Fetal liver development requires a paracrine action of oncostatin M through the gp130 signal transducer
1999; Springer Nature; Volume: 18; Issue: 8 Linguagem: Inglês
10.1093/emboj/18.8.2127
ISSN1460-2075
Autores Tópico(s)PI3K/AKT/mTOR signaling in cancer
ResumoArticle15 April 1999free access Fetal liver development requires a paracrine action of oncostatin M through the gp130 signal transducer Akihide Kamiya Akihide Kamiya Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Taisei Kinoshita Taisei Kinoshita Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Yoshiaki Ito Yoshiaki Ito Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Takaaki Matsui Takaaki Matsui Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Yoshihiro Morikawa Yoshihiro Morikawa Department of Anatomy and Neurobiology, Wakayama Medical School, 640-8155 Wakayama, Japan Search for more papers by this author Emiko Senba Emiko Senba Department of Anatomy and Neurobiology, Wakayama Medical School, 640-8155 Wakayama, Japan Search for more papers by this author Kinichi Nakashima Kinichi Nakashima Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, 101-0062 Tokyo, Japan Search for more papers by this author Tetsuya Taga Tetsuya Taga Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, 101-0062 Tokyo, Japan Search for more papers by this author Kanji Yoshida Kanji Yoshida Department of Molecular Immunology, Research Institute for Microviral Diseases, Osaka University, 565-0871 Suita, Japan Search for more papers by this author Tadamitsu Kishimoto Tadamitsu Kishimoto Department of Medicine III, Osaka University Medical School, 565-0871 Suita, Japan Search for more papers by this author Atsushi Miyajima Corresponding Author Atsushi Miyajima Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Akihide Kamiya Akihide Kamiya Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Taisei Kinoshita Taisei Kinoshita Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Yoshiaki Ito Yoshiaki Ito Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Takaaki Matsui Takaaki Matsui Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Yoshihiro Morikawa Yoshihiro Morikawa Department of Anatomy and Neurobiology, Wakayama Medical School, 640-8155 Wakayama, Japan Search for more papers by this author Emiko Senba Emiko Senba Department of Anatomy and Neurobiology, Wakayama Medical School, 640-8155 Wakayama, Japan Search for more papers by this author Kinichi Nakashima Kinichi Nakashima Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, 101-0062 Tokyo, Japan Search for more papers by this author Tetsuya Taga Tetsuya Taga Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, 101-0062 Tokyo, Japan Search for more papers by this author Kanji Yoshida Kanji Yoshida Department of Molecular Immunology, Research Institute for Microviral Diseases, Osaka University, 565-0871 Suita, Japan Search for more papers by this author Tadamitsu Kishimoto Tadamitsu Kishimoto Department of Medicine III, Osaka University Medical School, 565-0871 Suita, Japan Search for more papers by this author Atsushi Miyajima Corresponding Author Atsushi Miyajima Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan Search for more papers by this author Author Information Akihide Kamiya1, Taisei Kinoshita1, Yoshiaki Ito1, Takaaki Matsui1, Yoshihiro Morikawa2, Emiko Senba2, Kinichi Nakashima3, Tetsuya Taga3, Kanji Yoshida4, Tadamitsu Kishimoto5 and Atsushi Miyajima 1 1Laboratory of Cellular Biosynthesis, Institute of Molecular and Cellular Biosciences, University of Tokyo, Bunkyo-ku, 113-0032 Tokyo, Japan 2Department of Anatomy and Neurobiology, Wakayama Medical School, 640-8155 Wakayama, Japan 3Department of Molecular Cell Biology, Medical Research Institute, Tokyo Medical and Dental University, Chiyoda-ku, 101-0062 Tokyo, Japan 4Department of Molecular Immunology, Research Institute for Microviral Diseases, Osaka University, 565-0871 Suita, Japan 5Department of Medicine III, Osaka University Medical School, 565-0871 Suita, Japan *Corresponding author. E-mail: [email protected] The EMBO Journal (1999)18:2127-2136https://doi.org/10.1093/emboj/18.8.2127 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Fetal liver, the major site of hematopoiesis during embryonic development, acquires additional various metabolic functions near birth. Although liver development has been characterized biologically as consisting of several distinct steps, the molecular events accompanying this process are just beginning to be characterized. In this study, we have established a novel culture system of fetal murine hepatocytes and investigated factors required for development of hepatocytes. We found that oncostatin M (OSM), an interleukin-6 family cytokine, in combination with glucocorticoid, induced maturation of hepatocytes as evidenced by morphological changes that closely resemble more differentiated hepatocytes, expression of hepatic differentiation markers and intracellular glycogen accumulation. Consistent with these in vitro observations, livers from mice deficient for gp130, an OSM receptor subunit, display defects in maturation of hepatocytes. Interestingly, OSM is expressed in CD45+ hematopoietic cells in the developing liver, whereas the OSM receptor is expressed predominantly in hepatocytes. These results suggest a paracrine mechanism of hepatogenesis; blood cells, transiently expanding in the fetal liver, produce OSM to promote development of hepatocytes in vivo. Introduction Liver development comprises multiple stages and is influenced by hormonal factors as well as intercellular and matrix–cellular interactions. In mice, the initial event of liver ontogeny occurs on embryonic day 9 (E9); the midgut endoderm commits to become the liver through the interaction with the cardiogenic mesoderm. Then, this liver primodium accepts an inductive signal from the adjacent mesenchym in the heart and invades the septum transversum to give rise to the hepatic codes and bud (Douarin, 1975; Houssaint, 1980). Following these differentiation processes, fetal hepatocytes proceed through a series of maturation steps which accompany autonomous proliferation, cellular enlargement and functional maturation. The level of hepatic maturation has been characterized by the expression of liver- and stage-specific genes (Derman et al., 1981; Panduro et al., 1987). Alpha-fetoprotein (AFP) is an early fetal hepatic marker and its expression decreases as the liver develops (Shiojiri et al., 1991). In contrast, expression of albumin, the most abundant protein synthesized by hepatocytes, starts in early fetal hepatocytes (E12) and reaches the maximal level in adult hepatocytes (Tilghman and Belayew, 1982). At a late gestation or perinatal stage, hepatocytes start producing a number of metabolic enzymes including glucose-6-phosphatase (G6Pase) and tyrosine aminotransferase (TAT) to prepare their metabolism for the change in the physiological role of the liver (Greengard, 1970; Haber et al., 1995). Finally, several days after birth, serine dehydratase (SDH) and tryptophan oxygenase (TO) are induced in hepatocytes (Nagao et al., 1986; Noda et al., 1990, 1994). Interestingly, expression of these metabolic enzymes is often lost in hepatocytic carcinomas or malignant hepatomas and, conversely, AFP expression resumes in these tumors (Abelev, 1971). Hence, AFP and metabolic enzymes are useful makers to monitor liver development as well as cellular malignancy. Proliferation and differentiation of hepatocytes are affected by extracellular signals such as hormones and cytokines. For example, glucocorticoid modulates proliferation and function of adult hepatocytes both in vivo and in vitro. In the fetal liver, physiological concentrations of dexamethasone (Dex), a synthetic glucocorticoid, suppress AFP production and DNA synthesis and up-regulate albumin production (Belanger et al., 1981; Nawa et al., 1986; de Juan et al., 1992). Transforming growth factor-β (TGF-β) is a potent inhibitor of hepatocyte proliferation (Nakamura et al., 1985; de Juan et al., 1992) and was shown to augment albumin production in prenatal hepatocytes, implicating TGF-β in the regulation of hepatic differentiation (Sanchez et al., 1995). Intriguingly, the regulation of TAT mRNA expression depends on the developmental stage of the liver. TAT mRNA, which is virtually absent in the early fetal liver, is induced by Dex in primary hepatocytes of late embryonic stage. In contrast, Dex does not regulate TAT levels at earlier stages (mid-gestation; E12–14), even though these cells are able to express albumin in response to Dex (Shelly et al., 1989). These observations suggest that there is a key maturation step between the mid-gestation and late gestation/perinatal stages. However, the mechanism underlying hepatic development, particularly at the molecular and biochemical level, remains unclear. Oncostatin M (OSM) is a member of the interleukin-6 (IL-6)-related cytokine family that includes IL-6, IL-11, leukemia inhibitory factor (LIF), ciliary neurotrophic factor and cardiotrophin-1 (Bazan, 1991; Rose and Bruce, 1991; Pennica et al., 1995). These cytokines often exhibit similar functions since their receptors utilize gp130 as a common signal transducer (reviewed in Taga and Kishimoto, 1997). In particular, human OSM (hOSM) (Malik et al., 1989) shares many biological functions with LIF, such as induction of differentiation in M1 monocytic cells (Rose and Bruce, 1991; Bruce et al., 1992) and of acute phase proteins in hepatocytes (Richards et al., 1992; Baumann et al., 1993). However, hOSM also possesses unique functions, e.g. growth stimulation of endothelial cells (Wijelath et al., 1997) and smooth muscle cells (Grove et al., 1993). Two types of hOSM receptors have been identified; the type I OSM receptor is identical to the LIF receptor composed of gp130 and the LIF-binding subunit (Gearing et al., 1991, 1992; Liu et al., 1992), and the type II receptor consists of gp130 and the OSM-specific subunit (Mosley et al., 1996). Thus, the common functions of LIF and OSM are mediated by the type I receptor, whereas the unique activities of OSM are transduced by the type II receptor (Thoma et al., 1994; Mosley et al., 1996). In contrast, mouse OSM utilizes only its specific receptor (the type II receptor) and does not react with the LIF receptor (Ichihara et al., 1997; Lindberg et al., 1998; Tanaka et al., 1999). Mouse OSM was cloned as an IL-3-inducible gene in hematopoietic cells (Yoshimura et al., 1996) and was shown to be expressed in various types of hematopoietic cells as well as in the bone marrow (Ichihara et al., 1997). Identification of mouse OSM made it possible to analyze OSM-specific actions using murine systems. For example, we recently showed that OSM stimulates generation of definitive hematopoietic cells from putative progenitors (hemangioblasts) which reside in the aorta–gonad–mesonephros (AGM) region of E11 mouse embryo (Mukouyama et al., 1998). Here we describe the establishment of an in vitro culture system of fetal hepatic cells and demonstrate that OSM induces development of hepatocytes in vitro. Furthermore, studies of gp130 knockout mice reveal an essential role for OSM/gp130 in hepatic maturation in vivo. We also show evidence that OSM is produced by hematopoietic cells expanding in the fetal liver. Our results indicate that OSM is a paracrine regulator which plays a pivotal role during fetal liver development. Results Morphological changes of fetal hepatocytes in primary culture upon oncostatin M stimulation In order to investigate the molecular basis of fetal hepatic development, we established a novel primary culture system of murine fetal hepatocytes derived from E14 embryos. Fetal liver cells were dissociated using enzyme-based digestion buffer and plated onto 0.1% gelatin-coated dishes. Several hours later, floating hematopoietic cells and dead cells were removed by extensive washing with culture media. The remaining adhesive cells formed a monolayer sheet (Figure 2A), with the majority of these cells appearing to be hepatocytes as assessed by epithelial morphology. Cultured cells were characterized further by their expression profiles of liver- and stage-specific genes and proteins (Figure 1). mRNAs for the embryonic hepatocyte-specific AFP and for liver-specific hepatic nuclear factors [hepatocyte nuclear factor (HNF)-1 and HNF-4] were expressed in cultured cells. In addition, Western blot analyses showed that cultured cells expressed albumin and E-cadherin proteins, which are present in the E14.5 embryonic liver. On the contrary, cultured cells did not express differentiation markers for the peri- or postnatal liver (G6Pase and TAT). Moreover, these cells grow autonomously without any growth factors such as epidermal growth factor (EGF) and hepatocyte growth factor (HGF) (data not shown). These observations indicate that the majority, if not all, of cultured cells are embryonic hepatocytes that retain many characteristics of fetal hepatocytes in vivo at this stage. Figure 1.Characterization of fetal hepatic cells in primary culture. (A) Comparison of the expression profiles of liver-specific genes between primary cultured fetal hepatic cells and developing liver tissues. Ten micrograms of total RNAs extracted from fetal hepatic cells free from hematopoietic cells and liver tissues (E14, neonatal and adult) were analyzed by Northern blot using DIG-labeled cDNA probes for AFP, G6Pase, TAT, HNF-1α and HNF-4. Fetal hepatic cells express genes specific for embryonic hepatocytes, while differentiation markers for the postnatal liver were not detected. (B) Expression of the albumin and E-cadherin proteins. Protein samples were extracted from fetal hepatic cells and liver tissues (E14, neonatal and adult). Expression of albumin and E-cadherin was examined by Western blot. Download figure Download PowerPoint Figure 2.Morphology of cultured fetal hepatic cells. Fetal hepatic cells derived from the E14 embryonic liver were cultured for 5 days in the absence (A) or presence of 10 ng/ml OSM (B and D–F) or 10 ng/ml TGF-β (C) in combination with glucocorticoid (10−7 M dexamethasone). Cultured hepatic cells show a clear epithelial sheet (A). OSM induced the formation of multiple clusters that closely resemble more differentiated hepatocytes (B). In contrast, TGF-β did not induce such a cluster (C). (D) Staining of the albumin protein in fetal hepatic cells cultured in the presence of OSM. (E) E-cadherin staining of the cells shown in (D). The E-cadherin protein was localized to the sites of homophilic cell adhesion. (F) Superimposition of (D) and (E), showing that cell clusters are positive for both albumin and E-cadherin. Download figure Download PowerPoint Using this system, we searched for a molecule which stimulates hepatic development in vitro. Among the various cytokines which we tested, including IL-6, IL-11, LIF, OSM and TGF-β, only OSM showed striking effects on fetal primary hepatocytes. First, addition of OSM resulted in the formation of multiple clusters that closely resemble mature hepatocytes (Figure 2B). These clusters exhibited tight cell–cell contact, highly condensed and granulated cytosol and clear round-shaped nuclei in a way similar to mature hepatocytes. To examine whether these cell clusters were differentiated hepatocytes, we stained them with anti-E-cadherin and anti-albumin antibodies. As shown in Figure 2E, E-cadherin was localized to the periphery of cells that constitute clusters induced by OSM, although the total E-cadherin level did not change before or after OSM stimulation (data not shown). Co-localization of E-cadherin and albumin in the same cells revealed that E-cadherin+ cell clusters are parenchymal hepatocytes (Figure 2D and F). We also found that the presence of a physiological concentration of glucocorticoid (10−7 M Dex) is absolutely required for the OSM action; addition of OSM (up to 100 ng/ml) failed to induce the cluster formation in the absence of Dex (data not shown). These results indicate that OSM and glucocorticoid collaborate to induce morphological changes in fetal hepatocytes. Regulation of albumin production by OSM and Dex E14 embryonic liver expresses a detectable amount of albumin, although the level of expression was significantly lower than that in adult hepatocytes (Figure 1B). Since albumin production is an important and characteristic liver-specific function, we examined whether OSM would regulate albumin production in cultured hepatocytes. After 6 days of incubation with or without cytokines, equal amounts of total cellular proteins were examined for albumin production by Western blot analysis. When fetal hepatocytes were cultured in the absence of exogenously added factors, albumin production gradually decreased to a low level within 2–3 days as shown in Figure 3A, consistent with the previous observation (Sanchez et al., 1995). In contrast, albumin production was maintained in the presence of both 10 ng/ml OSM and 1.0×10−7 M Dex (Figure 3A). The effect was dose dependent, ranging from 0 to 10 ng/ml for OSM and from 0 to 10−6 M for Dex (Figure 3B and C). Again, the presence of both factors was necessary to maintain albumin production. TGF-β was shown previously to maintain the albumin level in cultured hepatocytes derived from E21 rat embryos (Sanchez et al., 1995). However, no such effect was observed in E14 mouse embryo-derived hepatocytes in our system. Probably, E14 embryo-derived hepatocytes are less mature and therefore not sensitive to TGF-β stimulation for the production of albumin. Figure 3.Regulation of albumin production by OSM and Dex in vitro. (A) Time course of albumin expression during primary culture in the presence of various combination of OSM (10 ng/ml), TGF-β (10 ng/ml) and Dex (10−7 M). (B and C) Dose dependence of albumin production on OSM in the presence of Dex (10−7 M) (B) or on Dex in the presence of OSM (10 ng/ml) (C). After a 7 day incubation, protein samples were extracted from hepatocytes and albumin production was examined by Western blot using the anti-albumin antibody. Download figure Download PowerPoint Gene induction of hepatic differentiation markers in vitro by OSM and Dex During hepatic development, expression of liver-specific genes is strictly regulated in accordance with the requirement for stage-specific liver functions (Panduro et al., 1987). Striking changes in gene expression occur around the perinatal stage, since the liver needs to assume metabolic functions soon after birth. In rats, for example, G6Pase begins to be produced at E20–21 (just before birth) and its expression reaches the maximum at 1 h after birth; likewise, TAT mRNA first appears in the postnatal liver and further increases with maturation. Similar expression profiles of mRNA for G6Pase and TAT were observed for murine hepatic development (Figure 4A). Thus, these genes are also useful makers to monitor the level of hepatic maturation in mice. Figure 4.Expression of differentiation markers in response to OSM (10 ng/ml) and Dex (10−7 M). (A) Developmental changes in expression patterns of liver-specific differentiation markers in vivo. Ten micrograms of total RNAs from E14, E18, neonatal and adult liver tissues were analyzed by Northern blot using DIG-labeled cDNA probes for G6Pase, TAT, TO and GAPDH (internal control). (B) Induction of mRNA for hepatic differentiation markers in vitro by OSM and Dex. (C) Time courses of G6Pase and TAT mRNA expression after OSM stimulation. (D) Time course of induction of haptoglobin mRNA after OSM stimulation. (E) Failure of other IL-6-related cytokines to induce G6Pase or TAT mRNA expression. Download figure Download PowerPoint To examine whether mRNA expression of G6Pase and/or TAT is inducible by OSM and Dex, fetal hepatic cultures were stimulated with either OSM or TGF-β (Figure 4B). In the absence of Dex, no expression of either mRNA was detected even when OSM was added. Stimulation by Dex alone induced G6Pase mRNA only slightly around days 4–8, and the expression decreased thereafter. In contrast, hepatocytes, cultured with both OSM and Dex, started expressing G6Pase mRNA from 4 days after stimulation, and the expression level continued to increase to reach a maximum after 8–10 days (Figure 4C). Induction of TAT mRNA expression was even more remarkable. OSM induced TAT mRNA expression after 4 days of culture, with a maximum obtained on day 8. As TAT mRNA was barely detectable in the absence of OSM, in vitro TAT induction strictly depends on OSM. In contrast, mRNA for an acute phase protein (haptoglobin) was induced more quickly by OSM than those for G6Pase and TAT (Figure 4D). These results suggest that molecular mechanisms underlying the induction of differentiation markers and acute phase proteins (Richards et al., 1992, 1997) are different, at least in part. Other IL-6-related cytokines (IL-6, LIF or IL-11) failed to stimulate induction of these enzymes; however, IL-6, when combined with the soluble IL-6 receptor (Yasukawa et al., 1992; Yawata et al., 1993), induced comparable levels of mRNA for both enzymes (Figure 4E). This suggests that the lack of an IL-6 response is due to lack of IL-6Rα expression and that OSM transduces differentiation signals through the common signaling receptor subunit, gp130. The continuous presence of OSM is necessary for induction of differentiation markers To determine the window of time during which OSM is required for hepatic maturation, we tested the effects of removal and delayed addition of OSM to the culture on in vitro maturation of fetal hepatocytes (Figure 5). Hepatocytes were first stimulated with OSM/Dex and then OSM was removed at various time points as indicated in the figure. Expression of differentiation markers was analyzed on day 7 post-stimulation. Figure 5A shows that removal of OSM before day 5 reduced the induction of differentiation markers. On the other hand, when OSM was removed on day 5, expression of differentiation markers was induced although the induction levels were slightly lower than those supported by continuous OSM stimulation. Conversely, delaying the addition of OSM reduced expression of differentiation markers; the later the addition of OSM, the lower the observed level of mRNA expression (Figure 5B). It is therefore likely that the induction of hepatic maturation in vitro requires continuous OSM stimulation. Figure 5.Continuous OSM stimulation is required for the in vitro hepatic maturation process. (A) The effect of OSM removal on gene activation of hepatic differentiation markers. Fetal hepatocytes were stimulated with OSM on day 0, and OSM was then removed from culture media on day 1, 2, 3 or 5 as indicated. Gene expression was examined on day 7. (B) OSM was added on day 1, 2, 3 or 5 after plating, and gene expression was examined on day 7. Download figure Download PowerPoint Stimulation of glycogenesis in vitro by OSM and Dex Regulation of the blood glucose level is another important function of the differentiated liver, and this is controlled by the rate of glycogenesis and glycogen breakdown (Nemeth et al., 1953; Foster et al., 1966). Glycogenesis first occurs during late fetal development, and both perinatal and more differentiated hepatocytes store a large reserve of glycogen (Yeung and Oliver, 1967; Philippidis and Ballard, 1969). To examine whether OSM/Dex induce functional maturation of hepatocytes to produce and store glycogen, we analyzed accumulation of intracellular glycogen in vitro by the Periodic acid–Schiff (PAS) staining method. Fetal hepatic cells cultured for 2 or 6 days in the presence of OSM were stained with the PAS reagent. No storage of glycogen was detected in cells incubated for 2 days regardless of the presence of OSM and/or Dex (data not shown). In contrast, after 6 days of incubation, we noted slight accumulation of glycogen in some hepatocytes cultured with Dex alone, and OSM/Dex strongly induced glycogen accumulation in the majority of the cells (Figure 6C and D). On the other hand, cells cultured with OSM alone or in the absence of both factors did not store a significant amount of glycogen even on day 6 (Figure 6A and B). These results indicate that, in addition to the morphological changes and induction of differentiation markers, OSM/Dex induce functional maturation of hepatocytes. Figure 6.Glycogenesis in cultured hepatocytes. Cultured fetal hepatocytes were incubated either with no factor (A), OSM alone (B), Dex alone (C) or OSM plus Dex (D). Cells were then stained with the PAS solution as described in Materials and methods. A number of PAS-positive cells appeared when stimulated with both OSM and Dex for 6 days. Download figure Download PowerPoint Defective development of the liver in gp130 knockout mice The results described above strongly suggest the importance of OSM and its receptor in hepatic development. To study further the role of the OSM/OSM receptor (OSMR) system in vivo, we analyzed gp130−/− mice. Although gp130 deficiency originally was described to be lethal around E14 in C57BL/6 mice (Yoshida et al., 1996), gp130−/− mice of the ICR genetic background often survive longer and die soon after birth (Kawasaki et al., 1997). These mice thus enabled us to analyze late fetal liver development. To investigate the liver function in gp130−/− mice, glycogen accumulation in the liver was examined. As shown in Figure 7, normal hepatocytes store a large amount of glycogen at both E17 and new-born stages (Figure 7A and C). In contrast, accumulation of intracellular glycogen was greatly reduced in the liver from gp130−/− mice (Figure 7B and D), although there were some PAS-positive cells along the periphery of liver lobules. This limited accumulation of glycogen in the gp130−/− liver was comparable with the small number of PAS-positive cells when fetal hepatocytes were cultured with Dex alone (Figure 6C). This suggests that although glucocorticoid induces limited glycogen accumulation, full metabolic function requires the OSM–gp130 signaling pathway. Defects found in gp130−/− mice are reminiscent of those in C/EBPα knockout mice (Wang et al., 1995). In C/EBPα−/− mice, hepatocytes are negative for PAS, although effects on differentiation markers are relatively modest. Likewise, TAT expression was reduced to some extent in the liver from gp130−/− mice (Figure 7E). Thus, while expression of these genes can be partly controlled by glucocorticoid or unknown factors, development of a functional liver appears to require the presence of both gp130 and C/EBPα signaling pathways. Figure 7.Defective maturation of the liver from gp130 knockout mice. (A–D) Glycogen storage in gp130−/− or control livers. Liver tissues from the wild-type (A and C) or gp130−/− mice (B and D) at E17 (A and B) or at a neonatal stage (C and D) were stained with the PAS solution. PAS staining of gp130−/− livers was much weaker than those of control mice at both stages. (E) Expression of a hepatic differentiation marker in gp130−/− and wild-type mice. mRNA samples were prepared from E17 embryonic livers and expression of TAT mRNA was examined by Northern blot. The level of TAT mRNA was significantly reduced in knockout livers. In some cases, gp130+/− mice showed a reduced level of TAT mRNA expression. This is consistent with the previous observation that gp130+/− mice show heterogeneous phenotypes in different assays (Yoshida et al., 1996). Download figure Download PowerPoint Expression of OSM and OSM receptor mRNAs in the developing liver Induction of differentiation markers, as well as the morphological and functional maturation specifically elicited by OSM/Dex, suggest that these molecules are involved in hepatic ontogeny in vivo. While glucocorticoid presumably is provided through the blood circulation, a cytokine such as OSM is often produced locally at the site of action. Accordingly, it is likely that both OSM and OSMR co-exist in the developing liver, if the OSM/OSMR system is really involved in the maturation process in vivo. We therefore investigated expression of OSM and OSMR mRNAs in developing liver tissues (Figure 8). OSM mRNA was clearly detected in the liver, starting from E12 through the neonatal stage. On the other hand, OSMR mRNA expression became appar
Referência(s)