Ule lab
RNA Networks Laboratory
: RNPs in disease

We study the biological functions of several RBPs, especially those that are implicated in amyotrophic lateral sclerosis (ALS) such as TDP-43, FUS and MATR3.

The cover of the October 2011 Genome Research issue, illustrated by James Tollervey, representing the topic studied by Tollervey et al paper in the issue.

Figure 1: The cover of the October 2011 Genome Research issue, illustrated by James Tollervey, representing the topic studied by Tollervey et al paper in the issue. The stars symbolize RNPs forming on transcripts within two individuals of different ages. The change in the pattern of stars indicates age-related changes in the levels of alternative transcript isoforms, as well as changes in RNPs themselves. A hallmark of many neurodegenerative diseases is the decreased dynamics of RNPs, and formation of aggregated RNPs, which are also represented by the appearance of larger star clusters.

Many neurodegenerative diseases are associated with formation of toxic aggregates of RNA-binding proteins, especially TDP-43, which result from the strong propensity of TDP-43 for molecular condensation. Moreover, mutations in the gene encoding TDP-43, as well as over a dozen other RNA-binding proteins, can cause ALS. These mutations are most common in the intrinsically disordered regions of these proteins, but we still have a limited understanding of how many of these regions contribute to RNP assembly and function. It is clear, however, that these regions are capable of forming multivalent contacts that usually promote the formation of RNP condensates.

We recently addressed the long-standing question of whether TDP-43 condensation is required for its RNA binding specificity and function. We created variants of TDP-43 with a gradient of condensation properties as evident by in vitro phase separation, and by imaging the condensates of TDP-43 in cells. We then used comparative iCLIP to find that the condensation of TDP-43 is required for its efficient binding to long RNA regions with widely dispersed binding motifs, which we refer to as ‘binding region condensates’. Thereby, TDP-43 regulates a select subset of 3’UTR isoforms, including autoregulation of TDP-43 itself. Our study thus showed that changes in TDP-43 condensation can deregulate a selective subset of RNAs, which could contribute to the early stages of neurodegenerative diseases. We are currently using human and mouse stem cells differentiated into motor neurons to study how the cell fate and activity influence the impact of mutations on the assembly, localisation and aggregation of mutant RNPs.

We have been working on TDP-43 and other ALS-relevant proteins for many years, and have examined how TDP-43, FUS and MATR3 regulate alternative splicing and 3’ end processing. We have use iCLIP to study the RNA partners of these proteins in various cell types and in brain, and have assessed how RNA binding properties of TDP-43 change in the brain tissue from patients with FTLD, which contains aggregates of TDP-43. This showed that the greatest changes in binding of TDP-43 occur on the long non-coding RNAs NEAT1 and MALAT1, which led us to a study examining cross-regulation of TDP-43 and NEAT1 during development.

We also used transcriptomic methods to study changes in RNA processing that are associated with brain aging and age-related neurodegenerative diseases. Interestingly, we found that the changes in glial-specific genes are particularly common in aging. Stratifying the gene expression by cell type, we found that genes specific for oligodendrocytes diminish their expression, whereas genes specific for endothelial cells and microglia increase expression during healthy brain aging. We also observed decreased region-specific expression of astrocyte and oligodendrocyte genes. To complement these insights from gene expression, we used a machine learning method to analyse high-resolution images of brain sections. This confirmed that the number of oligodendrocytes decreases in aging, and the neurons with largest cell bodies also appear to decline in number.