The risk for several common cancers is influenced by the transcriptomic landscape of the respective tissue-of-origin. Vitamin D influences in vitro gene expression and cancer cell growth. We sought to determine whether oral vitamin D induces beneficial gene expression effects in human rectal epithelium and identify biomarkers of response. Blood and rectal mucosa was sampled from 191 human subjects and mucosa gene expression (HT12) correlated with plasma vitamin D (25-OHD) to identify differentially expressed genes. Fifty subjects were then administered 3200IU/day oral vitamin D3 and matched blood/mucosa resampled after 12 weeks. Transcriptomic changes (HT12/RNAseq) after supplementation were tested against the prioritised genes for gene-set and GO-process enrichment. To identify blood biomarkers of mucosal response, we derived receiver-operator curves and C-statistic (AUC) and tested biomarker reproducibility in an independent Supplementation Trial (BEST-D). Six hundred twenty-nine genes were associated with 25-OHD level (P < 0.01), highlighting 453 GO-term processes (FDR<0.05). In the whole intervention cohort, vitamin D supplementation enriched the prioritised mucosal gene-set (upregulated gene-set P < 1.0E−07; downregulated gene-set P < 2.6E−05) and corresponding GO terms (P = 2.90E−02), highlighting gene expression patterns consistent with anti-tumour effects. However, only 9 individual participants (18%) showed a significant response (NM gene-set enrichment P < 0.001) to supplementation. Expression changes in HIPK2 and PPP1CC expression served as blood biomarkers of mucosal transcriptomic response (AUC=0.84 [95%CI 0.66–1.00]) and replicated in BEST-D trial subjects (HIPK2 AUC=0.83 [95%CI 0.77–0.89]; PPP1CC AUC=0.91 [95%CI 0.86–0.95]). Higher plasma 25-OHD correlates with rectal mucosa gene expression patterns consistent with anti-tumour effects, and this beneficial signature is induced by short-term vitamin D supplementation. Heterogenous gene expression responses to vitamin D may limit the ability of randomised trials to identify beneficial effects of supplementation on CRC risk. However, in the current study blood expression changes in HIPK2 and PPP1CC identify those participants with significant anti-tumour transcriptomic responses to supplementation in the rectum. These data provide compelling rationale for a trial of vitamin D and CRC prevention using easily assayed blood gene expression signatures as intermediate biomarkers of response.

中文翻译：

---

口服维生素 D 补充剂可诱导与抗肿瘤作用相关的直肠粘膜转录组学变化

几种常见癌症的风险受各自组织来源的转录组学景观的影响。维生素 D 影响体外基因表达和癌细胞生长。我们试图确定口服维生素 D 是否会在人类直肠上皮细胞中诱导有益的基因表达效应，并确定反应的生物标志物。从 191 名人类受试者中采集血液和直肠粘膜样本，并将粘膜基因表达 (HT12) 与血浆维生素 D (25-OHD) 相关联，以确定差异表达的基因。然后对 50 名受试者进行 3200IU/天的口服维生素 D3 给药，并在 12 周后重新采样匹配的血液/粘膜。针对基因集和 GO 过程富集的优先基因测试补充后的转录组学变化 (HT12/RNAseq)。为了识别粘膜反应的血液生物标志物，我们推导了接受者-操作者曲线和 C 统计量 (AUC)，并在独立补充试验 (BEST-D) 中测试了生物标志物的再现性。629 个基因与 25-OHD 水平相关（P < 0.01），突出显示 453 个 GO-term 过程（FDR <0.05）。在整个干预队列中，补充维生素 D 丰富了优先粘膜基因组（上调基因组 P < 1.0E-07；下调基因组 P < 2.6E-05）和相应的 GO 项（P = 2.90E-02） ), 突出了与抗肿瘤作用一致的基因表达模式。然而，只有 9 名个体参与者 (18%) 对补充剂表现出显着反应（NM 基因集富集 P < 0.001）。HIPK2 和 PPP1CC 表达的表达变化作为粘膜转录组反应的血液生物标志物 (AUC=0.84 [95%CI 0.66–1. 00]）并在 BEST-D 试验对象中重复（HIPK2 AUC=0.83 [95%CI 0.77–0.89]；PPP1CC AUC=0.91 [95%CI 0.86–0.95]）。较高的血浆 25-OHD 与与抗肿瘤作用一致的直肠粘膜基因表达模式相关，并且这种有益特征是由短期维生素 D 补充引起的。对维生素 D 的异质基因表达反应可能会限制随机试验确定补充剂对 CRC 风险的有益影响的能力。然而，在目前的研究中，HIPK2 和 PPP1CC 的血液表达变化确定了那些对直肠补充剂具有显着抗肿瘤转录反应的参与者。这些数据为使用易于测定的血液基因表达特征作为反应的中间生物标志物的维生素 D 和 CRC 预防试验提供了令人信服的理由。