背景和目的：癌症相关成纤维细胞（Cancer-associated fibroblasts，CAFs）在肿瘤治疗和患者预后中起着关键作用。CAFs在肿瘤微环境中呈现出功能和表型的异质性，从而促进了癌症的发生和发展。肿瘤干细胞（Cancer stem cells, CSCs）是一类具有干细胞特征的肿瘤细胞，在多种癌症的治疗耐受和复发中发挥着重要作用，但在食管鳞癌中缺乏特异标志物靶向CSCs。本研究目的在于探究食管鳞癌中低氧微环境维持CSCs生态位的具体机制，探寻食管鳞癌新的潜在治疗靶点及治疗反应预测的生物标志物以改善治疗疗效。

方法： 本研究利用60例食管鳞癌患者的单细胞转录组测序数据挖掘与肿瘤干性相关的基因特征及细胞亚群。通过构建常氧及低氧条件下成纤维细胞与食管鳞癌细胞共培养模型，在体外实验探究低氧微环境下食管鳞癌细胞干性具体的支持性生态位，利用了定量RT-PCR和Western blot去检测实验处理后食管鳞癌细胞干性及分化靶点的变化，流式分析探究实验处理后肿瘤干性样细胞的比例变化，成球实验验证肿瘤干性的变化，凋亡实验及克隆形成实验探究实验处理后肿瘤细胞对化疗的耐受的变化。运用慢病毒表达载体及短干扰RNA的基因编辑技术进行目的基因表达的下调探寻调控肿瘤干性的关键靶点。此外，构建CAFs与肿瘤细胞共注射的免疫缺陷小鼠的细胞移植瘤模型进行小鼠体内实验探究目的基因对肿瘤细胞成瘤能力及化疗耐受性的影响。此外，构建常氧及低氧诱导CAFs极化的共培养模型，获得了hsCAFs用于后续实验并利用此模型探究了低氧对hsCAFs的调控机制。趋化实验用于验证低氧环境中CAFs的趋化能力的改变。β-半乳糖苷酶实验及γH2AX的免疫荧光染色验证CAFs的衰老表型，ELISA实验验证CAF的衰老相关分泌表型及IGF1的分泌变化。最后为了探究hsCAF的临床相关性，收集了两组食管鳞癌患者队列分别用于验证hsCAF与食管鳞癌患者预后的相关性及与化疗疗效的相关性。

结果： 我们发现了在训练队列中食管鳞癌干性与低氧微环境显著相关，进一步发现了以衰老、分泌IGF1和低氧环境广泛存在为特征hsCAF及其与食管鳞癌的干性呈正相关。我们利用CAF与食管鳞癌细胞低氧共培养模型证实了低氧条件下CAFs促进了食管鳞癌细胞的干性，hsCAFs维持了食管鳞癌肿瘤干细胞的生态位，且促进了肿瘤的化疗耐受。我们验证了多种hsCAF与肿瘤干性相关因子，结果发现IGF通路在低氧条件下CAF促肿瘤干性中起着关键作用，验证了低氧条件下CAFs通过分泌IGF1促进食管鳞癌细胞的干性，并利用体外CAFs与食管鳞癌细胞共注射的小鼠成瘤模型验证了敲除IGF1的CAFs抑制了食管鳞癌细胞的成瘤能力及化疗耐受性。在机制上，我们发现了IGF1通过结合IGF1R-ITGA6/ITGB4受体抑制了AMPK活性进而促进了食管鳞癌细胞的干性和化疗耐药性。在此基础上，我们发现hsCAFs的形成及其IGF1的分泌上调是由低氧条件下食管鳞癌细胞激活IL1信号介导的，其中，CAFs中NFIA在低氧条件下上调进而调控IGF1表达上调。此外，我们进一步发现了低氧条件下食管鳞癌NRF2的激活诱导了CAF的衰老及衰老相关分泌表型，即hsCAF的形成，其机制是通过NRF2调控IL-1α分泌介导的。临床上，我们验证了hsCAF与食管鳞癌患者的预后不良及化疗耐药相关，阐明了靶向hsCAF这一治疗策略的临床价值。

结论： 食管鳞癌组织中存在一种低氧诱导的衰老成纤维细胞亚群。HsCAFs通过分泌IGF1抑制AMPK活性维持肿瘤干细胞的生态位进而促进了食管鳞癌的化疗耐药。低氧条件下，hsCAF的形成是低氧与癌细胞协同驯化CAFs的结果，低氧的食管鳞癌细胞中NRF2的激活上调IL-1α介导了CAF的衰老，激活NFIA促进了IGF1的分泌。食管鳞癌患者中hsCAFs提示预后不良与化疗耐药，这为临床个体化治疗及开发新的抗肿瘤策略提供了依据。

Background & Aims: Cancer-associated fibroblasts (CAFs) play a key role in tumor therapy and patient prognosis. CAFs exhibit functional and phenotypic heterogeneity in the tumor microenvironment, thus promoting the development of cancer. Cancer stem cells (CSCs) are a subpopulation with stem cell characteristics in cancer cells, which play an important role in the treatment resistance and recurrence of various cancers, but there is a lack of specific markers to target CSCs in esophageal squamous-cell carcinoma (ESCC). The aim of this study was to explore the specific mechanism of hypoxic microenvironment maintaining CSCs niche in ESCC, and to explore new potential therapeutic targets and biomarkers for treatment response prediction in ESCC.

Methods: In this study, single-cell RNA sequencing data of 60 patients with ESCC were used to explore the gene characteristics and cell subsets related to tumor stemness. By constructing a co-culture model of CAFs and ESCC cells under normoxia and hypoxia, the specific supportive niche of cancer stemness under hypoxia microenvironment was explored in vitro. Quantitative RT-PCR and Western blot were used to detect the changes of stemness-related markers with corresponding treatment. Flow analysis was used to investigate the change of cancer stem-like cells. Sphere formation assay was used to verify the function of stemness, Apoptosis experiment and clonal formation assay was conducted to investigate chemoresistance. Lentiviral expression vectors and short interfering RNA gene editing techniques were used to down-regulate the expression of genes to explore the key marker for stemness. In addition, a xenograft model co-injected with CAFs and ESCC cells was constructed to explore tumorigenic ability and chemoresistance in vivo. In addition, a co-culture model of CAFs polarization induced by normal and hypoxia was constructed to obtain hsCAFs and explore the regulatory mechanism of hypoxia on hsCAFs. The transwell assay was used to verify the chemotactic ability. The senescence phenotype of CAFs was verified by β-galactosidase assay, immunofluorescence staining of γH2AX, and ELISA assay. Finally, two ESCC patient cohorts were collected to verify the clinical correlation of hsCAF.

Results: In this study, we found a significant correlation between stemness and hypoxia in training cohort, and reported an unpublished type of CAFs, hsCAF, was positively correlated with tumor stemness. We used hypoxic co-culture model of CAF and ESCC cells to confirm that CAFs promoted tumor stemness under hypoxia, and hsCAFs maintained a supportive niche of CSCs for chemoresistance. Furthermore, we found that IGF1 within IGF pathway played a key role in promoting tumor stemness under hypoxia. IGF1-knockout CAFs inhibited the tumor formation ability and chemoresistance of ESCC. Mechanistically, IGF1 inhibited AMPK activity by binding to IGF1R-ITGA6/ITGB4 receptors to function. Moreover, the formation of hsCAFs was mediated by the activation of IL1 signaling by ESCC cells under hypoxia and NFIA was activated to regulates the IGF1 expression. Additionally, the activation of NRF2 in ESCC under hypoxia induced the senescence and senescence associated secretory phenotype (SASP) of CAFs via IL-1α. Clinically, we verified that hsCAF was associated with poor prognosis and chemotherapy resistance and clarified the clinical value of targeting hsCAF as a therapeutic strategy.

Conclusion: HsCAFs play a key role in maintenance of tumor stemness under hypoxia in ESCC. Mechanistically, hsCAFs inhibit AMPK activity by secreting IGF1, thereby promoting chemoresistance. The formation of hsCAF is the result of synergistic effect from hypoxia and cancer cells. Activation of NRF2 in ESCC promotes IL-1α to regulate CAF senescence and IGF1 expression via NFIA. High proportion of hsCAFs suggests poor prognosis and chemotherapy resistance, which provides a basis for development of individualized therapies and new anti-tumor strategies in ESCC.