Simultaneous Immobilization of Bioactives During 3D Powder Printing of Bioceramic Drug‐Release Matrices

Advanced Functional Materials (Impact Factor: 11.81). 05/2010; 20(10):1585 - 1591. DOI: 10.1002/adfm.200901759
ABSTRACT
The combination of a degradable bioceramic scaffold and a drug-delivery system in a single low temperature fabrication step is attractive for the reconstruction of bone defects. The production of calcium phosphate scaffolds by a multijet 3D printing system enables localized deposition of biologically active drugs and proteins with a spatial resolution of approximately 300 µm. In addition, homogeneous or localized polymer incorporation during printing with HPMC or chitosan hydrochloride allows the drug release kinetics to be retarded from first to zero order over a period of 3–4 days with release rates in the range 0.68%–0.96% h−1. The reduction in biological activity of vancomycin, heparin, and rhBMP-2 following spraying through the ink jet nozzles is between 1% and 18%. For vancomycin, a further loss of biological activity following incorporation into a cement and subsequent in vitro release is 11%. While previously acknowledged as theoretically feasible, is its shown for the first time that bone grafts with simultaneous geometry, localized organic bioactive loading, and localized diffusion control are a physical reality. This breakthrough offers a new future for patients by providing the required material function to match patient bone health status, site of repair, and age.

Full-text

Available from: Elke Vorndran
Simultaneous Immobilization of Bioactives During 3D
Powder Printing of Bioceramic Drug-Release Matrices
By Elke Vorndran, Uwe Klammert, Andrea Ewald, Jake E. Barralet, and
Uwe Gbureck*
1. Introduction
Rapid prototyping enables the layer-by-layer creation of near net
shape 3D objects in a variety of materials. The general underlying
principle is spatial control of powder bonding by methods such as
localized laser sintering and liquid solvent or binder application to
form a ‘‘green body’’ prior to sintering. Recently, custom made
scaffolds for bone repair and tissue engineering have been created
by selective laser sintering
[1]
or slip cast-
ing,
[2]
and in the production of porous
polyethylene implants for craniofacial
recontouring.
[3]
The performance of porous
scaffolds made by these approaches has
recently been reviewed.
[4]
Many of these
procedures involve a heat treatment step
such that the incorporation of organic,
biologically active, or hydrated molecules
within the bulkof the implant is impossible.
Recently, we reported the development of a
rapid 3D printing technique that utilized a
calcium phosphate cement setting reaction
to create bioceramic implants at ambient
temperature.
[5–7]
This one nozzle system
could precisely control liquid deposition
such that hitherto impossible geometries
could be fabricated with the formation of
precisely defined channels for controlled
tissue growth.
[8]
Biomaterials with precise composition
and bioactive concentration gradients can
help exploit cell chemotaxis and haptotaxis,
attachment, and differentiation to optimize
healing.
[9–13]
Localized delivery of therapeutic substances
can reduce the dose required to achieve a biological response
compared with systemic delivery. In this way, both the risk of
side effects
[14]
and cost of treatment can be significantly reduced.
By using a multijet 3D rapid prototyping machine we report
the first low temperature synthesis of bioceramic implants
while simultaneously depositing bioactive compounds with high
spatial accuracy for localized delivery. Three-dimensional macro-
porous architectures and discrete drug modification of the
implants was achieved following the strategy summarized
in Figure 1.
Initially we sought to determine the compatibility and accuracy
of the multijet printing system with bioactives; recombinant bone
morphogenic protein 2 (rhBMP-2), heparin (a model polysacchar-
ide), and vancomycin (an antibiotic glycopeptide) were evaluated.
Secondly, the activity of the bioactives was determined following
immobilization onto the calcium phosphate implant during the
setting reaction. In addition, the effect of bioactive localization
within the implant structure and polymer modification to the
ceramic on bioactive release kinetics was investigated.
Biocompatible polymers were incorporated by either using a
blend of hydroxypropylmethylcellulose (HPMC) and tricalcium
phosphate (TCP) powders to result in a homogeneous polymer
FULL PAPER
www.MaterialsViews.com
www.afm-journal.de
[*] Dr. U. Gbureck, E. Vorndran, Dr. A. Ewald
Department for Functional Materials in Medicine and Dentistry
University of Wu
¨
rzburg, 97070 Wu
¨
rzburg (Germany)
E-mail: uwe.gbureck@fmz.uni-wuerzburg.de
Dr. U. Klammert
Department of Cranio-Maxillo-Facial Surgery
University of Wu
¨
rzburg, 97070 Wu
¨
rzburg (Germany)
Dr. J. E. Barralet
Faculty of Dentistry, McGill University
Strathcona Anatomy & Dentistry Building
Montreal, Quebec H3A 2B2 (Canada)
DOI: 10.1002/adfm.200901759
The combination of a degradable bioceramic scaffold and a drug-delivery
system in a single low temperature fabrication step is attractive for the
reconstruction of bone defects. The production of calcium phosphate
scaffolds by a multijet 3D printing system enables localized deposition of
biologically active drugs and proteins with a spatial resolution of
approximately 300 mm. In addition, homogeneous or localized polymer
incorporation during printing with HPMC or chitosan hydrochloride allows
the drug rele ase kinetics to be retarded from first to zero order over a period of
3–4 days with release rates in the range 0.68%–0.96% h
1
. The reduction in
biological activity of vancomycin, heparin, and rhBMP-2 following spraying
through the ink jet nozzles is between 1% and 18%. For vancomycin, a further
loss of biological activity following incorporation into a cement and
subsequent in vitro release is 11%. While previously acknowledged as
theoretically feasible, is its shown for the first time that bone grafts with
simultaneous geometry, localized organic bioactive loading, and localized
diffusion control are a physical reality. This breakthrough offers a new future
for patients by providing the required material function to match patient bone
health status, site of repair, and age.
Adv. Funct. Mater. 2010,20,1585–1591
ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1585
Page 1
FULL PAPER
distribution (types I–III), or by simultaneously printing chitosan
and crosslinker solutions during implant fabrication and drug
deposition for a surface localized polymer modification (types IV).
2. Results
The fabrication of brushite samples by 3D
powder printing is based on the hydraulic
setting reaction of TCP powder and a phos-
phoric acid binder solution according to
Equation (1):
Ca
3
ðPO
4
Þ
2
þ H
3
PO
4
þ 6H
2
O
! 3CaHPO
4
2H
2
O (1)
Printed samples consisted predominately of
the crystalline phase brushite (CaHPO
4
2H
2
O,
51%) and contained a smaller proportion of
monetite (CaHPO
4
, 12%) and unreacted a/b-
TCP (Ca
3
(PO
4
)
2
, 37%) with a total porosity of 45.5% and a median
pore size of 27 mm as previously established.
[5,8]
The deviation of
the printed structure dimension was less than 5% and we
observed that this resulted from the liquidbinder bleeding through
the porous calcium phosphate (CaP) matrix slightly. This bleeding
presented a problem for the defined deposition of bioactives. The
printable accuracy of liquid additives was compared by printing a
parallel series of fine lines (200–1000 mm) with colored solutions
on polymer-modified and unmodified brushite samples. Polymer
modification was either performed homogeneously by mixing
TCP and HPMC powders, or locally by printing chitosan
simultaneously with the colored lines on both TCP and TCP that
contained HPMC (Fig. 2). Pure brushite samples showed the
highest dimensional deviation (Fig. 2a) with a line width of
1523 130 mm compared with the intended 200 mm (Fig. 2b-I),
followed by chitosan-modified brushite (Fig. 2b-II) with a line
width of 1238 133 mm, and HPMC/brushite with a line width of
440 46 mm (Fig. 2b-III), and the best results were achieved with
HPMC and chitosan-modified brushite that showed a line width of
272 19 mm (Fig. 2b-IV). Lines of 350 mm thickness or more were
faithfully reproduced for the latter material combination.
The in vitro release kinetics of spherical (11 mm diameter)
polymer-modified and pure brushite samples loaded with
vancomycin according to loading strategies Type I–IV was
determined for up to 100 h (Fig. 3). Table 1 shows the fitting
parameters of the experimental data with the Korsemeyer–Peppas
Equation (2):
[15]
M
t
=M
1
¼ k t
n
(2)
where M
t
/M
1
is the cumulative amount of released drug [%],
k [% s
1
] is the release constant, and n is the release exponent that
indicates the release process. In the case of spherical samples,
n 0.43 indicates Fickian diffusion processes, while
0.43 < n < 0.85 characterizes a combination of diffusion and
swelling controlled (anomalous) transport.
[16]
Drug release of
Type I homogeneously loaded samples (Fig. 3a,d,f) agreed well
with the Korsemeyer–Peppas model, which indicates a diffusion
controlled process (n ¼ 0.31–0.45) with an initial burst and a total
release within 7 days for pure brushite samples ( n ¼ 0.33; Fig. 3d)
and 9 days for chitosan-modified brushite samples (n ¼ 0.31;
Fig. 3f). The release from HPMC-modified brushite samples
www.afm-journal.de
www.MaterialsViews.com
Figure 1. The multijet 3D printing strategy of brushite-based bioceramic
implants with controlled bioactive loading. Bioactives were either dispersed
homogeneously through the ceramic (Type I), concentrated in the center
(Type II and IV) or dispersed along a concentration gradient from a high
concentration in the center to low at the exterior (Type III). Polymer
modification was achieved either by mixing tricalcium phosphate (TCP)
powder with hydroxypropylmethylcellulose (HPMC) or by printing chitosan
polymer solutions directly onto TCP powder, either homogeneously or
localized to the surface to create a polymer barrier (Type IV). Colors
represent: orange-yellow: bioactive (color intensity indicates drug concen-
tration), blue: polymer barrier.
Figure 2. a) Degree of bleeding of colored lines of different thicknesses for polymer-modified
and pure brushite cuboids (n ¼ 6). b) Printed samples with colored lines: I: pure brushite,
II: brushite with co-printed chitosan, III: brushite with homogeneously distributed HPMC, and
IV: brushite with both chitosan and HPMC modifications.
1586 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010,20,1585–1591
Page 2
FULL PAPER
(Fig. 3a) showed anomalous transport (n ¼ 0.45) and was
completed within 7 days. Pure brushite samples with Type II
depot loading (Fig. 3e) also followed the Korsemeyer–Peppas
equation (n ¼ 0.38), however, polymer-modified structures
showed a zero-order release with a constant release rate (a)of
0.96% h
1
(Fig. 3b) for up to 3–4 days. Inclusion of a polymer
barrier (Type IV loading) decreased the release rate to 0.84% h
1
(Fig. 3g); zero-order release over a period of 4 days was also be
achieved by Type III graduated loading of HPMC/brushite
samples with a constant release rate of 0.68% h
1
(Fig. 3c). By
graduating the deposition of vancomycin in brushite/HPMC, the
total release time was extended from 7 to 11 days (Fig. 3a and c).
The deviation between experimental data and fitting curves
resulted from the limited area of validity (60% cumulative release)
of the Korsemeyer–Peppas model.
Physical and chemical stresses imposed on bioactives during
printing include high shear, the potential for partial thermal
evaporation during purging through the print-head, as well as
contact with the phosphoric acid binder during hardening of the
structures (Fig. 4a). However, bulk drug activity was found to be
retained after the purging of liquids through the print head, which
demonstrated good resilience to this process (Fig. 4b). In
comparison to the control solutions, the biological activity of the
purged solutions were reduced to 82% 1% for heparin,
91% 18% for rhBMP-2, and 99% 7% for vancomycin.
Furthermore, the vancomycin activity after printing and in vitro
www.MaterialsViews.com
www.afm-journal.de
Figure 3. Cumulative release of vancomycin from polymer-modified and brushite spherical samples depending on polymer modification and drug loading.
Type I homogeneously loaded samples (a: brushite/HPMC, d: pure brushite, f: brushite/chitosan) and Type II depot loaded brushite structures (e) followed
the Korsemeyer–Peppas equation, while Type II depot loaded brushite/HPMC (b), Type VI depot-chitosan barrier (g), and Type III graded brushite/HPMC
structures (c) showed a linear release.
Table 1. Fitting parameters of the experimental data of Figure 3 with the Korsemeyer–Peppas model M
t
/M
1
¼ k t
n
or linear fit M
t
/M
1
¼ a t.
Model Brushite/polymer Vancomycin distribution type na
Korsmeyer–Peppas (first-order release) Brushite Homogeneous I 0.33 0.03
Brushite Depot II 0.38 0.05
Brushite/cellulose Homogeneous I 0.45 0.01
Brushite/chitosan Homogeneous I 0.31 0.02
Linear fit (zero-order release) Brushite/cellulose Depot II 0.96 0.04
Brushite/cellulose Graded III 0.68 0.02
Brushite/chitosan barrier Depot IV 0.84 0.02
Adv. Funct. Mater. 2010,20,1585–1591 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1587
Page 3
FULL PAPER
release from homogeneous-loaded scaffolds was quantified. The
eluate of vancomycin obtained after performing the whole printing
process followed by in vitro release showed an activity of 87% 9%
(Fig. 4b).
3. Discussion
The ever growing and inextricable link between medicine and
modern technology has had far reaching consequences for
biomaterials science, which traditionally revolved
around a microstructure–property–cost balance
and now is increasingly directed by advances in
the knowledge of cell–protein–material surface
interactions. Regenerative products based on
tissue induction prompted by protein release
from polymeric matrices are now a reality in the
field of bone and periodontalsurgery.
[16–18]
Three-
dimensional printing combined with calcium
phosphate cement chemistry enables the pre-
paration of low temperature, microporous
calcium phosphate bioceramics, which can be
adapted to irregular bone defects using com-
puted tomography (CT) scan data from a
patient
[19]
and can now also release biologically
active substances in a controlled manner.
3.1. Polymer Modification
The theoretically achievable resolution of the
printer used 600 540 dpi (spot sizes:
42.33 mm 47.04 mm), however, this degree
of resolution was not attained when printing
liquid bioactives on pure brushite samples
because the bioactive solution diffused within
the TCP powder. By mixing TCP powder with
HPMC or printing chitosan hydrochloride,
bioactive diffusion through the cement powder
was greatly reduced as a result of polymer
swelling, and drug localization was retained.
Powders used for 3D printing must fulfill two
crucial criteria, firstly they must allow the
formation of thin powder layers 100–200 mm
in thickness with a smooth surface (to obtain
high printing quality) and secondly they must
harden with the binder solution or not interfere
with the setting reaction of TCP with acid during
printing. The first criterion is associated with
the particle size distribution of the powder, it has
been demonstrated by our group that ideal
particle sizes are in the range of 20–50 mm with
the absence of small particle fractions <5 mm.
[5]
The limiting factors for applicable polymeric
printing solutions are the compatibility of the
solution to the print-head and purging process
as well as a fast gelling reaction. In order to
purge a solution through thin channels
(20 mm) of the print heads, a low viscosity (< 5 mPa s, approx.
1 wt% chitosan hydrochloride) is crucial. To act as an efficient
diffusion inhibitor the gelling reaction has to take place within
seconds to avoid diffusion of bioactive substances in the structures
as well as to develop a water-soluble gel. Low viscosity chitosan
hydrochloride solutions fulfill this role since they undergo a fast
gelling reaction by the reaction of negatively charged polymers and
polyanions in contact with an aqueous environment (Fig. 5).
[20–22]
By modifying printed brushite samples with HPMC and/or
chitosan hydrochloride, the liquid printing resolution was
www.afm-journal.de
www.MaterialsViews.com
Figure 4. a) Physical and chemical stresses that have the potential to deteriorate the biological
activity of additives from the fabrication process of brushite samples to biological testing.
b) Biological activities of heparin, rhBMP-2, and vancomycin after the printing process and after
release of brushite sample for vancomycin. The biological activity of heparin was defined by
activated partial thromboplastin times (aPTT) [s]: standard human plasma: 37.67 1.45, control:
70.27 2.61, and printed solution: 57.57 0.47. The biological activity of rhBMP-2 was defined
by ALP activity expressed as 10
6
M of p-nitrophenol produced per minute per well: control:
15.53 2.60 and printed solution: 14.11 2.56. The biological activity of vancomycin was
defined by the diameter of the bacterial inhibition zone [mm]: control solution: 26.20 0.84,
printed solution: 26.00 1.87, and eluate: 21.33 2.08.
1588 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010,20,1585–1591
Page 4
FULL PAPER
improved as demonstrated by the deposition of toluidine blue
solution in defined line thicknesses (Fig.2a and b) such that lines of
350 mm thickness or more were faithfully reproduced. The highest
resolution was obtained with a HPMC-modified TCP powder in
addition to chitosan hydrochloride printing. Only a marginally
lower resolution was obtained by HPMC modification alone,
which indicated that the resolution was mainly improved by the
high water uptake of the HPMC powder. Modification of brushite
samples with chitosan hydrochloride alone resulted in only a
marginally higher resolution than pure brushite because of the
limited water uptake ability of the locally applied gel.
3.2. Dependency of Release Profile and Drug/Polymer
Modification
By discrete bioactive localization and printing of graded drug
distributions in the structures it was possible to control the release
kinetics. Investigations of the release kinetics included pure and
polymer-modified brushite spheres with a Type I homogeneous,
Type II and IV depot, or Type III graded vancomycin distribution
(Fig. 1). Unmodified brushite samples with different bioactive
loading strategies showed no considerable differences in release
kinetics (Fig. 3d and e) because of unhindered drug diffusion
within the structures during the printing process. Modification of
brushite with HPMC led to a delayed drug release, whereas the
release kinetics were strongly dependant on the drug localization
(Fig.3a–c). Release fromType I-loaded brushite/HPMC (Fig. 3a) or
brushite/chitosan (Fig. 3f) structures showed first-order kinetics.
In comparison to Type I-loaded pure brushite the release is
retarded as a result of polymer swelling that forms a diffusion
barrier. This is also indicated by the release exponent n, which
characterizes the release as a diffusion and swelling controlled
process (Table 1). Zero-order release was achieved from Type II and
III loaded HPMC/brushite structures (Fig. 3b and c). According to
Ficks first law, drug diffusion depends on both the concentration
gradient and matrix properties. In the case of Type II loading
(Fig. 3b), diffusion is limited by polymer swelling, and since drug
diffusion through a drug free polymer barrier occurs before
release, the drug concentration gradient is decreased in
comparison to Type I-loaded scaffolds. In contrast, graded drug
release depends predominantly on the creation of a drug
concentration distribution tapered towards the periphery (Type III),
which reduces the concentration gradient and hence the release
rate. A zero-order release was also obtained by creating a diffusion
barrier with chitosan in depot-loaded brushite structures (Type IV,
Fig.3g). The variation of drug amount and local drug concentration
in homogeneously polymer-modified structures by graded or
depot drug modification or a local polymer modification by
printing polymer solution was beneficial to adjust the release rate
as well as the total release time. Furthermore, simultaneous
modification of bioceramic scaffolds with different drugs that
induce, for example, specific biological reactions in different
regions of the implant would be possible. Stimulation of
osteointegration would be feasible by modifying the implant
surface with osteoinductive additives such as bone morphogenic
protein (BMP)
[23–25]
and promoting vascularization in scaffolds by
loading with angiogenic factors such as vascular endothelial
growth factor (VEGF)
[26]
or copper ions.
[27,28]
3.3. Biological Activity of Printed Drugs
To remain biologically active, printed drugs must be resilient to the
printing process (Fig. 4a). The slight loss of biological activity
observed in printed bioactives is likely a result of the heat developed
during (thermal) purging through the narrow print heads,
whereby a small fraction of the printing liquid is evaporated.
Encouragingly, the bulk activity (>82%) of drug solutions was
retained during the printing process even for heat-sensitive
proteins like rhBMP-2 (Fig. 4b). Of course, the printing process
also includes further steps such as acid treatment for the setting
reaction, which may also lower the biological activity of rhBMP-2 or
heparin and have to be tested in further studies. The activity of
vancomycin was only reduced by a further 12% following the
complete printing process, which indicates this effect may only be
minor.
4. Conclusions
Polymer modification of brushite samples reduced the diffusion of
drug solutions within the structures during the printing process
and enabled faithful reproduction of micrometer-scale features.
Consequently release kinetics can be controlled by locally
modifying drug loading within the structures by 3D printing,
which results in a zero-order release with release rates in the range
of 0.68%–0.96% h
1
. Bioceramic scaffolds produced by multijet
3D printing now allow the discrete deposition of pharmaceutical
agents. Using these strategies it should now be possible to induce
localized biological reactions to enhance healing since there was
only a marginal loss in biological activity of the bioactive additive
following printing.
5. Experimental
Tricalcium phosphate (TCP) was synthesized by heating a 2:1 molar
mixture of anhydrous dicalcium phosphate (DCPA, CaHPO
4
, monetite)
and calcium carbonate (both Merck, Darmstadt, Germany) followed by
quenching to room temperature. The sintered cake was crushed with a
pestle and mortar, passed through a 160 mm sieve and ground for 10 min in
a ball mill.
www.MaterialsViews.com
www.afm-journal.de
Figure 5. Scheme of ionotropic crosslinking between chitosan and tripoly-
phosphate.
Adv. Funct. Mater. 2010,20,1585–1591 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1589
Page 5
FULL PAPER
Printing of cement samples was performed with a multicolor 3D-
powder printing system (spectrum Z510, Z-Corporation, Burlington, USA)
using TCP powder (polymer-modified or unmodified) and a binder solution
of 20% phosphoric acid (Merck, Darmstadt, Germany). For the printing
parameters, a thickness of 125 mm per layer was chosen and a binder/
volume ratio of 0.371. Polymer modification of printed samples was
performed either by blending TCP powder with a) 5 wt% of HPMC (Fluka,
Buchs, Switzerland) or b) by printing with pure TCP powder and applying a
polymer solution of 1 wt% chitosan hydrochloride (Kreaber GmbH & Co,
Ellerbek, Germany) in water that was crosslinked with 0.5 wt%
trisodiumpolyphosphate (TPP, Sigma–Aldrich, Steinheim, Germany)
through two different print heads.
The binder solution was stored in the first liquid reservoir that is linked
to the inkjet print head 0 (white) and was always involved to the printing
process at a certain percentage rate. For the additives three color reservoirs
(yellow, magenta, cyan) are available. The drug solutions (5% vancomycin
hydrochloride in distilled water (Abbott, Wiesbaden, Germany), 0.1%
toluidine blue in water, 5 10
6
M rhBMP-2 (Department of Physiological
Chemistry II, Biocenter, University of Wuerzburg, Germany) [29] or
10 IU mL
1
of heparin (Meduna GmbH, Isernhagen, Germany)) were
applied through print head 1 (yellow), while chitosan hydrochloride
solution was printed with print head 3 (cyan) and TPP was printed with
print head 2 (magenta). Structure geometries were designed using
Think3design (thinkiD Design Xpressions for Windows 2000/XP/Vista
Version 2007.1.106.49 SP2, think3 Inc., USA) software. To demonstrate the
achievable resolution a cuboid (h ¼ 2 mm, l ¼ 20 mm) with 5 colored lines
of different width (1000, 750, 500, 350, and 200 mm) were printed with
HPMC-modified and pure TCP powder.
The release kinetics of vancomycin were studied on polymer-modified or
pure brushite spherical samples (r ¼ 5.56 mm), whereas the structures
were modified with vancomycin according to Figure 1: homogeneous,
depot, or graded. For graded modification, the amount of printed drug
solution was tapered towards the periphery. The amount of drug in the
printed samples was determined by dissolving the loaded samples in 2.4
M
HCl following absorbance measurements against standard solutions at
237 nm using an UV/vis spectrometer (Cary 13, Varian, Australia). The drug
desorption kinetics were measured after immersion of the loaded samples
(n ¼ 3) in double distilled H
2
O (10 mL) at 37 8C for up to 18 days and
changing the immersion liquid after 1 h, the first 5 h, and after 1, 2, 3, and
4 days, and then every third day up to 18 days. The possible influence of
interfering polymer compounds was accounted for by using unloaded
polymer-modified samples as reference. Adsorption and desorption
profiles were modelled using the Korsmeyer–Peppas equation or by linear
fitting as pharmaceutical release models.
The biological activity after the thermal purging process through the
print-head was analyzed for vancomycin, heparin, and rhBMP-2.
Furthermore, the biological activity of vancomycin after the whole printing
process and in vitro release was analyzed. Therefore, vancomycin-loaded
cylindrical (h ¼ 5 mm, r ¼ 4 mm) scaffolds (2.5 mg sample
1
) were
submerged and eluted in static conditions for 2 h in H
2
O (1 mL) in a
15 mL centrifuge tube. The eluate was exposed to gram positive
Staphylococcus epidermidis (strain RP62A) in an agar diffusion assay.
Agar (2%) was suspended in LB medium (1% w/v tryptone, 0.5% w/v yeast
extract, 0.5% w/v NaCl) and autoclaved (120 8C, 20 min). The hot agar
solution was poured into sterile Petri dishes. S. epidermidis from an over
night culture (100 mL) was plated onto the solidified agar by the use of
sterile glass spheres. A freshly prepared 5% (4 mL) and a 0.1% (50 mL)
vancomycin solution as well as a 5% (4 mL) vancomycin solution purged
through the print head and the eluate (50 mL) were placed onto a filter piece
(1 cm in diameter). This was placed on top of solidified agar on which
S. epidermidis had been plated. Each test sample was prepared five-fold. The
Petri dishes were incubated at 37 8C for 48 h. After this time the diameter of
the zone of inhibition was determined by using a sliding rule. The average
and standard deviations were calculated. Statistical analysis was performed
by using the Anova t-test.
The anti-coagulation activity of heparin was analyzed after printing a
10 IU mL
1
heparin solution and purging through a print head. The original
10 IU mL
1
heparin solution (5 mL) as well as the purged heparin solution
(5 mL) were mixed with standard human plasma (250 mL; Siemens,
Marburg, Germany) and the activated partial thromboplastin time (aPTT)
was measured with a coagulation analyzer (Sysmex CA-540, Siemens,
Marburg, Germany), n ¼ 3.
The biological activity of the purged 5 10
6
M rhBMP-2 solution was
determined by alkaline phosphatase (ALP) activity. Therefore, cells of the
mouse myoblast cell line C2C12 [30] were maintained in growth medium
consisting of Dulbecco’s modified eagle medium (DMEM) supplemented
with 10% fetal calf serum (FCS) at 37 8C in a humidified atmosphere of 5%
CO
2
. Cells were seeded into 96-well microtiter plates at a concentration of
5 10
4
cells per well (to survey the rhBMP-2 solutions) and cultured with
growth medium (200 mL). After 48 h, the growth medium was removed and
replaced with equivalent low nitogen medium (DMEM supplemented with
5% FCS). Simultaneously 10% of a rhBMP-2 solution (5 10
6
M, non-
treated and treated by purging through the print head) was added. After
incubation for a further 72 h the ALP activity was measured. Therefore,
the cell layers were washed with PBS. Next the lysis buffer 1 (0.1
M glycine
(Carl Roth, Karlsruhe, Germany), 1 10
3
M MgCl
2
(Merck, Darmstadt,
Germany), 1 10
3
M ZnCl
2
(Sigma–Aldrich, Taufkirchen, Germany), 1%
Nonidet P40 Substitute (Sigma–Aldrich, Taufkirchen, Germany)) was
added (96-well plates: 100 mL per well) and incubated at room temperature
for one hour in a shaker. Afterwards an equivalent quantity of lysis buffer 2
(0.1
M glycine (Carl Roth, Karlsruhe, Germany), 1 10
3
M MgCl
2
(Merck,
Darmstadt, Germany), 1 10
3
M ZnCl
2
(Sigma–Aldrich, Taufkirchen,
Germany), p-nitrophenyl phosphate (pNPP) 20 mg per mL of H
2
O (Sigma–
Aldrich, Taufkirchen, Germany)) was added in the ratio of 1:10 immediately
prior to use. The enzymatic conversion of pNPP to p-nitrophenol was
measured photometrically at 405 nm after 5 min. Data were normalized by
a standard curve using various concentrations of p-nitrophenol. The
protein content of the cell cultures was determined using the DC Protein
Assay (Biorad, Munich, Germany) following the producer’s manual and
using various concentrations of BSA (bovine serum albumin) as the
standard. The ALP activity was expressed as 10
6
M of p-nitrophenol
produced per minute per well. The biological activity of the printed solution
was compared with the activity of the control solution.
Acknowledgements
The authors sincerely thank the Deutsche Forschungsgemeinschaft (DFG
GB1/7-1), the Quebec-Bavarian support grant from the Ministe
`
re
De
´
velopement e
´
conomique, innovation et Exportation of Quebec, Canada
and the Canadian research chair (JB) for their financial support. The
company Kraeber GmbH is acknowledged for providing the chitosan
material and the Department of Physiological Chemistry II, Biocenter,
University of Wuerzburg, for providing the rhBMP-2.
Received: September 17, 2009
Revised: January 23, 2010
Published online: April 6, 2010
[1] Z. Sadeghian, J. G. Heinrich, F. Moztaradeh, CFI, Ceram. Forum Int. 2004,
81, E39.
[2] C. E. Wilson, J. D. de Bruijn, C. A. van Blitterswijk, A. J. Verbout,
W. J. A. Dhert, J. Biomed. Mater. Res. 2004, 68A, 123.
[3] J. K. Liu, O. N. Gottfried, C. D. Cole, W. R. Dougherty, W. T. Couldwell,
Neurosurg. Focus 2004, 16,1.
[4] S. J. Hollister, Nat. Mater. 2005, 4, 518.
[5] U. Gbureck, T. Ho
¨
lzel, U. Klammert, K. Wu
¨
rzeler, F. A. Mu
¨
ller, J. E. Barralet,
Adv. Funct. Mater. 2007, 17, 3940.
[6] E. Vorndran, M. Klarner, U. Klammert, L. M. Grover, S. Patel, J. E. Barralet,
U. Gbureck, Adv. Eng. Mater. 2008, 10, B67.
[7] U. Gbureck, E. Vorndran, F. A. Mu
¨
ller, J. E. Barralet, J. Controlled Release
2007, 122, 173.
[8] U. Gbureck, T. Ho
¨
lzel, C. J. Doillon, Adv. Mater. 2007, 19, 795.
www.afm-journal.de
www.MaterialsViews.com
1590 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2010,20,1585–1591
Page 6
FULL PAPER
[9] W. J. Rosoff, R. McAllister, M. A. Esrick, G. J. Goodhill, J. S. Urbach,
Biotechnol. Bioeng. 2005, 91, 754.
[10] L. Formigli, G. Fiorelli, S. Benvenuti, A. Tani, G. E. Orlandini, M. L. Brandi,
S. ZecchiOrlandini, Cell. Tissue Res. 1997, 288, 101.
[11] M. D. McKee, A. Nanci, Microsc. Res. Tech. 1996, 33, 141.
[12] J. E. Barralet, S. Aldred, A. J. Wright, A. G. A. Coombes, J. Biomed. Mater.
Res. 2002, 60, 360.
[13] B. G. Chung, L. A. Flanagan, S. W. Rhee, P. H. Schwartz, A. P. Lee,
E. S. Monuki, N. L. Jeon, Lab Chip 2005, 5, 401.
[14] P. Wu, D. W. Grainger, Biomaterials 2006, 27, 2450.
[15] J. Siepmann, N. A. Peppas, Adv. Drug Delivery Rev. 2001, 48, 139.
[16] M. Boakye, P. V. Mummaneni, M. Garrett, G. Rodts, R. Haid, J. Neurosurg.
Spine 2005, 2, 521.
[17] A. R. Vaccaro, T. Patel, J. Fischgrund, D. G. Anderson, E. Truumees,
H. N. Herkowitz, F. Phillips, A. Hilibrand, T. J. Albert, T. Wetzel,
J. A. McCulloch, Spine 2004, 29, 1885.
[18] L. Heijl, G. Heden, G. Svardstrom, A. Ostgren, J. Clin. Periodontol. 1997, 24,705.
[19] U. Gbureck, T. Ho
¨
lzel,R.Thull,F.A.Mu
¨
ller,J.E.Barralet,Cytotherapy 2006, 8,14.
[20] S. Chen, M. Liu, S. Jin, B. Wang, Int. J. Pharm. 2008, 349, 180.
[21] X. Z. Shu, K. J. Zhu, Int. J. Pharm. 2002, 233, 217.
[22] F. L. Mi, H. W. Sung, S. S. Shyu, C. C. Su, C. K. Peng, Polymer 2003, 44,
6521.
[23] K. K. Huang, C. Shen, C. Y. Chiang, Y. D. Hsieh, E. Fu, J. Periodont. Res.
2005, 40,1.
[24] R. S. Thies, M. Bauduy, B. A. Ashton, L. Kurtzberg, J. M. Wozney, V. Rosen,
Endocrinology 1992, 130, 1318.
[25] T. Katagiri, A. Yamaguchi, T. Ikeda, S. Yoshiki, J. M. Wozney, V. Rosen,
E. A. Wang, H. Tanaka, S. Omura, T. Suda, Biochem. Biophys. Res. Commu n.
1990, 172, 295.
[26] T. J. Poole, E. B. Finkelstein, C. M. Cox, Dev. Dyn. 2001, 220,1.
[27] C. K. Sen, S. Khanna, M. Venojarvi, P. Trikha, E. C. Ellison, T. K. Hunt,
S. Roy, Am. J. Physiol. Heart Circ. Physiol. 2002, 282, H1821.
[28] D. Rajalingam, T. K. Kumar, C. Yu, Biochemistry 2005, 44, 14431.
[29] N. R. Kuebler, J. F. Reuther, G. Faller, Int. J. Oral Maxillofac. Surg. 1998, 27,
307.
[30] H. M. Blau, C. P. Chiu, C. Webster, Cell 1983, 32, 1171.
www.MaterialsViews.com
www.afm-journal.de
Adv. Funct. Mater. 2010,20,1585–1591 ß 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1591
Page 7
    • "In contrast, low temperature 3D printing allowed spatial distribution of vancomycin within the printed geometry (Vorndran et al., 2010). In a more recent work, Inzana et al. (2015) didn't used a highly acidic binder (20 wt% phosphoric acid) as Vorndran et al. (2010), which may result in residual acidity that can be highly cytotoxic (Inzana et al., 2014), and can significantly degrade the bactericidal activity of the incorporated drug. They utilized a low acidity binder solution to enable lowtemperature 3D printing of vancomycin-and rifampin-laden CaP scaffold while maintaining the bactericidal activity (Inzana et al., 2015). "
    [Show abstract] [Hide abstract] ABSTRACT: Bone tissue engineered 3-D constructs customized to patient-specific needs are emerging as attractive biomimetic scaffolds to enhance bone cell and tissue growth and differentiation. The article outlines the features of the most common additive manufacturing technologies (3D printing, stereolithography, fused deposition modeling, and selective laser sintering) used to fabricate bone tissue engineering scaffolds. It concentrates, in particular, on the current state of knowledge concerning powder-based 3D printing, including a description of the properties of powders and binder solutions, the critical phases of scaffold manufacturing, and its applications in bone tissue engineering. Clinical aspects and future applications are also discussed.
    No preview · Article · Apr 2016 · Biotechnology advances
  • Source
    • "As an example, it has been established that it is possible to produce customized food with 3D Hydrocolloid Printing [10, 11], though many challenges in regard to materials and construction remain. It has also been recently demonstrated that 3D printing can produce bone grafts with localized organic bioactive loading and diffusion control, possibly offering a breakthrough for patients by delivering the necessary material function to match human bone health status, site of repair, and age [12]. Click Chemistry and 3D drug printing: Click chemistry was a term coined to address the possibility of developing synthetic strategies that enable much more rapid discovery and production of molecules with a desired profile of properties [13]. "
    [Show abstract] [Hide abstract] ABSTRACT: Genomics is a disruptive technology, having revealed that cancers are tremendously complex and differ from patient to patient. Therefore, conventional treatment approaches fit poorly with genomic reality. Furthermore, it is likely that this type of complexity will also be observed in other illnesses. Precision medicine has been posited as a way to better target disease-related aberrations, but developing drugs and tailoring therapy to each patient's complicated problem is a major challenge. One solution would be to match patients to existing compounds based on in silico modeling. However, optimization of complex therapy will eventually require designing compounds for patients using computer modeling and just-in-time production, perhaps achievable in the future by three-dimensional (3D) printing. Indeed, 3D printing is potentially transformative by virtue of its ability to rapidly generate almost limitless numbers of objects that previously required manufacturing facilities. Companies are already endeavoring to develop affordable 3D printers for home use. An attractive, but as yet scantily explored, application is to place chemical design and production under digital control. This could be accomplished by utilizing a 3D printer to initiate chemical reactions, and print the reagents and/or the final compounds directly. Of interest, the Food and Drug Administration (FDA) has recently approved a 3D printed drug-levetiracetam-indicated for seizures. Further, it is now increasingly clear that biologic materials-tissues, and eventually organs-can also be "printed." In the near future, it is plausible that high-throughput computing may be deployed to design customized drugs, which will reshape medicine.
    Preview · Article · Jan 2016 · Oncotarget
  • Source
    • "The main benefits of low temperature 3D printing is its capability to construct polymer/mineral composites with increased material properties to include growth factors and drugs to enhance bone regeneration or combat infection.[89] Fused deposition modeling has frequently utilized biomedical polymers with low melting temperatures.[90] "
    [Show abstract] [Hide abstract] ABSTRACT: Current methods in handling maxillofacial defects are not robust and are highly dependent on the surgeon's skills and the inherent potential in the patients' bodies for regenerating lost tissues. Employing custom-designed 3D printed scaffolds that securely and effectively reconstruct the defects by using tissue engineering and regenerative medicine techniques can revolutionize preprosthetic surgeries. Various polymers, ceramics, natural and synthetic bioplastics, proteins, biomolecules, living cells, and growth factors as well as their hybrid structures can be used in 3D printing of scaffolds, which are still under development by scientists. These scaffolds not only are beneficial due to their patient-specific design, but also may be able to prevent micromobility, make tension free soft tissue closure, and improve vascularity. In this manuscript, a review of materials employed in 3D bioprinting including bioceramics, biopolymers, composites, and metals is conducted. A discussion of the relevance of 3D bioprinting using these materials for craniofacial interventions is included as well as their potential to create analogs to craniofacial tissues, their benefits, limitations, and their application.
    Full-text · Article · Oct 2015 · Journal of Prosthodontics
Show more