bioRxiv ScienceSearch

Biology subjects

Dai, N.

Publications and source records attributed to Dai, N..

2 recordsLinked to original sources

Using selective lung injury to improve murine models of spatially heterogeneous lung diseases

Many lung diseases, such as acute respiratory distress syndrome (ARDS), display significant regional heterogeneity, with patches of severely injured tissue adjacent to apparently healthy tissue. Current mouse models that aim to mimic ARDS generally produce diffuse injuries that cannot reproducibly generate ARDSs regional heterogeneity. This deficiency prevents the evaluation of how well therapeutic agents reach the most injured regions, and precludes many regenerative medicine studies, since it is not possible to know which apparently healing regions suffered severe injury initially. Finally, these diffuse injury models must be mild to allow for survival, as their diffuse nature does not allow for residual healthy lung to keep an animal alive long enough for many drug and regenerative medicine studies. To solve all of these deficiencies of current animal models, we have created a simple and reproducible technique to selectively induce lung injury in specific areas of the lung. Our technique, catheter-in-catheter selective lung injury (CICSLI), involves guiding an inner catheter to a particular area of the lung and delivering an injurious agent mixed with nanoparticles (fluorescently and/or radioactively labeled) that can be used to track the location and extent of where the initial injury was, days later. Further, we demonstrate that CICSLI can produce a more severe injury than diffuse models, yet has much higher survival since CICSLI intentionally leaves undamaged lung regions. Collectively, these attributes of CICSLI will allow better study of how drugs act within heterogeneous lung pathologies and how regeneration occurs in severely damaged lung tissue, thereby aiding the development of new therapies for ARDS and other lung diseases.

pharmacology and toxicology

Pancreatic islet chromatin accessibility and conformation defines distal enhancer networks of type 2 diabetes risk

The gene targets of enhancer activity in pancreatic islets are largely unknown, impeding discovery of islet regulatory networks involved in type 2 diabetes (T2D) risk. We mapped chromatin state, accessibility and conformation using ChIP-seq, ATAC-seq and Hi-C in human pancreatic islets, which we integrated with T2D genetic fine-mapping and islet expression QTL data. Active islet regulatory elements preferentially interacted with other active elements, often at distances over 1MB, and we identified target genes for thousands of distal islet enhancers. A third of T2D risk signals mapped in islet enhancers, and target genes regulated by these signals were specifically involved in processes related to protein transport and secretion. Among implicated target genes of T2D islet enhancer signals with no prior known role in islet function, we demonstrated that reduced IGF2BP2 activity in mouse islets leads to impaired glucose-stimulated insulin secretion. These results link distal islet enhancer regulation of protein secretion and transport to genetic risk of T2D, and highlight the utility of high-throughput chromatin conformation maps to uncover the gene regulatory networks of complex disease.

genomics