Research Article

Intramuscular transplantation of engineered hepatic tissue constructs corrects acute and chronic liver failure in mice Nalu Navarro-Alvarez1, , Alejandro Soto-Gutierrez1,à, Yong Chen2, Jose Caballero-Corbalan3, Wael Hassan4, Satoru Kobayashi5, Yoshitaka Kondo1, Masaya Iwamuro6, Kazuhide Yamamoto6, Eisaku Kondo7, Noriaki Tanaka1, Ira J. Fox8, Naoya Kobayashi1,* 1 Department of Surgery, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8558, Japan; Marion Bessin Liver Research Center, Albert Einstein College of Medicine of Yeshiva University, 1300 Morris Park Avenue, Ullmann 523, Bronx, NY 10461, USA; 3Departments of Oncology, Radiology & Clinical Immunology, Division of Clinical Immunology, The Rudbeck Laboratory, Uppsala University, Uppsala SE-751 85, Sweden; 4Department of Pathophysiology-Periodontal Science, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8558, Japan; 53-D Matrix Japan, Ltd., 3-2-4 Kojimachi, Chiyoda-ku, Tokyo 102-0083, Japan; 6 Department of Internal Medicine, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8558, Japan; 7Department of Pathology, Okayama University Graduate School of Medicine and Dentistry, 2-5-1 Shikata-cho, Okayama 700-8558, Japan; 8Department of Surgery, University of Pittsburgh, Pittsburgh, PA 15213, USA 2

See Editorial, pages 150–152

Background & Aims: Transplantation of isolated hepatocytes holds great promise as an alternative to whole organ liver transplantation. For treatment of liver failure, access to the portal circulation has significant risks and intrahepatic hepatocyte engraftment is poor. In advanced cirrhosis, transplantation into an extrahepatic site is necessary and intrasplenic engraftment is short-lived. Strategies that allow repeated extrahepatic infusion of hepatocytes could improve the efficacy and safety of hepatocyte transplantation for the treatment of liver failure. Methods: A non-immunogenic self-assembling peptide nanofiber (SAPNF) was developed as a three-dimensional scaffold and combined with growth factors derived from a conditionally immortalized human hepatocyte cell line to engineer a hepatic tissue graft that would allow hepatocyte engraftment outside the liver.

Results: The hepatic tissue constructs maintained hepatocytespecific gene expression and functionality in vitro. When transplanted into skeletal muscle as an extrahepatic site for engraftment, the engineered hepatic grafts provided life-saving support in models of acute, fulminant, and chronic liver failure that recapitulates these clinical diseases. Conclusions: SAPNF-engineered hepatic constructs engrafted and functioned as hepatic tissues within the muscle to provide life-sustaining liver support. These engineered tissue constructs contained no animal products that would limit their development as a therapeutic approach. Ó 2010 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.

Introduction Keywords: Hepatocytes; SAPNF; Extracellular matrix; Hepatocyte transplantation; Acute liver failure; Chronic liver failure; Hepatic tissue engineering. Received 1 February 2009; received in revised form 16 July 2009; accepted 20 July 2009; available online 8 December 2009 * Corresponding author. Address: Department of Gastroenterological Surgery, Transplant and Surgical Oncology, Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences, 2-5-1 Shikata-cho, Okayama 700-8558, Japan. Tel./fax: +81 86 235 7485. E-mail address: [email protected] (N. Kobayashi).   Present address: Bone Marrow Transplantation Section, Transplantation Biology Research Center, Massachusetts General Hospital, Harvard Medical, Boston, MA 02129, USA. à Present address: Department of Surgery, Center for Engineering in Medicine, Massachusetts General Hospital, Harvard Medical School and the Shriners Hospitals for Children, Boston, MA 02114, USA. Abbreviations: SAPNF, self-assembling peptide nanofiber; ELISA, enzyme-linked immunoabsorbent assay; ECM, extracellular matrix; HTX, hepatocyte transplantation; ALF, acute liver failure; FLF, fulminant liver failure; CLF, chronic liver failure; SCID/mice, severe combined immunodeficiency mice; 3-D, three-dimensional; OLT, orthotopic liver transplantation; CM, condition media; IM, intramuscular.

Orthotopic liver transplantation is considered the only definitive treatment for patients with liver failure [1,2]. Transplant surgery, however, is complex, risky, and costly. Transplantation of isolated hepatocytes holds great promise as an alternative to organ transplantation because it can be performed as a minimally invasive procedure [3,4]. In fulminant liver failure and liver-based metabolic diseases, the hepatic architecture is intact, and isolated hepatocytes can be transplanted into the liver. However, for patients with cirrhosis, the altered architecture of the liver is not considered an appropriate site for hepatocyte transplantation. Intrahepatic transplantation in this setting leads to poor engraftment, and risks worsening portal hypertension. Thus, transplantation into extrahepatic sites is necessary for treatment of most forms of chronic liver disease. Hepatocyte transplantation is not without risks. Access to the portal circulation through a percutaneous transhepatic route can

Journal of Hepatology 2010 vol. 52 j 211–219

Research Article lead to laceration of the liver and infusion of cells through the portal circulation can lead to donor hepatocyte embolization into the lungs, portal hypertension, and possible portal vein thrombosis. Thus, relatively few donor cells can be infused through the portal system at any one time. For patients with cirrhosis, the spleen is often considered the extrahepatic engraftment site of choice for hepatocyte transplantation, but with similar risks of bleeding [5], hepatocyte embolization to the lungs [6,7], and limitations in the cell number that can be transplanted. While other ectopic sites for hepatocyte engraftment have been examined, they have usually been associated with insufficient engraftment and short-term graft survival [8–16]. It is possible that engraftment at such sites may also be affected by the process of hepatocyte isolation. The extracellular matrix (ECM) of the liver provides hepatocytes with the tensile scaffold necessary for assembly into three-dimensional macroscopic structures and can serve as a reservoir for soluble and insoluble signaling molecules that play a key role in tissue maintenance and liver regeneration [17–19]. To overcome the effects of hepatocyte isolation, animal derived or synthetic extracellular matrices have been used to support cell growth and to maintain hepatocyte-specific activity in culture [20,21]. We examined the extent to which a non-immunogenic self-assembling peptide nanofiber (SAPNF) used in combination with conditioned media (CM) derived from an immortalized human hepatocyte cell line could provide a three-dimensional (3-D) scaffold that would allow hepatocytes to maintain their function in culture. We found that formation of this in vitro hepatic tissue construct re-established cell–cell and cell–matrix interactions and allowed maintenance of hepatocyte function equivalent to that associated with hepatocytes grown on Matrigel. Skeletal muscle is easily accessible, it has an abundant blood supply [22,23] and it has the capacity to accommodate large scale cell implantation [24]. We thus performed intramuscular (IM) transplantation of the SAPNF-derived hepatic tissue construct in newly developed, clinically relevant, mouse models of acute, fulminant, and irreversible end-stage liver failure. In these studies we demonstrated significant improvement in liver function and survival in animals transplanted with the SAPNF/CM-hepatic tissue constructs. Since the SAPNF components are immunologically inert and are degraded naturally over time to produce only amino acids that can be reused by the cells, this strategy has great potential for clinical application.

with 3.3 mg/ml BD Matrigel (Cat. No. 354234) or 0.5% (v/v) SAPNF (3-D Matrix Japan, Ltd., Tokyo, Japan), as previously described [19]. Hepatocytes were cultured in CM derived from CYNK-10 or in serum free Williams’ medium E (Sigma, St. Louis, MO) supplemented with 10 mM nicotinamide, 1% DMSO, 107 M Dexamethasone (Sigma), 5 mM Hepes (Gibco BRL), 100 U/ml penicillin, and 100 lg/ml streptomycin (Sigma) at 37 °C and 5% CO2. Evaluation of clearance and secretory capacity by engineered hepatic tissues At different time points, mouse hepatocytes, under the various culture conditions, were assessed for clearance activity. Ammonium sulfate (0.56 mM), lidocaine (1 mg/ml), and diazepam (1 mg/ml) were added to individual hepatic tissue cultures, at 2  105 hepatocytes/well and the amount of each substrate remaining in the medium after 24-h culture was measured. Ammonia concentration was determined using a Fuji Dri-Chem slide (Fuji) and lidocaine and diazepam were measured by SRL. Twenty-four-hours albumin secretion into the culture medium was assessed using a mouse albumin enzyme-linked immunosorbent assay (ELISA) kit (Shibayagi, Cat. No. AKRAL-121) according to the manufacturer’s instructions.

RT-PCR analysis Total RNA was extracted from hepatocytes under the various culture conditions using TRIzol (Invitrogen) reagent according to the manufacturer’s instructions. Reverse transcription (RT) was performed at 22 °C for 10 min, followed by 42 °C for 20 min, using specific primers with 1.0 lg of RNA per reaction as previously described [25]. For primer sequences, see Supplementary methods. PCR products were resolved on 1% agarose gels and visualized by ethidium bromide staining. Mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as an internal control for cDNA synthesis. Gene expression was determined based on the mRNA level relative to that obtained from normal adult mouse liver.

Animal models of fulminant and chronic liver failure Please see Supplementary methods. Hepatic tissue transplantation Hepatic tissues cultured for 3 days containing 5  106 mouse hepatocytes which is equivalent to 5% of the total mouse liver mass, were resuspended in 500 ll Matrigel or SAPNF, and injected in mice with liver failure into the muscles of both legs (Supplementary Fig. 1). Five hundred microliters of medium containing isolated hepatocytes cultured on ultra-low attachment plastic dishes for 4-h were used as controls. The experimental groups included: Group 1, Matrigel-hepatic tissue; Group 2, SAPNF/CM-hepatic tissue; Group 3, 4-h cultured hepatocytes; Group 4, saline. In mice with acute liver failure, 70% hepatectomy was performed just prior to transplantation. Body weight was measured and blood was collected weekly for plasma levels of ammonia, bilirubin, and glucose. Hepatic encephalopathy was measured weekly by means of a 6-point coma scale as previously described [26]. Histological tissue analysis

Materials and methods Preparation of conditioned medium (CM) derived from the CYNK-10 human hepatocyte cell line For human cell line establishment, see Supplementary methods. CYNK-10 cells were washed twice with PBS and cultured overnight in serum free Williams’ medium E (Sigma, St. Louis, MO) supplemented with 10 mM nicotinamide, 1% DMSO, 107 M Dexamethasone (Sigma), 5 mM Hepes (Gibco BRL), 100 U/ml penicillin, and 100 lg/ml streptomycin (Sigma) at 37 °C and 5% CO2. Conditioned medium (CM) was then collected and used after filtration through a 0.45-lm filter membrane (Sartorius, Hannover, Germany).

At the time of death or sacrifice, liver and muscle tissues containing hepatic tissue grafts were recovered, fixed in 20% formalin, and embedded in paraffin. Serial tissue sections were cut and stained with hematoxylin and eosin or Sirius Red for collagen deposition (Sigma–Aldrich, Direct Red 80, Cat. No. 36-5548). Sections were also prepared for immunofluorescence staining for albumin using a goat polyclonal antibody to albumin (Bethyl Laboratories, Cat. No. A80-229A) followed by a biotin anti-goat IgG (Histofine, Cat. No. 424012). Samples were examined under a confocal laser scanning microscope (LSM510, Carl Zeiss, Germany). Liver specimens from cirrhotic mice were prepared for collagen staining with Picro-Sirius Red. Statistical analyses

Hepatic tissue engineering For hepatocyte isolation, please see Supplementary methods. 2  105 hepatocytes were inoculated onto ultra-low attachment 6-well chambers (Corning), or cultured in 6-well plastic polystyrene plates (BD Falcon, Franklin Lakes, NJ) coated

212

Mean values are presented with standard deviations (SD). ANOVA was used to calculate the significance of difference in mean values. Multiple comparisons were done by Tukey method. The statistical analysis of survival time was done by the Kaplan–Meier survival test. A p value 0.5 mg/dl (normal range, 30 mg/ml) and an increase in plasma ammonia to >300 lg/dl (Normal range