Expert Opinion on Drug Delivery

ISSN: 1742-5247 (Print) 1744-7593 (Online) Journal homepage: http://www.tandfonline.com/loi/iedd20

Injectable hybrid delivery system composed of

gellan gum, nanoparticles and gentamicin for the
localized treatment of bone infections

Urszula Posadowska, Monika Brzychczy-Włoch, Anna Drożdż, Małgorzata

Krok-Borkowicz, Małgorzata Włodarczyk-Biegun, Piotr Dobrzyński, Wojciech
Chrzanowski & Elżbieta Pamuła

To cite this article: Urszula Posadowska, Monika Brzychczy-Włoch, Anna Drożdż, Małgorzata
Krok-Borkowicz, Małgorzata Włodarczyk-Biegun, Piotr Dobrzyński, Wojciech Chrzanowski
& Elżbieta Pamuła (2016) Injectable hybrid delivery system composed of gellan gum,
nanoparticles and gentamicin for the localized treatment of bone infections, Expert Opinion on
Drug Delivery, 13:5, 613-620, DOI: 10.1517/17425247.2016.1146673

To link to this article: http://dx.doi.org/10.1517/17425247.2016.1146673

Accepted author version posted online: 25

Jan 2016.
Published online: 16 Feb 2016.

Submit your article to this journal

Article views: 62

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=iedd20

Download by: [149.156.96.151] Date: 22 April 2016, At: 08:41

 EXPERT OPINION ON DRUG DELIVERY, 2016
VOL. 13, NO. 5, 613–620
http://dx.doi.org/10.1517/17425247.2016.1146673

ORIGINAL RESEARCH

Injectable hybrid delivery system composed of gellan gum, nanoparticles and

gentamicin for the localized treatment of bone infections
Urszula Posadowskaa, Monika Brzychczy-Włochb, Anna Drożdża, Małgorzata Krok-Borkowicza,
Małgorzata Włodarczyk-Biegunc, Piotr Dobrzyńskid, Wojciech Chrzanowskie and Elżbieta Pamułaa
a
Department of Biomaterials, Faculty of Materials Science and Ceramics, AGH University of Science and Technology, Krakow, Poland; bDepartment
of Microbiology, Medical College, Jagiellonian University, Krakow, Poland; cLaboratory of Physical Chemistry and Soft Matter, Wageningen
University, Wageningen, The Netherlands; dCenter of Polymer and Carbon Materials, Polish Academy of Sciences, Zabrze, Poland; eFaculty of
Pharmacy, The University of Sydney, Sydney, Australia

ABSTRACT ARTICLE HISTORY

Objectives: Bone infections are treated with antibiotics administered intravenously, antibiotic-releasing Received 24 October 2015
bone cements or collagen sponges placed directly in the infected area. These approaches render Accepted 22 January 2016
limited effectiveness due to the lack of site specificity and invasiveness of implanting cements and Published Online
12 February 2016
sponges. To address these limitations, we developed a novel polysaccharide hydrogel-based injectable
system that enables controlled delivery of gentamicin (GENT). Its advantages are minimal invasiveness, KEYWORDS
Downloaded by [149.156.96.151] at 08:41 22 April 2016

and localized and finely regulated release of the drug. Controlled drug delivery;
Methods: GENT was incorporated both directly within the gellan gum hydrogel and into poly(L-lactide- gellan gum; gentamicin
co-glycolide) nanoparticles embedded into the hydrogel. (GENT); nanoparticles;
Results: We confirmed the injectability of the system and measured extrusion force was 15.6 ± 1.0 N, osteomyelitis;
which is suitable for injections. The system set properly after the injection as shown by rheological poly(lactide-co-glycolide)
(PLGA); biocompatible
measurements. Desired burst release of the drug was observed within the first 12 h and the dose
materials
reached ~27% of total GENT. Subsequently, GENT was released gradually and sustainably: ~60% of
initial dose within 90 days. In vitro studies confirmed antimicrobial activity of the system against
Staphylococcus spp. and cytocompatibility with osteoblast-like cells.
Conclusions: Developed injectable system enables minimally invasive, local and sustained delivery of
the pharmaceutically relevant doses of GENT to combat bone infections.

1. Introduction it has been widely used for the production of several medical
devices (e.g. sutures, osteosynthesis screws and plates, scaf-
Intravenous administration of gentamicin (GENT) is considered
folds for bone tissue regeneration).[8,9]
as the first-line treatment of bone infections leading to osteo-
More recently, different classes of hydrogels emerge as
myelitis. Typically, a high dose of the drug is required for a
promising systems for drug delivery and tissue regeneration.
prolonged period (at least 6 weeks) to be effective.[1,2]
[10–13] In our experiments, we used gellan gum, a water-
However, prolonged administration of the drug may cause
soluble linear-anionic hydrogel that is produced during a fer-
serious side effects such as nephrotoxicity and ototoxicity,
mentation process by Pseudomonas elodea. It is composed of
which is a major drawback of the therapy.[3] The second line
[β-D glucose–D-glucuronic acid–β-D glucose–L-rhamnose].[14]
of osteomyelitis treatment uses solid implants placed locally at
gellan gum is used as a food supplement and has been
the infected site: bone cements or collagen sponges saturated
previously shown to be effective in delivery of amoxicillin
with GENT.[4,5] For both types of implants, invasive surgery is
and theophylline.[15,16] gellan gum was also used to produce
required to place them at the specific site in the body. To
scaffolds for cartilage and bone regeneration.[17,18] Taken
overcome the limitations of aforementioned approaches, drug
together, the ability to form hydrogel, biocompatibility and
releasing injectable systems have been proposed.[6]
tunable physical properties of gellan gum raise tantalizing
Here, we developed a new class of injectable system com-
possibility to formulate injectable drug delivery systems for
posed of GENT-loaded nanoparticles (NPs), embedded in the
localized and controlled delivery of antibiotics for osteomyeli-
polysaccharide hydrogel gellan gum (GG). We hypothesized
tis treatment.
that NPs would ensure controlled and sustained drug release,
The combination of these favorable characteristics of gellan
while the hydrogel would enable localized drug release by
gum with drug-releasing NPs, which are contained in the
maintaining NPs at the infected site. FDA and EMA-approved
hydrogel and which allow for localized and sustained delivery
poly(lactide-co-glycolide) (PLGA) was chosen to produce NPs.
of antibiotics at the site of infection, is a primary novelty.
PLGA has tunable degradation rate by altering molar ratio of
Furthermore, injectability and the easy of application is a
lactide to glycolide, chain structure, molecular weight,[7] and

CONTACT Elżbieta Pamuła epamula@agh.edu.pl Faculty of Materials Science and Ceramics, Department of Biomaterials, AGH University of Science and
Technology, Al. A. Mickiewicza 30, 30-059 Kraków, Poland.
© 2016 Informa UK Limited, trading as Taylor & Francis Group
 614 U. POSADOWSKA ET AL.
significant advantage of the system which adds to the novelty
and to its clinical relevance. Ultimately, injectable system that
enables effective treatment of bone infections represents a
significant cost reduction and improves comfort for patients
including shorter time to recovery.
The primary objective of our study was to develop a new
minimally invasive hydrogel-based drug delivery system that is
injectable and releases the drug locally with predefined
release profile. The injectability of the hydrogel is required to
enable the local delivery of the system to the infected site,
while temporal stability of the system is necessary to sustain
drug release at this site. To achieve these objectives, we
examined the drug release from (1) NPs only and (2) a system
containing gellan gum and NPs. The influence of NPs and
dissolved GENT on injectability and rheological properties
was investigated. Importantly, we interrogated antimicrobial
activity toward the bacterial strains causing osteomyelitis
(Staphylococcus spp.). Cytotoxicity of the extracts from the
system in contact with osteoblast-like cells was examined to
confirm its biocompatibility.
Downloaded by [149.156.96.151] at 08:41 22 April 2016
2. Materials and methods
2.1. Materials
GENT (Gentamicini sulfas, C21H43N5O7·32SO4) was purchased from
Galpharm, Poland. Gellan gum (GelzanTM, low-acyl according to
the manufacturer) was purchased from Sigma-Aldrich, Poland.
Methanol, mercaptoethanol, isopropanol, orthophthaldialdehyde
(OPA), phosphate-buffered saline concentrate (PBS buffer), and
calcium chloride were all of analytical grade and were purchased
from POCh, Poland. Ultrahigh-quality water (UHQ-water) was
produced in a UHQ-PS purification system (Elga, UK).
GENT-loaded nanpoarticles (NPs) were prepared using
PLGA (85:15, Mn = 80 kDa, d = 1.9, synthesized with the use
of zirconium acetylacetonate as an initiator) via double emul-
sification solid–oil–water (S/O/W) as described previously.
[19,20] In brief: 5 mg of GENT (solid – S) was emulsified in a
1.67% w/v PLGA solution in dichloromethane (6 ml, oil – O),
producing the primary emulsion (S/O). Then S/O was intro-
duced dropwise into an aqueous solution of PVA (4% w/v,
15 ml) on ice during ultrasonication (3 min, 40% of the cycle,
Sonics Vibra CellTM, USA) producing the secondary phase (S/
O/W). Next, the emulsion was stirred at 500 rpm overnight to
allow dichloromethane evaporation. The dispersion of solid
NPs was centrifuged (14,000 rpm, 4°C, 15 min) and thoroughly
flushed three times with UHQ-water. Obtained NPs were
freeze-dried for 48 h and then stored at 4°C. Before further
analyses, NPs were redispersed in UHQ-water using ultrasound
probe. We have previously reported [19] that NPs had dia-
meter of 240 ± 11 nm and ζ potential of −6.0 ± 0.5 mV. Drug
loading efficiency was 3.3 ± 0.6%, as evaluated by OPA assay
by the measurement of fluorescence (λex. – 340 nm, λem. –
455 nm; FLUOstar Omega, BMG Labtech).[21]
2.2. Preparation of hydrogel-based injectable system
Gellan gum was dissolved in UHQ-water at an initial concentra-
tion of 1.4% w/v at 90°C. The temperature of the solution was
decreased to 50°C and 3.0% w/v suspension of NPs (encapsu-
lated antibiotic 0.1% w/v; denoted as GENTE) combined with
dissolved GENT (antibiotic concentration 0.1% w/v; denoted as
GENTD) were added. Final antibiotic concentration in the sample
was 0.2% w/v.
To modulate the viscoelastic properties, the system was
cross-linked using CaCl2 (0.6% w/v) and such samples were
denoted as GG/GENTD+E. Additionally, control samples con-
taining only hydrogel and 0.6% w/v CaCl2 were prepared
(denoted as GG). In all samples, the final concentration of
gellan gum was 0.7% w/v.
2.3. Injectability
The injectability of the samples (GG, GG/GENTD+E) was charac-
terized as described previously.[22] Force curves and the max-
imal force (FMax) required to extrude 1 ml of a sample from a
5 ml syringe equipped with a standardized needle (18 G,
0.85 mm inner diameter and 1.45 mm outer diameter, Becton
Dickinson) were measured in a compression test (Zwick 1435,
Germany). The crosshead speed of the testing machine was
50 mm/min.
2.4. Rheological characteristics
To characterize rheological properties of GG and GG/GENTD+E
samples, strain sweep test was performed (Physica MCR 501
Rheometer, Anton Paar, Graz, Austria; Couette CC10/T200
coaxial geometry with a bob diameter 10.002 mm, cup dia-
meter 10.845 mm). Samples were placed in the measuring cell
and covered with paraffin oil to avoid drying. Gelation was
analyzed by measuring storage and loss moduli (G' and G'') as
a function of time during application of oscillatory sinusoidal
deformation (0–30 min, frequency f = 1 Hz, strain c = 0.01%).
During the second time interval (30–60 min), gels were broken
by logarithmically increasing strain (c = 0.01–100%, f = 1 Hz).
In the third time interval (60 – 90 min), behavior of gel after
breaking was analyzed (f = 1 Hz, c = 0.01%). All measurements
were performed at 37°C.
2.5. Drug release
The release profiles of antibiotic from NPs and GG/GENTD+E
were analyzed using diffusion method. Dialysis bags
(ZelluTransRoth, MWCO 12 kDa) were filled with 1 ml of NPs
suspension (30 mg of NPs, containing in total 1 mg of GENT)
or 1 ml of GG/GENTD+E (containing in total 2 mg of GENT, i.e.
1 mg of dissolved GENT and 1 mg of GENT encapsulated in
NPs), immediately immersed in the vials with 20 ml of PBS at
pH = 7.2 and stirred (50 rpm) at 37°C. At predetermined time
intervals, 0.5 ml of sample were collected up to 90 days and
GENT concentration was quantified by OPA assay as described
in Section 2.1.
2.6. Antimicrobial activity
In order to assess antimicrobial activity of GG/GENTD+E, agar
diffusion tests according to the Kirby–Bauer method were
conducted. The growth inhibition of two Gram-positive

pathogens Staphylococcus aureus (DSM 24167 from Deutsche
Sammlung von Mikroorganismen und Zellkulturen) and
Staphylococcus epidermidis (ATCC 700296 (American Type
Culture Collection) was observed. The clinical strains of S.
aureus (SA1-KCR) and S. epidermidis (SE1-KCR) isolated from
patients with joint infections (Krakow Centre of Rehabilitation
and Orthopedics, 2012) were also used. Aforementioned
strains (reference and clinical isolates) were incubated in
5 mL of Bacto™ Tryptic Soy Broth (Becton Dickinson) for
16 h at 37°C and prepared at the concentration 0.5 on the
McFarland scale (1.5 × 108 CFU/ml) in 0.7% w/v NaCl solution.
Then pathogens were seeded on Mueller–Hinton agar (Difco,
UK). To check if method of samples placement influences
detected antibacterial activity, the plates were divided into
two sectors. On the first sector, a well 5 mm in diameter
was made with a hole punch, into which a tube shaped GG/
GENTD+E (2 mm in height, 5 mm in diameter, ~40 µl) was
introduced. On the second sector, a similar GG/GENTD+E tube
shaped sample was placed. As positive controls, the GENT
antimicrobial susceptibility standard discs (CN 10 µg, Oxoid,
Downloaded by [149.156.96.151] at 08:41 22 April 2016
UK) were applied. GG samples were tested as negative con-
trols. The plates were then incubated at 37°C for 18 h and the
diameter of inhibition zone of the bacterial growth was mea-
sured using Calibrating Viewer.
2.7. In vitro cytocompatibility
Gellan gum solution was prepared in sterile conditions, CaCl2
solution was sterile-filtered, and NPs were treated with UV-
light for 40 min. MG-63 cells (European Collection of Cell
Cultures, Salisbury, UK) were cultured in an extract from sam-
ples placed in Eagle’s minimal essential medium (EMEM, PAN
BIOTECH, Germany) supplemented with 10% fetal bovine
serum, 1% penicillin–streptomycin, 0.1% sodium pyruvate
(PAA, Austria). To obtain the extract, tube-shaped samples
were incubated in EMEM for 24 h at 37°C (0.1 g of material
per 1 ml of medium). Next, the extract was diluted in EMEM by
factors of 1/1 (undiluted), 1/2 and 1/8. MG-63 cells (1.5 × 10-
4
cells/cm2) were cultured in a 48-well plate (Nunclon) for 24 h
and then incubated in undiluted and diluted extracts (1 ml).
Cells cultured in EMEM acted as reference. Subsequently
0.05 ml Alamar Blue reagent (10% w/v resazurin sodium salt
Figure 1. Force changes as a function of piston distance during injection of gellan gum hydrogel (GG) (A) and the system composed of gellan gum with dissolved and
encapsulated gentamicin (GG/GENTD+E) (B); the values of maximal extrusion force (FMax, mean ± S.E.M.) are shown at top right corners.
EXPERT OPINION ON DRUG DELIVERY 615
solution in PBS, Sigma-Aldrich) was added, cells were incu-
bated for 4 h at room temperature and then fluorescence was
measured using FLUOstar Omega, BMG Labtech (λex. – 530 nm,
λem. – 590 nm). In Alamar Blue test, resazurin acts as an
oxidation–reduction indicator of cellular metabolic activity.
The reduced form resorufin is highly fluorescent, and the
intensity of fluorescence produced is proportional to the num-
ber of living cells respiring. Thus, this test measures directly
and quantitatively cell viability.
Reduction of Alamar Blue reagent was calculated according
to the formula:


Sx  Scontrol
% Reduction of Alamar Blue ¼ 100%reduced 100%
ðS  Scontrol Þ
where Sx is the fluorescence of samples, Scontrol is the fluores-
cence of EMEM without cells, S100%reduced is the fluorescence
of reagent reduced in 100% (reagent with EMEM was auto-
claved for 15 min at 121°C).
Cell attachment, spreading and viability were evaluated
24 h and 6 days post-seeding using optical (interference con-
trast and surface area analysis) and fluorescence microscopy
(Axiovert 40 Carl Zeiss, Germany) after live/dead staining
(10 min incubation in calcein-AM and propidium iodide, 2
and 4 µM in PBS, respectively, at 37°C).
2.8. Statistics
All results were presented as a mean ± standard error of the
mean (S.E.M) of three individual measurements performed in
triplicate (n = 3). Statistical analysis was performed using the
unpaired t-test. Significant differences were assumed at
*p < 0.05; **p < 0.01; ***p < 0.001.
3. Results
3.1. Injectability
For the reference GG samples the force to initiate extrusion
(Figure 1A) was 7.5 N, and the force to sustain the extrusion
varied from 2 to 11 N; clogging of the needle was observed.
For the GG/GENTD+E (Figure 1B), the force to initiate the
extrusion was 12 N, and to sustain the extrusion, the force
was higher but more uniform and fluctuated in the range
 616 U. POSADOWSKA ET AL.

between 10 and 16 N. For the GG/GENTD+E samples no clog-

ging was observed. The maximal force to extrude GG samples
was 10.3 ± 0.7 N, whereas it increased in the case of GG/GENTD
+E up to 15.6 ± 1.0 N.

3.2. Rheological properties

To determine elastic and viscous properties both storage (G')
and loss (G'') moduli were measured. In the first time interval
(assembling), the elastic response immediately dominated
over the viscous behavior for both GG and GG/GENTD+E sys-
tems (Figure 2A and B). For GG, G' and G'' were 10 kPa and
0.5 kPa, respectively, and for the GG/GENTD+E system, G' and
Figure 3. Drug release profiles from gentamicin-loaded nanoparticles (NPs) and
G'' were 3.0 kPa and 0.5 kPa, respectively. During breaking by the system composed of gellan gum with dissolved and encapsulated genta-
applying increasing strain, both G' and G'' dropped and from micin (GG/GENTD+E); 1 ml of NPs suspension in UHQ-water (30 mg of NPs,
containing in total 1 mg gentamicin) and 1 ml of GG/GENTD+E (containing in
circa c = 10%; G'' dominated G' (Figure 2C and D). No strain
total 2 mg gentamicin); cumulative curves were fitted by linear regression
stiffening effect as a result of strain was observed. model; a, b and R2 parameters of the linear function are displayed at the top
During the last interval (reassembling), GG/GENTD+E recov- right corner.
ery was complete, whereas for GG storage modulus after
Downloaded by [149.156.96.151] at 08:41 22 April 2016

breaking was smaller than the value reached during the

assembling stage (2 kPa instead of 10 kPa) (Figure 2A).
3.3. Drug release study
Drug release studies showed that in the first 12 h the drug was
released rapidly from NPs (initial burst release) and reached
305 ± 24 μg (intercept of NPs release equation), which corre-
sponded to ~30% of total encapsulated drug. Next, a sustained
release phase (18.7 ± 1.4 μg/day, slope of the NPs release
equation) was observed that continued up to day 35, when
the entire amount of the drug was released (Figure 3).
Similarly, burst release was also observed for GG/GENTD+E, and
533 ± 14 μg of GENT was released in the first 12 h (intercept of
GG/GENTD+E equation), which corresponded to ~27% of total
GENT in the system (both encapsulated and dissolved portions).
Afterward, GENT was released in a sustained manner at the rate
of 8.0 ± 0.3 μg/day (slope of the GG/GENTD+E release equation).
Sustained release profile was observed until day 90 and cumu-
latively ~60% of the drug was eluted (~1.2 mg) (Figure 3).
Statistical analysis showed that initial burst release of drug was

Figure 2. Rheological characteristics of gellan gum hydrogel (GG) and the system composed of gellan gum with dissolved and encapsulated gentamicin (GG/GENTD+E).
Storage modulus G’ (A) and loss modulus G’’ (B) changes during assembling (0 – 30 min, f = 1 Hz, s = 0.01%), breaking (30 – 60 min, f = 1 Hz, c = 0.01% – 100%) and
reassembling (60 – 90 min, f = 1 Hz, c = 0.01%). Storage modulus G’ (C) and loss modulus G’’ (D) changes as a function of strain during breaking interval (30 – 60 min,
f = 1 Hz, c = 0.01% – 100%).

Figure 4. Growth inhibition of Staphylococcus spp. incubated with the system composed of gellan gum with dissolved and encapsulated gentamicin (GG/GENTD+E):
reference strain of S. aureus DSM 24167 (A), clinical isolate of S. aureus SA1-KCR (B), reference strain of S. epidermidis ATCC 700296 (C), and clinical isolate of S.
epidermidis SE1-KCR (D). Microphotographs were taken on 9 cm Petri dish, with a digital camera (Olympus); on the left sector of the Petri dish the sample was
introduced inside the agar well, on the right the sample was placed on the agar. Precise values of growth inhibition zones diameters are shown in graph (E). Mean ±
S.E.M.; *p < 0.05; **p < 0.01; ***p < 0.001.
significantly higher from GG/GENTD+E than from NPs
(533 ± 14 µg vs. 305 ± 24 µg, p-value = 0.00124, **p < 0.01).
Downloaded by [149.156.96.151] at 08:41 22 April 2016
During sustained release phase up to day 90, GG/GENTD+E deliv-
ered much lower doses of the drug than NPs (8.0 ± 0.3 µg/day
vs. 18.7 ± 1.4 µg/day, p-value = 0.01802, *p < 0.05).
3.4. Antimicrobial effect
Antimicrobial tests showed the formation of inhibition zones
with the diameter between 27 and 39 mm for all used
Staphylococcus spp. strains after exposure to GG/GENTD+E
(Figure 4A–D). The diameters of inhibition zones were the
same or significantly higher as compared to 10 μg GENT
standard (Figure 4E). Some differences in the size of the
inhibition zones were observed depending on the strain and
the way of sample placement (in agar or on agar). For S. aureus
strains, inhibition zones were higher when the samples were
placed in agar; this was not the case for S. epidermidis strains
(Figure 4E). To exclude a possible intrinsic impact of gellan
gum on bacteria growth, GG samples were tested and no
inhibition zones were observed (negative control, data not
shown).
3.5. In vitro cytocompatibility
The Alamar Blue assay performed after day 1 and 6 of MG-63
cell culture (Figure 5A) revealed that metabolism of cells was
not significantly affected by the presence of GG/GENTD+E
extracts. After day 6 in culture, cell metabolic activity increased
as compared to day 1 (from 10.2 ± 2.6% to 37.8 ± 4.3%). Cell
spreading area (Figure 5B) measured after day 1 in culture did
not depend on the concentration of the extracts.
For all samples at day 1 (Figure 5C: a, e, i), cells showed
spread and spindle-shaped morphology comparable to con-
trol samples (Figure 5C: m). In both, experimental and control
conditions, the majority of cells was viable and dead cells were
not detected (Figure 5C: b, f, j, n). At day 6, a small number
(less than 2%) of dead cells were observed (Figure 5C: d, h, l,
p). Overall, both metabolic activity and live/dead assay con-
firmed high viability of cells after the exposure to the extracts
EXPERT OPINION ON DRUG DELIVERY 617
(Figure 5C: c, g, k, o), thus demonstrated no cytotoxicity of the
drug delivery system.
4. Discussion
In this study, we developed and investigated a new injectable
system for the treatment of bone infections. The system is
based on gellan gum hydrogel containing dissolved GENT,
NPs with encapsulated GENT and calcium ions that, by cross-
linking, provided structural integrity. Handling, behavior dur-
ing extrusion, mechanical stability, and drug release kinetics
were tested. Importantly, we demonstrated antimicrobial effi-
cacy of the system against pathogens (Staphylococcus spp.,
the main cause of osteomyelitis) and its cytocompatibility.
4.1. Injectability and handling
There is a strong demand for drug delivery systems intended
for site-specific action. Additionally, it is advantageous if their
administration is minimally or non-invasive, that is injection.
[23] GG/GENTD+E system was injectable and the force required
to sustain the extrusion was uniform (see Figure 1). In contrast,
forces required to sustain extrusion for GG only varied with
the extrusion time. This phenomenon may be attributed to
the fact that hydrogels exhibit spatial inhomogeneities that
result from inhomogeneous distribution of cross-links. The
inhomogeneities arise during ionotropic gelation of hydrogel
by calcium ions (the higher degree of cross-linking, the more
non-uniform the structure is).[24] Such differences in force
during injection resulted in undesired drop-wise flow accom-
panied by clogging of the syringe. Consequently, low accuracy
of dosing excludes materials with such variable force–distance
characteristics as a therapeutic injection.
GG/GENTD+E was paste-like in handling. Forces to initiate
and sustain extrusion for GG/GENTD+E were higher but more
uniform than for GG. This can be ascribed either to static
friction within the GG/GENTD+E sample or cohesiveness of
the sample and the syringe material (more sticky than slippery
character).[25] To assure easy handling, the maximal extrusion
force of the system for surgical manipulation should be in the
 618 U. POSADOWSKA ET AL.
Downloaded by [149.156.96.151] at 08:41 22 April 2016

Figure 5. Metabolic activity of MG-63 cells reflected by reduction of Alamar Blue on day 1 and day 6 (A). Spreading area of cells on day 1 cultured in pure Eagle’s
minimal essential medium EMEM (Control) or the extracts from the system composed of gellan gum with dissolved and encapsulated gentamicin (GG/GENTD+E)
diluted in EMEM by the factors of 1/1 (undiluted), 1/2 or 1/8 (B). Mean ± S.E.M.; ***p < 0.001. Interference contrast and live/dead staining of MG-63 cells on 1 day
(1st, 2nd column) or day 6 (3rd, 4th column) in contact with GG/GENTD+E extracts diluted in EMEM by the factor of 1/1 (a, b, c, d), 1/2 (e, f, g, h) or 1/8 (i, j, k, l) and
pure EMEM – Control (m, n, o, p); green are viable cells, red are dead cells, red arrows show red dead cells. Scale bar = 50 μm for interference contrast images (a, e, i,
m) and 100 μm for live/dead staining (b, c, d, f, g, h, j, k, l, n, o, p) (Olympus Carl Zeiss Axiovert, Germany).

range 5–20 N [26] and for our system extrusion force
(15.6 ± 1.0 N) fell in this range.
4.2. Mechanical properties
To characterize GG/GENTD+E before, during and after injection,
rheological tests were performed to study gelation (assembling),
breaking of the gel (disassembling e.g. by injection), and reas-
sembling of the structure. Both GG/GENTD+E and GG systems
displayed a considerably higher elastic than viscous response
indicating strong and fast gel formation. The dependence, that
is G’ ˃> G”, showed that both samples were mainly elastic and
rubbery-like (see Figure 2A and B, time 0–30 min). The elasticity
of GG originates from double-helical aggregation of hydrogel
chains that are bound by calcium ions.[27] The addition of
antibiotic molecules and NPs reduced stiffness of GG, possibly
due to the hindrance of the aggregation of double helices.
To simulate the deformations occurring when the injection is
performed, GG/GENTD+E and GG were disassembled by applying
a high-shear stress (see Figure 2A and B, time 30–60 min). As a
result, both moduli dropped and G'' exceeded G'. This indicates
that the viscoelastic structure of GG transformed into a flowable
sol. Because the G' drop was faster for GG/GENTD+E than for GG it
seems that additives reduced mechanical resistance of the gel.
Also for GG/GENTD+E the drop in G' modulus was followed by its
increase. This is a typical behavior of emulsions and resulted
from energy dissipation associated with microstructure reorga-
nization before collapsing.[28]
As soon as the strain sweep phase was finished, reassembly
of the structure of both samples occurred, G' > G'' was restored
(see Figure 2A and B, time 60–90 min). Demonstrated self-
healing behavior combined with lack of strain stiffening effect
is desirable [29] and indicates that GG/GENTD+E is a good
candidate for injectable materials.
4.3. Controlling GENT release
Despite significant improvements in surgical techniques,
nearly 5% of bone graft implantations result in infection.[30]
To treat infected tissue locally, a burst release is essential

because bacteria tend to create a biofilm that is resistant to a
moderate dose of drugs.[31] Furthermore, because microbes
tend to rebuild the biofilm, a long sustained phase is also
desirable.[32] To address these constrains, we developed a
hydrogel-based system that combined a portion of dissolved
and encapsulated GENT. For NPs only a burst release of the
drug (~30%) was observed, and it was followed by sustained
release up to 35 days (see Figure 3). Although NPs on their
own assure suitable drug release, their clinical application
seems to be controversial due to possible phagocytic clear-
ance.[33,34]
A burst release was also observed from GG/GENTD+E (~27%)
followed by a very prolonged delivery phase. We speculate
that GENT released in the initial stage originated mainly from
its dissolved part contained in the hydrogel. It was shown
previously [27] that lower burst occurred for NPs containing
sodium alendronate embedded within gellan gum and that
only some amount of the drug was released. This effect may
be explained by the fact that GENT is positively charged and it
can interact with negatively charged hydrogel chains. In other
Downloaded by [149.156.96.151] at 08:41 22 April 2016
studies, the charge balance between ionic molecules (e.g.
GENT sulphate) and the collagen hydrogel was also found to
have influence on the release kinetics.[35]
Due to the fact that the amount of released drug from 1 ml
of GG/GENTD+E (which contained 2 mg of GENT in total) was
1.2 mg, it is apparent that the drug was released cumulatively
from both dissolved and encapsulated cargos. However, the
release rate for GG/GENTD+E (~8.0 µg/day) was lower as com-
pared to NPs (~20 µg/day). Presumably, hydrolytic degrada-
tion of PLGA within the hydrogel was reduced due to lower
availability of water molecules as shown in another study.[27]
Second, GENT released from the NPs might immediately inter-
act with hydrogel chains, which could hinder the drug release
into the external environment.
4.4. Antimicrobial properties
The bactericidal activity of the GG/GENTD+E was assessed by
the measurement of the inhibition zone of Staphylococcus spp.
(S. aureus and S. epidermidis; both reference strains and clinical
isolates), which both cause osteomyelitis.[36] Often the refer-
ence strains are more prone to antibiotics whereas for clinical
isolates antibiotics susceptibility is reduced.[36] According to
European Committee on Antimicrobial Susceptibility Testing,
values of inhibition zones indicating microbial sensitivity for S.
aureus and coagulase-negative staphylococci, for example S.
epidermidis, are 18 mm and 22 mm, respectively.[37] Here, the
application of our formulations resulted in higher inhibition
zones (27–39 mm and 27–32 mm, respectively, for S. aureus
and S. epidermidis). The growth of bacteria was successfully
inhibited (see Figure 4), and the effect varied depending on
the sample placement. Diameters obtained for S. aureus
strains were higher for samples placed inside the agar wells
than for those placed directly on the agar surface, showing
that the antimicrobial effect might be enhanced by higher
contact area of GG/GENTD+E with the agar. For samples
directly placed on agar, the contact area enabling diffusion
was only the bottom part of the sample 19.6 mm2, while for
samples placed inside the agar total contact area was
EXPERT OPINION ON DRUG DELIVERY 619
31.4 mm2. Thus, for samples placed inside the agar, antimicro-
bial effect was significantly higher than for the samples placed
on the agar. Similar positive correlation between contact area
and antibacterial effect was already documented in literature
for antibacterial agents such as silver nanoparticles.[38]
4.5. Cytocompatibility of injectable system
In vitro studies did not reveal any differences in cell morphol-
ogy, growth, and metabolic activity when MG-63 cells were
cultured in the extracts or under control conditions (see
Figure 5). This indicated that the GG/GENTD+E system was
not cytotoxic.
5. Conclusion
Based on the conducted studies it was concluded that the
combination of drug, drug-loaded NPs, gellan gum and cross-
linking agent resulted in injectable, easy-to-use paste-like
material that displayed self-healing behavior without strain-
stiffening effect. The system ensured low-invasive administra-
tion and local delivery of initial high doses of GENT followed
by its uniform and sustained delivery. The system showed
antimicrobial activity against osteomyelitis-causing pathogens
and did not affect bone forming cells. Developed drug deliv-
ery hydrogel-based system is promising for localized treat-
ment of bone diseases caused by microorganisms.
Declaration of interest
The study was financially supported by the Polish National Science Centre
(Grant no: 2012/05/B/ST8/00129). The authors have no other relevant
affiliations or financial involvement with any organization or entity with
a financial interest in or financial conflict with the subject matter or
materials discussed in the manuscript apart from those disclosed.
References
Papers of special note have been highlighted as: • of interest •• of con-
siderable interest
1. Calhoun J, Manring MM, Shirtliff M. Osteomyelitis of the long bones.
Semin Plast Surg. 2009;23(2):59–72.
2. Griffis CD, Metcalfe S, Bowling FL, et al. The use of gentamicin-
impregnated foam in the management of diabetic foot infections:
a promising delivery system? Expert Opin Drug Deliv. 2009;6(6):693–
42.
•• This paper presented a novel system for the delivery of genta-
micin that was biocompatible and does not require removal.
These features make the gentamicin-collagen implant a superior
drug delivery system.
3. Nnamani PO, Kenechukwu FC, Dibua EU, et al. Transdermal micro-
gels of gentamicin. Eur J Pharm Biopharm. 2013;84(2):345–354.
4. Virtoa MR, Frutosa P, Torradoa S, et al. Gentamicin release from
modified acrylic bone cements with lactose and hydroxypropyl-
methylcellulose. Biomaterials. 2003;24(1):79–87.
5. Nandi SK, Mukherjee P, Roy S, et al. Local antibiotic delivery systems
for the treatment of osteomyelitis – A review. Mater Sci Eng C. 2009;29
(8):2478–2485.
6. Hatefi A, Amsden B. Biodegradable injectable in situ forming drug
delivery systems. J Contr Release. 2002;80(1–3):9–28.
• This is a comprehensive review paper which critically describes
various strategies to prepare in situ setting drug delivery sys-
tems, and presents their advantages and disadvantages as loca-
lized drug delivery systems.
 620 U. POSADOWSKA ET AL.
7. Anderson JM, Shive MS. Biodegradation and biocompatibility of PLA
and PLGA microspheres. Adv Drug Deliv Rev. 1997;28(1):5–24.
8. Pamula E, Menaszek E. In vitro and in vivo degradation of poly(L:
-lactide-co-glycolide) films and scaffolds. J Mater Sci Mater Med.
2008;19(5):2063–2070.
9. Pamula E, Filová E, Bacáková L, et al. Resorbable polymeric scaffolds for
bone tissue engineering: the influence of their microstructure on the
growth of human osteoblast-like MG 63 cells. J Biomed Mater Res A.
2008;89(2):432–443.
10. Coviello T, Matricardi P, Marianecci C, et al. Polysaccharide hydrogels
for modified release formulations. J Contr Release. 2007;119(1):5–24.
11. Shu XZ, Ahmad S, Liu Y, et al. Synthesis and evaluation of injectable,
in situ crosslinkable synthetic extracellular matrices for tissue engi-
neering. J Biomed Mater Res A. 2006;79(4):902–912.
12. Nguyen KT, West JL. Photopolymerizable hydrogels for tissue engi-
neering applications. Biomaterials. 2002;23(22):4307–4314.
13. Kolambkar Y, Dupont K, Boerckel J, et al. An alginate-based hybrid
system for growth factor delivery in the functional repair of large
bone defects. Biomaterials. 2011;32(1):65–74.
14. Fernández-Ferreiro A, González Barcia M, Gil-Martínez M, et al. In
vitro and in vivo ocular safety and eye surface permanence determi-
nation by direct and magnetic resonance imaging of ion-sensitive
hydrogels based on gellan gum and kappa-carrageenan. Eur J Pharm
Biopharm. 2015;94:342–351.
Downloaded by [149.156.96.151] at 08:41 22 April 2016
15. Babu RJ, Sathigari S, Kumar MT, et al. Formulation of controlled
release gellan gum macro beads of amoxicillin. Curr Drug Deliv.
2010;7(1):36–43.
•• This paper explored and confirmed the applicability of gellan
gum macro beads for the gastroretentive-controlled delivery of
a sparingly soluble drug, amoxicillin.
16. Miyazaki S, Aoyama H, Kawasaki N, et al. In-situ gelling gellan
formulation for oral drug delivery. J Contr Release. 1999;60(2–3):
287–295.
• One of the first papers that demonstrated the in situ gelation of
the gellan gum in the stomach. The gelation was caused by the
release of calcium ions which were contained in the formulation
and release at low pH (acidic environment of stomach).
Furthermore, it was demonstrated that the bioavailability of
theophylline that was encased in the gels and then released
from gellan gels due to the gelation in the animal stomach was
increased by four–five-fold in rats and three-fold in rabbits
compared with that from the commercial oral formulation.
17. Oliveira JT, Santos TC, Martins L, et al. Gellan gum injectable hydrogels
for cartilage tissue engineering applications: in vitro studies and pre-
liminary in vivo evaluations. Tissue Eng Part A. 2010;16(1):343–353.
• This paper demonstrated the suitability of gellan gum hydrogels
for noninvasive injectable formulation that support the func-
tional cartilage tissue regeneration.
18. Douglas TE, Wlodarczyk M, Pamula E, et al. Enzymatic mineraliza-
tion of gellan gum hydrogel for bone tissue-engineering applica-
tions and its enhancement by polydopamine. J Tissue Eng Regen
Med. 2014;8(11):906–918.
19. Posadowska U, Brzychczy-Włoch M, Pamuła E. Gentamicin loaded
PLGA nanoparticles as local drug delivery system for the osteomye-
litis treatment. Acta Bioeng Biomech. 2014;17(3):41–48.
20. Dobrzynski P, Kasperczyk J, Janeczek H, et al. Synthesis of biode-
gradable copolymers with the use of low toxic zirconium com-
pounds. 1. Copolymerization of glycolide with L-Lactide initiated
by Zr(Acac)4. Macromolecules. 2001;34(15):5090–5098.
21. Anhalt JP. Assay of gentamicin in serum by high-pressure liquid
chromatography. Antimicrob Agents Ch. 1977;11(4):651–655.
22. Ghadiri M, Chrzanowski W, Lee WH, et al. Alginate/Clay
(Laponite®) composite with enhanced properties for biomedical
applications. Appl Clay Sci. 2013;85:64–73.
23. Wang F, Li Z, Khan M, et al. Injectable, rapid gelling and highly
flexible hydrogel composites as growth factor and cell carriers.
Acta Biomater. 2010;6(6):1978–1991.
24. Gerlach G, Arndt K-F, editors. Hydrogel sensors and actuators.
Springer series on chemical sensors and biosensors 6. Berlin:
Springer-Verlag; 2009. doi:10.1007/978-3-540-75645-3_1
25. Duan Y, Cai X, Du H, et al. Novel in situ gel systems based on P123/
TPGS mixed micelles and gellan gum for ophthalmic delivery of
curcumin. Colloids Surf B, Biointerfaces. 2015;128:322–330.
26. PN EN ISO 7886-1 Sterile hypodermic syringes for single use – Part 1:
Syringes for manual use [Internet]. (ISO 7886-1:1993, including
Technical Corrigendum 1:1995). [cited 2015 Oct 20]. Available from:
http://www.iso.org/iso/iso_catalogue/catalogue_tc/catalogue_detail.
htm?csnumber=14835
27. Posadowska U, Parizek M, Filova E, et al. Injectable nanoparticle-
loaded gellan gum-based system for the local delivery of sodium
alendronate. Int J Pharm. 2015;485(1–2):31–40.
•• This paper demonstrated for the first time the applicability of
gellan gum to formulate injectable system for the delivery of
bisphosphonates (sodium alendronate) for the treatment of
osteoporosis.
28. Garcia MC, Alfaro MC, Calero N, et al. Influence of gellan gum concentra-
tion on the dynamic viscoelasticity and transient flow of fluid gels.
Biochem Eng J. 2011;55(2):73–81.
29. Włodarczyk-Biegun MK, Werten MW, De Wolf FA, et al. Genetically
engineered silk–collagen-like copolymer for biomedical applications:
Production, characterization and evaluation of cellular response.
Acta Biomater. 2014;10(8):3620–3629.
30. Clauss M, Trampuz A, Borens O, et al. Biofilm formation on bone
grafts and bone graft substitutes: Comparison of different materials
by a standard in vitro tests and micrcalorimetry. Acta Biomater.
2010;6(9):3791–3797.
31. Schaer TP, Stewart S, Hsu BB, et al. Hydrophobic polycationic coatings
that inhibit biofilms and support bone healing during infection.
Biomaterials. 2012;33(5):1245–1254.
• This paper presents an effective approach to functionalize
orthopedic implants with hydrophobic polycations which pre-
vent biofilm formation on implant surface and promote bone
healing.
32. Tafin-Kampe K, Kamus-Fogumen B. Acute osteomyelitis due to
Staphylococcus aureus in children: What is the status of treatment
today? Pediatr Infect Dis J. 2013;5(3):122–126.
33. Inchaurraga L, Martín-Arbella N, Zabaleta V, et al. In vivo study of the
mucus-permeating properties of PEG-coated nanoparticles following
oral administration. Eur J Pharm Biopharm. doi:10.1016/j.ejpb.2014.
12.021.
34. Karmali PP, Simberg D. Interactions of nanoparticles with plasma
proteins: implication on clearance and toxicity of drug delivery
systems. Expert Opin Drug Deliv. 2011;8(3):343–357.
35. Singh P, Stefko J, Lumpkin JA, et al. The effect of electrostatic charge
interactions on release rates of gentamicin from collagen matrices.
Pharm Res. 1995;12(8):1205–1210.
36. Uskoković V, Desai TA. Nanoparticulate drug delivery platforms for
advancing bone infection therapies. Expert Opin Drug Deliv. 2014;11
(12):1899–1912.
•• This is very important review which very comprehensively pre-
sents discrepancies between in vitro and in vivo bioactivity
assessments and highlights the inherent imperfectness of the
former. It also details challenges in the design of ‘future’ nano-
particulate therapeutic carriers which should be multifunctional,
enable synergies and be capable of simultaneously targeting,
monitoring, and treating internal organismic disturbances.
37. European Committee on Antimicrobial Susceptibility Testing.
Breakpoint tables for interpretation of MICs and zone diameters.
Version 4.0. 2014. [cited 11 August 2015]. Available from: http://
www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/
Breakpoint_tables/Breakpoint_table_v_4.0.pdf.
38. Zarei M, Jamnejad A, Khajehali E. Antibacterial effect of silver nano-
particles against four foodborne pathogens. Jundishapur J Microbiol.
2014;7(1):e8720. doi:doi.10.5812/jjm.8720.