archives@tulane.edu / 1 / Anna Smith

Down syndrome (DS) is caused by the triplication of chromosome 21, but science is still investigating the precise mechanisms by which this results in the various phenotypes, such as anatomical abnormalities, intellectual deficits, and early development of Alzheimer’s disease (AD). The global changes in transcriptional activity and the altered expression of genes not transcribed from chromosome 21 point to changes to the epigenetic landscape. One of the candidate genes for this global gene dysregulation is High Mobility Group Nucleosome Binding Domain 1 (HMGN1) which is triplicated in DS. While investigating DS-associated B-cell acute lymphoblastic leukemia (B-ALL), researchers found the triplication of HMGN1 alone led to many of the same transcriptional and phenotypic changes that marked DS-associated B-cells from a mouse model with all 31 genes orthologous to human chromosome 21 genes triplicated. Amongst the pathways most affected by triplication, enrichment was greatest for targets of the Polycomb Repressive Complex 2 (PRC2) and sites of the transcriptionally repressive mark it catalyzes, H3K27me3. HMGN1 instead, promotes transcriptional activation and its overexpression leads to a global increase in RNA transcript levels. Therefore, overexpression of HMGN1 in DS may cause an increase in transcriptional activity and prevent the silencing of genes normally silenced by PRC2, with downstream effects on neurogenesis and gliogenesis, abnormal cellular migration, and deviant developmental timing that result in known DS phenotypes. With this hypothesis, we first wanted to quantify the levels of acetylation versus methylation at H3K27 in trisomy 21 induced pluripotent stem (iPS) cell-derived cellular models: neural progenitor cells (NPCs) and cortical organoids and to determine if there are measurable differences between the genotypes. We found a decrease in H3K27me3 in 130-day-old organoids, but not in NPCs. No changes were detected in the levels of H3K27ac. With the high comorbidity between DS and AD, and changes to the epigenome found in both diseases, we wondered whether there were specific alterations at H3K27 in DS-AD. To determine this, we performed an analysis of human postmortem brain tissue from individuals with DS-AD, AD, and control and quantified H3K27me3 and H3K27ac marks. Our data indicated that there are marginally significant changes in H3K27me3 that are unique to DS-AD as compared to control and AD samples. Encouraged by this data, we next measured gene expression levels of specific PRC2 target genes increased in trisomy. Our goal was to identify the causative relationship between the increased expression of HMGN1 in trisomy and upregulation of specific PRC2 target genes with known brain-related functions. We found that enhanced expression of particular PRC2 target genes in trisomic cells could be normalized with the short-hairpin RNA (shRNA)-induced knockdown of HMGN1 expression in trisomic NPCs. This implicates HMGN1 overexpression in DS in the dysregulation and overexpression of particular genes involved in morphogenesis, neurogenesis, neuronal migration, apoptosis, and cell viability through the antagonism of the PRC2 activity. We provided novel evidence for a possible mechanism for the cellular, molecular, and transcriptomic changes originating from the triplication of HMGN1 that can potentially lead to DS-related phenotypes such as intellectual disability and AD-related pathology. Furthermore, our findings suggest a possible therapeutic avenue to mitigate these phenotypes by regulating HMGN1 expression. Taken together, our work is the first to causatively connect HMGN1-induced epigenetic changes to DS-related brain cell phenotypes and to point out to a potential approach for correcting them. .