Suppression of Cancer Cell Stemness and Drug Resistance via MYC Destabilization by Deubiquitinase USP45 Inhibition with a Natural Small Molecule

Simple Summary Cancer stem cells (CSCs) play significant roles in cancer development, drug resistance and cancer recurrence. Thus, it is of great importance to study and target the mechanism by which CSCs are regulated. On the basis of our investigations, we have discovered that USP45 as a new deubiquitinase of MYC significantly promoted cervical cancer development, stemness and drug resistance. Our findings have established the close connection among USP45, MYC and CSCs. Moreover, we have identified that USP45 can be specifically bound and inhibited by a natural small molecule (α-mangostin), in turn significantly suppressing the USP45-induced stemness and drug resistance of CSCs. On the basis of our USP45 discoveries, a new window has opened for suppressing CSCs development, stemness and drug resistance. Our exciting discovery will attract a broad audience in clinical CSCs target therapy, signaling pathways, natural products, drug discovery and drug development. Abstract Cancer stem cells (CSCs) play significant roles in cancer development, drug resistance and cancer recurrence. In cancer treatments based on the CSC characteristics and inducing factors, MYC is a promising target for therapeutic molecules. Although it has been regarded as an undrugable target, its stability tightly regulated by the ubiquitin–proteasome system offers a new direction for molecule targeting and cancer treatment. Herein we report our discoveries in this research area, and we have found that deubiquitinase USP45 can directly bind with MYC, resulting in its deubiquitination and stabilization. Further, USP45 overexpressing can upregulate MYC, and this overexpressing can significantly enhance cancer development, cancer cell stemness and drug resistance. Interestingly, without enhancing cancer development, MYC silencing with shRNA can only suppress USP45-induced stemness and drug resistance. Moreover, we have identified that USP45 can be specifically bound and inhibited by a natural small molecule (α-mangostin), in turn significantly suppressing USP45-induced stemness and drug resistance. Since USP45 is significantly expressed in cervical tumors, we have discovered that the combination of α-mangostin and doxorubicin can significantly inhibit USP45-induced cervical tumorigenesis in an animal model. In general, on the basis of our USP45 discoveries on its MYC deubiquitination and α-mangostin inhibition, suppressing USP45 has opened a new window for suppressing cancer development, stemness and drug resistance.


Introduction
Although great progress has been made in understanding the mechanisms of cancer development, the disease's treatments are still unsatisfactory due to many scientific and promoted MYC degradation and, in turn, reduced MYC-mediated stemness and drug resistance in cervical cancer. Our results highlight USP45 as a critical factor in deciding cervical CSC fate. In conclusion, we have demonstrated that USP45 may play a critical role in enhancing cell survival after cancer treatments, and USP45 inhibition causing MYC destabilization may provide a new strategy for clinical CSC treatments.

Plasmids, siRNA and shRNA Information
The plvx-puro vector was used for all protein expressing, and purchased from Shanghai Haijihaoge Biotechnology. The pCMV-Ub and pCMV-Ub-K 0 vector was purchased from Shanghai Haijihaoge Biotechnology. The pLkO.1-MYC-shRNA was from Cheng-hua Li laboratory (Sichuan University, College of Life Science), and pLKO.1-GFP-shRNA were purchased from Biofeng Biotechnology. The mutants (USP45 C199A , K 48 and K 63 ) were generated using a QuickMutation™ Plus kit (D0208S, Beyotime, Shanghai, China). All plasmids used in this research were confirmed by DNA sequencing.

Bioinformatics Analyses
Most of the bioinformatics analysis in this research could be implemented via these public database platforms, the functions of which are listed below: The "TIMER2.0" database (http://timer.cistrome.org/, accessed on 9 May 2021) was utilized for USP45 pan-cancer expression analysis.
The "GSE63514" dataset from GEO database was processed using R studio, and the mRNA levels of USP45 and MYC in each sample were statistically displayed using Graphpad prism 6.0.

Plasmid Transfection, RNA Interference and Lentiviral Infection
Cells at 50% confluence were transfected with plasmids or siRNA using Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA), cells were collected at 48-60 h after transfection.

Western Blot
Cells were collected, washed with PBS and re-suspended in RIPA lysis buffer. Equal amounts of total proteins were loaded, separated by SDS-PAGE, transferred to PVDF membranes (ISEQ00010, Millipore, Germany), and hybridized to an appropriate primary antibody and HRP-conjugated secondary antibody for subsequent detection by enhanced chemiluminescence. Western Blot data were quantitated and processed via Image Lab 5.0 software (Bio-Rad, www.bio-rad.com, accessed on 18 May 2018). Original blots can be found at Supplementary Figures S5-S8.

Ubiquitination Assays
HEK293T cells at 50% confluence were transfected with MYC-HA, Ub-Flag and USP45-His using Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA), cells were treated with MG132 (10 µM) at 42 h and collected at 48 h after transfection. Protein was extracted using a Western/IP lysis buffer (P0013, Beyotime, China). HA-beads were washed twice with lysis buffer and incubated with protein for 6-9 h in a 4 • C shaker. HA-beads were collected, washed twice with PBS and then added with protein loading buffer for 10 min at 95 • C. Ubiquitination was detected by Western Blot.

Flow Cytometry Assay
CD133+ cell analysis: cells were harvested and washed twice with PBS, then were re-suspended with the staining buffer (00440942, Invitrogen, Carlsbad, CA, USA). The cells were incubated with CD133-PE antibody (12-1338-42, Invitrogen, Carlsbad, CA, USA) for 30 min at 4 • C. The cells were then washed twice with staining buffer and re-suspended using staining buffer, and analyzed by FACS machine (BD LSRfortessa, Franklin Lakes, NJ, USA).

Colony Formation Assay
A total of 2 × 10 3 cells were cultured on a 6-well plate. After 14 days, the colonies were washed with PBS, treated with 4% paraformaldehyde, and visualized using Diff-Quik stain (G1540, Solarbio, Beijing, China). Only those colonies containing more than 20 cells were counted.

MTT Assay
Cell viability was analyzed using the MTT agent (C0009S, Beyotime, Shanghai, China). A total of 5 × 10 3 cells were cultured on the 96-well plates. Cells were treated with doxorubicin, α-mangostin, or their combination for 48 h, then 200 µL of MTT (5 mg/mL) was added to each well. The samples were incubated at 37 • C for 4 h, and then sub-cultured in medium with 200 µL of DMSO (ST038, Beyotime, Shanghai, China). The absorbance of each well was determined at 492 nm. Three independent experiments were performed.

Hoechst 33,342 Assay
A total of 5 × 10 3 cells were cultured on the 96-well plates. After 12 h, the cells were washed with PBS, treated with Hoechst 33,342 (C0030, Solarbio, Beijing, China), photographed with UV light under a light microscope at high magnification, and the absorbance of each well was determined with a microplate reader at the excitation wavelength (350 nm) and the emission wavelength (461 nm).

Virtual Docking
The USP45 protein structure was simulated using the I-TASSER model [47] and converted to PDBQT format. The α-mangostin structure was downloaded from the PubChem database (Compound ID = 5281650). AutoDock Tools software was used for pretreatment and construction of protein-ligand complexes. AutoDock qVina2 was used for virtual docking between USP45 and α-mangostin. The docking region was framed as the whole USP45 protein, with x = 75.677, y = 78.918 and z = 79.734 axes as the center, and size x = 80, size y = 96 and size z = 76 grids. In virtual docking, 20 docking binding modes were generated, the docking affinities of the top 3 were presented and further analyzed. In addition, virtual docking between USP45 and α-mangostin can be finished directly with CB-DOCK2 [48].

USP45 Expression and Purification
BL21 expressing USP45-His was cultured in TB medium and incubated in a 200 rpm shaker (at 37 • C) until OD = 0.6-0.8. IPTG was added to the medium with the final concentration of 1 mM and was placed in a shaker at 200 rpm at 18 • C overnight. E. coli cells were harvested and resuspended with 10 mL (per 1 g bacteria) of the binding buffer (500 mM NaCl, 50 mM Tris-HCl, 5 mM imidazole, 1 mM PMSF, pH 8.0). After adding into the homogenizer and cracking for 20 min, the supernatant was collected by centrifugation at 20,000× g. The impurity particles in the supernatant were removed with 0.22 µm filter, and the protein was purified by AKT-Ni column and eluted by the elution buffer (100 mM NaCl, 20 mM Tris-HCl, 200 mM imidazole). The protein samples were detected using SDS-PAGE and Western Blot.

Alpha-Mangostin/USP45 Affinity Assay (BLI-OCTET K2 Assay)
The USP45 protein samples were concentrated via the ultrafiltration tube and centrifugation, followed by the protein sample transfer to the PBS buffers. Then, the USP45 protein solution and activated biotin (1:100) were incubated together at 4 • C for 4 h. The extra biotin was removed via the ultrafiltration tube and centrifugation, followed by the protein sample transfer to the PBST buffer (0.02% Tween20, pH = 7.4). The interaction between USP45 (50 µg/mL) and α-mangostin (20 µM) was analyzed using BLI-OCTET K2 (Sartorius, Gottingen, Germany).

Assessment of Tumorigenesis via Xenograft Model
All animals were cared for in accordance with the Animal Welfare Act guidelines under an animal protocol approved by Animal Care and Use Committee (Sichuan University, College of Life Science, Chengdu, China). Five-week-old female nude mice (DOSSY experiment animals company, Chengdu, China) were maintained with a regular 12 h light/dark cycle and raised on a standard rodent diet. The different number of cells (5 × 10 4 , 1 × 10 5 , 5 × 10 5 and 1 × 10 6 ) were resuspended in 100 µL of sanitary saline. Mice were randomly divided into 8 groups (n = 3 for each group), and xenograft-subcutaneously transplanted with cells. Tumor sizes were measured in half-week intervals. Tumor volume was calculated with the formula 0.5 × L × W 2 , where L and W represent the long and short axes of the tumor. The mice were euthanized at the end of the experiments.

The Combination of α-Mangostin and Doxorubicin Inhibits Tumorigenesis
Five-week-old female nude mice were divided into 4 groups (n = 3 for each group) and xenograft subcutaneously transplanted with 1.5 × 10 6 USP45-induced stem-like SiHa cells. Doxorubicin and α-mangostin were dissolved in corn oil (5% DMSO). The mice were fed via oral gavage every 3 days with corn oil, doxorubicin (5 mg/kg), α-mangostin (40 mg/kg) or their combination. Tumor sizes were measured in each half-week interval. At the end of 28 days, the mice were euthanized, tumors were obtained by skin cutting, and the tumor sizes were measured, weighed and photographed.

Statistical Analysis
GraphPad Prism 6.0 was used for data recording, collection, processing and calculation. All cell-based experiments were performed at least two times in duplicates. Data were presented as means ± SD, and quantitative data were analyzed statistically using Student's t-test to assess their significance.

USP45 Upregulates MYC Protein Level and Stability
The "TIMER2.0" database [49] was utilized for analyzing the USP45 expression level in various cancers, the result showed that the USP45 expression in partial cancer tissues was higher than normal ( Figure S1A). To predict the biological function of USP45 in cancer, Hallmark Gene Enrichment analysis analyzed the effects of USP45 associated with cancer signaling pathways. The results from the "GTBAdb" database [50] suggested that high USP45 expressing activated the cancer signaling, including MYC pathways ( Figure S1B,C). The survival analysis from the "GEPIA2" database [51] revealed that only cervical cancer patients with high USP45 expression exhibited poor overall survival (OS) and disease-free survival (DFS) ( Figure S2 and Figure S3). These bioinformatics analysis data revealed that USP45 might play a critical role in cervical cancer and activate MYC target signaling pathways. Therefore, we explored the regulatory mechanism of USP45-MYC in cervical cancer. The "GSE63514" dataset [52] showed that USP45 expression in the cancer tissues was elevated, but MYC was not significantly changed in cervical cancer tissues, compared with the normal tissues or stage CIN I-III ( Figure S4A,B). There was no correlation between the USP45 and MYC mRNA levels ( Figure S4C). Herein, we have hypothesized that USP45 may be a new deubiquitinase of MYC.
To investigate the hypothesis, USP45 and its mutant (C199A, the inactive single mutant) were overexpressed in the different cells. We found that USP45 overexpressing significantly upregulated MYC ( Figure 1A), and USP45 inhibition with the serval siRNAs significantly decreased MYC ( Figure 1B). The MYC decrease by USP45 inhibition with siRNA was reversed by overexpression of USP45 ( Figure 1C) or by the addition of the proteasome inhibitor MG132 ( Figure 1D). MYC regulation by USP45 was dependent on its DUB activity ( Figure 1E), and the USP45 resulted in MYC elevation in a dose-dependent manner ( Figure 1F), while its C199A mutant did not have any impact on the MYC level. We treated cells with protein synthesis inhibitor (cycloheximide, CHX) and found that the half-life of MYC was prolonged by USP45 overexpression (Figure 1G), or shortened by USP45 depletion ( Figure 1H). The decreases of MYC with USP45 siRNA were reversed by overexpression of USP45 or addi a proteasome inhibitor (MG132): (C) USP45-expressing plasmid was introduced into cells to with USP45 siRNA, followed by MYC level measurement with Western Blot, plvx-puro an siRNA as the negative control; (D) Cells transfected with USP45 siRNA were treated with or w MG132 (20 µM, 12 h), followed by MYC level measurement with Western Blot, GFP siRNA negative control. (E) USP45 overexpression enhanced MYC level, while its C199A mutant d Cells transfected with the control or USP45 (WT or the C199A mutant) were assayed by W Blot, plvx-puro vector as the negative control. (F) USP45 overexpression resulted in MYC ele in a dose-dependent manner, while its C199A mutant did not. USP45 or C199A with the inc amounts were transfected into cells, and MYC levels were detected, plvx-puro vector as the ne control. (G,H) USP45 enhanced the stability of MYC: (G) SiHa cells transfected with USP4 C199A were treated with cycloheximide (CHX, 50 µg/mL), and collected at the indicated tim Western Blot, plvx-puro vector as the negative control; (H) SiHa cells transfected with USP45 were treated with cycloheximide (CHX, 50 µg/mL) and collected at the indicated times for W Blot, GFP siRNA as the negative control. Experiments were conducted at least two times in cates.

USP45 Interacts with and Deubiquitinates MYC
Co-IP assays demonstrated that USP45 interacted directly with MYC, and the interaction was independent to the USP45 catalytic activity (Figure 2A-C). The deletion analysis demonstrated that the C-terminal domain of USP45 (190-814 domain) mediated its physical interaction with MYC ( Figure 2D). Ubiquitination assays demonstrated that USP45 overexpression directly reduced the ubiquitination of MYC ( Figure 2E). Conversely, when USP45 was inhibited with siRNA, the ubiquitination of MYC was increased ( Figure 2F), and the result was reversed via USP45 overexpression ( Figure 2G). Further, the C199A mutant lost the ability to deubiquitinate MYC ( Figure 2H). We discovered that the K 48 -link pattern of the ubiquitin chain on MYC was regulated by USP45, while the K 63 -link pattern was not ( Figure 2I).

USP45 Interacts with and Deubiquitinates MYC
Co-IP assays demonstrated that USP45 interacted directly with MYC, and the interaction was independent to the USP45 catalytic activity (Figure 2A-C). The deletion analysis demonstrated that the C-terminal domain of USP45 (190-814 domain) mediated its physical interaction with MYC ( Figure 2D). Ubiquitination assays demonstrated that USP45 overexpression directly reduced the ubiquitination of MYC (Figure 2E). Conversely, when USP45 was inhibited with siRNA, the ubiquitination of MYC was increased ( Figure 2F), and the result was reversed via USP45 overexpression ( Figure 2G). Further, the C199A mutant lost the ability to deubiquitinate MYC ( Figure 2H). We discovered that the K 48 -link pattern of the ubiquitin chain on MYC was regulated by USP45, while the K 63 -link pattern was not ( Figure 2I). (C) USP45 interaction with MYC was independent from its catalytic activity: HEK293T transfected with USP45-Flag (WT and C199A) were assayed by co-IP with anti-Flag, plvx-puro vector as the negative control. (D) The C-terminal domain of USP45 was required for binding to MYC: HEK293T cells transfected by MYC-HA together with control or USP45-Flag (WT or deletion mutants) were assayed by co-IP with anti-Flag, plvx-puro vector as the negative control. (E-H) USP45 deubiquitinating MYC was dependent on its catalytic activity: HEK293T cells, transfected by the indicated plasmids with USP45 (E), USP45 siRNA (F), USP45 and USP45 siRNA (G), or USP45 and the C199A mutant (H), were treated with MG132 for 6 h before harvesting, and the cells were collected for IP experiments, followed by analyzing the ubiquitination of MYC, plvx-puro vector or GFP siRNA as the negative control. (I) USP45 deubiquitinated MYC via specifically recognizing the K48 ubiquitin chain: HEK293T cells, transfected with MYC-HA, USP45-His and Ub-Flag (WT, K 0 , K 48 and K 63 ) were collected for the MYC ubiquitination analysis, plvx-puro vector as the negative control. Experiments were conducted at least two times in duplicates.

USP45 Promotes Cancer Proliferation, Stemness and Drug Resistance
We established the cervical cancer SiHa cells stably expressing USP45 or C199A. Western Blot assays showed that USP45, but not C199A, upregulated the MYC-targeting proteins (CDK1, CDK2 and CDK4), CSC proteins (Nanog, SOX2, OCT4 and BCL2) and EMT marker proteins (Vimentin and N-cadherin) ( Figure 3A). USP45 overexpression enhanced tumorsphere formation ( Figure 3B) and increased the population of CD133+ cells ( Figure 3C). In addition, the colony formation and proliferation of SiHa cells were significantly increased by USP45 overexpression ( Figure 3D,E). Interestingly, we found that USP45 promoted SiHa cells to efflux the Hoechst 33,342 ( Figure 3F,G). The cell viability and apoptosis analysis demonstrated that USP45 high expression enhanced drug resistance to doxorubicin ( Figure 3H,I). In summary, we demonstrated that USP45 overexpression promoted cancer proliferation, stemness, drug resistance, and transformation of SiHa cells into CSC-like cells.

MYC Is a Downstream Effector in the USP45-Induced Cancer Stemness and Drug Resistance
To investigate the causative roles of MYC in the proliferation, stemness and drug resistance of the USP45-induced CSC-like SiHa cells, we performed MYC deletion. In our results, knocking down MYC with shRNA reduced USP45-upregulated proteins ( Figure 4A), the tumorsphere number as well as CD133+ cell populations ( Figure 4B,C). Interestingly, the colony formation and proliferation of the cells were not changed by knocking down MYC ( Figure 4D,E). Further, knocking down MYC inhibited the Hoechst 33,342 efflux capacity of the cells ( Figure 4F,G). DOX resistance was significantly reduced by silencing MYC (Figure 4H,I). These results suggested that MYC was a critical downstream effector in USP45-induced cervical cancer stemness and drug resistance, but not proliferation.

Alpha-Mangostin Is an Inhibitor of USP45
We demonstrated that USP45 inhibition via destabilizing MYC suppressed the stemness and drug resistance of cervical cancer cells. As indirect MYC regulation by targeting USP45 can be used as a new strategy for cervical CSC treatments, we screened many USP45 inhibitors. Since the structure of USP45 has not been reported, we firstly used the I-TASSER program to simulate the structure ( Figure 5A). Through a series of virtual and experimental screenings, we identified a natural small molecule (α-mangostin, AMG) which interacted with USP45 specifically and tightly bound to the USP45 activity site ( Figure 5B-D). To investigate their interaction, USP45 was expressed and purified from E. coli ( Figure 5E). The purified USP45 was incubated with purified MYC from HEK293T cells, followed by the pulling down experiments, which demonstrated the direct interaction between USP45 and MYC ( Figure 5F). Further, our experimental binding study (BLI-OCTET K2 assay) with USP45 and AMG demonstrated their affinity with Kd of 19.3 µM ( Figure 5G). As expected, we observed its strong anticancer effect on cervical cancer cells ( Figure 5H) and its downregulation affecting the MYC protein levels without interfering with the mRNA levels ( Figure 5I). In summary, we found that the natural small molecule AMG is an USP45 inhibitor.

Alpha-Mangostin Inhibits the USP45-Induced Cancer Stemness, Drug Resistance and Tumorigenesis
To further verify the inhibition of USP45 functions with AMG, we performed the following experiments: Western Blot experiments demonstrated the AMG inhibition on the USP45-induced signaling pathways (MYC targeting, CSC and EMT signaling pathways; Figure 6A). With AMG, the tumorsphere formation, CD133+ cells population and Hoechst 33,342 efflux of USP45-induced CSC-like SiHa cells were significantly inhibited ( Figure 6B-E). Interestingly, the cancer cells were not significantly resistant to α-mangostin ( Figure 6F). The combination of AMG and DOX exhibited a synergistic effect on inhibiting cancer cell viability and proliferation and promoting apoptosis ( Figure 6G-I). Since the USP45-induced CSC-like SiHa cells have strong capacities in tumorigenesis ( Figure 7A) and drug resistance, their suppressions were investigated with AMG, DOX and their combination. Our results suggested that their combination had the strongest inhibition on tumorigenesis and drug resistance ( Figure 7B-E). In summary, AMG, as the inhibitor of USP45, suppressed cancer development, stemness and drug resistance ( Figure 7F).

Discussion
Our results are consistent with the literature's studies and demonstrates a new strategy for CSCs treatment. CSCs are intimately involved in cancer recurrence and metastasis, being responsible for the failure of anticancer therapies. It is of great importance to understand the molecular basis of CSC regulation in order to develop new

Discussion
Our results are consistent with the literature's studies and demonstrates a new strategy for CSCs treatment. CSCs are intimately involved in cancer recurrence and metastasis, being responsible for the failure of anticancer therapies. It is of great importance to understand the molecular basis of CSC regulation in order to develop new drugs and discover mechanisms of existing drugs for establishing better anticancer strategies.
The discovered direct relationship between USP45 and MYC can be used for suppressing MYC and cancer stemness. Since directly targeting MYC with small molecules and siRNA is difficult, reshaping the regulators of MYC signaling pathways to indirectly control MYC has become a new strategy for cancer treatments. Disrupting MYC-MAX heterodimerization suppresses MYC activity [53]. Further, indirect downregulation of MYC via ubiquitin-proteasome mechanisms is an excellent strategy for targeting CSCs, for instance activating SCF FBXW7 [54] and inhibiting deubiquitinase USP28 of MYC suppresses cell proliferation in diverse cancer [55]. The reported deubiquitinases of MYC include USP22 [42], USP28 [37], USP29 [39], USP36 [41] and USP37 [40], and they enhance cancer stemness via BMI1, LSD1 Snail and CEP63 stabilization, respectively. Although these MYC DUBs were reported previously, the direct connection among MYC, DUBs and CSCs was unknown. Clearly, our study established the direct connection between USP45, CSCs and MYC.
Further, we have discovered a USP45 inhibitor (α-mangostin) and a new mechanism for targeting USP45 and CSCs. Alpha-mangostin (AMG), isolated from the mangosteen, is an inhibitor of IDH1 R132H (isocitrate dehydrogenase-1 R132H mutant) with a Kd of 2.85 µM [56], and it was reported to interfere with all the major stages of carcinogenesis: initiation, promotion and progression [57]. AMG attenuates stemness and enhances cisplatin-induced stem-like cell death in cervical cancer [58]. However, the mechanisms of its activities remain unknown. Via our structural and functional studies, we have found that AMG can directly inhibit USP45 (with Kd of 19.3 µM), causing MYC destabilization and, in turn, suppressing the stemness and drug resistance of cervical CSCs. This small molecule may be a promising adjuvant drug for clinical CSC treatments.

Conclusions
On the basis of our investigations, we have discovered that USP45 can directly interact with MYC and is a new deubiquitinase of MYC. Further, we have discovered that increased USP45 expression significantly promoted cervical cancer cell proliferation, stemness and drug resistance, while USP45-enhanced MYC stability facilitated only stemness and drug resistance. These findings have established the close connection among USP45, MYC and CSCs. Interestingly, decreased USP45 activity by AMG inhibition significantly suppressed cancer cell proliferation, stemness and drug resistance. In general, on the basis of our USP45 discoveries on its MYC deubiquitination and α-mangostin inhibition, suppressing USP45 has opened a new window for suppressing cancer development, stemness and drug resistance.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/cancers15030930/s1, Figure S1: USP45 pan-cancer expression and function analysis, Figure S2: Overall survival of cancer patients was stratified by USP45 expression levels, Figure S3: Disease-free survival of cancer patients was stratified by USP45 expression levels, Figure S4: The "GSE63514" dataset was used to analyze clinical correlation between USP45 and MYC expression levels, Figure S5