RESUMO
Obesity has been associated with dysbiosis, but innate mechanisms linking intestinal epithelial cell subsets and obesity remain poorly understood. Using mice lacking Paneth cells (Sox9 ΔIEC mice), small intestinal epithelial cells specialized in the production of antimicrobial products and cytokines, we show that dysbiosis alone does not induce obesity or metabolic disorders. Loss of Paneth cells reduced ILC3 and increased ILC2 numbers in the intestinal lamina propria. High-fat diet (HFD) induced higher weight gain and more severe metabolic disorders in Sox9 ΔIEC mice. Further, HFD enhances the number of ILC1 in the intestinal lamina propria of Sox9 ΔIEC mice and increases intestinal permeability and the accumulation of immune cells (inflammatory macrophages and T cells, and B cells) in abdominal fat tissues of obese Sox9 ΔIEC . Transplantation of fecal materials from Sox9 ΔIEC mice in germ-free mice before HFD further confirmed the regulatory role of Paneth cells for gut ILC subsets and the development of obesity.
RESUMO
Single-stranded DNA (ssDNA) templates along with Cas9 have been used for gene insertion but suffer from low efficiency. Here, we show that ssDNA with chemical modifications in 10-17% of internal bases (eDNA) is compatible with the homologous recombination machinery. Moreover, eDNA templates improve gene insertion by 2-3 fold compared to unmodified and end-modified ssDNA in airway basal stem cells (ABCs), hematopoietic stem and progenitor cells (HSPCs), T-cells and endothelial cells. Over 50% of alleles showed gene insertion in three clinically relevant loci (CFTR, HBB, and CCR5) in ABCs using eDNA and up to 70% of alleles showed gene insertion in the HBB locus in HSPCs. This level of correction is therapeutically relevant and is comparable to adeno-associated virus-based templates. Knocking out TREX1 nuclease improved gene insertion using unmodified ssDNA but not eDNA suggesting that chemical modifications inhibit TREX1. This approach can be used for therapeutic applications and biological modeling.