Copyright © 2014 Cognizant Communication Corporation DOI: 10.3727/096368914X680082 CT-1081 Accepted 03/07/2014 for publication in “Cell Transplantation”
Characterization of liver-specific functions of human fetal hepatocytes in culture
Cinzia Maria Chinnicia, Francesca Timoneria, Giandomenico Amicoa, Giada Pietrosib, Giovanni Vizzinib, Marco Spadac, Duilio Paganoc, Bruno Gridellia,c, Pier Giulio Conaldia
a
Fondazione Ri.MED, Regenerative Medicine and Biomedical Technologies Unit, Department of Laboratory Medicine and Advanced Biotechnologies, ISMETT Via Tricomi 5, 90127 Palermo, Italy. b
Hepatology Unit, Department of Medicine, ISMETT, Via Tricomi 5, 90127 Palermo, Italy
c
Department of Surgery, ISMETT, Via Tricomi 5, 90127 Palermo, Italy
Running Header: Human Fetal Hepatocytes in Primary Cultures
Corresponding author: Cinzia Maria Chinnici, Ph.D Fondazione Ri.MED Via E. Tricomi 5 90127 Palermo, Italy Phone +39 091 21 92 496 Fax: +39 091 21 92 422 email: [email protected]
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
1
Copyright © 2014 Cognizant Communication Corporation ABSTRACT This study was designed to assess the metabolic activities of human fetal liver cells proposed as a potential source for hepatocyte transplantation. Fetal liver cells were isolated from livers of different gestational ages (16-22 weeks), and the functionality of the cell preparations was evaluated by establishing primary cultures. We observed that 20-22 week-gestation fetal liver cell cultures contained a predominance of cells with hepatocytic traits that did not divide in vitro, but were functionally competent. Fetal hepatocytes performed liver-specific functions at levels comparable to those of their adult counterparts. Moreover, exposure to dexamethasone in combination with oncostatin M promptly induced further maturation of the cells through the acquisition of additional functions (i.e., ability to store glycogen and up-take of indocyanine green). In some cases, particularly in cultures obtained from fetuses of earlier gestational ages (e.g., 18 weeks gestation), cells with mature hepatocytic traits proved to be sporadic, and the primary cultures were populated by clusters of proliferating cells. Consequently, the values of liver-specific functions detected in these cultures were low. Interestingly, we observed that a high cell-density culture system in combination with medium supplemented with dexamethasone and epidermal growth factor stabilized the hepatocytic monolayer for several weeks. This provides a useful in vitro model for studies requiring long-term culturing of human hepatocytes, and offers insight on how to preserve the functionality of hepatocytes. We found that human fetal liver cells can be cryopreserved without significant loss of viability and function, and evaluated up to 1 year in storage in liquid nitrogen. They might, therefore, be suitable for cell banking, and allow for the transplantation of large numbers of cells, thus improving clinical outcomes. Overall, our results indicate that fetal hepatocytes could be used as cell source for hepatocyte transplantation to treat liver failure and liver-based metabolic diseases.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
2
Copyright © 2014 Cognizant Communication Corporation Keywords: human fetal hepatocytes; primary cultures; liver function; cell transplant; cryopreservation.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
3
Copyright © 2014 Cognizant Communication Corporation
INTRODUCTION Human adult hepatocyte transplantation to treat some liver-based inborn errors of metabolism and liver failure entered clinical practice approximately 20 years ago (30,43) as an alternative to orthotopic liver transplantation (OLT). The procedure was shown to improve and/or partially correct metabolic defects (10,37,40). However, it was also shown to offer no sustained benefits in the majority of cases (38). A major challenge of hepatocyte transplantation is the limited supply of donor organs for isolating relevant numbers of good quality cells (40). Hepatocytes for transplantation are obtained from rejected organs for transplantation, such as steatotic livers, or those that have undergone a long cold ischemia time, but also from surgical leftovers after liver transplantation (31). Our group carried out a Phase I clinical study in which transplantation of freshly isolated human fetal liver cells (FLC) was used to bridge patients with end-stage liver disease to liver transplantation. The isolation procedure of FLCs, including a case report has been recently published (12). The procedure was found to be safe and also led to the improvement of clinical scores, thus confirming previous reports (22). In this study, fetal hepatocytes are proposed as an alternative source of cells for transplantation. Our aim was to more precisely evaluate phenotype and liver-specific functions of the FLCs obtained after collagenase perfusion of a human fetal liver, and determine whether they can support impaired or lost hepatic functions. With the exception of a few reports (13,20-22) there is little documentation in the literature of transplantation of human FLCs in clinical practice. In addition, extensive characterization of human liver cells of fetal origin is reported in only a few studies (7,24,28), which mainly focus on hepatic stem cell characterization (24), and cell immunophenotype (7), rather than on liver-specific functions of fetal hepatocytes. While the liver of adult organisms performs a multitude of functions related to metabolism and detoxification, the fetal liver is a transient site of hematopoiesis, and serves as a primary source of CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
4
Copyright © 2014 Cognizant Communication Corporation red blood cells up to the 24th week of gestation, while defined hepatic functions are fully acquired near birth (17,19). Erythroid cells have been shown to reside within the fetal liver parenchyma from mid-gestation to the time of birth, and a function of the fetal liver as a niche for maturation of erythroid cells has been suggested (16). As cells expressing hepatic markers begin to appear in second trimester human fetal livers (34), a considerable number of cells populating the fetal liver parenchyma are expected to be hematopoietic. Typically, hepatic functions are tested in vitro on primary cultures of hepatocytes obtained after collagenase digestion of the liver. Hepatocytes are plated onto collagen- or other extracellular matrix protein-coated dishes, and maintained in hormonally defined serum-free culture media (3,36). Because of the limited replicative capacity of the cells, and rapid loss of the hepatocytic phenotype, culturing of hepatocytes has proven to be difficult (2). However, the in vitro assessment of hepatic functions does not require long-term culturing of the cells (6,11,36) because some liver-specific functions can be evaluated just a few hours after plating. We focused our attention on the most representative hepatic functions, such as albumin secretion (a key marker of hepatic functioning), urea synthesis, and activity of cytochrome P450 (CYP) enzymes from the subfamily CYP3A, particularly CYP3A7 and CYP3A4 isoforms, which are known to be the most abundant cytochromes in the fetal and the adult liver, respectively. We also evaluated glucose 6-phosphatase (G6Pase) enzyme activity, glycogen storage, and uptake of the vital stain indocyanin green (ICG). We used commercially available human adult hepatocytes and human fetal liver mesenchymal-stromal cells (FL-MSCs) as positive and negative control, respectively. In addition to cell function, we investigated the proliferative capacity of fetal hepatocytes by culturing them in low cell-density. A critical issue during the transplantation of human FLCs is the limited number of hepatocytes obtained from a single donor fetus. Augmenting cell numbers would require pooling thawed cell suspensions from different donors to be transplanted in a single recipient, a strategy that is currently CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
5
Copyright © 2014 Cognizant Communication Corporation employed in adult-hepatocyte transplant programs (32,33). In view of this, we tested the effects of cryopreservation on cell viability, ability to adhere to culture dishes, and cell function. If cryopreservation proves successful it will allow for the banking of FLC suspensions to be used for programmed infusions of larger numbers of cells.
MATERIALS AND METHODS Tissue Procurement Human fetal livers were taken from 16 to 22 weeks gestation human fetuses obtained in cases of therapeutic abortions according to a protocol approved by ISMETT Institutional Research Review Board (IRRB). Data on gestational age are shown in Table 1.
Flow Cytometry Analysis To assess the immunophenotype of the different cell fractions in our FLC preparations, a flow cytometric analysis was performed as previously described (8). The primary antibodies used were: IgG1-FITC, IgG2-PE, CD45-FITC, CD34-PE, CD90-PE, Glycophorin A (Gly-A)-PE Ki-67-FITC, CD106-APC, EpCAM-FITC (all BD Pharmingen, Franklin Lakes, NJ), CK19-PE (Santa Cruz, Santa Cruz, CA), CK18-FITC (Abcam Cambridge, UK).
Cell Cultures and Media Composition FLCs were isolated, as previously described (12), by collagenase (NB 1 GMP grade; Serva, Electrophoresis GmbH, Heidelberg, Germany) perfusion of human fetal livers. Data on cell yield and viability are showed on Table 1. FLCs were plated onto collagen I-coated 6-well plates (BD CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
6
Copyright © 2014 Cognizant Communication Corporation Biosciences) in Kubota’s medium (KM; 45) containing 5% fetal bovine serum (FBS) (Lonza, Basel, Switzerland). After cell attachment (approximately 8 hours), cultures were switched to serum-free medium. KM reagent sources were: RPMI 1640, glutamine and antibiotics (all from Lonza), bovine serum albumin (BSA), nicotinamide, transferrin, hydrocortisone, 2-mercaptoethanol, selenium, zinc sulfate, oleic acid, linoleic acid, linolenic acid, and palmitic acid, palmitoleic acid (all from SigmaAldrich, St. Louis, MO), insulin (Humulin; Lilly, Indianapolis, IN). Furthermore, we used two different culture systems: a high cell-density (approximately 1.8 x 105 cells/ cm2), and a low cell-density system (approximately 0.6 x 105 cells/cm2). The low-cell density system was used to evaluate the proliferative activity of FLCs. We also cultured fetal hepatocytes in a differentiation medium (DM), which consisted of KM supplemented with dexamethasone (10-7 M; Hospira, Lake Forest, Illinois) and oncostatin M (OSM; 10 ng/ml; R&D Systems, Minneapolis, MN).
Isolation of Human Fetal Liver Mesenchymal Stromal Cells (FL-MSCs) from Primary Cultures of Fetal Hepatocytes After 10-15 days of culture, fetal hepatocytes dyed and spindle-shaped cells, termed fetal liver mesenchymal stromal cells (FL-MSCs), were harvested by trypsinization (0.05% trypsin 0.02% EDTA; Lonza), and transferred to untreated 6-well plastic dishes (SARSTEDT, Numbrecht, Germany).
Immunofluorescence Analysis For immunofluorescence staining, the cells were fixed according to the antibody manufacturer’s instructions and processed as previously described (8). The used primary antibodies were: anti-CK18
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
7
Copyright © 2014 Cognizant Communication Corporation (Abcam), anti-albumin, anti-alpha 1-antitrypsin (both from Sigma-Aldrich), anti-alpha fetoprotein (AFP), and anti-transferrin (both from Santa Cruz).
Immunocytochemistry Analysis Cells were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA), and incubated with methanol 3% H2O2 (both from Sigma-Aldrich) for 30 minutes to deactivate the endogenous peroxidase. After blocking with phosphate buffered saline (PBS; Lonza) 0.5% Tween 20 (Sigma-Aldrich) 3% BSA for one hour, cells were incubated with the primary antibody anti-factor VIII (gift) overnight in a humid chamber at 4°C. The cells were then stained with the secondary antibody using the Vectastain Elite ABC Kit (Vector Laboratories, Burlingame, CA), and incubated with 3,3’diaminobenzidine (DAB) chromogen substrate using the DAB Substrate Kit for peroxidase (Vector Laboratories).
In Vitro Functional Analysis Functional assays were carried out in both high and low density cultured cells. Cells were plated onto 6-well collagen I-coated dishes and cultured in 1 ml of KM. Culture supernatants were collected 1 hour, 6 hours and 24 hours after cell attachment, and used for albumin secretion and urea synthesis testing. For albumin secretion, samples were diluted 1:4 to fit the assay concentration range. CYP activity was performed 1 hour after cell attachment. The remaining functional assays (G6pase activity, glycogen storage and ICG uptake) were carried out after 48 hours of culturing. Plated human adult hepatocytes (Lonza), and cryopreserved human adult hepatocytes (Life Technologies, Carlsbad, CA) were used as positive control cells, while the FL-MSCs were used as negative control cells:
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
8
Copyright © 2014 Cognizant Communication Corporation 1. Albumin secretion and urea production The secretion of albumin was measured using the enzyme-linked immunosorbent assay (ELISA) Kit (Abnova, Heidelberg, Germany) according to manufacturer’s instructions. The absorbance was measured on a microplate reader (Model 680, Bio-Rad, Hercules, CA). Urea production was determined by blood urea nitrogen (BUN) test using an automated system (Dimension RxL Max, Siemens Healthcare Diagnostics, Tarrytown, NY). The results of both assays were normalized to the total number of attached cells and expressed as g/ml/106 cells/hour. Cells were detached by trypsinization and counted by Trypan Blue (Lonza) exclusion test. 2. Activity of cytochrome P450 (CYP assay) Activities of isoforms -3A7 and -3A4 of CYP3A were assessed by using the P450-Glo Assay (LuciferinPFBE and Luciferin-IPA, respectively; Promega, Madison, WI) according to manufacturer’s instructions for non-lytic CYP. Cultured cells were incubated with medium supplemented with pGlo substrates. One hour (Luciferin-IPA) and four hours (Luciferin-PFBE) after exposure, 25 l of medium and 25 l of luciferin detection reagent were added to a white opaque 96-well plate (Corning Inc., Corning, NY) and read on a luminometer (Glomax, Promega) after 20’ incubation at room temperature. To determine the inducible activity of CYP3A4, cells were treated with the specific inducer rifampicin (10 M for 48 hours; Sigma-Aldrich) (27). The results were normalized to the total number of attached cells and expressed as relative light units (RLU)/106 cells/hour. 3. Glucose-6-phosphatase (G6Pase) activity The activity of the enzyme G6Pase was assessed according to methods described by Sokal et al. (41), with minor modifications. Briefly, cultured cells were incubated with Tris-acetate buffer pH 6.5 containing 2.08 mM glucose-6-phosphate and 2.4 mM nitric lead (all from Sigma-Aldrich), for 20 minutes, at 37°C in a humidified atmosphere of 5% CO2. After rinsing in distilled water, cells were incubated with 1% ammonium sulfide (Sigma-Aldrich) for 15 seconds at room temperature to CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
9
Copyright © 2014 Cognizant Communication Corporation convert lead nitrate into brown lead sulfate, and rinsed carefully with distilled water. The reaction was evaluated under an inverted CKX41 microscope (Olympus) coupled with a DS-Fi1 camera and Digital Sight (Nikon) for image acquisition. 4. Periodic acid-Schiff (PAS) stain for glycogen accumulation Cultured cells were fixed with 4% paraformaldehyde for 10 minutes and incubated with 0.5% periodic acid (PA) (Sigma-Aldrich) solution for 5 minutes at room temperature. After washing with distilled water, cells were covered with Schiff reagent (Sigma-Aldrich) for 15-20 minutes at room temperature, followed by three washes with tap water. Cells were examined under an inverted microscope, and image acquisition was done as described in the G6Pase activity subsection, above. 5. Indocyanine green (ICG) uptake assay Cultured cells were rinsed carefully with sterile PBS, the medium switched to prewarmed KM 1 mg/ml ICG (Cardiogreen, Sigma-Aldrich), and incubated for 15 minutes at 37°C, in the dark, in an incubator with 5% CO2. After rinsing with PBS, cells were observed under an inverted microscope, and image acquisition was done as described in the G6Pase activity subsection, above.
In Vitro Maturation of Fetal Hepatocytes: Dexamethasone and Oncostatin M Treatment To induce maturation of fetal hepatocytes (18), high cell-density plated cells were cultured for 6-10 days in DM. Maturation of fetal hepatocytes was evaluated by microscopic observation and by comparing gene expression and functional activities of undifferentiated and differentiated cells.
Gene Expression: Reverse Transcription Polymerase Chain Reaction (RT-PCR)
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
10
Copyright © 2014 Cognizant Communication Corporation Total RNA extraction, cDNA synthesis and RT-PCR were performed as previously described (8). The primer sequences (Invitrogen) are listed in Table 2.
Cryopreservation/Thawing of FLC Suspensions The cryopreservation protocol for FLCs was based on the use of dimethyl sulfoxide (DMSO; SigmaAldrich), which has been shown to be the ideal cryoprotectant in most studies (42). William’s E-based Heparmed medium (Biochrom, Berlin, Germany) was supplemented with insulin, transferrin, and glucagon (ITG, Biochrom), 30% FBS, and 10% DMSO. We used a freezing protocol with a freezing density of 5 x 106 cells/ml, as described by Mitry et al. (33), with minor modifications. Cells were stored at -80°C overnight and then in liquid nitrogen for long-term storage. For thawing, cells were removed from the liquid nitrogen tank and immediately placed in a water bath at 37°C. Cells were suspended in an ice-cold thawing solution (Heparmed 5% FBS) at 1:20 dilution and washed twice at 50 x g, 4°C for 5 minutes. Viability of thawed FLCs was assessed by the Trypan blue exclusion method and compared to that of freshly isolated FLCs. Flow cytometry analysis was used to evaluate the immunophenotype of thawed FLCs. Thawed cells were also plated onto collagen I-coated 6-well dishes, and tested for their normal culture behavior and functions.
Plating Efficiency of FLCs Due to the predominance of erythroid GlyA+ cells in FLC suspensions, the calculation of plating efficiency was based on the number of CK18+ epithelial/hepatocytic cells, by counting the number of non-adherent CK18+ cells evaluated by flow cytometry. The result was expressed as a percentage of the initially plated CK18+ cells. We also evaluated the ability of thawed cells to form a compact hepatocytic monolayer.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
11
Copyright © 2014 Cognizant Communication Corporation Statistics Data analyses were performed by using SigmaStat for Windows. All data were expressed as average ± standard deviation (SD). Data from two different groups were compared using the Student’s t-test. Differences between the groups were considered significant at a p-value of < 0.05.
RESULTS Cell Yield, Viability and Plating Efficiency of Freshly Isolated FLCs The average viability of freshly isolated FLCs was 82.3 3.9%, with a cell yield of 0.9 x 109 0.39 cells, and a plating efficiency of 71.6 12.9% (mean SD representing 13 FLC samples) (Table 1).
Immunophenotype of Freshly Isolated FLC Suspensions We used early (16-19 weeks gestation; n = 6) and late (20-22 weeks gestation; n = 7) FLC samples. Significant differences between early and late FLCs were the high number of Ki-67+ proliferating cells and CK18+/CK19+ cells observed in early groups. The percentage of Ki-67+ cells was 57.3 4.6% in early vs. 3.7 4.7% in late FLCs (p
Characterization of liver-specific functions of human fetal hepatocytes in culture
Cinzia Maria Chinnicia, Francesca Timoneria, Giandomenico Amicoa, Giada Pietrosib, Giovanni Vizzinib, Marco Spadac, Duilio Paganoc, Bruno Gridellia,c, Pier Giulio Conaldia
a
Fondazione Ri.MED, Regenerative Medicine and Biomedical Technologies Unit, Department of Laboratory Medicine and Advanced Biotechnologies, ISMETT Via Tricomi 5, 90127 Palermo, Italy. b
Hepatology Unit, Department of Medicine, ISMETT, Via Tricomi 5, 90127 Palermo, Italy
c
Department of Surgery, ISMETT, Via Tricomi 5, 90127 Palermo, Italy
Running Header: Human Fetal Hepatocytes in Primary Cultures
Corresponding author: Cinzia Maria Chinnici, Ph.D Fondazione Ri.MED Via E. Tricomi 5 90127 Palermo, Italy Phone +39 091 21 92 496 Fax: +39 091 21 92 422 email: [email protected]
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
1
Copyright © 2014 Cognizant Communication Corporation ABSTRACT This study was designed to assess the metabolic activities of human fetal liver cells proposed as a potential source for hepatocyte transplantation. Fetal liver cells were isolated from livers of different gestational ages (16-22 weeks), and the functionality of the cell preparations was evaluated by establishing primary cultures. We observed that 20-22 week-gestation fetal liver cell cultures contained a predominance of cells with hepatocytic traits that did not divide in vitro, but were functionally competent. Fetal hepatocytes performed liver-specific functions at levels comparable to those of their adult counterparts. Moreover, exposure to dexamethasone in combination with oncostatin M promptly induced further maturation of the cells through the acquisition of additional functions (i.e., ability to store glycogen and up-take of indocyanine green). In some cases, particularly in cultures obtained from fetuses of earlier gestational ages (e.g., 18 weeks gestation), cells with mature hepatocytic traits proved to be sporadic, and the primary cultures were populated by clusters of proliferating cells. Consequently, the values of liver-specific functions detected in these cultures were low. Interestingly, we observed that a high cell-density culture system in combination with medium supplemented with dexamethasone and epidermal growth factor stabilized the hepatocytic monolayer for several weeks. This provides a useful in vitro model for studies requiring long-term culturing of human hepatocytes, and offers insight on how to preserve the functionality of hepatocytes. We found that human fetal liver cells can be cryopreserved without significant loss of viability and function, and evaluated up to 1 year in storage in liquid nitrogen. They might, therefore, be suitable for cell banking, and allow for the transplantation of large numbers of cells, thus improving clinical outcomes. Overall, our results indicate that fetal hepatocytes could be used as cell source for hepatocyte transplantation to treat liver failure and liver-based metabolic diseases.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
2
Copyright © 2014 Cognizant Communication Corporation Keywords: human fetal hepatocytes; primary cultures; liver function; cell transplant; cryopreservation.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
3
Copyright © 2014 Cognizant Communication Corporation
INTRODUCTION Human adult hepatocyte transplantation to treat some liver-based inborn errors of metabolism and liver failure entered clinical practice approximately 20 years ago (30,43) as an alternative to orthotopic liver transplantation (OLT). The procedure was shown to improve and/or partially correct metabolic defects (10,37,40). However, it was also shown to offer no sustained benefits in the majority of cases (38). A major challenge of hepatocyte transplantation is the limited supply of donor organs for isolating relevant numbers of good quality cells (40). Hepatocytes for transplantation are obtained from rejected organs for transplantation, such as steatotic livers, or those that have undergone a long cold ischemia time, but also from surgical leftovers after liver transplantation (31). Our group carried out a Phase I clinical study in which transplantation of freshly isolated human fetal liver cells (FLC) was used to bridge patients with end-stage liver disease to liver transplantation. The isolation procedure of FLCs, including a case report has been recently published (12). The procedure was found to be safe and also led to the improvement of clinical scores, thus confirming previous reports (22). In this study, fetal hepatocytes are proposed as an alternative source of cells for transplantation. Our aim was to more precisely evaluate phenotype and liver-specific functions of the FLCs obtained after collagenase perfusion of a human fetal liver, and determine whether they can support impaired or lost hepatic functions. With the exception of a few reports (13,20-22) there is little documentation in the literature of transplantation of human FLCs in clinical practice. In addition, extensive characterization of human liver cells of fetal origin is reported in only a few studies (7,24,28), which mainly focus on hepatic stem cell characterization (24), and cell immunophenotype (7), rather than on liver-specific functions of fetal hepatocytes. While the liver of adult organisms performs a multitude of functions related to metabolism and detoxification, the fetal liver is a transient site of hematopoiesis, and serves as a primary source of CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
4
Copyright © 2014 Cognizant Communication Corporation red blood cells up to the 24th week of gestation, while defined hepatic functions are fully acquired near birth (17,19). Erythroid cells have been shown to reside within the fetal liver parenchyma from mid-gestation to the time of birth, and a function of the fetal liver as a niche for maturation of erythroid cells has been suggested (16). As cells expressing hepatic markers begin to appear in second trimester human fetal livers (34), a considerable number of cells populating the fetal liver parenchyma are expected to be hematopoietic. Typically, hepatic functions are tested in vitro on primary cultures of hepatocytes obtained after collagenase digestion of the liver. Hepatocytes are plated onto collagen- or other extracellular matrix protein-coated dishes, and maintained in hormonally defined serum-free culture media (3,36). Because of the limited replicative capacity of the cells, and rapid loss of the hepatocytic phenotype, culturing of hepatocytes has proven to be difficult (2). However, the in vitro assessment of hepatic functions does not require long-term culturing of the cells (6,11,36) because some liver-specific functions can be evaluated just a few hours after plating. We focused our attention on the most representative hepatic functions, such as albumin secretion (a key marker of hepatic functioning), urea synthesis, and activity of cytochrome P450 (CYP) enzymes from the subfamily CYP3A, particularly CYP3A7 and CYP3A4 isoforms, which are known to be the most abundant cytochromes in the fetal and the adult liver, respectively. We also evaluated glucose 6-phosphatase (G6Pase) enzyme activity, glycogen storage, and uptake of the vital stain indocyanin green (ICG). We used commercially available human adult hepatocytes and human fetal liver mesenchymal-stromal cells (FL-MSCs) as positive and negative control, respectively. In addition to cell function, we investigated the proliferative capacity of fetal hepatocytes by culturing them in low cell-density. A critical issue during the transplantation of human FLCs is the limited number of hepatocytes obtained from a single donor fetus. Augmenting cell numbers would require pooling thawed cell suspensions from different donors to be transplanted in a single recipient, a strategy that is currently CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
5
Copyright © 2014 Cognizant Communication Corporation employed in adult-hepatocyte transplant programs (32,33). In view of this, we tested the effects of cryopreservation on cell viability, ability to adhere to culture dishes, and cell function. If cryopreservation proves successful it will allow for the banking of FLC suspensions to be used for programmed infusions of larger numbers of cells.
MATERIALS AND METHODS Tissue Procurement Human fetal livers were taken from 16 to 22 weeks gestation human fetuses obtained in cases of therapeutic abortions according to a protocol approved by ISMETT Institutional Research Review Board (IRRB). Data on gestational age are shown in Table 1.
Flow Cytometry Analysis To assess the immunophenotype of the different cell fractions in our FLC preparations, a flow cytometric analysis was performed as previously described (8). The primary antibodies used were: IgG1-FITC, IgG2-PE, CD45-FITC, CD34-PE, CD90-PE, Glycophorin A (Gly-A)-PE Ki-67-FITC, CD106-APC, EpCAM-FITC (all BD Pharmingen, Franklin Lakes, NJ), CK19-PE (Santa Cruz, Santa Cruz, CA), CK18-FITC (Abcam Cambridge, UK).
Cell Cultures and Media Composition FLCs were isolated, as previously described (12), by collagenase (NB 1 GMP grade; Serva, Electrophoresis GmbH, Heidelberg, Germany) perfusion of human fetal livers. Data on cell yield and viability are showed on Table 1. FLCs were plated onto collagen I-coated 6-well plates (BD CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
6
Copyright © 2014 Cognizant Communication Corporation Biosciences) in Kubota’s medium (KM; 45) containing 5% fetal bovine serum (FBS) (Lonza, Basel, Switzerland). After cell attachment (approximately 8 hours), cultures were switched to serum-free medium. KM reagent sources were: RPMI 1640, glutamine and antibiotics (all from Lonza), bovine serum albumin (BSA), nicotinamide, transferrin, hydrocortisone, 2-mercaptoethanol, selenium, zinc sulfate, oleic acid, linoleic acid, linolenic acid, and palmitic acid, palmitoleic acid (all from SigmaAldrich, St. Louis, MO), insulin (Humulin; Lilly, Indianapolis, IN). Furthermore, we used two different culture systems: a high cell-density (approximately 1.8 x 105 cells/ cm2), and a low cell-density system (approximately 0.6 x 105 cells/cm2). The low-cell density system was used to evaluate the proliferative activity of FLCs. We also cultured fetal hepatocytes in a differentiation medium (DM), which consisted of KM supplemented with dexamethasone (10-7 M; Hospira, Lake Forest, Illinois) and oncostatin M (OSM; 10 ng/ml; R&D Systems, Minneapolis, MN).
Isolation of Human Fetal Liver Mesenchymal Stromal Cells (FL-MSCs) from Primary Cultures of Fetal Hepatocytes After 10-15 days of culture, fetal hepatocytes dyed and spindle-shaped cells, termed fetal liver mesenchymal stromal cells (FL-MSCs), were harvested by trypsinization (0.05% trypsin 0.02% EDTA; Lonza), and transferred to untreated 6-well plastic dishes (SARSTEDT, Numbrecht, Germany).
Immunofluorescence Analysis For immunofluorescence staining, the cells were fixed according to the antibody manufacturer’s instructions and processed as previously described (8). The used primary antibodies were: anti-CK18
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
7
Copyright © 2014 Cognizant Communication Corporation (Abcam), anti-albumin, anti-alpha 1-antitrypsin (both from Sigma-Aldrich), anti-alpha fetoprotein (AFP), and anti-transferrin (both from Santa Cruz).
Immunocytochemistry Analysis Cells were fixed with 4% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA), and incubated with methanol 3% H2O2 (both from Sigma-Aldrich) for 30 minutes to deactivate the endogenous peroxidase. After blocking with phosphate buffered saline (PBS; Lonza) 0.5% Tween 20 (Sigma-Aldrich) 3% BSA for one hour, cells were incubated with the primary antibody anti-factor VIII (gift) overnight in a humid chamber at 4°C. The cells were then stained with the secondary antibody using the Vectastain Elite ABC Kit (Vector Laboratories, Burlingame, CA), and incubated with 3,3’diaminobenzidine (DAB) chromogen substrate using the DAB Substrate Kit for peroxidase (Vector Laboratories).
In Vitro Functional Analysis Functional assays were carried out in both high and low density cultured cells. Cells were plated onto 6-well collagen I-coated dishes and cultured in 1 ml of KM. Culture supernatants were collected 1 hour, 6 hours and 24 hours after cell attachment, and used for albumin secretion and urea synthesis testing. For albumin secretion, samples were diluted 1:4 to fit the assay concentration range. CYP activity was performed 1 hour after cell attachment. The remaining functional assays (G6pase activity, glycogen storage and ICG uptake) were carried out after 48 hours of culturing. Plated human adult hepatocytes (Lonza), and cryopreserved human adult hepatocytes (Life Technologies, Carlsbad, CA) were used as positive control cells, while the FL-MSCs were used as negative control cells:
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
8
Copyright © 2014 Cognizant Communication Corporation 1. Albumin secretion and urea production The secretion of albumin was measured using the enzyme-linked immunosorbent assay (ELISA) Kit (Abnova, Heidelberg, Germany) according to manufacturer’s instructions. The absorbance was measured on a microplate reader (Model 680, Bio-Rad, Hercules, CA). Urea production was determined by blood urea nitrogen (BUN) test using an automated system (Dimension RxL Max, Siemens Healthcare Diagnostics, Tarrytown, NY). The results of both assays were normalized to the total number of attached cells and expressed as g/ml/106 cells/hour. Cells were detached by trypsinization and counted by Trypan Blue (Lonza) exclusion test. 2. Activity of cytochrome P450 (CYP assay) Activities of isoforms -3A7 and -3A4 of CYP3A were assessed by using the P450-Glo Assay (LuciferinPFBE and Luciferin-IPA, respectively; Promega, Madison, WI) according to manufacturer’s instructions for non-lytic CYP. Cultured cells were incubated with medium supplemented with pGlo substrates. One hour (Luciferin-IPA) and four hours (Luciferin-PFBE) after exposure, 25 l of medium and 25 l of luciferin detection reagent were added to a white opaque 96-well plate (Corning Inc., Corning, NY) and read on a luminometer (Glomax, Promega) after 20’ incubation at room temperature. To determine the inducible activity of CYP3A4, cells were treated with the specific inducer rifampicin (10 M for 48 hours; Sigma-Aldrich) (27). The results were normalized to the total number of attached cells and expressed as relative light units (RLU)/106 cells/hour. 3. Glucose-6-phosphatase (G6Pase) activity The activity of the enzyme G6Pase was assessed according to methods described by Sokal et al. (41), with minor modifications. Briefly, cultured cells were incubated with Tris-acetate buffer pH 6.5 containing 2.08 mM glucose-6-phosphate and 2.4 mM nitric lead (all from Sigma-Aldrich), for 20 minutes, at 37°C in a humidified atmosphere of 5% CO2. After rinsing in distilled water, cells were incubated with 1% ammonium sulfide (Sigma-Aldrich) for 15 seconds at room temperature to CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
9
Copyright © 2014 Cognizant Communication Corporation convert lead nitrate into brown lead sulfate, and rinsed carefully with distilled water. The reaction was evaluated under an inverted CKX41 microscope (Olympus) coupled with a DS-Fi1 camera and Digital Sight (Nikon) for image acquisition. 4. Periodic acid-Schiff (PAS) stain for glycogen accumulation Cultured cells were fixed with 4% paraformaldehyde for 10 minutes and incubated with 0.5% periodic acid (PA) (Sigma-Aldrich) solution for 5 minutes at room temperature. After washing with distilled water, cells were covered with Schiff reagent (Sigma-Aldrich) for 15-20 minutes at room temperature, followed by three washes with tap water. Cells were examined under an inverted microscope, and image acquisition was done as described in the G6Pase activity subsection, above. 5. Indocyanine green (ICG) uptake assay Cultured cells were rinsed carefully with sterile PBS, the medium switched to prewarmed KM 1 mg/ml ICG (Cardiogreen, Sigma-Aldrich), and incubated for 15 minutes at 37°C, in the dark, in an incubator with 5% CO2. After rinsing with PBS, cells were observed under an inverted microscope, and image acquisition was done as described in the G6Pase activity subsection, above.
In Vitro Maturation of Fetal Hepatocytes: Dexamethasone and Oncostatin M Treatment To induce maturation of fetal hepatocytes (18), high cell-density plated cells were cultured for 6-10 days in DM. Maturation of fetal hepatocytes was evaluated by microscopic observation and by comparing gene expression and functional activities of undifferentiated and differentiated cells.
Gene Expression: Reverse Transcription Polymerase Chain Reaction (RT-PCR)
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
10
Copyright © 2014 Cognizant Communication Corporation Total RNA extraction, cDNA synthesis and RT-PCR were performed as previously described (8). The primer sequences (Invitrogen) are listed in Table 2.
Cryopreservation/Thawing of FLC Suspensions The cryopreservation protocol for FLCs was based on the use of dimethyl sulfoxide (DMSO; SigmaAldrich), which has been shown to be the ideal cryoprotectant in most studies (42). William’s E-based Heparmed medium (Biochrom, Berlin, Germany) was supplemented with insulin, transferrin, and glucagon (ITG, Biochrom), 30% FBS, and 10% DMSO. We used a freezing protocol with a freezing density of 5 x 106 cells/ml, as described by Mitry et al. (33), with minor modifications. Cells were stored at -80°C overnight and then in liquid nitrogen for long-term storage. For thawing, cells were removed from the liquid nitrogen tank and immediately placed in a water bath at 37°C. Cells were suspended in an ice-cold thawing solution (Heparmed 5% FBS) at 1:20 dilution and washed twice at 50 x g, 4°C for 5 minutes. Viability of thawed FLCs was assessed by the Trypan blue exclusion method and compared to that of freshly isolated FLCs. Flow cytometry analysis was used to evaluate the immunophenotype of thawed FLCs. Thawed cells were also plated onto collagen I-coated 6-well dishes, and tested for their normal culture behavior and functions.
Plating Efficiency of FLCs Due to the predominance of erythroid GlyA+ cells in FLC suspensions, the calculation of plating efficiency was based on the number of CK18+ epithelial/hepatocytic cells, by counting the number of non-adherent CK18+ cells evaluated by flow cytometry. The result was expressed as a percentage of the initially plated CK18+ cells. We also evaluated the ability of thawed cells to form a compact hepatocytic monolayer.
CT-1081 Cell Transplantation Epub; provisional acceptance 03/04/2014
11
Copyright © 2014 Cognizant Communication Corporation Statistics Data analyses were performed by using SigmaStat for Windows. All data were expressed as average ± standard deviation (SD). Data from two different groups were compared using the Student’s t-test. Differences between the groups were considered significant at a p-value of < 0.05.
RESULTS Cell Yield, Viability and Plating Efficiency of Freshly Isolated FLCs The average viability of freshly isolated FLCs was 82.3 3.9%, with a cell yield of 0.9 x 109 0.39 cells, and a plating efficiency of 71.6 12.9% (mean SD representing 13 FLC samples) (Table 1).
Immunophenotype of Freshly Isolated FLC Suspensions We used early (16-19 weeks gestation; n = 6) and late (20-22 weeks gestation; n = 7) FLC samples. Significant differences between early and late FLCs were the high number of Ki-67+ proliferating cells and CK18+/CK19+ cells observed in early groups. The percentage of Ki-67+ cells was 57.3 4.6% in early vs. 3.7 4.7% in late FLCs (p
