ChIP-seq of HTT protein in mouse cortex following HTT knockdown

Progressive striatal gene expression changes and epigenetic alterations are a prominent feature of Huntington’s disease (HD), but direct relationships between the huntingtin (HTT) protein and chromatin remain poorly described. Here, using chromatin immunoprecipitation and sequencing (ChIP-seq), we show that HTT reproducibly occupies specific locations in the mouse genome, including thousands of genomic loci that are differentially occupied in striatal tissue from a knock-in mouse model of the HD mutation (B6.HttQ111/+) versus wildtype controls. ChIP-seq of histone modifications, generated in parallel, revealed genotype-specific colocalization of HTT with trimethylation of histone 3 lysine 27 (H3K27me3), a repressive chromatin mark. Near genes that are differentially regulated in HD, greater HTT occupancy in HttQ111/+ vs. wildtype mice predicted increased H3K27me3, reduced histone 3 lysine 4 (H3K4me3), a marker of poised and active promoters, and down-regulated gene expression. Altered huntingtin-chromatin interactions may therefore play a direct role in driving transcriptional dysregulation in HD. Four-month-old male HttQ111/+ and age-matched Htt+/+ mice underwent unilateral intracerebroventricular injection of 500ug pan-huntingtin antisense oligonucleotide (Ionis #444652) or received no treatment.  Fresh-frozen cortical tissue ipsilateral to the injection was collected for ChIP at four weeks post injection. For each ChIP preparation, chromatin DNA was prepared using ipsilateral cortical tissue from three mice. Tissue was transferred to a glass dounce on ice and homogenized in cold PBS with protease inhibitors. High-resolution X-ChIP-seq was performed as described (Skene and Henikoff, 2015), with slight modifications. IPs were performed using Abcam Anti-huntingtin antibody EPR5526 (#ab109115). ChIP-Seq library preparation and sequencing reactions were conducted at GENEWIZ, Inc. (South Plainfield, NJ, USA) using NEB NextUltra DNA Library Preparation kit following the manufacturer’s recommendations (Illumina, San Diego, CA, USA). Sequencing was performed using 2x150 Paired End (PE) configuration. Sequencing reads were aligned to the mouse genome (mm9) using HISAT2. Aligned reads from the four ChIP libraries were down-sampled to the size of the smallest sequencing library using samtools view -s. Aligned reads from the Input genomic DNA of all four samples were merged into a single control BAM file. Peak-calling was then performed with MACS2.2, scaling to the size of the input control library. The final set of reproducible peaks was obtained using the dba() function in the DiffBind R package, retaining peaks that were reproducible at FDR < 0.01 in at least two samples and excluding artifactual blacklist regions from ENCODE