Abstract
We previously reported that cell-based therapies using isolated hepatocytes including hepatocyte transplantation and liver tissue engineering approaches provide therapeutic benefits to hemophilia. For clinical application of these approaches, it is important to establish an active hepatocyte proliferation system that enables providing a sufficient number of hepatocytes. We also reported that human hepatocytes, which were transplanted into the liver of urokinase-type plasminogen activator transgenic severe combined immunodeficiency (uPA/SCID) mice, were able to proliferate while retaining their ability to produce coagulation factor IX. The objective of this study was to explore the functionalities of other coagulation and anticoagulation factors of the propagated human hepatocytes in uPA/SCID mice. Human hepatocytes were transplanted into the liver of uPA/SCID mice, and the propagation status of human hepatocytes in the mice was monitored by the increase in serum human albumin levels and immunohistochemical evaluation on the liver sections. Using uPA/SCID livers with various stages of human hepatocyte propagation, we analyzed the gene expression levels of coagulation factors (prothrombin, factor VII, factor X, and factor VIII) and anticoagulation factors (protein C and protein S) by real-time polymerase chain reaction (PCR) using human-specific primers. As a result, the total amount of raw messenger RNA expression levels increased in all genes analyzed according to the progress of hepatocyte propagation and proliferation. Except for factor VIII, the gene expression levels of the highly repopulated uPA/SCID mouse livers with human hepatocyte showed higher levels than those of normal human livers, indicating that propagated human hepatocytes in the uPA/SCID system possess full functions to produce most of the coagulation-related factors. The current work demonstrated that human hepatocytes can be propagated in experimental animals while maintaining normal gene expression levels of coagulation-related factors. It could be speculated that the propagated cells serve as a cell source for the treatment of various types of coagulation factor deficiencies.
Keywords
Introduction
Production of coagulation and anticoagulation factors is one of the important functions of the liver, and most of these coagulation-related factors are produced by hepatocytes (4,6,30,31,33). There are various types of congenital bleeding disorders that lack a production of coagulation factor in the liver, showing a symptomatic bleeding tendency. Hemophilia A or B is well known as a representative bleeding disorder, which is caused by a failure in the production of functional coagulation factor VIII (FVIII) or factor IX (FIX) from the liver. Although the ultimate cure for hemophilia patients could be obtained by liver transplantation (10,13,14), world-wide donor organ shortage is the most critical obstacle. For patients with hemophilia and other congenital coagulation factor deficiencies, the elevation of the responsible factor level to 1–2% of normal levels can provide a phenotypic change from severe to moderate form, resulting in a marked improvement in the symptomology and the quality of life (5). Cell-based approaches using isolated hepatocytes could be a feasible therapeutic option toward these coagulation factor deficiencies (18,20,21,24).
Proof-of-concept studies for hepatocyte-based approaches have been accomplished in both laboratory animals and humans. We recently reported that hepatocyte transplantation provided an increase of 1–2% of coagulation activities in a mouse model of hemophilia B, FIXknockout mice (31). We also reported that engineering functional liver tissues beneath the kidney capsule were able to provide therapeutic effects in the mouse model of hemophilia A as well as hemophilia B (20,21). As reported by Dhawan et al. (7), hepatocyte transplantation was also successful in the clinic for the treatment of congenital factor VII deficiency. In utero liver cell transplantation was also investigated by Rosen et al. (25). They described phenotypic improvements in the mouse model of factor × deficiency. It is important to note that these hepatocyte-based approaches could be employed with a simple procedure and in a less invasive manner compared with organ transplantation (18,22).
One of the major hurdles in establishing hepatocyte-based approaches is the limited availability of biologically functional hepatocytes. At present, the number of donor livers for hepatocyte isolation remains severely limited. In most of cases, donor livers are of marginal quality that makes it difficult to obtain functional hepatocytes (23). An additional issue is that current technology for hepatocyte primary culture appears to be unable to support extensive cell proliferation (19). Under these circumstances, we previously proposed urokinase-type plasminogen activator transgenic severe combined immunodeficiency (uPA/SCID) mice as a feasible in-mouse hepatocyte propagation tool. uPA/SCID mice have a feature to develop an active damage of their own hepatic parenchymal cell and subsequent occurrence of continuous release of regenerative stimulus. Because of this nature, uPA/SCID mice provide a hepatic environment that is more conducive to the engraftment of human hepatocytes and a selective advantage for transplanted cells to proliferate (29). An important property of uPA/SCID mice was recently reported to allow human hepatocytes transplanted into the liver of uPA/SCID mice to actively propagate while retaining their ability to produce and secrete human FIX (30). From these data, it was reasonably speculated that the propagated human hepatocytes can serve as a cell source for future hepatocyte-based therapies toward hemophilia B. The functional preservation of FIX production of the propagated human hepatocytes encourages us to further assess functionalities for the production of other coagulation or anticoagulation factors.
We hypothesized that propagated human hepatocytes in the uPA/SCID mouse livers retained a normal gene expression of other coagulation and anticoagulation factors including prothrombin, factor VII, factor VIII, factor X, protein C, and protein. This report documents the first comprehensive analyses of coagulation factor-related gene expressions during in-mouse propagation status of human hepatocytes.
Materials and Methods
Animals
Recipient uPA/SCID mice were generated at Phoenix Bio (Higashihiroshima, Hiroshima, Japan) as described previously (29). Genotyping for the presence of uPA transgene in SCID mice was confirmed by polymerase chain reaction assay of isolated genomic DNA as described previously (11,29). Experimental protocols were developed in accordance with the guidelines of the local animal committees located at both PhoenixBio and Nara Medical University.
Transplantation of Human Hepatocytes for Propagation in the uPA/SCID Livers
Human hepatocytes, isolated from a 1-year-old white male and a 6-year-old Afro-American female were purchased from In Vitro Technologies (Baltimore, MD). The cryopreserved hepatocytes were thawed and suspended in transplant medium (9,29). The cell viability of the human hepatocytes was determined to be 64.4% and 49.2% by trypan blue exclusion test, respectively. One day prior to the transplantation and 1 week after the transplantation, uPA/SCID mice, 20–30 days old, received intraperitoneal injections of 0.1 mg of anti-asialo GM1 rabbit serum (Wako Pure Chemical Industries, Osaka, Japan) to inhibit recipient natural killer cell activity against the transplanted hepatocytes. Viable human hepatocytes (0.75 × 106) were transplanted using an infusion technique into the inferior splenic pole in which the transplanted cells flow from the spleen into the liver via the portal system (n = 18). After the transplantation, uPA/SCID mice were treated with nafamostat mesilate to inhibit complement factors activated by human hepatocytes as described elsewhere (29).
Determination of Replacement Ratio
Blood samples were collected periodically from the tail vein, and the levels of human albumin were determined with a Human Albumin ELISA Quantitation kit (Bethyl Laboratories, Montgomery, TX) to estimate the status of proliferation and propagation of the transplanted human hepatocytes as previously described (29). For accurate determination of the ratio that transplanted human hepatocytes occupied in the recipient mouse livers (the replacement ratio), the harvested liver section were stained with antibodies against human-specific cytokeratins 8 and 18 (hCK8/18), as described elsewhere (29). The replacement ratios of the mouse liver that received human hepatocytes were calculated as the ratio of area occupied by hCK8/18-positive hepatocytes to the entire area examined immunohistochemical sections of six lobes.
RNA Isolation and Quality Controls
Total RNA was extracted from the liver of the recipient mice with various stages of the replacement and normal human liver tissue samples by a RNeasy Mini Kit (QIAGEN, Hilden, Germany) following the manufacturer's instructions. Normal human liver tissue portions were obtained from surgical specimens during liver surgery for metastatic liver tumors after acquiring a written informed consent for the experimental use of harvested liver samples. DNase I was used to eliminate genomic DNA contamination, and the concentration of the RNA was determined by UV spectrometry. All of the RNA samples used in this study had an absorbance ratio (260/280 nm) between 1.9 and 2.1, and the integrity of RNA samples was confirmed by electrophoresis on a 1% agarose gel.
Reverse Transcription Coupled to Quantitative Real-Time PCR (Real-Time RT-PCR)
Total RNA (1 μg) was reverse-transcribed using oligo d(T)16 primers as described by the manufacturer (Omniscript RT Kit; QIAGEN). First-strand cDNA samples were subjected to quantitative PCR amplification using a StepOne Real-time PCR system (Applied Biosystems, Tokyo). For this experiment, we examined the following gene groups: 1) seven housekeeping reference genes, including glyceraldehyde-3-phosphate dehydrogenase (Gapdh), β-actin (Actb), peptidylprolyl isomerase A (Ppia), ribosomal protein L4 (Rpl4), transferrin receptor (Tfrc), β-glucuronidase (Gusb), and hypoxanthine phosphoribosyltransferase (Hprt1); 2) the genes of five vitamin K-dependent coagulation factors, including prothrombin (F2), factor VII (F7), factor × (F10), protein C (Prosc), and protein S (Pros1); and 3) factor VIII (F8) gene. TaqMan probes and primers for these genes were chosen from a TaqMan Gene Expression Assay (Applied Biosystems), and the information regarding these primer sets are listed in Table 1. All PCR analyses were performed using the following cycling conditions: 10 min at 95°C, followed by 40 cycles of 15 s at 95°C, and 1 min at 60°C. The specificity of the primers was verified by 2% agarose gel electrophoresis of the amplicons derived from naive human liver cDNAs. The PCR primers used in this study were confirmed to be human specific and to have no cross-reactions with mouse-derived genes. For quantification of gene expression, the cDNAs derived from total RNA extracted from normal human liver tissues were serially diluted and used to generate calibrations.
Primers Used in This Study
Statistical Analysis
Correlation coefficients between the repopulation rate and each gene expression were determined using Excel (Microsoft).
Results
Propagation of Human Hepatocytes in uPA/SCID Mouse Liver
Human hepatocytes were transplanted to uPA/SCID mice, which were sacrificed to excise the liver tissues at various time periods after the transplantation with monitoring the levels of blood human albumin. Using the collected liver samples, hCK8/18 immunostaing on the liver sections was performed to accurately assess the replacement ratios with human hepatocytes as described in Materials and Methods. As a result, the repopulation ratios ranged from 0% to 98%, and the number of mice in each repopulation category was 2, 4, 4, 4, and 4 in 0–20%, 21–40%, 41–60%, 61–80%, and 81–100%, respectively.
Selection of an Appropriate Reference Gene
The expressions of seven commonly used housekeeping genes (HKG) specific to human cells were evaluated in the recipient uPA/SCID mouse livers. Figure 1 shows that the raw expression levels of all seven HKGs increased in parallel to the replacement ratios. The correlation coefficiency with the replacement ratios of each gene was 0.68 for hGapdh; 0.86 for hActb; 0.72 for hPpia; 0.82 for hRpl4; 0.68 for hTfrc; 0.75 for hGusb; and 0.78 for hHprt1. When the replacement ratios exceeded 80%, the expression levels of all HKGs but Hprt1 became higher than those of normal human liver samples. hGapdh expression levels in the repopulated uPA/SCID livers were also observed to be beyond the levels of normal human livers at the repopulation ratio as low as 40%. Eventually, hGapdh expression levels reached approximately sixfold of normal human liver levels when the repopulation ratios exceeded 80%. In contrast, gene expression levels of hActb failed to reach to comparable levels with normal human liver until the repopulation ratios were close to 100%. Under the condition of varied gene expression levels of HKGs, it is important to select the most stably expressed HKG to assess the expression of target human genes in the uPA/SCID livers. For achieving this, Actb gene, which demonstrated the best correlation coefficient with the replacement ratios, was selected as a reference normalizing gene in the following gene expression analyses.
The raw gene expression levels of human housekeeping genes in the human hepatocytes repopulated in human hepatocytes were transplanted into urokinase-type plasminogen activator transgenic sever combined immunodeficient uPA/SCID mouse livers. Isolated uPA/SCID mouse livers (n = 18), and the livers were excised at various points of repopulation ratios determined by blood human albumin levels and human-specific cytokeratins 8 and 18 (hCK8/18) immunohistochemistry on the liver sections. The repopulation ratio ranged from 0% to 98% (0–20%, two mice; 21–40%, four mice; 41–60%, four mice; 61–80%, four mice; and 81–100%, four mice). Gene expression levels of commonly used seven housekeeping genes [glyceraldehyde-3-phosphate dehydrogenase (Gapdh), β-actin (Actb), peptidylprolyl isomerase A (Ppia), ribosomal protein L4 (Rpl4), transferrin receptor (Tfrc), β-glucuronidase (Gusb), and hypoxanthine phosphoribosyltransferase (Hprt1)] in human–chimeric mouse liver samples were quantified by real-time PCR with human-specific primers and expressed as relative values against the control normal human liver tissue (defined as 1.0). The correlation coefficient of each gene was expressed as an R value.
Expression of Coagulation Factor Genes
Human-specific coagulation-related gene expression levels were assessed on human hepatocyte-repopulated uPA/SCID mouse livers. The genes analyzed were: vitamin K-dependent coagulation factors (prothrombin, factor VII, and factor X) and anticoagulation factors (protein C and protein S), in addition to factor VIII. Raw expression levels of all vitamin K-dependent coagulation and anticoagulation factor genes showed a positive correlation with the repopulation ratios (Fig. 2). The correlation coefficient between the gene expression levels and the repopulation ratios were 0.78, 0.74, 0.80, 0.80, and 0.82 in F2, F7, F10, Prosc, and Pros1, respectively. The raw gene expression levels of all but F8 were beyond the levels of the normal human liver samples (defined as 1.0) as the repopulation ratios increased. In marked contrast, F8 gene expression levels were less than 40% of normal human liver tissues even though the repopulation ratios reached approximately 100%. The low levels of F8 gene expression failed to show a significant correlation with the repopulation ratios (R = 0.66).
The raw gene expression levels of human coagulation and anticoagulation factors in the human hepatocytes repopulated in the uPA/SCID mouse livers. Isolated human hepatocytes were transplanted into uPA/SCID mouse livers (n = 18), and the livers were excised at various points of repopulation ratios determined by blood human albumin levels and hCK8/18 immunohistochemistry on the liver sections. The repopulation ratio ranged from 0% to 98% (0–20%, two mice; 21–40%, four mice; 41–60%, four mice; 61–80%, four mice; and 81–100%, four mice). Gene expression levels of coagulation factors (prothrombin, F2; factor VII, F7; factor X, F10; and factor VIII, F8) and anticoagulation factors (protein C, Prosc; protein S, Pros1) in the human–chimeric mouse liver samples were quantified by real-time PCR with human-specific primers and expressed as relative values against the control normal human liver tissue (defined as 1.0). The correlation coefficient of each gene was expressed as an R value.
In order to evaluate the gene expression levels per human hepatocytes, the gene expression levels were normalized by ACTB gene expression levels in each sample (Fig. 3). As a result, normalized gene expression values of all the analyzed coagulation-related factor genes showed constant expression levels regardless of the repopulation ratios, demonstrating that the human hepatocytes in the uPA/SCID livers stably express coagulation-related factor genes throughout the repopulation stages.
The relative gene expressions levels of human coagulation and anticoagulation factors in the human hepatocytes repopulated in the uPA/SCID mouse livers. The raw gene expression levels of human coagulation and anticoagulation factors show in Figure 2 were normalized with Actb gene expression levels in each human–chimeric mouse liver samples and plotted the repopulation ratios.
Discussion
Propagation of primary human hepatocytes that possess hepatocyte-specific functionalities including blood clotting factor production has been one of the major paradigms in liver regenerative medicine (18,24). In the present study, we transplanted primary human hepatocytes to the liver of uPA/SCID mice and succeeded in propagating the human hepatocytes in the mouse livers. We then investigated mRNA expression levels of human-specific vitamin K-dependent coagulation factors (prothrombin, factor VII, and factor X), anticoagulation factors (protein C and protein S), and factor VIII of the propagated hepatocytes at various stages of propagation. The results showed that mRNA expression levels per human hepatocyte of all the analyzed genes were maintained through the propagation stage, indicating that the uPA/SCID in-mouse hepatocyte propagation system is a viable method to propagate hepatocytes that are intact in coagulation factor productions.
Coagulation factors are produced mainly by hepatocytes, and the long-term synthesis of these factors from primary human hepatocytes in vitro have been recently achieved by plating cells inside a 3D collagen gel matrix together with hormonally enriched culture medium under chemically defined conditions (6), indicating the pivotal role of the extracellular environment for coagulation factor production. However, the current procedure for the culture of primary hepatocytes appears to be difficult to support extensive cell proliferation (19), still remaining the problem of donor cell shortage unresolved. It is true that isolated hepatocytes could obtain proliferating ability and long-term survival in vitro by immortalization (26,34,35) or by selective culture of small hepatocyte population (27), but there is no report for studying the gene expression and production of coagulation factors including factor IX in these cell types. On the other hand, embryonic stem (ES) cells and induced pluripotent stem (iPS) cells have been intensively investigated as an attractive cell source for liver regenerative medicine, and differentiation technologies of these stem cells into hepatocyte-like cells have been improved (12). In these circumstances, Basma et al. (3) recently succeeded in differentiating human ES cells into hepatocytes-like cells maintaining the ability of human factor VII production; however, the expression of coagulation factors other than factor VII were undocumented. In contrast to these in vitro cell culture systems, there are several in vivo hepatocyate propagating systems, in which transplanted hepatocytes can efficiently proliferate in mouse livers, such as uPA/SCID mice (29) or fumarylacetoacetate hydrolase-/-/recombination activating gene-/-/gamma chain of the interleukin-2 receptor-/- (Fah-/-/Rag2-/-/Il2rg-/-) mice (2).
In our previous series of experiments, we found that human hepatocytes that were transplanted into the liver of uPA/SCID mouse perform active cell proliferation leading to a nearly total repopulation of the liver (17, 29,30,36) and confirmed that those proliferated hepatocytes maintained their ability to produce and secrete biologically functional human factor IX (30). In addition to human hepatocytes, we also reported that primary canine hepatocytes could proliferate in uPA/SCID mouse liver while retaining functions for canine factor IX production (30). At least, this in-mouse hepatocyte propagating system is only an available and promising procedure for proliferating factor IX-producing hepatocytes at the present time. In this regard, it is important to clarify gene expression of other coagulation factors as well as factor IX in propagated human hepatocyte in uPA/SCID mouse liver, because to establish a method for hepatocytes proliferation while retaining ability for coagulation factor production is indispensable for clinical cell therapy toward coagulation disorders.
We first assessed the expression levels of seven commonly analyzed HKGs for human genome in the human hepatocyte-repopulated uPA/SCID mouse livers. The gene expression levels of all seven HKGs increased in parallel to the increase of the repopulation ratio with a high correlation coefficient, providing direct evidence that the transplanted human hepatocytes progressively proliferated in uPA/SCID mouse livers (Fig. 1). Interestingly, as for six out of seven HKGs analyzed, the gene expression levels of samples with more than 80% repopulation ratios surpassed the levels of normal human livers (arbitrarily defied as 1.0). It was reported that the upregulation of certain HKGs, especially Gapdh, was closely associated with the events of DNA synthesis and cell division (1). Such HKG upregulation profiles associated with the hepatocyte proliferation were also observed in our previous works where hepatocyte proliferation in mouse livers was induced through the mode of compensatory regeneration (32) or direct hyperplasia (28). These findings suggest that human hepatocytes progress their cell cycling events in the uPA/SCID livers while upregulating their structural proteins necessary for cell proliferation. Since HKGs are used as internal reference gene(s) in gene expression analyses, it is essential to identify appropriate HKGs that are stably expressed during examined pathological process. During the human hepatocyte repopulation process in the uPA/SCID livers, we found that the expression levels of Actb showed the highest correlation coefficient (R = 0.86) with the repopulation ratios among seven HKGs analyzed. Therefore, we concluded that the use of Actb for normalization of gene expression levels was appropriate for obtaining accurate gene expression values.
To elucidate gene expression levels of coagulation and anticoagulation factors in propagating human hepatocytes in the liver of uPA/SCID mice was the main objective in this study. Using human-specific PCR primers, we found that raw expression levels of all genes increased in parallel to the increase of repopulation ratios with a high correlation coefficient. As the repopulation ratios increased higher than 80%, the expression levels of all genes except F8 surpassed the levels of normal human livers (Fig. 2). We also investigated the gene expression levels per human hepatocyte by normalizing the expression levels of each gene by the levels of Actb. Results clearly showed that the gene expression levels of all analyzed coagulation and anticoagulation factors were stably maintained throughout the in-mouse repopulation process (Fig. 3). Overall, the present study combined with our previous investigation (30) demonstrated that human primary hepatocytes were able to proliferate in the liver of uPA/SCID mice while retaining the cellular machinery for expressing F2, F7, F9, F10, Prosc, and Pros1. Since human hepatocyte propagated in uPA/SCID mice are able to be isolated and purified by cell-sorting technology (37), the present in-mouse propagated human hepatocytes is a feasible candidate cell source for a future therapeutic use toward coagulation factor deficiencies. Furthermore, it was reported that efficient gene transduction into proliferated human hepatocytes in uPA/SCID mice was possible by using retroviral vector system (9), indicating the capability of obtaining hepatocytes that were genetically modified in vivo for the purpose to achieve higher expression levels of target proteins (15).
In the present study, gene expression levels of F8 were remarkably low compared with those of other factors assessed. Even the liver samples showing the repopulation ratio more than 80% demonstrated only less than 40% of the control normal human liver tissues in human F8 gene expression (Fig. 2). uPA/SCID mice have a characteristic to allow their own hepatocytes to be replaced by repopulated human hepatocytes, but all other intravital environment as well as nonparenchymal liver cells remain predominantly host origin. As a result, it is likely that various signal cross-talks between repopulated human hepatocytes and host cells or humoral factors might become dysfunctional. For example, although hepatocytes require growth hormone (GH) for their active cell proliferation, human hepatocytes transplanted into mouse livers failed to be the target for such growth advantage because rodent GH is unable to bind to human GH receptors (17). We have indeed established that this functional deficiency could be fully recovered by administration of human GH to the human hepatocyte-repopulated uPA/SCID mice (17). The fact that human F8 gene expression level failed to increase may be explained by some mechanism similar to this type of interspecies incompatibility. An alternative possibility is that nonparenchymal cells, but hepatocytes, are the main responsive cells for human factor VIII productions. Although the liver has been shown to be the major site of factor VIII production as evidenced by previous liver transplantation clinical experiences (10,13,33), the precise type of liver cells contributing to factor VIII production has not been fully identified (4,6,8,16). In either the case, the present in-mouse hepatocyte propagation system using uPA/SCID mouse might be a valuable tool for the elucidation of cellular mechanism of factor VIII synthesis and production.
In conclusion, the present study provides encouraging evidence that uPA/SCID mouse system supported the active proliferation of human hepatocytes while maintaining cellular machinery to produce vitamin K-dependent coagulation (prothrombin, factor VII, and factor X) and anticoagulation factors (protein C and protein S) in addition to factor VIII. The current work thus can serve as a basis to create a hepatocyte propagation system to prepare sufficient amount of cells for the therapeutic purposes for deficiencies of these factors as well as for the research purpose to investigate the hepatocyte-specific production mechanisms of coagulation factors.
Footnotes
Acknowledgments
The authors would like to thank Dr. Norio Ueno (Tokyo Women's University) for his critical reading of the manuscript. This study was supported by grants for AIDS Research from the Ministry of Health, Labor and Welfare of Japan (M.S.), Grant-in-Aid (K.O. No. 21300180) and Special Coordination Funds for Promoting Science and Technology (K.O. and T.O.), and Global Center of Excellence Program (K.O.) from the Ministry of Education, Culture, Sports and Science and Technology of Japan, and Bayer Hemophilia Award Program (K.O.). The authors declare no conflict of interest.