The Merlin/NF2 tumor suppressor functions through the YAP oncoprotein to regulate tissue homeostasis in mammals

Abstract

The conserved Hippo signaling pathway regulates organ size in Drosophila and mammals. While a core kinase cascade leading from the protein kinase Hippo (Hpo) (Mst1 and Mst2 in mammals) to the transcription coactivator Yorkie (Yki) (YAP in mammals) has been established, upstream regulators of the Hippo kinase cascade are less well defined, especially in mammals. Using conditional knockout mice, we demonstrate that the Merlin/NF2 tumor suppressor and the YAP oncoprotein function antagonistically to regulate liver development. While inactivation of Yap led to loss of hepatocytes and biliary epithelial cells, inactivation of Nf2 led to hepatocellular carcinoma and bile duct hamartoma. Strikingly, the Nf2-deficient phenotypes in multiple tissues were largely suppressed by heterozygous deletion of Yap, suggesting that YAP is a major effector of Merlin/NF2 in growth regulation. Our studies link Merlin/NF2 to mammalian Hippo signaling and implicate YAP activation as a mediator of pathologies relevant to Neurofibromatosis 2.

Figure 1. Impaired liver function and bile duct development in Yap-deficient livers

(A) Gross image of control and Yap mutant livers from 1-year-old littermates. Note the pale and enlarged mutant liver. Scale bar = 1 cm. (B) Increased liver-to-body weight ratio in Yap mutant mice. Mutant and control littermates were analyzed at indicated ages (values are means ± SEM): 1–2 months, n = 6; 3–5 months, n = 20; 6–12 months, n = 14. *p < 0.05, **p < 0.01, t-test. (C) Steatosis in Yap mutant livers, as revealed by H&E staining. Note the presence of fat accumulation in mutant hepatocytes (yellow arrows). Scale bar = 50 μm. (D) Fibrosis in Yap mutant livers, as evaluated by Sirius Red staining, a specific method for collagen detection. Control and mutant livers were analyzed at the indicated ages. Scale bar = 100 μm. (E) Increased serum bilirubin and alanine aminotransferase (ALT) levels in Yap mutant mice. Control and mutant livers were analyzed at the indicated ages (values are means ± SEM): 1–2 months, n = 6; 3–5 months, n = 14; 6–12 months, n = 9. *p < 0.05, **p < 0.01, t-test. (F) Impaired bile duct development in Yap mutant livers, as revealed by cytokeratin (CK) staining at the indicated ages. While CK-positive BECs in normal livers progressed from ductal plates to tubular structures around the portal veins from E18.5 to P7, the CK-positive BECs in Yap mutant livers were scattered around the portal veins and did not form typical tubular structures, leading to an absence of bile ducts at P14. An overabundance of CK-positive cells was present in 8-week-old Yap mutant livers, but these cells appeared disorganized and not incorporated into the portal mesenchyme, and did not persist in older mice (30-week-old). Scale bar = 50 μm. See Figure S1 for data supplemental to Figure 1.

Figure 2. Increased hepatocyte turnover in Yap-deficient livers

(A) H&E staining of control and Yap mutant livers, showing apoptotic areas of hepatocytes in the mutant liver (asterisks). Scale bar = 100 μm. (B) Cleaved Caspase-3 staining (red), showing apoptotic areas of hepatocytes in the mutant liver. Scale bar = 100 μm. (C) Quantification of TUNEL-positive hepatocytes in control and Yap mutant livers. Values are means ± SEM (n = 3). *p < 0.05, **p < 0.01, t-test. (D) Increased mitosis in Yap mutant livers, as evaluated by H&E staining. Arrowheads point to mitotic hepatocytes, which are shown in higher magnification in the insets. Scale bar = 50 μm. (E) Quantification of Ki67-positive hepatocytes in control and Yap mutant livers. Values are means ± SEM (n = 4). **p < 0.01, ***p < 0.001, t-test. (F) Decreased viability of Yap mutant hepatocytes. Hepatocytes isolated from control and Yap mutant livers were cultured in vitro. Viable cell numbers were measured at indicated time points and plotted as percentage of viable cells relative to day 0. Values are means ± SEM (n=3). See Figure S2 for data supplemental to Figure 2.

Figure 3. Loss of Yap suppresses Nf2 mutant phenotypes

Unless otherwise indicated, livers from 1-year-old mice of the indicated genotypes were analyzed. (A) Gross liver images. Note that the Nf2 mutant liver shows massive overgrowth and formation of hepatocellular carcinoma (HCC) (indicated by arrows and confirmed by H&E staining in D). Scale bar = 1 cm. (B–C) H&E staining (B) and CK staining (C) showing the prevalence of bile duct hamartoma in Nf2 mutant livers and the absence of bile duct hamartoma in Nf2 Yap double mutant livers. Insets in C are higher magnifications of representative portal areas: normal bile duct in control liver; multiple bile ducts in Nf2 mutant liver; no bile duct in Yap mutant liver; some portal areas have no bile ducts, whereas others have CK-positive cells but do not form tubular structures in Nf2 Yap double mutant liver. Scale bars = 250 μm. (D) H&E staining showing HCC in Nf2 mutant livers and the absence of HCC in Nf2 Yap double mutant livers. Scale bar = 60 μm. (E) Comparison of bile ducts around portal veins in E18.5 livers by CK staining. Note the overabundance of CK-positive BECs and multiple focal dilations of ductal plates in Nf2 mutant livers, and the diminished BECs in Yap and Nf2 Yap double mutant livers. Scale bar = 30 μm. (F) Quantification of liver-to-body weight ratio. Note the severe hepatomegaly of Nf2 mutant livers and the similar size of Yap and Nf2 Yap double mutant livers. Values represent means ± SEM (n = 3). (G) Quantification of CK-positive BECs and ductule structures for E18.5 livers. Values represent means ± SEM (n = 3). (H) Quantitative real-time PCR analysis of selected genes from livers of control, Nf2, Yap and Nf2 Yap mice at the age of E18.5. OPN and EpCAM expression was increased in Nf2 livers, but decreased in Yap and Nf2 Yap livers. Jagged 1, Notch2 and Hes1 expression was not affected in any of the genetic background. Values are means ± SEM (n=3). See Figure S3 for data supplemental to Figure 3.

Figure 4. Heterozygosity of Yap suppresses Nf2 mutant liver phenotypes

(A) Gross liver images from two littermates at 16 months of age. Left: Alb-Cre; Nf2^flox2/flox2; Yap^+/+. Right: Alb-Cre; Nf2^flox2/flox2; Yap^flox/+. Note the massive overgrowth and tumor formation in Nf2 mutant liver (left) and the normal appearance of Nf2, Yap+/− liver (right). Scale bar = 1 cm. (B–C) H&E staining (B) and CK staining (C) showing the prevalence of bile duct hamartoma in Nf2 mutant livers (left) and the greatly diminished bile duct hyperplasia in Nf2 Yap+/− livers (right). Scale bars = 125 μm. (D) H&E staining showing HCC in Nf2 mutant liver (left) and the absence of HCC in Nf2 Yap+/− liver (right). Scale bar = 60 μm. (E) Control (top), Nf2 (middle) and Nf2 Yap+/− (bottom) livers from P0 animals were analyzed for Ki67 staining (green) and counterstained for the BEC marker cytokeratin (red). Note the abundance of Ki67-positive BECs in Nf2 livers, but not in control or Nf2 Yap+/− livers. Scale bar = 20 μm. (F) Similar to (E) except that apoptosis was analyzed by TUNEL staining (green). Note the general absence of TUNEL-positive BECs in all genotypes. (G) Quantification of Ki67- and TUNEL-positive BECs from (E–F). Values are means ± SEM (n = 5). ***, p < 0.001, t-test. See Figure S4 for data supplemental to Figure 4.

Figure 5. Yap heterozygosity suppresses cataracts induced by loss of Nf2

(A–C) H&E staining of lens from 2-month old littermates with the following genotypes: (A) Nf2^flox2/flox2 (as control), (B) Nes-Cre; Nf2^flox2/flox2, and (C) Nes-Cre; Nf2^flox2/flox2; Yap^flox/+. Note the general loss of structural integrity in (B), but not in (A) or (C). In (A), the anterior, equator and posterior regions of lens are marked by A, Eq and P, respectively. (D–L) High magnification views of lens in (A–C) highlighting anterior, equator and posterior regions of the lens. Note the disorganization of anterior lens epithelium in Nes-Cre; Nf2^flox2/flox2, and the normal monolayer cuboidal epitheliaium in Nes-Cre; Nf2^flox2/flox2; Yap^flox/+ lens (compare arrowheads in D–F). Also note the accumulation of ectopic cells and capsular material at the equator region of Nes-Cre; Nf2^flox2/flox2 (asterisk, H), but not Nes-Cre; Nf2^flox2/flox2; Yap^flox/+ (I) lens. Furthermore, the posterior lens, which is normally accellular (J; appearing as hematoxylin-negative), shows abberant accumulation of cells in Nes-Cre; Nf2^flox2/flox2 (asterisk, K), but not Nes-Cre; Nf2^flox2/flox2; Yap^flox/+ (L) lens. Scale bars = 100 μm.

Figure 6. NF2 is required for Hippo signaling and physically associates with canonical pathway components

(A) Western blot analysis of liver lysates from 2-month-old mice. Liver lysates from 3 mice of each indicated genotype were probed with antibodies against P-YAP, YAP, PLats, Lats1, P-Erk1/2, Erk1/2, NF2 and Actin. Note the decreased levels of YAP and Lats1/2 phosphorylation in Nf2 mutant livers. (B) Confocal images of periportal areas from control and Nf2-deficient livers (at 2-month of age), stained for YAP (red) and DAPI (blue). Note that in wildtype hepatocytes (top), YAP staining shows a honeycomb-like pattern, with less YAP in the nucleus than in the cytoplasm (arrows). In contrast, Nf2-deficient cells (bottom) commonly show uniform distribution of YAP inside and outside of the nucleus. Representative cells are denoted by arrows. Scale bar = 20 μm. (C) Physical association between NF2 and KIBRA. Cell lysate from HEK293 cells without (lane 1) or with (lane 2) T7-KIBRA expression was immunoprecipitated (IP) with α-T7 antibody and probed with antibody against endogenous NF2 and α-T7 antibody. Endogenous NF2 was detected in T7-KIBRA immunoprecipitates. (D) Physical association between NF2 and WW45. α-WW45 antibody was used to immnoprecipitate endogenous WW45 from untransfected HEK293 cells and probed with α-NF2 or α-WW45 antibody. Endogenous NF2 was detected in WW45-immnoprecipitates, not in immunoprecipitates with control IgG. (E) Lysates from HEK293 (left) or ACHN (right) cells expressing Myc-Lats2 with various combinations of Myc-WW45, T7-KIBRA and HA-NF2 were probed with the indicated antibodies. Note the stimulation of Lats2 phosphorylation by KIBRA-NF2 (lane 3) or KIBRA-NF2-WW45 (lane 4), but not WW45 (lane 2), in HEK293 cells. Also note the failure of KIBRA-NF2 (lane 7) to stimulate Lats2 phosphorylation, and the rescue of this defect by WW45 co-expression (lane 8). * denotes a non-specific band in α-KIBRA (T7) blots.