A Transcriptional Failure Leads to Systemic Inflammation
- RNA and DNA don't always interact properly.
- RNA can bind to DNA to form structures called R-loops.
- R-loops can be ejected into the cytoplasm, and become part of cytoplasmic chromatin fragments there; this is more common in senescent cells.
- This process activates an inflammatory pathway within the cell, causing the expression of factors related to senescence.
- Halting this process hides senescent cells from immune cells but reduces total inflammation.
Researchers have found that bound pieces of RNA and DNA in the cytoplasm of senescent cells encourage these cells to secrete inflammatory factors.
When transcription gets sticky
As part of the transcription process that produces necessary proteins, RNA must chemically interact with DNA. When it binds to DNA and integrates itself into the genome, it forms a three-stranded loop called an R-loop, and these are commonly found on active genes. R-loops can also form as reactions to DNA damage, allowing for RNA-mediated repair [1].
In healthy nuclei, the formation and dissolution of R-loops are tightly regulated processes that are both controlled by enzymes called helicases [2]. When these processes go wrong and R-loops stick around longer than they should, this leads to genomic instability [3]. This triggers internal inflammatory processes, which lead to chronic inflammation [4], a condition well-known in the context of age-related diseases as inflammaging.
Fewer but in the wrong place
These researchers began with a population of embryonic human lung fibroblasts (IMR90), some of which had been driven senescent through the activation of cancer-related genes, and used an established tool to map R-loops throughout the genome [5]. As R-loops are closely associated with gene transcription [6], and senescent cells transcribe fewer genes than proliferating cells [7], the researchers’ first finding was entirely expected: senescent cells have fewer R-loops than proliferating ones.
However, the senescent cells were found to have more R-loops in the cytoplasm outside the nucleus. Most of these R-loops had come from the nucleus, but others had come from the mitochondria, which have their own DNA (mtDNA). These cytoplasmic R-loops were largely derived from alpha-satellite repeats, regions that commonly form R-loops in the nuclei of senescent cells.
This export of R-loops from the nucleus to the cytoplasm is governed by two endonucleases that normally perform repair functions: XPF and XPG. Both of these endonucleases were found to be upregulated in senescent IMR90 cells, and knocking down either of them reduced the number of cytoplasmic R-loops; administering KPT-330, which inhibits this transport, achieved similar results, as did knocking down the related export protein XPO1.
A direct link to inflammation
The researchers then caused senescent IMR90 cells to express another compound that increases the lysosomal digestion of cytoplasmic R-loops, RH-NES. While it had no effect on senescence itself, RH-NES expression reduced the amount of inflammatory factors and so reduced the expression of senescence-associated secretory phenotype (SASP)-related genes, which promote inflammation in other cells when released. Reducing the number of cytoplasmic R-loops through other methods decreased SASP production as well.
This was due to these R-loops forming cytoplasmic chromatin fragments (CCFs), which have been previously documented to encourage SASP production [8]; these researchers found significant co-localization of R-loops and CCFs, which were also co-located with the DNA damage marker γH2AX. Suppressing CCF formation with trichostatin A also reduced SASP production. Further work confirmed that R-loop production and CCF formation are entirely independent processes that occur in separate parts of the cell.
This binding of R-loops and CCFs was found to be due to DDX1, an RNA helicase that is part of the DNA damage response [9]. DDX1 interacted with XPO1 in senescent cells more than in proliferating cells, and senescence caused more of it to migrate into the cytoplasm. The precise molecular residues that DDX uses to bind to XPO1 were elucidated, and variants of DDX1 that could not bind to XPO1 in this way were unable to be exported into the cytoplasm. Senescent cells with less DDX1 in the cytoplasm, as expected, produced less of the SASP. This increase in SASP activity was also found to involve the well-known DNA sensor cGAS, which was required for the co-location of R-loops into CCFs.
The SASP makes cells a target
The researchers tested the effects of DDX1 in vivo, finding both positive and negative effects. Knocking down DDX1 in senescent fibroblasts co-located with ovarian cancer cells reduced the growth of tumors when this combination was injected into mice, as the SASP encourages tumor growth.
On the other hand, knocking down DDX1 in premalignant hepatocytes injected into mice reduced the immune system’s ability to identify and destroy these cells. Immune cells use the SASP as a targeting vector, and suppressing it in this way allowed these senescent cells to evade immune clearance.
Still, treating 22-month-old female mice with KPT-330 significantly increased their lifespan. The researchers focused on the liver, as this organ’s senescence is strongly linked to SASP production. KPT-330 treatment reduced liver damage, liver fibrosis, total cholesterol, and overall inflammation in these animals.
The researchers hold that their work has created a new potential treatment for fighting systemic inflammation brought about by senescence. While this may allow senescent cells to evade immune clearance, the excessive circulating SASP is well-known to cause inflammaging and its attendant problems. Further work will determine whether or not this trade-off is truly beneficial.
Literature
[1] Petermann, E., Lan, L., & Zou, L. (2022). Sources, resolution and physiological relevance of R-loops and RNA–DNA hybrids. Nature reviews Molecular cell biology, 23(8), 521-540.
[2] Yang, S., Winstone, L., Mondal, S., & Wu, Y. (2023). Helicases in R-loop formation and resolution. Journal of Biological Chemistry, 299(11), 105307.
[3] Hamperl, S., & Cimprich, K. A. (2014). The contribution of co-transcriptional RNA: DNA hybrid structures to DNA damage and genome instability. DNA repair, 19, 84-94.
[4] Chatzidoukaki, O., Stratigi, K., Goulielmaki, E., Niotis, G., Akalestou-Clocher, A., Gkirtzimanaki, K., … & Garinis, G. A. (2021). R-loops trigger the release of cytoplasmic ssDNAs leading to chronic inflammation upon DNA damage. Science advances, 7(47), eabj5769.
[5] Yan, Q., Shields, E. J., Bonasio, R., & Sarma, K. (2019). Mapping native R-loops genome-wide using a targeted nuclease approach. Cell reports, 29(5), 1369-1380.
[6] Niehrs, C., & Luke, B. (2020). Regulatory R-loops as facilitators of gene expression and genome stability. Nature reviews Molecular cell biology, 21(3), 167-178.
[7] Papantonis, A., Antebi, A., Partridge, L., & Beyer, A. (2024). Age-associated changes in transcriptional elongation and their effects on homeostasis. Trends in Cell Biology.
[8] Dou, Z., Ghosh, K., Vizioli, M. G., Zhu, J., Sen, P., Wangensteen, K. J., … & Berger, S. L. (2017). Cytoplasmic chromatin triggers inflammation in senescence and cancer. Nature, 550(7676), 402-406.
[9] Li, L., Germain, D. R., Poon, H. Y., Hildebrandt, M. R., Monckton, E. A., McDonald, D., … & Godbout, R. (2016). DEAD Box 1 facilitates removal of RNA and homologous recombination at DNA double-strand breaks. Molecular and cellular biology, 36(22), 2794-2810.
About the author
Josh Conway
