The group of Hana Cahova in Prague has discovered a new group of 5‘-RNA caps in bacteria. They are using this newly acquired knowledge to seek a better understanding of the molecular processes connected to RNA, the naturally ubiquitous dinucleoside polyphosphates, and stress in bacteria and higher organisms.
At present, there are more than 170 known modifications of RNA. We still do not understand what role many of them play. The least explored RNA modifications are 5’-RNA caps. A typical example is the m7Gp3Nm cap (cap I) found in eukaryotes.1) This chemical structure is essential for eukaryotic translation and is recognised by translational factors.
For a long time, it was believed that caps of this type exist only in eukaryotic organisms. In 2009, two new structures – nicotinamide adenine dinucleotide polyphosphate (NAD)2) and coenzyme A (CoA)3) – were detected as 5’ non-canonical caps in bacteria by LC-MS.
While many biologists were surprised by the presence of protein cofactors in RNA, chemists were excited: Both cofactors have, in principle, an RNA structure (adenosine, ribose, phosphates).
The essential step towards gaining a better understanding of the role of these caps was the development of a capturing technique enabling the sequencing of capped RNA. So far, only the method for NAD-RNA sequencing had been developed.4) Thanks to that, we know that NAD-RNA is present in all types of cells including bacteria,4,5) humans,6) plants7) and Archaea8). Nevertheless, the role of the NAD-RNA cap is still unclear.
More non-canonical RNA caps
In my laboratory, we envisaged that NAD and CoA are not the only non-canonical RNA caps in bacteria and other organisms. The naturally present dinucleoside polyphosphates (NpnNs), which were discovered more than 50 years ago in almost all types of cells,9) have a chemical structure similar to the canonical eukaryotic RNA cap.
NpnNs contain two nucleosides (usually adenosine or guanosine) that are interconnected by a 5’,5’- polyphosphate bridge with a length of two to six phosphates. These molecules are often called alarmones, as their concentrations increase under stress.10)
We proposed that these molecules might be incorporated into RNA by RNA polymerases during transcription, where they can serve as first non-canonical initiating nucleotides. Testing of various RNA polymerases showed that NpnNs are good substrates and that RNA is easily capped with them.
Free NpnNs are usually cleaved by enzymes from the NudiX family (with NudiX = Nucleoside diphosphate linked to another moiety X). This family is ubiquitous. For instance, Escherichia coli (E. coli) encodes 13 genes for NudiX enzymes and humans 24 NudiX genes. Therefore, we tested whether some E. coli NudiX enzymes can cleave NpnN-capped RNA. Besides a NudiX enzyme, RppH, we identified another enzyme, ApaH (diadenosine tetraphosphatase), that can efficiently cleave all NpnN-capped RNAs in vitro.
NpnN-capped RNAs in vivo?
All these findings led to the hypothesis that NpnN-capped RNAs also exist in vivo. Therefore, we isolated a fraction of short RNA from the model organism E. coli. We digested it into the form of nucleotides, and we established an LC-MS technique to identify dinucleoside polyphosphates. We found nine new RNA caps with the structure of dinucleoside polyphosphates. Some of the detected NpnNs were mono- or dimethylated.
As the concentration of free NpnNs in the cell increases under stress, we also observed a higher amount of NpnNs RNA caps in RNA isolated from late stationary phase in comparison with exponential phase.
This observation led us to the theory that the presence of NpnN RNA caps may protect RNA from degradation under stress conditions and that this could be the main role of NpnNs in the cell. In particular, methylated RNA caps may be more resistant to cleavage by RNA decapping enzymes. Thus, we tested whether RppH and ApaH can also cleave methylated NpnN capped RNAs:
While ApaH cleaves all the methylated NpnN RNA caps, RppH cleaves only unmethylated ones. On the basis of the bacterial transcription profile, we proposed that under normal conditions (sufficient nutrients) the cell produces RNA with various NpnN RNA caps that are cleaved by RppH, and that the RNA degradation is triggered when the cell does not need particular RNAs any more.
The cell flexibly reacts to changes of the environment and constantly produces required RNA. When the cell starts to starve, some unknown methyltransferases are expressed and methylate NpnN-capped RNAs. Thus, methylated NpnN-RNAs cannot be decapped by RppH and further degraded – so the cell can preserve the current pool of RNA-macromolecules, as it does not have enough building blocks to produce new RNA. Once the situation of the cell improves, ApaH is expressed and the old methylated NpnN-RNAs are decapped and further degraded.11)
How NpnNs are incorporated in RNA
To understand the mechanism of NpnNs incorporation in RNA by RNA polymerase, we also studied the production of capped RNA using HPLC and molecular dynamic simulations:12) Particularly the addition of ApnGs to in vitro transcription with T7 RNA polymerase can lead to ten times higher production of ApnG-RNA than a reaction without added NpnNs.
Molecular dynamic simulations showed that the adenosine moiety can non-canonically pair with the thymine in –1 position of the DNA template. Particularly, the thymine turns around its axis and pairs with N3 and O2 of the adenosine moiety from ApnG. This shows: free NpnNs can lead to the preferential production of NpnN-capped RNAs by an RNA polymerase once they are present in a higher local concentration in its proximity.
As NpnN-RNA caps have the potential to pair with template DNA, it is possible that they also contribute to the binding of, e.g., regulatory RNAs to its target RNA or protein. Furthermore, they may somehow regulate the translation of mRNA (even though translation is cap-independent in bacteria) or navigate RNA into various cellular loci.
Free Ap4A and Ap3A were also detected in higher organisms, including within human cells, in quite high (micromolecular) concentrations. Their increased concentration was also connected with some stress or immune stimuli, and they were detected mainly in the nucleus.13) Therefore, the existence of NpnN-capped RNAs is presumably not limited to bacteria, and these types of RNA may also exist in eukaryotes.
Nevertheless, without understanding what types of RNA bear these non-canonical RNA caps, we cannot focus on the studies of their biological roles. Therefore, our first and most challenging task is to develop selective capturing or profiling techniques, which will make it possible to identify RNAs bearing NpnNs.
Since NpnNs have a similar chemical structure to canonical RNA building blocks, the application of classical chemical reactions is limited. Therefore, specific enzymes have to be identified that selectively interact with, cleave, or modify only particular NpnN-capped RNAs. Once we can selectively capture NpnN-capped RNAs from cellular isolated RNA, we will sequence it using either direct nanopore sequencing or deep sequencing with the Illumina sequencing platform.
Our studies have revealed a new class of 5’-RNA caps in bacteria and opened new questions, for example:
Why are there so many types of non-canonical RNA caps in bacteria?What types of RNA (tRNA, mRNA, regulatory sRNA, etc.) are capped with NpnNs?Does NpnN-RNA also exist in higher organisms?What is the main function of non-canonical RNA caps?
Answering these questions will help to understand the molecular processes connected to RNA, NpnNs, and stress in bacteria and higher organisms.
Three questions to the backer of the author: Ullrich Jahn
What‘s the difference between research in Germany and the Czech Republic?
In principle not much, the structure is similar. The quality of research has improved significantly and has become competitive with leading European nations.
Why did you suggest Hana for an article?
I have been knowing Hana from her time as a PhD student and am impressed by the way she has taken up to now.
What‘s your relationship with Prague?
I’ve been living in Prague for 15 years.
Ullrich Jahn is a senior research group leader at IOCB Prague.
Hana Cahova is head of Junior Research Group of Chemical Biology of Nucleic Acids at IOCB Prague since 2016. She received PhD at UCT Prague and IOCB Prague in organic chemistry. In 2011, she was awarded Alexander von Humboldt fellowship and joined the laboratory of Andres Jäschke at Heidelberg University. Currently, she and her team focus on understanding the role of RNA modifications in viruses and other model organisms. In 2022, she received ERC Starting grant for the project StressRNaction.
1 J. M. Adams, S. Cory, Nature 1975, 255, 28–33
2 Y. G. Chen, W. E. Kowtoniuk, I. Agarwal, Y. Shen, D. R. Liu, Nat. Chem. Biol. 2009, 5, 879
3 W. E. Kowtoniuk, Y. Shen, J. M. Heemstra, I. Agarwal, D. R. Liu, Proc. Natl. Acad. Sci. 2009, 106, 7768–7773
4 H. Cahová, M.-L. Winz, K. Höfer, G. Nübel, A. Jäschke, Nature 2014, 519, 374
5 H. G. Morales-Filloy, Y. Zhang, G. Nübel, S. E. George, N. Korn, C. Wolz, A. Jäschke, J. Bacteriol. 2019, JB.00591–00519
6 X. Jiao, S. K. Doamekpor, J. G. Bird, B. E. Nickels, L. Tong, R. P. Hart, M. Kiledjian, Cell 2017, 168, 1015–1027.e1010
7 Y. Wang, S. Li, Y. Zhao, C. You, B. Le, Z. Gong, B. Mo, Y. Xia, X. Chen, Proc. Natl. Acad. Sci. USA 2019, 116, 12094–12102
8 O. Ruiz-Larrabeiti, R. Benoni, V. Zemlianski, N. Hanišáková, M. Schwarz, B. Brezovská, B. Benoni, J. Hnilicová, V. R. Kaberdin, H. Cahová, M. Vítězová, M. Převorovský, L. Krásný, bioRxiv 2021, 2021.2012.2014.472595
9 P. G. Zamecnik, M. L. Stephenson, C. M. Janeway, K. Randerath, Biochem. Biophys. Res. Co. 1966, 24, 91–97
10 R. A. VanBogelen, P. M. Kelley, F. C. Neidhardt, J. Bacteriol. 1987, 169, 26–32
11 O. Hudeček, R. Benoni, P. E. Reyes-Gutierrez, M. Culka, H. Šanderová, M. Hubálek, L. Rulíšek, J. Cvačka, L. Krásný, H. Cahová, Nat. Commun. 2020, 11, 1052
12 R. Benoni, M. Culka, O. Hudeček, L. Gahurova, H. Cahová, ACS Chem. Biol. 2020, 15, 1765–1772
13 Y.-N. Lee, H. Nechushtan, N. Figov, E. Razin, Immunity 2004, 20, 145–151