PO.MCB06.01 · 分子与细胞生物学
与老年女性相比,年轻女性原发性乳腺肿瘤中ESR1的表观遗传沉默:一项整合甲基化与表达的分析
Epigenetic silencing of ESR1 in primary breast tumors from young women compared to those from older women: An integrative methylation and expression analysis
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
患原发性乳腺癌的年轻女性(<40岁)常表现为更具侵袭性的疾病,且雌激素受体α(ESR1)表达低于老年女性(≥60岁),但其潜在机制尚不清楚。我们假设DNA甲基化参与年轻女性肿瘤中ESR1的抑制。我们分析了来自108例原发性肿瘤(54例年轻,54例老年)的TCGA-BRCA HumanMethylation450数据,这些样本按PCAPAM50亚型和种族进行匹配。预处理后,数据集包含363,870个CpG位点。使用limma进行差异甲基化分析,并校正亚型和种族。基因集富集分析(GSEA)使用从CpG映射的21,582个基因的预排序t值。使用匹配的TCGA RNA-seq数据评估ESR1甲基化与表达之间的相关性。使用Lund队列(GSE75067;n=61)进行验证;由于样本量较小,全基因组建模不可行,因此采用Wilcoxon检验评估ESR1相关CpG。我们鉴定出3,012个差异甲基化CpG(FDR<0.05,倍数变化>2),其中36个在年轻女性肿瘤中高甲基化,2,976个低甲基化,表明老年肿瘤存在整体甲基化更高的趋势。在TCGA的53个ESR1相关CpG中,7个仅在年轻管腔型肿瘤中显著高甲基化,且全部7个在管腔A/B型中均显著高甲基化。值得注意的是,老年女性中没有任何ESR1 CpG发生高甲基化。这7个位点的ESR1甲基化与表达呈负相关(ρ=−0.35至−0.51),与转录沉默一致。在Lund数据集中,61个ESR1 CpG中,2个在年轻管腔A型中高甲基化,16个在管腔B型肿瘤中高甲基化,而1个在管腔B型中呈低甲基化。在TCGA中高甲基化的7个CpG中,年轻LumA和LumB各有1个显著;其余虽显示相似趋势但无统计学意义,可能因样本量小所致。基于甲基化的GSEA显示老年组早期雌激素反应通路中高甲基化富集,与既往报道年轻女性雌激素通路活性增强的研究一致。我们报道年轻乳腺癌中存在一致的ESR1启动子高甲基化,尤其在管腔A/B亚型中,并得到与转录本水平负相关的支持。这些发现提示ESR1沉默是年轻乳腺肿瘤的一个反复出现的特征,并强化了基于甲基化的生物标志物在按年龄分层的癌症分析中的潜在应用价值。免责声明:本出版物内容仅由作者负责,不一定反映USUHS、HJF、国防部或陆军、海军、空军各部门的观点、意见或政策。提及商品名、商业产品或组织并不意味着美国政府的认可。
查看英文原文 English abstract
Young women with primary breast cancer (<40 years) often present with more aggressive disease and show lower estrogen receptor alpha (ESR1) expression than older women (≥60 years), but the underlying mechanisms remain unclear. We hypothesized that DNA methylation contributes to ESR1 repression in tumors from young women. We analyzed TCGA-BRCA HumanMethylation450 data from 108 primary tumors (54 young, 54 older) matched by PCAPAM50 subtype and race. After preprocessing, the dataset contained 363,870 CpG sites. Differential methylation was performed using limma, adjusting for subtype and race. Gene set enrichment analysis (GSEA) used preranked t-values for 21,582 genes mapped from CpGs. Matched TCGA RNA-seq data were used to evaluate correlations between ESR1 methylation and expression. Validation was conducted using the Lund cohort (GSE75067; n = 61); genome-wide modeling was not feasible due to smaller sample size, so Wilcoxon tests assessed ESR1-associated CpGs. We identified 3,012 differentially methylated CpGs (FDR < 0.05, fold change > 2), with 36 hypermethylated and 2,976 hypomethylated in tumors from young women, indicating a global trend toward higher methylation in older tumors. Among 53 ESR1-associated CpGs in TCGA, seven were significantly hypermethylated exclusively in young luminal tumors and all 7 are significantly hypermethylated in both luminal A/B. Notably, none of the ESR1 CpGs were hypermethylated in older women. ESR1 methylation of these 7 sites was inversely correlated with expression (ρ = −0.35 to −0.51), consistent with transcriptional silencing. In the Lund dataset, of 61 ESR1 CpGs, 2 were hypermethylated in young Luminal A and 16 in Luminal B tumors, whereas 1 showed hypomethylation in Luminal B. Of the 7 CpGs hypermethylated in TCGA, one each was significant in young LumA and LumB; others showed similar trends, though not statistically significant, likely due to small sample size. Methylation-based GSEA showed enrichment of hypermethylation within the early estrogen response pathway in the older group, consistent with previous studies reporting heightened estrogen pathway activity in younger women. We report consistent ESR1 promoter hypermethylation in young breast cancer, especially in Luminal A/B subtypes, supported by inverse correlation with transcript levels. These findings suggest that ESR1 silencing is a recurrent feature of young breast tumors and reinforce the potential utility of methylation-based biomarkers for age-stratified cancer profiling. Disclaimer: The contents of this publication are the sole responsibility of the author(s) and do not necessarily reflect the views, opinions or policies of USUHS, HJF, the DoW or the Departments of the Army, Navy or Air Force. Mention of trade names, commercial products, or organizations does not imply endorsement by the U.S. Government.
利益披露 Disclosure
P. Raj Kumar, None..
A. Praveen Kumar, None..
M. Russo, None..
J. Liu, None.
H. Hu,
miRoncol Diagnostics, Inc. co-founder and shareholder.
X. Lin, None.