MDM2 Binding Protein Induces the Resistance of Hepatocellular Carcinoma Cells to Molecular Targeting Agents via Enhancing the Transcription Factor Activity of the Pregnane X Receptor

The MDM2 binding protein (MTBP) has been considered an important regulator of human malignancies. In this study, we demonstrate that the high level of MTBP’s endogenous expression is correlated with poor prognosis of advanced hepatocellular carcinoma (HCC) patients who received sorafenib. MTBP interacted with the Pregnane X receptor (PXR) and enhanced the transcription factor activity of PXR. Moreover, MTBP enhanced the accumulation of PXR in HCC cells’ nuclear and the recruitment of PXR to its downstream gene’s (cyp3a4’s) promoter region. Mechanically, the knockdown of MTBP in MHCC97-H cells with high levels of MTBP decelerated the clearance or metabolism of sorafenib in HCC cells and led to the resistance of HCC cells to sorafenib. Whereas overexpression of MTBP in in MHCC97-L cells with low levels of MTBP showed the opposite trend. By establishing the interaction between MTBP and PXR, our results indicate that MTBP could function as a co-activator of PXR and could be a promising therapeutic target to enhance the sensitivity of HCC cells to molecular targeting agents.


INTRODUCTION
The hepatocellular carcinoma (HCC) has been a fatal threat to the health of the people of China due to the high rates of hepatitis and the resistance or insensitivity to HCC cells in relation to molecular targeting agents. This has been considered an important influencing factor for poor prognoses of patients with advanced HCC (1)(2)(3)(4). The PXR (pregnane X receptor) is a member of the nuclear receptors super-family and mainly expresses in the liver and gastrointestinal tract tissues (5,6). It has long been considered to be the regulatory center for the metabolism and detoxification of exogenous drugs and toxicants; results of recent studies show that it is also a key regulator for anti-tumor drug resistance in cancerous cells (7)(8)(9)(10). Upon binding and activating by ligands, PXR can recruit the promoter or enhancer regions to mediate the transcription of its downstream drug-resistance-related genes, inducing cyp3a4 and abcb1 (also named multi-drug resistance 1 [mdr-1]/ATP binding cassette subfamily B member 1 [ABCB1]). It does this by encoding the P-glycoprotein [P-gp]) to mediate the elimination of sorafenib in cancerous cells via the N-oxide metabolism of sorafenib or by decreasing the intracellular accumulation of it, which in turn reduces the clinical efficacy or resistance of the drug during treatment (5)(6)(7)(8)(9)(10).
The MDM2 binding protein (MTBP) has a molecular weight of 104 kDa and was originally found to interact with MDM2 (murine double minute 2) (11,12). Recently, MTBP has been regarded as an important regulator of human cancer cells' proliferation or metastasis (13,14). MTBP could promote human cancer cells' proliferation or metastasis by functioning as a co-activator of c-MYC (cellular-myelocytomatosis viral oncogene) or ZEB2 (Zinc Finger E-Box Binding Homeobox 2) (13). In this study, MTBP was found to function as a co-activator of PXR, further enhancing PXR's downstream drug-resistancerelated genes. MTBP was highly expressed in HCC clinical specimens compared to those of the paired non-tumor tissues. The knockdown of MTBP in MHCC97-H with high levels of endogenous MTBP could have decelerated the elimination or clearance of sorafenib and enhanced the antitumor effect of sorafenib in MHCC97-H cells. Whereas overexpression of MTBP in in MHCC97-L cells with low levels of MTBP showed the opposite trend.

Clinical Specimens and Agents
The use of clinical specimens and cell lines were approved by the ethics committee of the Fifth Medical Center, General Hospital of Chinese PLA (People's Liberation Army). All methods and experiments were carried out in accordance with the Helsinki Declaration. A total number of 52 patients with advanced HCC received sorafenib as described in our previous publications (Feng et al. [2018] and Shao et al. [2018]) (7,8). The cDNA samples were conserved at -80°C (7,8). The expression of MTBP in the cDNA of the patients was examined by quantitative polymerase chain reaction (qPCR). The hepatic cell lines L-02, HepG2, MHCC97-H, MHCC97-L, BEL-7402, SMMC-7721 and Hu7 cells were conserved in our lab and cultured in Dulbecco's Modified Eagle Medium (DMEM, Hyclone, Thermo Fisher Scientific Corporation, Waltham, MA, USA). We added Fetal Bovine Serum (FBS, Invitrogen, Thermo Fisher Scientific Corporation, Waltham, MA, USA) at 37°C in DMEM to culture cells. The antitumor agents (the molecular targeted agents), sorafenib, regorafenib, lenvatinib, apatinib and anlotinib, were chemically synthesized by Dr. and Prof. Shuang Cao in Wuhan Institute of Technology. Rifampicin, a typical agonist of PXR, and ketoconazole, a typical antagonist of PXR, were purchased from Selleck Corporation.
For cell-based experiments, the molecular targeting agents were dissolved in DMSO (dimethyl sulfoxide) and diluted with phenol red-free DMEM (Thermo Fisher Scientific Corporation) (15,16). For the animal experiments, the formulation (oral liquid for the oral administration) of molecular targeting agents was prepared by using PEG400 and Twenn80 (17,18). The lentivirus particles with the full length of MTBP and the siRNA of MTBP were purchased from the Vigene Corporation, Jinan City, Shandong Province, China.

Luciferase Examination
The luciferase reporters of PXR (PXRE-Luc, XREM-Luc, DR3-Luc and ER6-Luc) were described in our previous publication (7,8). The cells were cultured and suspended using the phenol redfree DMEM supplemented with 0.5% charcoal-stripped fetal bovine serum (FBS; Hyclone, Logan, Utah, USA). Cells were transfected with the MTBP and siMTBP and analyzed by luciferase activation or b-gal activation following the instructions provided in methods section of the manuscript by Yang et al., .

Western Blot Experiments
Total protein samples were extracted from HCC cells and the subcutaneous tumor tissues; western blot experiments were performed following the methods described in our previous publication. The antibodies against P-GP (P-glycoprotein, encoded by abcb1), CYP3A4, Lamin A, PXR and b-Actin were purchased from Abcam Corporation, Cambridge, UK. The expression levels of proteins were examined by their antibodies. The blots were visualized via the chemiluminescence by use of an ECL kit (Amersham Biosciences, Piscataway, NJ, USA).

Pharmacokinetic Examination
The sustaining of sorafenib in HCC cells and tumor tissues was examined by liquid chromatography-mass spectrometry/mass spectrometry (LC-MS/MS) methods to reveal the elimination or clearance of sorafenib in HCC cells. Pharmacokinetic experiments were performed according to the protocols described in our previous work (7,8).

ChIP Experiments
The HCC cells were transfected with the siRNA of MTBP and MTBP were treated with rifampicin for 30min. Then, cells were harvested for ChIP analysis and the experiments were performed

The Subcellular Sub-Fraction Analysis
The HCC cells were transfected with the siRNA of MTBP and MTBP were treated with rifampicin for 30min to 1h. Next, the cells were harvested and the subcellular sub-fraction was performed following the methods provided by Yang et al. and Lu et al. (26,27). Lamin A was used as the indicator of the nuclear sub-fraction and b-Actin was chosen as the indicator of the cytoplasm sub-fraction. The protein levels of MTBP, PXR, Lamin A and b-Actin were examined by their antibodies.

Cell Survival Examination
The HCC cells were transfected with the siRNA of MTBP and MTBP were treated with the indicated concentration of molecular targeting agents (10mmol/L, 3mmol/L, 1mmol/L, 0.3mmol/L, 0.1mmol/L, 0.03mmol/L and 0.01mmol/L) for 48h.

The In Vivo Antitumor Effect of Molecular Targeting Agents on HCC Cells
The animal experiments were reviewed approved by the Institutional Animal Care and Use Committee, the Fifth Medical Center, Chinese PLA. All animal experiments were performed in accordance with the UK Animals (Scientific Procedures) Act, 1986 and the associated guidelines. Nude mice ages 4 to 6 weeks were purchased from the Si-Bei-Fu Corporation (Beijing, China). For the subcutaneous tumor model, HCC cells were cultured and harvested to prepare the single cell suspension. Then, the cell-suspension was injected into the subcutaneous positions of the mice (about 1×10 6 cells for each injection point). Three to four days after injection, mice received the molecular targeting agents via oral administration once every two days. After three to four weeks' treatment, the mice were collected and their tumor volumes were calculated as width 2 *length/2 (31,32). Their tumor weights were measured with a precise balance. The inhibition rates of molecular targeting agents on HCC cells' subcutaneous growth were calculated based on tumor weights and tumor volumes. For the intrahepatic tumor model, HCC cells were first injected into the mice to form tumor tissues (7,8). Next, the tumors were collected to prepare micro-blocks of tumor tissues which were mixed with medical hydrogel to form hydrogel drips with tumor tissues. Supplemental Table 1 shows the weights of the microblocks formed by the MHCC97-H cells and the MHCC97-L cells and transplanted into the mice. The hydrogel drips were adhered onto the surface of the liver organs of the mice. Three to four days after injection, the mice received the molecular targeting agents via oral administration once every two days. After three to four weeks' treatment, mice received a micro-PET screening to examine the intrahepatic growth of HCC cells. After micro-PET screening, the mice were harvested and their livers with nodules formed by HCC cells were collected (7,8). The results were shown as micro-PET images, the quantitative results of micro-PET, the images of liver organs with nodules or the quantitative results of liver organs' images with nodules formed by HCC.

The Statistical Analysis
Statistical significance analyses were analyzed by using the statistical software (software version: SPSS 9.0, the IBM corporation, Armonk, New York, USA). The IC50 values or the Statistical significance was analyzed by Bonferroni correction with or without two ways ANOVA.

MTBP Enhanced the Transcription Factor Activation of PXR
First, the expression of MTBP in hepatic cells was examined. As shown in Figure 1, among hepatic cell lines, MHCC97-H, a highly aggressive HCC cell line, expressed the highest level of endogenous MTBP, whereas MHCC97-L, an HCC cell line with low aggressiveness, expressed the lowest level of endogenous MTBP. Therefore, MHCC97-H cells were used to knockdown MTBP's expression and MHCC97-L cells were used to overexpress MTBP.
Next, the effect of MTBP on PXR's transcription factor activation was examined. As shown in Table 1, rifampicin, a typical agonist of PXR, induced the transcription factor activation of PXR. The overexpression of MTBP enhanced the activation of PXR induced by rifampicin and the EC 50 values decreased; the knockdown of MTBP decreased the activation of PXR induced by rifampicin and the EC 50 values increased ( Table 1).
Moreover, the overexpression of MTBP enhanced the mRNA or protein levels of PXR's downstream genes, cyp3a4 and abcb1, induced by rifampicin. The knockdown of MTBP decreased the mRNA and protein levels of PXR's downstream genes, cyp3a4 and abcb1, induced by rifampicin ( Table 1 and Figure 2). Therefore, MTBP enhanced the transcription factor activation of PXR.

MTBP Interacted With PXR and Promoted the Accumulation of PXR in Nuclear and the Recruitment of PXR to the Promoter of Its Downstream Gene cyp3a4
To further examine the effect of MTBP on PXR, the protein interaction between MTBP and PXR was examined by coimmunoprecipitation (co-IP). As shown in Figure 3, MTBP interacted with PXR in MHCC97-H cells. Additionally, the overexpression of MTBP promoted the accumulation of PXR in nuclear induced by rifampicin, whereas the knockdown of MTBP deceased the accumulation of PXR in nuclear ( Figure 4). The recruitment of PXR to the promoter of its downstream gene, cyp3a4, was examined by ChIP analysis (Figure 4). As shown in Figure 4, the overexpression of MTBP promoted the recruitment of PXR to the promoter region of cyp3a4. The knockdown of MTBP decreased the recruitment of PXR to the gene's promoter region ( Figure 4). Therefore, MTBP interacted with PXR and promoted the accumulation of PXR in nuclear and the recruitment of PXR to the promoter of its downstream gene.

MTBP Accelerated the Elimination of Sorafenib in HCC Cells
These results indicate that MTBP enhanced the activation of PXR and enhanced the expression of PXR's downstream genes which mediated the metabolism of sorafenib. Therefore, the effects of MTBP on the elimination or clearance of sorafenib in HCC cells were examined by the LC-MS/MS methods. As shown in Figure 5 and Table 2, the overexpression of MTBP in

MTBP Enhanced the Resistance of HCC Cells to Molecular Targeting Agents
The antitumor effect of molecular targeting agents on HCC cells was examined by MTT experiments. As shown in Supplemental Table 2, sorafenib inhibited the survival of HCC cells in a dose-dependent manner. The overexpression of MTBP enhanced the resistance of MHCC97-L cells to sorafenib and the IC 50 values of sorafenib on MHCC97-L cells were increased (Supplemental Table 2). The knockdown

MTBP Repressed the In Vivo Antitumor Effect of Sorafenib on HCC Cells' Subcutaneous Growth and the Intrahepatic Growth
To further examine the effect of MTBP on the in vivo antitumor activation of molecular targeting agents, the subcutaneous tumor model was used. As shown in Figure 6 and Table 2, the molecular targeting agents inhibited the subcutaneous growth of HCC cells in nude mice. The overexpression of MTBP decreased the antitumor effect of molecular targeting agents and the IC 50 values increased ( Figure 6 and Table 3). The knockdown of MTBP enhanced the sensitivity of MHCC97-H cells to molecular targeting agents and the IC 50 values decreased ( Figure 6 and Table 3). The results were shown as the images of subcutaneous tumor tissue formed by MHCC97-H from the mice received sorafenib treatment and the IC 50 values of molecular targeting agents sorafenib, regorafenib, lenvatinib, anlotinib or apatinib ( Figure 6 and Table 3). The expression level of the related factors (MTBP, PXR, abcb1 or cyp3a4) in the subcutaneous tumors ( Figure 6A) was shown as Figures 6F-I. Next, the intrahepatic tumor models were used. As shown in Figure 7, the intrahepatic growth of HCC cells could form the intrahepatic nodules (tumor lesions) in the livers of the mice. The intrahepatic growth of HCC cells were shown as the micro-PET images (live imaging of small animals) and the images of livers with nodules ( Figure 7). Treatment with sorafenib inhibited the intrahepatic growth of HCC cells in the liver of each of the mice (Figure 7). The overexpression of MTBP repressed the antitumor effect of sorafenib on both the micro-PET intensity images of the livers and the nodules formed by MHCC97-L (Figure 7). The knockdown of MTBP enhanced the antitumor effect of sorafenib both on the micro-PET intensity images of the livers and the nodules formed by MHCC97-H ( Figure 7). Therefore, MTBP enhanced the resistance of HCC cells to molecular targeting agents.

MTBP Is Associated With the Poor Prognosis of HCC Patients
The above results indicate that MTBP could interact with PXR and enhance the transcription activation of PXR to mediate the resistance of HCC cells to sorafenib. To confirm the specificity of MTBP and the clinical significance of MTBP, the expression of MTBP in advanced HCC clinical specimens was examined and the patients were divided into two groups (MTBP-high group and MTBP-low group) according to the median value of the endogenous MTBP level (Figures 8A-D). As shown in Figures 8C, D and    Moreover, the relationship between the expression of MTBP with the downstream genes of PXR, cyp3a4 and ABCB1 in clinical specimens was examined. As shown in Figures 8E, F, the expression of MTBP-1 was positively related to the expression of cyp3a4 (P<0.0001) and ABCB1 (P<0.0001) in clinical specimens. As expectative, the expression of MTBP in HCC specimens did not associate with PXR ( Figure 8G). Therefore, the data from the clinical specimens confirmed MTBP's pro-proliferative and oncogenic activation and the effect of MTBP on PXR's activation.

The Specificity of MTBP on PXR's Activation
Further experiments were performed in Huh-7 cells (P53deficient HCC cells), HEK293 (PXR-negative cell line), and using PXR antagonist ketoconazole treating cells. The activation of PXR was revealed by the EC 50 values of rifampicin on luciferase reporters or mRNA level of PXR's downstream genes. As shown in Supplemental Table 3, overexpression of MTBP enhanced the activation of luciferase reporters or the mRNA level of PXR's downstream genes, cyp3a4   Table 4). Therefore, The specificity of MTBP on PXR's activation was confirmed by using the multi-assays.

DISCUSSION
The molecular targeting agents represented by sorafenib are of great significance for the treatment of advanced HCC but it has been clearly reported that patients' sensitivity to the drug varies according to each individual. During the treatment, patients are prone to developing resistance to molecular targeting agents (33). Some cellular signaling pathways, including CSCs (cancer stem cells), the Notch pathway, the EMT (epithelial-mesenchymal transition) and mTOR (mammalian target Rapamycin) can affect the antitumor activity of molecular targeting agents (34)(35)(36)(37)(38)(39)(40). Results from recent studies have shown that metabolic factors are also important mechanisms for the resistance of various molecular targeted therapies. Yin et al. reported that the use of the SREBP-1 inhibitor Betulin down-regulated the transcription factor activity of SREBP-1 to increase the sensitivity of HCC cells to sorafenib (23). Ma et al. reported that miR-6077 can promote the antitumor effect of anlotinib on NSCLC (non-small cell lung cancer) cells by downregulating the expression level of GLUT-1 (glucose transporter 1) sensitivity (41). SREBP-1 (Sterol Regulatory Element Binding Protein-1) and GLUT-1 are important regulators of cellular material and energy metabolism (42)(43)(44)(45) for modulating the microenvironment of tumor cells. Therefore, it is necessary to further examine the participation of the aberrant metabolism of malignant tumor cells in the progress or resistance of cancer cells to antitumor agents (46). PXR can act as a regulatory hub for the elimination and detoxification of exogenous drugs and toxicants through its downstream resistance-related genes, known to induce the metabolism and elimination of molecularly-targeted drugs, resulting in the resistance of HCC cells to the drugs (47,48). In this study, MTBP interacted with PXR and enhanced the transcription factor activation of PXR which eventually caused the accelerated elimination or clearance of sorafenib. These results not only extend our knowledge about PXR's regulation but also our knowledge about resistance to the drug. MTBP was first identified by an MDM2-interacting protein (49). Recently, multiple points of evidence have indicated the pro-proliferative and oncogenic activation of MTBP. This activation has been carried out by multiple mechanisms: (1) MTBP is co-amplified with c-MYC and functions as c-Myc's coactivator (50); (2) MTBP could enhance the activation of ZEB2 to induce the EMT process of cancerous cells (13); (3) MTBP inhibits the apoptosis of cancer cells via suppression of the MDM2/P53 axis (14); and (4) MTBP can enhance the metastasis of HCC cells via the MDM2-Mediated E-Cadherin Degradation (11).
In this study, MTBP interacted with PXR and functioned as the co-activator of PXR to enhance the resistance of HCC cells to molecular targeting agents. Our results are consistent with those of our previous publication in which we mentioned that MTBP can function as an activator of ZEB or c-MYC (50). ZEB is the important transcription factor mediating the EMT process of human cancer cells; c-MYC not only promotes the proliferation of cancer cells, it also participates in the aberrant metabolism of cancer cells (51)(52)(53)(54)(55). Therefore, in the future it would be valuable to examine the interaction between MTBP and the transcription factors. MTBP has also been found to act as a tumor suppressor in some cases. Findings from our previous work and related data indicate that MTBP may not significantly affect the proliferation of HCC cells but it does inhibit the metastasis of HCC cells by inhibiting the activation of ACTN (alpha-actinin 4) (56)(57)(58)(59). It is possible that ACTN is the main mechanism by which MTBP functions and the different expressions of ACTN, c-MYC and ZEB in different cells may affect the role of MTBP (56)(57)(58)(59).
In this study we not only detected the expression of the drugresistant genes CYP3A4 and ABCB1, we also directly observed the effects of MTBP on the metabolism and clearance of the molecularly-targeted drug sorafenib. The overexpression of MTBP can accelerate the metabolism and clearance of sorafenib in HCC cells and tissues while the knocking down of MTBP expression can prolong the retention time of sorafenib in   Cells were first injected into mice to form subcutaneous tumor tissues; the micro-blocks of subcutaneous tissues were transplanted into the liver of each mouse to form intrahepatic tumor lesions (nodules). Then, the mice were given a sorafenib (0.5mg/kg) treatment via oral administration. The results were shown as the images of micro-PET, the quantitative results of micro-PET images (A) or the images of liver organs with nodules formed by HCC cells, the radio-activation of liver organs to blood, the intensity of liver regions to body-background (micro-PET's images) or the relative area of the nodules. (B) MHCC97-L cells were cultured and transfected with empty vectors or the vectors of MTBP. Cells were first injected into mice to form subcutaneous tumor tissues and the micro-blocks of subcutaneous tissues were transplanted into the each liver to form the intrahepatic tumor lesions (nodules). Then, the mice were given sorafenib (3.0mg/kg) treatment via oral administration. The results were shown as the images of micro-PET, the quantitative results of micro-PET images (A) or the images of liver organs with nodules formed by HCC cells, the radio-activation of liver organs to blood, the intensity of liver regions to body-background (micro-PET's images) or the relative area of nodules. *P < 0.05 The white arrow in Figure 7 indicated the liver region from mirco PET examination; whereas the black arrow in Figure 7 indicated the nodules or lesions formed by HCC cells in the liver organs.
HCC cells and tissues. Because the drug has been used in clinical treatments for a long time, its drug metabolism-related characteristics, such as those noted in our previous results, have established an LC-MS/MS method for detecting it. The application time of several other new molecules is shorter than that of sorafenib and data related to LC-MS/MS methodologies for these drugs is rarely reported. It will be valuable to establish these methods in the future for detecting the effects of MTBP on the metabolism and clearance of regorafenib, lenvatinib, apatinib and anlotinib in HCC cells.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by The Fifth Medical Center of Chinese PLA General Hospital. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by The Fifth Medical Center of Chinese PLA General Hospital. Written informed consent was obtained from the owners for the participation of their animals in this study.