Andreas Beyer / Argyris Papantonis - C 4

Contribution of deteriorating RNA biosynthesis to cellular ageing


Using transcriptome data from 5 animal species we have shown that the speed of RNA polymerase II (Pol-2) elongation increases with ageing. In this project we plan to further analyse this phenomenon and to better elucidate molecular mechanisms. Using additional molecular profiling data and integration with published data we are going to elucidate the mechanisms underlying the Pol-2 speed changes. 


Multiple cellular processes have been implicated in age-associated degeneration of tissues, such as cellular senescence, depletion of stem cells, or declining cell-cell interactions. However, we know surprisingly little about the molecular causes leading to these changes. Whereas molecular pathways responsible for some of these changes are well studied (such as DNA repair, TOR signalling, apoptosis, and others), we lack sufficient understanding of the factors impacting on or altering these pathways. The advent of high-throughput RNA quantification techniques (especially RNA-sequencing, RNA-seq) has greatly contributed to our understanding of the molecular mechanisms underlying ageing and age-associated tissue degeneration. However, previous research has almost exclusively focused on the identification of genes that are differentially expressed in response to aging or lifespan-extending conditions. The goal of this previous research has been to identify transcriptional programs that either trigger ageing or which are activated in response to ageing. Little attention has been payed to the process of transcription itself and its role in ageing-associated decline of cellular function and tissue integrity. We recently showed that the quality of transcription itself declines during aging, which has substantial impact on the quality of the transcripts generated from e.g. RNA polymerase II (Pol-2). By analysing transcriptome data from 5 species (C. elegans, D. melanogaster, M. musculus, R. novegicus, H. sapiens) we showed that the elongation speed of Pol-2 increases with age, thereby reducing splicing efficiency. Although the absolute effects were subtle, they affected hundreds of genes and they were strikingly consistent across all of the species studied. Importantly, we could show that lifespan-extending conditions such as dietary restriction or inhibition of insulin signalling reduced Pol-2 elongation speed and improved splicing efficiency. We hypothesize that these changes (which affected particularly regulatory proteins such as transcriptional regulators) have widespread impact on tissue integrity and therefore contribute to age-associated phenotypes.

Nucleosome profiling

We started to investigate the link between nucleosome positioning and changes in Pol-2 elongation speed and to quantify and analyse the variability of elongation speed and splicing efficiency between single cells. This will consolidate our hypothesis and help to better understand how changes in RNA biosynthesis contribute to the loss of tissue integrity during ageing and disease. 

Nucleosome positioning along the DNA fiber affects both Pol-II elongation rate and splicing efficiency. Furthermore, older eukaryotic cells display reduced nucleosomal density in chromatin and “noisier” core nucleosome positioning. Thus, age-associated changes in chromatin structure might contribute to the changes in Pol-II elongation speed and splicing efficiency. To probe this hypothesis, we performed micrococcal nuclease  (MNase) digestion of chromatin from early (proliferating) and late-passage (senescent) human IMR90 cells followed by ~400 million paired-end read sequencing (required to make accurate estimates of nucleosome positioning) of the mononucleosomal DNA fragments (MNase-seq). Following mapping, we explored changes in nucleosome occupancy and positioning over exons and exon-intron boundaries. We observed that nucleosome peaks moved away from exon centers and from exon-intron boundaries in senescent versus proliferating cells (Fig. 1).

Single cell transcriptome profiling

Profiling Pol-II at the level of single cells is challenging. Our Pol-II elongation speed measure relies on the detection of a sufficient number of reads from intronic regions in nascent (unspliced) transcripts. Standard single cell sequencing protocols rely on poly-A enrichment of mature transcripts, which precludes nascent transcripts. We therefore devised a different protocol: our protocol extracts single nuclei, rather than single whole cells. Subsequently, all transcripts in the nucleus are poly-A labelled, i.e. even immature nascent transcripts can subsequently be processed with standard protocols. We have performed a first test of this protocol using proliferating (‘young’) and senescent (‘old’) HUVEC cells as a model. After filtering for quality we retrieved sufficient data for 186 young cells and 139 senescent cells.

After processing the data with ZINB-WaVE we obtained a clear clustering distinguishing ‘old’ from ‘young’ cells, which underlines the quality and utility of the data (Figure 2). Further, we could confirm that on average, the Pol-II speed was higher in the senescent cells compared to the proliferating cells. Currently, we are in the process of developing analysis tools to quantify Pol-II speed at the level of single cells. The complication in this case is that any given gene is only transcribed by zero, one or very few polymerases, which prohibits the use of established methods that are based on computing a slope over the read distribution in introns. We are therefore considering to combine information from multiple introns across genes.

Selected publications

Zirkel A, […], Papantonis A: HMGB2 Loss upon Senescence Entry Disrupts Genomic Organization and Induces CTCF Clustering across Cell Types. Molecular Cell (2018) 70(4):730-744.

Liu Y*, Beyer A*#, Aebersold R#: On the dependency of cellular protein levels on mRNA abundance (Review). Cell (2016) 165(3):535-50. * equal contribution. # corresponding author.

Valenzano D, […], Beyer A, Johnson EA, Brunet A: The African turquoise killifish genome provides insights into evolution and genetic architecture of lifespan. Cell (2015) 163(6):1539-54.

Prof. Dr. Andreas Beyer

CECAD Cologne / RG location

Prof. Dr. Andreas Beyer

Principal Investigator C 4

Publications - Andreas Beyer

Link to PubMed

Prof. Dr. Argyris Papantonis

Center for Molecular Medicine Cologne

Prof. Dr. Argyris Papantonis

Co-Principal Investigator C 4 /
Principal Investigator CMMC JRG VIII

Publications - Argyris Papantonis

Link to PubMed

Group Members

Antonios Papadakis (doctoral student)

Figure 1

CMMC Research Beyer
Nucleosome profiling of fetal lung cells (IMR90) at early (proliferating; grey) and late (senescent; blue) passages. In senescent cells nucleosomes are positioned broader relative to exon-intron junctions, suggesting noisier chromatin structure that could contribute to Pol-II elongation speed changes.

Figure 2

CMMC Research Beyer
Single-cell transcriptome analysis. Clustering of single cells based on expression similarity using ZINB-WaVE. Proliferating (‘young’) and senescent (‘old’) cells are clearly separated.