20th-banner-rev.png

Albany 2019: 20th Conversation - Abstracts

category image Albany 2019
Conversation 20
June 11-15 2019
Adenine Press (2019)

Can R-loops themselves be the mysterious R-lesions?

All free-living organisms have two kinds of RNase H enzymes (activities that remove the RNA nucleotides from the RNA:DNA hybrid duplexes): "type I" attacks R-loops and ≥4nt long R-tracts, whereas "type II" attacks 5'RNA-DNA3' junctions, mostly at single RNA nucleotides in DNA. Type II mutants (rnhB in E. coli) behave like WT, suggesting that single rNs in DNA are innocuous, while type I mutants (rnhA in E. coli) are inhibited, suggesting that not only R-loops form frequently in the chromosomal DNA, but that they are also quite stable and, unless actively hydrolyzed, interfere with replication and transcription. Surprisingly, the combined type I/II RNase H defect reveals synergy: the double rnhAB mutant E. coli barely grows and forms long filamentous cells that have problems with nucleoid segregation. We have argued that the chromosomal problems of the double rnhAB mutant reflect formation of R-lesions: RNA-containing DNA lesions that interfere with chromosomal replication and segregation.

Since the only common substrate for the two bacterial RNase H types is R-tract, we propose that the double rnhAB mutants accumulate R-tracts in their DNA that eventually lead to irreparable double-strand breaks. This idea is consistent with the following phenotypes of the double rnhAB mutants over the single rnhA or rnhB mutants: 1) increased SOS response; 2) dependence on recombinational repair; 3) formation of double-strand breaks in the chromosomal DNA; 4) sensitivity to inhibition of translation (should induce R-loops); 5) sensitivity to inhibition of chromosomal replication (should preserve R-loops); 6) RNase-treatment-induced fragmentation of the chromosomal DNA. The emerging mechanism envisions that in the absence of RNase H, a fraction of stabilized R-loops is transformed into R-tracts, that are converted into R-gaps by replication, the latter being attacked by cellular RNases, yielding double-strand DNA breaks.

However, studying one peculiar phenotype of the rnhAB mutants shows how an R-loop could itself be an integral part of an R-lesion. The rnhAB mutants are extremely UV-sensitive, but at the same time are not generally sensitive to other forms of DNA damage. We show that in UV-ed rnhAB mutants: 1) while no additional double-strand breaks are formed, post-UV replication recovery is slow and incomplete, suggesting replication fork impediments; 2) it is inhibition of transcription by pyrimidine dimers (PDs) in DNA, rather than inhibition of translation by PDs in mRNA, that is responsible for the sensitivity; 3) promotion/stabilization of R-loops exacerbates the sensitivity, while their destabilization reduces the sensitivity; 4) PDs are rapidly removed from the bulk DNA, yet ~10% of PDs in the chromosome are removed slowly; 5) RNA:DNA hybrids accumulate massively in genomic DNA after UV. We are currently testing whether the chromosomal regions containing UV-induced RNA:DNA hybrids are enriched for slowly-removable PDs. If yes, then UV-induced R-lesions could be RNA polymerases stalled at PDs in template DNA, while anchored on their 5'-side by R-loops, an arrangement that should block PD removal.

This research is supported by grant GM 073115 from the National Institutes of Health.

References

Kouzminova EA, Kadyrov FF and Kuzminov A. (2017) RNase HII saves rnhA mutant Escherichia coli from R-loop-associated chromosomal fragmentation. J. Mol. Biol. 429: 2873-2894. PMID: 28821455

Elena A. Kouzminova
Andrei Kuzminov

Department of Microbiology
University of Illinois at Urbana-Champaign
B103 CLSL
601 South Goodwin Avenue
Urbana, IL 61801

kuzminov@illinois.edu