Cellular gp96 upregulates AFP expression by blockade of NR5A2 SUMOylation and ubiquitination in HCC

doi:10.21203/rs.3.rs-1977718/v2

Download PDF

Research Article

Cellular gp96 upregulates AFP expression by blockade of NR5A2 SUMOylation and ubiquitination in HCC

https://doi.org/10.21203/rs.3.rs-1977718/v2

This work is licensed under a CC BY 4.0 License

Version 2

posted

You are reading this latest preprint version

AFP is the most widely used biomarker for the diagnosis of hepatocellular carcinoma. However, a substantial proportion of HCC patients have either normal or marginally increased AFP levels in serum, and the underlying mechanisms are not fully understood. In the present study, we provided in vitro as well as in vivo evidence that heat shock protein gp96 promoted AFP expression at the transcriptional level in HCC. NR5A2 was identified as a key transcription factor regulated by AFP and its stability was enhanced by gp96. A further mechanistic study by CO-IP, GST-pull down and molecular docking showed the competitive binding of gp96 and SUMO E3 ligase RanBP2 to NR5A2 at the sites spanning from aa 507 to 539. The binding of gp96 inhibited SUMOylating, ubiquitination, and subsequent degradation of NR5A2. In addition, clinical analysis of HCC patients indicated that gp96 expression was positively correlated to serum AFP levels in tumors. Therefore, our study uncovered the novel regulatory mechanism of gp96 on the stability of its client proteins by directly affecting their SUMOylation and ubiquitination. These findings will help in designing more accurate AFP-based HCC diagnosis and progression monitoring approaches.

Gp96

AFP

NR5A2

RanBP2

SUMOylation

Alpha-fetoprotein, a 70 kDa glycoprotein, is the main serum protein produced during fetal life[1]. Unlike other members of the serum-albumin family, the AFP gene is transcriptionally silent in the adult liver but is reactivated in hepatocellular carcinoma (HCC), where elevated serum levels of AFP have been found in about 70–85% of HCC patients[2]. Since serum AFP is mainly produced by hepatic tumor cells, a concentration of 20 ng/mL is generally used as a pathological threshold value, and > 400 ng/mL is usually considered diagnostic for HCC[3, 2]. Although the discovery of new oncologic markers has been carried out for decades, AFP remains the most widely used biomarker in HCC diagnosis[3, 2].

Several underlying regulatory mechanisms of pathological AFP expression have been reported in HCC[3]. Numerous transcription factors are crucial for the regulation of the AFP promoter[3–5]. The human AFP gene is comprised of five regulatory regions: a tissue-specific promoter, a repressor upstream of the promoter, and the three enhancers located at 2.5, 5.0, and 6.5 kb upstream of the AFP promoter, respectively[3, 6]. The potential AFP transcription factors include hepatocyte nuclear factor-1 (HNF-1), liver receptor homolog-1 (LRH-1, also known as FTF or NR5A2), retinoid x receptor (RXR), CAAT/enhancer binding protein (C/EBP), homeobox protein NK-2 (Nkx2.8), zinc-fingers and homeoboxes 2 (Zhx2), and zinc finger and BTB domain containing 20 (Zbtb20)[4, 6, 5, 3]. Zhx2 and Zbtb20 bind specifically to the AFP promoter and inhibit its activity; their levels are increased with the decrease in AFP expression after birth[4, 7]. AFP expression could also be regulated post-transcription by miRNAs including miRNA-122, miRNA-620, miRNA-1236, miRNA-1270, and miRNA-329, which directly target the 3’-UTR region of the AFP mRNA[8, 9]. In addition, epigenetic regulation like methylation, and acetylation status of the AFP gene also influence AFP expression[10, 11]. Nevertheless, serologic surveillance showed that a significant proportion (around 15–30%) of HCC patients have either normal or minimally increased AFP levels[12]. Meanwhile, other liver diseases such as hepatitis and cirrhosis may also lead to elevated serum AFP levels. Therefore, the exact mechanisms of AFP expression in HCC are not well understood and need further investigation[3, 2].

Heat shock protein (HSP) gp96 is the most abundant protein in the endoplasmic reticulum (ER)[13]. Similar to most of HSPs, gp96 acts as a molecular chaperone in protein folding and facilitates the degradation of misfolded proteins. Unlike its cytosolic counterpart HSP90, gp96 has a relatively strict binding selectivity and only binds certain client proteins, including Toll receptor (Toll-like receptors), the majority of α and β integrin subunits, Wnt co-receptor low-density lipoprotein receptor-related protein 6, glycoprotein A repetitions predominant, insulin-like growth factor, binding immunoglobulin protein (BiP), epidermal growth factor receptor-2 (HER2), and uPAR[14–19, 13, 20, 21]. These client proteins are involved in balancing cancer-induced ER stress responses and regulating inflammation in the tumor microenvironment[22]. An elevated level of gp96 correlates with the development, invasion, and metastasis of tumors, which validates the pro-oncogenic function of cellular gp96[13, 18, 15, 16, 19, 17]. Furthermore, cellular and cell membrane gp96 has been shown to promote anti-apoptotic characteristics and metastasis of HCC via interaction with p53 and uPAR concluding the association of high gp96 expression with tumor metastasis and recurrence[18, 16].

The present study aimed to comprehensively examine the effects of gp96 on AFP expression and its serum levels in HCC by generating hepatic gp96 knock-out mice and clinical sample analysis. The underlying mechanism of the gp96-mediated regulatory network was further dissected. The results uncovered a novel regulatory pathway of AFP expression and suggested that detection of gp96 levels may increase the sensitivity of HCC diagnosis by AFP test.

Animal studies

Liver-specific gp96 KO mice were presented by Professor Zihai Li and generated by crossing Albumin-Cre mice with Hsp90b1flox/flox mice to obtain Albumin-Cre Hsp90b1flox/flox mice [23] . Genotyping was done by PCR analysis of genomic DNA obtained from the tail, and the primer sequences used were 5'-TGCCAGAGACTACAATTCCCAGCA-3’,5'-AAACACGAACTCACCAATCGTGCC-3’ (Hsp90b1flox/flox mice), 5'-TGG CAA ACA TAC GCA AGG G-3’, and 5'-CGG CAA ACG GAC AGA AGC A-3’ (Albumin-Cre mice). Fifteen days after birth, gp96 KO and heterozygous (Het) mice were intraperitoneally injected with diethyl-nitrosamine (DEN) at a dose of 25 mg/kg. Mice were then sacrificed eight months after injection, and livers were excised. Externally visible tumors (≥ 1 mm) in the liver were counted and the liver tissues were stored at-80 °C.

Cell culture and transfection

The human hepatoma cell lines Huh7 and HepG2 were purchased from Cell Resource Center, IBMS, CAMS/PUMC, China. Cells were maintained in DMEM medium and incubated at 37 °C in a 5% CO₂atmosphere. Cell transfections were performed using TransIT-X2 Dynamic Delivery System (MIR6000) according to the manufacturer’s recommendations. TransIT-X2 was purchased from Mirus Bio LLC (Madison, WI USA). For all in vitro experiments, cells were detached from flasks using PBS buffer containing 0.05% trypsin and 0.2 mM EDTA and seeded onto dishes, 2~3 days before treatment.

Reagents and antibodies

The following antibodies and reagents were obtained as indicated: gp96 antibody (sc 32249), NR5A2 antibody (sc398349), C/EBPα antibody (sc365318), RXRα antibody (sc515929), ZBTB20 antibody (sc515370), and ZHX2 antibody (sc393399) from Santa Cruz Biotechnology (Dallas, TX, USA); AFP antibody (ab169552) and SUMO2+3 antibody (ab81371) from Abcam (Cambridge, MA, USA); RanBP2 antibody (NB100-74480) and NR5A2 antibody (PP-H2325-00) from R&D Systems (Minneapolis, MN, USA); TOP2A antibody (20233-1-AP), HDAC4 antibody (17449-1-AP) and SUMO1 antibody (10329-1-AP) from Proteintech (Wuhan, Hubei, P.R. China); ubiquitin antibody (BS91400), Nkx2.8 antibody (BS9251), and HNF4α antibody (BS2983) from Bioworld technology (District Nanjing, P. R. China; GAPDH antibody (TA-08) and horseradish peroxidase (HRP)-conjugated secondary antibodies from Zhongshan Golden bridge Biotechnology (Beijing, P. R. China); MG132 (S2619) from Selleck Chemicals LLC (Houston, TX, USA); and the ECL Plus chemiluminescence system from Beyotime Biotechnology (Shanghai, P. R. China). All small interfering RNA (siRNA) were designed and synthesized by Ribobio Co, Ltd (Guangzhou, P. R. China). Antibodies and siRNA were used according to the manufacturer’s instructions. The oligonucleotide sequences of siRNA are shown below indicated in Supplementary Table S1.

Plasmid constructs and protein purification

Plasmid pcDNA 3.1-gp96, PGEX6p-1-gp96 (aa 22-803), PGEX6p-1-N-gp96 (aa 22-376), PGEX6p-1-M-gp96 (aa 337-594) and PGEX6p-1-C-gp96 (aa 561-803) were maintained in our lab. Flag or GST-tagged NR5A2 (aa 299-536 or 299-506) was cloned into the pcDNA 3.1 vector or PGEX6p-1 vector. The GST-tagged gp96 or segments of gp96 (gp96-N, gp96-M, and gp96-C) and GST-NR5A2 were expressed in Escherichia coli (BL21 DE3) and purified by passing the lysates through a Glutathione Sepharose 4 Fast Flow column according to the manufacturer’s instructions. Human gp96 was subcloned into the pFastBac1 vector. Recombinant gp96 was expressed in the Bac-to-Bac Baculovirus expression system and purified by Ni-affinity, anion exchange, and size-exclusion chromatography.

The human AFP promoter sequences (-240/+29) were amplified by PCR and inserted into the pGL3 basic vector (Promega, Madison WI, USA). The luciferase vectors were named pGL3-AFP (-240/+29). The binding sites for NR5A2, Nkx2.8, HNF1, or C/EBPα in the pGL3-AFP construct were mutated and the mutant plasmids were acquired. The point mutation primers are shown in Supplementary Table S2.

Lentivirus vector construction and establishment of stable cell line

The small hairpin oligonucleotide for gp96 or luciferase was cloned into PLKO.1 lentivirus vector, and named PLKO.1-gp96 or PLKO.1-Luci, respectively. The vectors of pMD2, psPAX, and PLKO.1-gp96 or PLKO.1-Luci were co-transfected into 293T cells. After 48 h, the supernatant was collected and concentrated by PEG8000. Lentivirus particles were concentrated to reach approximately 1×10¹⁰PFU. Stable cell lines were selected with 1.0 µg/mL puromycin for 14 days.

To overexpress human gp96 protein, the gp96 gene (NCBI: ENSMUSG00000020048) was cloned into the PCDH-EF1α-MCS-T2A-puro vector named PCDH-EF1α-MCS-T2A-gp96. The vectors of pMD2, psPAX, and PCDH-EF1α-MCS-T2A-gp96 or empty vector PCDH-EF1α-MCS-T2A-puro were co-transfected into 293T cells. Stable cell lines were selected with 1.0 µg/mL puromycin for 14 days.

Protein identification by nano LC-MS/MS analysis

To identify proteins bound with NR5A2, Huh7 cell lysates were subjected to immunoprecipitation with NR5A2 antibody. Precipitated protein was separated by electrophoresis and stained with Coomassie Blue dye. The targeted protein bands were cut from the gel and digested with trypsin to obtain candidate peptides. The peptides were loaded onto a trap column (C18, 5 µm particles, 100 µm ID, 2 cm length, Dr. Maisch GmbH) and separated using an analytical column (C18, 3 µm particles, 75 µm ID, 15 cm length, Dr. Maisch GmbH) at a flow rate of 400 nL/min with a 60 min LC gradient composed of Solvent A (0.1% formic acid (v/v)) and Solvent B (acetonitrile, 0.1% formic acid (v/v)). The separation was done using a linear gradient of 3-10% B for 2 min; 10-25% B for 38 min; 25-40% B for 15 min; 40-90% B for 3 min; and finally, 90% B for 2 min. The mass spectrometric data were analyzed using the Mascot database search engine. Peptide sequences were interpreted from the MS/MS spectra by searching Homo sapiens in the SwissProt protein database.

Quantitative Real-time PCR

Total RNA was extracted with TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). Real-time PCR was performed using TB Green Premix Reagent (Takara Bio Inc., Shiga, Japan) according to the manufacturer’s protocol. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal standard for quantification. Primers used in RT-PCR were chemically synthesized and are shown in the Supplementary tables. Relative expression was quantified by the comparative threshold cycle (CT) method. The primers used in Real-time PCR are indicated in Supplementary Table S3.

Luciferase reporter assay

The pGL3-Basic plasmid containing a luciferase ORF under the AFP promoter(-240/+29) (pAFP-Luci) was constructed. Huh7 cells were co-transfected with pRL-TK and firefly luciferase reporter plasmid carrying either the wild-type or mutant AFP promoter. Firefly luciferase and Renilla luciferase activities were measured with the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA). The firefly luciferase activity was normalized to Renilla luciferase activity.

In vitro GST pull-down assay

Five micrograms of GST-tagged proteins were incubated with 500 µg cell lysates at 4 °C for 2~3 h. Then, GST beads (GE Healthcare) were added and incubated overnight at 4 °C. The beads were collected by brief centrifugation and subjected to western blot analysis.

ELISA assay

AFP levels in the supernatant were measured by Human AFP ELISA kit (Multisciences, Zhejiang, P, R. China) according to the manufacturer’s instructions.

Co-immunoprecipitation (co-IP)

The human hepatoma cells were lysed with ice-cold NP40 Cell Lysis Buffer (50 mmol/L Tris (pH 7.4), 150 mmol/L NaCl, sodium pyrophosphate, β-glycerophosphate, sodium orthovanadate, sodium fluoride, EDTA, and leupeptin). Equal amounts of total protein were incubated with 2 µL primary antibodies or IgG as a control for 2~3 h at 4 °C. Then, protein A&G Sepharose beads (Santa Cruz Biotech, Dallas, TX, USA) were added and incubated with the cell lysates overnight at 4 °C. The beads were washed four times with cell lysis buffer and resuspended in 5X SDS-PAGE loading buffer. The proteins were separated by SDS-PAGE and analyzed by western blot.

Western blot analysis

The cell lysates were denatured in 5X SDS-PAGE loading buffer, and the proteins were separated with SDS-PAGE and transferred to polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA). The membranes were blocked and incubated with primary antibodies and the corresponding HRP-conjugated secondary antibodies. Protein was visualized in the film using the enhanced chemiluminescence (ECL) substrate (Beyotime Bio, Shanghai, P, R. China). GAPDH served as the loading control.

Immunohistochemistry (IHC)

Immunohistochemical staining was performed as described previously. The liver tissues were fixed in formalin and cut into four-µm-thick sections. Then, the paraffin sections were immunostained using the primary antibodies and the corresponding HRP-conjugated secondary antibodies and were visualized with 3’3-diaminobenzidine tetrahydrochloride (DAB).

Molecular modeling docking

The crystal structures of NR5A2-LDB (ligand-binding domain) from the docking model of NR5A2 DBD (DNA binding domain)-LBD, were obtained from PDB-Dev (PDBDEV_00000035) [24] . The crystal structure of full-length gp96 (PDB code 5ULS, residues 48-754) contained three major domains: the N-terminal, middle, and C-terminal domains [25] . The crystal structure of the SUMO-RanGAP1-Ubc9-Nup358 complex has been reported previously (PDB code 1Z5S) containing RanBP2 residues 2631-2711 [26, 27] . The gp96-NR5A2 and NR5A2-RanBP2 protein-protein docking models were created with ZDOCK Server (version 3.0.2) protein-protein docking; each with a centroid phase with rigid-body docking and an all-atom phase [28] . PyMoL was used for further visualization and figure preparation.

Patient sample and specimens

Paraffin-embedded hepatic tumor sections of 63 HCC patients were obtained from Department of Pathology and Hepatology, the fifth Medical Centre, Chinese PLA General Hospital, between January 2019 and August 2020. The studied clinical characteristics of the subjects are listed in Table 1. The serological AFP level included in this analysis was obtained from the preoperative measurement before and closest to the date of resection. The endogenous gp96 staining was categorized as low (score 1+), medium (score 2+), and high (score 3+) expression based on the staining intensity and immune reactive cell percentage according to a widely used scoring method (slightly modified) [29] . The intracellular g96 scoring is described as follows: score (+), slight and incomplete cell staining of hepatoma cells; Score (2+), medium cytoplasm staining in at least 50% of hepatocyte cells; Score (3+), high cytoplasm staining in at least 85% of hepatocyte cells. The assessments were blindly scored by two independent observers.

Table 1 The associated of gp96 protein level with Serum AFP levels in patients with primary HCC

characteristic	Gp96 expression		P value
characteristic	Low(13/63)	High(50/63)
Age(years)			0.5018
<50	6	18
>=50	7	32

Gender			0.2699
male	12	38
Female	1	12

Tumor size			0.5078
<3cm	5	14
>=3cm	8	36

Tumor number			0.739
=1	8	35
>1	5	15

HBsAg			0.1893
Positive	6	33
Negative	7	17


HBcAg			0.8006
Positive	1	5
Negative	12	45

T stage			0.2063
I-II	1	0
III-IV	12	50

Differentiation grade			0.631
Well differentiated	0	3
Moderatedly differentiated	12	42
Poorly differentiated	1	5

Extrahepatic spread			NA
Yes	0	0
No	13	50

Recurrence			NA
positive	0	0
negative	13	50

Statistical analysis

All data were analyzed using SPSS software (SPSS Science, Chicago, IL) and R software (version4.1.2; http://Rproject.org). Results are reported as the means ± standard deviations. Differences between mean values were analyzed using the Student’s t-test. P-values < 0.05 were considered statistical significant.

Hepatic knock-out of gp96 dramatically decreases AFP levels in DEN-induced HCC

In this study, we used the well-established DEN-induced HCC mouse model, where liver-specific gp96-KO mice and Het mice were intraperitoneally injected with DEN to induce hepatic tumors [23] . Deletion of gp96 (Hsp90b1) in the liver of gp96-KO mice was confirmed by Real-time PCR and Western blotting (Fig. 1a) [23] . Eight months after DEN-injection, both gp96-KO and Het mice developed visible liver tumors; tumor histological features were shown by H&E staining (Fig. 1b) [23] . Compared to the DEN-untreated Het mice liver tissue, there was a drastic increase in gp96 and AFP levels in DEN-induced liver tumors (Fig. 1c). Consistent with a previous study, fatty nodules were observed in tissue sections of gp96-KO mice tumors (Fig. S1a). Gp96-KO mice showed a decrease in tumor weight and low tumor burden compared to Het mice under DEN treatment (Fig. 1d). Global gene expression in liver tumor tissues of gp96-KO and Het mice was studied by transcriptome analysis. Compared to the Het group, gp96-KO mice showed increased expression of 5413 genes and decreased expression of 4488 genes in HCC (Fig 1e, and additional file 1; fold change (log2FC) > 0.5), among which AFP expression was reduced by around 2.488-fold. The decrease in AFP gene expression was confirmed by Real-time PCR (Fig. 1f), Western blotting (Fig. 1g), and IHC analysis (Fig. 1 h).

Cellular gp96 promotes AFP expression mainly through its transcription factor NR5A2 in hepatoma cells

Two hepatoma cells, Huh7 and HepG2, with stable knockdown of gp96 were used to determine gp96-mediated AFP expression. Compared to the mock, knockdown of gp96 caused an obvious decrease in AFP mRNA as well as protein levels (Fig. 2a and 2b). In the meantime, gp96 knockdown also caused a pronounced decrease in AFP levels by 98% and 95% in the supernatant of Huh7 and HepG2 cells, respectively (Fig. 2c). In contrast, overexpression of gp96 in Huh7 cells resulted in increased levels of cellular AFP mRNA and protein (Fig. 2d and 2e), as well as elevated AFP secretion (Fig. 2f). Overall, the data suggested that gp96 promotes AFP expression at the transcriptional level.

The effect of gp96 on the expression of AFP transcription factors

The potential binding sites of AFP transcription factors in the promoter region are shown in Fig. 3a [3-5] . As seen in Fig. 3b and 3c, knockdown of gp96 caused a sharp decrease in HNF1, HNF4, and C/EBPα, but an increase in Nkx2.8 mRNA as well as protein levels. For NR5A2, gp96 knockdown only decreased its protein level but did not affect its mRNA level (Fig. 3b and Fig. 3c). Furthermore, an AFP promoter luciferase reporter assay was performed to determine the key transcriptional factors involved in gp96-induced AFP expression. Compared to the wild-type promoter, mutations within the binding sites of Nkx2.8 or NR5A2/HNF4 significantly suppressed the AFP promoter activity in Huh7 cells. In addition, mutations within the binding sites of HNF1 also led to a slight decrease in AFP promoter activity. Gp96 knockdown-induced decrease in promoter activity was only mildly affected by mutations in HNF1 binding sites but was abolished by mutations in Nkx2.8 or NR5A2/HNF4 binding sites (Fig. 3e).Nkx2.8, NR5A2 and HNF4 were then selected for RNAi treatment. Only NR5A2 siRNA treatment largely abolished the effect of gp96 knockdown on promoter activity. Decrease of AFP expression induced by gp96 knockdown was inhibited by NR5A2 siRNA treatment(Fig. 3f, 3g and Fig. S2). Collectively, these data suggested that cellular gp96 regulates AFP expression mainly via NR5A2.

Cellular gp96 interacts with NR5A2 and enhances its stability

Next, the effect of gp96 on the expression of NR5A2 was determined. A sharp increase in protein level (around 3.6 folds) but not in mRNA level of NR5A2 was detected in gp96 transfected Huh7 cells compared to mock (Fig. 4a and 4b). In addition, the gp96-mediated increase in NR5A2 levels was gp96 dose-dependent (Fig. 4c). Conversely, a drastic reduction of 87% in NR5A2 protein level but not mRNA level was observed in Huh7-gp96KD cells (Fig. 4d and 4e). Decreased NR5A2 protein levels were also observed in DEN-induced liver tumors of gp96-KO mice (Fig. S3) [23] . As seen in Fig. 4f, the knockdown of gp96 caused a dramatic reduction in the protein stability of NR5A2.

As illustrated in Fig. 4 g, the co-IP assay showed that endogenous gp96 is associated with NR5A2 in Huh7 cells. To identify the interaction regions between gp96 and NR5A2, GST-tagged full-length gp96, its N-terminal, M-domain, and C-terminal domain fragments, GST-NR5A2, and His-gp96 were expressed (Fig. S4a-c). GST pull-down assay demonstrated that the full-length gp96 and its C-terminal domain were able to interact with NR5A2 (Fig. 4 h). Meanwhile, the GST-tagged NR5A2 pull-down assay also indicated a direct interaction between gp96 and NR5A2 (Fig. 4i and 4j).

Cellular gp96 inhibits the interaction of RanBP2 with NR5A2 and reduces RanBP2-mediated NR5A2 SUMOylation and subsequent ubiquitination

Knockdown of gp96 in Huh7 cells significantly increased NR5A2 ubiquitination, while gp96 overexpression suppressed the ubiquitination dramatically (Fig. 5a). Since, post-transcriptional processing in nuclear receptors of the NR5A subfamily mainly occurs via the SUMOylated modification, we quantified SUMO1, SUMO2/3, and their related conjugates [30] . Our results showed that knockdown or overexpression of gp96 in Huh7 cells affected protein SUMOylation via SUMO1 but not via SUMO2/3 (Fig. S5). Also, knockdown of gp96 promoted NR5A2 SUMOylation by SUMO1 modification, whereas overexpression of gp96 inhibited its SUMOylation (Fig. 5b). Moreover, gp96 knockdown-induced NR5A2 ubiquitination was highly attenuated by SUMO1 siRNA, suggesting that SUMO1 SUMOylation is involved in NR5A2 ubiquitination (Fig. 5c).

To identify a potential SUMO1 E3 ligase, mass spectrometry (LC-MS/MS) was performed on NR5A2-associated proteins (additional file 2). Among the detected proteins, nucleoporin RanBP2 has been reported to possess SUMO1 E3 ligase activity [26, 31] . The results of the co-IP assay validated the interaction of NR5A2 with RanBP2 (Fig. 5d). It was found that their interaction was enhanced in gp96 knockdown cells, whereas it was suppressed by gp96 overexpression (Fig. 5e). Unlike NR5A2, gp96 knockdown did not affect RanBP2 levels (Fig. S6). Similarly, gp96 knockdown increased the SUMOylation of NR5A, and RanBP2 RNAi inhibited this effect (Fig. 5f). Cellular gp96 inhibited the interaction of NR5A2 with RanBP2 in a dose-dependent manner (Fig. 5g). GST pull-down assay further demonstrated that gp96 was competitively binding against RanBP2 to NR5A2 in a dose-dependent manner (Fig. 5h and 5i).

We obtained similar results for SUMO modification of another RanBP2 substrate, DNA topoisomerase 2-alpha (TOP2A), to which gp96 showed binding (Fig. S7a-c and S7e) [32] . In contrast, such regulation was not observed for histone deacetylase 4 (HDAC4), a RanBP2 substrate, which could not bind to gp96 (Fig. S7a, S7b, S7d, S7f) [33] . Altogether, our results indicated that cellular gp96 interacts with its client protein NR5A2, which might specifically block the binding of its SUMO1 E3 ligase RanBP2, thereby suppressing the RanBP2-mediated SUMOylation and subsequent ubiquitination of NR5A2.

The binding sites overlap for gp96 and RanBP2 in NR5A2

Next, molecular docking was performed with ZDOCK Server (version 3.0.2) to crystal structures obtained from PDB [28] . The structural model of full-length NR5A2 has a highly conserved DNA binding domain (DBD), a conserved 12-α-helical bundle ligand-binding domain (LBD), and a SUMO-interactive motif (SIM) [24, 30] . NR5A2 LDB (299-536) was docked and modeled in complex with gp96 (aa 48-754) or RanBP2 (aa 2631-2711) (Fig. 6a and 6b) [25, 24] . As shown in Fig. 6c, the potential binding sites for gp96 and RanBP2 in NR5A2 were highly overlapped and located within sequences of aa 515-536 and aa 507-536, respectively. Both Co-IP and His-pull down assays demonstrated that truncated NR5A2 (aa 299-506Δ507-536), with depletion of the binding sites, lost its capacity to interact with gp96 or RanBP2 (Fig. 6d and 6e, and 6f). These data verified that the aa 507-536 sequence of NR5A2 is necessary for its binding to both gp96 and RanBP2.

High gp96 expression in liver tumors correlates with serum AFP levels in HCC patients

To address the clinical correlation of gp96 expression in tumor tissue and serum AFP in HCC patients, we analyzed gp96 expression in 63 primary liver tumors by IHC (Fig. 7a). IHC staining of intracellular gp96 in liver tissues was shown as faint/medium/strong and qualitatively scored as categories 1+/2+/3+, respectively [29] . As seen in Fig. 7b, gp96 expression levels were positively correlated with serum AFP levels in HCC patients (R²=10.177; P = 0.001422). In addition, patients with elevated gp96 expression also displayed higher serum AFP levels (all P < 0.005 among patients with gp96 expression of categories 1+, 2+, and 3+) (Fig. 7c).

In this study, we investigated the regulation of AFP expression by gp96 in HCC. Firstly, comparative transcriptome analysis of DEN-induced tumors obtained from the liver of gp96-KO and Het mice showed that cellular gp96 increased AFP expression and supported tumor formation. Next, the mechanistic study revealed that gp96 upregulated NR5A2 at the protein level, which is a key transcription factor for the AFP gene. Further, RanBP2 was identified as an NR5A2-associated protein having the SUMO1 E3 ligase activity, and gp96 could block their interaction by competitively binding to the same site in NR5A2. Finally, the clinical data from HCC patients showed that gp96 expression in the tumor was positively correlated with the AFP levels in serum. Figure 8 illustrates the gp96-NR5A2-RanBP2 regulatory pathway that elevates AFP expression in HCC. According to this model, RanBP2 promotes the ubiquitination and proteasomal degradation of NR5A2 by SUMOylation in normal liver tissue and certain HCC with low cellular gp96 expression, thereby halting the NR5A2-induced AFP expression. On the contrary, in HCC with high gp96 expression, gp96 increases AFP transcription by blocking the interaction between NR5A2 and RanBP2. Our results provide new insight into understanding the distinct status of AFP expression in HCC and might help to design AFP-based HCC diagnosis and disease monitoring approaches with enhanced sensitivity and specificity.

NR5A2 (also known as LRH-1and FTF) is a member of the nuclear receptors NR5A subfamily[34, 35]. It contains several lysine residues that are modified with SUMOylation by E3-SUMO ligases[36, 30]. SUMOylation regulates the transcriptional activity and subnuclear localization of NR5A2. SUMO modification might also regulate the stability of the transcription factors by affecting the subsequent ubiquitination process through cross-talk with ubiquitins[36, 37]. However, SUMO E3 ligases involved in the SUMOylation of NR5A2 are still unknown. In our present study, we demonstrated that RanBP2 acts as a SUMO E3 ligase that binds to and target NR5A2 for SUMOylation modification. We further defined its binding site in NR5A2. RanGAP, HDAC4, and Borealin proteins (TOP2A) have been reported as the SUMOylation targets of RanBP2 in previous studies[38, 33, 32]. Since, all the identified RanBP2 target proteins, including RanGAP, HDAC4, TOP2A, and NR5A2 are closely associated with cancer development and progression, the exact role of RanBP2 in HCC deserves further investigation[39, 40].

Heat shock protein gp96 belongs to the HSP90 family and is a major chaperone protein in the ER. Gp96 acts as a master ER chaperone in response to stress, inflammation, and cancer[41, 13, 19]. It binds to a variety of client proteins facilitating their folding and directing their maturation, assembly, and export from the ER to the cell surface[22, 42]. It plays a critical role in maintaining ER homeostasis and protein quality control[22]. In addition, ER-resident gp96 can be translocated to the cell membrane or cytosol. Similar to other studies, we found that cell membrane or cytosolic gp96 binds to and stabilizes multiple cancer-related proteins, including pro-ADAMTS9, uPAR, HER2, EGFR, ER-α36, and p53, thereby promoting tumor growth, anti-apoptosis, and invasion[43, 16, 15, 44, 45, 18, 19]. In this study, we uncovered the underlying mechanism of gp96-mediated interaction and stabilization of NR5A2 by sterically blocking the SUMO E3 ligase RanBP2 binding to NR5A2, and protecting SUMOylation-induced NR5A2 degradation. A similar regulatory mechanism by gp96 was also observed for TOP2A, which is a RanBP2 substrate as well as gp96 client protein. However, this regulatory mechanism was not observed for HDAC4, which is a RanBP2 substrate but does not interact with gp96. Since ubiquitin and SUMO-related modifications determine the fate of modified proteins, including their proteasomal degradation, it is worthwhile to explore the effects of gp96 on the stability of its client proteins. We found that gp96 directly controls the interaction between the bound proteins and their E3 ubiquitin/SUMO ligases via the same or overlapped binding sites.

Cellular gp96 expression is controlled mainly by heat shock transcription factors (HSFs), e.g., ATF4[46]. Our previous study found that HBV x protein (HBx) could increase gp96 expression by enhancing the binding of NF-KB to the gp96 promoter. [47]. Expression of gp96 increases in malignant hepatocytes with the progression of HCC[47]. Higher levels of gp96 in tumors are associated with intrahepatic metastasis, degree of tumor differentiation, tumor size, and poor survival of patients[48, 49]. High expression of cell membrane gp96 in tumors correlates with pathogenesis and poor prognosis of HCC in patients[16]. These studies indicate the potential of gp96 as a prognostic indicator for HCC. Simultaneously, AFP expression levels also increase with the progression of HCC[2, 3]. AFP levels in serum have been correlated with HBV infection, tumor size, poor differentiation, and invasion of blood vessels[50, 3]. HCC patients with tumor metastasis have the highest levels of serum AFP[51, 52]. Thus, it can be concluded that AFP has a close association with the poor prognosis of HCC in patients[3, 2, 53]. The identified AFP transcription factors include NR5A2, Nkx2.8, and RAR, while transcriptional repressors include ZHX2 and ZBTB20[3–5]. In the present study, we demonstrated that gp96 upregulates AFP gene expression via NR5A2-induced transcription. Clinical data of HCC patients showed that AFP levels in serum are positively correlated with gp96 expression in tumor tissues, and the majority of HCC patients with low gp96 expression exhibited normal AFP levels (see Fig. 7). Hence, we tentatively defined two distinct HCC subtypes: low gp96 expression with normal AFP, and high gp96 expression with elevated AFP. This finding indicates that low gp96 expression in liver tumors attributes to normal or minimally increased AFP levels in a substantial proportion of HCC patients.

In conclusion, the current study uncovered a novel mechanism of gp96-mediated upregulation of AFP transcription and expression via NR5A2 in HCC. This mechanism involves the binding of cellular gp96 to NR5A2 and sterically blocking the interaction between NR5A2 and RanBP2, thereby inhibiting NR5A2 SUMOylation, ubiquitination, and subsequent degradation. Furthermore, the clinical data of HCC patients revealed two distinct HCC subtypes: high and low gp96 expression, which is positively correlated with serum AFP levels. Considering that AFP is the most widely used biomarker in HCC diagnosis and prognosis, our data provide new insights into the regulatory network of AFP expression and might help in designing more precise monitoring approaches for HCC diagnosis and progression.

In this study, our indicates that low gp96 expression in liver tumors attributes to minimally increased or even normal AFP levels in a substantial proportion of HCC patients. In addition, our results uncover a novel mechanism that gp96, as a chaperon, affects its client proteins stability by directly control the interaction between the bound proteins and their E3 ubiquitin/SUMO ligases via binding the same or overlapped sites.

HCC, hepatocellular carcinoma; AFP, Alpha-fetoprotein; NR5A2, Nuclear receptor subfamily 5 group A member 2; DEN, diethyl-nitrosamine; TOP2A, DNA topoisomerase 2-alpha; HDAC4, Histone deacetylase 4; SIM, SUMO-interactive motif; DBD, DNA binding domain; LBD, ligand-binding domain; HNF-1, Hepatocyte Nuclear Factor-1; RXR, Retinoid x receptor; C/EBP, CAAT/enhancer binding protein; Nkx2.8, Homeobox protein NK-2; Zhx2, Zinc-fingers and homeoboxes 2; Zbtb20, Zinc finger and BTB domain containing 20.

Ethical Approval

All animals received human care, and the mice study was in strict accordance with the “Institute of Microbiology, Chinese Academy of Sciences of Research Ethics Committee”. The mice study protocol was approved by the Research Ethics Committee (permit number PZIMCAS2011001).

Consent for publication

Consent to publish has been obtained from all authors.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding authors on reasonable request.

Competing interests

The authors declare that there are no conflicts of interest.

Funding

This work is supported by the funding from: a grant from the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB29040000), the Industrial innovation team grant from Foshan Industrial Technology Research Institute, Chinese Academy of Sciences, the National Natural Science Foundation of China (32070163,81871297,81903142), the China ATOMIC energy authority, Foshan High-level Hospital construction DengFeng plan and Guangdong Province biomedical innovation platform construction project tumor immunobiotherapy.

Authors' contributions

S.M, L.Q, and C.L. concepted and designed the study. L.Q, and Z.L performed the experiments. L.Q, and Z.L collected data and samples. L.Q, S.M and C.L. made the statistical analysis and wrote the manuscript. C.L. and S.M supervised this work. All authors read and approved the final manuscript.

Acknowledgments

The authors thank Donghong Wang, Xiuge Hou, Zihao Wang, Han Zhang, Yang Li, Xiaolan Zhang, Tong Zhao, Fang Cheng, Shuman Xie, for technical help and advices. We thank Editeg (www.editeg.com) for its linguistic assistance during the preparation of this manuscript.

Deutsch HF. Chemistry and biology of alpha-fetoprotein. Adv Cancer Res. 1991;56:253–312. doi:10.1016/s0065-230x(08)60483-2.
Galle PR, Foerster F, Kudo M, Chan SL, Llovet JM, Qin S et al. Biology and significance of alpha-fetoprotein in hepatocellular carcinoma. Liver Int. 2019;39(12):2214–29. doi:10.1111/liv.14223.
Sauzay C, Petit A, Bourgeois AM, Barbare JC, Chauffert B, Galmiche A et al. Alpha-foetoprotein (AFP): A multi-purpose marker in hepatocellular carcinoma. Clin Chim Acta. 2016;463:39–44. doi:10.1016/j.cca.2016.10.006.
Peterson ML, Ma C, Spear BT. Zhx2 and Zbtb20: novel regulators of postnatal alpha-fetoprotein repression and their potential role in gene reactivation during liver cancer. Semin Cancer Biol. 2011;21(1):21–7. doi:10.1016/j.semcancer.2011.01.001.
Kajiyama Y, Tian J, Locker J. Characterization of distant enhancers and promoters in the albumin-alpha-fetoprotein locus during active and silenced expression. J Biol Chem. 2006;281(40):30122–31. doi:10.1074/jbc.M603491200.
Bernier D, Thomassin H, Allard D, Guertin M, Hamel D, Blaquière M et al. Functional analysis of developmentally regulated chromatin-hypersensitive domains carrying the alpha 1-fetoprotein gene promoter and the albumin/alpha 1-fetoprotein intergenic enhancer. Mol Cell Biol. 1993;13(3):1619–33. doi:10.1128/mcb.13.3.1619-1633.1993.
Xie Z, Zhang H, Tsai W, Zhang Y, Du Y, Zhong J et al. Zinc finger protein ZBTB20 is a key repressor of alpha-fetoprotein gene transcription in liver. Proc Natl Acad Sci U S A. 2008;105(31):10859–64. doi:10.1073/pnas.0800647105.
Kojima K, Takata A, Vadnais C, Otsuka M, Yoshikawa T, Akanuma M et al. MicroRNA122 is a key regulator of α-fetoprotein expression and influences the aggressiveness of hepatocellular carcinoma. Nat Commun. 2011;2:338. doi:10.1038/ncomms1345.
Gao R, Cai C, Gan J, Yang X, Shuang Z, Liu M et al. miR-1236 down-regulates alpha-fetoprotein, thus causing PTEN accumulation, which inhibits the PI3K/Akt pathway and malignant phenotype in hepatoma cells. Oncotarget. 2015;6(8):6014–28. doi:10.18632/oncotarget.3338.
Xue J, Cao Z, Cheng Y, Wang J, Liu Y, Yang R et al. Acetylation of alpha-fetoprotein promotes hepatocellular carcinoma progression. Cancer Lett. 2020;471:12–26. doi:10.1016/j.canlet.2019.11.043.
Chen W, Peng J, Ye J, Dai W, Li G, He Y. Aberrant AFP expression characterizes a subset of hepatocellular carcinoma with distinct gene expression patterns and inferior prognosis. J Cancer. 2020;11(2):403–13. doi:10.7150/jca.31435.
Gurakar A, Ma M, Garonzik-Wang J, Kim A, Anders RA, Oshima K et al. Clinicopathological Distinction of Low-AFP-Secreting vs. High-AFP-Secreting Hepatocellular Carcinomas. Ann Hepatol. 2018;17(6):1052–66. doi:10.5604/01.3001.0012.7206.
Wu BX, Hong F, Zhang Y, Ansa-Addo E, Li Z. GRP94/gp96 in Cancer: Biology, Structure, Immunology, and Drug Development. Adv Cancer Res. 2016;129:165–90. doi:10.1016/bs.acr.2015.09.001.
Marzec M, Hawkes CP, Eletto D, Boyle S, Rosenfeld R, Hwa V et al. A Human Variant of Glucose-Regulated Protein 94 That Inefficiently Supports IGF Production. Endocrinology. 2016;157(5):1914–28. doi:10.1210/en.2015-2058.
Li X, Sun L, Hou J, Gui M, Ying J, Zhao H et al. Cell membrane gp96 facilitates HER2 dimerization and serves as a novel target in breast cancer. Int J Cancer. 2015;137(3):512–24. doi:10.1002/ijc.29405.
Hou J, Li X, Li C, Sun L, Zhao Y, Zhao J et al. Plasma membrane gp96 enhances invasion and metastatic potential of liver cancer via regulation of uPAR. Mol Oncol. 2015;9(7):1312–23. doi:10.1016/j.molonc.2015.03.004.
Ansa-Addo EA, Thaxton J, Hong F, Wu BX, Zhang Y, Fugle CW et al. Clients and Oncogenic Roles of Molecular Chaperone gp96/grp94. Curr Top Med Chem. 2016;16(25):2765–78. doi:10.2174/1568026616666160413141613.
Wu B, Chu X, Feng C, Hou J, Fan H, Liu N et al. Heat shock protein gp96 decreases p53 stability by regulating Mdm2 E3 ligase activity in liver cancer. Cancer Lett. 2015;359(2):325–34. doi:10.1016/j.canlet.2015.01.034.
Duan X, Iwanowycz S, Ngoi S, Hill M, Zhao Q, Liu B. Molecular Chaperone GRP94/GP96 in Cancers: Oncogenesis and Therapeutic Target. Front Oncol. 2021;11:629846. doi:10.3389/fonc.2021.629846.
Liu B, Yang Y, Qiu Z, Staron M, Hong F, Li Y et al. Folding of Toll-like receptors by the HSP90 paralogue gp96 requires a substrate-specific cochaperone. Nat Commun. 2010;1(6):79. doi:10.1038/ncomms1070.
Hong F, Liu B, Chiosis G, Gewirth DT, Li Z. α7 helix region of αI domain is crucial for integrin binding to endoplasmic reticulum chaperone gp96: a potential therapeutic target for cancer metastasis. J Biol Chem. 2013;288(25):18243–8. doi:10.1074/jbc.M113.468850.
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. 2012;1823(3):774–87. doi:10.1016/j.bbamcr.2011.10.013.
Rachidi S, Sun S, Wu BX, Jones E, Drake RR, Ogretmen B et al. Endoplasmic reticulum heat shock protein gp96 maintains liver homeostasis and promotes hepatocellular carcinogenesis. J Hepatol. 2015;62(4):879–88. doi:10.1016/j.jhep.2014.11.010.
Seacrist CD, Kuenze G, Hoffmann RM, Moeller BE, Burke JE, Meiler J et al. Integrated Structural Modeling of Full-Length LRH-1 Reveals Inter-domain Interactions Contribute to Receptor Structure and Function. Structure. 2020;28(7):830 – 46.e9. doi:10.1016/j.str.2020.04.020.
Huck JD, Que NL, Hong F, Li Z, Gewirth DT. Structural and Functional Analysis of GRP94 in the Closed State Reveals an Essential Role for the Pre-N Domain and a Potential Client-Binding Site. Cell Rep. 2017;20(12):2800–9. doi:10.1016/j.celrep.2017.08.079.
Reverter D, Lima CD. Insights into E3 ligase activity revealed by a SUMO-RanGAP1-Ubc9-Nup358 complex. Nature. 2005;435(7042):687–92. doi:10.1038/nature03588.
Gareau JR, Reverter D, Lima CD. Determinants of small ubiquitin-like modifier 1 (SUMO1) protein specificity, E3 ligase, and SUMO-RanGAP1 binding activities of nucleoporin RanBP2. J Biol Chem. 2012;287(7):4740–51. doi:10.1074/jbc.M111.321141.
Pierce BG, Hourai Y, Weng Z. Accelerating protein docking in ZDOCK using an advanced 3D convolution library. PLoS One. 2011;6(9):e24657. doi:10.1371/journal.pone.0024657.
Hicks DG, Tubbs RR. Assessment of the HER2 status in breast cancer by fluorescence in situ hybridization: a technical review with interpretive guidelines. Hum Pathol. 2005;36(3):250–61. doi:10.1016/j.humpath.2004.11.010.
Liu W, Zeng M, Fu N. Functions of nuclear receptors SUMOylation. Clin Chim Acta. 2021;516:27–33. doi:10.1016/j.cca.2021.01.007.
Pichler A, Gast A, Seeler JS, Dejean A, Melchior F. The nucleoporin RanBP2 has SUMO1 E3 ligase activity. Cell. 2002;108(1):109–20. doi:10.1016/s0092-8674(01)00633-x.
Dawlaty MM, Malureanu L, Jeganathan KB, Kao E, Sustmann C, Tahk S et al. Resolution of sister centromeres requires RanBP2-mediated SUMOylation of topoisomerase IIalpha. Cell. 2008;133(1):103–15. doi:10.1016/j.cell.2008.01.045.
Kirsh O, Seeler JS, Pichler A, Gast A, Müller S, Miska E et al. The SUMO E3 ligase RanBP2 promotes modification of the HDAC4 deacetylase. Embo j. 2002;21(11):2682–91. doi:10.1093/emboj/21.11.2682.
Sun Y, Demagny H, Schoonjans K. Emerging functions of the nuclear receptor LRH-1 in liver physiology and pathology. Biochim Biophys Acta Mol Basis Dis. 2021;1867(8):166145. doi:10.1016/j.bbadis.2021.166145.
Fayard E, Auwerx J, Schoonjans K. LRH-1: an orphan nuclear receptor involved in development, metabolism and steroidogenesis. Trends Cell Biol. 2004;14(5):250–60. doi:10.1016/j.tcb.2004.03.008.
Lee J, Yang DJ, Lee S, Hammer GD, Kim KW, Elmquist JK. Nutritional conditions regulate transcriptional activity of SF-1 by controlling sumoylation and ubiquitination. Sci Rep. 2016;6:19143. doi:10.1038/srep19143.
Rosonina E, Akhter A, Dou Y, Babu J, Sri Theivakadadcham VS. Regulation of transcription factors by sumoylation. Transcription. 2017;8(4):220–31. doi:10.1080/21541264.2017.1311829.
Werner A, Flotho A, Melchior F. The RanBP2/RanGAP1*SUMO1/Ubc9 complex is a multisubunit SUMO E3 ligase. Mol Cell. 2012;46(3):287–98. doi:10.1016/j.molcel.2012.02.017.
Liu X, Liu J, Xiao W, Zeng Q, Bo H, Zhu Y et al. SIRT1 Regulates N(6) -Methyladenosine RNA Modification in Hepatocarcinogenesis by Inducing RANBP2-Dependent FTO SUMOylation. Hepatology. 2020;72(6):2029–50. doi:10.1002/hep.31222.
Blondel-Tepaz E, Leverve M, Sokrat B, Paradis JS, Kosic M, Saha K et al. The RanBP2/RanGAP1-SUMO complex gates β-arrestin2 nuclear entry to regulate the Mdm2-p53 signaling axis. Oncogene. 2021;40(12):2243–57. doi:10.1038/s41388-021-01704-w.
Hoter A, El-Sabban ME, Naim HY. The HSP90 Family: Structure, Regulation, Function, and Implications in Health and Disease. Int J Mol Sci. 2018;19(9). doi:10.3390/ijms19092560.
Kim JW, Cho YB, Lee S. Cell Surface GRP94 as a Novel Emerging Therapeutic Target for Monoclonal Antibody Cancer Therapy. Cells. 2021;10(3). doi:10.3390/cells10030670.
Koo BH, Apte SS. Cell-surface processing of the metalloprotease pro-ADAMTS9 is influenced by the chaperone GRP94/gp96. J Biol Chem. 2010;285(1):197–205. doi:10.1074/jbc.M109.039677.
Niu M, Xu J, Liu Y, Li Y, He T, Ding L et al. FBXL2 counteracts Grp94 to destabilize EGFR and inhibit EGFR-driven NSCLC growth. Nat Commun. 2021;12(1):5919. doi:10.1038/s41467-021-26222-x.
Hou J, Deng M, Li X, Liu W, Chu X, Wang J et al. Chaperone gp96 mediates ER-α36 cell membrane expression. Oncotarget. 2015;6(31):31857–67. doi:10.18632/oncotarget.5273.
Fan Q, Mao H, Wu C, Liu Y, Hu Y, Zhong B et al. ATF4 (activating transcription factor 4) from grass carp (Ctenopharyngodon idella) modulates the transcription initiation of GRP78 and GRP94 in CIK cells. Fish Shellfish Immunol. 2014;38(1):140–8. doi:10.1016/j.fsi.2014.03.004.
Fan H, Yan X, Zhang Y, Zhang X, Gao Y, Xu Y et al. Increased expression of Gp96 by HBx-induced NF-κB activation feedback enhances hepatitis B virus production. PLoS One. 2013;8(6):e65588. doi:10.1371/journal.pone.0065588.
Ji F, Zhang Y, Zhu ZB, Guo Y, Shen SL, Cao QH et al. Low levels of glycoprotein 96 indicate a worse prognosis in early-stage hepatocellular carcinoma patients after hepatectomy. Hum Pathol. 2019;86:193–202. doi:10.1016/j.humpath.2018.11.025.
Chen WT, Tseng CC, Pfaffenbach K, Kanel G, Luo B, Stiles BL et al. Liver-specific knockout of GRP94 in mice disrupts cell adhesion, activates liver progenitor cells, and accelerates liver tumorigenesis. Hepatology. 2014;59(3):947–57. doi:10.1002/hep.26711.
Zheng Y, Zhu M, Li M. Effects of alpha-fetoprotein on the occurrence and progression of hepatocellular carcinoma. J Cancer Res Clin Oncol. 2020;146(10):2439–46. doi:10.1007/s00432-020-03331-6.
Piñero F, Dirchwolf M, Pessôa MG. Biomarkers in Hepatocellular Carcinoma: Diagnosis, Prognosis and Treatment Response Assessment. Cells. 2020;9(6). doi:10.3390/cells9061370.
Trevisani F, D'Intino PE, Morselli-Labate AM, Mazzella G, Accogli E, Caraceni P et al. Serum alpha-fetoprotein for diagnosis of hepatocellular carcinoma in patients with chronic liver disease: influence of HBsAg and anti-HCV status. J Hepatol. 2001;34(4):570–5. doi:10.1016/s0168-8278(00)00053-2.
Berry K, Ioannou GN. Serum alpha-fetoprotein level independently predicts posttransplant survival in patients with hepatocellular carcinoma. Liver Transpl. 2013;19(6):634–45. doi:10.1002/lt.23652.

No competing interests reported.

Fig.S1.tiff
Fig. S1 Oil red O staining of a tumor section from Het and KO mice liver.
Fig.S2.tiff
Fig. S2 Huh7 mock cells were transfected with siRNAs specific for Nkx2.8, NR5A2, or HNF4, control siRNA, or untransfected as control. At 48 h post-transfection, the protein levels of NR5A2 (a), Nkx2.8 (b), or HNF4 (c) were determined by western blot assay. Huh7 mock cells were transfected with NR5A2 siRNAs and control siRNA as mock. At 48 h post-transfection, gp96 and NR5A2 mRNA levels were measured by Real-time PCR. Average values with standard deviations are shown. **P < 0.01 compared with the control group.
Fig.S3.tiff
Fig. S3 NR5A2 mRNA or protein levels were measured by Real-time PCR (a) or Western blotting (b) analysis in tumor tissues of gp96-KO and Het mice liver treated with DEN (n = 5/group). Average values with standard deviations are shown. (c) IHC analysis of NR5A2 expression in paraffin tumor tissue sections of Het and gp96-KO mice. **P < 0.01 compared with the control group.
Fig.S4.tiff
Fig. S4 Recombinant GST-gp96, its N-terminal, M-domain, and C-terminal fragments (a), GST-NR5A2 (aa 299-536) (b), and His-gp96 (c) were expressed in E. coli. The recombinant proteins were purified by GST Sepharose 4FF or Ni-NTA column, and assessed by SDS-PAGE followed by Coomassie Blue staining.
Fig.S5.tiff
Fig. S5 SUMO1 (a) or SUMO2/3 (b) in whole cell lysates of Huh7-gp96KD and Huh7-gp96 were detected by western blotting.
Fig.S6.tiff
Fig. S6 RanBP2 was assessed by western blotting in Huh7-gp96KD, Huh7-Luci (mock), or Huh7 control cells.
Fig.S7.tiff
Fig. S7 (a)TOP2A and HDAC4 were analyzed by Western blotting in Huh7-gp96KD, Huh7-Luci (mock), or Huh7 control cells. (b-c) Co-IP assays were performed with anti-gp96 (b), anti-TOP2A (c), or anti-HDAC4 (d) antibodies in Huh7 cells followed by the Western blotting analysis of indicated proteins. IgG as the isotype control for Co-IP. (e and f) TOP2A or HDAC4 was immunoprecipitated with the TOP2A or HDAC4 antibody in Huh7-gp96KD or mock cells. RanBP2 levels in the immunoprecipitates were measured by Western blotting. Whole-cell lysates (WCL) were also blotted for respective proteins.
Fig.S8.tiff
Fig. S8 The recombinant GST-NR5A2 (aa 299-536) and its C-terminal truncated GST-NR5A2 (aa 299-506) were expressed in E. coli, purified, and assessed by SDS-PAGE followed by Coomassie Blue staining.
Additionalfile1transcriptome.xlsx
Additionalfile2mass.xlsx
Additionalfile3docking.zip
SupplementaryTables.docx

Download PDF

Version 2

posted

You are reading this latest preprint version

Cellular gp96 upregulates AFP expression by blockade of NR5A2 SUMOylation and ubiquitination in HCC

Status:

Version 2

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 2