Jiang Kaiyu, Fu Yao, Kelly Jennifer A, Gaffney Patrick M, Holmes Lucy C, Jarvis James N
Department of Pediatrics, University of Buffalo Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY, USA.
Genes and Human Disease Research Program, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA.
bioRxiv. 2023 Sep 14:2023.09.11.557171. doi: 10.1101/2023.09.11.557171.
BACKGROUND/PURPOSE: Knowledge of the 3D genome is essential to elucidate genetic mechanisms driving autoimmune diseases. The 3D genome is distinct for each cell type, and it is uncertain whether cell lines faithfully recapitulate the 3D architecture of primary human cells or whether developmental aspects of the pediatric immune system require use of pediatric samples. We undertook a systematic analysis of B cells and B cell lines to compare 3D genomic features encompassing risk loci for juvenile idiopathic arthritis (JIA), systemic lupus (SLE), and type 1 diabetes (T1D).
We isolated B cells from healthy individuals, ages 9-17. HiChIP was performed using CTCF antibody, and CTCF peaks were identified. CTCF loops within the pediatric were compared to three datasets: 1) self-called CTCF consensus peaks called within the pediatric samples, 2) ENCODE's publicly available GM12878 CTCF ChIP-seq peaks, and 3) ENCODE's primary B cell CTCF ChIPseq peaks from two adult females. Differential looping was assessed within the pediatric samples and each of the three peak datasets.
The number of consensus peaks called in the pediatric samples was similar to that identified in ENCODE's GM12878 and primary B cell datasets. We observed <1% of loops that demonstrated significantly differential looping between peaks called within the pediatric samples themselves and when called using ENCODE GM12878 peaks . Significant looping differences were even less when comparing loops of the pediatric called peaks to those of the ENCODE primary B cell peaks. When querying loops found in juvenile idiopathic arthritis, type 1 diabetes, or systemic lupus erythematosus risk haplotypes, we observed significant differences in only 2.2%, 1.0%, and 1.3% loops, respectively, when comparing peaks called within the pediatric samples and ENCODE GM12878 dataset. The differences were even less apparent when comparing loops called with the pediatric vs ENCODE adult primary B cell peak datasets.The 3D chromatin architecture in B cells is similar across pediatric, adult, and EBVtransformed cell lines. This conservation of 3D structure includes regions encompassing autoimmune risk haplotypes.
Thus, even for pediatric autoimmune diseases, publicly available adult B cell and cell line datasets may be sufficient for assessing effects exerted in the 3D genomic space.
背景/目的：了解三维基因组对于阐明驱动自身免疫性疾病的遗传机制至关重要。三维基因组在每种细胞类型中都是独特的，目前尚不确定细胞系是否能忠实地重现原代人类细胞的三维结构，也不确定儿科免疫系统的发育方面是否需要使用儿科样本。我们对B细胞和B细胞系进行了系统分析，以比较包含青少年特发性关节炎（JIA）、系统性红斑狼疮（SLE）和1型糖尿病（T1D）风险位点的三维基因组特征。
我们从9至17岁的健康个体中分离出B细胞。使用CTCF抗体进行HiChIP，并鉴定CTCF峰。将儿科样本中的CTCF环与三个数据集进行比较：1）在儿科样本中自行调用的CTCF共有峰，2）ENCODE公开可用的GM12878 CTCF ChIP-seq峰，以及3）来自两名成年女性的ENCODE原代B细胞CTCF ChIPseq峰。在儿科样本和三个峰数据集中的每一个中评估差异环化。
儿科样本中调用的共有峰数量与ENCODE的GM12878和原代B细胞数据集中鉴定的数量相似。我们观察到，在儿科样本自身调用的峰与使用ENCODE GM12878峰调用的峰之间，显示出显著差异环化的环不到1%。将儿科调用峰的环与ENCODE原代B细胞峰的环进行比较时，显著的环化差异甚至更小。当查询在青少年特发性关节炎、1型糖尿病或系统性红斑狼疮风险单倍型中发现的环时，在比较儿科样本中调用的峰与ENCODE GM12878数据集时，我们分别仅在2.2%、1.0%和1.3%的环中观察到显著差异。当比较儿科与ENCODE成年原代B细胞峰数据集调用的环时，差异甚至不太明显。B细胞中的三维染色质结构在儿科、成人和EBV转化的细胞系中相似。这种三维结构的保守性包括包含自身免疫风险单倍型的区域。
因此，即使对于儿科自身免疫性疾病，公开可用的成人B细胞和细胞系数据集可能足以评估在三维基因组空间中发挥的作用。
