Cell Transplantation, Vol. 15, pp. 921–927, 2006 0963-6897/06 $90.00 + .00Printed in the USA. All rights reserved. E-ISSN 1555-3892Copyright 2006 Cognizant Comm. Corp. www.cognizantcommunication.com

Self-Assembling Peptide Nanofiber as a Novel Culture Systemfor Isolated Porcine Hepatocytes

Nalu Navarro-Alvarez,* Alejandro Soto-Gutierrez,* Jorge D. Rivas-Carrillo,*Yong Chen,* Tsuyoshi Yamamoto,* Takeshi Yuasa,* Haruo Misawa,† Jiro Takei,‡

Noriaki Tanaka,* and Naoya Kobayashi*

*Department of Surgery, Okayama University Graduate School of Medicine and Dentistry, Okayama 700-8558, Japan†Department of Orthopedic Surgery, Okayama University Graduate School of Medicine and Dentistry, Okayama 700-8558, Japan

‡3-DMatrix Japan, Ltd., Tokyo 102-0083, Japan

Freshly isolated porcine hepatocytes are a very attractive cell source in the cell-based therapies to treat liverfailure because of unlimited availability. However, due to the loss of hepatocyte functions in vitro, there isa need to develop a functional culture system to keep the cells metabolically active. Here we compared theeffect of a self-assembling peptide nanofiber (SAPNF) as an extracellular matrix (ECM) with collagen typeI on hepatocyte metabolic and secretion activities following hepatocyte isolation. Isolated porcine hepato-cytes were cultured in SAPNF and collagen type I. Morphological assessment at different time points wasperformed by using SEM and phase contrast microscope. Metabolic and secretion activities were compara-tively performed in the groups, by means of ammonia, lidocaine, and diazepam as well as albumin. Hepato-cytes cultured on SAPNF revealed a three-dimensional spheroidal formation, thus maintaining cell differenti-ation status during 2 weeks of culture. On the other hand, hepatocytes in collagen revealed a spread shape,and by day 14 no hepatocyte-like cells were observed, but cells with long shape were present, thus revealinga degree of dedifferentiation in collagen culture. Hepatocytes in SAPNF were capable of drug-metabolizingactivities and albumin secretion in higher ratio than those cultured on collagen. The present work clearlydemonstrates the usefulness of SAPNF for maintaining differentiated functions of porcine hepatocytes inculture.

Key words: Porcine hepatocytes; Self-assembling peptide nanofiber; Extracellular matrix;Hepatocyte culture

INTRODUCTION hepatocytes has inevitable drawbacks, because the cellslose their polarity during the isolation procedure and thefunctions decrease rapidly in the subsequent in vitro cul-Mortality of acute liver failure (ALF) remains high

despite intensive care treatment (19). Thus, there is a ture (26). Importantly, hepatocytes are anchorage depen-dent and sensitive to environmental factors such as cellcompelling need to develop alternative therapies to treat

this life-threatening condition. Bioartificial liver (BAL) density and nature of the neighboring cells (9). There-fore, in the development of hepatocyte-based therapies,can be used in order to bridge the liver function until a

suitable donor organ is available for transplantation or the provision of an in vivo microenvironment where canbe attained cell-to-cell contact as well as cell-to-matrixthe patient’s own organ has recovered (7,16). Healthy

human hepatocytes are an ideal source for BAL; how- interactions is of extreme importance to maintain theliver-specific functions (26).ever, their relative scarcity is one of the major disadvan-

tages, compounded by the competing demands of ortho- Toward that goal, we have focused on the use of anovel biomaterial through molecular self-assemblingtopic liver transplantation (OLTX). Therefore, researchers

have focused on the use of porcine hepatocytes, because peptide nanofiber RADA-16 (arginine–alanine–asparticacid) (30), composed by four repeats of alternating hy-of unlimited availability, as a xenograft for temporary

support (20) and extracorporeal perfusion (4), as well a drophilic (aspartic acid negatively charged, and argi-nines positively charged) and hydrophobic amino acidsbiological component in BALs (22). However, indepen-

dent of the cell source, the use of primarily isolated (alanines) [COCH3]-RADARADARADARADA-[CONH2]

Received July 7, 2006; final acceptance August 3, 2006.Address correspondence to Naoya Kobayashi, M.D., Ph.D., Department of Surgery, Okayama University Graduate School of Medicine and Den-tistry, 2-5-1 Shikata-cho, Okayama, 700-8558, Japan. Tel: (+81) 86-235-7255; Fax: (+81) 86-235-7485; E-mail: immortal@md.okayama-u.ac.jp

921

922 NAVARRO-ALVAREZ ET AL.

(29). The peptide forms a nanofiber scaffold of about 10 hepatocytes were incubated in a humidified atmosphereof 5% CO2 and 95% air at 37°C with daily mediumnm in diameter with pores between 5 and 200 nm,

namely hydrogels with extremely high water content changes. In the following experiments, hepatocyte func-tions were evaluated per unit of time per milligram of[>99.5% (w/v) water]. The RAD motif may function in

a similar way to the ubiquitous integrin receptor binding cellular protein at each of time points. Cellular proteinconcentrations were determined by Multiscan JX (Thermosite RGD (arginine–glycine–aspartate) (28). Within this

framework, the cells are surrounded in a manner compa- Electron Co.,Yokohama, Japan).rable to the native extracellular matrix (29).

Here we report the development of a novel cell cul- Morphological Assessmentture system by promoting spheroid formation of porcine

During the culture time, morphological appearance ofhepatocyte using the self-assembling peptide nanofiber,porcine hepatocytes was observed using a phase contrastRADA-16, to enable the cells to recover from isolationmicroscope (Olympus CK40-SL Japan) and scanninginjury, restore their functions, and maintain differenti-electron microscope (SEM) (Hitachi S-2300, Hitachiated functions in vitro.Co. Ltd., Tokyo, Japan). For SEM, the samples werewashed with PBS followed by fixation with 2% glutaral-MATERIALS AND METHODSdehyde for 2 h at 37°C, and gently washed with PBS.Isolation and Culture of Porcine HepatocytesThe samples were then postfixed with osmium tetraox-

All procedures performed on the pigs were approvedide for 2 h and dehydration was accomplished using a

by the Okayama University Institutional Animal Caregraded series of ethanol (50%, 60%, 70%, 80%, 90%,

and Use Committee and thus within the guidelines ofand 99%). The samples were then dried at critical point

laboratory animals. Under general anesthesia, pigs (JAfor 2 h in absolute alcohol, mounted on aluminum stub,

West, Okayama, Japan) weighing 20 kg underwent up-and sputter-coated with gold before viewing under SEM.

per middle incision and the left lateral lobes, weighing80 g, were surgically removed for hepatocyte isolation.

Evaluation of Metabolic Capacity and AlbuminHepatocytes were isolated with a four-step retrogradeSecretion by Porcine Hepatocytesdispase/collagenase perfusion method, as previously re-

ported (18). Cell viability was assessed by a trypan blue For the measurement of metabolic capacity, 2 × 105

dye exclusion method. In the hepatocyte isolation and porcine hepatocytes were inoculated in each well of aculture experiments, the isolated hepatocytes with via- 12-well plate and ammonia-, lidocaine-, and diazepam-bilities of more than 90% were used. metabolizing activities of the hepatocytes per milligram

of cellular protein were evaluated over time in bothSelf-Assembling Peptide Nanofiber Preparation

groups. Ammonium sulfate (Nakarai Tesuku Co., Ky-Self-assembling peptide nanofiber (SAPNF) hydrogel oto, Japan), lidocaine (Fujisawa Pharmaceutical Co.,

1% (kindly provided by 3-DMatrix Japan Ltd., Tokyo, Osaka, Japan), or diazepam (Takeda Chemical Indus-Japan) was sonicated for 30 min before use and diluted tries, Osaka, Japan) was added to the culture medium atwith an equal volume of sterile water; 200 µl of 0.5% the final concentration of 0.56 mM, 1 mg/ml, or 1 µg/SAPNF was added to a well of 12-well plates (BD Bio- ml, respectively. The concentration of each reagent wassciences, San Jose, CA). Culture medium was added on measured 4 h later to compare the metabolic capacitiesthe top of SAPNF, which resulted in instantaneous gela- between the groups. Ammonia concentration was deter-tion of the material. The medium was changed twice mined using a FujiDri-Chem slide (FujiCo., Tokyo, Ja-every 15 min to equilibrate pH of the culture medium, pan). Concentrations of lidocaine and diazepam wereand SAPNF-coated culture well was incubated in a hu- measured by SRL Co. (Okayama, Japan). Twenty-four-midified atmosphere of 5% CO2 and 95% air at 37°C hour albumin secretion into the culture medium wasfor porcine hepatocyte culture. Collagen type I-coated measured over time per milligram of cellular protein byplates (Biocoat, Becton Dickinson) were used as a con- a pig albumin enzyme-linked immunoabsorbent assaytrol (Fig. 1). quantitation kit (Bethyl, Laboratories, Inc., Montgom-

ery, TX).Cell Seeding and Culture

Porcine hepatocytes were seeded in the wells of 12-Statistical Analyseswell SAPNF-coated or collagen-coated plates to a final

density of 2 × 105 per well. Williams’ E medium (Sigma, Mean values are presented with standard deviations(SDs). A two-tailed Student’s t-test was used to calcu-St. Louis, MO) supplemented with 10% fetal bovine se-

rum (Sigma), 100 U/ml penicillin, and 100 µg/ml strep- late the significance of difference in mean values. Avalue of p < 0.05 was considered statistically significant.tomycin (Sigma) was used in all the experiments. The

SELF-ASSEMBLING PEPTIDE NANOFIBER FOR ISOLATED PORCINE HEPATOCYTES 923

Figure 1. Schematic drawing of the experimental design. Hepatocytes were isolated from porcine livers and cultured in SAPNF orin collagen type I. They were maintained for 14 days, for an in vitro assessment of the functionality and morphology.

RESULTS hepatocytes in collagen type I culture never formedspheroid structures. The hepatocytes showed a flattenedSAPNF Facilitated Spheroidal Aggregationmorphology (Fig. 3C, D).of Porcine Hepatocytes

Freshly isolated porcine hepatocytes cultured onPorcine Hepatocytes in SAPNF Culture EfficientlySAPNF-coated plates started to aggregate and formedMetabolized Ammonia, Lidocaine, and Diazepamspheroids of compact morphology within 7 days (Fig. 2A,and Produced Albumin FavorablyB). The formed spheroids maintained their smooth, trans-

parent, and round shape for at least 14 days (Fig. 2C). Metabolic rates of ammonia, lidocaine, and diazepamOn the other hand, porcine hepatocytes cultured on of porcine hepatocytes per 24 h/mg of cellular protein

collagen type I-coated plates showed a flattened and ex- were comparatively analyzed between the groups (n =tended shape within 24 h (Fig. 2D) and then spread out, 3). Under collagen type I culture conditions, the hepato-and the appearance of cells with a long shape started to cytes rapidly lost their ability to perform differentiatedappear (Fig. 2E, F). functions. There was a significantly higher metabolism

of ammonia observed in porcine hepatocyte culture ofSEM Revealed That Porcine Hepatocytes in SAPNF SAPNF in comparison with the hepatocytes culturedCulture Developed Spheroid Structures with collagen type I (day 1: SAPNF 58.6 ± 2%, collagen

53.1 ± 2%; day 7: SAPNF 43.8 ± 2.2%, collagen 16.6 ±SEM at different times after cell seeding revealed thatporcine hepatocytes in SAPNF aggregated into increas- 1.5%; day 14: SAPNF 26.4 ± 3%; collagen 9 ± 1.3%)

(Fig. 4A). Metabolism of lidocaine was significantlyingly larger cell clumps that eventually formed intotightly packed spheroid structures of around 100 µm di- higher in porcine hepatocytes in SAPNF than the hepa-

tocytes in collagen type I culture (day 1: SAPNF 40 ±ameter, which is an adequate size to avoid central necro-sis of the cells (11) (Fig. 3A, B). In contrast, porcine 1.78%, collagen 38.8 ± 3.4%; day 7: SAPNF 31.8 ± 4%,

924 NAVARRO-ALVAREZ ET AL.

Figure 2. Morphological appearance of porcine hepatocytes. Morphological study showed that the porcine hepatocytes cultured inSAPNF maintained well-developed round-like appearance, starting with dispersed cells (A), which aggregated into doublets andtriplets within the first 24 h after inoculation. Larger aggregates appeared in the culture by day 2, reaching an average size of 100µm in diameter by day 7 (B, C). In contrast, collagen type I-cultured hepatocytes revealed a flattened appearance (D). Collagencultures on day 7 showed very few hepatocyte-like cells (E), and some cells with fibroblast-like morphology appeared thereafter(F). Original magnification: ×200. Scale bars: 50 µm; (A–C) and 20 µm (D–F).

collagen: 11.6 ± 1%; day 14: SAPNF 19.05 ± 3%, colla- ilarity to human hepatocytes. However, once the hepato-cytes are isolated from the liver, the cells start to losegen 8.1 ± 1.7%) (Fig. 4B). Diazepam was significantly

better metabolized in the porcine hepatocytes in SAPNF many of the differentiated functions in a considerablyshorter time, mainly due to the disrupting of cell-to-cellthan the hepatocytes in collagen type I culture (day 1:

SAPNF 39.9 ± 2.8%, collagen 37.2 ± 1.9%; day 7: or cell-to-matrix interactions (26). To engineer an ade-quate environment for hepatic function preservation inSAPNF 34.6 ± 4%, collagen 13.9 ± 2%; day 14: SAPNF

24.08 ± 3.5%, collagen 5.78 ± 1.7%) (Fig. 4C). vitro, we made use of a self-assembling peptide nanofi-ber, SAPNF, for porcine hepatocyte culture.The ability to produce albumin was maintained in the

porcine hepatocytes in SAPNF culture for up to 14 days Different scaffolds as extracellular matrices (ECM)for hepatocyte culture have been used. Some of them are(Fig. 4D), with a slight increase in the production on

day 5, due to the completion of spheroidal formation by made of synthetic polymers; however, while hepatocytesretained viability using synthetic materials, such asthe cells, where functions were increased. However,

after this, a slight decrease in albumin production was PLLA (poly-L-lactic acid) (14), PLGA (poly-lactic acid-co-glycolic acid) (10), and alginate (25), it has beenobserved, but constant levels remained until day 14. In

contrast, the capacity to produce albumin by the porcine demonstrated that the cells are not able to retain liver-specific gene expression for a long time (15). The fiberhepatocytes cultured in collagen type I was impaired in

a time-dependent manner, in which a dramatic decrease size of these materials is quite larger than the cells, thusfailing to mimic the in vivo environment of ECM. Inwas observed on day 7 and by day 14 no albumin was

detected. (Fig. 4D). addition, it has been reported that these materials some-times cause cell cytotoxicity and inflammatory responses

DISCUSSION (3). In contrast, the fiber size of SAPNF (10 nm) is smallenough for the cells to mimic in vivo ECM.The use of primary porcine hepatocytes is increas-

ingly favored for clinical application as a source of BAL Natural derived matrices for the maintenance of he-patic functions, such as the two-layer collagen gels (8)because of the unlimited availability and functional sim-

SELF-ASSEMBLING PEPTIDE NANOFIBER FOR ISOLATED PORCINE HEPATOCYTES 925

and Matrigel derived from Engelberth-Holm Swarm by using collagen type I (Fig. 2E, F). Our findings areconsistent with previous reports indicating that hepato-mouse sarcoma (1), have also been used. However, the

major limitations of these materials are host reaction and cyte cultures on collagen type I were maintained nolonger than 1 week (24).zoonosis. The animal origin of the products would make

their clinical application very difficult (12,15,21). On The maintenance of porcine hepatocyte functions inSAPNF could be explained by the fact that the influencethe other hand, the components of SAPNF can be natu-

rally degraded over time, producing only amino acids, of the hydrophilic/hydrophobic properties of SAPNFand the occurrence of charged functional chemicalwhich can be reused by the cells with no adverse effects.

Thus, its clinical application seems to be highly possi- groups could reconstitute the interactions of the cellswith these polypeptides (31,32). It is well known that able. In addition, SAPNF has been successfully tested in

animal experiments with no immunological reaction (6). variety of signals for the regulation of growth and differ-entiation of the cells can be provided by intercellularConsidering the clinical setting, one of the pivotal

functions of BAL is the ability to eliminate ammonia contacts, as well as the contact with adequate ECM(13,23). Despite the fact that the hepatocytes werefrom the patient’s blood to reduce the risk of encepha-

lopathy (17,22). Remarkably, porcine hepatocytes cul- highly functional during the first week of culture whenusing SAPNF, a slight decrease in function was ob-tured in SAPNF demonstrated a high degree of ammo-

nia-metabolizing capacity in vitro for the first 7 days served thereafter, but was still maintained up until 14days. We assumed that the diminishment of hepatocyte(Fig. 4A). Similar results were observed with lidocaine

and diazepam in the porcine hepatocytes cultured in functions in a time-dependent manner was partially dueto the mechanical disruption of SAPNF induced by me-SAPNF (Fig. 4B, C). In contrast, porcine hepatocytes

cultured in collagen type I failed to maintain the hepatic dium change. To reduce such a damage to the SAPNF,we are now investigating the use of a culture insert.functions even for a week. In addition, we observed the

presence of cells with long shape, fibroblast-like cells, Other possibilities for the decreased hepatic functionswould be culture stress on the cells and dedifferentiationprobably due to dedifferentiation of original hepatocytes

Figure 3. Structural examination of hepatocytes by SEM. Porcine hepatocytes cultured in SAPNFrevealed a well-developed round microscopic appearance at 24 h after inoculation (A) and formedspheroidal aggregation on day 7 (B). In contrast, hepatocytes cultured in collagen type I showedflattened microscopic appearance at 24 h (C). Then the cell number decreased and an overgrowthof fibroblast-like cells was observed by day 7 (D). Scale bars: 50 µm.

926 NAVARRO-ALVAREZ ET AL.

Figure 4. Ammonia and drug-metabolizing capacities and albumin-producing ability of porcine hepatocytes. Metabolic capacityper milligram of cellular protein was comparatively analyzed among the groups. Porcine hepatocytes cultured in SAPNF signifi-cantly increased ammonia metabolism (A) 4 h after loading ammonium sulfate in the culture medium at a final concentration of0.56 mM ammonia, in comparison with hepatocytes cultured in collagen type I (A). (B, C) After loading lidocaine (1 mg/ml) ordiazepam (1 µg/ml) in the culture medium, metabolic rates were compared 4 h later. Porcine hepatocytes cultured with SAPNFsignificantly metabolized both lidocaine (B) and diazepam (C), in higher ratio compared to collagen. (D) Albumin production wasalso significantly better in SAPNF-cultured hepatocytes than in collagen. The data are representative of three independent experi-ments. *p < 0.01.

cultured rat hepatocytes. Role of basement-membrane ma-of the cells (2). To overcome these issues, we can incor-trix. Mol. Biol. Med. 7:187–197; 1990.porate several molecules, such as a deleted variant of

2. Block, G. D.; Locker, J.; Bowen, W. C.; Petersen, B. E.;hepatocyte grown factor (dHGF) (5) or antiapoptotic Katyal, S.; Strom, S. C.; Riley, T.; Howard, T. A.; Micha-molecules such as V5 (27), in the SAPNF culture sys- lopoulos, G. K. Population expansion, clonal growth, and

specific differentiation patterns in primary cultures of he-tem, based on our previous investigations. Such strate-patocytes induced by HGF/SF, EGF and TGF alpha ingies are with the purpose of preservation of the naturala chemically defined (HGM) medium. J. Cell Biol. 132:platform provided by the SAPNF and enhancing the1133–1149; 1996.

functional capacities of the isolated hepatocytes. 3. Canaple, L.; Nurdin, N.; Angelova, N.; Saugy, D.; Hun-In summary, the present study demonstrates a new keler, D.; Desvergne, B. Maintenance of primary murine

hepatocyte functions in multicomponent polymer cap-tissue culture system that may provide a practical andsules—in vitro cryopreservation studies. J. Hepatol. 34:clinically applicable method to engineer a hepatic tissue11–18; 2001.using SAPNF as a base for BAL construction.

4. Chari, R. S.; Collins, B. H.; Magee, J. C.; DiMaio, J. M.;Kirk, A. D.; Harland, R. C.; McCann, R. L.; Platt, J. L.;REFERENCES Meyers, W. C. Brief report: treatment of hepatic failure

1. Bissell, D. M.; Caron, J. M.; Babiss, L. E.; Friedman, with ex vivo pig-liver perfusion followed by liver trans-plantation. N. Engl. J. Med. 331:234–237; 1994.J. M. Transcriptional regulation of the albumin gene in

SELF-ASSEMBLING PEPTIDE NANOFIBER FOR ISOLATED PORCINE HEPATOCYTES 927

5. Chen, Y.; Kobayashi, N.; Suzuki, S.; Soto-Gutierrez, A.; 18. Kunieda, T.; Maruyama, M.; Okitsu, T.; Shibata, N.;Takesue, M.; Totsugawa, T.; Kosaka, Y.; Arata, T.;Rivas-Carrillo, J. D.; Tanaka, K.; Navarro-Alvarez, N.;

Fukazawa, T.; Narushima, M.; Miki, A.; Okitsu, T.; Kobayashi, K.; Ikeda, H.; Oshita, M.; Nakaji, S.; Ohmoto,K.; Yamamoto, S.; Kurabayashi, Y.; Kodama, M.;Amemiya, H.; Tanaka, N. Transplantation of human hepa-

tocytes cultured with deleted variant of hepatocyte growth Tanaka, N.; Kobayashi, N. Cryopreservation of primarilyisolated porcine hepatocytes with UW solution. Cellfactor prolongs the survival of mice with acute liver fail-

ure. Transplantation 79:1378–1385; 2005. Transplant. 12:607–616; 2003.19. Lee, W. M. Acute liver failure. N. Engl. J. Med. 329:6. Davis, M. E.; Motion, J. P.; Narmoneva, D. A.; Taka-

hashi, T.; Hakuno, D.; Kamm, R. D.; Zhang, S.; Lee, 1862–1872; 1993.20. Makowa, L.; Cramer, D. V.; Hoffman, A.; Breda, M.;R. T. Injectable self-assembling peptide nanofibers create

intramyocardial microenvironments for endothelial cells. Sher, L.; Eiras-Hreha, G.; Tuso, P. J.; Yasunaga, C.;Cosenza, C. A.; Wu, G. D.; et al. The use of a pig liverCirculation 111:442–450; 2005.

7. Demetriou, A. A.; Brown, Jr., R. S.; Busuttil, R. W.; Fair, xenograft for temporary support of a patient with fulmi-nant hepatic failure. Transplantation 59:1654–1659; 1995.J.; McGuire, B. M.; Rosenthal, P.; Am Esch, 2nd, J. S.;

Lerut, J.; Nyberg, S. L.; Salizzoni, M.; Fagan, E. A.; 21. Mooney, D. J.; Sano, K.; Kaufmann, P. M.; Majahod, K.;Schloo, B.; Vacanti, J. P.; Langer, R. Long-term engraft-de Hemptinne, B.; Broelsch, C. E.; Muraca, M.; Salmeron,

J. M.; Rabkin, J. M.; Metselaar, H. J.; Pratt, D.; De La ment of hepatocytes transplanted on biodegradable poly-mer sponges. J. Biomed. Mater. Res. 37:413–420; 1997.Mata, M.; McChesney, L. P.; Everson, G. T.; Lavin,

P. T.; Stevens, A. C.; Pitkin, Z.; Solomon, B. A. Prospec- 22. Morsiani, E.; Pazzi, P.; Puviani, A. C.; Brogli, M.; Valieri,L.; Gorini, P.; Scoletta, P.; Marangoni, E.; Ragazzi, R.;tive, randomized, multicenter, controlled trial of a bioartif-

icial liver in treating acute liver failure. Ann. Surg. 239: Azzena, G.; Frazzoli, E.; Di Luca, D.; Cassai, E.; Lom-bardi, G.; Cavallari, A.; Faenza, S.; Pasetto, A.; Girardis,660–670; 2004.

8. Dunn, J. C.; Tompkins, R. G.; Yarmush, M. L. Long-term M.; Jovine, E.; Pinna, A. D. Early experiences with a por-cine hepatocyte-based bioartificial liver in acute hepaticin vitro function of adult hepatocytes in a collagen sand-

wich configuration. Biotechnol. Prog. 7:237–245; 1991. failure patients. Int. J. Artif. Organs 25:192–202; 2002.23. Nagaki, M.; Shidoji, Y.; Yamada, Y.; Sugiyama, A.;9. Dvir-Ginzberg, M.; Gamlieli-Bonshtein, I.; Agbaria, R.;

Cohen, S. Liver tissue engineering within alginate scaf- Tanaka, M.; Akaike, T.; Ohnishi, H.; Moriwaki, H.; Muto,Y. Regulation of hepatic genes and liver transcription fac-folds: Effects of cell-seeding density on hepatocyte viabil-

ity, morphology, and function. Tissue Eng. 9:757–766; tors in rat hepatocytes by extracellular matrix. Biochem.Biophys. Res. Commun. 210:38–43; 1995.2003.

10. Fiegel, H. C.; Havers, J.; Kneser, U.; Smith, M. K.; Moel- 24. Saavedra, Y. G.; Mateescu, M. A.; Averill-Bates, D. A.;Denizeau, F. Polyvinylalcohol three-dimensional matricesler, T.; Kluth, D.; Mooney, D. J.; Rogiers, X.; Kaufmann,

P. M. Influence of flow conditions and matrix coatings on for improved long-term dynamic culture of hepatocytes. J.Biomed. Mater. Res. A 66:562–570; 2003.growth and differentiation of three-dimensionally cultured

rat hepatocytes. Tissue Eng. 10:165–174; 2004. 25. Shapiro, L.; Cohen, S. Novel alginate sponges for cell cul-ture and transplantation. Biomaterials 18:583–590; 1997.11. Folkman, J.; Hochberg, M. Self-regulation of growth in

three dimensions. J. Exp. Med. 138:745–753; 1973. 26. Sivaraman, A.; Leach, J. K.; Townsend, S.; Iida, T.;Hogan, B. J.; Stolz, D. B.; Fry, R.; Samson, L. D.; Tan-12. Glicklis, R.; Shapiro, L.; Agbaria, R.; Merchuk, J. C.;

Cohen, S. Hepatocyte behavior within three-dimensional nenbaum, S. R.; Griffith, L. G. A microscale in vitro phys-iological model of the liver: Predictive screens for drugporous alginate scaffolds. Biotechnol. Bioeng. 67:344–

353; 2000. metabolism and enzyme induction. Curr. Drug Metab. 6:569–591; 2005.13. Hamilton, G. A.; Jolley, S. L.; Gilbert, D.; Coon, D. J.;

Barros, S.; LeCluyse, E. L. Regulation of cell morphology 27. Tanaka, K.; Kobayashi, N.; Gutierrez, A. S.; Rivas-Carrillo, J. D.; Navarro-Alvarez, N.; Chen, Y.; Narushima,and cytochrome P450 expression in human hepatocytes by

extracellular matrix and cell–cell interactions. Cell Tissue M.; Miki, A.; Okitsu, T.; Noguchi, H.; Tanaka, N. Pro-longed survival of mice with acute liver failure with trans-Res. 306:85–99; 2001.

14. Jiang, J.; Kojima, N.; Guo, L.; Naruse, K.; Makuuchi, M.; plantation of monkey hepatocytes cultured with an antiap-optotic pentapeptide V5. Transplantation 81:427–437;Miyajima, A.; Yan, W.; Sakai, Y. Efficacy of engineered

liver tissue based on poly-L-lactic acid scaffolds and fetal 2006.28. Yamada, K. M. Adhesive recognition sequences. J. Biol.mouse liver cells cultured with oncostatin M, nicotin-

amide, and dimethyl sulfoxide. Tissue Eng. 10:1577– Chem. 266:12809–12812; 1991.29. Yokoi, H.; Kinoshita, T.; Zhang, S. Dynamic reassembly1586; 2004.

15. Kaufmann, P. M.; Heimrath, S.; Kim, B. S.; Mooney, of peptide RADA16 nanofiber scaffold. Proc. Natl. Acad.Sci. USA 102:8414–8419; 2005.D. J. Highly porous polymer matrices as a three-dimen-

sional culture system for hepatocytes. Cell Transplant. 6: 30. Zhang, S. Fabrication of novel biomaterials through mo-lecular self-assembly. Nat. Biotechnol. 21:1171–1178;463–468; 1997.

16. Kjaergard, L. L.; Liu, J.; Als-Nielsen, B.; Gluud, C. Artifi- 2003.31. Zhang, S.; Holmes, T.; Lockshin, C.; Rich, A. Spontane-cial and bioartificial support systems for acute and acute-

on-chronic liver failure: A systematic review. JAMA 289: ous assembly of a self-complementary oligopeptide toform a stable macroscopic membrane. Proc. Natl. Acad.217–222; 2003.

17. Kundra, A.; Jain, A.; Banga, A.; Bajaj, G.; Kar, P. Evalua- Sci. USA 90:3334–3338; 1993.32. Zhang, S.; Holmes, T. C.; DiPersio, C. M.; Hynes, R. O.;tion of plasma ammonia levels in patients with acute liver

failure and chronic liver disease and its correlation with Su, X.; Rich, A. Self-complementary oligopeptide matri-ces support mammalian cell attachment. Biomaterials 16:the severity of hepatic encephalopathy and clinical fea-

tures of raised intracranial tension. Clin. Biochem. 38: 1385–1393; 1995.696–699; 2005.