Biocompatibility of choline salts as crosslinking agents for collagen based biomaterials

Article abstract of DOI:10.1039/b910601d

A series of novel choline based salts, some of which can be described as ionic liquids, are prepared and evaluated for their biocompatibility; when combined with collagenous biomaterials they exhibit good cell viability and adhesion properties as required for biomedical implant applications.

Full text of DOI:10.1039/b910601d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Biocompatibility of choline salts as crosslinking agents for
collagen based biomaterialsw
R. Vijayaraghavan,a B. C. Thompson,a D. R. MacFarlane,*a Ramadhar Kumar,b
M. Surianarayanan,b S. Aishwaryab and P. K. Sehgalb
Received (in Cambridge, UK) 29th May 2009, Accepted 3rd November 2009
First published as an Advance Article on the web 26th November 2009
DOI: 10.1039/b910601d
A series of novel choline based salts, some of which can be
described as ionic liquids, are prepared and evaluated for their
biocompatibility; when combined with collagenous biomaterials
they exhibit good cell viability and adhesion properties as
required for biomedical implant applications.
collagen based wound dressings are advantageous over other
systems because of unique characteristics such as weak
antigenicity,8 biodegradability,9 and superior biocompatibility.10
Collagen materials for long-term biomedical implant applications
need to be crosslinked to improve the mechanical stability
of the biopolymer and induce controlled biodegradability
because, when not crosslinked, collagen membranes exhibit
poor mechanical properties and rapid degradation limits their
application as a biomaterial.11 Glutaraldehyde (GA) is the
most widely used chemical crosslinking agent11 because it
stabilizes collagen efficiently. The crosslinking is thought to
involve the formation of Schiff bases.12 However, GA-crosslinked
biomaterials are poorly biocompatible with some cell lines
including human fibroblasts, osteoblasts, Chang cells, and
endothelial cells.13–15 The side effects of GA treatment were
attributed to the degradation of the GA-derived crosslinks and
the continuous release of aldehydes contributing to prolonged
toxic effects.11,16 Several alternative crosslinking agents such
as diphenylphosphoryl azide (DPPA) have been developed to
minimize the toxicity.17 Crosslinking with DPPA reduces
degradation of collagen biomaterials and produces a more
biocompatible material than crosslinking with GA.11,18
However, there is a need to develop crosslinkers which are
completely biocompatible and produce no inflammatory
response for long-term implant applications. In this study
choline salts have been used as crosslinkers with collagen based
biomaterials and the biocompatibility of the resultant materials
tested.
Tissue engineering is now gaining importance as an alternative
to tissue or organ transplantation. With this technology, tissue
loss or organ failure can be treated either by implantation of
an engineered biological substitute or, alternatively, by ex vivo
perfusion methods. The tissue-engineered products may be
fully operational at the time of treatment or have the potential
to integrate and form the expected functional tissue upon
implantation.1,2 In many cases, biomaterials are modified to
repair or replace the damaged tissue. For example, progress
has been made with engineered cardiovascular tissues such as
heart valves3,4 and blood vessels.5,6 The development of
biomaterials poses significant challenges since the implanted
tissue is greatly influenced by the composition, architecture,
three-dimensional environment of the implanted ‘‘scaffold’’
and the biocompatibility of the material. The material porosity
and pore size distribution and continuity greatly influence
the attachment of specific cell types and interaction of the
biomaterials with the host. The inflammatory response to
biomaterials has also been a major safety problem and many
of the currently utilized biomaterials elicit inflammatory
responses upon implantation.7 Hence there is a need for stable
biocompatible materials that are bio-acceptable and that,
preferably, having induced new tissue growth, will degrade
via normal biochemical pathways.
In recent years, ionic liquids (ILs) have been extensively
investigated for use as replacement solvents for many chemical
and biochemical reactions.19–21 It has been shown that enzymes
Biomaterials based on natural collagen are emerging as a
potential source of such biocompatibility. Collagen is a
protein mainly existing in human skin, skeleton, cartilage,
tendons, ligaments and blood vessels, and it is a very important
structural protein which supports and protects the organs and
body. Collagen makes up 25–33% of total proteins in the body
and 80% of the skin composition which is almost equal
to 6% of bodyweight. It can be harvested from a number
of sources including animals and processed by dissolution and
re-precipitation, and hence is readily available as a biomaterial.
It has appeared in a number of applications, for example
can be used to carry out bio-transformations in ILs.22,23
A
number of biocompatible hydrated ionic liquids have been
identified, for example choline dihydrogenphosphate (CDP)
with 20% dissolved water, and shown to be good solvents for
proteins.24,25 In the ionic solution some proteins are tremendously
stabilized as compared to aqueous solutions,26,27 thus leading
to a variety of applications in drug delivery and sensor
development. The presence of a solute amount of water in
these systems is probably an important factor in supporting
solubility and stability.
In the present study we have synthesized a number of novel
choline salts, based on other anions from the ‘‘Generally
Recognized as Safe’’ list or known Active Pharmaceutical
Ingredients (Fig. 1), including choline lactate (CL), choline
levulinate (CLu), and choline tartarate (CT). They have then
been used as replacements for the most commonly used
a Monash University, Wellington Road, Clayton Campus, Clayton,
Victoria, 3800, Australia. E-mail: d.macfarlane@sci.monash.edu.au
b CISRA and Biproducts Lab., CLRI Chennai 600020, India
w Electronic supplementary information (ESI) available: ESI available
includes IR spectroscopic data and CD spectra, thermogravimetric
traces and biocompatibility assay details. See DOI: 10.1039/b910601d
ꢀc
This journal is The Royal Society of Chemistry 2010
294 | Chem. Commun., 2010, 46, 294–296

View Article Online
not a significant process. IR and circular dichroism spectro-
scopy data (ESIw, Table S1 and Fig. S1) indicate that the triple
helical structure of the collagen strand is retained in the
crosslinked materials. Thermogravimetric analysis (ESIw,
Fig. S2) shows that the materials are more thermally stable
than the GA crosslinked materials.
Fig. 1 Chemical structures of cation and anions.
The crosslinking involved here is probably hydrogen
bonding based since the anions all share a feature of having
multiple hydrogen bonding sites, both donor and acceptor.
Natural collagen in the body is held insoluble by the action of
such hydrogen bonds, between glycine and hydroxyproline
units in particular. These units are disrupted by the acid
treatment that creates soluble collagen. Hence our hypothesis
is that the role of the crosslinker here is to provide an alternate
means of forming these strong hydrogen bonds. The ionic
nature of the crosslinker probably further strengthens the
intermolecular interactions. The IR data (ESIw, Table S1)
support this hypothesis. As described by Usha29et al. the
collagen amide A band shifts to higher frequencies as a result
of crosslinking, the Cr(III) ion producing the strongest effect.
The data in ESIw Table S1 show that the shifts observed in the
present materials are in some cases almost as substantial as
Cr(III). This shift may stem from coordinate or H-bonding
interactions of the amide nitrogen, hence in the present case
where all of the anions are capable of multiple H-bonds, it
seems likely that this is the origin of the crosslinking effect.
The choline based salts and the crosslinked collagen films
were characterized for biocompatibility using the 3-[4,5-
dimethyltriazol-2-y1]-2,5-diphenyltetrazolium (MTT) assay
with cardiac fibroblast cells, as described in the literature30
and the results were compared with control samples. The
controls used were the neat medium (without collagen or
crosslinker) and collagen crosslinked with GA. Further details
of the MTT assay and method are provided in the ESI.w The
results show (Fig. 2 and 3) the ability of the salts to support
cell viability and proliferation. All the samples exhibited
comparable biocompatibility to GA. In addition, CLu and
CT showed better biocompatibility over GA in this assay.
Microscopic observation of rat fibroblastoma cells (L-929)
confirmed that the cells in all cases were well grown and
attached to the surface, although the CLu and CT crosslinked
materials produce better growth compared to the sample
crosslinked with GA. Cell spreading and attachment was
monitored by fluorescence microscopy of fixed and fluorescently
labelled cells (details provided in ESIw) and typical micro-
graphs of the control (GA crosslinked) and CT crosslinked
films are shown in Fig. 4. The results show that the cells were
chemical crosslinkers for collagen, such as GA, to reduce the
toxicity of collagen based biomaterials.
Choline hydroxide (45% in methanol), L(+) lactic acid
(98%), levulinic acid (98%), D(À) tartaric acid (99%) and
bovine achilles tendon were purchased from Aldrich while
phosphoric acid (85%) aqueous solution was purchased from
Merck. The general synthesis involves (1 : 1 mole ratio)
neutralization reaction of choline hydroxide and the corres-
ponding acids (such as lactic acid, phosphoric acid, etc.) and
dehydrating the reaction mixture at high temperature (70 1C)
and reduced pressure to obtain the pure salt. For example,
choline lactate (CL) (10 g) was made by a slow addition of
lactic acid (4.66 g) to 45% methanolic choline hydroxide
solution (13.93 g) in an ice bath and stirring it for about
2 hours at room temperature. Then the reaction mixture was
evaporated at reduced pressures to obtain crude CL. To
the crude compound, activated charcoal (approx. 1 g) was
added, stirred with water and filtered. The filtrate was again
evaporated to obtain the pure liquid in 98% yield. The other
choline derivatives were prepared in a similar manner. All
were characterized by electrospray mass spectroscopy and
only the expected anion and cation were observed in each
case. Electrospray mass spectroscopy analysis (cone Æ35 V) : CL,
m/z (relative intensity, %): ES+, 103.7 (Me3N+CH2CH2OH,
100); ESÀ, 89.8 (lactate, 100); CDP, ES+, 103.7
(Me3N+CH2CH2OH, 100); ESÀ, 96.7 (dihydrogenphosphate,
100); CLu, ES+, 103.7 (Me3N+CH2CH2OH, 100); ESÀ, 114.9
(levulinate, 100); CT, ES+, 103.7 (Me3N+CH2CH2OH, 100);
ESÀ, 149.3 (tartarate, 100). CL and CLu are liquids at room
temperature. CDP and CT are solids with melting points
185 1C and 150 1C, respectively. The latter two easily become
liquid on addition of small amounts of water. In both cases
20% (w/w) is sufficient to produce a clear, fluid liquid at room
temperature.
The collagen was isolated from bovine achilles tendon as an
aqueous solution as per the literature procedure.28 The
crosslinking procedure typically involves the mixing of 5 g
collagen solution (2 mg mlÀ1) with 7.5 ml of an 8 mM aqueous
solution of the crosslinker at room temperature. The mixture
was then cast on a Teflon sheet and dried at room temperature
overnight to produce a thin film. GA is well known to cross-
link the collagen fibrils, via Schiff base formation, rendering
the material water insoluble, as observed in this work with the
ionic crosslinkers. In a certain respect this replicates the
properties of the natural material, which in vivo must be
insoluble. Water uptake tests show that the dry materials take
up water to the extent of about 70–200% w/w after 24 hours,
depending on the ionic liquid involved. There was little further
uptake and after 6 months continuous immersion in a
relatively large volume (20 ml) of pure water the samples
remain intact. Since the original collagen is water soluble, this
indicates that leaching of the ionic liquid from the material is
Fig. 2 MTT assay of choline based salts as compared to controls.
Chem. Commun., 2010, 46, 294–296 | 295
ꢀc
This journal is The Royal Society of Chemistry 2010

View Article Online
Notes and references
1 W. Yongzhong, J. B. Dominick, K. Hyeon-Joo, S. K. Hyun and
L. K. David, Biomaterials, 2006, 27, 4434.
2 C. S. Cho, S. J. Seo, I. K. Park, S. H. Kim, T. H. Kim, T. Hoshiba,
I. Harada and T. Akaike, Biomaterials, 2006, 27, 576.
3 T. C. Flanagan and A. S. Pandit, Eur. Cells Mater., 2003, 6, 28.
4 L. Xue and H. P. Greisler, J. Vasc. Surg., 2003, 37, 472.
5 L. E. Niklason, J. Gao, W. M. Abbott, K. Hirschi, S. Houser,
R. Marini and R. Langer, Science, 1999, 284, 489.
6 P. J. Schaner, N. D. Martin, T. N. Tulenko, I. M. Shapiro,
N. A. Tarola, R. F. Leichter, R. A. Carabasi and P. J. DiMuzio,
J. Vasc. Surg., 2004, 40, 146.
7 J. M. Anderson, Trans. Am. Soc. Artif. Intern. Organs, 1988, 34,
101.
Fig. 3 MTT assay of the choline salt crosslinked biomaterials.
8 F. L. Nicolas and C. H. Gagnieu, Biomaterials, 1997, 18, 815.
9 G. Goissis, E. Marcantoniojunior, R. A. C. Marcantonio, R. C.
C. Lia, D. C. J. Cancian and W. M. de Carvalho, Biomaterials,
1999, 20, 27.
10 S. Eun Song, K. Yeon, C. Taehoon, B. Hyun-Jung and L. Young
Moo, Biomaterials, 2006, 27, 2951.
11 E. Jorge-Herrero, P. Fernandez, J. Turnay, N. Olmo, P. Calero,
R. Garcia, L. Freile and J. J. Castillo-Olivares, Biomaterials, 1999,
20, 539.
12 L. H. H. Olde Damink, M. Dijkstra, M. J. A. Van Luyn, P. B. Van
Wachem, P. Nieuwenhuis and J. Feijen, J. Mater. Sci.: Mater.
Med., 1995, 6, 460.
13 L. L. H. Huang-Lee, D. T. Cheung and M. Nimni, J. Biomed.
Mater. Res., 1990, 24, 1185.
14 M. J. A. Van Luyn, P. B. Van Wachen, L. H. H. Olde Damink,
P. J. Dijkstra, J. Feijen and P. Nieuwenhuis, Biomaterials, 1992, 13,
1017.
15 J. E. Gough, C. A. Scotchford and S. Downes, J. Biomed. Mater.
Res., 2002, 61(1), 121.
16 C. E. Schmidt and J. M. Baier, Biomaterials, 2000, 21, 2215.
17 H. Petite, V. Frei, A. Huc and D. Herbage, J. Biomed. Mater. Res.,
1994, 28, 150.
18 S. Roche, M. C. Ronzie‘re, D. Herbage and A. M. Freyria,
Biomaterials, 2001, 22, 9.
19 J. S. Wilkes, Green Chem., 2002, 4, 73.
20 J. Golding, S. Forsyth, D. R. MacFarlane, M. Forsyth and
G. B. Deacon, Green Chem., 2002, 4, 223.
21 P. Wasserscheid and T. Welton, Ionic Liquids in Synthesis, Wiley,
Weinheim, 2003.
22 U. Krugle, M. Eckstein and N. Kaftzik, Biocatalytic reactions in
ionic liquids, Wiley, Weinheim, 2003, p. 336.
Fig. 4 Fluorescence micrographs of L-929 cells growing on GA and
CT crosslinked materials. Actin cytoskeleton is stained with phalloidin
(green), and the nuclear stain DAPI shows the cell nuclei (blue). Scale
bars represent 50 mm.
23 J. L. Kaar, J. Am. Chem. Soc., 2003, 125, 4125.
24 K. Fujita, D. R. MacFalane and M. Forsyth, Chem. Commun.,
2005, 4804–3043; O. Winther-Jensen, R. Vijayaraghavan,
J. Z. Sun, B. Winther-Jensen and D. R. MacFarlane, Chem.
Commun., 2009, 3041–3043.
25 K. Fujita, M. Forsyth, D. R. MacFarlane, R. W. Reid and
G. D. Elliott, Biotechnol. Bioeng., 2006, 94, 1209–1213;
K. Fujita, D. R. MacFalane, M. Forsyth, M. Y. Fujita,
K. Murata, N. Nakamura and H. Ohno, Biomacromolecules,
2007, 8(7), 2080–1213.
26 N. Byrne and C. A. Angell, J. Mol. Biol., 2008, 378, 707.
27 N. Byrne, L. M. Wang, J. P. Belieres and C. A. Angell, Chem.
Commun., 2007, 2714.
28 Z. Deyl and M. Adam, The methodology of connective tissue
research, Joynson-Bruvvers, Oxford, 1st edn, 1976, p. 1.
29 R. Usha, A. Dhathathreyan, A. B. Mandal and T. Ramasami,
J. Polym. Sci., Part B: Polym. Phys., 2004, 42, 3859–3865.
30 T. J. Mossmann, J. Immunol. Methods, 1983, 65(1–2), 55.
found to be healthy, capable of proliferation, forming a good
coverage on the material and attached very well to the surface.
Overall these results show that these choline based salts, in
particular those based on the levulinate and tartarate anions, can
be useful for biomedical implant applications. Studies pertaining
to the mechanical properties of the salt crosslinked biomaterials
will be reported separately and the in vivo studies of the
biomaterial films for wound healing applications are underway.
DRM is grateful to the Australian Research Council for his
Federation Fellowship.
ꢀc
This journal is The Royal Society of Chemistry 2010
296 | Chem. Commun., 2010, 46, 294–296