Biomaterials 24 (2003) 2331–2337
Galactosylated chitosan as a synthetic extracellular matrix for hepatocytes attachment In-Kyu Parka, Jun Yangb, Hwan-Jeong Jeongc, Hee-Seung Bomd, Ichiro Haradab, Toshihiro Akaikeb, Su-Il Kima, Chong-Su Choa,* a
School of Agricultural Biotechnology, Seoul National University, 103 Serdun-dong, Kwonsun-gu, Suwon 441-744, South Korea b Department of Biomolecular Engineering, Tokyo Institute of Technology, Yokohama 226-8501, Japan c Department of Nuclear Medicine, Wonkwang University, Iksan 570-711, South Korea d Department of Nuclear Medicine, Chonnam National University Medical School, Gwangju 501-757, South Korea Received 29 January 2002; accepted 19 January 2003
Abstract Galactose moiety as the hepatocyte anchorage was covalently coupled with chitosan for the development of synthetic extracellular matrix. Hepatocytes adhesion to galactosylated chitosan (GC)-coated polystyrene (PS) dish became as high as 94.7% after 2 h incubation whereas the hepatocytes adhesion to chitosan-coated PS dish was 69.1%, indication of galactose-specific recognition between GC molecules and asialoglycoprotein receptors of hepatocytes. The DNA synthesis of the hepatocytes adhered to GCcoated dish was increased in the presence of epidermal growth factor (EGF) at low concentration of GC (0.05 mg/ml) whereas the DNA synthesis of the hepatocytes adhered to GC-coated dish was decreased in the presence of EGF at high concentration of GC (5 mg/ml). The spreading shapes of the hepatocytes adhered to the surface in the presence of EGF at low concentration of GC (0.05 mg/ml) were enhanced than in the absence of EGF. The hepatocytes adhered to the surface at high concentration of GC (5 mg/ ml) showed round shapes and exhibited many spheroid formation after 24 h in the presence of EGF. r 2003 Elsevier Science Ltd. All rights reserved. Keywords: Galactosylated chitosan; Extracellular matrix; Hepatocyte; Spheroid formation
1. Introduction Cellular growth and differentiation, in two-dimensional cell culture as well as in the three-dimensional space of the developing organism requires extracellular matrix (ECM) with which the cells can interact [1]. Proliferation and differentiation of hepatocytes are also regulated by the ECM. The natural ECMs such as collagen, laminin and fibronectin have been revealed to control spreading, migration, adhesion and proliferation of hepatocytes since they have several globular domains which are specialized for binding to the cell [2–3]. Synthetic ECMs replace many functions of the natural ECM, organizing cells into a three-dimensional architecture, providing mechanical integrity to the new tissue, and providing a space for the diffusion of *Corresponding author. Tel.: +82-31-290-2494; fax: +82-31-2962495. E-mail address:
[email protected] (C.-S. Cho).
nutrients and metabolites to and from the cell [4–5]. Alginates are naturally derived polysaccharides and have been extensively used as one of synthetic ECMs [6]. Akaike et al. reported that a galactosecarrying polystyrene (PS), poly(N-p-vinylbenzyl-4-o-bd-galactopyranosyl-d-gluconamide) (PVLA) as the artificial ECM regulated the proliferation, differentiation, and shapes of hepatocytes [7–9]. The behaviors such as morphology and differentiated functions of the hepatocytes on the artificial PVLA were distinct from those on natural ECM. The hepatocytes adhesion on the PVLA was mediated by the galactose-specific interactions between asialoglycoprotein receptors (ASGR) of the hepatocytes and galactose residues of the PVLA. Chitosan as a binary heteropolysaccharide containing [1–4] linked 2-acetamide-2-deoxy-b-d-glucopyranose and 2-amino-2-deoxy-b-d-glucopyranose residues [10] has characteristics similar to glycosaminoglycans and is biodegradable and non-toxic [11]. The chitosan
0142-9612/03/$ - see front matter r 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0142-9612(03)00108-X
2332
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
molecule has amino and hydroxyl groups which can be modified chemically. In this study, galactose moiety as the hepatocyte anchorage was covalently coupled with chitosan for the development of synthetic ECM which could control spreading, adhesion and proliferation of the hepatocytes. Hepatocyte anchorage is a strict requirement for survival of hepatocytes, and it orchestrates critical roles in many cellular functions including albumin secretion and ammonia elimination [12]. In previous studies, viability, spheroid formation and functions of hepatocytes of the alginate/galactosylated chitosan (GC) sponge were higher than those of the alginate one [13–14]. The emphasis of this study was placed on galactose moiety in the GC for the specific adhesive ligand to ASGR in the two-dimensional cell culture.
CH2 CH
n
HO O OH OH
OH
NH
CH2
O
HO
H OH H
O
OH
H H
OH Fig. 1. Chemical structure of PVLA.
for FTIR examination. 1H-NMR spectra were measured using AVANCE 600 spectrometer (Bruker, Germany). 2. Experimental 2.1. Materials Lactobionic acid (LA) was purchased from TCI (Tokyo, Japan). Chitosan (viscosity: 50B150 cP) and 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide (EDC) were purchased from Wako Pure Chemical Industrial, Ltd. (Tokyo, Japan). Biotin-labeled canavalia ensiformis lectin (Con A) and allomyrina dichotoma lectin (Allo A) were supplied by EY laboratories (CA, USA). N-hydroxylsuccinimide (NHS) was purchased from Sigma Chemical Co. (MO, USA). N,N,N0 N0 tetramethylethylenediamine (TEMED) was supplied by Aldrich (WI, USA). 2.2. Synthesis of GC and PVLA Coupling of LA with the chitosan was performed using EDC and NHS as coupling agents. Briefly, 2.3 g (6.35 mmol) of LA dissolved in 50 ml of TEMED/HCl buffer solution (pH 4.7) was activated with a mixture of NHS (0.14 g) and EDC (0.6 g). Subsequently, 2.2 g of chitosan (13.7 mmol) was added into the solution at an equivalent molar ratio to LA. The reaction was performed for 72 h at room temperature. The resulting product was purified using a dialysis tube (12,000 MWCO) against distilled water for 4 days, followed by lyophilization. The PVLA was prepared by the same method previously reported [15] and its structure was shown in Fig. 1. 2.3. Measurement of FTIR and NMR FTIR spectra were measured using M series (MIDAC Corporation) FTIR spectrometer. Dried samples were grounded with KBr powder and compressed into pellets
2.4. Lectin-enzyme assay Exactly 100 ml of 0.1 wt% GC was poured into 96-well PS plate overnight and blocked with 0.1 wt% BSA dissolved in WE medium for 30 min at room temperature. The plate was incubated with 50 ml of 1 mg/ml biotin-labeled Con A or Allo A in 0.9% NaCl solution for 1 h at room temperature, and then washed 5 times with 0.1 wt% Tween 20 dissolved in 0.9 wt% NaCl (NaCl-T). Then, 1/5000 diluted peroxidase-conjugated streptavidin (Vector Laboratories, CA, USA) in 0.1 wt% BSA/NaCl was poured into the wells followed by 1 h incubation at room temperature. After washing 7 times with NaCl-T, orthophenylenediamine (OPD) solution (0.4 mg/ml OPD and 0.4 ml H2O2 in citratephosphate buffer, pH 5.0) was used as a substrate of peroxidase. Absorbency was measured at 492 nm by a microplate reader (MTP-32, Corona Electric Co., Ibaraki, Japan). 2.5. Hepatocyte attachment Primary hepatocytes were isolated from ICR mouse (5–7 weeks old, male) by the method of in situ collagenase perfusion method as described by Seglen [16]. Cell viability measured by Trypan Blue exclusion was about 90%. The viable primary hepatocytes were suspended in Williams’ E medium (Gibco BRL), containing antibiotics (50 mg penicillin/ml, 50 mg streptomycin/ml and 100 ml neomycin/ml), HEPES (18 mm), epidermal growth factor (EGF, 20 ng EGF/ml) and insulin (100 nm). Isolated hepatocytes having an initial viability of 90% checked by trypan blue exclusion were suspended in GC-coated 24-well PS dish at 5 104 cells/ ml in Williams’ E medium without serum. The cultures were incubated in a humidified air/CO2 incubator (95/5,
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
2333
v/v). After a prescribed time, the medium including free non-adhered cells was thoroughly washed with PBS solution. The number of collected free cells was counted using a Coulter counter. 2.6. DNA synthesis A 8 103 cells/0.1 ml of isolated hepatocytes were placed into 96-well PS dish precoated with GC and incubated at 37 C in a humidified air/CO2 (95/5 vol%) incubator for 48 h. Then, DNA synthesis as the quantitative assay of the cell proliferation was performed by the incorporation of 5-bromo-20 -deoxyuridine (BrdU) [17]. The incorporation of BrdU was measured by the cell proliferation ELISA system version 2.
5
4
3
2
(a)
3. Results and discussion ppm
3.1. Synthesis of GC
5.0
4.5
4.0
3.5
3.0
2.5
2.0
(b) Fig. 3. 1H NMR spectra of chitosan (a) and GC (b).
100 90
Hepatocyte Adhesion (%)
GC was prepared by the similar method previously reported [14]. The synthesis scheme was shown in Fig. 2. From measurement of FTIR spectra of GC, the carbonyl stretching of LA disappeared due to the amide bond formation between carboxylic groups of LA and amine groups of chitosan. All peaks of amides I and II of GC slightly shifted from 1664 and 1594 to 1640 and 1572 cm1, respectively, when compared with those of chitosan, an indication of the conformational change of
80 70 60 50
(a) (b) (c)
40 10 0 0
20
40
60
80
100
120
Time (min) Fig. 4. Hepatocyte adhesion onto chitosan- (a), GC- (b), and PVLA(c) coated PS dishes against time at 37 C.
chitosan after reaction with LA (data not shown). Fig. 3 shows NMR spectra of chitosan (a) and GC (b). The substitution values of LA in GC was calculated by comparing the characteristic peak areas of LA group (4.1 ppm) with that of 2.0 ppm peak attributed to the original acetamide group of chitosan. The substitution values of LA coupled with chitosan in GC was estimated by 12.8 mol%. 3.2. Hepatocyte attachment
Fig. 2. Synthesis scheme of GC.
Fig. 4 shows hepatocytes adhesion into chitosan- (a), GC- (b) and PVLA-coated (c) PS dishes against time at 37 C. The results indicated that hepatocytes adhesion to
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
the GC-coated PS dish became as high as 94.7% after 120 min incubation, which was almost similar to PVLAcoated PS dish. On the other hand, hepatocytes adhesion to chitosan-coated PS dish was low (about 69.1%). It was already reported that hepatocytes adhesion to PVLA-coated dish was galactose-specific recognition between PVLA molecules and the ASGR of hepatocytes [7]. Therefore, it can be said that hepatocytes adhesion to GC-coated dish was galactose-specific recognition between GC molecules and ASGR of hepatocytes. The interaction between ASGR and galactose ligands requires Ca2+ ions. We examined the effect of divalent cations such as Ca2+ and Mg2+ on hepatocyte adhesion to dishes, which was shown in Fig. 5. It was found that the hepatocyte adhesion to GC-coated dishes was not dependent on the presence or absence of Ca2+ ions due to the non-specific interaction between cationic ions of chitosan and anionic ions of hepatocyte cell membrane although effective hepatocyte adhesion onto GC-coated dishes could not be induced by Mg2+ ions alone. Also, inhibition of hepatocyte adhesion onto GC-coated dishes was detected in the presence of PVLA as shown in Fig. 5, indication of specific interaction between galactose moieties of GC and ASGR of the hepatocytes. It was reported that the galactose chains can be recognized by the Allo A lectin as the carbohydrate binding protein [18]. The binding of GC with Allo A (or Con A) was shown in Fig. 6. The results indicated that GC-coated dish bound with Allo A three times than Con A although there was not much difference of binding between Allo A and Con A in the chitosancoated dish, indicating that the lectin Allo A specifically bound to galactose moieties in the GC. 3.3. Proliferation of hepatocyte DNA synthesis as the index of hepatocyte proliferation was shown in Fig. 7. The results indicated that the
Hanks (-)
2+
Hanks (Ca )
2+
Hanks (Mg )
GC/PVLA 0
20
40
60
80
100
Adhered cells (%)
Fig. 5. The effect of divalent cations (Ca2+ and Mg2+) on hepatocyte adhesion to dishes.
0.40
Allo A Con A
0.35 0.30 0.25
OD492nm
2334
0.20 0.15 0.10 0.05 0.00
Chitosan
GC
Fig. 6. Direct lectin-enzyme assay of galactose residues on coated dishes. Con A and Allo A were used as lectin specificity for glucose and galactose, respectively. The value was an average of six tests.
GC ( 0.05 µg/ml ) (-) EGF GC ( 0.05 µg/ml ) (+) EGF GC ( 5 µg/ml ) (-) EGF GC ( 5 µg/ml ) (+) EGF 0.0
0.1
0.2
0.3
0.4
DNA Synthesis
Fig. 7. DNA synthesis of hepatocytes cultured on GC (0.05 or 5 mg/ ml)-coated PS surface after 72 h. Concentration of EGF was 50 ng/ml. EGF () and EGF (+) represent without EGF and with EGF, respectively.
DNA synthesis of the hepatocytes adhered to GCcoated dish was increased in the presence of EGF at low concentration of GC (0.05 mg/ml) whereas the DNA synthesis of the hepatocytes adhered to GC-coated dish was decreased in the presence of EGF at high concentration of GC (5 mg/ml). It is reasonable to assume that the DNA synthesis as a measure of proliferation is dependent on the shape of adhered hepatocytes. We assume that the hepatocytes adhered at the lower galactose densities exhibited the higher DNA synthesis under the spreading morphology, and the lower DNA synthesis under the round morphology. Also, the spreading morphology of adhered hepatocytes at the lower galactose densities was enhanced in the presence of EGF, and the spheroid formation of adhered hepatocytes at the higher galactose densities was increased in the presence of EGF. The morphology
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
of the adhered hepatocytes will be discussed in the following part. 3.4. Morphology of adhered hepatocytes Fig. 8 shows phase-contrast microphotographs of hepatocytes adhered to GC-coated dish at low concentration of GC (0.05 mg/ml). Cells adhered to the surface showed spreading shapes after 24 h in the absence of EGF whereas cells adhered to the surface showed spreading shapes after 10 h in the presence of EGF, indicating that the spreading shapes of adhered hepatocytes in the presence of EGF were enhanced than in the absence of EGF. Fig. 9 shows phase-contrast microphotographs of hepatocytes adhered to GC-coated dish at high concentration of GC (5 mg/ml). Cells adhered to the surface
EGF(-)
2335
showed round shapes and exhibited spheroid formation after 24 h in the absence of EGF whereas cells adhered to the surface showed round shapes and exhibited many spheroid formations after 24 h in the presence of EGF. It was reported that shapes of the hepatocytes were regulated by changing the coating densities of the ECM such as laminin, fibronectin and collagen [3]. At low coating densities, the hepatocytes had a round morphology. At their higher coating densities, the hepatocytes had a spreading morphology, which is the integrinmediated recognition mechanism. On the other hand, the hepatocytes adhesion to GC-coated dish is galactose-specific recognition between GC and the ASGR of hepatocytes. Yang et al. already reported that differentiated functions of hepatocyte in alginate/GC sponges were more closely related to spheroid formation rather than to the hepatocytes adhesion [13]. Therefore, it has
EGF(+)
(a)
(b)
(c)
(d) Fig. 8. Phase-contrast microphotographs of hepatocytes adhered to GC-coated dish at low concentration of GC (0.05 mg/ml); (a) 3 h, (b) 10 h, (c) 24 h, and (d) 48 h. EGF () and EGF (+) represent without EGF and with EGF, respectively.
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
2336
EGF(-)
EGF(+)
(a)
(b)
(c)
(d) Fig. 9. Phase-contrast microphotographs of hepatocytes adhered to GC-coated dish at high concentration of GC (5 mg/ml); (a) 3 h, (b) 10 h, (c) 24 h, and (d) 48 h. EGF () and EGF (+) represent without EGF and with EGF, respectively.
been proposed that galactose moieties of GC and cationic groups of chitosan enhance hepatocyte spheroid formation because spheroid formation of hepatocytes was observed in the chitosan-coated dish in the presence of EGF after 24 h although the cells adhered to the chitosan-coated surface showed spreading shapes in the absence of EGF after 24 h (data not shown).
Acknowledgements
4. Conclusions
References
GC obtained by reaction of LA as galactose moieties with chitosan had the excellent adhesion and spheroid formation of hepatocytes due to the galactose-specific recognition between GC molecules and ASGR of hepatocytes. Therefore, GC has potency as one of synthetic ECMs for liver tissue engineering.
This work was supported by the grant of Nuclear Energy R&D Program (M20203200028-02A0702-00411) from the Ministry of Science and Technology of Korea.
[1] Olsen BR. Matrix molecules and their ligands. In: Lanza R, Langer R, Chick W, editors. Principles of tissue engineering, Austin: Academic Press, 1997. [2] Sawada N, Tomomura A, Sattler CA, Sattler GL, Kleinman HK, Pitot HC. Exp Cell Res 1986;167:458. [3] Mooney D, Hansen L, Vacanti J, Langer R, Farmer S, Ingber D. Switching from differentiation to growth in
I.-K. Park et al. / Biomaterials 24 (2003) 2331–2337
[4] [5]
[6] [7]
[8]
[9]
[10]
hepatocytes: control by extracellular matrix. J Cell Physiol 1992;151:497. Putnam AJ, Mooney D. Tissue engineering using synthetic extracellular matrices. Nat Med 1996;2:824. Shoichet MS, Li RH, White ML, Winn SR. Stability of hydrogels used in cell encapsulation: an in vitro comparison of alginate and agarose. Biotechnol Bioeng 1996;50:374. Smidsr^d O, Skjak-Br!ck G. Alginate as immobilization matrix for cells. TIBTECH 1990;8:71. Kobayashi A, Goto M, Kobayashi K, Akaike T. Receptormediated regulation of differentiation and proliferation of hepatocytes by synthetic polymer model of asialoglycoprotein. J Biomater Sci Polym Ed 1994;6:325. Kobayashi K, Kobayashi A, Tobe S, Akaike T. Carbohydratecontaining polystyrenes. In: Lee YC, Lee RT, editors. Neoglycoconjugates: preparation and applications. New York: Academic Press, 1994. p. 261–84. Kobayashi K, Kobayashi A, Akaike T. Culturing hepatocytes as on lactose-carrying polystyrene layer asialoglycoprotein receptormediated interactions. Method Enzymol 1994;247:409. Draget KI, Varum KM, Moen E, Gynnild H, Smidsrod O. Chitosan crosslinked with Mo[VI] polyoxyanions: a new gelling system. Biomaterials 1992;13:635.
2337
[11] Muzzarelli R, Baldassare V, Conti F, Ferrara P, Biagini G, Gazzanelli G, Vasi V. Biological activity of chitosan: ultrastructural study. Biomaterials 1988;9:247. [12] Reid LM. Stem cell biology, hormone/matrix synergies and liver differentiation. Curr Opin Cell Biol 1990;2:121. [13] Yang J, Chung TW, Nagaoka M, Goto M, Cho CS, Akaike T. Hepatocyte-specific porous polymer-scaffolds of alginate/galactosylated chitosan sponge for liver-tissue engineering. Biotechnol Lett 2001;23:1385. [14] Chung TW, Yang J, Akaike T, Cho KY, Nah JW, Kim SI, Cho CS. Preparation of alginate/galactosylated chitosan scaffold for hepatocyte attachment. Biomaterials 2002;23:2827. [15] Kobayashi A, Sumitomo H, Ina Y. Synthesis and functions of polystyrene derivatives having pendant oligosaccharides. Polym J 1985;17:567. [16] Seglen PO. Preparation of isolated rat liver cells. Meth Cell Biol 1976;13:23–83. [17] Muir D, Varon S, Manthorpe M. An enzyme-linked immunosorbent assay for bromodeoxyuridine incorporation using fixed microcultures. Anal Biochem 1990;185:377. [18] Lee RT, Meyers RW, Lee YC. Further studies on the binding characteristics of rabbit liver galactose/N-acetylgalactosaminespecific lectin. Biochemistry 1982;21:6292–8.