1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

Liver cancer is one of the most common solid tumors, with poor prognosis and high mortality. Mutation or deletion of the tumor suppressor phosphatase and tensin homolog deleted on chromosome 10 (PTEN) is strongly correlated with human liver cancer. Glucose-regulated protein 94 (GRP94) is a major endoplasmic reticulum (ER) chaperone protein, but its in vivo function is still emerging. To study the role of GRP94 in maintaining liver homeostasis and tumor development, we created two liver-specific knockout mouse models with the deletion of Grp94 alone, or in combination with Pten, using the albumin-cre system. We demonstrated that while deletion of GRP94 in the liver led to hyperproliferation of liver progenitor cells, deletion of both GRP94 and PTEN accelerated development of liver tumors, including both hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC), suggestive of progenitor cell origin. Furthermore, at the premalignant stage we observed disturbance of cell adhesion proteins and minor liver injury. When GRP94 was deleted in PTEN-null livers, ERK was selectively activated. Conclusion: GRP94 is a novel regulator of cell adhesion, liver homeostasis, and tumorigenesis. (Hepatology 2014;59:947–957)






alkaline phosphatase


alanine aminotransferase


α-smooth muscle actin




cytokeratin 19


central vein




epithelial cadherin




extracellular matrix


epithelial cell adhesion molecule


endoplasmic reticulum


extracellular signal-regulated kinase


glucose-regulated protein 78


glucose-regulated protein 94


glutamine synthetase


hepatocellular carcinoma


Hepatocyte Paraffin 1


high-mobility group protein B1


hematopoietic stem cell


heat shock protein 90


insulin-like growth factor


liver progenitor cell




phosphatase and tensin homolog deleted on chromosome 10


portal vein


tumor necrosis factor-α

The two most common types of liver cancer are hepatocellular carcinoma (HCC) and cholangiocarcinoma (CC), arising from hepatocytes and cholangiocytes (bile duct cells), respectively.[1] Liver tumor with liver progenitor cell (LPC) characteristics is particularly aggressive, with poor prognosis.[1] Genetically defined liver cancer mouse models have provided important experimental tools to study the role of LPCs in live tumorigenesis. For example, the conditional knockout of the Pten tumor suppressor gene by albumin-Cre (Alb-Cre) leads to liver injury, LPC proliferation, and liver cancer development, including both HCC and CC.[2] LPCs are bipotential and quiescently reside in the stem cell niche, located in the most peripheral branches of the biliary tree. During chronic or massive liver injury, LPCs are activated and differentiate into hepatocytes and cholangiocytes. Deregulated LPCs can give rise to liver cancer.[1] Nevertheless, the regulation of LPC proliferation and its role in liver tumorigenesis are not well understood.

Glucose-regulated protein 94 (GRP94) is a major endoplasmic reticulum (ER) chaperone protein, assisting protein folding, processing, and secretion, and is the ER counterpart of HSP90.[3] Client proteins of GRP94 include cell adhesion and signaling molecules such as integrins, Toll-like receptors, and insulin-like growth factors (IGFs), suggesting that GRP94 has unique functions controlling specific pathways critical for cell adhesion, immune modulation, and growth signaling.[4] Hepatocytes communicate with each other through gap junctional channels composed of connexin proteins, of which Cx26 and Cx32 are most abundant.[5] Furthermore, disruption of cell adhesion mediated by connexins, adherens junction protein E-cadherin, and integrin β1 has been linked to tumorigenesis.[5-7] However, the functional interaction between GRP94 and these adhesion molecules is unknown.

Recently, we discovered that while homozygous mutation of Grp94 in mice results in embryonic lethality,[8, 9] inducible knockout of Grp94 in adult mice leads to the loss of attachment of the hematopoietic stem cells (HSCs) in the bone marrow niche and increased HSC proliferation.[10, 11] This suggests that GRP94 may also regulate other stem cell pools and tumorigenesis arising from deregulated stem cell proliferation. Here we report the creation of two liver-specific knockout mouse models with the deletion of Grp94 alone or in combination with Pten. Our studies revealed that GRP94 deficiency led to hyperproliferation of LPCs, correlating with impaired cell adhesion. Deletion of both GRP94 and PTEN accelerated HCC and CC development with minor liver injury and that ERK was selectively activated. These studies uncover a novel role of GRP94 in regulating liver physiology and tumorigenesis.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

Grp94f/f mice on a mixed C57BL/6; 129/Sv background[8] were crossed with Ptenf/f mice on a C57BL/6; 6xDBA2; 129 background[12] to generate Ptenf/fGrp94f/f mice, which were mated with the transgenic Alb-Cre; Ptenf/f mice on a C57BL/6; J129svj background[2] to generate Alb-Cre; Ptenf/fGrp94f/f and Alb-Cre; Grp94f/f mice. Genotyping by polymerase chain reaction (PCR) was previously described.[12] Blood samples were collected through retro-orbital bleeding. All protocols for animal use were reviewed and approved by the USC Institutional Animal Care and Use Committee.

Plasma Biochemistry

Plasma alanine aminotransferase (ALT) was determined using ALT Reagent (Raichem, San Diego, CA). Plasma total bilirubin and alkaline phosphatase (ALP) were measured following the manufacturer's instruction (Thermo Scientific, Waltham, MA).


Apoptosis was determined using terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining (Roche Diagnostics, Mannheim, Germany).

Real-Time Quantitative PCR

RNA was extracted from mouse livers and reverse-transcription and real-time PCR were performed as described.[10] Primers used for AFP, EpCAM, CK19, and 18S RNA have been described.[2, 10]

Statistical Analysis

Statistical significance was assayed by 2-tailed Student t test and the error bars reflect the standard error (SE).

Histology and Immunostaining

See the Supporting Information.

Western Blot Analysis

See the Supporting Information.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information
Hyperproliferation of Liver Progenitor Cells in cGrp94f/f Mice

To study the requirement of GRP94 in liver homeostasis, we created a liver-specific Grp94 knockout mouse model (Alb-Cre; Grp94f/f or cGrp94f/f). Littermates lacking Alb-Cre served as wild-type (WT) controls. In livers isolated from 2-month-old cGrp94f/f mice, Grp94 allele deletion was confirmed by PCR (Fig. 1A). Loss of GRP94 protein expression was confirmed by western blot, which also revealed a mild (1.3-fold) compensatory increase of another ER chaperone GRP78 (Fig. 1B). cGrp94f/f livers appeared normal except that they were about 25% smaller than the WT, and in some mice the surface acquired nodular appearance by 9 months (Fig. 1C). At both 2 and 9 months, hematoxylin and eosin (H&E) staining revealed an increase of mononuclear cells that fit the morphological description of mouse LPCs in cGrp94f/f livers compared to the WT (Fig. 1D). Ki67 staining further demonstrated progressive increase of highly proliferative cells adjacent to portal veins (PVs) in cGrp94f/f livers, while WT livers were relatively quiescent (Fig. 1D). Double staining of Ki67 and LPC marker A6 confirmed significantly increased Ki67-positive cells and more double-positive cells around PVs, such that in cGrp94f/f livers at 9 months, about 48% of the Ki67-positive cells were costained with A6 (Fig. 1E,F). Double staining of Ki67 with the mesenchymal cell marker α-smooth muscle actin (α-SMA) yielded a few double-positive cells, whereas about 38%-50% of the Ki67-positive cells were costained with the hematopoietic cell marker CD45 in both WT and cGrp94f/f livers (Supporting Fig. S1). For cGrp94f/f livers at 9 months, the small nodules visible in some mice were not tumors by histological analysis, and glutamine synthetase (GS) staining revealed similar liver zonation pattern but some GS-marked hepatocytes were not around central veins (Fig. S2).


Figure 1. Hyperproliferation of cells adjacent to portal veins in cGrp94f/f livers. (A) Liver PCR genotyping and (B) western blot of liver lysates at 2 months. (C) Liver appearance and weight at indicated ages. (D) Liver H&E staining and Ki67 staining of proliferative cells (black arrows) at 2 months and 9 months. Insets show 2× magnification. PV: portal vein. (E) Immunofluorescence staining with LPC marker A6 (green) and Ki67 (red) in cGrp94f/f frozen liver sections. White arrows denote double-positive cells. Nuclei were stained with DAPI (blue). (F) Quantitation of Ki67+ cells (left) (Student t test) and A6, Ki67 double-positive cells (right) (χ2 test). Scale bar: 25 μm. All data are presented as mean ± SE (*P < 0.05, **P < 0.01, and ***P < 0.001).

Download figure to PowerPoint

Consistently, immunofluorescence staining showed LPC markers A6 and pan-cytokeratin (panCK) labeling multiple layers of periportal cells in cGrp94f/f livers, whereas in WT livers these same markers were restricted to bile duct epithelial cells (Fig. 2A). Quantitation of A6-positive cells showed progressive expansion from 2 to 9 months for both genotypes, with cGrp94f/f livers containing a higher percentage of PVs bearing large numbers (over 50) of A6-positive cells at both 2 and 9 months (Fig. 2B). Additionally, more A6-positive cells extending from PVs into liver parenchyma, indicative of migratory capability, were observed in cGrp94f/f livers (Fig. 2C).


Figure 2. Increased LPC pool in cGrp94f/f mice. (A) Immunofluorescence staining with LPC markers A6 (green) and panCK (red) in frozen liver sections at 2 months. Nuclei were stained with DAPI (blue). (B) Quantitation of PVs bearing over 50 A6+ cells (**P < 0.01 and ***P < 0.001; χ2 test). (C) Quantitation and examples of free A6+ cells (white arrows) in cGrp94f/f livers. (D) Plasma ALT, total bilirubin, and ALP measurements (*P < 0.05 and **P < 0.01). (E) TUNEL staining (red) of liver sections for apoptotic cells (white arrows) at 2 months. (F) Necrosis marker HMGB1 staining. PV: portal vein. Scale bar: 50 μm. All data are presented as mean ± SE.

Download figure to PowerPoint

Liver injury is a common cause for LPC proliferation. We detected no increase in plasma alanine aminotransferase (ALT) levels at both 2 and 9 months, but ∼2-fold and 1.2-fold increase of cholestatic markers, plasma bilirubin, and alkaline phosphatase (ALP), respectively, in cGrp94f/f mice at 2 months, which reverted back to WT levels at 9 months (Fig. 2D). Although the percentage of apoptotic cells increased from 0.2% to 0.5% in cGrp94f/f livers, the overall apoptosis was minimal (Fig. 2E). Necrosis was not detected by histological examination and the lack of cytosolic HMGB1 staining (Fig. 2F).

Disorganization of Cell Adhesion Molecules in cGrp94f/f Livers

To test whether the LPC proliferation was induced by a disruption in GRP94-mediated cell adhesion, we examined the expression pattern of two prominent liver gap junction proteins, Cx26 and Cx32. In WT livers, Cx26 and Cx32 exhibited well-structured long patches at hepatocyte cell membranes; in contrast, both Cx26 and Cx32 staining in cGrp94f/f livers appeared disorganized and dispersed (Fig. 3A,B). Western blot analysis showed slight Cx26 and more prominent Cx32 reduction in cGrp94f/f livers (Fig. S3A).


Figure 3. Disruption of cell adhesion molecules in cGrp94f/f livers at 2 months. (A,B) Immunofluorescence staining of Cx26 (A) and Cx32 (B) in frozen liver sections. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. (C) Schematic drawing of E-cadherin zonal expression pattern (red) in WT and diffuse expression (pink) in cGrp94f/f livers. Immunofluorescence costaining of integrin β1 (green) and E-cadherin (red) of the boxed areas (U1,2 and L1,2) is shown below. White arrows denote clusters of liver progenitor/bile duct cells expressing high levels of E-cadherin near PVs. (D) Immunofluorescence staining of E-cadherin (green) and panCK (red). Double-positive cells are indicated by white arrows. PV: portal vein; CV: central vein. Scale bar (C,D): 50 μm.

Download figure to PowerPoint

Cx26 regulates integrin β1 expression in breast cancer cells,[13] and E-cadherin colocalizes with Cx26 and Cx32 during gap junction reappearance in regenerating mouse hepatocytes.[14] To examine whether Cx26 and Cx32 disorganization affects expression patterns of integrin β1 and E-cadherin in cGrp94f/f livers, double staining of integrin β1 and E-cadherin were performed. First, we noted that while E-cadherin expression in WT livers showed zonal patterns where expression was restricted to periportal areas as expected, it was homogeneously expressed throughout cGrp94f/f livers (data not shown and Fig. 3C). In periportal areas of WT livers, integrin β1 and E-cadherin mostly colocalized at hepatocyte cell membranes (Fig. 3C, U1 panel). In contrast, in cGrp94f/f livers, integrin β1 was largely in the cytoplasm and the colocalization with E-cadherin was lost (Fig. 3C, U2 panel). In areas right adjacent to PVs, we detected more cells morphologically resembling progenitor/bile duct cells expressing strong E-cadherin in cGrp94f/f livers (Fig. 3C, L1 and L2 panels). Double staining of E-cadherin and LPC marker panCK confirmed that the cell clusters near PVs were LPCs/bile duct cells (Fig. 3D). Collectively, these results suggest that the cell adhesion in cGrp94f/f livers was impaired, correlating with LPC activation.

Accelerated Liver Tumorigenesis by Biallelic Deletion of Pten and Grp94 in the Liver

To test whether cell adhesion disturbance and LPC hyperproliferation in cGrp94f/f mice affect tumor development, we generated a novel hepatic Pten and Grp94 knockout mouse model (Alb-Cre; Ptenf/fGrp94f/f or cPtenf/fGrp94f/f). Deletion of Pten and Grp94 alleles and loss of protein expression in cPtenf/fGrp94f/f livers were validated by PCR (data not shown) and western blot (Fig. 4A).


Figure 4. Biallelic deletion of Pten and Grp94 in the liver accelerates liver tumorigenesis. (A) Western blot of liver lysates at 2 months. (B) Liver H&E and Ki67 staining at 2 months showing proliferative cells (black arrows). Arrowheads denote bile ducts. Scale bar: 25 μm. (C) Liver appearance and weight at 8-9 months presented as mean ± SE (**P < 0.01 and ***P < 0.001). (D) Same as B but at 8-9 months.

Download figure to PowerPoint

At 2 months, cPtenf/f livers devoid of PTEN were enlarged (data not shown) with fat accumulation and 1-2 bile ducts around each PV, and limited proliferation as revealed by Ki67 staining (Fig. 4B). In contrast, cPtenf/fGrp94f/f livers were normal in size (data not shown), with prominent bile duct proliferation. Additionally, multiple layers of LPCs surrounding bile ducts were observed with high proliferative activity (Fig. 4B). At 8-9 months, cPtenf/fGrp94f/f mice showed visible liver tumor formation and significantly increased liver/body weight (Fig. 4C). Histological analysis indicated that cPtenf/f mice developed fatty liver, whereas cPtenf/fGrp94f/f mice had bile duct dysplasia with papillary growth pattern (Fig. 4D), correlating with markedly higher Ki67 staining in the liver (Fig. 4D).

Development of HCC and CC in cPtenf/fGrp94f/f Mice

Whereas no tumor was detected in cPtenf/f mice until 12 months, as previously reported,[2] 80% of cPtenf/fGrp94f/f mice developed visible liver tumors by 6 months (not prior to 4.5 months), and 100% of cPtenf/fGrp94f/f mice showed tumors by 8-9 months (Fig. 5A). Liver tumor H&E from cPtenf/fGrp94f/f mice revealed the presence of HCC and CC (Fig. 5B). Immunofluorescence staining of HepPar1, a protein that is highly expressed in HCC,[2] and cholangiocyte marker panCK validated that cPtenf/fGrp94f/f mice developed both HCC and CC (Fig. 5C), suggesting a common LPC origin of the mixed lineage tumors.


Figure 5. HCC and CC formation in cPtenf/fGrp94f/f mice. (A) Liver tumor spectrum. Each circle represents one mouse. The solid and open circles represent mice with and without tumors, respectively. (B) H&E staining of liver tumors in cPf/f94f/f mice (8 months) showed compact trabecular growth structures of HCC and altered tubule structures protruding to the duct lumen characteristic of CC. Right panels represent 2× magnification of the boxed regions. (C) Immunofluorescence staining with HepPar1 (green) and panCK (red) identified HCC and CC, respectively, in cPf/f94f/f livers (9 months). Nuclei were stained with DAPI (blue). PV: portal vein. Scale bar: 50 μm.

Download figure to PowerPoint

Expansion of the Liver Progenitor Cell Pool in cPtenf/fGrp94f/f Mice

Quantitative PCR analysis of LPC markers showed significantly higher levels of EpCAM and CK19, but not AFP, in cPtenf/fGrp94f/f livers at both 2 and 9 months (Fig. 6A). LPCs give rise to hepatocytes and cholangiocytes and express markers for both.[2] Such bipotential LPCs expressing both lineage markers, HepPar1 and panCK, were detected in cPtenf/fGrp94f/f livers (Fig. 6B). cPtenf/f mice are known to exhibit chronic liver injury which leads to hepatocyte death prior to LPC proliferation.[2] In agreement, a gradual increase in plasma ALT levels was observed in cPtenf/f mice starting at 3.5 months (Fig. 6C). In contrast, although we observed LPC expansion in cPtenf/fGrp94f/f mice, ALT levels in cPtenf/fGrp94f/f mice did not increase at 2 and 3.5 months (Fig. 6C). Nonetheless, correlating with the onset of tumor formation, high ALT was observed in cPtenf/fGrp94f/f mice at 6 and 8-9 months (Fig. 6C). Moreover, despite a mild but detectable increase in apoptosis in cPtenf/f, which was further elevated in cPtenf/fGrp94f/f livers, overall apoptosis was low (0.4% or less) in all three genotypes (Fig. 6D).


Figure 6. Expansion of LPCs in cPtenf/fGrp94f/f mice. (A) Quantitative PCR analysis of LPC markers EpCAM, AFP, and CK19. (B) Immunofluorescence staining of cPf/f94f/f livers at 9 months with HepPar1 (green) and panCK (red) identifies bilineage LPCs (white arrows). Nuclei were stained with DAPI (blue). PV: portal vein. Scale bar: 25 μm. (C) Plasma ALT measurements. (D) Quantitation of liver TUNEL staining at 2 months. All data are presented as mean ± SE (*P < 0.05 and **P < 0.01).

Download figure to PowerPoint

Disturbance of Cell Junctions in cPtenf/fGrp94f/f Livers

Immunofluorescence staining revealed that gap junction plaques assembled by Cx26 and Cx32 in cPtenf/fGrp94f/f livers at 2 months were largely disorganized compared to the WT, with an intermediate phenotype observed for cPtenf/f livers (Fig. 7A). cPtenf/fGrp94f/f livers also expressed reduced levels of both connexins (Fig. S3B). Altered integrin β1 and E-cadherin patterns in cPtenf/fGrp94f/f livers were similar to that of cGrp94f/f livers in periportal areas, namely, cytoplasmic rather than membrane distribution of integrin β1, disjointed E-cadherin expression and loss of zonal patterns, and increased clusters of E-cadherin-positive cells around PVs (Fig. 7B). In contrast, cPtenf/f livers showed a similar expression level and pattern of E-cadherin as the WT; however, integrin β1 was mostly localized in the cytoplasm (Fig. 7B).


Figure 7. Disturbance of cell junctions in cPtenf/fGrp94f/f livers at 2 months. Immunofluorescence staining of Cx26 (A) and Cx32 (B) in frozen liver section. Nuclei were stained with DAPI (blue). Scale bar: 20 μm. (B) Schematic drawings of liver E-cadherin expression patterns. PV: portal vein; CV: central vein. Double staining of integrin β1 (green) and E-cadherin (red) of the boxed areas (U1,2,3 and L1,2,3) is shown below. White arrows indicate clusters of liver progenitor/bile duct cells with strong E-cadherin. Scale bar: 50 μm.

Download figure to PowerPoint

ERK Activation in Premalignant cPtenf/fGrp94f/f Livers

In examining proliferative pathways linked to liver cancer at the early age, we observed strong AKT activation in cPtenf/f livers (Fig. 8A). Interestingly, both p-AKT and total AKT levels were lower in cPtenf/fGrp94f/f livers, thus the p-AKT/AKT ratio was similar to cPtenf/f. p-AKT was barely detectable in both WT and cGrp94f/f livers (Fig. 8A). Activation of SRC, p38 as measured by p-SRC and p-p38 levels, showed no difference, and the levels of β-catenin, an effector of canonical Wnt signaling, were similar in all genotypes (Fig. 8A). Strikingly, p-ERK was markedly up-regulated in cPtenf/fGrp94f/f, but not in WT, cPtenf/f, or cGrp94f/f livers (Fig. 8B). p-ERK activation was observed in hepatocytes and in some cells around PVs morphologically resembling LPCs in cPtenf/fGrp94f/f livers at 2 months (Fig. 8C).


Figure 8. ERK activation in premalignant cPtenf/fGrp94f/f livers at 2 months. (A) Representative western blots of liver lysates for the indicated proteins. (B) Same as (A) except ERK was analyzed. (C) Staining of p-ERK in the liver. The boxed area was enlarged with black arrows denoting positive cells morphologically similar to LPCs. Scale bar: 50 μm. (D) Summary model on how hepatic GRP94 depletion promotes PTEN-null induced tumorigenesis. PV: portal vein.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

GRP94, which is only expressed in multicellular organisms, plays an essential role in cell-cell and cell-matrix interactions.[4] Conditional knockout of GRP94 in the muscle and gut revealed how GRP94 affects organ homeostasis could be context-dependent, involving pathways such as IGF-1 and Wnt signaling.[9, 15] As summarized in Fig. 8D, our studies identify GRP94 as a novel regulator of liver homeostasis and tumorigenesis. We propose that in the normal liver, while hepatocytes are highly polarized by cell adhesion proteins, LPCs are quiescent, localized in their niche. Upon GRP94 deletion, cell-cell/cell-matrix interaction is disrupted, which activates LPCs to proliferate and migrate. When GRP94 is depleted in the PTEN-null liver, besides loss of cell adhesion, ERK is activated in both hepatocytes and LPCs, further transforming LPCs to become tumor-initiating cells, giving rise to HCC and CC. Interestingly, this occurs with limited early liver injury, minimal apoptosis, and no detectable necrosis.

Expansion of LPCs occurs when the replication of hepatocytes/cholangiocytes is inhibited,[1] and there could be multiple mechanisms. One explanation is that cGrp94f/f livers are smaller, probably due to decreased capacity of hepatocytes to grow, thus LPC proliferation is an SOS pathway to maintain liver size. Here we provide evidence that disruption of cell adhesion and ERK activation may also contribute to LPC proliferation in cGrp94f/f livers.

Connexins, integrin β1, and E-cadherin have been reported to regulate stem cell-niche interactions in different tissues.[16-18] Moreover, Cx26, Cx32, and integrin β1 in the liver are exclusively expressed in hepatocytes/cholangiocytes,[5, 19] which intimately associate with LPCs under normal conditions. Integrins regulate ECM composition,[20] such as laminin, which maintains LPCs at an undifferentiated state.[21] Disrupted cell adhesion in hepatocytes might also affect their contact with nonparenchymal cells (hepatic stellate and Kupffer cells), which closely surround activated LPCs during liver injury and are important for LPC expansion/invasion.[21-23] Therefore, disorganization of these cell adhesion proteins at an early stage might disrupt the niche, allowing LPCs to overcome contact inhibition, resulting in hyperproliferation and migration. While our results are consistent with the tumor suppressor roles of E-cadherin in epithelial cancer,[6] connexins in liver tumors,[5] and the report that integrin β1 deletion correlates with prostate cancer progression,[7] future studies are required to establish the causative role of cell adhesion molecules in GRP94-mediated liver tumorigenesis.

Upon examination of major proliferative pathways, we identified selective activation of the ERK pathway in cPtenf/fGrp94f/f livers. ERK activation has been reported in HCC associating with aggressive tumor behavior.[24] LSP1, a scaffold protein associated with RAF/MEK/ERK pathway, is the most commonly deleted gene in HCC.[25] Additionally, RAF/MEK/ERK pathway promotes proliferation of Sca-1-positive LPCs.[26] We showed that p-ERK was mainly up-regulated in hepatocytes and in some LPCs at the premalignant stage. Therefore, while ERK signaling may directly induce LPC proliferation, ERK activation in hepatocytes might also stimulate the release of diffusible factors that are mitogenic to LPCs. Indeed, ERK signaling mediates inflammatory cytokine induction, such as tumor necrosis factor-α (TNF-α), which can regulate LPC proliferation.[27, 28] However, while our studies suggest LPCs to be the origin of tumors in cPtenf/fGrp94f/f mice, mature hepatocytes may also give rise to HCC and CC.[29, 30]

No studies to date have examined the in vivo function of GRP94 in tumor development. Here we investigated the consequence of loss of GRP94 function in the liver and uncovered its novel role in maintaining cell adhesion, and in the context of PTEN-null livers selectively activating ERK pathway and accelerating tumorigenesis. While the generality of these observations in other liver tumor models remains to be determined, cGrp94f/f mice fed for 3 weeks with a 3,5-diethoxycarbonyl-1,4-dihydrocolidine (DDC) diet, known to induce liver injury and LPC proliferation,[31] showed higher proliferation of LPCs/bile duct cells, compared to the WT (Fig. S4). GRP94 overexpression has been reported in various cancers, including HCC.[3, 32, 33] GRP94, as an antiapoptotic protein,[34] may well protect tumor cells from host cell defense and promote tumor progression. Thus, while loss of GRP94 function may accelerate tumorigenesis, gain of GRP94 function could offer protection against stress in a growing tumor, which awaits further investigation.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

We thank Drs. Valentina Factor and Parkash Gill for the generous gifts of antibodies, members in Lee lab and Stiles lab for helpful discussions, and Tony Li for assistance with bilirubin measurement.


  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information
  • 1
    Roskams T. Liver stem cells and their implication in hepatocellular and cholangiocarcinoma. Oncogene 2006;25:3818-3822.
  • 2
    Galicia VA, He L, Dang H, Kanel G, Vendryes C, French BA, Zeng N, et al. Expansion of hepatic tumor progenitor cells in Pten-null mice requires liver injury and is reversed by loss of AKT2. Gastroenterology 2010;139:2170-2182.
  • 3
    Ni M, Lee AS. ER chaperones in mammalian development and human diseases. FEBS Lett 2007;581:3641-3651.
  • 4
    Eletto D, Dersh D, Argon Y. GRP94 in ER quality control and stress responses. Semin Cell Dev Biol 2010;21:479-485.
  • 5
    Vinken M, Henkens T, De Rop E, Fraczek J, Vanhaecke T, Rogiers V. Biology and pathobiology of gap junctional channels in hepatocytes. Hepatology 2008;47:1077-1088.
  • 6
    Pecina-Slaus N. Tumor suppressor gene E-cadherin and its role in normal and malignant cells. Cancer Cell Int 2003;3:17.
  • 7
    Moran-Jones K, Ledger A, Naylor MJ. beta1 integrin deletion enhances progression of prostate cancer in the TRAMP mouse model. Sci Rep 2012;2:526.
  • 8
    Mao C, Wang M, Luo B, Wey S, Dong D, Wesselschmidt R, et al. Targeted mutation of the mouse Grp94 gene disrupts development and perturbs endoplasmic reticulum stress signaling. PLoS One 2010;5:e10852.
  • 9
    Marzec M, Eletto D, Argon Y. GRP94: An HSP90-like protein specialized for protein folding and quality control in the endoplasmic reticulum. Biochim Biophys Acta Mol Cell Res 2012;1823:774-787.
  • 10
    Luo B, Lam BS, Lee SH, Wey S, Zhou H, Wang M, et al. The endoplasmic reticulum chaperone protein GRP94 is required for maintaining hematopoietic stem cell interactions with the adult bone marrow niche. PLoS One 2011;6:e20364.
  • 11
    Luo B, Tseng CC, Adams GB, Lee AS. Deficiency of GRP94 in the hematopoietic system alters proliferation regulators in hematopoietic stem cells. Stem Cells Dev 2013 [Epub ahead of print].
  • 12
    Fu Y, Wey S, Wang M, Ye R, Liao CP, Roy-Burman P, et al. Pten null prostate tumorigenesis and AKT activation are blocked by targeted knockout of ER chaperone GRP78/BiP in prostate epithelium. Proc Natl Acad Sci U S A 2008;105:19444-19449.
  • 13
    Kalra J, Shao Q, Qin H, Thomas T, Alaoui-Jamali MA, Laird DW. Cx26 inhibits breast MDA-MB-435 cell tumorigenic properties by a gap junctional intercellular communication-independent mechanism. Carcinogenesis 2006;27:2528-2537.
  • 14
    Fujimoto K, Nagafuchi A, Tsukita S, Kuraoka A, Ohokuma A, Shibata Y. Dynamics of connexins, E-cadherin and alpha-catenin on cell membranes during gap junction formation. J Cell Sci 1997;110(Pt 3):311-322.
  • 15
    Liu B, Staron M, Hong F, Wu BX, Sun S, Morales C, et al. Essential roles of grp94 in gut homeostasis via chaperoning canonical Wnt pathway. Proc Natl Acad Sci U S A 2013;110:6877-6882.
  • 16
    Gonzalez-Nieto D, Li L, Kohler A, Ghiaur G, Ishikawa E, Sengupta A, et al. Connexin-43 in the osteogenic BM niche regulates its cellular composition and the bidirectional traffic of hematopoietic stem cells and progenitors. Blood 2012;119:5144-5154.
  • 17
    Loulier K, Lathia JD, Marthiens V, Relucio J, Mughal MR, Tang SC, et al. beta1 integrin maintains integrity of the embryonic neocortical stem cell niche. PLoS Biol 2009;7:e1000176.
  • 18
    Karpowicz P, Willaime-Morawek S, Balenci L, DeVeale B, Inoue T, van der Kooy D. E-Cadherin regulates neural stem cell self-renewal. J Neurosci 2009;29:3885-3896.
  • 19
    Shiojiri N, Sugiyama Y. Immunolocalization of extracellular matrix components and integrins during mouse liver development. Hepatology 2004;40:346-355.
  • 20
    Yue J, Zhang K, Chen J. Role of integrins in regulating proteases to mediate extracellular matrix remodeling. Cancer Microenviron 2012;5:275-283.
  • 21
    Lorenzini S, Bird TG, Boulter L, Bellamy C, Samuel K, Aucott R, et al. Characterisation of a stereotypical cellular and extracellular adult liver progenitor cell niche in rodents and diseased human liver. Gut 2010;59:645-654.
  • 22
    Roskams T. Relationships among stellate cell activation, progenitor cells, and hepatic regeneration. Clin Liver Dis 2008;12:853-860, ix.
  • 23
    Van Hul N, Lanthier N, Espanol Suner R, Abarca Quinones J, van Rooijen N, Leclercq I. Kupffer cells influence parenchymal invasion and phenotypic orientation, but not the proliferation, of liver progenitor cells in a murine model of liver injury. Am J Pathol 2011;179:1839-1850.
  • 24
    Schmitz KJ, Wohlschlaeger J, Lang H, Sotiropoulos GC, Malago M, Steveling K, et al. Activation of the ERK and AKT signalling pathway predicts poor prognosis in hepatocellular carcinoma and ERK activation in cancer tissue is associated with hepatitis C virus infection. J Hepatol 2008;48:83-90.
  • 25
    Nalesnik MA, Tseng G, Ding Y, Xiang GS, Zheng ZL, Yu Y, et al. Gene deletions and amplifications in human hepatocellular carcinomas: correlation with hepatocyte growth regulation. Am J Pathol 2012;180:1495-1508.
  • 26
    Jin C, Samuelson L, Cui CB, Sun Y, Gerber DA. MAPK/ERK and Wnt/beta-Catenin pathways are synergistically involved in proliferation of Sca-1 positive hepatic progenitor cells. Biochem Biophys Res Commun 2011;409:803-807.
  • 27
    Li DQ, Luo L, Chen Z, Kim HS, Song XJ, Pflugfelder SC. JNK and ERK MAP kinases mediate induction of IL-1beta, TNF-alpha and IL-8 following hyperosmolar stress in human limbal epithelial cells. Exp Eye Res 2006;82:588-596.
  • 28
    Tanaka M, Itoh T, Tanimizu N, Miyajima A. Liver stem/progenitor cells: their characteristics and regulatory mechanisms. J Biochem 2011;149:231-239.
  • 29
    Fan B, Malato Y, Calvisi DF, Naqvi S, Razumilava N, Ribback S, et al. Cholangiocarcinomas can originate from hepatocytes in mice. J Clin Invest 2012;122:2911-2915.
  • 30
    Sekiya S, Suzuki A. Intrahepatic cholangiocarcinoma can arise from Notch-mediated conversion of hepatocytes. J Clin Invest 2012;122:3914-3918.
  • 31
    Wang X, Foster M, Al-Dhalimy M, Lagasse E, Finegold M, Grompe M. The origin and liver repopulating capacity of murine oval cells. Proc Natl Acad Sci U S A 2003;100:11881-11888.
  • 32
    Yao DF, Wu XH, Su XQ, Yao M, Wu W, Qiu LW, et al. Abnormal expression of HSP gp96 associated with HBV replication in human hepatocellular carcinoma. Hepatobil Pancreat Dis Int 2006;5:381-386.
  • 33
    Fu Y, Lee AS. Glucose regulated proteins in cancer progression, drug resistance and immunotherapy. Cancer Biol Ther 2006;5:741-744.
  • 34
    Reddy RK, Lu J, Lee AS. The endoplasmic reticulum chaperone glycoprotein GRP94 with Ca2+-binding and antiapoptotic properties is a novel proteolytic target of calpain during etoposide-induced apoptosis. J Biol Chem 1999;274:28476-28483.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

hep26711-sup-0001-suppinfo01.doc50KSupporting Information
hep26711-sup-0002-suppfig1.tif1026KSupporting Information Figure 1.
hep26711-sup-0003-suppfig2.tif4360KSupporting Information Figure 2.
hep26711-sup-0004-suppfig3.tif1414KSupporting Information Figure 3.
hep26711-sup-0005-suppfig4.tif3883KSupporting Information Figure 4.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.