10

Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org

Volume 5, Issue 4 - 2010
Antimicrobial Properties and Release Profile of Ampicillin
from Electrospun Poly(-caprolactone) Nanofiber Yarns
Hang Liu
1
, Karen K. Leonas
1
, Yiping Zhao
2


1
Department of Apparel, Merchandising, Design and Textiles, Washington State University,
Pullman, WA 99164, USA

2
Department of Physics and Astronomy, University of Georgia, Athens, GA 30602, USA

Correspondence to:
Hang Liu, e-mail: hangliu@wsu.edu

ABSTRACT
Poly(-caprolactone) (PCL) electrospun fibers
containing ampicillin sodium salt have been
produced and twisted into nanofiber yarns. The fiber
diameters and crystallinity, the in vitro antimicrobial
properties of the yarns, and the in vitro release of
ampicillin from yarns containing various ampicillin
concentrations are studied. Decreased fiber
diameters and reduced diameter variation are
observed with the addition of ampicillin salt into the
polymer solution. The results from the zone of
inhibition test of the yarns against both gram-positive
Staphylococcus aureus and gram-negative Klebsiella
pneumoniae indicate that the released ampicillin
retains its effectiveness after the production
processes, therefore the as-spun yarns are
antimicrobial active. A burst release of ampicillin
from the yarns has been observed in the first hour,
and the release is almost completed in 96 hours. The
burst release is believed to be due to the low
compatibility of ampicillin with PCL, the
accumulation of ampicillin on fiber surface and the
small fiber diameters. An empirical release model is
developed to describe the release profile. The results
indicate that the electrospun nanofibers yarns will
have a great potential to be used for biomaterials,
such as surgical sutures, to decrease the surgical site
infection rate.

INTRODUCTION
With the wide use of implantable biomaterials, the
associated surgical site infections (SSIs) have been
receiving more and more attention. Approximately
15% of the 2.4 million cases of hospital-acquired
infections (HAIs) each year in the United States are
SSIs, which accounts for the second most commonly
reported cause of HAI [1-3]. Although advanced
infection control practices have been recommended
by the Centers for Disease Control (CDC), the HAI
rate has not decreased, and, instead, has increased by
36% in the past 20 years [4]. A completely sterile
surgery is almost impossible and there are several
sources for microorganisms to access the surgical site
and cause infections [5]. In addition, in the new era
of biomaterial-implants, a number of factors make
the infection more likely to occur. First, some non-
pathogenic normal skin flora becomes a potential
pathogen with the presence of biomaterials [6].
Second, the use of biomaterials has shown to
significantly lower the critical number of
microorganism required to cause infection. An
example is that the number of bacteria required to
cause infection with the presence of silk sutures was
reduced 10,000-fold of that needed without sutures as
reported by Elek and Conen [7]. Third, relapses to
antibiotic therapy are frequent after an initial
response due to the formation of bacterial biofilms on
the surface of biomaterials [6, 8]. The consequence
of SSIs includes longer hospital stays, extra financial
costs, increased physiological and psychological
burdens on patients and their families, and sometimes
requiring an additional surgery to remove the infected
implant [1, 5, 9].

The traditional oral antibiotic therapy is not effective
all the time, and it can have adverse side effects. In
order to reduce SSIs when bio-implants are involved,
a growing interest among material scientists is to
provide antimicrobial properties to the implantable
biomaterials. Therefore, in addition to serving its
primary function, the implant will also help prevent
the formation of bacterial biofilms and the released
11
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
antimicrobial agent would kill or inhibit the growth
of bacteria, thereby reducing the potential of SSIs.

Currently, the primary method of providing
antimicrobial functions to fiber-based biomaterials is
by topical coating, a finishing process that occurs
after fiber production. Due to the high temperature
used in melt spinning, it is not practical to
incorporate antimicrobial agents into the fibers
during the production and keep their effectiveness.
Decamethoxin (DMO) (a quaternary ammonium
compound), triclosan, silver ions and silver
compounds are among the antimicrobial agents used
to coat surgical sutures [4, 10-14]. Kovtun et al. [11]
studied DMO release of coated polycaproamide
threads in water and the results showed that 70% to
100% of the DMO was released in 13 days. In the
first 24 hours the release rate was the highest.
However, the selection of a proper coating material to
adhere the fiber and the antimicrobial agent together
and the application of the coating onto sutures make
the method complicated. A recent fiber fabrication
technique, electrospinning which has the advantages
of producing polymer fibers and of combining two or
more different materials at room temperature, makes
this mission less complicated [15]. Simple
equipment, structural integrity of electrospun
products, the elimination of using high temperature,
and being able to produce filament nanofibers are
other features that enable electrospinning to be an
attractive technique to produce biomaterials with
additives [15, 16]. Previous research has studied the
incorporation of drugs, DNA and proteins, into
electrospun fibers and their release profiles [17-25].
These studies have shown various release profiles as
the polymers, additives and electrospinning
processing parameters varied. For example, in a
study on the release of Mefoxin from poly(lactide-co-
glycolide) electrospun scaffold, Kim et al. found that
the maximum release of the drug in vitro was at one
hour incubation and all the drug was completely
released after six hours [20]. In the study of the
release of heparin from electrospun PCL mats by
Luong-Van et al., the results showed that half of the
encapsulated heparin was released from electrospun
PCL mats after 14 days [22]. In electrospinning,
depending on the properties of polymers, additives in
polymer solution and spinning parameters, the fiber
properties, including fiber diameter, fiber
crystallinity, and molecular orientation, differ
dramatically [26-36]. These fiber properties
influence the release profile of the added chemicals.
To date, all reported studies on controlled release
focus on the preparation of and release from
electrospun scaffolds/webs, in which the fibers are
randomly oriented. However, to fabricate fibers into
yarns is important to those biomaterials based on
yarn form, such as surgical sutures and woven wound
dressings. The alignment of fibers in a yarn is
essential to maintain the integrity of the yarn as a
continuous strand; this is critical to provide the yarn
with adequate mechanical properties needed when
the end product is in use. In this paper, poly(-
caprolactone) (PCL) (a biodegradable polymer used
in devices approved by the Food and Drug
Administration) electrospun fiber yarns containing an
antimicrobial agent, ampicillin sodium salt has been
prepared. The electrospun fibers were collected as
well-aligned fiber bundles instead of fiber webs using
a rotating fiber collector. The in vitro antimicrobial
properties have been evaluated by a zone of
inhibition method against Staphylococcus aureus and
Klebsiella pneumonia. The in vitro release of
ampicillin from yarns into phosphate buffer solution
(PBS) is examined using UV-VIS spectrophotometer.

MATERIALS AND METHODS
Materials
The polymer, PCL (Mn = 80,000) (Sigma-Aldrich
Chemical), the solvents, chloroform and methanol
(Fisher Chemical), and the antimicrobial agent,
ampicillin sodium salt (Mediatech Inc.), were all used
as received. PCL is a widely used biodegradable
polymer for biomedical applications, such as surgical
sutures and drug delivery devices. Ampicillin is a
broad-spectrum antimicrobial agent with high
selectivity and is in the beta-lactam antibiotics class,
which inhibit bacteria cell wall synthesis by binding
to specific penicillin-binding proteins (PBPs) located
inside the bacterial cell wall.

Electrospinning of Nanofibers and Yarn
Preparation
Nanofibers were electrospun from PCL solutions and
PCL solutions containing various concentrations of
ampicillin under ambient conditions using a home-
made electrospinning system. PCL solutions were
prepared by stirring the mixture of PCL pellets,
solvent(s) and/or ampicillin for one hour (using a
stirring bar on a NUOVA hot plate with a stirring
speed of two) to obtain a homogeneous solution.
Chloroform was used as the solvent for the pure PCL
solution. The hydrophilic ampicillin sodium salt did
not dissolve in chloroform, therefore to obtain a
uniform solution with ampicillin, a mixture of polar
methanol and non-polar chloroform (volume ratio of
methanol to chloroform was 1:9) was used. The PCL
concentration in the polymer solution was 14% (w/v).
12
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
The relative concentrations of ampicillin to PCL
varied from 0 to 20% by weight. The prepared
polymer solution was fed into a syringe (Monoject
Sterile 3cc Luer-lok) with a right angle-shaped metal
needle (Monoject Disposable 21 GA 1 IN). The
syringe was connected to a high voltage power
supply (Gamma, ES50P - 10W) and was mounted on
a syringe pump (KD Scientific, KDS100). A custom-
designed rotating drum of 1 cm wide and 10 cm in
diameter was used to collect well-aligned fibers. The
rotating speed of the collector was 1500 revolutions
per minute (rpm). The distance between the needle
tip to the drum surface was fixed at 8 cm. Fibers
produced were wound onto an aluminum foil strip
attached to the circumference of the drum, which was
grounded, under the conditions of 9 kV voltage and
0.3 ml/h feedrate to form fiber bundles. After being
stored for overnight to allow for complete
evaporation of the solvent(s), the fiber bundles were
removed from the aluminum strip and twisted into
yarns (5 twists per inch) using a hand-driven twist
tester (Alfred Suter Co. Inc). Each yarn was made of
fibers spun from 0.1 ml of polymer solution. The
overall length of the yarns was approximately 31 cm.

Fiber Characterization
The surface morphology of the fibers was
characterized using a Zeiss 1450EP Scanning
Electron Microscopy (SEM). Fiber samples were
sputter coated with gold for 60 seconds using a SPI
ModuleTM Sputter Coater to make the surface
conducting for SEM observations. The thickness of
the gold coating was estimated to be 15.3 nm. The
average diameter and angular variation of the fibers
were analyzed using ImageTool 3.0 based on 30
measurements of each sample. The angular variation
was a measurement of the fiber alignment.

The percentage of fiber crystallinity was measured
using Differential Scanning Calorimetry (Mettler
Toledo DSC821e) with temperature ranging from
60C to 100C with ambient air. The heating rate
was 10C/min. The enthalpy of fusion value for PCL
with 100% crystallinity was 139.5 J/g as reported in
the literature [37]. This value was used to calculate
the percentage of crystallinity of the nanofibers.

Zone of Inhibition Test
The in vitro antimicrobial properties of the
electrospun yarns containing ampicillin were
evaluated by the zone of inhibition method against
two potential human pathogens, the Gram-negative
Klebsiella pneumoniae (American Type Culture
Collection No. 4352) and the Gram-positive
Staphylococcus aureus (American Type Culture
Collection No. 6538). The two organisms are
recommended by American Association of Textile
Chemists and Colorist (AATCC) for antibacterial
activity assessment of textile materials in AATCC
Standard Test Method 147. The yarns were folded
and loosely hand-twisted into approximately 30 mm
long and placed in the center of a Nutrient Agar plate
coated with the challenge bacteria. The
concentration of the challenge bacteria was
approximately 10
5
CFU/ml. The plates were
incubated at 37C for 24 hours, and the zone of
inhibition was measured (details see Fig. 4). PCL
yarns without ampicillin addition were used as
control and the result showed there were no clear
zones formed for the control samples. Three
replications for yarns with the same ampicillin
concentration were examined.

Ampicillin Release Test
The in vitro release of ampicillin from the
electrospun yarns was investigated by measuring the
concentration of ampicillin released in the phosphate
buffer solution (PBS) (with a pH of 7.4). Yarns were
immersed into 25 ml of PBS in centrifuge tubes,
which were placed in a 37C shaking bath (Precision
Reciprocal Shaking Bath) with a shaking speed of 50
rpm. At specific time intervals, the absorption of the
PBS solution was determined by a UV-VIS
spectrophotometer (Shimadzu UV-2401 PC UV-VIS
Recording Spectrophotometer) at the absorbance
peak with a wavelength of 220 nm. The absorbance
peak correlated very well with the concentration of
ampicillin. A linear calibration curve was obtained
by measuring the absorption of solutions with
predetermined ampicillin concentrations as shown in
Figure 1. The obtained relationship is shown in the
following equation (Eq. (1)):

Solution concentration = 57.39 x Absorption (1)
The chi-square was 0.00063.
FIGURE 1. Calibration curve showing the relationship between
ampicillin concentration in PBS and absorbance.

13
http://www.jeffjournal.org Journal of Engineered Fibers and Fabrics
in
bers. The release behaviors of three samples with
were measured.
ESULTS AND DISCUSSION
For all the measurements in this study, the
absorbance readings were within the calibration
range. The percentage of released ampicillin was
determined by dividing the amount of ampicillin
released to the PBS by the amount of ampicillin
originally contained in the yarn sample. The original
amount of ampicillin in the fiber was determined by
the prepared mixture for fiber spinning and was
calculated by multiplying the concentration of
ampicillin in the polymer solution before spinning
and the amount of polymer solution used to sp
fi
the same ampicillin concentration

R
Fiber Characterization

Volume 5, Issue 4 - 2010
FIGURE 2. Representative SEM images of fibers spun with 14%
CL and various ampicillin concentrations: (a) 0%; (b) 8%; (c)
n enlarged fiber
segment, the fiber surface is smooth without pores,
cracks, or other physical defects.
th t [31].
ers is
eneral,
th slightly
lo e without
a picillin
m ains
ng the
e and
rength
an
to obtain yarns with
eal mechanical properties.


P
12%; (d) 16% and (e) 20%. (The scale bar represents 5 m.)


Figure 2 shows representative SEM images of
electrospun fibers with varying ampicillin
concentrations. All the images show that under the
current electrospinning conditions, uniform fibers are
obtained. All the fibers have good alignment along
the rotating direction of the collector. As shown in
Figure 3, an SEM image of a



FIGURE 3. Representative SEM image showing the smooth
surface of a fiber (The scale bar represents 500 nm.)
Table I provides information on fiber diameters,
angular variations and percentages of crystallinity.
The angular variations are between 5
o
and 9
o
, which
indicates the effectiveness of using a rotating drum to
collect well-aligned fibers. The fiber diameters are
between 260 nm to 330 nm. The results show that
the addition of ampicillin sodium salt to the polymer
solution results in reduced fiber diameters and
diameter variation, which is believed to be due to the
ionic nature of the ampicillin salt. The ions in the
solution contribute to the charge building at the tip of
the needle, which magnifies the effect of the electric
field, resulting in smaller diameters. Furthermore,
the enhanced conductivity of the solution with the
salt reduces the fiber diameter variation as a result of

e sufficient transfer of charges into the je
The percentage of crystallinity in the fib
measured to be between 40% and 50%. In g
e fibers spun with ampicillin salt have a
wer percentage of crystallinity than thos
mpicillin salt as the existence of am
olecules in between the PCL polymer ch
prevent their close packing, thereby reduci
crystallinity. These ratios of crystallin
amorphous regions provide fibers with both st
d elongation. Some preliminary tensile tests
showed that the load-elongation curve of the
FIGURE 4. Representative load-elongation curve of PCL yarns

electrospun yarns is similar to that of conventional
yarns in shape. Figure 4 is a load-elongation curve
of electrospun yarns containing 8% ampicillin. The
load at break is approximately 3 N and the elongation
at break is 65%. Further tests are undertaken for the
tensile strength of the yarns. It is believed that by
adjusting the spinning parameters, the fiber
icrostructure can be tailored m
id


14
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010



TABLE I. Physical properties of fibers with various concentrations of ampicillin

Ampicillin Concentration 0% 8% 12% 16% 20%
Fiber Diameter (nm)
(Mean Deviation)
330 250 320 90 300 110 280 170 260 70
Angular Variation (degree) 6.3 5.4 8.3 8.7 6.8
Percentage of Crystallinity (%) 49.5 43.8 41.0 43.1 43.5



TABLE II. The Zone of Inhibition

Zone of Inhibition D (mm)

0% 2% 5% 8% 12% 16% 20%

Staphylococcus aureus 0 0 38 1.5 49 0.6 50 0.6 52 0.6 52 1.0 53 0.6
Klebsiella pneumoniae 0 0 0 0 14 1.0 18 1.0 22 0.6 24 0.6 26 0.6


Zone of Inhibition
Figure 5 shows some representative images of the
zone of inhibition test results for Klebsiella
pneumonia and Staphylococcus aureus. All samples
except the 2% ampicillin concentration against
Klebsiella pneumoniae result in a zone of inhibition,
which indicates that the released ampicillin retains its
efficacy after being dissolved in the polymer
solution, spun into fibers and twisted into yarns.
Neither the solvent nor the spinning process
influences the antimicrobial efficacy. The clear area
around the yarn shown in Figure 5(c) without
bacteria growth is defined as the zone of inhibition,
and its size is labeled as D. The D values of yarns
containing 2, 5, 8, 12, 16 and 20% ampicillin are
listed in Table II. The size of the clear zone increases
monotonically with the ampicillin concentration in
the fibers for the two different bacteria study here.
However, the increment in D value is more
pronounced at the low ampicillin concentrations from
2% to 5%, and 5% to 8%; while from 8% to 20% and
all levels in between, the increment is not as
remarkable. This is probably due to the limit
penetration and dispersion characteristics of
ampicillin in agar.





oniae; (d) 2%, (e) 12%, and (f) 20% against
taphylococcus aureus. (The definition of the clear zone size, D, is
so illustrated.)


FIGURE 5. Zone of inhibition photos of yarns containing
ampicillin concentration of(a) 2%, (b) 12%, and (c) 20% against
Klebsiella Pneum
S
al


Comparing the effects of the ampicillin salt on the
two bacteria studied, the results show that
Staphylococcus aureus is more sensitive to ampicillin
than Klebsiella pneumoniae. When the ampicillin
concentration is 2%, no zone of inhibition is observed
for Klebsiella pneumoniae; while for Staphylococcus
aureus, D = 38 mm. When the concentration is
increased to 20%, the clear zone for Klebsiella
D
15
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
pneumoniae (D = 26 mm) is less than half of that for
Staphylococcus aureus (D = 53 mm). The resistance
f Klebsiella pneumonia to beta-lactam antibiotics
to the bla
SHV
gene [38].
o
has been reported due

Ampicillin Release
The percentage of ampicillin released versus the
release time from yarns containing 8, 12, 16, and
20% ampicillin is shown in Figure 6. A burst
depletion of more than 50% is observed upon the
immersing of yarns into PBS for all different
ampicillin concentrations. The release rate slows
down dramatically after the first hour. After 24
hours, more than 95% of the ampicillin has been
released from the yarns and is dissolved in the PBS
solution; and the release process is almost complete
within 96 hours. The overall quick release of
mpicillin from PCL yarns is thought to be due to the
) The lower compatibility of ampicillin in PCL
n nearly 3 months was less than 3%. Other
udies also showed that PCL degrades slowly [40,
om saturated. Therefore, the high solubility of the
, ampicillin is
hysically trapped in the fibers and ready to dissolve
in water when the water penetrates.

a
following three reasons:

1
fibers than that in PBS

There are two mechanisms of drug release from
biodegradable polymers, small-molecule-diffusion-
controlled release and polymer-degradation-driven
release [39]. In the diffusion-controlled release
process, the aqueous medium first penetrates into the
matrix of the polymer and the drug. In the matrix,
the drug dissolves in the medium, diffuses within the
matrix, and finally migrates from the polymer to the
medium because of the concentration gradient
through the pores or capillaries formed by the
medium. The polarities of drugs and release medium
are decisive in the diffusion-controlled release
process. This release mechanism is therefore
applicable only to those drugs that have the same
polarity as the aqueous medium. For those drugs that
do not dissolve in the release medium, polymer
degradation is the primary driving force for drug
release. In this study, the ampicillin salt is
hydrophilic, while PCL is lipophilic. The quick
release rate suggests that the diffusion of ampicillin
through the PCL fibers is the primary releasing
mechanism as degradation of the polymer has not
begun in this limited time period. Study on the
degradation of the PCL yarns showed that the weight
loss i
st
41].

There are several factors influencing the rate of
diffusion-controlled drug release: 1) dissolvability of
the drug in the release medium (the solubility of
ampicillin in PBS in this study); 2) ease of the
medium to penetrate the polymer and drug matrix
(the penetration of PBS into fibers); and 3) physical
attractions or chemical bonds between the polymer
and the drug (attraction between ampicillin and
PCL). The dissolvability of the ampicillin sodium
salt in water is approximately 50 mg/ml (based on
MSDS of ampicillin sodium salt). For the samples
used in this study, less than 3 mg of ampicillin is
contained in all of the yarn samples. With all the
ampicillin in the fibers dissolving in the 25-ml PBS,
the possible maximum ampicillin concentration is far
fr
ampicillin salt in water is in favor of its fast release.

In terms of the migration of ampicillin inside the
fiber structure, the unordered amorphous regions
provide more open space than the crystalline regions;
thus, it is easier for water to penetrate and small
molecules to move in the amorphous regions. The
crystallinity of electrospun fibers with ampicillin is
no more than 50%, leaving more than 50% of the
fiber in amorphous phase, which facilitates the
release of ampicillin. Theoretically, there are no
primary bonds or hydrogen bonds formed between
PCL and ampicillin. Therefore
p


6. Release FIGURE of ampicillin from yarns containing different
mpicillin concentrations measured by UV-VIS
ophotometer.
a
spectr






16
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
2) The accumulation of ampicillin on the surface of
the fibers

In addition to the low compatibility of ampicillin in
PCL fibers, the accumulation of ampicillin on the
surface or near-surface of the PCL fibers could result
in the burst release during the first hour. Upon
immersing the samples into PBS, more than 50% of
the ampicillin contained in fibers is depleted. The
possible explanation for this quick release during
such a short time period is the accumulation of
ampicillin on the fiber surface. After dissolving the
ampicillin on or near the fiber surface, the rate of
release then decreases dramatically. For example, for
yarns containing 12% ampicillin, after the yarn was
immersed into PBS, in 20 minutes 73.6% of
ampicillin was released from the yarn. From 20
minutes to two hours, only 14.3% more was released
hydrochloride is mostly on the surface of the PLLA
PCL, PLA and their
copolymer electrospun fiber mats. These studies
is.
n the release study are between 260 nm and 330
er model and the Higuchi model, we
ropose the eq (
file
rst
E III. The fitting paramete lease data to
. (1).
ing P ers
in this time period. The accumulation of ampicillin
on fiber surfaces could be understood from the
spinning process.
First, the ampicillin ions migrate to the surface of the
polymer jet when it is flying from the needle tip to
the collector due to the ionic strength [20]. Second,
since the ampicillin sodium salt and PCL polymer
have different polarities, they are incompatible with
each other. Therefore, during electrospinning, with
the evaporation of methanol, two processes might
happen at the same time: one is the phase separation
of the polymer and the ampicillin sodium salt. The
ampicillin is carried by methanol toward the direction
of its evaporation, which is toward fiber surfaces.
The other process is the aggregation of adjacent
ampicillin molecules into larger particles. Both
processes take place simultaneously during the jet
flying and after the deposition of fibers on the
collector until all solvents are evaporated. Both
processes result in quick release. A study by Zeng et
al. [25] on the distribution of various drugs in poly(l-
lactic acid) (PLLA) fibers has demonstrated the
influence of the compatibility of the drug-polymer-
solvent system on the distribution of drugs in
electrospun fibers. They have found that lipophilic
rifampin and paclitaxel distribute evenly in PLLA
electrospun fibers, while hydrophilic doxorubicin
fibers. Similar results were reported by Buschle-
Diller et al. [42] in their study of
support the authors hypothes

3) The small fiber diameter

The release of ampicillin is associated with the
penetration of water into the fibers and the
dissolution of ampicillin in water. The smaller the
fiber diameter, the shorter the time needed for water
to penetrate the fiber. The average fiber diameters
used i
nm. This could be one reason for the high release
rate.

An initial burst depletion followed by a slow release,
the similar release behavior, has been reported by
several other studies, including the release studies of
Mefoxin from PDLA electrospun film by Zhong et al.
[43] and Mefoxin from a PLGA nanofibrous scaffold
by Kim et al. [20]. There are three models
commonly used to explain drug release profiles, the
first-order Ficks equation, the semi-empirical
Higuchi model and theoretical Roseman and Higuchi
model [44-48]. The Ficks first-order equation is
based on the non-steady state diffusion; the Higuchi
model describes the drug release from a single face of
a non-swelling tablet; and the Roseman and Higuchi
model describes the release of drug from a cylindrical
polymer matrix. However, none of the models fit the
data obtained from this study. By combining the
Fickian first-ord
p following uation (Eq. 2)) as the
: release pro

ln (A
1
A
2
M
t
/M
o
)=-kt
1/2
(2)

where A1 and A2 are constants, related to the bu
release at the moment of the yarns immersed into
PBS; k is a constant reflecting the overall release

TABL
Eq
rs of the ampicillin re

Fitt aramet
Ampicillin
centration con 8% 12% 16% 20%
A
1
2.1 0.1 2.4 0.2 2.2 0.1 2.3 0.1
A
2
2.1 0.1 0.2 0.1 0.1
0.75 0.08 0.8 0.1 1.0 0.1 0.94 0.08
R
2
0.98 0.97 0.98 0.99
2.4 2.3 2.3
k
17
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010

cal
llin released is higher than
eters, fiber crystallinity and
itial ampicillin load.
iber diameters and improves the
iameter variation.
t an
dditional after-production finishing process.
Antonio for providing
e free ImageTool software.


IGURE 7. The ampicillin release data fitted to the empiri F
model for yarns with different ampicillin concentrations.

rate; M
0
is the total amount of drug in the sample;
and M
t
is the amount of drug released at time t. The
release data are fitted by the above equation and the
solid curves in Figure 7 are the results of the fitting
(The release curves are intentionally offset for
different ampicillin concentrations to show all the
fittings). The fitting parameters for different
ampicillin concentrations are summarized in Table
III. The goodness of the fitting, R
2
, is better than
0.97 for all of the different ampicillin concentrations.
The rate constant k increases with the increasing in
ampicillin concentration in fibers from 8% to 12%
and 16%. The parameters are valid when the
percentage of ampici
approximately 60%.

The fitting parameters shown in Table III
demonstrate that the release profiles for different
ampicillin concentrations only differ slightly
regardless of the structural difference. This implies
that the primary release mechanism for drug release
is the same for fibers containing different ampicillin
concentrations. The slight difference is due to the
variations in fiber diam
in




CONCLUSION
Electrospun poly(-caprolactone) nanofibers
containing ampicillin have been produced and
collected as well-aligned fiber bundles using a
custom-made electrospinning system. The fiber
bundles are then twisted into yarns. The alignment
and size of the fibers, the antimicrobial effectiveness
of the yarns on two microorganisms, Gram-positive
Staphylococcus aureus and Gram-negative Klebsiella
pneumoniae, and the release profile of ampicillin
from yarns are evaluated. The addition of ampicillin
sodium salt reduces f
d

The zone of inhibition results suggest that the
ampicillin retains its antimicrobial efficacy after
being incorporated into the polymer solution, spun
into fibers and twisted into yarns, and released from
the fibers. This antimicrobial agent is more
effectively against Staphylococcus aureus than
Klebsiella pneumoniae. The ampicillin is almost
completely released from the fibers with different
ampicillin concentrations in 96 hours. A burst
depletion of more than 75% is observed in the first
hour for fibers with all different ampicillin
concentrations. The burst depletion is probably due
to the low compatibility of ampicillin sodium salt and
the PCL polymer, the accumulation of ampicillin on
the fiber surface and the small fiber diameter. The
burst release is favorable to reduce SSIs since most of
the bacteria are introduced to the surgical site at the
time of surgery or in the immediate after the surgery
[1]. The ampicillin release profile could be fitted by
an empirical model, and the release mechanisms of
fibers with different ampicillin concentrations are
very similar. The study suggests that there is a great
potential to fabricate biomaterials, such as surgical
sutures, using electrospinning and provide inherent
antimicrobial properties to the materials withou
a


ACKNOWLEDGEMENT
The authors would like to thank the American
Association of Textile Chemists and Colorists
Research Foundation for providing partial financial
support for this project and the University of Texas
Health Science Center at San
th



18
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
REFERENCES
[1] Inserting and replacing GORE DUALMESH PLUS
biomaterial inhibits implant infections by drug-resistant
'Superbugs" and resists biofilm development, Vol. 2008,
Business Wire, 2007.
[2] Bide, M., et al., Fiber-antibiotic interactions in the
development of infection resistant sutures, AATCC Review, 7,
No.9, 2005, 44-48.
[3] Mangram, A.J., et al., Guideline for prevention of surgical
site infection 1999, Infect. Control. Hosp. Epidemiol., 20,
1999, 247-278.
[4] Ethicon: a johnson & johnson company, Ethicon receiveds
FDA clearance to market Vicryl* Plus, first ever antibacterial
suture, Vol. 2006, 2002.
[5] Maathuis, P.G.M., S.K. Bulstra, in: G. Rakhorst and R. J.
Ploeg (Eds.), Biomaterials in modern medicine, World
Scientific Publishing Company, New Jersey, 2008, pp. 97-
112.
[6] Costerton, J.W., et al., Bacterial biofilms in nature and
disease, Annu. Rev. Microbiol., 41, 1987, 435-464.
[7] Elek, S.D., P.E. Conen, The virulence of S. pyogenes for
man. A study of the problems of wound infection Br. J. Exp.
Pathol., 38, 1957, 573-586.
[8] Ratner, B.D., Biomaterials Science, 2nd ed., Elsevier Inc.,
San Diego, 2004.
[9] Chu, C.C., et al., Newly made antibacterial braided nylon
sutures. I. In vitro qualitative and in vivo preliminary
biocompatibility study, J. Biomed. Mater. Res., 21, No.11,
1987, 1281-1300.
[10] Blaker, J.J., et al., Development and characterisation of silver
doped bioactive glass coated sutures for tissue engineering
and wound healing applications, Biomaterials, 25, 2004,
1319-1329.
[11] Kovtun, E.A., et al., Study of the kinetics of release of
decametoxin from modified surgical sutures in aqueous
medium, Fibre Chemistry, 31, No.5, 1999, 374-379.
[12] Kovtun, E.A., et al., Creation of sutures with antimicrobial
properties, Fibre Chemistry, 32, No.4, 2000, 287-292.
[13] Kovtun, E.A., Plygan, E. P., and Sergeev, V.P., Study of the
stability of modified sutures in biological medium, Fibre
Chemistry, 32, No.6, 2000, 441-443.
[14] Sergeev, V.P., Development of method of manufacturing
modified sutures corresponding to current requirements in
medicine, Fibre Chemistry, 34, No.6, 2002, 433-439.
[15] Huang, Z.M., et al., A review on polymer nanofibers by
electrospinning and their applications in nanocomposites,
Compos. Sci. Technol., 63, No.15, 2003, 2223-2253.
[16] Sill, T.J., H.A. von Recum, Electro spinning: Applications in
drug delivery and tissue engineering, Biomaterials, 29,
No.13, 2008, 1989-2006.
[17] Jia, H., et al., Enzyme-carring polymeric nanofibers prepared
via electrospinning for use as unique biocatalysts, Biotechnol.
Prog., 18, 2002, 1027-1032.
[18] Katti, D.S., et al., Bioresorbable nanofiber-based systems for
wound healing and drug delivery: optimization of fabrication
parameters, J. Biomed. Mater. Res., 70B, 2004, 286-296.
[19] Kenawy, E.R., et al., Release of tetracycline hydrocloride
from electrospun poly(ethylene-co-vinylacetate), poly(lactic
acid), and a blend, J. Controlled Release, 81, 2002, 57-64.
[20] Kim, K., et al., Incorporation and controlled release of a
hydrophilic antibiotic using poly(lactide-co-glycolide)-based
electrospun nanofibrous scaffolds, J. Controlled Release, 98,
2004, 47-56.
[21] Li, M.Y., et al., Electrospinning polyaniline-contained gelatin
nanofibers for tissue engineering applications, Biomaterials,
27, No.13, 2006, 2705-2715.
[22] Luong-Van, E., et al., Controlled release of heparin from
poly(epsilon-caprolactone) electrospun fibers, Biomaterials,
27, No.9, 2006, 2042-2050.
[23] Luu, Y.K., et al., Development of a nanostructured DNA
delivery scaffold via electrospinning of PLGA and PLA-PEG
block copolymer, J. Controlled Release, 89, 2003, 341-353.
[24] Verreck, G., et al., Incorporation of drugs in an amorphous
state into electrospun nanofibers composed of a water-
insoluable, non-biodegradable polymer, J. Controlled
Release, 92, 2003, 349-360.
[25] Zeng, J., et al., Biodegradable electrospun fibers for drug
delivery, J. Controlled Release, 92, 2003, 227-231.
[26] Ding, B., et al., Preparation and characterization of a
nanoscale poly(vinyl alcohol) fiber aggregate produced by an
electrospinning method, J. Polymer. Sci. B Polymer. Phys.,
40, 2002, 1261-1268.
[27] Fong, H., et al., Beaded nanofibers formed during
electrospinning, Polymer, 40, 1999, 4548-4592.
[28] Huang, L., et al., Engineered collagen-PEO nanofibers and
fabrics, J. Biomater. Sci. Polym. Ed., 12, No.9, 2001, 979-
993.
[29] Ishii, Y., et al., A new electrospinning method to control the
number and a diameter of uniaxially aligned polymer fibers,
Materials Letters, 62, 2008, 3370-3372.
[30] Katta, P., et al., Continuous electrospinning of aligned
polymer nanofibers onto a wire drum collector, Nano Lett., 4,
No.11, 2004, 2215-2218.
[31] Lee, C.K., et al., The influence of added ionic salt on
nanofiber uniformity for electrospinning of electrolyte
polymer, Synthetic Metals, 154, 2005, 209-212.
[32] Lee, J.S., et al., Role of molecular weight of atactic
poly(vinyl alcohol) (PVA) in the structure and properties of
PVA nanofabric prepared by electrospinning, Journal of
Applied Polymer Science, 93, 2004, 1638-1646.
[33] Rutledge, G.C., S.V. Fridrikh, Formation of fibers by
electrospinning, Advance Drug Delivery Reviews, 59, 2007,
1384-1391.
[34] Thompson, C.J., et al., Effects of parameters on nanofiber
diameter determined from electrospinning model, Polymer,
48, 2007, 6913-6922.
[35] You, Y., et al., Effect of solution properties on nanofiber
structure of electrospun poly(lactic-co-glycolic acid), Journal
of Applied Polymer Science, 99, 2006, 1214-1221.
[36] Pan, H., et al., Continuous aligned polymer fibers produced
by a modified electrospinning method, Polymer, 47, 2006,
4901-4904.
19
Journal of Engineered Fibers and Fabrics http://www.jeffjournal.org
Volume 5, Issue 4 - 2010
[37] Crescenzi, V., et al., Thermodynamics of fusion of poly--
propiolactone and poly--caprolactone. Comparative analysis
of the melting of aliphatic polylactone and polyester chains,
Euro. Polym. J., 8, No.3, 1972, 449-463.
[38] Fu, Y.M., et al., Differential expression of bla(SHV) related
to susceptibility to ampicillin in Klebsiella pneumoniae, Int.
J. Antimicrob. Agents, 29, No.3, 2007, 344-347.
[39] Miyajima, M., et al., Factors influencing the diffusion-
controlled release of papaverine from poly(l-lactic acid)
matrix, J. Controlled Release, 56, 1998, 85-94.
[40] Huang, M.H., et al., Synthesis and degradation of PLA-PCL-
PLA triblock copolymer prepared by successive
polymerization of epsilon-caprolactone and (DL)-lactide,
Polymer, 45, No.26, 2004, 8675-8681.
[41] Little, U., et al., Accelerated degradation behaviour of
poly(epsilon-caprolactone) via melt blending with
poly(aspartic acid-co-lactide) (PAL), Polym. Degrad. Stab.,
94, No.2, 2009, 213-220.
[42] Buschle-Diller, G., et al., Release of antibiotics from
electrospun biocomponent fibers, Cellulose, 14, 2007, 553-
562.
[43] Zong, X., et al., Structure and process relationship of
electrospun bioabsorbable nanofiber membranes, Polymer,
43, 2002, 4403-4412.
[44] Ford, J.L., et al., Formulation of sustained release
promethazine hydrochloride tablets using
hydroxypropylmeethycellulose matrices, Int. J. Pharm., 24,
1985, 327-338.
[45] Hsu, Y.Y., et al., In vitro controlled release of isoniazid from
poly(lactide-co-glycolide) matrices, J. Controlled Release,
31, 1994, 223-228.
[46] Miyajima, M., et al., Mechanism of drug release from
poly(L-lactic acid) matrix containing acidic or neutral drugs,
J. Controlled Release, 60, 1999, 199-209.
[47] Ritger, P.L., N.A. Peppas, A simple equation for description
of solute release I. Fickian and non-Fickian release from non-
swellable devices in the form of slabs, spheres, cylinders or
discs, J. Controlled Release, 5, 1987, 23-36.
[48] Zurita, R., et al., Triclosan release from coated polyglycolide
threads, Macromol. Biosci., 6, No.1, 2006, 58-69.











AUTHORS ADDRESSES
Hang Liu
Karen Leonas
Washington State University
Department of Apparel, Merchandising, Design and
Textiles
51 Kruegel Hall
Pullman, WA 99164, USA

Yiping Zhao
University of Georgia
Department of Physics and Astronomy
Athens, GA 30602, USA