Abstract
目的
基于生物信息学分析探究吡咯啉-5-羧酸还原酶-1（PYCR1）作为泛癌生物标志物的潜力，并探究其在膀胱癌（BLCA）中的表达、功能及临床意义。
方法
通过生物信息学分析PYCR1与泛癌患者预后、免疫微环境重塑、肿瘤突变负荷及微卫星不稳定性的关联。在TCGA-BLCA数据集中通过单因素和多因素回归分析PYCR1作为BLCA患者独立预后风险因素的潜力，并构建临床决策模型。利用IMvigor210中的BLCA队列鉴定PYCR1作为免疫治疗效果评估独立因子的潜力。基于pRRophetic药物库筛选PYCR1高表达时BLCA治疗耐受的潜在化疗药物。CMap-XSum算法和分子对接技术用于筛选并验证小分子PYCR1抑制剂。
结果
PYCR1高表达与多种肿瘤不良预后、免疫细胞浸润、肿瘤突变负荷及微卫星不稳定性显著相关（r>0.3）。PYCR1在BLCA中过表达，PYCR1高表达与BLCA预后差密切相关（HR：1.14，95% CI： 1.02-1.68，P=0.006）。PYCR1高表达时西妥昔单抗、5-氟尿嘧啶、多柔比星等抗肿瘤药物IC50提高（P<0.0001）。
结论
PYCR1是癌症潜在的预后生物标志物和治疗靶点，PYCR1高表达是BLCA患者不良预后的独立危险因素，其具有良好的临床决策能力，是预测化疗药物敏感性和免疫治疗效果的指标。
Keywords: 膀胱癌, 吡咯啉-5-羧酸还原酶-1, 免疫治疗, 化疗药物敏感性, 小分子PYCR1抑制剂
Abstract
Objective
To explore the potential of pyrroline-5-carboxylate reductase 1 (PYCR1) as a pan-cancer biomarker and investigate its expression, function, and clinical significance in bladder cancer (BLCA).
Methods
Bioinformatics analysis was conducted to evaluate the associations of PYCR1 with prognosis, immune microenvironment remodeling, tumor mutation burden (TMB), and microsatellite instability (MSI) in cancer patients. Using the TCGA-BLCA dataset, univariate and multivariate regression analyses were performed to assess the potential of PYCR1 as an independent prognostic risk factor for BLCA, and a clinical decision model was constructed. The IMvigor210 cohort was utilized to evaluate the potential of PYCR1 for independently predicting the efficacy of immunotherapy. The pRRophetic was employed to screen candidate chemotherapeutic agents for treating BLCA with high PYCR1 expression. The CMap-XSum algorithm and molecular docking techniques were used to explore and validate small molecule inhibitors of PYCR1.
Results
A high expression of PYCR1 was significantly associated with poor prognosis, immune cell infiltration, TMB and MSI in various tumors (r>0.3). PYCR1 was overexpressed in BLCA, and high PYCR1 expression was closely related to poor prognosis in BLCA patients (HR: 1.14, 95% CI: 1.02-1.68, P=0.006). The IC50 of the anti-cancer drugs cetuximab, 5-fluorouracil, and doxorubicin increased significantly in BLCA cell lines with high PYCR1 expressions (P<0.0001).
Conclusion
High PYCR1 expression is an independent risk factor for poor prognosis in BLCA patients and can serve as a significant indicator for clinical decision-making as well as a marker for predicting sensitivity to chemotherapeutic agents and the efficacy of immunotherapy.
Keywords: bladder cancer, pyrroline-5-carboxylate reductase 1, immunotherapy, chemotherapy drug sensitivity, small molecule PYCR1 inhibitors
脯氨酸代谢与癌症的发生和发展密切相关，吡咯啉-5-羧酸还原酶-1（PYCR1）是脯氨酸生物合成中的最后一种酶，其过表达与癌细胞的生长、侵袭以及对化疗药物耐受的产生有关［1］。PYCR1能够诱导癌症干细胞代谢物循环改变和扩增及促进肿瘤相关成纤维细胞（CAFs）在癌症基质中沉积，进而促进肿瘤免疫微环境重塑（TIME）［2， 3］。调节TIME重塑可逆转肿瘤对顺铂、吉西他滨等化疗药物的耐受［4］。研究表明，抑制PYCR1可通过影响TIME重塑逆转肿瘤免疫治疗和化疗的耐药性，PYCR1具有作为肿瘤治疗靶点的潜力［5］。然而，PYCR1表达水平与泛癌中肿瘤患者预后生存、肿瘤免疫浸润、肿瘤突变负荷（TMB）、微卫星不稳定性（MSI）之间的联系及其在肿瘤中调控的分子功能尚不清楚。
以顺铂和吉西他滨为基础的联合化疗是改善晚期膀胱癌（BLCA）患者预后的重要手段。然而，耐药性的产生常导致BLCA的复发和转移［6， 7］。TIME重塑诱导的免疫治疗和化疗耐受导致BLCA复发和转移仍是临床治疗的难题［8， 9］。最新研究显示，抑制PYCR1能够通过影响TIME重塑来增强肾细胞癌和肌成纤维细胞癌对化疗药物的敏感性［10， 11］。此外，PYCR1在BLCA中显示出作为生物标志物的潜力［12］。然而，PYCR1是否作为BLCA的独立风险因素促进肿瘤进展尚不清楚。在本文研究了泛癌中PYCR1表达与患者预后、TIME重塑、TMB和MSI的关联，并基于大数据分析并鉴定了PYCR1作为BLCA的独立预后风险因子的潜力。建立PYCR1评分对BLCA免疫治疗效果及化疗药物敏感性进行预测，为BLCA患者提供新型免疫治疗和化疗策略。随着高通量技术的发展，药物转录组学研究的规模激增，XSum和CMap是最先进的肿瘤学特征搜索方法［13］。小分子药物的开发已经展现出广泛的潜力，为肿瘤的有效治疗提供了新的方向［14， 15］。本研究拟通过基于CMap大型药物集和XSum算法发掘一种新型的PYCR1小分子抑制剂通过抑制PYCR1来治疗BLCA。
1. 材料和方法
1.1. 材料
1.1.1. 数据来源
泛癌及BLCA临床数据收集自TCGA数据库（https：//portal.gdc.cancer.gov）和基因表达总库（Gene Expression Omnibus， GEO，http：//www.ncbi.nlm.nih.gov/geo/），BLCA队列为TCGA-BLCA、GSE3167、GSE37817、GSE166716和GSE13507。免疫治疗组收集自泌尿系统肿瘤IMvigor210队列及泛癌GSE78220、GSE67501队列。化疗药物集源自Paul Geeleher等人的研究［16］，BLCA相关免疫组化图谱和苏木精-伊红染色（Hematoxylin-Eosin staining， HE）图谱取自HPA数据库（https：//www.proteinatlas.org/）。小分子抑制剂和药物转录组数据集来自连接图谱（Connectivity Map， CMap，https：//clue.io/about），小分子抑制剂结构来自PubChem数据库（https：//pubchem.ncbi.nlm.nih.gov/）。通路和功能富集注释基因集来自分子特征数据库（Molecular Signatures Database， MSig DB，https：//www.gsea-msigdb.org/gsea/msigdb）。蛋白受体PYCR1人源晶体结构来自RCSB PDB数据库（https：//www.rcsb.org/）。
1.1.2. 细胞与试剂
人BLCA细胞系（T24、5637、UM-UC-3）和人输尿管上皮永生化细胞SV-HUC-1（中科院上海细胞库）；胎牛血清、RPMI 1640培养基、DMEM培养基、F-12 K培养基（Gibco）；PYCR1-siRNA1、siRNA2和阴性对照（NC）由苏州吉玛基因股份有限公司设计和合成；Lipofectamine RNAiMax（Invitrogen）；RNAiso Plus、PrimeScript RT Master Mix、SYBR Premix Ex TaqTM II kit（宝生物工程）；CCK-8工作液购自同仁化学研究所。
1.1.3. 仪器
QuantstudioTM DX实时荧光定量PCR仪（Applied Biosystems）；多功能酶标仪（Multiscan MK3， Thermo Fisher Scientific）。
1.2. 方法
1.2.1. 基于泛癌研究数据挖掘PYCR1对肿瘤患者的预后意义
基于获取的泛癌临床数据和“limma”、“GSEABase”和“GSVA”程序包对PYCR1 mRNA表达水平进行分析，“survival”程序包用于分析PYCR1表达水平与癌症患者的无疾病生存期（DFS）、总生存期（OS）、无进展生存期（PFS）和疾病特异性生存期（DSS）的关系，绘制Kaplan-Meier生存曲线。基于“org.Hs.eg.db”、“clusterProfiler”程序包进行GSEA信号通路顶端富集和底部富集。
1.2.2. 构建集成机器学习网络探究PYCR1对TIME重塑和突变景观的影响
使用R软件搭载免疫算法CIBERSORT-ABS和CIBERSORT构建集成机器学习网络对PYCR1和免疫细胞浸润水平、基质评分、免疫评分、ESTIMATE评分和肿瘤纯度进行关联性模拟，其中ESTIMATE评分=基质评分+免疫评分，与肿瘤纯度成反比［17］。基于MSI和TMB评分对PYCR1在泛癌中的突变景观进行解析，并使用“fmsb”程序包绘制雷达图。
1.2.3. 揭示PYCR1预测癌症免疫治疗响应的效力
基于IMvigor210、GSE78220和GSE67501免疫治疗队列，使用“limma”程序包根据PYCR1 mRNA表达水平预测癌症患者PD-1和PD-L1免疫治疗响应率。
1.2.4. 使用HPA数据库鉴定PYCR1表达
从HPA数据库中获取了正常膀胱组织和BLCA组织中PYCR1表达水平免疫组化图谱（×1000），并获取了PYCR1高表达和低表达水平BLCA组织中肿瘤包膜侵袭和血管侵犯的局部HE染色图谱（Patient ID：264，265）。
1.2.5. 细胞培养
UM-UC-3细胞在DMEM培养基中培养，T24和5637细胞在RPMI 1640培养基中培养，SV-HUC-1细胞在F-12 K培养基中培养，3种培养基中均添加10%的FBS和1%的青霉素-链霉素，细胞在37 ℃、5% CO2的培养箱中培养。使用显微镜监测细胞生长，细胞达到80%时传代或冷冻保存。实验使用传代3~4代内的细胞进行。
1.2.6. 细胞转染
将T24和UM-UC-3细胞以50 000/孔的密度接种于6孔培养板中。当细胞长到培养板的80%左右时，使用Lipofectamine RNAiMax进行细胞转染。siRNA序列如下：PYCR1-siRNA1：正义，5'GCCC ACAAGAUAAUGGCUATT3'；反义，5'UAGCCAUUA UCUUGUGGGCTT3'；PYCR1-siRNA2：正义，5'GGU GGAAGAGGACCUGAUUTT3'；反义，5'AAUCAGG UCCUCUUCCACCTT 3'；NC：正义，5'UUCUCCGAA CGUGUCACGUTT3'；反义，5'ACGUGACACGUUC GGAGAATT。
1.2.7. 提取RNA和qRT-PCR
PYCR1和GAPDH基因引物均购自苏州吉玛基因股份有限公司。使用RNAiso Plus提取总 RNA，用PrimeScript RT Master Mix合成cDNA。使用SYBR Premix Ex TaqTM II试剂盒在QuantstudioTM DX系统上进行PCR扩增。扩增程序如下：95 ℃预变性1 min，95 ℃变性30 s，60 ℃退火30 s，72 ℃延伸60 s，40个循环。GAPDH用作内部对照，用2-ΔΔCt 法计算相对表达量，引物序列如下：PYCR1：正义，5'GAAGATGGGGGTGAAGTTGA3'；反义，5'C TCAATGGAGCTGATGGTGA3'；GAPDH：正义，5'GG AGCGAGATCCCTCCAAAAT3'；反义：5'GGCTGTT GTCATACTTCTCATGG3'。
1.2.8. CCK-8检测
在T24和UM-UC-3细胞中瞬时转染si-PYCR1和NC。培养24 h后，以每孔1000个细胞的密度将细胞播种到96孔板中。在0、24、48、72、96 h的时间点分别用10 μL CCK8溶液处理细胞，并在37 ℃孵育2 h，然后用多功能酶标仪测量吸光度A 450 nm。
1.2.9. 基于GEO数据集鉴定PYCR1作为BLCA独立风险因子
对BLCA患者GEO数据库数据集（GSE3167， GSE37817，GSE166716，GSE236292）进行基于PYCR1表达水平的Kaplan-Meier、ROC分析，数据均进行BATCH去批次处理。
1.2.10. 单因素和多因素回归检验BLCA独立预后风险因素
基于TCGA数据提取BLCA患者年龄、性别、拷贝数变异、单核苷酸变异等临床性状及PYCR1在样本中的表达水平，使用“limma”程序包进行单因素和多因素回归分析，控制变量以临床总生存期作为参考值，以患者生存状况作为临床结局，并通过“rmda”程序包构建BLCA临床决策曲线（DCA）用于判断BLCA患者临床预后和治疗策略。DCA可以帮助我们做出更好临床决策的风险模型［18］。
1.2.11. PYCR1用于预测BLCA免疫治疗和化疗疗效
使用3组共计387名经过免疫治疗患者的队列建立PYCR1评分模型探究PYCR1高表达对BLCA免疫治疗效果的预测价值，依托“pRRophetic”程序包在BLCA患者队列中发掘高敏感和具有PYCR1特异性的化疗药物。
1.2.12. PYCR1在BLCA中的功能富集分析
基于PYCR1 mRNA表达水平筛选其在BLCA中的共表达基因，设置条件为logFC>1，P<0.05。基因本体论（GO）和京都基因和基因组百科全书（KEGG）富集分析依托“org.Hs.eg.db”、“clusterProfiler”程序包完成并进行可视化，用于揭示PYCR1调控BLCA进展的信号通路。
1.2.13. 基于XSum算法筛选小分子抑制剂
利用Xsum算法基于PYCR1 mRNA表达水平对BLCA进行药物筛选［19］。通过分析癌和癌旁组织之间的表达差异，提取该疾病的分子特征。根据PYCR1特征进行CMap分析，以找到可能对抗BLCA的药物［20］。该过程依赖“PharmacoGx”和“CoreGx”程序包完成，并使用“ggplot2”进行可视化。
1.2.14. 分子对接筛选PYCR1潜在抑制剂
使用AutoDockTools Vina 1.2.5（CCSB， USA）对确定的靶标进行去溶剂化处理，并对侧链氨基酸残基进行补充。在CHARMM36力场中对能量进行最小化，并模拟各种小分子抑制剂构象。为模拟生理环境，还加入TIP3P和SPC/E溶剂模型。使用Vina score评估了结合亲和力，并确定目标蛋白质上的相互作用位点。每次模拟均重复五次，计算的结合能确保标准偏差低于2 kcal/mol，以保证可靠性。
1.2.15. 数据分析
公共数据库数据处理、分析和绘图由R Project（Lucent Technologies，v4.1.3，v4.3.1）及其对应的配套程序包完成。PyMOL软件（Schrödinger， Inc.， USA）用于可视化蛋白受体-小分子配体复合物三维互作模式。qRT-PCR、CCK-8和GEO独立队列中PYCR1的mRNA表达水平数据以均数±标准差表示。组间分析使用GraphPad Prism 8.0（La Jolla，CA），采用Shapiro-Wilk检验检测正态分布。qRT-PCR数据量较小且不符合正态分布，采用Mann-Whitney U检验进行统计分析；CCK-8和PYCR1的mRNA表达水平数据符合正态分布，采用独立样本t检验进行统计分析。Excel软件用于绘制柱状图，P<0.05表明差异具有统计学意义。
2. 结果
2.1. 泛癌中PYCR1的表达水平
在LUSC、BRCA、STAD和LUAD等肿瘤中，PYCR1表达水平显著高于正常组织（P<0.001，图1A、B）。在PRAD、KIRC、KIRP、KICH和BLCA中，PYCR1在肿瘤组织中的表达水平显著提高（P<0.001，图1C、 D）。
图1.
泛癌中PYCR1的表达
Fig.1 Expression of PYCR1 in pan-cancer. A, B: mRNA and protein expression levels of PYCR1 in UCSC Xena and GTEx pan-cancer data. C, D: Differences in mRNA and protein expression levels of PYCR1 between cancer and adjacent tissues in 33 cancers. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001 vs control group.
2.2. 泛癌中PYCR1表达水平与患者预后的关联
PYCR1表达水平与泛癌患者OS、PFS、DFS和DSS风险关联分析显示，PYCR1高表达与ESCA、ACC等肿瘤患者较差的OS相关（P<0.05）。PYCR1高表达与KIRC、KIRP等肿瘤患者较差的OS相关（P<0.01）。在RCC和PRAD中，PYCR1过表达与患者较差的DSS和PFS有关（P<0.05）。在ACC中，PYCR1的高表达均提示较差的OS、PFS、DFS和DSS（P<0.05，表1）。
表1.
泛癌中PYCR1的预后意义
Tab.1 Prognostic significance of PYCR1 in pan-cancer
| Index | Cancers | HR | 95% CI | P |
|---|---|---|---|---|
| OS |
SARC KIRP KIRC ESCA ACC LGG |
1.29 2.01 1.59 1.22 1.68 0.78 |
1.13-1.47 1.58-2.54 1.40-1.80 0.94-1.59 1.28-2.21 0.60-1.00 0.96-3.35 0.67-1.01 1.37-2.09 1.44-1.87 1.18-4.23 1.25-1.92 1.12-2.47 0.74-4.82 1.27-2.24 1.59-2.15 0.90-1.65 1.09-1.45 2.92-1003.87 2.27-4.20 |
<0.001 <0.01 <0.01 0.022 0.022 0.017 |
| PFS |
UVM LGG KIRP KIRC KICH ACC |
1.79 0.82 1.69 1.64 2.24 1.54 |
0.032 0.013 <0.01 <0.01 0.020 <0.01 |
|
| DFS | ACC | 1.66 | 0.024 | |
| DSS |
PCPG ACC KIRC ESCA SARC PRAD KIRP |
1.88 1.69 1.85 1.22 1.26 54.17 3.09 |
0.022 0.028 <0.01 0.043 0.047 0.030 <0.01 |
2.3. 泛癌中PYCR1调控的分子功能
GSEA富集分析表明，PYCR1在泛癌中主要富集在代谢重编程信号通路的顶端以及免疫调节信号的底部，包含药物代谢-细胞色素P450信号、卟啉和叶绿素的代谢、糖代谢和自噬调节等，提示PYCR1可能与肿瘤TIME抑制和药物耐受有关（表2）。
表2.
泛癌中PYCR1调控的分子功能（前3）
Tab.2 Top 3 molecular functions regulated by PYCR1 in pan-cancer
| Cancers | Description |
|---|---|
| CESC |
Maturity-onset diabetes of the young, Glutathione metabolism, Melanoma Olfactory Transduction, Maturity-onset diabetes of the young, Ascorbic Acid Metabolism Taurine and Hypotaurine Metabolism, Taste transduction, Hedgehog signaling pathway Ascorbic Acid Metabolism, Autophagy, Pentose and Glucuronate Interconversions Lysine Degradation, Arginine and Proline Metabolism, Basal Cell Carcinoma Folate biosynthesis, Autophagy, Cytosolic DNA sensing by cGAS Other glycan degradation, Autophagy, Butanoate Metabolism Primary immunodeficiency, Viral myocarditis, Graft-Versus-Host Disease Viral myocarditis, Maturity-onset diabetes of the young, Glutathione metabolism |
| LUSC | |
| LIHC | |
| THYM | |
| LAML | |
| CHOL | |
| UCS | |
| THCA | |
| KICH | |
| HNSC | Ascorbic Acid Metabolism, Porphyrin and Chlorophyll Metabolism, Metabolism of xenobiotics by cytochrome P450 |
| UCEC | Olfactory Transduction, Autophagy, Autoimmune thyroid disorders |
| STAD | Dilated cardiomyopathy, Arrhythmogenic Right Ventricular Cardiomyopathy, Hypertrophic cardiomyopathy |
| ACC | Glycosaminoglycan degradation, Hedgehog signaling pathway, Olfactory Transduction |
| SKCM | Primary immunodeficiency, Linoleic acid metabolism, Graft-Versus-Host Disease |
| LUAD | Olfactory Transduction, Autophagy, RIG-I-like Receptor signaling pathway |
| TGCT | Allograft rejection, Graft-Versus-Host Disease, Primary immunodeficiency |
| READ | Cytosolic DNA sensing by cGAS, Autoimmune thyroid disorders, Autophagy |
| LGG | Pentose and Glucuronate Interconversions, Ascorbic Acid Metabolism, Porphyrin and Chlorophyll Metabolism |
| BRCA | Olfactory Transduction, Cytosolic DNA sensing by cGAS, Autoimmune thyroid disorders |
| COAD | Taste transduction, Olfactory Transduction, Cytosolic DNA sensing by cGAS |
| BLCA | Maturity-onset diabetes of the young, Valine, Leucine, and Isoleucine Degradation, Graft-Versus-Host Disease |
2.4. 泛癌中PYCR1与免疫评分的关联
在异质性较高肿瘤如STAD、TGCT、PRAD和CESC中，PYCR1高表达与肿瘤纯度呈正相关关系（r>0.3，P<0.05，图2）。在LUAD、LUSC和GBM等肿瘤中，PYCR1高表达与免疫评分呈负相关（r<-0.3，P<0.05），与基质评分呈负相关关系（r<-0.3，P<0.05，图2）。
图2.
泛癌中PYCR1与肿瘤免疫评分的关联分析
Fig.2 Correlation analysis between PYCR1 and tumor immunity score in pan-cancer.
2.5. 泛癌中PYCR1与免疫细胞浸润水平的关联
免疫细胞浸润分析结果显示，在泛癌TIME景观中，PYCR1高表达主要与M0和M1型巨噬细胞浸润呈正相关关系（r>0.3，P<0.05），与肿瘤相关巨噬细胞（TAMs）的浸润呈负相关关系（r<-0.3，P<0.05），提示PYCR1高表达可能与TIME中炎症反应维持有关。在TGCT中，PYCR1则介导TAMs的浸润（r>0.3，P<0.05）。在LAML中，PYCR1与B淋巴细胞的活化有关（图3）。
图3.
泛癌中PYCR1调节免疫细胞浸润水平
Fig.3 PYCR1 regulates the level of immune cell infiltration in pan-cancer.
2.6. 泛癌中PYCR1与TMB、MSI和免疫治疗的潜在关联
PYCR1对肿瘤的TMB和MSI具有显著的影响，促进ACC、STAD和BLCA中TMB和MSI的升高（图4A、B）。在GSE78220免疫治疗队列中，PYCR1对抗PD-1治疗应答无预测价值。在GSE67501免疫治疗队列中，PYCR1高表达易造成抗PD-L1治疗耐受和无应答（P<0.05，图4B、C）。在IMvigor210免疫治疗队列中，与低表达组患者相比，PYCR1高表达BLCA患者接受免疫检查点抑制剂治疗预后效果更好（P<0.01，图4D）。
图4.
泛癌中PYCR1评估TMB、MSI以及免疫治疗获益的潜力
Fig.4 Potential of PYCR1 for assessing tumor mutation burden (TMB), microsatellite instability (MSI) and immune therapy benefits in pan-cancer. A: Correlation of PYCR1 with TMB and MSI. B: Correlation of PYCR1 with benefits of anti-PD-1 immunotherapy in SKCM patients in the SKCM immunotherapy cohort. C: Correlation of PYCR1 with benefits of anti-PD-L1 immunotherapy in RCC patients in the metastatic RCC immunotherapy cohort. D: Correlation of PYCR1 with immune checkpoint inhibitor therapy benefit in patients with BLCA in the IMvigor210 cohort.
2.7. 在BLCA中PYCR1的表达
与正常膀胱尿路上皮组织对比，PYCR1在BLCA组织中的mRNA表达水平显著提高（P<0.05，图1C、 D）。在4组外部BLCA mRNA表达独立验证集中，PYCR1在肿瘤组织的mRNA表达量均高于正常组织（P<0.05，图5A）。IHC图谱显示PYCR1蛋白在1名55岁男性（ID： 1798）正常膀胱尿路上皮组织中表达较弱，在低级别BLCA（89岁男性，ID： 2704）中呈中等强度染色，而在高级别BLCA（63岁女性，ID： 3780）中呈强阳性染色，主要表达于细胞质和细胞膜（图5B）。HE染色提示，PYCR1高表达水平的BLCA组织中包膜和血管侵袭较PYCR1低表达水平的BLCA组织更明显（A04/B03， A05/B01），平均PYCR1每百万个RNA分子中的转录本数（nTPM）更高（图6）。
图5. 在BLCA中PYCR1的mRNA和蛋白质表达水平.
图6.
PYCR1高表达和低表达的BLCA组织学特征
Fig.6 Histological features of BLCA with high and low PYCR1 expressions. A: HE staining profile of BLCA tissue with high PYCR1 expression level (Original magnification: ×1000). B: HE staining profile of BLCA tissue with low PYCR1 expression level (×1000).
Fig.5 mRNA and protein expression levels of PYCR1 in BLCA. A: Differences in the expression levels of PYCR1 between BLCA tissue and normal bladder and urinary tract epithelial tissue in the external GEO validation sets GSE3167, GSE166716, GSE13507, and GSE37817. B: Expression of PYCR1 in normal bladder and urinary tract epithelial tissue, low-grade BLCA tissue, and high-grade BLCA tissue. *P<0.05 vs Normal group.
2.8. 沉默PYCR1抑制BLCA细胞的增殖能力
qRT-PCR检测结果显示，与SV-HUC-1相比，PYCR1在BLCA细胞系中的表达水平明显上调（P<0.01，图7A）。将si-PYCR1转染到T24和UM-UC-3细胞中，PYCR1在细胞系中表达水平明显下降（P<0.01，图7B），细胞的增殖活力明显下降（P<0.01，图7C）。
图7.
沉默PYCR1对BLCA细胞系增殖能力的影响
Fig.7 Effect of PYCR1 silencing on proliferative capacity of BLCA cell lines. A: mRNA expression levels of PYCR1 in BLCA cell lines and SV-HUC-1. **P<0.01 vs SV-HUC-1. B: Verification of PYCR1 silencing efficiency. C: Effect of PYCR1 silencing on proliferation of T24 and UM-UC-3 cells. *P<0.05, **P<0.01, ***P<0.001 vs si-NC.
2.9. 在BLCA中PYCR1的预后意义
在GEO数据库GSE13507队列中，与PYCR1高表达组患者相比，低表达组患者具有更好的预后（P<0.001），1、3、5、8年AUC曲线分别为0.678、0.791、0.824、0.875，提示基于PYCR1 mRNA表达水平来预测BLCA患者预后具有较高的检验效力（图8）。
图8.
高PYCR1表达水平对BLCA患者预后的影响
Fig.8 Effect of high PYCR1 expression level on prognosis of BLCA patients.
2.10. PYCR1可作为BLCA的独立预后风险因素
单因素回归分析结果显示，BLCA患者的年龄、肿瘤分型、PYCR1 mRNA表达水平、单核苷酸变异、肿瘤低甲基化水平及肿瘤分期是影响BLCA患者临床预后的危险因素（P<0.05，表3）。多因素回归分析结果显示，BLCA患者的高龄（HR：1.05，95% CI：1.02-1.07，P=0.001）、PYCR1 mRNA过表达（HR：1.14，95% CI： 1.02-1.68，P=0.006）、单核苷酸变异位点增多（HR： 1.00，95% CI： 0.99-1.01，P=0.019）及肿瘤临床分期III-IV（HR：1.49，95% CI： 1.10-2.04，P=0.011）与BLCA患者不良预后有关（表4）。
表3.
单因素回归分析识别BLCA危险因素
Tab.3 Univariate analysis for identifying prognostic risk factors for BLCA
| Characteristics | HR | 95% CI | P |
|---|---|---|---|
| Age (year) | 1.05 | 1.02-1.07 | <0.001 |
| Tumor classification | 1.25 | 1.04-1.49 | 0.016 |
| Copy number variation | 0.97 | 0.79-1.19 | 0.764 |
| Gender | 0.95 | 0.55-1.65 | 0.858 |
| Grade | 2.02 | 0.49-8.29 | 0.327 |
| Hypermethylation | 0.93 | 0.79-1.09 | 0.377 |
| Hypomethylation | 0.92 | 0.78-1.08 | 0.044 |
| Tumor mutation burden | 1.10 | 0.85-1.44 | 0.457 |
| PYCR1 | 1.23 | 0.99-1.87 | 0.032 |
| Race | 0.91 | 0.67-1.24 | 0.544 |
| Single nucleotide variation | 1.00 | 0.99-1.01 | 0.027 |
| Stage | 1.60 | 1.19-2.15 | 0.002 |
表4.
多因素回归分析识别BLCA独立预后风险因素
Tab.4 Multivariate analysis for identifying independent prognostic risk factors for BLCA
| Characteristics | HR | 95% CI | Assignment and Attributes | P |
|---|---|---|---|---|
| Age (year) | 1.04 | 1.02-1.07 | Continuous variable | 0.001 |
| Tumor classification | 1.16 | 0.95-1.41 | 0= LumP,1= LumNS,2= LumU,3= Stroma-rich,4= Ba/Sq,5= NE-like | 0.135 |
| Hypomethylation | 0.89 | 0.75-1.05 | Continuous variable | 0.162 |
| PYCR1 | 1.14 | 1.02-1.68 | Continuous variable | 0.006 |
| Single nucleotide variation | 1.00 | 0.99-1.01 | Continuous variable | 0.019 |
| Stage | 1.49 | 1.10-2.04 | 0= I-II,1= III-IV | 0.011 |
2.11. 比较PYCR1与临床常用指标的决策能力
DCA模型显示，在12项指标中，PYCR1 mRNA表达水平、肿瘤分期、肿瘤分型及患者年龄具有较强的临床决策能力，要优于OS模型，其中，PYCR1 mRNA表达水平具有最高的BLCA临床决策效力，可用于独立评估BLCA患者的临床预后并辅助做出合适的治疗决策（图9）。
图9.
基于临床性状和PYCR1构建BLCA临床DCA模型
Fig.9 Construction of BLCA clinical DCA model based on clinical traits and PYCR1 expression levels.
2.12. PYCR1在BLCA中调控的分子功能
GO富集分析结果显示，PYCR1主要参与DNA结合的负调控、对神经生长因子的反应、淀粉样细胞迁移和上皮细胞迁移等生物学功能（P<0.05，Q<0.05，图10A）。KEGG富集分析显示，PYCR1可能与Hippo信号通路、细胞衰老、细胞周期和干细胞多能性等功能的调控有关（P<0.05，Q<0.05，图10B）。
图10.
PYCR1促进BLCA进展调控的分子机制
Fig.10 Molecular mechanism of PYCR1 for promoting BLCA progression. A, B: GO and KEGG signaling pathway enrichment analysis circle diagram of PYCR1 co-expressed genes.
2.13. 基于PYCR1的BLCA化疗药物敏感性分析
药物敏感性分析提示，与PYCR1低表达组相比，BLCA等尿路上皮癌常用药物如5-氟尿嘧啶、多柔比星、西妥昔单抗及丝裂霉素C等抗肿瘤药物在PYCR1高表达组中IC50显著提高（P<0.0001）。此外，戈沙妥珠单抗（SN-38）、曲美替尼和CUDC-101在PYCR1高表达患者中药物IC50显著提高（P<0.0001，图11）。
图11.
基于PYCR1高表达的BLCA药物敏感性分析
Fig.11 Drug sensitivity analysis of BLCA with high PYCR1 expressions.
2.14. 基于CMap XSum算法筛选PYCR1特异性抑制剂
CMap分析表明， Exisulind是最有可能通过特异性抑制PYCR1治疗BLCA的药物。Fasudil、W-13、Butein和Iloprost有较高的潜力成为PYCR1小分子抑制剂（图12）。
图12.
基于PYCR1表达谱的小分子药物虚拟筛选
Fig.12 Virtual screening of small molecule drugs based on PYCR1 expression profiling. A: Screening for small molecule drugs in BLCA patients based on PYCR1 expression profile using XSum algorithm. B: Likely compound structures of candidate small molecule inhibitors of PYCR1. C: Known PYCR1 inhibitor Pycr1-IN-1 is used as the positive control.
2.15. 分子对接验证小分子抑制剂-PYCR1复合物稳定性
Pycr1-IN-1等六种小分子配体均能与PYCR1蛋白受体结合形成稳定的复合物，其中Exisulind-PYCR1复合物最为稳定（表5）。Exisulind主要依靠氢键相互作用力与PYCR1侧链氨基酸残基缬氨酸（VAL）-70、天冬酰胺（ASN）-123和苏氨酸（THR）-124形成稳定复合物（图13A）。此外，PYCR1蛋白氨基酸残基VAL-70、丙氨酸（ALA）-272具有很强的成药活性（图13A~F）。
表5.
小分子抑制剂-PYCR1蛋白复合物构象结合能（前3）
Tab.5 Conformational binding energies of small molecule inhibitor-PYCR1 protein complexes （Top 3）
| Ligands | ID | Cavity volume (Å3) | Vina score (kcal/mol) | Center (x, y, z) |
|---|---|---|---|---|
| Pycr1-IN-1 | C1 | 436 | -5.2 | 34, 58, 14 |
| C2 | 377 | -4.5 | 31, 63, 1 | |
| C3 | 316 | -4.7 | 9, 60, -6 | |
| Exisulind | C2 | 377 | -7.1 | 31, 63, 1 |
| C4 | 232 | -7.1 | 1, 66, -5 | |
| C3 | 316 | -6.8 | 9, 60, -6 | |
| Fasudil | C1 | 436 | -7 | 34, 58, 14 |
| C4 | 232 | -6.5 | 1, 66, -5 | |
| C5 | 142 | -6.4 | 40, 83, -2 | |
| Butein | C4 | 232 | -6.6 | 1, 66, -5 |
| C2 | 377 | -6.5 | 31, 63, 1 | |
| C1 | 436 | -6.3 | 34, 58, 14 | |
| Iloprost | C1 | 436 | -6.3 | 34, 58, 14 |
| C2 | 377 | -6.2 | 31, 63, 1 | |
| C4 | 232 | -6.1 | 1, 66, -5 | |
| W-13 | C1 | 436 | -6.2 | 34, 58, 14 |
| C2 | 377 | -6 | 31, 63, 1 | |
| C4 | 232 | -5.9 | 1, 66, -5 |
图13.
候选化合物与蛋白受体互作模式图
Fig.13 Interaction pattern of candidate compounds with protein receptors. A-F: 3D interaction pattern diagram between candidate small molecule inhibitors and receptor protein PYCR1. Gray stick model represents subsequent small molecule inhibitors, the blue stick model represents PYCR1 side chain amino acid residues, the pink dashed line represents hydrophobic interactions, the green dashed line represents hydrogen bonding interactions, and the yellow dashed line represents alkyl interactions.
3. 讨论
PYCR1通过氧化NAD（P）H将Δ1-吡咯啉-5-羧酸盐（P5C）还原为脯氨酸，从而催化脯氨酸循环的生物合成半反应，在肿瘤的发生和发展过程中具有重要作用［21］。本研究发现PYCR1 的mRNA和蛋白质表达在BLCA组织中上调，并与癌症分期、组织学类型、TMB和MSI评分以及TIME重塑相关。PYCR1表达水平高提示BLCA患者的不良预后。既往的研究表明PYCR1是BLCA的生物标志物［12， 22］。本研究表明，PYCR1可以作为BLCA的独立危险因子，在BLCA的诊断、患者预后评估及治疗决策中具有潜在的临床价值。体外实验结果显示，PYCR1在BLCA细胞系中过表达，沉默PYCR1抑制了BLCA细胞的活力。Xiao S和Du S等人的研究同样表明，PYCR1过表达促进BLCA的恶性生物学行为［23， 24］，表明PYCR1与BLCA发生和发展密切相关。
研究表明，BLCA接受铂类化疗或5-氟尿嘧啶、多西他赛、紫杉醇治疗时易出现耐药，但具体分子机制并不清楚［25， 26］。敲除PYCR1可显著减少体内肿瘤的TMB并增加其对药物的敏感性［27］。本研究表明，PYCR1高表达时BLCA具有更高的TMB和MSI。此外，BLCA化疗常用药物如铂类、吉西他滨、5-氟尿嘧啶和多西他赛等IC50明显提高，提示化疗药物耐受。此外，戈沙妥珠单抗、曲美替尼、CUDC-101等新型药物在PYCR1高表达时也出现耐受。戈沙妥珠单抗是源自伊立替康的新型抗体偶联药物，已有研究表明其在BLCA中具有很好的安全性和有效性［28， 29］。曲美替尼是用于BLCA治疗的小分子蛋白激酶抑制剂［30， 31］。CUDC-101是一种靶向EGFR/HDAC/HER-2治疗BLCA的新型靶向治疗药物［32］。因此，PYCR1可能是导致BLCA化疗耐药的潜在因素，探明其诱导BLCA化疗药物耐受的分子机制具有重要意义。此外，PYCR1的过表达介导了肿瘤TIME重塑和免疫抑制的发生，提示PYCR1过表达的患者接受免疫治疗更有可能获益［10］。本研究表明，PYCR1高表达时，BLCA患者使用免疫检查点抑制剂治疗时可能具有更好的效果。因此，PYCR1可以作为BLCA患者化疗药物选择和免疫治疗方案决策的潜力指标。
计算机辅助药物设计能够加快药物研发和降低药物开发过程的成本［33］。虚拟筛选和分子对接被广泛用于小分子抑制剂及靶向药物的发掘［34， 35］。本研究基于CMap-XSum算法和分子对接技术筛选并验证了Exisulind作为新型PYCR1抑制剂用于治疗BLCA的潜在可能，Exisulind能够很好的与PYCR1蛋白受体保持稳定结合。已有研究表明，Exisulind是一种靶向促凋亡药物，在泛癌中诱导肿瘤细胞凋亡且不影响正常细胞［36］。此外，Exisulind展现出了很好的靶向特异性，在胃癌和HCC中具有显著的抗肿瘤效果［37， 38］。
综上所述，PYCR1是癌症潜在的预后生物标志物和治疗靶点，与BLCA恶性生物学行为以及化疗药物耐受的产生有关。小分子抑制剂Exisulind可能通过靶向PYCR1抑制BLCA进展，但有待后续实验进一步验证。
基金资助
广东省基础与应用基础研究基金（2024A1515012742）；广东医科大学大学生创新创业训练计划项目（GDMU2023048，GDMU2023355，S202410571048）
参考文献
- 1. Li Y, Xu J, Bao P, et al. Survival and clinicopathological significance of PYCR1 expression in cancer: A meta-analysis[J]. Front Oncol, 2022, 12: 985613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Luo Y, Yu J, Lin Z, et al. Metabolic characterization of sphere-derived prostate cancer stem cells reveals aberrant urea cycle in stemness maintenance[J]. Int J Cancer, 2024, 155(4): 742-55. [DOI] [PubMed] [Google Scholar]
- 3. Kay EJ, Paterson K, Riera-Domingo C, et al. Cancer-associated fibroblasts require proline synthesis by PYCR1 for the deposition of pro-tumorigenic extracellular matrix[J]. Nat Metab, 2022, 4(6): 693-710. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Zhou P, Du X, Jia W, et al. Engineered extracellular vesicles for targeted reprogramming of cancer-associated fibroblasts to potentiate therapy of pancreatic cancer[J]. Signal Transduct Target Ther, 2024, 9(1): 151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Fang K, Sun M, Leng Z, et al. Targeting IGF1R signaling enhances the sensitivity of cisplatin by inhibiting proline and arginine metabolism in oesophageal squamous cell carcinoma under hypoxia [J]. J Exp Clin Cancer Res, 2023, 42(1): 73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Li F, Zheng Z, Chen W, et al. Regulation of cisplatin resistance in bladder cancer by epigenetic mechanisms[J]. Drug Resist Updat, 2023, 68: 100938. [DOI] [PubMed] [Google Scholar]
- 7. Lopez-Beltran A, Cookson M S, Guercio B J, et al. Advances in dia-gnosis and treatment of bladder cancer[J]. Bmj, 2024, 384: e076743. [DOI] [PubMed] [Google Scholar]
- 8. Rouanne M, Adam J, Radulescu C, et al. BCG therapy down-regulates HLA-I on malignant cells to subvert antitumor immune responses in bladder cancer[J]. J Clin Invest, 2022, 132(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Al Hussein Al Awamlh B, Chang SS. Novel Therapies for High-Risk Non-Muscle Invasive Bladder Cancer[J]. Curr Oncol Rep, 2023, 25(2): 83-91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Xu X, Wang Y, Hu X, et al. Effects of PYCR1 on prognosis and immunotherapy plus tyrosine kinase inhibition responsiveness in metastatic renal cell carcinoma patients[J]. Neoplasia, 2023, 43: 100919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Fan G, Yu B, Tang L, et al. TSPAN8(+) myofibroblastic cancer-associated fibroblasts promote chemoresistance in patients with breast cancer[J]. Sci Transl Med, 2024, 16(741): eadj5705. [DOI] [PubMed] [Google Scholar]
- 12. Li Z, Jiang Y, Liu J, et al. Exosomes from PYCR1 knockdown bone marrow mesenchymal stem inhibits aerobic glycolysis and the growth of bladder cancer cells via regulation of the EGFR/PI3K/AKT pathway[J]. Int J Oncol, 2023, 63(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Zhang J, Yuan S, Cao W, et al. Signature Search Polestar: a comprehensive drug repurposing method evaluation assistant for customized oncogenic signature[J]. Bioinformatics, 2024, 40(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Huang X, Tan J, Chen M, et al. Prognostic, Immunological, and Mutational Analysis of MTA2 in Pan-Cancer and Drug Screening for Hepatocellular Carcinoma[J]. Biomolecules, 2023, 13(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Huang J, Zhang J L, Ang L, et al. Proposing a novel molecular subtyping scheme for predicting distant recurrence-free survival in breast cancer post-neoadjuvant chemotherapy with close correlation to metabolism and senescence[J]. Front Endocrinol (Lausanne), 2023, 14: 1265520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Geeleher P, Cox NJ, Huang RS. Clinical drug response can be predicted using baseline gene expression levels and in vitro drug sensitivity in cell lines[J]. Genome Biol, 2014, 15(3): R47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Yoshihara K, Shahmoradgoli M, Martínez E, et al. Inferring tumour purity and stromal and immune cell admixture from expression data [J]. Nat Commun, 2013, 4: 2612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Van Calster B, Wynants L, Verbeek JFM, et al. Reporting and Interpreting Decision Curve Analysis: A Guide for Investigators[J]. Eur Urol, 2018, 74(6): 796-804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Yang C, Zhang H, Chen M, et al. A survey of optimal strategy for signature-based drug repositioning and an application to liver cancer [J]. Elife, 2022, 11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Subramanian A, Narayan R, Corsello SM, et al. A Next Generation Connectivity Map: L1000 Platform and the First 1, 000, 000 Profiles [J]. Cell, 2017, 171(6): 1437-52.e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Christensen EM, Bogner AN, Vandekeere A, et al. In crystallo screening for proline analog inhibitors of the proline cycle enzyme PYCR1[J]. J Biol Chem, 2020, 295(52): 18316-27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Liu Z, Sun T, Zhang Z, et al. An 18-gene signature based on glucose metabolism and DNA methylation improves prognostic prediction for urinary bladder cancer[J]. Genomics, 2021, 113(1 Pt 2): 896-907. [DOI] [PubMed] [Google Scholar]
- 23. Xiao S, Chen J, Wei Y, et al. BHLHE41 inhibits bladder cancer progression via regulation of PYCR1 stability and thus inactivating PI3K/AKT signaling pathway[J]. Eur J Med Res, 2024, 29(1): 302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Du S, Sui Y, Ren W, et al. PYCR1 promotes bladder cancer by affecting the Akt/Wnt/β‑catenin signaling[J]. J Bioenerg Biomembr, 2021, 53(2): 247-58. [DOI] [PubMed] [Google Scholar]
- 25. Deng D, Liu F, Liu Z, et al. Robust pyroptosis risk score guides the treatment options and predicts the prognosis of bladder carcinoma [J]. Front Immunol, 2022, 13: 965469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Xiong S, Li S, Zeng J, et al. Deciphering the immunological and prognostic features of bladder cancer through platinum-resistance-related genes analysis and identifying potential therapeutic target P4HB[J]. Front Immunol, 2023, 14: 1253586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Shenoy A, Belugali Nataraj N, Perry G, et al. Proteomic patterns associated with response to breast cancer neoadjuvant treatment[J]. Mol Syst Biol, 2020, 16(9): e9443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Loriot Y, Petrylak DP, Rezazadeh Kalebasty A, et al. TROPHY-U-01, a phase II open-label study of sacituzumab govitecan in patients with metastatic urothelial carcinoma progressing after platinum-based chemotherapy and checkpoint inhibitors: updated safety and efficacy outcomes[J]. Ann Oncol, 2024, 35(4): 392-401. [DOI] [PubMed] [Google Scholar]
- 29. Goldenberg DM, Sharkey RM. Antibody-drug conjugates targeting TROP-2 and incorporating SN-38: A case study of anti-TROP-2 sacituzumab govitecan[J]. MAbs, 2019, 11(6): 987-95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Elbadawy M, Sato Y, Mori T, et al. Anti-tumor effect of trametinib in bladder cancer organoid and the underlying mechanism[J]. Cancer Biol Ther, 2021, 22(5-6): 357-71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Plumber SA, Tate T, Al-Ahmadie H, et al. Rosiglitazone and trametinib exhibit potent anti-tumor activity in a mouse model of muscle invasive bladder cancer[J]. Nat Commun, 2024, 15(1): 6538. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Wang Z, Li L, Chu C, et al. CUDC‑101 is a potential target inhibitor for the EGFR‑overexpression bladder cancer cells[J]. Int J Oncol, 2023, 63(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Vemula D, Jayasurya P, Sushmitha V, et al. CADD, AI and ML in drug discovery: A comprehensive review[J]. Eur J Pharm Sci, 2023, 181: 106324. [DOI] [PubMed] [Google Scholar]
- 34. Zhang Y, Luo M, Wu P, et al. Application of Computational Biology and Artificial Intelligence in Drug Design[J]. Int J Mol Sci, 2022, 23(21). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Li Y, Miao J, Liu C, et al. Kushenol O Regulates GALNT7/NF-κB axis-Mediated M2 Macrophage Polarization and Efferocytosis in Papillary Thyroid Carcinoma[J]. Phytomedicine, 2025: 156373. [DOI] [PubMed] [Google Scholar]
- 36. Goluboff E T. Exisulind, a selective apoptotic antineoplastic drug [J]. Expert Opin Investig Drugs, 2001, 10(10): 1875-82. [DOI] [PubMed] [Google Scholar]
- 37. Geng H, Huang C, Xu L, et al. Targeting cellular senescence as a therapeutic vulnerability in gastric cancer[J]. Life Sci, 2024, 346: 122631. [DOI] [PubMed] [Google Scholar]
- 38. Huang Y, Wang S, Zhang X, et al. Identification of Fasudil as a collaborator to promote the anti-tumor effect of lenvatinib in hepatocellular carcinoma by inhibiting GLI2-mediated hedgehog signaling pathway[J]. Pharmacol Res, 2024, 200: 107082. [DOI] [PubMed] [Google Scholar]