Reinforced Hybrid Collagen Sponges for Tissue Engineering

Abstract

We created an anisotropic material based on collagen sponge and reactive polylactide structured by laser photopolymerization. The combination of collagen with reactive polylactide improves the resistance of the formed matrices to biodegradation in comparison with collagen sponge, while the existence of sites with different mechanical characteristics and cell affinity on the matrix provides directed cell growth during their culturing. It was shown that reinforcement of the collagen sponges 7-fold increased the mean Young’s modulus for the hybrid matrix without affecting its cytotoxicity. The developed matrix provides cell adhesion and proliferation along reinforcement lines and can be used for fabrication of tissue engineering constructs.

Full text

142
0007 -4888/18/16510142 © 2018 Springer Science+Business Media, LLC
Reinforced Hybrid Collagen Sponges for Tissue Engineering
K. N. Bardakova
1,2
, E. A. Grebenik
2
, E. V. Istranova
2
, L. P. Istranov
2
,
Yu. V. Gerasimov
3
, A. G. Grosheva
3
, T. M. Zharikova
2
, N. V. Minaev
1
,
B. S. Shavkuta
1,2
, D. S. Dudova
1
, S. V. Kostyuk
4
, N. N. Vorob’eva
1
,
V. N. Bagratashvili
1
, P. S. Timashev
1,2
, and R. K. Chailakhyan
1,3
Translated from Kletochnye Tekhnologii v Biologii i Meditsine, No. 1, pp. 31-37, January, 2018
Original article submitted December 11, 2017
We created an anisotropic material based on collagen sponge and reactive polylactide struc-
tured by laser photopolymerization. The combination of collagen with reactive polylactide
improves the resistance of the formed matrices to biodegradation in comparison with col-
lagen sponge, while the existence of sites with different mechanical characteristics and cell
afnity on the matrix provides directed cell growth during their culturing. It was shown that
reinforcement of the collagen sponges 7-fold increased the mean Young’s modulus for the
hybrid matrix without affecting its cytotoxicity. The developed matrix provides cell adhesion
and proliferation along reinforcement lines and can be used for fabrication of tissue engineer-
ing constructs.
Key Words: collagen matrix; laser photopolymerization; directed cell growth; branched
polylactide; multipotent stromal cells
1Institute of Photonics Technologies, Federal Research Centre “Crys-
tallography and Photonics”, Russian Academy of Sciences; 2Institute
of Regenerative Medicine, I. M. Sechenov First Moscow State Medi-
cal University, Ministry of Health of the Russian Federation; 3N. F.
Gamaleya National Research Center of Epidemiology and Microbio-
logy, Ministry of Health of the Russian Federation, Moscow, Russia;
4Research Institute for Physical Chemical Problems, Belarusian State
University, Minsk, Belarus. Address for correspondence: ruben-
chail@yandex.ru. R. K. Chailakhyan
The main requirements for tissue engineering matrices
include among other things biocompatibility, ability to
promote cell adhesion and proliferation, optimal rate
of biodegradation, and mechanical properties of native
tissue [4]. The materials based on collagen, the main
organic component of the intercellular matrix, largely
meet the above requirements and are of great interest.
Collagen scaffolds can be obtained by either chemi-
cal, mechanical, or enzymatic removal of cells from
tissues (decellularization) or extraction of collagen
from different tissues followed by creation of collagen
structures from the solution [3]. The second method
is used for preparing hydrogels, lms, and sponges
that are often characterized by relatively high rate of
biodegradation and insufcient mechanical strength
for their application in surgery. These problems can be
solved by proper choice of collagen type, variation of
collagen concentration, structuring by cross-linking [5],
spatial ordering of collagen bers [7], and modication
with synthetic polymers and inorganic agents [6].
For a number of medical tasks (e.g. muscle tis-
sue repair), the use of traditional biomaterials is also
limited due to their insufcient mechanical strength
and elasticity. These properties can be improved by
different methods. For instance, a hybrid scaffold was
prepared on the basis of collagen sponge treated with
glutaraldehyde and reinforced with polylactide bers
[10] Another strategy of modication is based on in-
clusion of bio-functional materials inducing certain
biological processes and improving mechanical prop-
erties of the original collagen matrix. For instance, hy-
droxyapatite nanoparticles were used as bioactive and
reinforcing additives [8]. Collagen scaffolds formed
by different methods can also be modied by different
polymer coatings. Thus, alginate coating of the col-
lagen scaffold formed by extrusion printing allowed
increasing Young’s modulus by 9 times [9]. The use
Cell Technologies in Biology and Medicine, No. 1, May, 2018
DOI 10.1007/s10517-018-4116-8

143
of polylactide mesh also improved mechanical char-
acteristics of collagen materials. The prospects of us-
ing such hybrid scaffolds based on collagen/gelatin
sponges for skin regeneration were discussed [11].
Reinforcement of the collagen sponge intended for
epithelial tissue repair with commercial polylactide
meshes was reported to increase scaffold resistance to
in vivo biodegradation [1].
Our aim was to create an anisotropic material
based on collagen sponge and reactive polylactide
structured by laser photopolymerization. Reinforce-
ment with polylactide is assumed to reduce the rate
of biodegradation [13,14] and stimulate directed cell
proliferation along the polylactide bers followed by
uniform population of the matrix.
MATERIALS AND METHODS
Preparation of the collagen sponge. The collagen
sponge was prepared from collagen solution (0.7 mg/ml;
pH 2.9) by the method of freeze-drying. For preparing
collagen solution, the middle layer of cattle dermis
was cut into 5×5 cm pieces that were incubated in
2.5 M NaOH with 0.85 M Na2SO4 for 48 h at 20oC
with periodic mixing. After 6-h washing in 0.85 M
Na2SO4, the fragments were incubated with 4% boric
acid until complete alkali neutralization on section
(qualitatively assessed by the reaction with phenol-
phthalein). Neutralized samples were washed with
distilled water until negative reaction for the pres-
ence of sulphate ions, after which they were placed in
0.5 M acetic acid to complete dissolution. The resultant
solution was puried by collagen precipitation with
12% NaCl followed by centrifugation at 3000 rpm for
20 min; the pellet was then dissolved in 0.25 M acetic
acid. To remove salts, the solution was dialyzed for
24 h against 0.25 M acetic acid. The collagen solution
brought to a concentration of 0.7 mg/ml was trans-
ferred to cuvettes with (layer thickness 3 mm), frozen
at -30oC, and lyophilized.
Preparation of photosensitive composition.
To obtain photoactive material, 800 µl of 50 wt%
branched polylactide solution (M(NMR)=749 g/mol
for one beam; proportion of reagents during synthesis:
[D,L-lactide]/[PE]=20/1; detailed synthesis procedure
described in [12]) was mixed with 3600 µl dichloro-
methane and 40 mg of photoinitiator 4,4’-bis (diethyl-
amino)benzophenone, the solution was mixed for 24 h
and then used for laser structuring.
Experiment on laser photopolymerization. The
collagen sponge (a square fragment with a side of 15
mm and thickness 2-4 mm) was evenly coated with
photosensitive composition (500 µl). In 10 min, the
excess of photoactive liquid was removed from super-
cially dry collagen sponge (to this end, the sponge
was squeezed for 2-3 sec between the coverslips);
then, a metal mold with a square window was placed
on top of the collagen sponge to x the scaffold on
the slide.
The reinforcement was carried out using a UV di-
ode laser at λ=405 nm and maximum power 100 mW.
Laser radiation was focused directly on the sample, the
diameter of the laser spot was 60-80 μ. Reinforcing
lines were formed at the following radiation param-
eters: laser power from 50-70 mW, scanning speed
on the surface of the sample 3-5 mm/sec, the distance
between the centers of neighboring lines varied from
200 to 250 μ. This laser treatment allowed reinforcing
the collagen lm without damaging its structure.
Removal of polymer excess. Immediately af-
ter structuring, the reinforced collagen scaffold was
placed in dichloromethane for 2-3 days (the solvent
was changed 3-4 times) and then washed with ethyl
alcohol (95 vol%) and deionized water for 7-10 days.
Characterization of hybrid scaffolds. The scaf-
fold morphology was studied by scanning electron
microscopy Phenom Pro X (Phenom). Collagen struc-
ture integrity after laser exposure was studied by re-
ectance IR spectroscopy on a Spectrum 100 FT-IR
spectrometer (Perkin Elmer).
Nanoindentation. The local mechanical proper-
ties of the prepared collagen sample were determined
using a Piuma nanoindenter (Optics11) that included
a controller, optical ber, and spherical probe used
for construction of the force—displacement curves.
The probe was xed to a exible beam of the canti-
lever. The contact of the cantilever with the sample
surface led to its deection that was measured inter-
ferometrically using an optical ber. To determine
Young’s modulus of the collagen scaffolds, the probe
was forced 5 µ deep into the sample at each measure-
ment point. The Young’s modulus was calculated from
the force—displacement curve using Piuma applied
software and Hertzian model for a sphere in contact
with a plane.
To study mechanical characteristics of colla-
gen scaffolds, a cantilever with a spring constant of
0.45 N/m and tip radius 26.5 μ was used. The samples
were xed on the bottom of a Petri dish. All mea-
surements were carried out in distilled water at 37oC.
During measurements, the cantilever probe always re-
mained in the liquid medium to minimize errors due to
adhesion forces at the air—water interface. The area for
Young’s modulus distribution maps was 1000×500 µ
with 25 µ increments along X and Y axes. Based on
the results of measurements, the effective Young’s
modulus was determined and the maps of Young’s
modulus distribution on the surface were constructed.
MTT test. 3T3 mouse broblasts were seeded
on a 96-well plate (5000 cells per well) and cultured
K. N. Bardakova, E. A. Grebenik, et al.

144
in a CO2 incubator at 37oC in DMEM/F-12 (PanEco)
supplemented with glutamine and 10% fetal calf serum
(FCS; HyClone). In 24 h, the culture medium was
replaced with successive dilutions of extracts of the
original collagen sponge and hybrid scaffold prepared
in a culture medium containing 5% FCS. The extracts
were obtained by incubating the samples (total area
6 cm2 each) in culture medium in a CO2 incubator at
37oC for 24 h. Sodium dodecyl sulfate in a concentra-
tion range of 0.013-0.500 mg/ml was used as the posi-
tive control. The sponges were preliminary sterilized
in 70% ethanol for 30 min at constant stirring, then
washed three times.
Isolation of human bone marrow mesenchymal
stromal cells (MSC). Human bone marrow was ob-
tained from a maxillofacial surgical clinic; informed
consent of the patient was obtained. The bone mar-
row specimen was transferred to a vial with fresh nu-
trient medium, suspended with a syringe, and ltered
through a 4-layer capron lter; the total number of
cells was counted. The cells were seeded into 80-cm2
plastic asks (Nunc) with 15 ml complete culture me-
dium containing 80% α-MEM (Sigma-Aldrich), 20%
FCS (HyClone), and antibiotics (100 U/ml penicillin
and 100 mg/ml streptomycin); seeding density 3.5-
4.0×104 cells/cm2. The cells were cultured at 37oC
and 5% CO2. The cells were rst passaged on days
12-14 of culturing, when discrete colonies of stro-
mal broblasts appeared, and then, upon attaining
conuence.
MSC culturing in scaffolds. For studying cell
adhesion and growth on the hybrid scaffold, passage
4 human bone marrow MSC were used. The scaffold
(reinforced collagen) was preliminary sterilized and
placed in the medium. Fragments 3×3 mm were dried
with a tampon for complete medium removal. Then,
the fragments were treated with FCS for 15 min and
then soaked with the tampon. The fragments were
placed onto the bottom of 25 cm2-culture ask tightly
to each other. The total area of this construction was
~1 cm2. The scaffold was lled with cells suspension
(concentration 106 in 0.5 ml culture medium) using a
syringe. The cells adhered to plastic for 1.5 h at 37oC
and 5% CO2. Then, the total volume of the medium
supplemented with serum was added to the culture
ask and the cells were cultured for 2 months; the
medium was changed regularly.
Confocal and light microscopy. Population
of the hybrid scaffold by cells was assessed using
a Nikon A1 Multiphoton confocal microscope and
Nikon Eclipse TS 100) light inverted microscope.
For confocal microscopy, the cells were fixed with
4% formalin and the nuclei were poststained with
ethidium bromide homodimer (Thermo Fisher Sci-
entific).
RESULTS
At the rst stage of the study, the regimens for form-
ing reinforcing strips of reactive polylactide on the
collagen sponge were chosen. The parameters of laser
exposure were as follows: power 70 mW, laser beam
speed 5 mm/sec, density 4 lines per mm. These condi-
tions prevented overheating of the collagen material
under the action of laser radiation and reinforced colla-
gen scaffolds after structuring retained exibility typi-
cal of collagen sponge. Reinforcement did not change
porosity of the collagen scaffold, but the proportion of
free pores in the material decreased (Fig. 1, a, b). At
the same time, the structured scaffolds retained their
conguration and manipulative properties after 2-week
incubation in PBS, in contrast to the original collagen
sponge that considerably swelled and disintegrated to
separate bers during this time.
This change in the physical properties of the scaf-
fold (reduced swelling in the solvent and reduced per-
meability) attests to increased concentration of cross-
links and reorganization of the 3D network. The length
of polylactide lines was 70-80 μ and the distance be-
tween them 140-150 µ.
Possible changes in the collagen structure after
laser exposure were assessed by IR spectroscopy. On
both IR spectra of collagen (Fig. 1, c), we observed
the appearance of typical oscillation frequencies of
hydroxyl groups (3200-3700 cm—1; not shown on our
spectra) and peptide bonds, amides I, II and III (1200-
1700 cm—1). During collagen denaturation, its triple
helix becomes unwound and forms random peptide
coils, which should affect uctuations of amides I
and III most sensitive to such conformational transi-
tions. Unchanged oscillation frequency can indicate
the absence of transformation of collagen bers dur-
ing laser heating. Insignicant shift of amide II band
towards the low-frequency region after reinforcement
and washing procedure (from 1546 to 1538 cm—1)
can indicate weakening of hydrogen bonds due to in-
creased humidity of the collagen scaffold. After rein-
forcement, the mean Young’s modulus for the collagen
scaffold increased by ~7 times: to 520.75±201.23 vs.
77.60±55.30 kPa for original scaffold (Fig. 1, d).
The obtained result was primarily determined by
the formation of photocured polylactide strips on the
collagen surface, which was conrmed by distribution
of the Young’s modulus values on the surface of the
reinforced scaffold (Fig. 1, d). The highest values of
Young’s modulus showed periodicity corresponding to
that of polylactide strips applied to the collagen scaf-
fold. Only one of the reinforcing lines is clearly seen
(Fig. 1, d). The non-uniform mechanical properties of
the obtained strips are determined by anisotropy of
the initial collagen scaffold, which leads to the sec-
Cell Technologies in Biology and Medicine, No. 1, May, 2018

145
Fig. 1. Microscopic, spectral, and mechanical characteristics of the original and reinforced matrix. a) Original collagen sponge; b) imme-
diately after laser structuring; c) IR spectra of reinforced (1) and original (2) collagen matrix; d) surface distribution of Young’s modulus
for the original and reinforced collagen sponge. Polylactide strips formed by laser irradiation on the surface of the collagen sponge (b, d).
K. N. Bardakova, E. A. Grebenik, et al.

146
Fig. 2. Biocompatibility of collagen sponges. a) MTT-analysis of cytotoxicity of extracts of the original and reinforced collagen sponge. So-
dium dodecyl sulfate (SDS) was used as the positive control. b, c) MSC cultured for 1 month (b) and 2 months (c). Staining with ethidium
bromide homodimer. Confocal microscopy.
Fig. 3. Vital light microscopy of human bone marrow MSC cultured on reinforced collagen for 1 week (a) or 4 weeks (b), ×20.
Cell Technologies in Biology and Medicine, No. 1, May, 2018

147
ond reason for Young’s modulus growth for reinforced
scaffold — the formation of spatial cross-linking in the
collagen material as a result of laser exposure. Con-
siderable macroscopic heterogeneity of the original
collagen scaffold is the cause of partial reection and
scattering of laser radiation and energy absorption by
the collagen scaffold because of diffuse scattering of
photons leads to the formation of cross-links in the
polymer network. The existence of areas with different
mechanical properties in the reinforced collagen scaf-
fold is very important for cell culturing on its surface.
The cytotoxicity of the reinforced sponge was as-
sessed by the MTT analysis (Fig. 2, a).
Addition of the polylactide component during re-
inforcement did not increase the cytotoxicity of the
collagen sponges. The observed decrease in prolifera-
tive activity of cells can be explained by high swell-
ing capacity of collagen structures and, consequently,
depletion of the growth medium with FCS present in
the scaffold.
The human bone marrow MSC were stained with
ethidium bromide homodimer (orange) (Fig. 2, b, c).
Blue color corresponds to scaffold autouorescence,
this parameter reects colocalization of cells and sites
modied by polylactide.
Vital light microscopy of human bone marrow
MSC on reinforced collagen showed active prolifera-
tion of cells by week 4; they populated the central part
of the scaffold fragment both on the surface and in the
depth (Fig. 3, a, b).
Thus, reinforcement of the collagen sponges
7-fold increased the mean Young’s modulus for the
hybrid scaffold without affecting its cytotoxicity. The
developed scaffold provides cell adhesion and proli-
feration along the reinforcement lines and can be used
for fabrication of tissue engineering constructs.
The study was supported by the Russian Science
Foundation (grant No. 16-15-00042; Development of
hybrid structures, cell experiments), Russian Founda-
tion for Basic research (grant No. 16-02-00248; physi-
cochemical and mechanical characterization of materi-
als, analysis of the cell morphology and distribution
by the method of confocal microscopy), and Federal
Agency for Scientic Organizations (FASO Russia)
(Agreement No. 007-GZ/Ch3363/26; analysis of cell
distribution by the method of confocal microscopy).
REFERENCES
1. Glybochko PV, Alyaev YuG, Shekhter AB, Vinarov AZ, Istra-
nov LP, Istranova EV, Aboyants RK, Lyundup AV, Krashenin-
nikov ME, Guller AE, Butnaru DV, Marisov LV, Kantimerov
DF, Kurkov AV, Safronova EI, Vorob’eva EA, Iritsyan MM.
PLGA mesh-collagen hybrid scaffold and tissue-engineered
product in substitution urethroplasty: experimental validation.
Urologiya. 2015;(6):5-13. Russian.
2. Timashev PS, Bardakova KN, Demina TS, Pudovkina GI,
Novikov MM, Markov MA, Asyutin DS, Pimenova LF, Svid-
chenko EA, Ermakov AM, Selezneva II, Popov VK, Kon-
ovalov NA, Akopova TA, Solov’eva AB, Panchenko VYa,
Bagrarashvili VN. A new biocompatible material based on
solid phase modified chitosan for laser stereolithography.
Sovremen. Tekhnol. Med. 2015;7(3):20-31. Russian.
3. Shekhter AB, Guller AE, Istranov LP, Istranova EV, Butnaru
DV, Vinarov AZ, Zakharkina OL, Kurkov AV, Kantimerov
DF, Antonov EN, Marisov LV, Glybochko PV. Morphology
of collagen matrices for tissue engineering (biocompatibility,
biodegradation, tissue response). Arkh. Patol. 2015;77(6):29-
38. Russian.
4. Cao X, Wang J, Liu M, Chen Y, Cao Y, Yu X. Chitosan-colla-
gen/organomontmorillonite scaffold for bone tissue engineer-
ing. Front. Mater. Sci. 2015;9(4):405-412.
5. Chan EC, Kuo SM, Kong AM, Morrison WA, Dusting GJ,
Mitchell GM, Lim SY, Liu GS. Three dimensional collagen
scaffold promotes intrinsic vascularisation for tissue engi-
neering applications. PLoS One. 2016;11(2):e0149799. doi:
10.1371/journal.pone.0149799.
6. Dong C, Lv Y. Application of сollagen scaffold in tissue en-
gineering: recent advances and new perspectives. Polymers.
2016;8(2):42. doi: 10.3390/polym8020042.
7. Grover CN, Cameron RE, Best SM. Investigating the morpho-
logical, mechanical and degradation properties of scaffolds
comprising collagen, gelatin and elastin for use in soft tissue
engineering. J. Mech. Behav. Biomed. Mater. 2012;10:62-74.
8. Kane RJ, Roeder RK. Effects of hydroxyapatite reinforcement
on the architecture and mechanical properties of freeze-dried
collagen scaffolds. J. Mech. Behav. Biomed. Mater. 2012;7:41-
49.
9. Lee H, Ahn SH, Kim GH. Three-dimensional collagen/al-
ginate hybrid scaffolds functionalized with a drug delivery
system (DDS) for bone tissue regeneration. Chem. Mater.
2011;24(5):881-891.
10. Liu X, Huang C, Feng Y, Liang J, Fan Y, Gu Z, Zhang X. Rein-
forcement of a porous collagen scaffold with surface-activated
PLA bers. J. Biomater. Sci. Polym. Ed. 2010;21(6-7):963-977.
11. Lu H, Oh HH, Kawazoe N, Yamagishi K, Chen G. PLLA-
collagen and PLLA-gelatin hybrid scaffolds with funnel-like
porous structure for skin tissue engineering. Sci. Technol. Adv.
Mater. 2012;13(6). ID 064210.
12. Timashev P, Kuznetsova D, Koroleva A, Prodanets N, Dei-
wick A, Piskun Y, Bardakova K, Dzhoyashvili N, Kostjuk S,
Zagaynova E, Rochev Y, Chichkov B, Bagratashvili V. Novel
biodegradable star-shaped polylactide scaffolds for bone re-
generation fabricated by two-photon polymerization. Nano-
medicine (Lond). 2016;11(9):1041-1053.
13. Velasco MA, Narváez-Tovar CA, Garzón-Alvarado DA. De-
sign, materials, and mechanobiology of biodegradable scaf-
folds for bone tissue engineering. Biomed. Res. Int. 2015;2015.
ID 729076. doi: 10.1155/2015/729076.
14. Vilay V, Todo M. Tensile deformation behavior of collagen/
PCL hybrid scaffolds for ligament tissue engineering. J. Solid
Mech. Mater. Eng. 2013;7(3):463-472.
K. N. Bardakova, E. A. Grebenik, et al.