Suppression of GCH1 Sensitizes Ovarian Cancer and Breast Cancer to PARP Inhibitor

Background Breast and ovarian cancers are common malignancies among women, contributing to a significant disease burden, and are characterized by a high level of genomic instability, owing to the failure of homologous recombination repair (HRR). Pharmacological inhibition of poly(ADP-ribose) polymerase (PARP) could elicit the synthetic lethal effect of tumor cells in patients with homologous recombination deficiency, ultimately achieving a favorable clinical benefit. However, primary and acquired resistance remain the greatest hurdle, limiting the efficacy of PARP inhibitors; thus, strategies conferring or augmenting tumor cell sensitivity to PARP inhibitors are urgently required. Methods Our RNA-seq data of niraparib-treated and -untreated tumor cells were analyzed by R language. Gene Set Enrichment Analysis (GSEA) was applied to assess the biological functions of GTP cyclohydrolase 1 (GCH1). Quantitative real-time PCR, Western blotting, and immunofluorescence were applied to confirm the upregulation of GCH1 upon niraparib treatment at transcriptional and translational levels. Immunohistochemistry of patient-derived xenograft (PDX)-derived tissue sections further validated that niraparib increased GCH1 expression. Tumor cell apoptosis was detected by flow cytometry, while the superiority of the combination strategy was confirmed in the PDX model. Results The expression of GCH1 was aberrantly enriched in breast and ovarian cancers and increased after niraparib treatment via JAK-STAT signaling. GCH1 was also demonstrated to be associated with the HRR pathway. Subsequently, the enhancement of the tumor-killing effect of PARP inhibitors induced by GCH1 suppression using siRNA and GCH1 inhibitor was validated by flow cytometry in vitro. Finally, using the PDX model, we further demonstrated that GCH1 inhibitors markedly potentiated PARP inhibitors' antitumor efficacy in vivo. Conclusion Our results illustrated that PARP inhibitors promote GCH1 expression via the JAK-STAT pathway. We also elucidated the potential relationship between GCH1 and the homologous recombination repair pathway and proposed a combination regimen of GCH1 suppression with PARP inhibitors in breast and ovarian cancers.


Introduction
Ovarian and breast cancers, regarded as the two major malignancies threatening women's health [1], have limited therapeutic options other than surgical resection combined with adjuvant chemotherapy and face more daunting challenges that the majority of patients will experience chemotherapy resistance and relapse [2,3]. With more targeted therapies springing up, the management of ovarian and breast cancer has been revolutionized and stepped into a new era.
Poly(ADP-ribose) polymerase (PARP), frst known for its catalytic enzyme activity in the process of poly(ADPribosyl)ation, was subsequently verifed to play crucial roles in a series of cellular processes such as timely and accurate DNA repair and cell death [4,5]. Pharmacological inhibitors of PARP have shown promising therapeutic potential in various cancers. Results from a phase 3 trial PRIMA/ ENGOT-OV26/GOG-3012 indicated a statistically signifcant extension of progression-free survival (PFS) in patients who received niraparib compared with placebo among newly diagnosed advanced ovarian cancer patients [6]. Olaparib, the frst PARP inhibitor, was demonstrated to provide a substantial PFS beneft for patients with BRCA1/2 mutations and HER2-negative breast cancer, reducing the risk of disease progression or death by 42% in comparison with standard therapy [7]. As of 2020, PARP inhibitors have gained approval in ovarian cancer, breast cancer, prostate cancer, and pancreatic adenocarcinoma [6][7][8][9], suggesting that PARP inhibitors might have the ability to shift the treatment paradigm for these cancers. However, the population sensitive to PARP inhibitors is mostly limited to patients with homologous recombination defciency (HRD) due to their synthetic lethal efects. Terefore, an improved understanding of the biology of cancer will facilitate the development of rational strategies to increase the sensitivity to PARP inhibitors and expand the applicable population.
Whilst PARP inhibitors initially present favorable tumor-killing efects, most patients would experience acquired drug resistance over time, triggering tumor progression. Resistance to PARP inhibitors is now recognized to be acquired primarily through the following mechanisms: restoration of homologous recombination function, stability and protection of replication forks, enhancement of drug efux transporters, and PARP-related mechanisms [10]. In response to these possible resistance mechanisms, many studies have focused on combination regimens in an attempt to combat the emergence of acquired resistance with the aim of amplifying the antitumor efects of PARP inhibitors. PI3K-AKT pathway inhibitors, which have been reported to impair homologous recombination repair (HRR), render the tumor cells more sensitive to PARP inhibitors [11][12][13][14]. Suppression of the NAD + salvage pathway using the NAMPT inhibitor FK866 led to inhibition of PARylation, resulting in potent enhancement of olaparib cytotoxicity in triple negative breast cancer [15]. Furthermore, the feld of immune checkpoints has been remarkably vigorous in the last decades, and its combination with PARP inhibitors is also an area of active investigation. Te benefts of PARP inhibitors in combination with anti-PD-1/anti-PD-L1 can manifest themselves via increased objective response rates in patients [16,17], probably related to the activation of cGAS-STING signaling and increased expression of PD-L1 [18,19]. Indeed, metabolic reprogramming is also closely associated with the sensitivity of PARP inhibitors. Suppression of an important glycolytic enzyme, phosphoglycerate mutase 1 (PGAM1), sensitized BRCA1/2-profcient breast cancer to PARP inhibitors by impairing homologous recombination repair [20]. Tese preclinical and clinical data given above have sparked a rapidly growing interest in the combination strategies of PARP inhibitors.
Moreover, we clarifed that the suppression of GCH1 could sensitize tumor cells to the PARP inhibitor, in order to better ameliorate clinical management of ovarian and breast cancers.

Materials and Methods
2.1. RNA Extraction and Library Construction. SKOV3 cells (human ovarian cancer cell line obtained from American Type Culture Collection, ATCC) were seeded in T75 culture fask at a suitable density and cultured overnight to adhere. Cells were treated with 10 μM niraparib (ZL-2306, Zai Lab, China) or 0.2% DMSO for 72 hours prior to collection. Each group contained three parallel replicates. Te total RNA was quantifed using a Nano Drop and Agilent 2100 bioanalyzer (Termo Fisher Scientifc, MA, USA) after being extracted using Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Purifed mRNA was used for cDNA synthesis, followed by the construction of the fnal library which was further sequenced on the MGISEQ 2000 platform (BGI Corporation, Shenzhen, China).

Cell Culture and Transfection.
Human breast cancer cell line MDA-MB-231 and HCC38 and ovarian cancer cell line A2780 and OVCAR8 were obtained from ATCC and cultured in DMEM or RPMI 1640 medium containing 10% FBS and 1% penicillin-streptomycin solution (G4003, Servicebio, China). GCH1 siRNAs and control siRNA were purchased from RiboBio Co. Ltd. (Guangzhou, China). Te transfection of siRNA was performed according to the manufacturer's instruction and knockdown efcacy was examined by real-time PCR after 48 hours.

Isolation of RNA and Quantitative Real-Time PCR.
Te total RNA was extracted using the FastPure Cell/Tissue Total RNA Isolation Kit (RC101, Vazyme), while cDNA was further synthesized using HiScript II Q RT SuperMix for qPCR (R223-01, Vazyme). Te ChamQ Universal SYBR qPCR Master Mix (Q711-02, Vazyme) was applied to quantitative real-time PCR. All primers used were as follows: GCH1-F GCCATGCAGTTCTTCACCAAGG GCH1-R ATGGAACCAAGTGATGCTCACAC GAPDH-FTGCACCACCAACTGCTTAGC GAPDH-RGGCATGGACTGTGGTCATGAG 2.5. Western Blotting. For in vitro cell line samples, optimized density of tumor cells was seeded prior to treatment. After being treated for the indicated time and concentration, cells were rinsed with ice-cold PBS and lysed in RIPA bufer with proteases and a phosphatase inhibitor (78440, Termo Fisher Scientifc). Denatured cell extracts were subjected to SDS-PAGE gel electrophoresis before being transferred to PVDF membranes. Blots were blocked and incubated with anti-GCH1 antibodies (ab236387, Abcam) or anti-GAPDH (GB11002, Servicebio, China) for 12-16 hours at 4°C, followed by incubation with appropriate horseradish peroxidase-conjugated antibodies at room temperature for 1 hour. Immunoblotting was detected by the ECL chemiluminescence system (Bio-Rad).

Immunohistochemical Staining of Tissue Sections.
After being fxed with formalin and embedded with parafn, tumor tissues were sliced into slides. Parafn sections were heated at 65-68°C for 2 hours before being deparafnized in a dewaxing agent and rehydrated sequentially. Antigen retrieval was performed using water bath heating and endogenous peroxidase was consumed in 3% H 2 O 2 . After blocking with 5% BSA for 30 min at 37°C, sections were incubated with primary antibodies against GCH1 (ab236387, Abcam) overnight at 4°C and secondary antibodies at room temperature for 1 hour. DAB was applied for color development. Sections were counterstained with hematoxylin, dehydrated by graded ethanol, and coverslipped before being visualized with an Olympus microscope. Te IHC score was assessed based on staining intensity score (0, no staining; 1, weak staining; 2, moderate staining; and 3, intense staining) and the proportion score (1, 0-25% of the tumor cells were stained; 2, 26-50%; 3, 51-75%; and 4, more than 75% of the tumor cells were stained). IHC score-= staining intensity score × proportion score.

Data Source and Preparation.
RNA expression profles and clinical information from breast invasive carcinoma (BRCA) project (include 292 normal and 1,099 cancerous tissues, and a total of 112 paired normal and tumor tissues) and ovarian serous cystadenocarcinoma (OV) project (include 88 normal and 427 cancerous tissues) were obtained from the Cancer Genome Atlas (TCGA) database. All raw RNA-seq data were transformed from FPKM format into transcripts per million reads (TPM) format for further studies.

Function
Analysis of GCH1-Associated DEGs. As for the identifcation of DEGs between high-and low-GCH1 expression groups, the threshold was determined according to the parameters, adjusted P value < 0.05, and the absolute log-fold change larger than 1.5. Te identifed DEGs were then processed for the Gene Set Enrichment Analysis (GSEA) using the R package clusterProfler (3.8.0) and the thresholds were set as adjusted P value <0.05 and FDR < 0.25. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were conducted to explore the potential function of GCH1associated DEGs. Te protein-protein interaction (PPI) network was constructed according to the online STRING database (https://string-db.org/). Te CHIP-seq data were collected from Cistrome Data Browser (http://cistrome. org/db).

Clinical Signifcance of GCH1.
Te relationship between the expression level of GCH1 and clinicopathological characteristics was assessed with the Wilcoxon signed-rank sum test and logistic regression. All the statistical analyses given above were performed in R (v3.6.3), and P values <0.05 were considered signifcant.

Colony Formation
Assay. 500 MDA-MB-231 or A2780 cells were seeded in each well of 6-well plates and allowed to adhere for 4 hours prior to treating with 0.2% DMSO, 5 μM DAHP, 10 μM niraparib, or a combination for 5-10 days. Cells were then fxed for 30 min and stained with crystal violet for 1 hour and counted.
2.13. Cell Viability Assay. Cells were seeded in a 96-well plate and treated with 10 μM niraparib for indicated times and wells were replaced with fresh medium containing CCK8 (A311-01, Vazyme). Te cell viability was measured according to the protocols.

Establishment of the Patient-Derived Xenograft (PDX)
Model. Fresh tumor tissue obtained from surgery was preserved in saline solution and immediately transported to the laboratory on ice within 4 hours. Te tumor tissues were trimmed and cut into small fragments. 6-8-week-old female NOD-Prkdc scid Il2rg em1 /Smoc micewere preanesthetized with 1%-1.5% isofurane (R510-22-10, RWD Life Cycle) and the tissue pieces were implanted subcutaneously on the dorsum of the mice to produce the frst generation xenografts. Te size of the tumor was measured every 2 days thereafter. When the maximum tumor diameter was about 15-20 mm, the micewere euthanized and the tumor tissue was transplanted into new NOD-Prkdc scid Il2rg em1 /Smoc mice as the next generation models.

In Vivo Animal Studies.
Te animal study was approved by the Animal Care and Use Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. Twenty 6-8 weeks-old female NOD-Prkdc scid Il2rg em1 /Smoc mice were anesthetized with 1%-1.5% isofurane (R510-22-10, RWD Life science, Shenzhen, China). Subsequently, patient-derived xenografts (PDXs) were engrafted subcutaneously on the dorsum. After inoculation, mice weights and tumor sizes were measured twice weekly. When the tumors reached 50-100 mm 3 , mice were randomized into four subgroups. Niraparib was dosed (20 mg/kg, intragastric administration) once daily for 5 days per week while DAHP was dosed (60 mg/ml, intraperitoneal administration) once daily. Tumor volume based on caliper measurements will be calculated using the modifed ellipsoidal formula: tumor volume � 1/2 length × width 2 . At experimental endpoints, the mice will be sacrifced after anesthesia. Tissue samples were taken for routine hematoxylin and eosin (HE) staining and immunohistochemistry.

Statistical
Analysis. GraphPad Prism 9 was employed in statistical analysis to graph our data. Student's t-test was used for comparison between two groups, and one-way ANOVA analysis was applied to the comparison of more than two groups. Time course data were analyzed by twoway ANOVA. All results are represented as the mean ± SD. P < 0.05 was considered statistically signifcant and denoted as follows: * P < 0.05, * * P < 0.01, and * * * P < 0.001.

PARP Inhibitor Promotes the Transcription and Translation of GCH1 in Breast and Ovarian Cancers.
To clarify the alterations of tumor cells after PAPR inhibitors administration, providing clues to the mechanism of drug resistance, niraparib-treated and -untreated SKOV3 cells were subjected to RNA-seq and the DEGs were identifed by Wilcoxon rank-sum test (Supplementary Table 1). We matched these DEGs with a list of 2,752 MRGs that encoded almost all known human metabolic enzymes and transporters. Considering the cutof criteria (adjust P value < 0.05 and log|FC| > 1.0), a total of 276 diferentially expressed MRGs containing 172 upregulated and 104 downregulated genes were extracted (Figure 1(a)). Representative DEGs were also illustrated in volcano plot (Figure 1(b)). Among the upregulated DEGs in the niraparib-treated group, SPHK1 was the most signifcantly changed gene, but the diference in expression between the tumor and the corresponding normal tissue was inconspicuous in breast and ovarian cancers according to the TCGA and GTEx databases ( Supplementary Figures 1(a) and 1(b)). So, we focused on GCH1, which had a relatively signifcant diference.
Consistent with the RNA-seq results, GCH1 mRNA levels were increased in multiple ovarian and breast cancer cell lines in response to niraparib treatment (Figure 2(a)). Te niraparib-induced additional expression of GCH1 was validated by immunoblotting ( Figure 2(b)). Likewise, immunofuorescence results confrmed that GCH1 was enriched in the presence of niraparib (Figure 2(c)). We further performed histochemical staining of clinical specimens derived from PDX models of breast and ovarian cancer and similarly verifed that GCH1 expression was substantially higher in the niraparib-treated group compared with the control group ( Figure 2(d)). In summary, the transcription and translation of GCH1 was remarkably elevated upon PARP inhibitor treatment.

Clinical Implications of GCH1 in Breast and Ovarian
Cancers. Given the limited research on the expression and function of GCH1 in breast and ovarian cancers, we initially explored GCH1 expression and its correlation with clinicopathological characteristics. Based on the TCGA and GTEx databases, we analyzed the expression of GCH1 mRNA. Te results demonstrated that GCH1 expression was much higher in breast and ovarian cancers ( Furthermore, logistic regression analysis revealed the statistically signifcant association between GCH1 high expression and histological type, PR status, ER status, and HER2 status in breast cancer (Supplementary Table 4). A prominent correlation of GCH1 expression with Federation International of Gynecology and Obstetrics (FIGO) stage was observed in ovarian cancer as well (Supplementary Table 5).

PARP Inhibitor Induces the Accumulation of GCH1
through the JAK-STAT Pathway. To further elucidate the potential mechanisms of GCH1 regulation, all the cases were divided into two groups and compared based on the median GCH1 expression in tumors. According to the threshold of | log2(FC)| > 1.5 and adjusted P value < 0.05, a total of 778 genes, covering 227 upregulated genes and 551 downregulated genes, were recognized as DEGs in breast cancer. While in ovarian cancer, 324 genes (including 252 increased and 72 decreased genes) were identifed as statistically signifcant between the two cohorts. Te top 50 genes positively correlated with GCH1 were illustrated by heatmaps (Figures 4(a) and 4(b)), and the top 50 DEGs negatively associated with GCH1 were shown in Supplementary Figures 2(a) and 2(b). Surprisingly, STAT1 ranked frst in both breast cancer and ovarian cancer, implying that there might be a strong correlation between STAT1 and GCH1. Te Gene Set Enrichment Analysis (GSEA) of all DEGs also showed signifcant enrichment of the JAK-STAT signaling pathway (Figures 4(c) and 4(d)). Further, the correlations of GCH1 and STAT1 were directly assessed. Given the Spearman correlation coefcients were 0.53 and 0.58, GCH1 and STAT1 were well-correlated in breast cancer (Figure 4(e)) and ovarian cancer (Figure 4(f )). Finally, we obtained CHIP-seq data from the Cistrome Data Browser showing that STAT1 could bind to the promoter region of GCH1 to play a regulatory role (Figure 4(g)). To validate the involvement of the JAK-STAT pathway in the regulation of GCH1, MDA-MB-231, A2780, and HCC38 cancer cell lines were subjected to JAK-STAT inhibitor ruxolitinib or/and niraparib. Consistent with the abovementioned fndings, the JAK-STAT inhibitor alone exhibited a modest suppressive impact on GCH1 expression. In addition, niraparib has been reported to activate the JAK-STAT signaling pathway [29]. Our results revealed that inactivation of JAK-STATsignaling resulted in the rollback of niraparib-induced upregulation of GCH1, indicating that niraparib might modulate GCH1 via the JAK-STAT pathway (Figures 4(h)-4(k)).

GCH1 Correlates with the Homologous Recombination
Repair (HRR) Pathway. We observed that the core molecules of HRR cluster were mainly enriched in the GCH1high expression group (P � 0.04) in breast cancer ( Figure 5(a)). Although, there was no statistical signifcance between the two groups in ovarian cancer (P � 0.05), the overall enrichment trend was similar to that of breast cancer ( Figure 5(b)). Further, we analyzed the association between GCH1 and 28 genes highly related to the HRR pathway (Figures 5(c) and 5(d)) and Spearman's correlation coefcients greater than 0.3 were considered to indicate a certain correlation. In breast cancer, GCH1 was signifcantly related to BRCA2, FANCA, FANCI, CDK12, FANCD2, BRIP1, and CHEK1 ( Figure 5

Inhibition of GCH1 Potentiates the Efcacy of PARP Inhibitor Terapy in Vivo.
On the basis of the beneft achieved by niraparib and the inhibition of GCH1 in vitro, we sought to explore the efcacy using the ovarian cancer PDX model in vivo. Tumors established in NOG mice previously were

Discussion
PARP inhibitors have been approved by the FDA as maintenance therapy for ovarian cancer and breast cancer, and the major obstacle to their clinical application is the inevitable occurrence of acquired drug resistance. An improved understanding of the biological changes of PARP inhibitor-treated tumors would facilitate the design of rational treatment strategies as powerful tools against tumor, postponing the onset of resistance and delaying disease progression. In this context, we screened diferentially expressed metabolic enzyme and transporter genes between the niraparib-treated group and the control group and observed signifcant upregulation of GCH1 expression. Our data demonstrate an increase in GCH1 induced by niraparib at both mRNA and protein levels in cancer cell lines, which was further validated using PDX-derived tissue sections. Moreover, JAK-STAT signaling mediated the upregulation of GCH1 in response to PARP inhibitors. Finally, the relationship of GCH1 to homologous recombination repair was revealed, and we proved the superiority of the combination of PARP inhibitors and GCH1 inhibition in tumor killing in vitro and in vivo. In summary, GCH1 may serve as a promising target to sensitize tumor cells to PARP inhibitors and enhance the antitumor efcacy. GCH1 is the major rate-limiting enzyme for BH4, an essential cofactor for the production of NOSs [37]. Te GCH1/BH4 metabolic pathway participates in the generation of reactive oxygen species and is important in many abnormal disease statuses including diabetes, hypertension, and Parkinson's disease [38][39][40][41][42]. GCH1 protects cells from ferroptosis through the synthesis of BH4/BH2 leading to lipid remodeling in a GPX4-independent manner [43]. Te inhibition of GCH1 was proved to be an efective way to manage cancer pain [44]. However, research on the expression and biological functions of GCH1 in cancer is still limited.
Tus, we frst evaluated the expression level of GCH1 and found that GCH1 was signifcantly elevated in breast invasive carcinoma and ovarian cancer. Consistent with our results, GCH1 was also upregulated in glioblastoma [24], indicating its potential hub role in tumorigenesis. We also explored clinical implications of GCH1 and found that higher GCH1 expression was proved to be correlated with several clinicopathological traits.
Despite its clinical value, data on the potential functions and regulating mechanisms involving GCH1 in tumors remain limited. A previous study has reported cytokinestimulated induction of GCH1 in endothelial cells that TNFα could upregulate GCH1 expression via the activation of IKK/NF-ĸB pathway, while IFN-c had a similar efect attributed to the JAK2/STAT pathway [45]. In this setting, functional annotation of GCH1-associated DEGs was performed. Te JAK-STAT signaling pathway, especially transcription factor STAT1, was verifed to be positively correlated with GCH1-high expression, highlighting the underlying role of STAT1 in the regulation of GCH1 in breast and ovarian cancers. Further, we observed that administration of niraparib boosted GCH1 expression, which    was reversed by inhibition of the JAK-STAT signaling. In spite of the fact that suppression of the JAK-STAT pathway led to a signifcant decrease of GCH1 expression in MDA-MB-231 cells, it was not evident in A2780 and HCC38 cell lines. Tis may be due to the inconsistencies in the basal activation levels of the JAK-STAT pathway in diferent cell lines. Based on the above results, we concluded that niraparib might elevate GCH1 expression via the activation of JAK-STAT pathway.
In the past decades, there has been a great increase in the number of preclinical studies focusing on approaches that could strengthen the PARP inhibitors' cytotoxicity. Te loss of the PAR-dependent nucleosome sliding enzyme ALC1/ CHD1L conferred up to 250-fold PARP inhibitor hypersensitivity on HR-defcient tumor cells [46]. High expression of CXorf67, a DNA damage response protein that competes with BRCA2 for PALB2 interaction to impair the HR pathway, could increase sensitivity to PARP inhibitors in PFA ependymomas [47]. Te autophagy inhibitor chloroquine triggered the deleterious NHEJ DNA repair process and synergistically augmented the anticancer efect of talazoparib in HR-profcient breast cancer [48]. Besides, in HR-profcient ovarian cancer, EZH2 inhibition could promote sensitivity to PARP inhibitors via upregulation of MAD2L2 in a CARM1-dependent manner [49]. Coincidentally, our results revealed that GCH1, which was remarkably upregulated upon niraparib treatment, might be involved in HRR, leading to the resistance to PARP inhibitors. Tese data provided proof of principle that the synergistic combination of GCH1 suppression and PARP inhibitors was superior, and the superiority was further demonstrated in vitro. Additionally, we have demonstrated in vivo using the PDX model that targeting GCH1 could potentiate the sensitivity of tumor cells to PARP inhibitors and markedly inhibit tumor growth without producing excessive toxic side efects. It is plausible to speculate that certain combinational strategies might provide breast and ovarian cancer patients with more benefts and improve their prognosis. Although we propose a rational therapeutic strategy to render niraparib more efective in killing tumor cells that can be further tested in the clinic, whether GCH1 could be a biomarker for sensitivity to PARP inhibitor treatment still warrants further investigations.
In spite of the merits of the current study, limitations should be addressed. First, part of the data included in the current study on assessing the biological functions of GCH1 was derived from the public TCGA database. Cohorts and studies from other centers are required to validate the fndings. Second, the synergistic efects in vivo of GCH1 inhibitors with PARP inhibitors in breast cancer require more detailed exploration due to the limitations of the PDX model.

Conclusions
In conclusion, we clarifed that niraparib induced the upregulation of GCH1, which was associated with sensitivity to PARP inhibitors in ovarian cancer and breast cancer for the frst time. Tis study provides new perspectives for enhancing the potency of PARP inhibitors and ofers novel strategies for clinical management of breast and ovarian cancer patients.

Data Availability
Te datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ethical Approval
Te studies involving human participants (Approval number: TJH2020S122-5) were approved by the Medical Ethics Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. Te animal study (Approval number: TJH-202111003) was approved by the Animal Care and Use Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology.
Data are presented as the mean ± SD. All experiments were performed at least three times. Statistical signifcance was determined by one-way ANOVA. * P < 0.05, * * P < 0.01, and * * * P < 0.001. n.s. Nonsignifcant Table 3. Te association between GCH1 expression and clinicopathological variables in ovarian cancer. Supplementary Table 4. Logistic regression analysis of association between clinicopathological characteristics and GCH1 expression in breast cancer patients. Supplementary Table 5. Logistic regression analysis of association between clinicopathological characteristics and GCH1 expression in ovarian cancer patients. (Supplementary Materials)