Preliminary Investigation and Characterization of
Electrospun Polycaprolactone and Manuka Honey
Scaffolds for Dermal Repair
Benjamin A Minden-Birkenmaier1, Rachel M Neuhalfen1, Blythe E Janowiak PhD2, Scott A. Sell, PhD1
1
Saint Louis University, Biomedical Engineering, St Louis, Missouri UNITED STATES
2
Saint Louis University, Department of Biology, St. Louis, Missouri UNITED STATES
Correspondence to:
Scott A. Sell email: [email protected]
ABSTRACT
This study focused on the characterization of Manuka
honey-containing
poly(ε-caprolactone)
(PCL)
nanofiber scaffolds with regards to wound healing.
Scaffolds were electrospun from 1, 5, 10, and 20%
v/v Manuka honey solutions. Scaffolds were
subjected to ethanol disinfection and soaked in
phosphate-buffered saline (PBS) for various
timepoints, and scaffold morphology and honey
release was quantified. Scaffolds showed increased
water vapor transmission rate (WVTR) with scaffold
soak time, indicating an increase in evaporation due
to enhanced osmotic potential of the scaffolds.
Mechanical testing indicated lower elasticity and
strength with honey incorporation, but showed no
significant change in material degradation rate with
the presence of honey over a 28 day PBS soak.
Fibroblast studies showed honey incorporation
increased cell infiltration into the scaffold, but
scaffold conditioned media did not induce significant
chemotaxis
towards
the
scaffold.
Honey
incorporation also demonstrated an increase in
fibroblast proliferation when in direct contact with
the scaffolds. Bacterial clearance from pure honey
was observed in both Gram positive Streptococcus
agalactiae (Group B Streptocococcus) and Gram
negative Escherichia coli (E. coli), but honey
scaffolds demonstrated significant clearance in only
the Gram negative E. coli. While further investigation
is needed, this preliminary study demonstrates the
wound-healing potential of Manuka honey-loaded
electrospun scaffolds.
bacterial inhibition [2], and cleansing the wound via
osmotic effects [3]. Such remedial effects are a result
of several intrinsic properties of honey which result
from the process of its creation [4].
Honey is created from nectar collected from flowers
by bees, processed in specialized pouches called
crops, and then regurgitated within the hive to be
concentrated by a drying process. During the internal
processing of the nectar, the bee’s hypopharyngeal
gland secretes certain enzymes into the crop to
facilitate its breakdown. One of these enzymes,
glucose oxidase, breaks down glucose into gluconic
acid and hydrogen peroxide, both of which are
important for honey’s wound healing properties. The
gluconic acid content of honey gives it a low pH,
ranging from 3.5 to 4.5, which in a wound reduces
protease activity, increases oxygen release from
hemoglobin, and stimulates macrophage and
fibroblast activity [1, 5, 6]. Meanwhile, the hydrogen
peroxide acts as an antiseptic, killing a wide variety
of bacteria, and also stimulates macrophage vascular
endothelial growth factor (VEGF) production, which
in turn increases angiogenesis in the area [7]. Many
of the floral sources for nectar contain flavenoids and
aromatic acids, which function as antioxidants in a
wound environment. These antioxidants neutralize
the free radicals generated by the hydrogen peroxide
activity. Left unchecked, the oxidative species would
activate the nuclear transcription factor NF-kB,
which would lead to the production of cytokines that
recruit and activate leukocytes, amplifying the
inflammation response and leading to a state of
chronic inflammation. This antioxidant content,
combined with honey’s low pH, helps mediate the
inflammatory response to avoid that chronic state [5].
INTRODUCTION
Many studies have demonstrated the beneficial
effects of honey on wound sites. These effects
include inducing wound closure [1], providing
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
126
http://www.jeffjournal.org
Another enzyme involved in the breakdown of
nectar, invertase, divides the nectar’s sucrose into
glucose and fructose. During the drying process, the
solution becomes supersaturated with these sugars,
giving the solution a highly negative osmotic
potential. This potential pulls lymph through the
wound area, flushing necrotic tissue, slough, and
other debris from the wound site [8]. The flow of
lymph also carries nutrients and oxygen from deeper
tissue to the wound area, supplying these active cells
with the resources to promote faster growth and
healing. It also removes excess fluid from deep in the
wound site, depriving bacteria of the water they need
to grow [9].
infiltration, and chemotaxis. The antimicrobial effects
of the scaffolds were quantified by measuring the
clearance potential with two types of bacteria:
Streptococcus agalactiae, commonly known as Group
B Streptococcus (GBS), and Escherichia coli.
MATERIALS AND METHODS
Unless otherwise indicated, all cell studies were done
using DMEM-F12 (Hyclone) supplemented with
10% fetal bovine serum (FBS) (Fisher Scientific
Company), and 1% penicillin/streptomycin (Hyclone)
(100 U/mL penicillin, 100 ug/mL streptomycin in the
final media mixture).
Electrospinning
PCL/Manuka honey solutions were prepared by
dissolving increasing volumes (1, 5, 10, and 20% v/v)
of Manuka honey (Medihoney, 100% active
Leptospermum honey, Derma Sciences) in HFP
(Oakwood Chemical), sonicating for 30 minutes in a
room temperature water bath (Branson 200
Ultrasonic Cleaner, Branson Ultrasonics) and then
dissolving 15 wt% of PCL (Polycaprolactone,
Mw=70,000-90,000, Sigma Aldrich). Scaffold
controls were created using an identical protocol with
DI water in place of Manuka honey at identical
volume ratios to HFP (0, 1, 5, 10, and 20% water
v/v), since a large percentage of Manuka honey is
water (13.4 - 22.9%) [19]. For each solution, PCL
was allowed to dissolve overnight, and then
electrospun.
Manuka honey is a specific honey variety made by
bees that collect nectar from Leptospermum
scopartum, a shrub native only to New Zealand. In
addition to the other honey components, Manuka
honey contains an inherent Unique Manuka Factor
(UMF). This UMF is believed to be based upon the
presence of methylglyoxal, and serves as an
additional antibacterial factor [10-12].
Extensive research has been performed on
electrospun poly(ε-caprolactone) (PCL) nanofibrous
sheets as cell culture scaffolds and wound dressings,
with favorable results [13, 14]. The electrospinning
process creates a nanofibrous structure of randomly
oriented fibers similar in morphology to the
biologically produced extracellular matrix (ECM) of
skin [14]. Electrospun scaffolds have been shown to
be effective at inducing cell infiltration and
proliferation [13, 14]. They have been shown to
promote fluid drainage and re-epithelialization in
animal wounds [14, 15]. PCL is a highly
biocompatible material that degrades due to a
hydrolysis reaction [16]. For solid PCL structures,
this degradation occurs over a period of years;
however, in porous scaffolds, breakdown happens
over the timespan of months. [17]. These properties,
coupled with its status as a low-cost, easily
electrospun polymer, make it an ideal test material.
Electrospun PCL has been tested as a wound
dressing/drug delivery device and shown to
effectively promote wound closure in animal models
[18].
Electrospinning was done using a syringe pump (7801001, Fisher Scientific) and a high-voltage DC
power supply (CZE1000PN30, Spellman High
Voltage Electronics Corp). 3.5 mL of each solution
was loaded in a 5 mL syringe tipped with a blunted
18 gauge needle (PrecisionGlide, Becton Dickinson).
Solution was extruded at 4 mL/hour at a voltage of
28 kV onto a spinning (400 rpm) rectangular (0.5cm
x 2.5cm x 9cm) stainless steel mandrel 12.5 cm away
from the needle tip. After the electrospinning
procedure, the scaffold was removed from the
mandrel using a razor blade and stored in a
desiccation chamber at -20ºC. Scaffolds varied in
thickness stored in a desiccation chamber from 0.1
mm to 0.8 mm. Samples for testing were taken from
the thicker parts of the scaffold, and in general varied
between 0.3 mm and 0.6 mm thick.
In the present study, PCL was electrospun from a
solution of mixed 1,1,1,3,3,3-hexafluoro-2-propanol
(HFP) combined with Manuka honey. These
electrospun scaffolds, along with corresponding
water controls, were characterized via SEM imaging
and mechanical testing. The response of human
dermal fibroblasts (HDFs) to these scaffolds were
ascertained with regards to proliferation, scaffold
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
Release and Morphology
10 mm diameter punches were taken of scaffolds
spun from pure HFP, 1%, 5%, 10%, and 20% honey,
and disinfected via ethanol (Fisher Chemical) soak
for 30 minutes followed by 3x 10 minute phosphate
buffered saline (PBS, Hyclone) rinses. They were
127
http://www.jeffjournal.org
Permeability = (Q*μ* δL) / (A * δP)
then incubated in 0.4 mL PBS for 1 hour, 24 hours, 4
days, 7 days, or 14 days at 37 ºC and 50% humidity.
Controls of each scaffold type which did not go
through the disinfection process were also incubated
in 0.4 mL PBS for 14 days. After incubation,
scaffolds were removed and imaged using a scanning
electron microscope (SEM) (Zeiss, Evo LS15).
Samples were sputter coated with gold prior to
imaging. Accelerating voltage of 20kV was applied
and scaffolds were imaged at 1500x magnification,
with Image J software used to measure fiber diameter
and pore area (each from a minimum of 240
individual pore measurements from 4 images taken of
separate areas of the scaffold). Images of scaffolds
which had not been disinfected or soaked, or which
had been disinfected but not soaked, were also taken
and analyzed in order to ascertain the effect of the
disinfection process on the fiber and pore size.
Releasate of soaked scaffolds was retained and
assayed for glucose (Glucose Assay Kit, Sigma
Aldrich) as an indicator of honey release.
where:
Q = flow rate
μ = viscosity of liquid
δL = thickness of scaffold
A = cross sectional area
δP = vacuum pressure
Mechanical Properties
6.2 mm wide x 18.6 mm long dogbone-shaped
punches with a width of 2.7 mm in the failure region
were taken of each scaffold and incubated in PBS for
1, 4, 7, 14, 21, and 28 days at 37 ºC and 50%
humidity. At each timepoint dogbones were subjected
to uniaxial tensile testing using a mechanical testing
system (MTS Criterion Model 42, MTS Systems
Corporation) fitted with a 100 N load cell. A gauge
length of 10 mm and a constant strain rate of 10
mm/min were used to test each sample to failure.
Data integration was done using MTS TW Elite
software.
Water Vapor Transmission Rate (WVTR)
Cuvettes (1.1 x 1.1 cm cross sectional area, 4.5cm
tall) were filled with 3 mL distilled water. A 1.5 x 1.5
cm piece of each scaffold (enough to cover the
cuvette top securely) was soaked in PBS (HyClone)
for either ten minutes or one hour, and then sealed
over the mouth of each filled cuvette with an
industrial sealant (CraftBond Acid-Free MultiPurpose Spray Adhesive, Elmers). Sealant was
sprayed on a Kimwipe, then rubbed across the mouth
of the cuvette three times before attaching the
scaffold square. The mass of each cuvette was then
measured and recorded, and the cuvettes were placed
in an incubator at 37ºC temperature and 50%
humidity for 4 days. Weight loss was plotted against
the elapsed time to get a slope (reduction in g/day).
WVTR was calculated by dividing slope by the
mouth area of dish [20].
WVTR = slope (g/day) / test area (m2).
Fibroblast Chemotaxis
10 mm diameter circular punches of each scaffold
type were disinfected by soaking in ethanol (Fisher
Chemical) for 30 minutes followed by 3x 10 minute
PBS rinses. Scaffolds were then placed in sterile 24
well plates. 500 µL of media was added, and then
transwell inserts (Corning Inc) with 8µm pores were
placed in each well. 50,000 fibroblasts (ATCC CRL2522,) were seeded in 100 µL media in each
transwell insert and then the plates were incubated at
37ºC, changing the media in the top well every 3
days. After 7 days, an MTS assay (CellTiter 96
Aqueous Non-Radioactive Cell Proliferation Assay,
Promega) was performed on both the bottom well
and the insert in order to distinguish chemotaxis from
proliferation. Controls of standard culture media, and
13% v/v honey in culture media (roughly the same
amount of honey estimated to be in each 10 mm
punch of 20% honey scaffold, delivered as a bolus
rather than as a controlled release from the scaffolds)
were run. In addition, a positive control of 10 mg/mL
preparation rich in growth factors (PRGF) in culture
media was used. The PRGF was prepared via a
previously described method [22, 23]. Briefly, human
blood from 3 donors was purchased (Biological
Specialty Corp.), pooled, and centrifuged to create
platelet rich plasma (PRP). PRP was then frozen in a
-80 ºC freezer, thawed, frozen again, and lyophilized
for 24 hours to create powdered PRGF. This PRGF
was then added to culture media at 10 mg/mL and
run alongside the other samples to establish a positive
control for fibroblast chemotaxis.
(1)
Permeability Testing
10 mm diameter circular punches of each scaffold
type were tested. Punches (n=3) were tested dry, after
a 10 minute PBS soak, and after a 1 hour PBS soak.
Scaffold thickness was measured using a digital
caliper (Absolute Digimatic Indicator, 543-Standard
Type, Mitutoyo). Each punch was loaded into a filter
holder, and 50 mL of water was passed through the
scaffold using a vacuum pump (UN810FTP, KNF
Laboport), recording the vacuum pressure and the
time elapsed. Scaffold permeability was calculated
using the following equation [21]:
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
(2)
128
http://www.jeffjournal.org
placed 5 cm apart on the bacteria plates. A sterile test
disc was spread with a thin layer of honey and placed
on the plates to serve as an indicator of the
antimicrobial effects of the honey when delivered as
a burst release. A 10 IU penicillin disc was also
placed on the plates to serve as a positive control.
Plates were then incubated at 37 ºC on brain heart
infusion agar for 24 hours. Images were taken using a
digital camera and bacterial clearance around each
scaffold punch was measured using Image J.
Fibroblast Proliferation
10 mm diameter circular punches of each scaffold
were disinfected via 30 minute ethanol soak followed
by 3x 10 minute PBS rinses. 30,000 HDFs were
seeded in 600 µL of DMEM-F12 media on top of the
scaffold punches and incubated at 37 ºC. At days 1
and 4 of the study, an MTS assay was performed to
measure cell proliferation.
Fibroblast Infiltration
10 mm diameter circular punches were taken of each
scaffold type and disinfected via 30 minute ethanol
rinse followed by 3x 10 minute PBS rinses. 100,000
fibroblasts were seeded on top of the scaffold
punches in 600 µL media and incubated at 37 ºC for
28 days, changing media every 3 days. After the
incubation period, scaffolds were removed and fixed
in formalin (Fisher Scientific Company). Samples for
histologic evaluation were cross-sectioned for slide
mounting and hematoxylin and eosin (H&E) staining
(Saint Louis University Research Microscopy Core).
Images of the furthest-infiltrated regions of H&E
slides were taken by optical light microscopy (Zeiss
Axiovert 200, 40X). The cellular infiltration depth
was determined via Image J by measuring the
distance from the scaffold edge to the 60 deepest
cells on each image.
Statistics
All statistical analyses were performed utilizing
Sigma Stat (Version 2.03; SPSS, Inc). For studies
with sixty or more replicates, data was evaluated
using a one-way analysis of variance (ANOVA) and
then subjected to a pair-wise multiple comparison
procedure (Tukey Test) with an a priori alpha value
of 0.05. Studies with fewer than 60 replicates were
subject to a nonparametric Kruskal-Wallis pair-wise
test with an a priori alpha value of 0.05
RESULTS
Morphology
The results of the fiber morphology assessments over
their release time are shown in Figure 1. Fiber
diameter measurements taken from the SEM images
of the scaffolds are shown in Figure 1B. These
results showed no meaningful significance between
scaffold types or between non-disinfected scaffolds,
disinfected scaffolds, or scaffolds soaked for various
timepoints. The pore size measurements, shown in
Figure 1C, show no meaningful significances
between scaffold types or soaking times. Some
scaffold types appear to show a non-significant trend
of decreasing pore size with soaking time, which
could be due to shifting of the fibers during the
soaking. Since these trends were not statistically
significant, however, they were not explored further.
Bacterial Clearance
6 mm diameter circular punches were taken of each
scaffold type (n=6) and disinfected via 30 minute
ethanol soak followed by 3x 10 minute PBS rinses.
Solutions of each test bacteria (GBS and Escherichia
coli) were spread onto agar plates using a sterile
swab. GBS was obtained from American Type Tissue
Culture (S. agalactiae 2603 V/R, was obtained from
ATCC: BAA-611). The E. coli strain used was K99
(ATCC: PTA-5951). These strains were chosen
because they are commonly used examples of both a
Gram positive and Gram negative bacterial strain,
respectively. Additionally, these bacteria are both
commonly found in human dermal wounds [9, 24,
25]. Using aseptic technique, the scaffold discs were
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
129
http://www.jeffjournal.org
FIGURE 1. (A) SEM images of electrospun scaffolds prior to disinfection. All images taken at 1500x. Mean fiber diameter (B) and pore area (C)
of non-disinfected (ND), just disinfected (JD), and disinfected electrospun scaffolds soaked for the indicated time. The ND D14 samples were
soaked for 14 days without being disinfected first. Measurements were taken from SEM images (n=240).
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
130
http://www.jeffjournal.org
percentage. This data showed no significant
differences in honey lost during disinfection across
scaffold type.
WVTR
The results of the WVTR test are shown in Figure 3.
WVTR of the scaffolds after either a 10 minute PBS
soak or a 1 hour PBS soak was observed over an
incubation period of 4 days, and WVTR values
ranged from 27 to 335 g/(m2*day). Statistical analysis
showed the WVTR of scaffolds spun from 5%, 10%,
and 20% honey to be significantly increased by the 1
hour PBS soak over the ten minute PBS soak. In
addition, the WVTR of these scaffolds was
significantly greater than the scaffold spun from pure
HFP for the one hour soak. Despite an increasing
trend, an alpha of 0.05 showed no significant
differences between these scaffolds and their
corresponding water controls; however, a more lax
alpha of 0.15 showed statistical significance from the
water controls and the 10% honey and 20% honey
scaffolds.
FIGURE 3. Results of the WVTR test of scaffolds soaked for 10
minutes and 1 hour.
FIGURE 2. (A) Glucose release of disinfected scaffolds at various
timepoints, (B) glucose release of non-disinfected scaffolds after
14 days, and (C) percent glucose loss in the disinfection.
Permeability Testing
As shown in Figure 4, the soaking period did not
significantly affect the permeability of the scaffold
for any of the sample types. Of the non-soaked
scaffolds, the 1% honey and 20% honey were
significantly more permeable than the pure HFP
control and their respective water controls. For both
the 10 minute PBS soaked and 1 hour PBS soaked
scaffolds, only 20% honey was significantly different
from the pure HFP control, but both the 1% honey
and 20% honey scaffolds were significantly more
permeable than their respective water controls.
Release
As shown in Figure 2, all honey-containing scaffolds
showed significant release of glucose over the 14 day
incubation time, and the amount of glucose released
by each timepoint significantly increased with the
amount of honey incorporated into the scaffold up
through the 10% honey samples. The amount of
glucose lost due to disinfection was calculated by
subtracting the 14 day disinfected scaffold releasate
from the non-disinfected scaffold releasate, and
dividing by the non-disinfected releasate to obtain a
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
131
http://www.jeffjournal.org
pure HFP samples, but the water control samples as
well. However, even in the 20% honey sample, there
was no significant decline in modulus over time,
suggesting that while the presence of the Manuka
honey decreased the tensile strength of the scaffold,
honey incorporation did not increase scaffold
degradation rate.
The peak stresses of each scaffold over their
incubation time in PBS are shown in Figure 6. While
the 1% and 5% honey scaffolds tended to be lower
than the pure HFP and water controls, statistical
analysis revealed only days 14 and 28 in the 1%
honey and only days 1 and 28 in the 5% honey to be
significantly different. However, the 10% and 20%
honey show a distinct, statistically significant drop
from both the pure HFP control and the water control
at all timepoints, indicating that the incorporation of
large amounts of honey into the scaffold does
significantly weaken that scaffold mechanically.
FIGURE 4. Results of permeability testing of scaffolds tested dry,
after a 10 minute PBS soak, and after a 1 hour PBS soak (n=3).
Mechanical Properties
As shown in Figure 5, the honey scaffolds had lower
elastic moduli than the pure HFP control at all
timepoints, and most of these timepoints showed a
statistically significant difference from not only the
1% Honey
5% Honey
10% Honey
20% Honey
FIGURE 5. Elastic moduli of each scaffold type during incubation in PBS measured over time (n=7).
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
132
http://www.jeffjournal.org
1% Honey
5% Honey
10% Honey
20% Honey
FIGURE 6. Peak stress in each scaffold type during incubation in PBS measured over time (n=7).
Fibroblast Chemotaxis
As shown in Figure 7, none of the honey scaffolds or
water/PCL controls showed any significant fibroblast
chemotaxis. The only sample to demonstrate
statistically significant chemotaxis was the positive
chemotaxis control of 10 mg/mL PRGF, which
showed that our fibroblasts had the potential to
migrate if stimulated with the correct chemokines.
All wells except the 13% honey/media showed
significant fibroblast proliferation, and the 20%
honey scaffold had the highest proliferation of any
sample, although not statistically significant. The
13% honey/media control showed a significant
decrease in cell number relative to the amount
seeded, indicating that when delivered as a bolus, the
Manuka honey induced cell death in vitro. However,
when delivered slowly via controlled release from an
electrospun scaffold, that same amount of honey
induced fibroblast proliferation, as shown by the high
total cell number in the 20% honey scaffold column.
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
FIGURE 7. Fibroblast chemotaxis test results (n=3).
Fibroblast Proliferation
As shown in Figure 8, all samples showed significant
fibroblast proliferation between days 1 and 4. At both
days one and four, the 20% honey scaffold had
significantly more fibroblast proliferation than any of
the other samples, although this difference was more
133
http://www.jeffjournal.org
pronounced on day 1 than on day 4, potentially due to
the cells reaching confluence and becoming contact
inhibited by this timepoint. None of the other honey
samples showed a significant rise in proliferation
over either the pure HFP control or the water controls
at either timepoint.
Bacterial Clearance
As shown in Figure 10, the only E. coli inhibition
was seen in the scaffolds containing the highest
content of honey, 10% and 20% honey, respectively.
While this clearance was statistically significant, it
was much lower than that created by the control
sterile disk coated with honey, which created the
highest clearance of any of the samples including the
penicillin disk. In contrast, all samples exhibited
some GBS clearance, although no statistical
difference was seen between any of the honey
samples and the pure HFP or water controls.
However, the sterile disk with honey exhibited much
higher GBS clearance than any of the other samples,
and was statistically significant from all of them,
including the penicillin control. This sterile disk with
honey control showed a statistically significant
increase in E. coli clearance over the GBS clearance,
although the honey scaffold results showed no such
trend between bacteria types. Overall, the 10% and
20% Manuka honey scaffolds proved to be as
effective in bacterial clearance as the penicillin
control.
FIGURE 8. Fibroblast proliferation when in direct contact with
scaffolds (n=3).
Fibroblast Infiltration
As shown in Figure 9, fibroblasts were able to
successfully infiltrate all scaffold samples. The 5%,
10%, and 20% honey samples were statistically
significantly higher than the pure HFP or water
controls, and there was a statistically significant trend
of increasing infiltration distance with increasing
honey content (Figure 9).
FIGURE 10. (A) Representative image from bacterial clearance
test. Counterclockwise, from top: 10% honey, 20% honey, 1%
honey not disinfected (ND), 5% honey ND, 10% honey ND, 20%
honey ND. Center: Sterile disk spread with honey. (B) Bacterial
clearance test results (n=6) including sterile disk coated with
Manuka honey
DISCUSSION
The effectiveness of honey and honey-based products
in treating a variety of dermal wounds has been
demonstrated in clinical settings [26-28]. However,
FIGURE 9. (A) Representative H&E stained images for fibroblast
infiltration into electrospun structures. All images taken at 40x. (B)
Fibroblast infiltration quantification (n=60).
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
134
http://www.jeffjournal.org
there remain issues with optimization of Manuka
honey delivery in a wound site. This preliminary in
vitro study focused on characterizing not only
scaffold morphology and material properties, but also
properties more directly applicable to wound care,
including WVTR, bacterial clearance, and fibroblast
chemotaxis, proliferation, and infiltration.
environment, it will form renewed skin, accelerating
the healing process [30, 31]. The WVTR of a wound
dressing should thus be tailored to maintain this wet
environment without causing an excessive buildup of
wound exudate, which could lead to infection. Lamke
et al. reported the WVTR for granulating wounds,
first degree burns, and normal skin to be 5,138 + 202
g/m2/day, 279 + 26 g/m2/day, and 204 + 12 g/m2/day
respectively [20, 32]. As shown in Figure 3, the
WVTR of our scaffolds ranged between 50 and 300
g/m2/day; on the order of natural skin but well below
the target of 2000-2500 g/m2/day for optimal WVTR.
However, the WVTR of the honey scaffolds did
significantly increase over the one hour soak time in
proportion to the honey content of the scaffolds. This
effect suggests that over the soaking time, the
inherent negative osmotic potential of the Manuka
honey becomes activated and contributes to vapor
transport. It may also be possible that as honey elutes
from the scaffold the WVTR will increase. Future
work will involve longer soak periods to ascertain the
upper limit of this trend.
Scaffold pore size, which is a function of fiber
diameter, is an important parameter in the promotion
of cell infiltration into the scaffold. As a general rule,
pore diameters on the scale of 10 to 100 µm
(corresponding to pore areas of 78 µm2 - 7,853 µm2)
allow for good cell infiltration and movement within
the scaffold [29]. As shown in Figure 1C, the honey
scaffolds and control scaffolds had pore sizes ranging
from 800 µm2 to 7000 µm2, with no significant
changes due to scaffold composition or soaking time.
It should be noted that our pore size range was within
the adequate range for cellular infiltration. The
glucose assay performed on the releasate showed that
the scaffolds exhibited a small burst release after 1
hour of soaking, and then continually released
glucose throughout the 14 day period that the test was
performed. As glucose is a large constituent of
honey, and glucose release strongly correlated with
honey concentration of the scaffold, it can be
assumed that glucose release is a strong indicator of
total honey release from the scaffold. The 10% honey
sample exhibited a higher honey release at every
timepoint, indicating that there is perhaps a
difference in the honey incorporation between it and
the 20% honey sample that enables the 10% honey
scaffold to better retain and release its honey, even
after ethanol disinfection. The 14 day soak releasate
of the non-disinfected samples showed much higher
honey release than the 14 day soak releasate from the
disinfected samples, indicating that the ethanol
disinfection process utilized throughout this study
does in fact leach much of the honey from the
scaffold. Nevertheless, these disinfected scaffolds
maintain a level of honey that allows for continued
release for at least 14 days post-disinfection. These
findings seem to indicate that there may be a coating
of honey on the surface of the fibers which is
released as a burst during the disinfection process and
the first hour of soak, but that there is also honey
incorporated into the fibers themselves which is
released more slowly over a period of weeks, and has
the ability to positively impact scaffold bioactivity.
The permeability results, as shown in Figure 4,
indicate little-to-no change in permeability with
regards to soaking period in any of the scaffolds,
meaning that honey elution does not significantly
affect the permeability of the scaffold. However, the
1% honey and 20% honey scaffolds were
significantly more permeable than the controls for
both the dry and the soaked scaffolds. This suggests
that honey incorporation has the potential to
significantly increase scaffold permeability, even
though it does not increase pore size. This could be
due to several effects, such as a difference in the
distribution of the fibers within the scaffold, or the
osmotic potential of the honey present in the
scaffolds. A larger sample size will be necessary to
determine if this permeability increase is present for
scaffolds of all honey concentrations.
As shown in Figures 5 and 6, honey incorporation
made the scaffolds weaker and more brittle.
However, the lack of a significant downward trend in
either elastic modulus or peak stress over incubation
time indicates that honey incorporation does not
increase scaffold degradation rate. This finding, in
conjunction with the SEM and glucose release data,
indicates that it is likely that the honey does not
replace or interrupt the polymeric backbone of the
fibers, but rather that it is integrated within the
polymer network in a manner that does not disrupt
chain and fiber formation.
The WVTR of a wound dressing is of utmost
importance. If the wound is directly exposed to air,
then water loss will be too high for proper healing,
and it will scab over. However, if the wound is
protected by a dressing that maintains a wet
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
As honey is eluted from the scaffold, it promotes cell
infiltration. This effect was quantified in the
135
http://www.jeffjournal.org
fibroblast infiltration test shown in Figure 9. These
data showed a trend of increasing cell infiltration
with increased honey incorporation, which can be
observed visually in the H&E images. These images
show large void spaces in the 5%, 10%, and 20%
honey scaffolds, with cells clustered around the edges
of these void spaces. Having demonstrated that
scaffold releasate had no significant chemotactic
effects on fibroblasts, as shown in Figure 7, it may be
that once within the scaffold, honey concentrations
are high enough to promote increases in cellular
proliferation within the structures in areas of close
proximity to regions of large honey deposition. In
conjunction, it may also be possible that the
incorporation of honey, which was shown to decrease
scaffold mechanical properties (Figures 5 and 6),
decreased the moduli of individual fibers to a point
that allowed for significantly increased cellular
infiltration. This premise will require further
investigation into the mechanical properties of
individual fibers within the honey-containing
scaffolds.
well as more comprehensive studies with expanded
Gram negative bacterial species such as Salmonella.
CONCLUSION
Electrospun Manuka honey scaffolds have enormous
potential as wound dressings and precursors to tissue
engineered skin. This preliminary research has
outlined the relevant characteristics of such scaffolds
and established their usefulness in promoting healing
and clearing bacteria from the wound environment.
Further research will be necessary to better
understand the limits of these characteristics and
explore other aspects of these scaffolds, such as their
ability to mediate the inflammatory response, and
their behavior in vivo.
ACKNOWLEDGMENT
The authors would like to acknowledge funding for
this research contributed by the Saint Louis
University Summer Undergraduate Research
Experience (SURE) Program, and John LaMaster,
Kayla Scott, Zachary Toth, and Emily GrowneyKalaf for their assistance with this research.
As with any porous material, the scaffold
proliferation assay might have been hampered by the
diffusion of the MTS solution into the scaffold. With
fibroblasts infiltrating into the scaffolds at different
rates, as shown in the cell infiltration graph in Figure
8, we postulate that scaffolds with more cell
infiltration will be disproportionately affected by the
resistance of the scaffold environment to MTS
diffusion. Even given this effect, the significant spike
in cell proliferation of the 20% honey scaffold
relative to the controls indicates that honey
incorporation in the scaffold does promote fibroblast
proliferation.
REFERENCES
[1]
Gethin, G.T., S. Cowman, and R.M. Conroy,
The impact of Manuka honey dressings on
the surface pH of chronic wounds.
International Wound Journal, 2008. 5(2): p.
185-194.
[2]
Cooper, R., P. Molan, and K. Harding, The
sensitivity to honey of Gram‐positive cocci
of clinical significance isolated from
wounds. Journal of Applied Microbiology,
2002. 93(5): p. 857-863.
[3]
Osato, M.S., S.G. Reddy, and D.Y. Graham,
Osmotic Effect of Honey on Growth and
Viability of Helicobacterpylori. Digestive
diseases and sciences, 1999. 44(3): p. 462464.
[4]
Abebayehu, D., A.J. Spence, and S.A. Sell,
The Present and the Potential of Manuka
Honey in Dermal Regenerative Medicine, in
Cellular and Molecular Mechanisms of
Honey Wound Healing, E. Ranzato and S.
Martinotti, Editors. 2014, Nova Science
Publishers, Inc.: New York. p. 79 - 106.
[5]
Molan, P., et al., Why honey works. Honey in
modern wound management. Aberdeen:
Wounds UK Ltd, 2009. 9: p. 36-37.
[6]
Al-Waili, N., K. Salom, and A.A. Al-Ghamdi,
Honey for wound healing, ulcers, and burns;
data supporting its use in clinical practice.
The scientific world journal, 2011. 11: p.
766-787.
As demonstrated by the high clearance of both
bacteria types by the sterile disk spread with honey in
Figure 10, Manuka honey is effective against both
Gram positive and Gram negative bacteria. However,
the results of the honey scaffold test indicate that the
controlled release of smaller amounts of honey by the
scaffolds is more effective against the Gram negative
E. coli, as the clearance of the Gram positive GBS
was not any greater around the honey scaffolds than
around the water and pure HFP controls. As
penicillin is more effective against Gram positive
GBS, these results suggest a unique role for Manuka
honey in the treatment of Gram negative bacteriainfected wounds. However, it should be noted that
this test represents a small snapshot of the bacterial
clearance potential of the honey. Further testing will
be necessary to quantify the hydrogen peroxide
production and sustained elution from the scaffold, as
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
136
http://www.jeffjournal.org
[7]
[8]
[9]
[10]
[11]
[12]
[14]
[15]
[16]
[17]
[18]
Cho, M., T.K. Hunt, and M.Z. Hussain,
Hydrogen peroxide stimulates macrophage
vascular endothelial growth factor release.
American Journal of physiology-heart and
circulatory physiology, 2001. 280(5): p.
H2357-H2363.
Molan, P.C., Re-introducing honey in the
management of wounds and ulcers-theory
and practice. 2002.
Chirife, J., G. Scarmato, and L. Herszage,
Scientific basis for use of granulated sugar
in treatment of infected wounds. The Lancet,
1982. 319(8271): p. 560-561.
Jenkins, R., N. Burton, and R. Cooper, Effect
of manuka honey on the expression of
universal stress protein A in meticillinresistant< i> Staphylococcus aureus</i>.
International journal of antimicrobial agents,
2011. 37(4): p. 373-376.
Leong, A.G., P.M. Herst, and J.L. Harper,
Indigenous New Zealand honeys exhibit
multiple anti-inflammatory activities. Innate
immunity, 2012. 18(3): p. 459-466.
Sell, S.A., et al., A preliminary study on the
potential of manuka honey and platelet-rich
plasma in wound healing. Int J Biomater,
2012. 2012: p. 313781.
Yoshimoto, H., et al., A biodegradable
biodegradable nanofiber scaffold by
electrospinning and its potential for bone
tissue engineering. Biomaterials, 2003.
24(12): p. 2077-2082.
Chong, E., et al., Evaluation of electrospun
PCL/gelatin nanofibrous scaffold for wound
healing and layered dermal reconstitution.
Acta biomaterialia, 2007. 3(3): p. 321-330.
Chen, J.-P., G.-Y. Chang, and J.-K. Chen,
Electrospun collagen/chitosan nanofibrous
membrane as wound dressing. Colloids and
Surfaces
A:
Physicochemical
and
Engineering Aspects, 2008. 313: p. 183-188.
Woodruff, M.A. and D.W. Hutmacher, The
return
of
a
forgotten
polymer—
polycaprolactone in the 21st century.
Progress in Polymer Science, 2010. 35(10):
p. 1217-1256.
Sung, H.-J., et al., The effect of scaffold
degradation rate on three-dimensional cell
growth and angiogenesis. Biomaterials,
2004. 25(26): p. 5735-5742.
Choi, J.S., K.W. Leong, and H.S. Yoo, In vivo
wound healing of diabetic ulcers using
electrospun nanofibers immobilized with
human epidermal growth factor (EGF).
Biomaterials, 2008. 29(5): p. 587-596.
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
[19]
[20]
[21]
[22]
[23]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
137
Pohl, P., Determination of metal content in
honey by atomic absorption and emission
spectrometries. TrAC Trends in Analytical
Chemistry, 2009. 28(1): p. 117-128.
Mi, F.L., et al., Fabrication and
characterization
of
a
sponge-like
asymmetric chitosan membrane as a wound
dressing. Biomaterials, 2001. 22(2): p. 16573.
O'Brien, F.J., et al., The effect of pore size on
permeability and cell attachment in collagen
scaffolds for tissue engineering. Technol
Health Care, 2007. 15(1): p. 3-17.
Sell, S.A., et al., Incorporating platelet-rich
plasma into electrospun scaffolds for tissue
engineering applications. Tissue Eng Part
A, 2011. 17(21-22): p. 2723-37.
Wolfe, P.S., et al., The Creation of
Electrospun Nanofibers from Platelet-Rich
Plasma. J Tissue Sci Eng, 2011. 2(2).
Citron, D.M., et al., Bacteriology of
moderate-to-severe diabetic foot infections
and in vitro activity of antimicrobial agents.
Journal of clinical microbiology, 2007.
45(9): p. 2819-2828.
Mhalu, F., Infection with Streptococcus
agalactiae in a London hospital. Journal of
clinical pathology, 1976. 29(4): p. 309-312.
Natarajan, S., et al., Healing of an MRSAcolonized, hydroxyurea-induced leg ulcer
with honey. Journal of Dermatological
Treatment, 2001. 12(1): p. 33-36.
Mohd Zohdi, R., et al., Gelam (Melaleuca
spp.) honey-based hydrogel as burn wound
dressing. Evidence-Based Complementary
and Alternative Medicine, 2011. 2012.
Subrahmanyam, M., Honey dressing versus
boiled potato peel in the treatment of burns:
a prospective randomized study. Burns,
1996. 22(6): p. 491-493.
Yannas, I.V. and J.F. Burke, Design of an
artificial skin .1. Basic design principles.
Journal Of Biomedical Materials Research,
1980. 14(1): p. 65-81.
Yoshii, F., et al., Electron beam crosslinked
PEO and PEO/PVA hydrogels for wound
dressing. Radiation Physics and Chemistry,
1999. 55(2): p. 133-138.
Winter, G.D. and J.T. Scales, Effect of air
drying and dressings on the surface of a
wound. 1963.
Lamke, L.-O., G. Nilsson, and H. Reithner,
The evaporative water loss from burns and
the water-vapour permeability of grafts and
artificial membranes used in the treatment
of burns. Burns, 1977. 3(3): p. 159-165.
http://www.jeffjournal.org
AUTHORS’ ADDRESSES
Benjamin A Minden-Birkenmaier
Rachel M Neuhalfen
Blythe E Janowiak PhD
Scott A. Sell, PhD
Saint Louis University
Biomedical Engineering
3507 Lindell Blvd
St Louis, MO 63103
UNITED STATES
Journal of Engineered Fibers and Fabrics
Volume 10, Issue 4 – 2015
138
http://www.jeffjournal.org