Torben Heick Jensen


The regulation and fidelity of gene expression is of paramount importance for the maintenance and differentiation of all living organisms. Our laboratory studies the production and quality control of RNA in eukaryotic cells (S. cerevisiae, mouse and human) and its contribution to gene expression regulation. A main focus of the laboratory is to understand the molecular principles dictating the sorting of newly transcribed RNA into a productive pathway involving its packaging with protein and cellular transport vs. a destructive pathway leading to RNA turnover. We believe that a thorough understanding of these relationships will also position us to better understand any putative function of the pervasive transcription of eukaryotic genomes.

Laboratory efforts can roughly be divided into four research topics:

From 2005-2015 Torben Heick Jensen was heading the Danish National Research Foundation-funded ‘Centre for mRNP Biogenesis and Metabolism’. These, and other, efforts are now continued via funding from the European Research Council (ERC), the Danish Council for Independent Research, the Novo Nordisk foundation, the Lundbeck Foundation and the Danish Cancer Society. The THJ laboratory is also part of the iSEQ Centre for Integrative Sequencing (

Nuclear Human RNA Decay

One outcome of our efforts is the realization that transcription initiation from human promoters largely occurs in a divergent fashion. Reverse-oriented RNAs, so-called PROMoter uPstream Transcripts (PROMPTs), are inherently unstable, possibly due to their short length and early transcription termination by promoter-proximal polyadenylation (pA) sites. For a further read please check out: Preker et al. Science 2008; Ntini et al. Nat. Struct. Mol. Biol. 2013; Andersen et al. Nat. Struct. Mol. Biol. 2013 ; Andersson et al. Mol Cell 2015.

Our laboratory is actively engaged in characterizing RNA decay pathways within human nuclei. These efforts involve the identification of new pathways components and their substrates. To this end, we use state-of-the-art affinity capture / mass spectrometry approaches and high throughput transcriptome-wide methodologies. However, decay of RNA is by no means just a ‘clean-up-act’ to remove worn out molecules. In addition to its essential role in the quality control of genome expression, RNA turnover is also at the core of gene expression regulation – forming intricate connections to RNA productive systems – thus, being centrally placed to determine RNA/RNP fate. Hence, we also work towards establishing models for how regulated RNA turnover helps control key biological processes.

Classifying non-coding RNA

Using HeLa cells as a model system, we have previously devised an approach to profile the susceptibility of lncRNA to rapid decay. Transcribed ENCODE HeLa promoters were grouped into five major classes via k-medoids clustering based upon exosome sensitivity, expression levels and transcriptional strand bias (directionality) and visualized by principal component analysis (PCA). For a further read please check out: Andersson et al. Nature Communications 2014.

Our laboratory aims to identify physiologically relevant long non-coding RNAs (lncRNAs) that are stably expressed in cells, and more generally deduce rules predicting the stability and half-life of any RNA. This is for example achieved by depleting degradation enzymes followed by obtaining various high throughput RNA sequencing data sets, which are analyzed using computational classification methods. In this way, we hope to pinpoint functional lncRNAs from the sea of pervasive, spurious and unstable transcripts that our cells constantly produce. With our work we also aim to expose when activities, other than the lncRNA itself (for example regulatory transcription events), are functionally relevant. In this way, our studies should reveal new physiologically meaningful RNAs and uncover new gene regulatory concepts created by the pervasive transcription of mammalian genomes.

Nonsense-Mediated Decay

snoRNA host genes are highly transcribed thereby producing large amounts of pre-mRNAs that are processed to spliced mRNAs and snoRNAs. snoRNA host genes are more prone to produce NMD-susceptible mRNAs than normal protein-coding genes. One interpretation of this is that snoRNA host genes can switch between the production of protein-coding mRNA and non-coding NMD-susceptible RNA by e.g. alternative splicing to adjust protein expression levels while maintaining an unaltered high level of snoRNA production.?

Our laboratory is studying the cytoplasmic RNA degradation pathway nonsense-mediated mRNA decay in human cells (NMD; Lykke-Andersen and Jensen, 2015). We are interested in both mechanistic aspects of NMD as well as its role in the general control of eukaryotic gene expression. In particular, we are studying the function of the endonuclease SMG6 and its interplay with other nucleases in NMD (Eberle et al., 2009; Lykke-Andersen et al., 2014). Additionally, we have recently taken an interest in the relationship between NMD and snoRNA host genes. These highly expressed genes encode mRNAs or mRNA-like transcripts from their exons and snoRNAs from one or more introns. Because maturation of intron-hosted snoRNAs generally depends on the splicing process, the spliced RNA can be considered as a by-product of snoRNA production. In many cases this ‘by-product’ acts as a normal mRNA and encodes functional protein. However, compared to general protein-coding genes, snoRNA host genes have a much more pronounced tendency to give rise to NMD-susceptible mRNAs (Lykke-Andersen et al., 2014). There are several possible explanations for this, which we are actively investigating. In combination, our studies of the mechanism and global impact of NMD may reveal novel insights about the regulation of eukaryotic gene expression.

RNA Synthesis and Decay in S. cerevisiae

Model based on (Schmid et al., Cell Reports 2015), depicting how the nuclear poly(A) binding protein Nab2p impacts mRNA production. In the presence of Nab2p in the nucleus, the protein binds and protects mRNAs, which can then get exported to the cytoplasm for translation (left panel). mRNAs that are not protected by Nab2p get degraded in the nucleus by the exonucleases Rrp6p and Dis3p (right panel).

In our lab we use the budding yeast Saccharomyces cerevisiae to study RNA metabolism in the cell nucleus and focus on the relationships between RNP synthesis, decay and export. To identify relevant substrates of key players, we use rapid nuclear protein depletion schemes (‘anchor-away’), coupled with genome-wide methodologies for measuring transcription and RNA synthesis as well as RNA fluorescense in-situ hybridization. Nuclear RNA metabolism is not only an intricate component of the gene expression machinery but also provides opportunities for quality control and regulation of transcript levels. Identification of such regulatory principles is the primary target of our yeast work. Indeed, our lab has previously identified new paradigms for gene expression control based on polyadenylation, poly(A) binding proteins and RNA 3’-5’ exonucleolytic decay (Saguez et al., 2008; Schmid et al., 2012; Schmid et al., 2015). Consequently, yeast work provides important insights in its own right, but also contributes important inspiration for projects in mammalian cells.