Additional file 1 of Dynamics of the gut microbiome, IgA response, and plasma metabolome in the development of pediatric celiac disease

Additional file 1: Figure S1. Flow chart describing CD progressors demographics and sample size. Figure S2. Violin plot representation of most enriched genus/species in CD progressors compared to healthy controls. A. Fold change in ASVs at age 2.5 CD progressors (left panel, n=15) and healthy controls (right panel, n=16). B. Fold change in ASVs at age 5 in CD progressors (left panel, n=10) and healthy controls (right panel, n=13). Figure S3. Gating strategy for IgA sequencing and analysis. A. Schematic overview of IgA-based fecal bacteria separation combined with 16S rRNA gene sequencing (IgA-seq) for stool samples from CD progressors and healthy controls. MACS: Magnetic-activated cell separation. B. Gating strategy for the isolation of IgA-/+ bacteria from the CD progressors and healthy controls’ fecal samples. C. Empirical Bayes quasi-likelihood F-tests analysis for comparing IgA-coated and non-coated gut microbiota ASVs in healthy controls (upper row) and CD progressors (lower row) at ages 2.5, and 5. Frequency: number of ASVs. FDR: False Discovery Rate. D. Empirical Bayes quasi-likelihood F-tests analysis for the comparisons of IgA-coated or non-coated gut microbiota ASVs between CD progressors and healthy controls (upper row: age 2.5 years old; lower row: age 5 years old). F. Box plots showing representative ASVs in which abundances were similar in the gut microbiota (presorting samples) but differently targeted by IgA at age 2.5. E. Violin plots showing representative ASVs in which abundances were similar in the gut microbiota (presorting samples) but differently targeted by IgA at age 5. Figure S4. Heat Map showing IgA target in CD progressors and healthy subjects. A. Heat map showing the relative abundance of the top ASVs significantly different between IgA+ and IgA- samples of CD progressors and healthy controls (ASVs=51, selected based on p-value) at age 2.5 B. at age 5. Each column represents an individual participant and each row represents an ASV. Figure S5. The cytokine and plasma metabolome profiles of CD progressors and CD patients. A. Comparison of all 48 cytokines analyzed in plasma samples obtained from CD progressors (n=10) and healthy controls (n=10) at age 5. Data were expressed as means ±SEM. *p<0.05, **p<0.01, ***p<0.001. Statistical analysis was performed by a two-tailed, unpaired student’s t-test. B. Violin plots showing the representative Clostridium XIVa bacteria abundance between CD progressors and healthy controls (Left: before separation by IgA coating, Right: in IgA+ bacteria). Figure S6. Gating Strategy for the Flow cytometry analysis. Strategy 1- Gating strategy for NK1.1 and Qa-1 expression in TCRβ+ cells. Strategy 2- Gating strategy for CD8, CD4, NKG2D, CD103, and NKp46. Figure S7. TDCA diet induces changes in T-cell composition in different cell subsets. A. H&E images of ileum tissue sections of control and TDCA treated female mice. Full image (upper panel) and image at high magnification (lower panel). Scale =20μm. B. Villi/ Crypt ratio in ileum tissue sections of control and the TDCA treated female mice. C. Number of plasma cells in the lamina propria of the ileum section of female mice. D. TCRβ+ cells as % of total CD45+ cells. E. NKG2D+ cells as % of total TCRβ+ CD45+ cells. F. CD103+ cells as % of total CD4+ cells. G. Qa-1+ cells as % of total CD4+ cells in the IELs, PP, LP, and spleen of female (left panel) and male (right panel) mice. H Relative gene expression of Qa-1 and IL-10 in the ileum tissue analyzed using qPCR. Female (left panel) and male (right panel) mice after 10 weeks of TDCA treatment compared to controls. Data were expressed as mean ± SEM. *p<0.05, **p <0.01, ***p<0.001. Statistical analysis was performed by a two-tailed unpaired student’s t-test.

补充文件1：图S1。描述克罗恩病（Crohn’s Disease，CD）进展者人口统计学特征与样本量的流程图。
图S2。相较于健康对照，克罗恩病（CD）进展者中富集度最高的菌属/物种的小提琴图展示。A. 2.5岁时克罗恩病（CD）进展者（左图，n=15）与健康对照（右图，n=16）的扩增子序列变体（Amplicon Sequence Variant，ASV）的倍数变化。B. 5岁时克罗恩病（CD）进展者（左图，n=10）与健康对照（右图，n=13）的扩增子序列变体（ASV）的倍数变化。
图S3。免疫球蛋白A（Immunoglobulin A，IgA）测序与分析的门控策略。A. 针对克罗恩病（CD）进展者与健康对照的粪便样本，基于IgA的粪便细菌分离结合16S核糖体RNA基因测序（IgA-seq）的示意图概览。磁激活细胞分选（Magnetic-activated cell separation，MACS）。B. 从克罗恩病（CD）进展者与健康对照的粪便样本中分离IgA结合与非结合细菌的门控策略。C. 分别在2.5岁和5岁时，对比健康对照（上行组）与克罗恩病（CD）进展者（下行组）的IgA结合与非结合肠道菌群扩增子序列变体（ASV）的经验贝叶斯拟似然F检验分析。Frequency：扩增子序列变体数量；错误发现率（False Discovery Rate，FDR）。D. 对比2.5岁（上行组）与5岁（下行组）克罗恩病（CD）进展者和健康对照的IgA结合或非结合肠道菌群扩增子序列变体（ASV）的经验贝叶斯拟似然F检验分析。F. 箱线图展示2.5岁时，肠道菌群（分选前样本）丰度相似但被IgA靶向存在差异的代表性扩增子序列变体（ASV）。E. 小提琴图展示5岁时，肠道菌群（分选前样本）丰度相似但被IgA靶向存在差异的代表性扩增子序列变体（ASV）。
图S4。展示克罗恩病（CD）进展者与健康受试者IgA靶向特征的热图。A. 2.5岁时，克罗恩病（CD）进展者与健康对照的IgA阳性与IgA阴性样本间存在显著差异的前51个扩增子序列变体（ASVs=51，基于p值筛选）的相对丰度热图。B. 5岁时的上述热图。每一列代表一名受试者，每一行代表一个扩增子序列变体（ASV）。
图S5。克罗恩病（CD）进展者与克罗恩病患者的细胞因子及血浆代谢组谱。A. 5岁时采集的克罗恩病（CD）进展者（n=10）与健康对照（n=10）的血浆样本中，全部48种细胞因子的分析对比。数据以均值±标准误（Standard Error of the Mean，SEM）表示。*p<0.05，**p<0.01，***p<0.001。统计分析采用双侧非配对学生t检验。B. 小提琴图展示克罗恩病（CD）进展者与健康对照的代表性梭菌XIVa群细菌丰度（左图：IgA包被分选前；右图：IgA结合细菌组分）。
图S6。流式细胞术分析的门控策略。策略1：TCRβ阳性细胞中NK1.1与Qa-1表达的门控策略。策略2：CD8、CD4、NKG2D、CD103及NKp46的门控策略。
图S7。牛磺脱氧胆酸（Taurodeoxycholic acid，TDCA）饮食诱导不同细胞亚群的T细胞组成发生改变。A. 对照组与TDCA处理的雌性小鼠回肠组织切片的苏木精-伊红（Hematoxylin-Eosin，H&E）染色图像：完整视野（上行组）与高倍放大图像（下行组），标尺=20μm。B. 对照组与TDCA处理的雌性小鼠回肠组织切片的绒毛/隐窝比值。C. 雌性小鼠回肠组织固有层的浆细胞数量。D. TCRβ阳性细胞占总CD45阳性细胞的百分比。E. NKG2D阳性细胞占总TCRβ阳性CD45阳性细胞的百分比。F. CD103阳性细胞占总CD4阳性细胞的百分比。G. 雌性小鼠（左图）与雄性小鼠（右图）的肠上皮内淋巴细胞（Intraepithelial Lymphocytes，IELs）、派尔集合淋巴结（Peyer's Patches，PP）、固有层（Lamina Propria，LP）及脾脏中，Qa-1阳性细胞占总CD4阳性细胞的百分比。H. 采用实时定量聚合酶链反应（quantitative real-time polymerase chain reaction，qPCR）分析回肠组织中Qa-1与白细胞介素10（Interleukin 10，IL-10）的相对基因表达：TDCA处理10周后的雌性小鼠（左图）与雄性小鼠（右图）相较于对照组。数据以均值±标准误（SEM）表示。*p<0.05，**p<0.01，***p<0.001。统计分析采用双侧非配对学生t检验。