Synthesis and characterization of novel antimicrobial polymers containing pendent triclosan moieties

Article abstract of DOI:10.1016/j.reactfunctpolym.2011.10.009

Novel antimicrobial copolymers were produced by first converting the commodity biocide, triclosan (TCS), to an epoxy-functional derivative, 2-((5-chloro-2-(2,4-dichlorophenoxy)phenoxy) methyl)oxirane (ETCS), and then reacting ETCS with polyethylenimine (P

Full text of DOI:10.1016/j.reactfunctpolym.2011.10.009

Reactive & Functional Polymers 72 (2012) 69–76
Contents lists available at SciVerse ScienceDirect
Reactive & Functional Polymers
journal homepage: www.elsevier.com/locate/react
Synthesis and characterization of novel antimicrobial polymers containing pendent
triclosan moieties
Alex J. Kugel a, Scott M. Ebert b, Shane J. Stafslien b, Ivan Hevus a, Ananiy Kohut a, Andriy Voronov a,
Bret J. Chisholm a,b,
⇑
a Department of Coatings and Polymeric Materials, North Dakota State University Fargo, ND 58102, United States
b Center for Nanoscale Science and Engineering, North Dakota State University Fargo, ND 58102, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 July 2011
Received in revised form 10 October 2011
Accepted 22 October 2011
Available online 30 October 2011
Novel antimicrobial copolymers were produced by first converting the commodity biocide, triclosan (TCS),
to an epoxy-functional derivative, 2-((5-chloro-2-(2,4-dichlorophenoxy)phenoxy) methyl)oxirane (ETCS),
and then reacting ETCS with polyethylenimine (PEI). While neither ETCS or PEI showed high antimicrobial
activity toward either the Gram-positive bacterium, Staphylococcus epidermidis, or the Gram-negative bac-
terium, Escherichia coli, some the copolymers showed very high activity toward both bacteria. Antimicro-
bial activity for these copolymers was found to be highly dependent on both the molecular weight of the
PEI utilized and the concentration of pendent groups derived from ETCS. In general, decreasing PEI molec-
ular weight and increasing TCS pendent group concentration increased antimicrobial activity. Surface ten-
sion measurements showed that the molecular parameters affecting antimicrobial activity also affected
surface activity in a similar fashion. Thus, it was speculated that the mechanism of antimicrobial activity
associated with these copolymers involves interaction of the copolymers with the bacterial cell wall. A
comparison of the antimicrobial activity of the most effective copolymers to TCS showed that the copoly-
mers were more effective toward E. coli than pure TCS when compared using an equivalent TCS content
(i.e. TCS pendent group content for the copolymers). This characteristic coupled with the fact that the
TCS-containing copolymers are highly aqueous soluble liquids as opposed to a crystalline solid of limited
solubility may afford utility of these copolymers for a variety of applications.
Keywords:
Functional polymers
Polyethylenimine
Triclosan
Biocide
Antimicrobial
Staphylococcus epidermidis
Escherichia coli
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
been found to be nontoxic orally as well as showing no mutagenic,
carcinogenic, or teratogenic properties [9,10].
Antimicrobial materials and biocides are compounds that kill or
prevent the growth of pathogenic and other unwanted microor-
ganisms [1]. This broad group of chemicals is used extensively in
the fields of medicine [2], water treatment [3], personal-care [4],
marine vessels [5], and architecture [6]. 2,4,40-Trichloro-20-hydrox-
ydiphenyl ether, typically referred to as triclosan (TCS), is a broad-
spectrum antimicrobial agent commonly used in personal-care
products that was initially developed by the Ciba-Geigy Company
in the 1960s [7]. TCS has been formulated into hand soaps, surgical
scrubs, shower gels, underarm deodorants, toothpastes, hand lo-
tion, and mouthwashes; incorporated into fabrics and plastics such
as children’s toys, surgical drapes, cutting boards and toothbrush
handles; and even infused into concrete for floors [8]. Extensive
investigations have been done on the toxicity of TCS and it has
There is some discrepancy as to the mode of action of TCS, but it
is generally accepted that it is bactericidal with multiple mecha-
nisms of action at higher levels and, at lower levels, inhibits a spe-
cific site in the fatty acid biosynthetic pathway [11]. A specific
targeted pathway at lower concentrations causes much concern
because it allows for bacteria to develop resistant strains when ex-
posed to sublethal concentrations of TCS [8,12,13]. This is particu-
larly concerning considering the prevalence of TCS use.
Several groups have studied different techniques to control the
release and biocidal natureof TCSby incorporatingthemolecule into
polymer compositions. All of the recent studies have involved the
incorporation of TCS into polymer compositions by either attaching
TCS directly to a polymer chain using a hydrolytically-labile
functional group or by physically blending it with polymers
[14,15]. For these previously studied polymeric compositions, anti-
microbial properties were obtained through the release of TCS
molecules.
⇑
Corresponding author at: Center for Nanoscale Science and Engineering, North
Dakota State University Fargo, ND 58102, United States. Tel.: +1 701 231 5328; fax:
+1 701 231 5325.
The authors have been interested in generating surface coatings
that provide antimicrobial activity without leaching toxic
E-mail address: bret.chisholm@ndsu.edu (B.J. Chisholm).
1381-5148/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.reactfunctpolym.2011.10.009

70
A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
components into the environment [16–21]. Most approaches for
producing non-leaching, contact-active antimicrobial surfaces in-
volve the generation of bound cationic functional groups at the sur-
face [22]. These bound cationic groups can effectively disrupt the
integrity of the bacterial cell wall through ionic interactions leading
to cell death. The authors have previously demonstrated that tether-
ing relatively high levels of TCS to a siloxane coating matrix using a
non-hydrolyzable linking group provides antimicrobial activity
without leaching toxic compounds from the coating [17,19,20]. This
result suggested that, at relatively high concentrations, the TCS mol-
ecule may have a cell membrane disruption affect. During the course
of conducting research focused on the development of contact-ac-
tive, non-leaching antimicrobial surface coatings, it was discovered
that polymers derived from the reaction of polyethylenimine (PEI)
and 2-((5-chloro-2-(2,4-dichlorophenoxy)phenoxy) methyl)oxi-
rane (epoxy triclosan) were biocidal. This document describes the
synthesis, physical properties, and antimicrobial properties of these
novel antimicrobial polymers.
reaction progress monitored using proton nuclear magnetic reso-
nance spectroscopy (1H NMR) by observing the disappearance of
peaks in the spectrum corresponding to protons associated with
the epoxy group of ETCS. Upon completion of the reaction, solvent
was removed at reduced pressure using a rotary evaporator. The
product, a yellow viscous liquid, was collected and characterized
using 1H NMR and differential scanning calorimetry (DSC). All
other PEI copolymers containing pendent triclosan moieties were
synthesized using the same synthetic procedure with the excep-
tion that PEI molecular weight and ETCS concentration was varied.
In addition, reaction time was adjusted based on reaction monitor-
ing results to ensure complete reaction. Table 1 lists the details of
each synthesis conducted.
2.3. Instrumentation
NMR spectra were obtained using a JEOL ECA400 400 MHz NMR
equipped with a 24-place autosampler carousel. Samples were
measured in deuterated chloroform using 16 scans for proton spec-
tra, pulse width of 14.003
of 45°, attenuation of 6 dB, pulse time of 7.0015
l
s, acquisition time of 2.18 s, pulse angle
2. Experimental
ls, receiver gain of
28, relaxation delay of 4 s, and repetition time of 6.18 s. Carbon
spectra were obtained in deuterated chloroform using 1000 scans,
2.1. Materials
pulse width of 10.309
30°, attenuation of 9 dB, pulse time of 3.4363
l
s, acquisition time of 1.04 s, pulse angle of
TCS was used as received from Alfa Aesar. PEIs of varying molecu-
lar weight (Mn = 423 g/mole, Mn = 600 g/mole, Mn = 10,000 g/mole),
epichlorohydrin, potassium hydroxide, isopropanol, toluene, mag-
nesium sulfate, and hexanes were used a received from Sigma–Al-
drich Chemical. ChromARÒ chromatography grade chloroform was
used as received from Mallinckrodt Chemicals. Marine broth (MB)
was prepared according to the manufacturer’s (Becton Dickinson
Labware) specifications.
ls, receiver gain of
58, relaxation delay of 2 s, and repetition time of 3.04 s. In addition,
carbon spectra were run with decoupling and NOE activated at an
NOE time of 2 s.
High performance liquid chromatograpy (HPLC) was conducted
using an Agilent 1100 Series HPLC equipped with an Agilent 1100
autosampler and diode array detector. The HPLC column was a Zor-
bax Eclipse XDB-C18 running a mobile phase of 30.0% water and
70.0% acetonitrile at 40 °C with a column flow rate of 1.8 mL/min.
2.2. Synthesis
20
l
L sample aliquots were injected using aspiration/dispense rate
settings of 200
l
L/min. The detected signal of the diode array detec-
Epoxy triclosan (ETCS) was synthesized as follows: 80 g of iso-
propanol and 19.4 g (0.346 mol) of potassium hydroxide were
charged to a 1 L, three-neck, round-bottom flask equipped with a
magnetic stir bar and the mixture stirred at room temperature un-
til the potassium hydroxide dissolved. In an 800 mL glass beaker,
100 g (0.346 mol) of TCS was dissolved at room temperature in
250 g of isopropanol using magnetic stirring. The TCS solution
was subsequently added to the round-bottom flask containing
the potassium hydroxide solution. The flask was then placed in a
temperature-controlled silicone oil bath and equipped with a con-
denser and a 250 mL addition funnel. A thermocouple was placed
into the reaction flask and the temperature controller set at
60 °C. Once the temperature had equilibrated, 95.9 g (1.036 mol)
of epichlorohydrin was added dropwise to the solution over the
course of 5 min using the addition funnel. During the course of
the reaction, a precipitate (potassium chloride) was formed. The
reaction was allowed to run for 16 h. Upon completion of the reac-
tion, the reaction mixture was transferred to a 1-L, single-neck,
round-bottom flask and then placed on a rotary evaporator to re-
move unreacted epichlorohydrin at reduced pressure. Further puri-
fication was done using solvent extraction with water and a 1/1 v/v
mixture of hexanes and toluene. The organic phase was washed
four times with water and dried over magnesium sulfate. Solvent
was removed at reduced pressure on a rotary evaporator and the
clear viscous liquid product was collected (yield: 88%).
tor was 280 nm. Data analysis was done using ChemStation for LC
3D Systems supplied by Agilent Technologies.
DSC thermograms were obtained using a TA Instruments Q1000
DSC. 7.12–18.5 mg of polymer were dispensed into hermetic alu-
minum pans and sealed using the TA instruments DSC Blue Sample
Press. The temperature profile consisted of equilibrating the sam-
ples at ꢀ90 °C for 2 min, heating from ꢀ90 °C to 90 °C at 10 °C/
min, cooling from 90 °C to ꢀ90 °C at 10 °C/min, and finally heating
from ꢀ90 °C to 90 °C at 10 °C/min. Glass transition temperatures
(Tgs) were taken as the inflection point of the last heating regime
with the assistance of TA Instruments Universal Analysis 2000
software.
2.4. Surface tension
Surface tension measurements were performed by drop shape
analysis on a pendant drop produced with a Contact Angle/Surface
Table 1
Compositions of reaction mixtures used for PEI copolymer synthesis.
Copolymer acronym
PEI Mn (g/mole)
Wt. PEI (g)
Wt. ETCS (g)
423PEI–5%TCS
423
423
10.0
7.5
5.1
7.6
423PEI–10%TCS
423PEI–20%TCS
600PEI–5%TCS
600PEI–10%TCS
600PEI–20%TCS
10,000PEI–5%TCS
10,000PEI–10%TCS
10,000PEI–20%TCS
423
600
600
5.0
10.0
7.5
10.2
5.1
7.6
A PEI copolymer, referred to in this document as 423PEI–5%TCS,
was synthesized as follows: 10.0 g (0.294equiv NH) of the 423
g/mole PEI and 5.1 g (0.015 mol) of ETCS were dissolved at room
temperature in 90 mL of chloroform using a 250 mL single-neck,
round-bottom flask and the mixture heated at 50 °C under a
blanket of nitrogen. The reaction was allowed to run for 64 h and
600
5.0
10.0
7.5
10.2
5.1
7.6
10,000
10,000
10,000
5.0
10.2

A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
71
Tension Analyzer from First Ten Angstroms (FTÅ125). Measure-
3. Results and discussion
ments were carried out at room temperature on polymer solutions
of varying concentration (10ꢀ4–10% w/v). All glassware was washed
in a 1 N NaOH bath and thoroughly rinsed with Millipore water be-
fore use.
A series of PEI copolymers containing pendent TCS moieties
were synthesized using the synthetic scheme shown in Fig. 1.
TCS (I) was first derivatized by reaction with epichlorohydrin (II)
to produce ETCS (III) and ETCS was subsequently reacted with
PEI (IV) to produce the copolymers of interest (V). Fig. 2 displays
the 1H NMR spectrum obtained for ETCS. The peaks in the region
of the spectrum ranging from 2.6 ppm to 4.3 ppm confirm success-
ful incorporation of the epoxide functionality into the compound.
Since trace amounts of TCS in the ETCS may result in false positives
with respect to the characterization of antimicrobial activity, HPLC
was used to identify trace impurities of TCS in the ETCS sample.
The elution time for TCS was determined to be 4.7 min while that
for ETCS was 6.8 min. For the ETCS sample, no peak at 4.7 min was
discernable indicating the lack of trace amounts of TCS. ETCS was
isolated as a liquid; however, crystallization was observed with
time at ambient temperature. The melting temperature of the crys-
tals was determined by DSC to be 43 °C.
TCS-containing PEI copolymers were readily synthesized by
simply heating solutions of PEI and ETCS at 50 °C. As shown in
Fig. 3, reaction progress was easily monitored using 1H NMR by
observing the disappearance of protons associated with the epoxy
ring of ETCS. The series of TCS-containing PEI copolymers produced
possessed variations in the number-average molecular weight (Mn)
of the PEI backbone and the concentration of TCS pendent groups.
The PEI Mns, as specified by the manufacturer, were 423, 600, and
10,000 g/mole. TCS pendent group concentration was varied at 5,
10, and 20 mol% based on total PEI repeat units assuming that
the nominal chemical structure of the PEIs was (CH2CH2NH)n. For
this document, the following formula was used to identify the dif-
ferent copolymers produced:
2.5. Antimicrobial properties
The antimicrobial properties of the polymers were determined
using an adaptation of the standard minimum inhibitory concen-
tration (MIC) test. MIC is typically reported as the lowest concen-
tration of antimicrobial that completely inhibits growth over a
set period of incubation relative to a control containing no antimi-
crobial [23]. For the polymers investigated, the amount of bacterial
growth at each polymer concentration was reported. The microor-
ganisms used for the study included the Gram-positive bacterium,
Staphylococcus epidermidis ATCC 35984, and the Gram-negative
bacterium, Escherichia coli ATCC 12435. 100 mg of each polymer
was dissolved into 10 mL of methanol to generate a 10 mg/mL
working solution. 10 mL of the appropriate growth medium, tryp-
tic soy broth (S. epidermidis) or Luria–Bertani broth (E. coli), was
spiked with 200 lL of the 10 mg/mL polymer working solution to
achieve a final concentration of 0.2 mg/mL. A 1:1 serial dilution
of the 0.2 mg/mL polymer suspension was then prepared and
0.5 mL of bacterial suspension in growth medium (1:1000 of over-
night culture) was then added to 0.5 mL of each dilution generating
polymer concentrations of 0.78–100 lg/mL. 0.2 mL of each poly-
mer solution (spiked with bacteria) was then added in triplicate
to a 96-well plate, placed in a 37 °C incubator for 24 h with shak-
ing, and measured for growth by taking absorbance measurements
at 600 nm with a multi-well plate spectrophotometer. 0.2 mL of
growth medium and 0.2 mL of growth medium with the appropri-
ate bacterium served as a negative and positive growth control,
respectively. The antimicrobial activity for each polymer concen-
tration was reported as a percent reduction in bacterial growth
as compared to the positive growth control. Each percent reduction
value was reported as the mean of three replicate samples.
xPEI ꢀ y%TCS
where x is the Mn of the PEI backbone and y is the mole percent of
TCS pendent groups based on the total number of PEI repeat units. A
O
Cl
OH
Cl
O
O
O
Cl
O
Cl
Cl
Cl
Cl
II
I
III
Cl
Cl
O
O
N
Cl
O
Cl
O
N
H
O
n
OH
IV
Cl
Cl
N
H
III
x
n-x
V
Fig. 1. The synthesis of ETCS (III) from epichlorohydrin (II) and triclosan (I), and the subsequent synthesis of PEI–TCS (V) from PEI (IV) and ETCS (III).

72
A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
i,j
423 g/mole
10
0
600 g/mole
O
h
10,000 g/mole
d,k
f,g
Cl
b
O
k
-10
-20
-30
-40
-50
-60
-70
O
a
a
c
e
c
e
d
Cl
Cl
i
b
j
f
g
h
7
6
5
4
3
ppm
Fig. 2. 1H NMR spectrum for ETCS with peak assignments. *This peak corresponds
0
5
10
15
20
25
to residual CHCl3 present in the CDCl3 used as the solvent.
TCS Content (mole %)
Fig. 4. Tg as a function of TCS pendent group concentration.
a,b
O
c
used to produce the TCS-containing PEI copolymers was deter-
mined and compared to that of pure triclosan. As shown in Fig. 5,
the PEIs and ETCS showed little to no biological activity toward
the two microorganisms used in this study, E. coli and S. epidermi-
dis, while pure triclosan showed a substantial decrease in growth
over the same concentration range. The effectiveness of triclosan
toward S. epidermidis and E. coli was consistent with expectations
based on previous results [7]. Considering that, at low concentra-
tions, triclosan is believed to inhibit a specific bacterial target with-
in the fatty acid biosynthetic pathway, it was not surprising that
altering the chemical structure of triclosan by capping the hydroxy
group with epichlorohydrin to produce ETCS inhibited antimicro-
bial effectiveness.
d,e
Cl
O
O
Cl
Cl
(c)
(b)
(a)
e
c
d
e
d
c
4
3.8
3.6
3.4
3.2
3
In aqueous media at essentially neutral pH, PEI is a polycation
[24] and, thus, would be expected to interact with the negatively
charged components at the outer surface of the cell envelope. It
has been shown that binding of PEI to the bacterial cell wall
through electrostatic interactions can increase the permeability
of the cell wall enabling the permeation of antibiotics into the inte-
rior of the cell that would normally be unable to penetrate the cell
wall [25]. Similar to the results shown in Fig. 5, Helander et al. [26]
found that PEI had no bactericidal effect when tested toward the
Gram-negative bacteria, E. coli, Pseudomonas aeruginosa, and
Salmonella typhimurium. This result was not surprising considering
the plethora of research that has demonstrated that cationic anti-
microbial agents require a minimum degree of lipophilicity to
compromise the integrity of the cell envelope to the extent that
cell death occurs [27–29]. According to the generally accepted
mechanism of antimicrobial action for cationic disinfectants, once
the cationic compound has adsorbed onto the surface of the bacte-
rial cell through electrostatic interactions between the positively
charged quaternary nitrogen and the head groups of acidic
-phospholipids, the lipophilic portion of the compound penetrates
into the hydrophobic core of the cell wall, resulting in a loss of the
integrity of the cell wall and osmoregulatory capability and subse-
quent cell death.
ppm
Fig. 3. 1H NMR spectra as a function of reaction time for the synthesis of a PEI–TCS
copolymer: (a) 0 h; (b) 4 h; and (c) 28 h. These spectra correspond to a reaction
involving the 423 g/mole PEI.
description of the nine different copolymers produced is provided
in Table 1.
As shown in Fig. 4, incorporation of TCS moieties as pendent
groups to the PEI backbone sharply increased glass transition tem-
perature (Tg). Tg increased with TCS pendent group concentration
for each of the PEIs used for the study. The variation in Tg for the
parent PEIs can be attributed to the variation in molecular weight.
Polymer end-groups possess more conformational freedom than
internal segments resulting in an overall increase in material free
volume with decreasing molecular weight. The magnitude of the
increase in Tg with TCS pendent group concentration was quite
high. This result can be understood by considering the bulkiness
and relatively high rigidity of the TCS structure, the relatively short
length of the linking group between the TCS moieties and the PEI
backbone, and the ability of the secondary hydroxyl present in
the linking group between TCS moieties and the PEI backbone to
undergo hydrogen bonding.
Figs. 6 and 7 display the antimicrobial properties of the TCS-
containing polymers toward S. epidermidis and E. coli, respectively.
From these figures, it can be easily seen that both copolymer
molecular weight and TCS moiety concentration strongly affected
antimicrobial activity. Copolymers based on the lowest molecular
3.1. Antimicrobial properties
Prior to determining the antimicrobial activity of the TCS-con-
taining PEI copolymers, the antimicrobial activity of the precursors

A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
73
100
80
60
40
20
0
100
(b)
(a)
80
PEI 423 g/mole
PEI 600 g/mole
PEI 10,000 g/mole
ETCS
PEI 423 g/mole
PEI 600 g/mole
PEI 10,000 g/mole
ETCS
60
40
20
0
TCS
TCS
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
Concentration (micrograms/ml)
Concentration (micrograms/ml)
Fig. 5. Antimicrobial activity of the precursor materials used in the study and TCS toward S. epidermidis (a) and E. coli (b).
100
100
80
60
40
20
0
(a)
(b)
80
60
40
20
0
423PEI-10%TCS
600PEI-10%TCS
10,000PEI-10%TCS
423PEI-5%TCS
600PEI-5%TCS
10,000PEI-5%TCS
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
Concentration (micrograms/ml)
Concentration (micrograms/ml)
100
(c)
80
60
40
20
0
423PEI-20%TCS
600PEI-20%TCS
10,000PEI-20%TCS
0
2
4
6
8
10
12
14
Concentration (micrograms/ml)
Fig. 6. Inhibitory response of the PEI–TCS copolymers toward S. epidermidis. PEI–TCS copolymers containing 5 (a), 10 (b) and 20 (c) mole percent pendent TCS moieties.

74
A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
weight PEI were very effective at inhibiting growth of both
S. epidermidis and E. coli, while copolymers based on the highest
molecular weight PEI were ineffective. For the copolymers based
on the medium molecular weight PEI (i.e. 600 g/mole), those con-
taining the low and medium levels of TCS pendent groups (i.e.
600PEI–5%TCS and 600PEI–10%TCS) showed no antimicrobial
activity, while the copolymer based on the highest concentration
of pendent TCS groups, 600PEI–20%TCS, showed high antimicrobial
activity toward both bacteria. While the mechanism of antimicro-
bial activity associated with these PEI–TCS copolymers is un-
known, the observation that antimicrobial activity varied with
polymer molecular weight and copolymer composition suggested
that adsorption and diffusion characteristics of the copolymers
with respect to the bacterial cell wall were important factors.
Stark variations in antimicrobial activity with molecular weight
have been previously observed by several others involved in the
study of antimicrobial polymers. For example, Ikeda et al. found
that biological activity of in-chain quaternary ammonium salts to-
ward Staphylococcus aureus depended strongly on molecular
weight with higher molecular weight polymers providing in-
creased activity [30]. Kanazawa et al. also observed increased anti-
microbial efficacy toward S. aureus with increasing molecular
weight for poly[tributyl(4-vinylbenzyl)phosphonium chloride]
cationic polymers that ranged in molecular weight from 16,000
to 94,000 g/mole [31]. Cationic poly(vinyl pyridine) was found to
be bactericidal at 160,000 g/mole but not at 60,000 g/mole [32],
and N-alkylated PEI showed excellent activity at 25,000 g/mole
but not at 2000 g/mole [33]. Chen and coworkers utilized quater-
nary ammonium-functionalized poly(propylene imine) dendri-
mers of very narrow molecular weight distribution and
controlled molecular weight to study the influence of molecular
size and charge density on antimicrobial activity and found a par-
abolic dependence of molecular weight on antimicrobial activity
[34]. Considering the mechanism of antimicrobial action by cat-
ionic materials, this complex relationship between molecular
weight and antimicrobial activity is most likely due to the inter-
play between charge density and diffusivity. With increasing poly-
cation molecular weight, the process of adsorption onto the
bacterial cell surface would be expected to be enhanced due to a
higher charge density, while the process of diffusion into the inte-
rior of the cell wall would be expected to be inhibited.
Considering the previous reports on polymeric biocides, it was
expected that antimicrobial activity would have increased with
increasing molecular weight for the series of PEI–TCS copolymers
investigated. The highest molecular weight PEI–TCS copolymer
was derived from a 10,000 g/mole PEI. This molecular weight is well
100
100
(a)
(b)
80
80
60
60
423PEI-5%TCS
600PEI-5%TCS
423PEI-10%TCS
600PEI-10%TCS
40
40
10,000PEI-5%TCS
10,000PEI-10%TCS
20
0
20
0
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
Concentration (micrograms/ml)
Concentration (micrograms/ml)
100
(c)
80
60
40
20
0
423PEI-20%TCS
600PEI-20%TCS
10,000PEI-20%TCS
0
2
4
6
8
10
12
14
Concentration (micrograms/ml)
Fig. 7. Inhibitory response of the PEI–TCS copolymers toward E. coli. PEI–TCS copolymers containing 5 (a), 10 (b), and 20 (c) mole percent pendent TCS moieties.

A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
75
below any molecular weights previously reported to be too high to
provide effective antimicrobial activity. To obtain a better under-
standing of this trend in molecular weight, the surface activity of
copolymers possessing 5 mol% pendent TCS groups was character-
ized by measuring surface tension as a function of polymer concen-
tration. As shown in Fig. 8, the two low molecular weight
copolymers, 423PEI–5%TCS and 600PEI–5%TCS, were significantly
more surface active than the higher molecular weight copolymer,
10,000PEI–5%TCS. The higher surface activity observed for the
two lower molecular weight copolymers would be expected to
provide enhanced interaction with the bacterial cell wall which
most likely contributed to the enhanced antimicrobial activity ob-
served for these copolymers. In addition, the surface activity of
600PEI–20%TCS was measured to understand the influence of TCS
content on surface activity. A comparison of the surface tension
data obtained for 600PEI–20%TCS to that of 600PEI–5%TCS, as
shown in Fig. 8, indicated that increasing TCS pendent group con-
centration increased surface activity. Thus, the higher antimicrobial
activity observed for 600PEI–20%TCS as compared to 600PEI–5%TCS
may be partly attributed to the higher surface activity of the former
resulting in an enhanced affinity for the bacterial cell wall.
To compare the overall effectiveness of the TCS-containing
copolymers to pure TCS, Fig. 9 was constructed. In contrast to
Figs. 6 and 7, which express concentration on the x-axis as polymer
concentration, the x-axis in Fig. 9 displays concentration as the
concentration of TCS-moieties. By expressing concentration in this
manner, it can be seen that attaching TCS moieties to the low
molecular weight PEI provides greater antimicrobial activity to-
ward E. coli than pure TCS. This characteristic coupled with the fact
that the TCS-containing copolymers are highly aqueous soluble liq-
uids as opposed to a crystalline solid of limited solubility may af-
ford utility of these copolymers for a variety of applications.
4. Conclusion
75
It was discovered that copolymers produced by reacting ETCS
with PEI produced antimicrobial polymers. Although the mecha-
nism of antimicrobial activity for these novel polymers is un-
known, the influence of molecular weight on antimicrobial
activity suggested that the mechanism is highly dependent on
the interaction of the polymer with the bacterial cell wall. For most
polymeric antimicrobials that function by disrupting the integrity
of the cell envelope, such as cationic polymers, a molecular weight
dependence on antimicrobial activity is often observed. This
molecular weight dependence stems from the interplay between
cationic charge density and diffusivity. High molecular weights
and high charge densities would be expected to enhance adsorp-
tion of a cationic polymer onto the bacterial cell wall, however,
these same characteristics would be expected to impede diffusion
of the polymer into the interior of the cell wall.
For the PEI–TCS copolymers produced, antimicrobial activity
decreased with increasing copolymer molecular weight which
was counter-intuitive considering previous work described in the
literature for polymeric antimicrobials. Surface tension measure-
ments made as a function of copolymer molecular weight showed
that surface activity at a constant TCS pendent group concentration
of 5 mol% increased with decreasing molecular weight. Since this
trend was similar to the trend found between molecular weight
423PEI-5%TCS
600PEI-5%TCS
10,000PEI-5%TCS
600PEI-20%TCS
70
65
60
55
50
45
0.001
0.01
0.1
1
10
100
Concentration (wt. %)
Fig. 8. Surface tension as a function of PEI–TCS copolymer concentration in water.
100
100
(b)
(a)
80
60
80
60
TCS
TCS
423PEI-5%TCS
423PEI-5%TCS
423PEI-10%TCS
423PEI-20%TCS
423PEI-10%TCS
423PEI-20%TCS
40
40
20
0
20
0
0
2
4
6
8
10
12
14
0
2
4
6
8
10
12
14
Concentration of TCS Moieties (micrograms/ml)
Concentration of TCS Moieties (micrograms/ml)
Fig. 9. A comparison of the antimicrobial activity of the copolymers derived from the lowest molecular weight PEI (i.e. 423 g/mole) to TCS. For this comparison, concentration
is expressed as the concentration of TCS pendent groups as opposed to copolymer concentration. The two bacteria used were S. epidermidis (a) and E. coli (b).

76
A.J. Kugel et al. / Reactive & Functional Polymers 72 (2012) 69–76
[15] B.D. Kalyon, U. Olgun, Am. J. Infect. Control 29 (2001) 124–125.
[16] P. Majumdar, E. Crowley, H. Maung, S.J. Stafslien, J. Daniels, L. Vander Wal, B.J.
Chisholm, ACS Combi. Sci. 13 (2011) 298–309.
[17] Z. Chen, B.J. Chisholm, S. Stafslien, J. He, S. Patel, J. Biomed. Mater. Res. 95A
(2010) 486–494.
and antimicrobial activity, it supports the suggestion that the
mechanism of antimicrobial activity involves interaction with the
bacterial cell wall.
[18] P. Majumdar, E. Lee, N. Gubbins, D.A. Christianson, S.J. Stafslien, J. Daniels, L.
Vander Wal, J. Bahr, B.J. Chisholm, J. Combi. Chem. 11 (2009) 1115–1127.
[19] B.J. Chisholm, D.A. Christianson, S.J. Stafslien, C. Gallagher-Lein, J. Daniels, in:
ACS Symp. Ser. 1002(Smart Coatings II), (2009) pp. 127–141.
[20] S. Alam, B.J. Chisholm, Polym. Prepr. (ACS, Div. Polym. Chem.) 50 (2009).
[21] P. Majumdar, E. Lee, N. Patel, K. Ward, S.J. Stafslien, J. Daniels, B.J. Chisholm, P.
Boudjouk, M.E. Callow, J.A. Callow, Biofouling 24 (2008) 185–200.
[22] R. Kugler, O. Bouloussa, F. Rondelez, Microbiology 151 (2005) 1341–1348.
[23] J.M. Andrews, J. Antimicrob. Chemother. 48 (2001) 5–16.
[24] J. Han, S.K. Kim, T.-S. Cho, J.-C. Lee, H.S. Joung, Macromol. Res. 12 (2004) 501–
506.
[25] H. Khalil, T. Chen, R. Riffon, R. Wang, Z. Wang, Antimicrob. Agents Chemother.
52 (2008) 1635–1641.
[26] I.M. Helander, H.-L. Alakomi, K. Latva-Kala, P. Koski, Microbiology 143 (1997)
3193–3199.
[27] P. Gilbert, L.E. Moore, J. Appl. Microbiol. 99 (2005) 703–715.
[28] P. Gilbert, A. Al-Taae, Lett. Appl. Microbiol. 1 (1985) 101–104.
[29] N.N. Daoud, N.A. Dickinson, P. Gilbert, Microbious 37 (1983) 73–85.
[30] T. Ikeda, H. Yamaguchi, S. Tazuke, J. Bioact. Compat. Polym. 5 (1990) 31–41.
[31] A. Kanazawa, T. Ikeda, T. Endo, J. Polym. Sci. Part A: Polym. Chem. 31 (1993)
1441–1447.
Acknowledgment
The authors acknowledge the Office of Naval Research for finan-
cial support through Grants N00014-05-1-0822 and N00014-06-1-
0952.
References
[1] E.R. Kenawy, S.D. Worley, R. Broughton, Biomacromolecules 8 (2007) 1359–
1384.
[2] A.L. Demain, L. Zhang, L. Nat. Prod. 1 (2005) 3–29.
[3] R. Eycott, Specialty Chem. 16 (1996) 156.
[4] G. Bognolo, Chim. Oggi. 23 (2005) 24–25.
[5] I. Omae, Chem. Rev. 103 (2003) 3431–3448.
[6] A. Ogaki, Techno-Cosmos 16 (2003) 32–37.
[7] R.D. Jones, H.B. Jampani, J.L. Newman, A.S. Lee, Am. J. Infect. Control 28 (2000)
184–196.
[8] H.P. Schweizer, FEMS Microbiol. Lett. 202 (2001) 1–7.
[9] H.N. Bhargava, P.A. Leonard, Am. J. Infect. Control 24 (1996) 209–218.
[10] L. Campbell, M.J. Zirwas, Dermatitis 17 (2006) 204–207.
[11] S.P. Yazdankhah, A.A. Scheie, E.A. Hoiby, B.T. Lunestad, E. Heir, T.O. Fotland, K.
Naterstad, H. Kruse, Microb. Drug Resist. 12 (2006) 83–90.
[12] P. Gilbert, A.J. McBain, Surg. Infect. 3 (Suppl. 1) (2002) S55–S63.
[13] A.D. Russell, J. Antimicrob. Chemother. 53 (2004) 693–695.
[14] J. Lu, M.A. Hill, M. Hood, D.F. Greeson, J.R. Horton, P.E. Orndorff, A.S. Herndon,
A.E. Tonelli, J. Appl. Polym. Sci. 82 (2001) 300–309.
[32] J.C. Tiller, C.-J. Liao, K. Lewis, A.M. Klibanov, Proc. Natl. Acad. Sci. USA 98 (2001)
5981–5985.
[33] J. Lin, S. Qiu, K. Lewis, A.M. Klibanov, Biotechnol. Bioeng. 83 (2003) 168–172.
[34] C.Z. Chen, N.C. Beck-Tan, P. Dhurjati, T.K. van Dyk, R.A. LaRossa, S.L. Cooper,
Biomacromolecules 3 (2000) 473–480.