- Email: [email protected]
A Cellular View of Nf2 in Liver Homeostasis and Tumorigenesis
Developmental Cell
Previews A Cellular View of Nf2 in Liver Homeostasis and Tumorigenesis Kai Breuhahn1 and Peter Schirmacher1,* 1Institute of Pathology, University Hospital Heidelberg, 69120 Heidelberg, Germany *Correspondence: [email protected] DOI 10.1016/j.devcel.2010.08.015
How liver adjusts and stabilizes its size is unsolved so far; the answers to this question may also provide insights into mechanisms of hepatocarcinogenesis. Two recent papers suggest a role for Merlin/Nf2 in control of liver cell turnover, but results appear conflicting at first glance. Maintenance of organ size through physiological cell turnover and rapid restoration of original weight and cellularity after resection is one of the most enigmatic abilities of the liver; it does not take more than 7 days to regain its original size after partial hepatectomy. Understanding the underlying mechanisms and their dysregulation is likely to provide key concepts for molecular liver carcinogenesis. While the essential role of growth factor-induced signaling in hepatocellular proliferation and hepatocarcinogenesis is well established (Breuhahn et al., 2006), the mechanism of liver size adjustment and cellularity (‘‘hepatostat’’) has escaped detection. Previous data demonstrated that the evolutionarily conserved Hippo kinase pathway represents a potent regulator of organ size in Drosophila (Hpo/ Wts/Yki) and mammals (Mst/Lats/YAP). Depending on cell density, its activation reduces cell proliferation and survival due to phosphorylation and cytoplasmic retention of the transcriptional coregulator Yki/YAP. Its dysregulation leads to tissue overgrowth and tumorigenesis in vivo (Dong et al., 2007; Camargo et al., 2007). Diverse physiological regulators of this pathway, including Merlin (Hamaratoglu et al., 2006), have been identified in Drosophila; however, the regulatory network controlling YAP activity in mammals is far from being understood. The mammalian Merlin homolog Nf2 (neurofibromin 2) is a tumor suppressor mutated in neurofibromatosis, a genetic disorder leading to multiple benign and finally malignant mesenchymal tumors. The role of Nf2 in liver homeostasis and tumorigenesis has recently been approached by two papers (Benhamouche et al., 2010; Zhang et al., 2010) using an identical model: a conditional, liver-specific
deletion of Nf2 (Alb-Cre;Nf2lox2/lox2), which results in hepatomegaly (liver overgrowth) and subsequent development of multiple malignant primary liver tumors, similar to phenotypes resulting from overexpression of YAP (Camargo et al., 2007; Dong et al., 2007) or the loss of Mst1/2 (Lu et al., 2010; Zhou et al., 2009) or WW45 (a homolog of Drosophila Salvador and adaptor for the Hippo kinase; Lu et al., 2010; Lee et al., 2010). But here the commonalities between the two studies end. In genetic experiments, Zhang et al. demonstrate that YAP heterozygosity (which on its own confers no obvious phenotypic effect) largely rescues the hepatic Nf2deletion phenotype, consistent with the Drosophila pathway. However, Benhamouche et al. use phosphorylation analyses and drug-based inhibition experiments to show a connection to EGFR signaling instead. Regarding Hippo pathway kinase activation and YAP regulation, both groups arrive at opposite conclusions. There is also discordance on the observed phenotypes: while Zhang et al. describe the development of HCCs and biliary hamartomas, Benhamouche et al. observe the whole spectrum of malignant hepatocellular, biliary, and mixed liver tumors. Finally, Zhang et al. describe a hepatocellular phenotype (increased turnover/proliferation of hepatocytes) that is not observed by Benhamouche et al.; instead, the latter group observes extensive proliferation of putative liver progenitor cells (‘‘oval cells,’’ see below) preceding liver tumor formation, suggesting tumor development from this cell population. The differences are quite significant and require a close look also from a cell biology point of view.
In the normal liver after partial hepatectomy, most differentiated hepatocytes enter one to two rounds of replication; this is sufficient for hepatocellular regeneration with no contribution from any other cell population. If hepatocyte replication is blocked (e.g., by drugs), there is a back-up mechanism of poorer efficiency based on proliferation of small round cells that although biased to biliary differentiation, seem to have some hepatocytic differentiation potential. Carcinogens induce morphologically similar ‘‘oval cells’’ (not necessarily the same cells) that have been extracted, propagated, and immunologically characterized. In rodents, oval cells may give rise to tumors of hepatocellular, cholangiocellular, and mixed differentiation. Both manuscripts demonstrate enlargement of the liver, due partly to extensive ductal proliferation. These lesions are progressive and take over the majority of liver tissue. Thus, calling them hamartomas (localized, stable, nonneoplastic malformations) is not correct, but frankly, calling them oval cells, implying a bipotent precursor with both hepatocytic and cholangiocytic potential, may also miss the mark. These ductal proliferations appear to be well delineated toward the parenchyma and confined to expanding portal tracts/septa and do not show a diffuse intercalating migratory phenotype frequently observed in rodent models with oval cell proliferation. Neither paper provides evidence that these ducts give rise to hepatocellular progeny, and as differentiated ducts with lumen, they seem committed to biliary differentiation. Varying differentiation of resulting tumors does not prove origin from a bipotent precursor cell, since secondary tumor cell plasticity (i.e., relaxation of a strict
Developmental Cell 19, September 14, 2010 ª2010 Elsevier Inc. 363
Developmental Cell
Previews differentiation pattern) is a common phenomenon in tumors potentially related to acquired epigenetic changes. How may the perceived differences in signaling be explained? It is important to note that the major premalignant phenotype appears to reside in a biliary cell type, while the primary target cell population both groups have aimed for with their Alb-Cre construct (hepatoblasts/hepatocytes) shows less, if any, phenotype. Conversely, YAP is required for biliary development (Zhang et al., 2010). These considerations leave room for alternative explanations: Nf2 deficiency may initially contribute to biliary differentiation of an albumin-expressing cell population (e.g., a subpopulation of hepatoblasts) via YAP, thus explaining the sensitivity of the Nf2 mutant phenotype to YAP levels. Once these cells have attained biliary differentiation, the Nf2/YAP dependency is terminated and Nf2/EGFR addiction is induced. EGFR signaling becomes responsible for continuous proliferation, rendering these cells susceptible to EGFR blockade. Alternatively, YAP may serve in this context not as an Nf2 effector but as a modulator of downstream (e.g., EGFR) signaling; such crosstalk would not be unprecedented, but would contrast with some cell culture experiments of Zhang et al. and the current understanding of Merlin signaling in Drosophila. What is the take-home message: (1) Nf2 deficiency leads to multiple malignant
liver tumors via a persistently proliferating yet rather differentiated ductal cell population that seems to have premalignant properties, and 2) there is evidence that in certain cell types and under certain conditions, YAP as well as EGFR signaling are relevant effectors of Nf2. Important questions are so far unsolved: which liver cell populations are autonomously affected by YAP- and Nf2-deletion effects, and is there a primary hepatocellular phenotype and thus a true hepatocyte function of Nf2? What are the relevant Nf2-induced downstream effector mechanisms in the different liver cell types? Does the proliferating ductal cell population have any hepatocellular differentiation potential in vitro and in vivo, as required for a bipotent progenitor, or does it show a largely biliary phenotype? Answers will largely depend on an unbiased and meticulous definition of the relevant liver cell types and their resulting alterations. Are these findings relevant to the human situation? So far, this is unclear; there is no defined human disease counterpart to autonomous ductal proliferation. Moreover, Nf2 patients are not prone to liver cancer despite living long enough to experience it; preliminary screens have failed to detect Nf2 mutations in hepatocellular carcinomas (Kanai et al., 1995). Irrespective of the direct applicability of these tumor phenotypes to the human condition, these papers clearly highlight the complexity and cell type-
364 Developmental Cell 19, September 14, 2010 ª2010 Elsevier Inc.
specificity of tumor suppressor effects on downstream signaling and point the way toward further study of this important issue. REFERENCES Benhamouche, S., Curto, M., Saotome, I., Gladden, A.B., Liu, C.H., Giovannini, M., and McClatchey, A.I. (2010). Genes Dev. 24, 1718– 1730. Breuhahn, K., Longerich, T., and Schirmacher, P. (2006). Oncogene 25, 3787–3800. Camargo, F.D., Gokhale, S., Johnnidis, J.B., Fu, D., Bell, G.W., Jaenisch, R., and Brummelkamp, T.R. (2007). Curr. Biol. 17, 2054–2060. Dong, J., Feldmann, G., Huang, J., Wu, S., Zhang, N., Comerford, S.A., Gayyed, M.F., Anders, R.A., Maitra, A., and Pan, D. (2007). Cell 130, 1120– 1133. Hamaratoglu, F., Willecke, M., Kango-Singh, M., Nolo, R., Hyun, E., Tao, C., Jafar-Nejad, H., and Halder, G. (2006). Nat. Cell Biol. 8, 27–36. Kanai, Y., Tsuda, H., Oda, T., Sakamoto, M., and Hirohashi, S. (1995). Jpn. J. Clin. Oncol. 25, 1–4. Lee, K.P., Lee, J.H., Kim, T.S., Kim, T.H., Park, H.D., Byun, J.S., Kim, M.C., Jeong, W.I., Calvisi, D.F., Kim, J.M., and Lim, D.S. (2010). Proc. Natl. Acad. Sci. USA 107, 8248–8253. Lu, L., Li, Y., Kim, S.M., Bossuyt, W., Liu, P., Qiu, Q., Wang, Y., Halder, G., Finegold, M.J., Lee, J.S., et al. (2010). Proc. Natl. Acad. Sci. USA 107, 1437–1442. Zhang, N., Bai, H., David, K.K., Dong, J., Zheng, Y., Cai, J., Giovannini, M., Liu, P., Anders, R.A., and Pan, D. (2010). Dev. Cell 19, 27–38. Zhou, D., Conrad, C., Xia, F., Park, J.S., Payer, B., Yin, Y., Lauwers, G.Y., Thasler, W., Lee, J.T., Avruch, J., et al. (2009). Cancer Cell 16, 425–438.
Previews A Cellular View of Nf2 in Liver Homeostasis and Tumorigenesis Kai Breuhahn1 and Peter Schirmacher1,* 1Institute of Pathology, University Hospital Heidelberg, 69120 Heidelberg, Germany *Correspondence: [email protected] DOI 10.1016/j.devcel.2010.08.015
How liver adjusts and stabilizes its size is unsolved so far; the answers to this question may also provide insights into mechanisms of hepatocarcinogenesis. Two recent papers suggest a role for Merlin/Nf2 in control of liver cell turnover, but results appear conflicting at first glance. Maintenance of organ size through physiological cell turnover and rapid restoration of original weight and cellularity after resection is one of the most enigmatic abilities of the liver; it does not take more than 7 days to regain its original size after partial hepatectomy. Understanding the underlying mechanisms and their dysregulation is likely to provide key concepts for molecular liver carcinogenesis. While the essential role of growth factor-induced signaling in hepatocellular proliferation and hepatocarcinogenesis is well established (Breuhahn et al., 2006), the mechanism of liver size adjustment and cellularity (‘‘hepatostat’’) has escaped detection. Previous data demonstrated that the evolutionarily conserved Hippo kinase pathway represents a potent regulator of organ size in Drosophila (Hpo/ Wts/Yki) and mammals (Mst/Lats/YAP). Depending on cell density, its activation reduces cell proliferation and survival due to phosphorylation and cytoplasmic retention of the transcriptional coregulator Yki/YAP. Its dysregulation leads to tissue overgrowth and tumorigenesis in vivo (Dong et al., 2007; Camargo et al., 2007). Diverse physiological regulators of this pathway, including Merlin (Hamaratoglu et al., 2006), have been identified in Drosophila; however, the regulatory network controlling YAP activity in mammals is far from being understood. The mammalian Merlin homolog Nf2 (neurofibromin 2) is a tumor suppressor mutated in neurofibromatosis, a genetic disorder leading to multiple benign and finally malignant mesenchymal tumors. The role of Nf2 in liver homeostasis and tumorigenesis has recently been approached by two papers (Benhamouche et al., 2010; Zhang et al., 2010) using an identical model: a conditional, liver-specific
deletion of Nf2 (Alb-Cre;Nf2lox2/lox2), which results in hepatomegaly (liver overgrowth) and subsequent development of multiple malignant primary liver tumors, similar to phenotypes resulting from overexpression of YAP (Camargo et al., 2007; Dong et al., 2007) or the loss of Mst1/2 (Lu et al., 2010; Zhou et al., 2009) or WW45 (a homolog of Drosophila Salvador and adaptor for the Hippo kinase; Lu et al., 2010; Lee et al., 2010). But here the commonalities between the two studies end. In genetic experiments, Zhang et al. demonstrate that YAP heterozygosity (which on its own confers no obvious phenotypic effect) largely rescues the hepatic Nf2deletion phenotype, consistent with the Drosophila pathway. However, Benhamouche et al. use phosphorylation analyses and drug-based inhibition experiments to show a connection to EGFR signaling instead. Regarding Hippo pathway kinase activation and YAP regulation, both groups arrive at opposite conclusions. There is also discordance on the observed phenotypes: while Zhang et al. describe the development of HCCs and biliary hamartomas, Benhamouche et al. observe the whole spectrum of malignant hepatocellular, biliary, and mixed liver tumors. Finally, Zhang et al. describe a hepatocellular phenotype (increased turnover/proliferation of hepatocytes) that is not observed by Benhamouche et al.; instead, the latter group observes extensive proliferation of putative liver progenitor cells (‘‘oval cells,’’ see below) preceding liver tumor formation, suggesting tumor development from this cell population. The differences are quite significant and require a close look also from a cell biology point of view.
In the normal liver after partial hepatectomy, most differentiated hepatocytes enter one to two rounds of replication; this is sufficient for hepatocellular regeneration with no contribution from any other cell population. If hepatocyte replication is blocked (e.g., by drugs), there is a back-up mechanism of poorer efficiency based on proliferation of small round cells that although biased to biliary differentiation, seem to have some hepatocytic differentiation potential. Carcinogens induce morphologically similar ‘‘oval cells’’ (not necessarily the same cells) that have been extracted, propagated, and immunologically characterized. In rodents, oval cells may give rise to tumors of hepatocellular, cholangiocellular, and mixed differentiation. Both manuscripts demonstrate enlargement of the liver, due partly to extensive ductal proliferation. These lesions are progressive and take over the majority of liver tissue. Thus, calling them hamartomas (localized, stable, nonneoplastic malformations) is not correct, but frankly, calling them oval cells, implying a bipotent precursor with both hepatocytic and cholangiocytic potential, may also miss the mark. These ductal proliferations appear to be well delineated toward the parenchyma and confined to expanding portal tracts/septa and do not show a diffuse intercalating migratory phenotype frequently observed in rodent models with oval cell proliferation. Neither paper provides evidence that these ducts give rise to hepatocellular progeny, and as differentiated ducts with lumen, they seem committed to biliary differentiation. Varying differentiation of resulting tumors does not prove origin from a bipotent precursor cell, since secondary tumor cell plasticity (i.e., relaxation of a strict
Developmental Cell 19, September 14, 2010 ª2010 Elsevier Inc. 363
Developmental Cell
Previews differentiation pattern) is a common phenomenon in tumors potentially related to acquired epigenetic changes. How may the perceived differences in signaling be explained? It is important to note that the major premalignant phenotype appears to reside in a biliary cell type, while the primary target cell population both groups have aimed for with their Alb-Cre construct (hepatoblasts/hepatocytes) shows less, if any, phenotype. Conversely, YAP is required for biliary development (Zhang et al., 2010). These considerations leave room for alternative explanations: Nf2 deficiency may initially contribute to biliary differentiation of an albumin-expressing cell population (e.g., a subpopulation of hepatoblasts) via YAP, thus explaining the sensitivity of the Nf2 mutant phenotype to YAP levels. Once these cells have attained biliary differentiation, the Nf2/YAP dependency is terminated and Nf2/EGFR addiction is induced. EGFR signaling becomes responsible for continuous proliferation, rendering these cells susceptible to EGFR blockade. Alternatively, YAP may serve in this context not as an Nf2 effector but as a modulator of downstream (e.g., EGFR) signaling; such crosstalk would not be unprecedented, but would contrast with some cell culture experiments of Zhang et al. and the current understanding of Merlin signaling in Drosophila. What is the take-home message: (1) Nf2 deficiency leads to multiple malignant
liver tumors via a persistently proliferating yet rather differentiated ductal cell population that seems to have premalignant properties, and 2) there is evidence that in certain cell types and under certain conditions, YAP as well as EGFR signaling are relevant effectors of Nf2. Important questions are so far unsolved: which liver cell populations are autonomously affected by YAP- and Nf2-deletion effects, and is there a primary hepatocellular phenotype and thus a true hepatocyte function of Nf2? What are the relevant Nf2-induced downstream effector mechanisms in the different liver cell types? Does the proliferating ductal cell population have any hepatocellular differentiation potential in vitro and in vivo, as required for a bipotent progenitor, or does it show a largely biliary phenotype? Answers will largely depend on an unbiased and meticulous definition of the relevant liver cell types and their resulting alterations. Are these findings relevant to the human situation? So far, this is unclear; there is no defined human disease counterpart to autonomous ductal proliferation. Moreover, Nf2 patients are not prone to liver cancer despite living long enough to experience it; preliminary screens have failed to detect Nf2 mutations in hepatocellular carcinomas (Kanai et al., 1995). Irrespective of the direct applicability of these tumor phenotypes to the human condition, these papers clearly highlight the complexity and cell type-
364 Developmental Cell 19, September 14, 2010 ª2010 Elsevier Inc.
specificity of tumor suppressor effects on downstream signaling and point the way toward further study of this important issue. REFERENCES Benhamouche, S., Curto, M., Saotome, I., Gladden, A.B., Liu, C.H., Giovannini, M., and McClatchey, A.I. (2010). Genes Dev. 24, 1718– 1730. Breuhahn, K., Longerich, T., and Schirmacher, P. (2006). Oncogene 25, 3787–3800. Camargo, F.D., Gokhale, S., Johnnidis, J.B., Fu, D., Bell, G.W., Jaenisch, R., and Brummelkamp, T.R. (2007). Curr. Biol. 17, 2054–2060. Dong, J., Feldmann, G., Huang, J., Wu, S., Zhang, N., Comerford, S.A., Gayyed, M.F., Anders, R.A., Maitra, A., and Pan, D. (2007). Cell 130, 1120– 1133. Hamaratoglu, F., Willecke, M., Kango-Singh, M., Nolo, R., Hyun, E., Tao, C., Jafar-Nejad, H., and Halder, G. (2006). Nat. Cell Biol. 8, 27–36. Kanai, Y., Tsuda, H., Oda, T., Sakamoto, M., and Hirohashi, S. (1995). Jpn. J. Clin. Oncol. 25, 1–4. Lee, K.P., Lee, J.H., Kim, T.S., Kim, T.H., Park, H.D., Byun, J.S., Kim, M.C., Jeong, W.I., Calvisi, D.F., Kim, J.M., and Lim, D.S. (2010). Proc. Natl. Acad. Sci. USA 107, 8248–8253. Lu, L., Li, Y., Kim, S.M., Bossuyt, W., Liu, P., Qiu, Q., Wang, Y., Halder, G., Finegold, M.J., Lee, J.S., et al. (2010). Proc. Natl. Acad. Sci. USA 107, 1437–1442. Zhang, N., Bai, H., David, K.K., Dong, J., Zheng, Y., Cai, J., Giovannini, M., Liu, P., Anders, R.A., and Pan, D. (2010). Dev. Cell 19, 27–38. Zhou, D., Conrad, C., Xia, F., Park, J.S., Payer, B., Yin, Y., Lauwers, G.Y., Thasler, W., Lee, J.T., Avruch, J., et al. (2009). Cancer Cell 16, 425–438.