Biodegradable soy wound dressings withcontrolled release of antibiotics: Results froma guinea pig burn model

Dana Egozi a,b, Maya Baranes-Zeevi c, Yehuda Ullmann b,d, Amos Gilhar b,Aviad Keren b,d, Elias Matanes d, Israela Berdicevsky b, Norberto Krivoy b,Meital Zilberman c,*

aDepartment of Plastic Surgery, Kaplan Medical Center, Rehovot, Israelb Faculty of Medicine, Technion – Israel Institute of Technology, Haifa 32000, IsraelcDepartment of Biomedical Engineering, Tel-Aviv University, Tel-Aviv 69978, IsraeldDepartment of Plastic Surgery and the Burn Unit, Rambam Health Care Campus, Haifa, Israel

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 7

a r t i c l e i n f o

Article history:

Accepted 27 March 2015

Keywords:

Gentamicin

Infection

Histology

Soy protein

Wound healing

a b s t r a c t

There is growing interest in the development of biodegradable materials from renewable

biopolymers, such as soy protein, for biomedical applications. Soy protein is a major fraction

of natural soybean and has the advantages of being economically competitive, biodegrad-

able and biocompatible. It presents good water resistance as well as storage stability. In the

current study, homogenous antibiotic-loaded soy protein films were cast from aqueous

solutions. The antibiotic drug gentamicin was incorporated into the films in order to inhibit

bacterial growth, and thus prevent or combat infection, upon its controlled release to the

surrounding tissue. The current in vivo study of the dressing material in contaminated deep

second-degree burn wounds in guinea pigs (n = 20) demonstrated its ability to accelerate

epithelialization with 71% epithelial coverage compared to an unloaded format of the soy

material (62%) and a significant improved epithelial coverage as compared to the conven-

tional dressing material (55%). Our new platform of antibiotic-eluting wound dressings is

advantageous over currently used popular dressing materials that provide controlled

release of silver ions, due to its gentamicin release profile, which is safer. Another advantage

of our novel concept is that it is based on a biodegradable natural polymer and therefore

does not require bandage changes and offers a potentially valuable and economic approach

for treating burn-related infections.

# 2015 Elsevier Ltd and ISBI. All rights reserved.

* Corresponding author at: Department of Biomedical Engineering, Faculty of Engineering, Tel.:-Aviv University, Tel-Aviv 69978, Israel.Tel.: +972 3 6405842; fax: +972 3 6407939.

E-mail address: meitalz@eng.tau.ac.il (M. Zilberman).

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/burns

http://dx.doi.org/10.1016/j.burns.2015.03.0130305-4179/# 2015 Elsevier Ltd and ISBI. All rights reserved.

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 71460

1. Introduction

1.1. Soy protein

There is growing interest in the development of biodegradable

materials from renewable biopolymers, such as soy protein,

for biomedical applications [1–7]. Soy protein is a major

fraction of natural soybean and can be found in several

soybean derivatives in different quantities, depending on the

extraction method: defatted soy flour contains 50–59% protein,

soy protein concentrate contains 65–72% protein and soy

isolate contains more than 90% protein [8]. Soy protein has

been explored mainly in the polymer, food and agriculture

fields. Use of soy protein as a food source is still increasing, due

to its functional and nutritional value, availability and low

price [9]. In the materials industry, soy protein was studied as

an adhesive and as a ‘‘green’’ plastic [1]. Soy protein is an

abundant plant protein and has the advantages of being

economically competitive, biodegradable and biocompatible.

It presents good water resistance as well as storage stability

[4]. Soy protein is also very versatile and its properties can be

tailored by physical, chemical, or enzymatic treatments [10],

such that it can provide diverse requirements for different

biomedical applications. It is known that local or systemic use

of soybean or its components has an Ig E induced immuno-

genic effect more in children and less in adults, but in spite of it

the soybean is used for the preparation of an intravenous

medication widely used in surgery. As always, the public

should be informed in the package insert. The statistics says

that soybean affects approximately 0.4% of the children’s

population. Also, approximately 50% of children with soy

allergy outgrew their allergy by age of 7 years. The soy protein

extract used in the current study was provided by a company

which sells it for the food industry and therefore it is

considered as safe.

The suitability of soy protein for biomedical applications

has been investigated, due to its low cost and bioactive

properties [11]. Soy thermoplastics were found to be non-

cytotoxic, and even encouraged cell proliferation during

in vitro tests [12]. Proposed applications range from bone

cement to hydrogel and membranes for wound-dressing

applications. The most promising recent applications include

drug delivery carrier films, temporary replacement implants

and scaffolds for tissue engineering [13].

Snyders et al. [14] and Shingel et al. [15] investigated hybrid

hydrogels made of poly(ethylene glycol) and soy protein for

moist wound-dressing applications and concluded that these

hydrogels can be considered as a safe, biocompatible and

inflammatory inert wound-dressing material. Santos et al. [16]

developed chitosan/soy (cht/soy)-based membranes as

wound-dressing materials and showed that these new

membranes possess the desired features for healing/repair

stimulation, ease of handling, and final esthetic appearance,

which are valuable properties for wound dressings.

1.2. Wound dressings

Various types of wounds result in tissue damage. These

include burns, pressure ulcers, diabetic ulcers and trauma.

The main goal in wound management is to achieve rapid

healing with good functional and esthetic results. An ideal

wound dressing can restore the milieu required for the healing

process, while simultaneously protecting the wound bed

against bacteria and environmental threats. The dressing

should also be easy to apply and remove. Most modern

dressings are designed to maintain a moist healing environ-

ment and to accelerate healing by preventing cellular

dehydration and promoting collagen synthesis and angiogen-

esis [17]. However, over-restriction of water evaporation from

the wound should be avoided, since accumulation of fluid

under the dressing may cause maceration and facilitate

infection. The physical and chemical properties of the

dressing should therefore be adapted to the type of wound

as well as to the degree of wound exudation.

Bacterial contamination of a wound seriously threatens its

healing. In burns, infection is the major complication after

the initial period of shock, and it is estimated that about 75%

of the mortalities following burns are related to infections

rather than to osmotic shock and hypovolemia [18]. This has

encouraged the development of improved wound dressings

that provide an antimicrobial effect by eluting germicidal

compounds such as iodine (Iodosorb1, Smith & Nephew),

chlorohexidine (Biopatch1, J&J), or, most frequently, silver

ions (e.g., Acticoat1 by Smith & Nephew, Actisorb1 by J&J and

Aquacell1 by ConvaTec). Such dressings are designed to

provide controlled release of the active agent through a slow

but sustained release mechanism which helps avoid toxicity

yet ensures delivery of a therapeutic dose into the wound.

1.3. About the current study

It can be said that there is an increasing need to develop new

biodegradable materials for use in wound-healing applica-

tions. Soy protein is a very promising natural biodegradable

material, due to the abovementioned advantages. We there-

fore chose to develop and study soy protein films as a platform

for wound-dressing applications, in all relevant aspects.

We believe that a high quality soy protein film which is

hydrophilic but also cross-linked and therefore exhibits

desired relevant properties may provide properties that are

usually achieved using the bilayer concept.

Homogenous antibiotic-loaded yellowish films were cast

from aqueous solutions. A detailed description of the

preparation process was published elsewhere [19]. The

antibiotic drug gentamicin was incorporated into the films

in order to inhibit bacterial infection upon its controlled

release to the surrounding tissue. The mechanical and

physical properties of these films, the antibiotic release

profiles and their antimicrobial effects as well as cellular

response results, were recently published by us [19,20]. Since

these films are crosslinked and also contain plasticizer

(glycerol), they are strong and flexible and can be easily

handled. The new gentamicin-eluting soy protein wound

dressing which exhibited the best properties in the above

study was selected for the current in vivo study and compared

to a neat soy protein wound dressing without the drug and to

a standard Melolin dressing.

The guinea pig is often used as a dermatological and

infection model [21–24]. Research on guinea pigs has included

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 7 1461

topical antibiotic treatment [25], delivery of delayed-release

antibiotics [26], and investigation of wound-dressing mate-

rials [27,28]. A deep partial skin thickness burn is an

excellent wound model for the evaluation of wound healing,

not only for contraction and epithelialization from the

peripheral area such as in third-degree burns, but also for

evaluation of the recovery of skin appendages that serve

as the main source for the re-epithelialization which

completes the healing process. The metabolic response to

severe burn in guinea pigs is very similar to the human post-

burn metabolic response [29]. Furthermore, bacterial coloni-

zation and changes within the complement component

of the immune system in human burn victims is analogous

to guinea pigs affected by severe burns [23]. Such a model

was therefore used in the current study to evaluate the

effectiveness of our novel composite antibiotic-eluting

wound dressing.

2. Materials and methods

2.1. Materials

Non-genetically modified soy protein (Solpro 910TM, minimum

90% (w/w) protein, on a dry weight basis) was obtained as a

donation from SolbarTM (Ashdod, Israel).

Glycerol (G-7893) was used as plasticizer, glyoxal (50650)

was used as the cross-linking agent and the antibacterial drug

gentamicin sulfate (G-1264) (450–477 g/mol, Tm = 218–237 8C)

was used as the incorporated active agent (all purchased from

Sigma–Aldrich, Rehovot, Israel).

Reagent kit (8-1P31-25) and calibration kit (8-1P31-01) were

purchased from Sigma–Aldrich, Rehovot, Israel for analysis of

gentamicin concentrations.

2.2. Preparation of the soy protein films

Films of soy protein were prepared using the solvent casting

method. Soy protein solution was prepared by dissolving soy

protein powder in preheated double distilled water at 55 8C

and thoroughly stirred. Plasticizer (glycerol) and cross-linker

(glyoxal) were then added. For the gentamicin-loaded films,

gentamicin was first dissolved in 10 mL of the preheated

double distilled water and added to the soy protein solution

(following the addition of the cross-linker). All film parameters

are presented in Table 1.

The solution was constantly stirred with a magnetic stirrer

and kept at 55 8C for 30 min to allow the removal of bubbles.

The solution was then stirred at room temperature for an

additional 30 min to enable further removal of bubbles and

initial cross-linking. Finally, the solution was cast into two

Table 1 – Parameters of the studied films.

Parameter Value Relative to

Soy protein 5% (w/v) DDW

Plasticizer (glycerol) 35% (w/w) Soy protein

Cross-linker (glyoxal) 1% (w/w) Soy protein

Gentamicin 0% or 3% (w/w) Soy protein

low-density polyethylene plates (8.74 cm diameter) and left to

dry under ambient conditions for 72 h. In order to maintain

uniform thickness of the films, exactly 45 mL of solution were

cast into each plate. This amount yielded films with a

thickness of approximately 0.5 mm.

When dried, the films were carefully filed off the dishes and

cut into 45 mm discs. Each disk was then positioned between

two glass Petri dishes and subjected to thermal treatment in

an oven for 24 h at 80 8C. The specimens were kept in a

desiccator (RT, 30% RH) at room temperature and constant

humidity until use. All specimens were sterilized with

ethylene oxide and sealed in an airtight plastic bag.

2.3. In vitro gentamicin release

The soy protein films were cut into round discs (1 cm

diameter), which were individually placed in glass tubes in

triplicates and 3 mL of the release medium were added to each

tube. The tubes with the immersed discs were sealed with

plastic cups and kept in a microbiological incubator (Heraeus,

B12) at 37 8C and 100% humidity under static conditions. The

entire release medium from each tube was removed at

predetermined time points and was replaced by fresh medium

(3 mL). The removed medium was kept refrigerated at 4 8C

until analyzed.

Immunoassay analysis: Gentamicin concentrations were

determined using a high-speed immunoassay with an

ARCHITECT i2000SR (Abbott) system. Samples were diluted

with PBS (pH 7) with 0.02% sodium azide, as necessary.

2.4. In vivo burn model: procedure and treatment groups

Twenty female guinea pigs (Harlan, NL) were allocated for the

study after authorization by the Technion Animal Care and

Use Committee (authorization IL-092-09-2012). The animals

(average weight 300 g) were raised in a pathogen-free animal

facility. Following five acclimatization days, the animals were

randomly divided into three groups. All animals were

anesthetized by intramuscular injection of ketamine (40–

50 mg/kg) and xylazine (4–5 mg/kg). After shaving the animals’

skin, a depilatory cream (Orna 19, Alpha Cosmetica Israel) was

applied to complete hair removal. Two standardized deep

second-degree burns were inflicted on the back of each animal

on either side of the spine according to a validated method

described by Kaufman et al. (1990). Iron templates (circle,

D = 40 mm) were immersed in water preheated to 75 8C and

then placed in perfect contact with the animal’s skin for

exactly 5 s by applying light pressure. The burn area was

traced onto a transparent paper as a reference for later follow-

up. After infliction of the burns, each animal was seeded with

0.5 mL broth containing 1 � 108 CFU/mL Pseudomonas aerugi-

nosa using a micropipette. Each group was then treated with

the relevant treatment option, as follows:

Group 1 (6 animals, 12 wounds) was treated with a neutral

non-adherent dressing material (Melolin1, Smith & Nephew).

Melolin1 consists of three layers: a low adherent perforated

film, a highly absorbent cotton/acrylic pad and a hydrophobic

backing layer. According to the manufacturer, it enables rapid

drainage of wound exudate, thus reducing trauma to the

healing tissue. The dressing was placed directly on the burn,

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 71462

and was secured by an elastic adhesive bandage (Tenso-

plastTM, Smith & Nephew).

Group 2 (7 animals, 14 wounds) was treated with our soy

dressing, without antibiotics. A round disk slightly larger than the

burn area (D �45 mm) was placed directly on the burn, covered

with Melolin1, and secured as described above. This dressing

material was tested in order to evaluate the effect of the dressing’s

texture,materials,anddegradationonthewound-healingprocess.

Group 3 (7 animals, 14 wounds) was treated with the soy

dressing which released gentamicin. The dressing materials

were placed and secured as described above. These dressing

materials were tested in order to evaluate the effect of

antibiotic release kinetics on the wound-healing process.

Each of the animals was placed in an individual cage with

food ad libitum and allowed to recover. The animals received

analgesic treatment prior to the procedure and for 5

consecutive days (buprenorphine, 5 mg/kg per os).

2.5. Wound closure

Twelve days following the burn, wound areas were traced on a

transparent paper and scale bar photographs were taken. The

burn wounds were evaluated macroscopically by two quanti-

tative parameters: (i) percentage of the original area subjected

to burn which was still an open wound, and (ii) wound

contraction as depicted by the total wound area (epithelialized

and non-epithelialized) as a percentage of the original area

subjected to burn. The measurements of the two burns within

one animal were averaged to produce a single data, from

which the total group average was calculated.

2.5.1. Post mortem examinationThe animals were sacrificed and 1 cm2 biopsies were taken

from the center of the wound and immediately fixed in

phosphate buffered formalin.

2.5.2. Biopsy evaluationsFollowing the fixation stage, biopsies were embedded in

paraffin using a standard embedding protocol, and 5 mm

sections were prepared using a Leica microtome. Sections

were stained with hematoxylin and eosin (H&E), observed and

photographed under �20 and �100 magnification using an

Olympus upright light microscope. Wound-healing analysis

was conducted in a blinded manner by two separate

evaluators using a semi-quantitative grading system. The

data obtained from the observers was analyzed and based on

the average of the two burns within one animal to produce a

single observation per animal. The sections were evaluated

based on structure and content. The wound-healing criteria

included epithelial coverage, epidermal thickness, epidermis-

dermis attachment, neoangiogenesis, number of follicles

(adnexa), dermis thickness, white blood cells (WBC) and

collagen. Grading was between 0 and 3: 0 absence, 1 minimal

presence, 2 moderate presence and 3 extensive presence,

except for WBC, where 0-extensive presence and 3-absence.

2.6. Statistical analysis

Means and standard deviation of sample (SD) were calculated

for each data set. Differences between means were analyzed

for statistical significance using a one-way ANOVA with the

Tukey-Kramer multiple comparisons posttest (SPSS version

17.0). p values �0.05 were considered significant.

3. Results

3.1. In vitro gentamicin release profile

Cumulative release of gentamicin from the soy protein wound

dressings presented an initial burst release (6 h) of approxi-

mately 54% followed by two phases of release. The first phase

exhibited a decreasing release rate for a period of 12 days,

followed by a second phase of a constant slow release rate of

0.15% drug/day (R2 = 0.9685) for more than 40 days (Fig. 1).

3.2. Wound closure

Second-degree burns in guinea pigs were used as a wound-

healing model for testing our novel soy wound dressing, which

is a membrane that covers the wound and allows slow release

of gentamycin. Pseudomonas was applied topically immedi-

ately after the creation of the burns in order to bypass the

sterile conditions of the pathogen-free room and to mimic

burn contamination that typically occurs in patients with

burns.

Twenty guinea pigs were included in the study and were

divided into three groups: Melolin-control group, soy protein

dressing without antibiotic, and soy protein dressing with

controlled release of gentamicin (Fig. 2). The closed (epithe-

lialized) wound area and the open (non-epithelialized) wound

area were traced on a transparent paper 12 days after creation

of the burns. The degree of healing was calculated by the

percentage of the epithelialized (closed) area on the 12th day

compared to the total burn area on the 12th day. On the 12th

day post-burn, epithelialization of the wounds that were

dressed with Melolin covered 55% � 15% of the burn area, the

average closure of the soy protein group was 62% � 13%,

whereas the closure area of the soy protein group with

gentamicin release reached 71% � 11% (Fig. 3A). The differ-

ence in epithelialization and wound closure between the soy

gentamicin group and the Melolin group was statistically

significant ( p = 0.04), while no significant difference was noted

between the soy group and the Melolin group.

3.3. Wound contraction

Wound contraction was calculated as 1 minus the total wound

area (epithelialized and non-epithelialized) divided by the

original area subjected to burn. A significant reduction was

found in the contraction rate between the soy treated groups

(36% � 8%, p < 0.02 for soy only, and 31% � 10%, p < 0.005 for

soy with gentamicin release) as compared to the Melolin group

(51% � 12%) (Fig. 3B).

3.4. Histological evaluation

Animals were sacrificed after 12 days, and 1 cm2 biopsies

were taken from the center of the wound and immediately

fixed in phosphate buffered formalin. Two different observers

Fig. 1 – In vitro cumulative release (mean W SD) of gentamicin during (A) 56 days and (B) the first 12 h of release. Second

phase of continuous release (14–56 days) was fitted to a linear line (C) to demonstrate a constant release rate.

Fig. 2 – Representative photographs of wounds 7 and 12

days post-treatment with MelolinW (A1 and A2,

respectively), soy dressing without antibiotics (B1 and B2,

respectively) and soy dressing with controlled gentamicin

release (C1 and C2, respectively). Bars = 1 cm.

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 7 1463

examined the biopsies that were stained with H&E (Fig. 4A–C),

according to the abovementioned criteria. The gentamicin

soy group was significantly superior to the other two groups

( p < 0.05) (Fig. 4D): better epithelialization and neoangiogen-

esis, thicker dermis, fewer WBC, more follicles, and greater

collagen content. The general scores differed significantly for

the gentamicin release soy dressing (10.8 � 1.4) versus the soy

and the Melolin treatment groups (7.2 � 1.7; 7.2 � 1.2, respec-

tively, p < 0.05). Wounds treated with controlled release of

gentamicin obtained significantly higher histological scores

than both controls after 12 days. This was mainly due to

significant re-epithelialization of the wound ( p = 0.02), better

epidermis-to-dermis adherence (p = 0.02), amount of collagen

(p = 0.01) and more skin appendages ( p = 0.03). The differ-

ences in the amount of neoangiogenesis, WBC and the

thickness of the dermis were not significant.

Furthermore, the soy dressing material was found to create

good contact with the wound, and the soy itself did not cause

any irritation to the animal’s skin.

4. Discussion

4.1. Gentamicin release profile

Commercial wound dressings that incorporate iodine (Iodo-

sorb1 by Smith & Nephew), chlorohexidime (Biopatch1 by J&J)

Fig. 3 – (A) Percentage of wound healing at 12 days post-

burn creation. (B) Wound contraction as percentage of total

measured wound area (epithelialized and non-

epithelialized) at 12 days post-burn creation relative to the

wound area at the time of wound creation. Each dot

represents the average of two burns on each animal

(mean W SD). Square (&) represents measurements of

superimposed dots from two different animals. Triangle

(~) represents measurements of superimposed dots from

three different animals. *Statistically significant as

compared to Melolin treated group.

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 71464

or most frequently silver ions (e.g., Acticoat1 by Smith &

Nephew, Actisorb1 by J&J and Aquacel1 by ConvaTec) as

active agents are available in the market. Such dressings are

designed to provide controlled release of the active agent

through a slow but sustained release mechanism which helps

avoid toxicity yet ensures delivery of a therapeutic dose to

the wound. Despite frequent usage, there is growing evidence

that silver is highly toxic to keratinocytes and fibroblasts and

may delay burn wound healing if applied indiscriminately

to healing tissue areas [30–32]. Furthermore, most of these

dressing materials undergo frequent changes. Bioresorbable

dressings may successfully address this shortcoming, since

they do not need to be removed from the wound surface once

they have fulfilled their role.

In our previous study of the soy protein wound dressings

loaded with gentamicin we in vitro studied the antimicrobial

effect of gentamicin release profiles on bacterial inhibition

and also evaluated the cytotoxicity on fibroblast cells. Our

main conclusions were that the gentamicin release from our

soy films demonstrated great efficiency toward the relevant

bacteria, which are abundant on human skin [20]. The films

could effectively inhibit S. aureus and S. albus infections for at

least two weeks and P. aeruginosa for three days. Filtered film

extracts released during the first 24 h and from the 24–48 h

showed no cytotoxic effect on the fibroblast cells, although the

gentamicin quantities released were much higher than these

of silver ion that are used in commercial wound dressings [20].

Based on these results, a selected soy protein film was used

in the current animal study.

The in vitro gentamicin release profile from our selected

film showed a moderate burst effect (54%) during the first 6 h,

followed by a continuous decrease in release rates during the

first two weeks. This was followed by a third stage of zero-

order release (constant release rate) that lasted until the end

of the experiment (8 weeks). It should be mentioned that the

specimens maintained their integrity throughout the entire

test period. The burst effect and release profile during the first

two weeks were typical of diffusion-controlled systems. The

third phase of constant release rate presumably involved

some degradation of the soy protein matrix combined with

diffusion of the remaining drug that was more firmly attached

to the protein chains.

The obtained release profile could be beneficial for

application as antibiotic-eluting wound dressings. During

the first hours after being wounded, a relatively high drug

release is essential to eliminate infections that may not have

been eliminated during wound cleansing. Later, a continuous

low release rate can keep the wound ‘‘infection free’’ for more

than 2 weeks, which is the time usually taken for wound

healing.

It should be noted that most of the gentamicin (75%) was

released within the first week of the study, due to the

hydrophilic nature of the antibiotic and the soy protein matrix.

The hydrophilic nature of soy protein enables relatively rapid

water intake, leading to full swelling of the matrix within a few

hours of immersion. Unfavorable more rapid drug release

rates have been reported in the literature for other antibiotic-

eluting systems [33]. Controlling the release of antibiotics from

these systems is challenging due to the hydrophilic nature of

both the drug and the host polymer. In most cases the drug

reservoir is depleted in less than 2 days, resulting in a very

short antibacterial effect. Thus, our new antibiotic-eluting soy

protein wound dressing is advantageous over other systems.

The obtained gentamicin release profile can be explained by

the fact that many antibiotic drugs bind proteins via van der

Waals or ionic interactions. The bonded portion may act as a

reservoir and be released more slowly than the unbound form.

0–30% of the adsorbed gentamicin binds to albumin [34].

Furthermore, gentamicin is a highly charged polycation (+3.5,

pH 7.4) [35], whereas soy protein has many negatively charged

Fig. 4 – Representative histological sections of wound areas taken from the center of the wound 12 days post-burn creation.

(A) Melolin treated group. (B) Soy dressing without antibiotics. (C) Soy dressing with controlled gentamicin release. (D)

Scoring of the histological sections of the wound area taken from the center of the wound 12 days post-burn creation. The

examined wound healing criteria included epithelial coverage, epidermal thickness, epidermis-dermis attachment,

neoangiogenesis, number of follicles (adnexa), dermis thickness, WBC and collagen. Grading was between 0 and 3: 0

absence, 1 minimal presence, 2 moderate presence and 3 extensive presence, except for WBC, where 0-extensive presence

and 3-absence (mean W SD).

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 7 1465

carboxyl groups. It is thus probable that some ionic bonding

took place. Such a bonding mechanism has been shown before

between deprotonated carboxyl groups of succinylated colla-

gen and positive anion groups of the gentamicin molecule [36].

4.2. In vivo study

The post-burn end-point was chosen to be twelve days. At this

stage, the wounds in all groups demonstrated epithelialization

of more than half of the original wound area. The 3 study

groups, with or without controlled release of gentamicin, were

found to have 29–45% non-epithelialized areas and 31–50%

contraction. A better epithelialization rate was observed for

the soy with gentamycin dressing compared to soy without

gentamycin and compared to the standard of care treatment

of Melolin. Furthermore, a significant decrease in contraction

rate was observed in the soy dressing with and without

controlled release of gentamicin as compared to the Melolin

dressing group (31% � 10% (p < 0.005), 36% � 8% ( p < 0.02) and

51% � 12, respectively). Wound contraction although it is an

inherent part of a normal wound healing we would like to keep

it to a minimum, since in humans it may lead to disfigurement

of the skin and poor esthetic results. It may also lead to loss of

the normal flexibility of the skin – a fixed deformity that entails

a functional disability. The reduced effect of the contraction,

which becomes a dominant mechanism when healing is

delayed, is achieved by the significantly accelerated epitheli-

alization as observed in the controlled release of gentamicin

from the wound dressing group. Furthermore, it should be

stressed that the soy dressing did not evoke a negative

response due to its chemical, structural or physical properties.

The controlled release of gentamicin dressing may have a

systemic effect, and thus each animal received the same

dressing from each treatment group to prevent cross

contamination.

A clear benefit for the soy with gentamycin dressing was

also found in the histological parameters compared to the

control groups, meaning that the quality of healing was better

when using the gentamicin dressing. Histological evaluation

of the wounds showed superior organization of the epithelium

and dermis in the gentamicin soy dressing compared to the

control groups. The general scores differed significantly

(10.8 � 1.4, for the gentamicin soy dressing compared to the

soy group, 7.2 � 1.7, and the Melolin group, 7.2 � 1.2, p < 0.05).

The quality of the epidermis-dermis junction was better and

there were more collagen fibers in the gentamicin dressing.

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 71466

Another important finding in groups treated with controlled

release of gentamicin was the dominance and good quality of

the hair follicles and adnexa which are the source for re-

epithelialization from within the burn area.

In conclusion, in vivo evaluation of the gentamicin-eluting

soy protein dressing material in a contaminated wound

demonstrated its advantage over the other investigated dres-

sings. We consider the better qualities of the healed woundinthe

gentamycin soy dressing to be one of the most important

advantages of this new dressing. The gold standard of local

treatment with topical antibacterial agents, e.g. Silverol1, is

changed daily or twice-daily, which is time consuming and

painful to the patient as well as less economical. Several of the

currently used dressing materials that provide controlled release

of silver ions as an antibacterial agent might have toxic effect on

cells, which can delay wound healing. The current dressing

material shows promising results, although it still needs

evaluation in a clinical trial.

Conflict of interest statement

All the authors state no conflict of interest.

Acknowledgements

The authors are grateful to the Israel Science Foundation (ISF,

grant no. 1055/11) for supporting this research. We would like

to thank SolbarTM (Ashdod, Israel) for kindly providing the soy

protein isolate used in this study.

r e f e r e n c e s

[1] Kumar R, Choudhary V, Mishra S, Varma IK, Mattiason B.Adhesives and plastics based on soy protein products.Ind Crops Prod 2002;16:155–72.

[2] Kumar R, Zhang L. Aligned ramie fiber reinforced arylatedsoy protein composites with improved properties. ComposSci Technol 2009;69:555–60.

[3] Lodha P, Netravali AN. Thermal and mechanical propertiesof environment-friendly green plastics from stearic acidmodified-soy protein isolate. Ind Crops Prod 2005;21:49–64.

[4] Zhang J, Mungara P, Jane J. Mechanical and thermalproperties of extruded soy protein sheets. Polymer2001;42:2569–78.

[5] Chen Y, Zhang L, Gu J, Liu J. Physical properties ofmicroporous membranes prepared by hydrolyzingcellulose/soy protein blends. J Membr Sci 2004;241:393–402.

[6] Chen L, Subirade M. Elaboration and characterization ofsoy/zein protein microspheres for controlled nutraceuticaldelivery. Biomacromolecules 2009;10:3327–34.

[7] Ciannamea EM, Federica SM, Saumell C, Stefani PM,Ruseckaite RA. Soybean protein films. Characterization andpotential as novel delivery devices of Duddingtonia flagransChlamydospores. Biol Control 2013;66:92–101.

[8] Liu K. Expanding soybean food utilization. Food Technol2000;54:46–58.

[9] Park SK, Hettiarachchy NS, Ju ZY, Gennadios A. Formationand properties of soy protein films and coatings. Protein-based films and coatings. CRC 2002;123–37.

[10] Cao N, Fu Y, He J. Preparation and physical properties of soyprotein isolate and gelatin composite films. Food Hydrocoll2007;21:1153–62.

[11] Santin M, Ambrosio L. Soybean-based biomaterials:preparation, properties and tissue regeneration potential.Expert Rev Med Dev 2008;5:349–58.

[12] Silva GA, Vaz CM, Coutinho OP, Cunha AM, Reis RL. In vitrodegradation and cytocompatibility evaluation of novel soyand sodium caseinate-based membrane biomaterials.J Mater Sci: Mater Med 2003;14:1055–66.

[13] Zilberman M, Peles Z, Soy Olami H. Protein as a biomaterialfor wound healing applications. Encycl Polym Mater PolymBiomater 2014 [in press].

[14] Snyders R, Shingel KI, Zabeida O, Roberge C, Faure MP,Martinu L, et al. Mechanical and microstructural propertiesof hybrid poly (ethylene glycol)-soy protein hydrogels forwound dressing applications. J Biomed Mater Res A2007;83:88–97.

[15] Shingel KI, Di Stabile L, Marty JP, Faure MP. Inflammatoryinert poly(ethylene glycol) protein wound dressingimproves healing responses in partial and full thicknesswounds. Int Wound J 2006;3:332–42.

[16] Santos TC, Horing B, Reise K, Marques AP, Silva SS, OliveiraJM, et al. In vivo performance of chitosan/soy-basedmembranes as wound-dressing devices for acute skinwounds. Tissue Eng A 2013;19:860–9.

[17] Field CK, Kerstein MD. Overview of wound healing in amoist environment. Am J Surg 1994;167:S2–6.

[18] Revathi G, Puri J, Jain BK. Bacteriology of burns. Burns1998;24:347–9.

[19] Peles Z, Zilberman M. Novel soy protein wound dressingswith controlled antibiotic release: mechanical and physicalproperties. Acta Biomater 2011;8:209–17.

[20] Peles Z, Binderman I, Berdicevsky I, Zilberman M. Soyprotein films for wound healing applications: antibioticrelease, bacterial inhibition and cellular response.J Tissue Eng Regen Med 2013;7:401–12.

[21] Kaufman T, Lusthaus SN, Sagher U, Wexler MR. Deeppartial thickness burns: a reproducible animal model tostudy burn wound healing. Burns 1990;16:13–6.

[22] Yannas IV, Burke JF, Orgill DP, Skrabut EM. Wound tissuecan utilize a polymeric template to synthesize a functionalextension of skin. Science 1982;215:174–6.

[23] Bjornson AB, Bjornson HS, Lincoln NA, Altemeier WA.Relative roles of burn injury, wound colonization, andwound infection in induction of alterations of complementfunction in a guinea pig model of burn injury. J Trauma1984;24:106–15.

[24] Orenstein A, Klein D, Kopolovic J, Winkler E, Malik Z, KellerN, et al. The use of porphyrins for eradication ofStaphylococcus aureus in burn wound infections. FEMSImmunol Med Microbiol 1997;19:307–14.

[25] Boon RJ, Beale AS, Sutherland R. Efficacy of topicalmupirocin against an experimental Staphylococcus aureussurgical wound infection. J Antimicrob Chemother1985;16:519–26.

[26] Galandiuk S, Wrightson WR, Young S, Myers S, Polk Jr HC.Absorbable, delayed release antibiotic beads reducesurgical wound infection. Am Surg 1997;63:831–5.

[27] Kawai K, Suzuki S, Tabata Y, Taira T, Ikada Y, Nishimura Y.Development of an artificial dermis preparation capable ofsilver sulfadiazine release. J Biomed Mater Res 2001;57:346–56.

[28] Mazurak VC, Burrell RE, Tredget EE, Clandinin MT, Field CJ.The effect of treating infected skin grafts with Acticoat onimmune cells. Burns 2007;33:52–8.

[29] Herndon DN, Wilmore DW, Mason AD. Development andanalysis of a small animal model simulating the humanpostburn hypermetabolic response. J Surg Res 1978;25:394–403.

b u r n s 4 1 ( 2 0 1 5 ) 1 4 5 9 – 1 4 6 7 1467

[30] Poon VK, Burd A. In vitro cytotoxity of silver: implicationfor clinical wound care. Burns 2004;30:140–7.

[31] Paddle-Ledinek JE, Nasa Z, Cleland HJ. Effect of differentwound dressings on cell viability and proliferation. PlastReconstr Surg 2006;117:110S–8S.

[32] Burd A, Kwok CH, Hung SC, Chan HS, Gu H, Lam WK, et al.A comparative study of the cytotoxicity of silver-baseddressings in monolayer cell, tissue explant, and animalmodels. Wound Repair Regen 2007;15:94–104.

[33] Sripriya R, Kumar MS, Sehgal PK. Improved collagen bilayerdressing for the controlled release of drugs. J Biomed MaterRes B Appl Biomater 2004;70(2):389–96.

[34] Wishart DS, Knox C, Guo AC, Shrivastava S, Hassanali M,Stothard P, et al. DrugBank: a comprehensive resource forin silico drug discovery and exploration. Nucl Acids Res2006;34(Database Issue):D668.

[35] Abraham AM, Walubo A. The effect of surface charge onthe disposition of liposome-encapsulated gentamicin tothe rat liver: brain, lungs and kidneys after intraperitonealadministration. Int J Antimicrob Agents 2005;25(5):392–7.

[36] Bloomfield SE, Miyata T, Dunn MW, Bueser N, Stenzel KH,Rubin AL. Soluble gentamicin ophthalmic inserts as a drugdelivery system. Arch Ophthalmol 1978;96(5):885–7.