbituric acid)benzene)-co-(tolulene-2,4-diisocyanate)] (8) was prepared using tolulene-2,4-diisocyanate, Scheme 2 [16]. Analysis of poly[(1,3-dihydroxy-4(5-azobarbituric acid)-benzene)-co-(tolylene-2,6-diisocyanate)] (4), FTIR (KBr): m (cm 1) 1640, 1700, 1660, 3429, 1135, 1471 and 2920. 1H NMR (DMSO, 500 MHz): d 2.2 (s, 3H), 4.8 (s, 1H), 4.2 (s, 1H), 6.8 (s, 2H), 10.5 (s, 1H) and 7.0–8.4 (s, 6H). 13C NMR (DMSO, 125 MHz): ppm 11.4, 49, 109.9, 111.6, 113.2, 116 117.9, 118.5, 120, 121, 125, 137.3, 137.7, 146, 150, 185, 163.0 and 153.2. Elemental analysis, found (%): C, 51.7; H, 3.2; N, 18.1. Calculated for C19H16N8O7 (%): C, 52.1; H, 3.2; N, 19.2 [16]. Analysis of poly[1,3-dihydroxy-4(5-azobarbituric acid)benzene)-co-(tolulene-2,4-diisocyanate)] (8), FTIR (KBr): m (cm 1) 1617, 1639, 1712, 3417, 3458, 3550, 1110, 1457.
1
H NMR (DMSO, 500 MHz): d 2.0 (s, 3H), 6.4 (s, 1H), 7.1 (s, 1H), 6.9 (s, 2H), 7.4 - 8.5 (m, 6H), and 9.4 (s, 1H). 13C NMR (DMSO, 125 MHz): ppm 12, 49, 106, 110, 113, 114, 115, 116, 119, 119.6, 120, 121, 123, 126, 128, 151, 154 and 160. Elemental analysis, found (%): C, 51.8; H, 3.2; N, 18.5. Calculated for C19H16N8O7 (%): C, 52.1; H, 3.2; N, 19.2 [16].
2.2.4. General procedure for halogenation Polyurethane (4) (0.44 g, 0.001 mol) was suspended in sodium hydroxide (0.20 g, 0.005 mol) solution and the halogen (chlorine, bromine or iodine) was added gradually until neutralization to pH 7. The mixture was stirred for 1 h during which the temperature was kept below 5 °C using an external ice bath. The resulting product was filtered, washed copiously with chlorine-free water, dried and weighed, Scheme 2 [16]. Polyurethane (8) (0.44 g, 0.001 mol) was chlorinated only in the same way, Scheme 2 [16].
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
2.3. Effect of the halogenated polymers on bacterial growth and viability A culture of E. coli was prepared by inoculating one bacterial colony into 20 ml of nutrient broth in a Universal bottle and incubating for 24 h at 37 °C. From the bacterial suspension, 0.1 ml was transferred to a 20 ml Universal bottle containing 10 ml of fresh medium. Six Universals were prepared; three used in testing the effect of the polymer on bacterial growth and the other three to test the effect of the polymer on the viability of the bacteria. To study the effect of the polymer on growth rate of E. coli, 0.5 g of the halogenated polymer was added to the first bottle of broth while 0.5 g of the control polymer (non-halogenated) was added to the second bottle, to act as a polymer control, and the third was left as a bacterial control without polymer. The three bottles were stirred at 37 °C and sampled at timed intervals for viable count, employing the ‘Miles and Misra’ technique [18]. To investigate the effect of the polymer on the viability of E. coli, the other three bottles, inoculated as above, were incubated for 17 h at 37 °C, and the number of bacteria determined by viable count. At this time 0.5 g of the halogenated polymer was added to one bottle; 0.5 g of the control polymer (non-halogenated) was added to the second, to act as a polymer control, and the third vessel was left as a bacterial control. The three bottles were stirred at room temperature, and samples from each culture taken for viable count at regular time intervals, as previously. The procedure was repeated to test the effect of the halogenated polymers on a Gram-positive bacterium (S. aureus). 2.4. Effect of the halogenated polymers on bacterial viability under halogen quenching The previous experiment, investigating the effect of the polymer on the bacterial viability, was repeated. During the viable counts 0.05 ml of 0.5 M sodium thiosulphate was added to each decimal dilution to quench any halogen which may have evolved during the reaction between the polymer and the bacteria. 2.5. Effect of the non-halogenated polymer on the liquid medium Non-halogenated polymer, 0.5 g (non-halogenated 4) was added to each of two Universal bottles each containing 10 ml of sterile liquid medium. One of these was stirred at ambient temperature and the other was stirred at 37 °C for 17 h. The polymer was allowed to settle in each vessel and 5 ml of the overlaying broth removed to a fresh, sterile, Universal bottle. Bacterial suspension, 0.05 ml (either E. coli or S. aureus prepared as described above) was added as inoculum and the growth of the cultures followed by viable count during incubation at 37 °C. Similarly, 5 ml of sterile liquid medium was inoculated as a bacterial control and a viable count performed at the same time intervals.
1451
2.6. Determination of the amount of halogen released from the chlorinated polymer during contact with water, nutrient broth and bacterial medium Chlorinated polymer (5), 0.5 g, was stirred with 10 ml of chlorine-free water. The polymer was filtered, dried and the degree of halogenation of the polymer was determined before and after the experiment using iodometric titration in order to determine the amount of halogen released from the polymer to the water [9,16]. The experiment was repeated using nutrient broth and bacterial suspensions (both E. coli and S. aureus) instead of chlorine-free water and in each case the amount of delivered halogen calculated. 2.7. Evaluation of polymer (5) (chlorinated) in a column water filter Polymer (5), 1.0 g, was packed loosely in a glass tube (15 cm length and 1 cm diameter) to a height of 4 cm. Bacterial suspension [prepared by inoculating one bacterial colony in 20 ml of liquid medium (Nutrient Broth, Oxoid) and incubated for 17 h at 37 °C] was passed through the column and the output from the column recycled through it again. Before recycling, 0.1 ml from the passed liquid was sampled for viable count. Five cycles were performed for each column. Two columns were prepared containing the original polymer (non-chlorinated form of 5) which acted as a control, one for S. aureus and the other for E. coli. A further two columns were prepared for the N-halamine polymer (5), one for S. aureus and the other for E. coli, therefore four columns in total. The number of viable cells in the original bacterial suspensions was determined [3] and the turbidity of the liquid before and after passing through the columns also determined spectrophotometrically at 540 nm [10–12,15].
3. Results and discussion The biological activity and mode of action of a new synthesized N-halamine biocidal polymer was evaluated. The effect of this N-halamine polymer on bacterial growth was examined and one of its halogen derivatives was evaluated for use in water filters. 3.1. Determination of the effect of the halogenated polymers on bacterial viability and growth rate 3.1.1. Growth rate No bacterial growth was recorded in presence of any of the halogenated polymers (5–7). Fig. 1a shows S. aureus does not grow in the presence of the chlorinated polymer (5) or in the presence of the non-halogenated polymer (4) (polymer control); which, as S. aureus is not motile, may be due its adsorption onto the polymer surface. From Fig. 1c, E. coli also did not grow in the presence of the chlorinated polymer. However, unlike S. aureus, it grew in the presence of the control polymer (non-halogenated polymer) but at a different rate and to a lower final population than the E. coli ‘bacterial
1452
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Fig. 1. (a) and (c) Log plot of viable counts (cfu/ml) vs time of S. aureus and E. coli, respectively, grown in the presence of the chlorinated polymer (5), the control polymer (4) and without polymer (bacterial control). (b) and (d). Log plot of viable counts (cfu/ml) vs time of S. aureus and E. coli, respectively, in nutrient broth after stirring with the chlorinated polymer (5), control polymer (4) and without polymer (bacterial control).Where T = Test (bacteria + chlorinated polymer). BC = Bacterial control (no treatment). PC = Polymer control (bacteria + non-halogenated polymer).
control’. This may be due to differences in the motility and surface composition of the two types of bacteria. The same behaviour was recorded for the brominated polymer (6) (Fig. 3a and c) and for the iodinated polymer (7) (Figs. 4a and c). As shown in Figs. 1a and c the non-halogenated polymer (4) inhibited the growth of S. aureus and had some limited effect on the growth of E. coli. A plausible explanation may be the removal of critical nutrient broth components by the polymer through adsorption onto its surface. To investigate this possibility we treated fresh broth with polymer (4) at different temperatures, removing the polymer by allowing it to settle, and then removing and using the overlying liquor. The isolated medium was inoculated with a fresh culture of S. aureus and the bacterial growth followed by viable counts. The results (Figs. 2a and b) show that treating the broth with non-halogenated polymer (4) adversely affects the rate of bacterial growth, even after removal of the polymer (although to a lesser extent than when the polymer is still present). This suggests that there may be a dual mode of action, one by direct contact polymer–bacteria (possibly there is significant bacterial adsorption in the case of S. aureus) and another by affecting the broth composition (through nutrient adsorption onto the polymer).
3.1.2. Viability The chlorinated polymer (5) achieved a 3 log reduction in the bacterial population in 7 min while no bacterial colonies were detected after 15 min (equivalent to a 9 log reduction) in the case of S. aureus, Fig. 1b. A similar behaviour was recorded with E. coli, a 3 log reduction was recorded in the first 7 min while no colonies were detected after 15 min (equivalent to a 10 log reduction), Fig. 1d. The brominated polymer (6) behaves similarly to the chlorinated polymer (5) for E. coli (Fig. 3b), and also for S. aureus but with a 9 log reduction in 15 min (Fig. 3d). The iodinated polymer showed the highest power of sterilization (Fig. 4b) showing a 10 log reduction in the population of E. coli after 7 min following contact between the iodinated polymer (7) and the bacteria. S. aureus behaves similarly; no colonies were detected after 7 min contact between the iodinated polymer (7) and the bacteria (Fig. 4d). To determine the rate of killing by the iodinated polymer (7) the amount of polymer in contact with the bacteria was reduced; the experiment was repeated, but using 0.25 g of polymer in contact with 10 ml of bacterial suspension and the time intervals were reduced to detect any viable bacterial colonies early in the culture. For E. coli, a 5 log reduction in the bacterial population was achieved in 1 min and no viable colonies detected (equivalent to 10
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
log reduction) after 5 min contact time. The results for S. aureus were unequivocal, the rate of killing by the iodinated polymer could not be determined due to the powerful effect of the polymer; no viable colonies were detected after 1 min contact time (equivalent to 9 log reduction) (Figs. 4e and f, respectively).
1453
3.2. Killing rate determination with halogen quenching Most of the research surrounding the N-halamine polymers states that the mode of action of these polymers depends only on the contact between the polymer particles and the bacterial cells [1–15]. However, we suggest that
Fig. 2. (a) and (b) Log plot of viable count (cfu/ml) vs time of S. aureus and E. coli, respectively, grown in liquid medium previously stirred with the nonhalogenated polymer (4). Where Growth of S. aureus (S25) E. coli (E25) in medium pre-treated with the non-halogenated polymer at 25 °C (or, S37 and E37 at 37 °C). Sc and EC are the respective control grown in untreated nutrient broth.
Fig. 3. (a) and (c) Log plot of E. coli and S. aureus viable count (cfu/ml) vs time grown in the presence of the brominated polymer (6), the control polymer (4) and without treatment (bacterial control). (b) and (d) Log no. of colonies (cfu/ml) vs time of E. coli and S. aureus, after exposure to the brominated polymer (6), the non-halogenated polymer (4) and without treatment (bacterial control). Where T = Cells treated with brominated polymer. PC = Cells treated with the control polymer (non-halogenated polymer). BC = Bacterial control (no treatment). BC and PC are superimposed in Fig. 3b.
1454
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
Fig. 4. (a) and (c) Log no. of E. coli and S. aureus viable count (cfu/ml) vs time during growth in nutrient broth; in the presence of the iodinated polymer (7), the non-halogenated polymer (4) and without treatment (bacterial control). (b) and (d) Log no. of E. coli and S. aureus, respectively, vs time after treating the bacterial cells with; the iodinated polymer (7) (0.5 g/10 ml bacterial suspension) the non-halogenated polymer and without treatment (bacterial control). (e) and (f) Log no. of E. coli and S. aureus, respectively, at timed intervals after stirring the bacterial cells with; iodinated polymer (7) (0.25 g/10 ml bacterial suspension) and without treatment (bacterial control). Where T = Cells treated with the iodinated polymer. PC = Polymer control, cells treated with nonhalogenated polymer. BC = Bacterial control, untreated cells.
the mechanism of killing, by halogenated polymers can also occur through the release into the medium of soluble halogen species; in addition to contact between the bacteria and polymers. To investigate this, the effect of adding a halogen quencher (sodium thiosulphate) to samples for viable counts immediately prior to counting was examined. Mixing was stopped and the polymer granules allowed to set-
tle before samples of the overlying bacterial suspension were taken for viable count with quenching as described earlier. Any differences in the observed killing rate would reflect a release of active species during mixing of polymer and bacteria. The chlorinated polymer (5) achieved a 2 log reduction in the E. coli population in 40 min and no viable colonies were detected after 1.5 h (9 log reduction). Similarly the
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
chlorinated polymer achieved a 1 log reduction in 40 min for the S. aureus population and no bacterial colonies were detected after 1.5hr (9 log reduction) (Figs. 5a and b, respectively). The brominated polymer (6) showed greater activity, achieving a 4 log reduction in the E. coli population in 40 min and no viable colonies detected after 1.5 h (9 log reduction) while it achieved a 4 log reduction in the S. aur-
1455
eus population in 15 min and no viable colonies detected after 40 min (9 log reduction) (Figs. 5c and d, respectively). For the Iodinated polymer (7), the experiment was performed with a 1 g:40 ml ratio between the polymer weight and the bacterial suspensions due to the high biocidal power of the iodinated polymer. In spite of this reduced quantity of polymer, no E. coli or S. aureus colonies were detected at 7 min (9 log reduction) (Figs. 5e and f, respectively).
Fig. 5. (a) and (b) Effect of the chlorinated polymer (5) on the viability of E. coli and S. aureus, respectively, under chlorine quenching. (c) and (d) Effect of the brominated polymer (6) on the viability of E. coli and S. aureus, respectively, under bromine quenching. (e) and (f) Effect of the iodinated polymer (7) on the viability of E. coli and S. aureus, respectively, under iodine quenching. Where T = Treated cells with the halogenated polymer. BC = Untreated cells (bacterial control).
1456
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
From the previous results (Figs. 1–5) it was clear that the most powerful biocidal effect is exhibited by the iodinated polymer. Without halogen quenching the rate of killing of the chlorinated and brominated polymers are very similar, while with halogen quenching, the brominated polymer shows a more powerful effect on each type of bacteria. This may be related to the stability of the halogen on the polymer; I–N bonds have the lowest stability therefore
the halogen can be easily exchanged between the polymer and bacteria. The Cl–N exhibits the lowest biological power but is the most stable bond; hence its use in water filters. The difference in the killing rate of the polymer with and without halogen quenching (a lower rate with quenching) indicates that halogen species (released from the polymers) are involved in the biocidal action, and suggesting the mechanism of killing is not by contact alone [1–15].
Fig. 6. (a) and (b) Effect of chlorinated polymer (9) on the viability of E. coli and S. aureus, respectively, under chlorine quenching. Where T = Cells treated with the chlorinated polymer. BC = Cells without treatment (bacterial control).
Fig. 7. (a) and (b) Log viable count (cfu/ml) of S. aureus and E. coli recovered from the eluate after each passage through the column containing control polymer. (c) and (d) Log viable counts (cfu/ml) of S. aureus and E. coli, respectively, after each cycle through the column containing chlorinated polymer (5).
1457
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
This dual mechanism of killing (contact + release of halogen species) is currently under investigation and will be reported in due course. As stated earlier there is also a possibility of halogen exchange between the polymer and the protein in the broth medium. This latter possibility can be examined by calculating the amount of released halogen in different liquid media. 3.3. Amount of released chlorine To confirm the presence of halogen species released from the polymer, the amount of chlorine delivered from the polymer to the bacteria was determined by iodometric titration of the halogenated polymer isolated after controlled contact with water, broth medium and the bacterial suspensions [9,16]. The amount of released chlorine was determined as the difference between original chlorine content in the polymer and chlorine content after contact with the different media, the ratio between the weight of polymer and the liquid was 1 g: 20 ml The amounts of delivered halogen ions to water, nutrient broth, E. coli (1.7 109 cfu/ml) in broth medium and S. aureus (3.1 108 cfu/ml) in broth medium were 0.19 ppm ± 0.05, 3.9 ppm ± 0.12, 8.5 ppm ± 0.32 and 5.9 ppm ± 0.24, respectively. The amount of delivered chlorine to water is very low but increased for Nutrient broth, possibly through chlorine exchange between the polymer and protein in the broth medium, which supports the possibility of bacterial death due to changing the nature of the nutrients in the medium, in addition to the release effect and a contact effect. In the presence of bacteria the amount of delivered chlorine increased indicating that more chlorine is released from the polymer when in contact with bacteria. The mechanism or the mode of action of this process is under further investigation but from these results we suggest that the mode of action of this kind of polymer is a combination of factors. 3.4. Determination of the killing rate of the chlorinated polymer (9) under quenching conditions and in comparison with chlorinated polymer (5) This experiment was performed to investigate the suitability of using the alternative commercial, but low purity, toluene-2,4-diisocyante instead of tolylene-2,6-diisocyante. From Fig. 6a, chlorinated polymer (9) achieved a 6 log reduction in the E. coli population in 40 min and no bacterial colonies were detected after 40 min (equivalent to 9 log reduction). For S. aureus, Fig. 6b, the chlorinated polymer (9) achieved a 6 log reduction in 40 min and no colonies detected after 90 min (equivalent to 9 log reduction). From Figs. 6a and b it was clear that using toluene-2,4-diisocyanate (polymer 9) instead of tolylene-2,6-diisocyante (polymer 5) gave very similar results and that the commercial product was equally effective. 3.5. Chlorinated polymer (5) evaluation in water filters This experiment investigated the potential to use one of these halogenated polymers (5) in water filters on a labo-
Table 1 Absorbance (at 540 nm, read against a nutrient broth blank) and viable counts before and after perfusing columns of non-halogenated and halogenated polymers Non-halogenated polymer (control) E. coli Before perfusing the column (viable counts (cfu/ml)) After the fifth cycle (viable counts (cfu/ml)) After incubating the fifth cycle for 24 h (viable counts (cfu/ml))
0.41 9
(2.6 10 ) 0.30 (8.4 108) 1.33
(2.5 109)
Halogenated polymer
S. aureus
E. coli
0.30
0.27 7
(8.5 10 ) 0.04
S. aureus 0.32 9
(1.8 10 ) 0.02
(3.1 108) 0.01
(Nd)
(Nd)
(Nd)
0.46
0.00
0.03
(7.5 107)
(Nd)
(Nd)
Nd = Non-detected.
ratory scale. Figs. 7a and b show the effect of the control polymer (4) contained in a 4 cm length 1 cm diameter column on suspensions of S. aureus and E. coli. After 3 cycles no viable colonies of S. aureus were detected in the eluate (Fig. 7a). Whereas for E. coli, although there was an initial reduction in the population, this recovered and there was no significant reduction overall in the bacterial population over the 5 cycles. This suggests that the control polymer is merely acting as a filter, allowing passage of the motile E. coli but not exhibiting any biocidal action. Results from columns containing the chlorinated polymer (5) are shown in Figs. 7c and d and Table 1. After one passage through the column the population of both types of bacteria was reduced to a non-detectable level (equivalent to an 8 log reduction). This was also reflected in the spectrophotometric measurements (Table 1); the negative values after 5 cycles are due to a bleaching effect on the broth medium. Further incubation of the samples post-passage confirmed the biocidal effect and demonstrates the potential application of the polymer in water filters. 4. Conclusions The rate of bacterial killing of new halamine polymers was examined and the effect of these polymers on bacterial growth determined. No bacterial growth was recorded in the presence of any of the halogenated polymers. The iodinated polymer showed greater biocidal power than chlorinated and brominated polymers. Investigations performed with and without halogen quenching indicate that there is a difference in killing-rate, suggesting that the mode of action of these polymers is dual; proceeding through both release of halogen species into the medium and through bacteria–polymer contact. In addition, a third possibility exists – changing the nature of nutrients around the bacteria through interactions between the halogenated polymer and medium proteins. The non-halogenated polymer also showed an inhibitory effect on growth of S. aureus. One of
1458
A.E.-S.I. Ahmed et al. / Reactive & Functional Polymers 68 (2008) 1448–1458
the prepared N-halamine polymers (5) (chlorinated) was used successfully in a laboratory column for sterilizing a bacterial culture. Acknowledgment The project is sponsored by the Egyptian government. References [1] G. Sun, L.C. Allen, E.P. Luckie, W.B. Wheatly, S.D. Worley, Ind. Eng. Chem. Res. 34 (1994) 4106. [2] G. Sun, W.B. Wheatly, S.D. Worley, Ind. Eng. Chem. Res. 33 (1994) 168. [3] G. Sun, T.Y. Chen, W. Sun, W.B. Wheatley, S.D. Worley, J. Bioact. Compat. Polym. 10 (1995) 135. [4] Y. Chen, L. Wang, H. Yu, Q. Shi, X. Dong, J. Mater. Sci. 42 (2007) 4018. [5] Y. Chen, S.D. Worley, T.S. Huang, J. Weese, J. Kim, C-I. Wei, J.F. Williams, J. Appl. Polym. Sci. 92 (2004) 368.
[6] Y. Chen, S.D. Worley, T.S. Huang, J. Weese, J. Kim, C.-I. Wei, J.F. Williams, J. Appl. Polym. Sci. 92 (2004) 363. [7] Y. Chen, S.D. Worley, J. Kim, C.-I. Wei, T.Y. Chen, Ind. Eng. Chem. Res. 42 (2003) 280. [8] Z. Chen, J. Luo, Y. Sun, Biomaterials 28 (2007) 1597. [9] Z. Chen, Y. Sun, Ind. Eng. Chem. Res. 45 (2006) 2634. [10] A.M. El-Masry, Pigment Resin Technol. 34 (2005) 265. [11] A.M. El-Masry, H.Y. Moustafa, A.I. Ahmed, A.F. Shaaban, Pigment Resin Technol. 33 (2004) 211. [12] A.M. El-Masry, H.Y. Moustafa, A.I. Ahmed, A.F. Shaaban, Pigment Resin Technol. 33 (2004) 75. [13] K. Barnes, J. Liang, R. Wu, S.D. Worley, J. Lee, R.M. Broughton, T.S. Huang, Biomaterials 27 (2006) 4825. [14] J. Liang, J.R. Owens, T.S. Huang, S.D. Worley, J. Appl. Polym. Sci. 101 (2006) 3448. [15] H.Y. Moustafa, Pigment Resin Technol. 35 (2006) 71. [16] A.E.I. Ahmed, J.N. Hay, M.E. Bushell, J.N. Wardell, G. Cavalli, React. Funct. Polym. 68 (2008) 248. [17] S.F. Brian, J.H. Antony, W.G.S. Peter, R.T. Austin, ‘Vogel’ Text Book of Practical Organic Chemistry, fifth ed., Longman Ltd, 1989. 1170. [18] A.A. Miles, S.S. Misra, J. Hygiene (London) 38 (1938) 732.