Response to: The Schizosaccharomyces pombe imprint — nick or ribonucleotide(s)?

Abstract

Vengrova and Dalgaard believe our observations are better explained by their ribonucleotide(s) replacement model [1xRNase-sensitive DNA modification(s) initiates S. pombe mating-type switching. Vengrova, S. and Dalgaard, J.Z. Genes Dev. 2004; 18: 794–804Crossref | PubMed | Scopus (66)See all References1] rather than by the dephosphorylated nick model we proposed [2xA programmed strand-specific and modified nick in S. pombe constitutes a novel type of chromosomal imprint. Kaykov, A. and Arcangioli, B. Curr. Biol. 2004; 14: R915–R917Abstract | Full Text | Full Text PDF | PubMed | Scopus (30)See all References2]. Here we wish to review the two models in the context of previously published work and discuss the observations made in the accompanying letter.Recently, Vengrova and Dalgaard stated that the imprint can be purified intact, or cleaved in some conditions. We believe it may be purified as cleaved in all conditions, since we were never able to isolate intact imprinted mat1 upper strand. Furthermore, they propose that the imprint is composed of either one or two ribonucleotides, and that starting from an intact or nicked molecule, treatments with RNase T2 or NaOH remove one ribonucleotide from the 5′-end mat1-distal upper strand, converting the imprint into a gapped molecule [1xRNase-sensitive DNA modification(s) initiates S. pombe mating-type switching. Vengrova, S. and Dalgaard, J.Z. Genes Dev. 2004; 18: 794–804Crossref | PubMed | Scopus (66)See all References1].Our recent work [2xA programmed strand-specific and modified nick in S. pombe constitutes a novel type of chromosomal imprint. Kaykov, A. and Arcangioli, B. Curr. Biol. 2004; 14: R915–R917Abstract | Full Text | Full Text PDF | PubMed | Scopus (30)See all References2] was designed to further analyze the molecular nature of the imprint and to directly challenge the nick and RNA models. A PstI site was inserted at mat1, and shows that a nick is located at a precise and fixed position, as observed in the wild-type strain, and is independent of RNase T2 treatment (yielding no gaps), which is incompatible with the presence of a 5′-end ribonucleotide(s). In addition, the potential caveat that our engineered strain containing PstI affects the number of nucleotides modified in a sequence-specific manner is not supported by the PstI series of mutant strains, which allows us to position the nick next to any of the four bases (stated as data not shown in [2xA programmed strand-specific and modified nick in S. pombe constitutes a novel type of chromosomal imprint. Kaykov, A. and Arcangioli, B. Curr. Biol. 2004; 14: R915–R917Abstract | Full Text | Full Text PDF | PubMed | Scopus (30)See all References2]). Altogether, our data support the simplest and most economical model, that the imprint at mat1 is a nick containing 3′OH and 5′OH termini and is resistant to RNase T2 treatments.Here, Vengrova and Dalgaard attempt a unifying explanation. They suggest that two populations of imprint exist, containing either one or two ribonucleotides, and propose that hydrolysis of a putative DNA–RNA–DNA hybrid molecule containing a single ribonucleotide would yield a nicked molecule in which the ribonucleotide will remain attached to the 3’ end of the nick. In principle, this new interpretation could allow our set of data [2xA programmed strand-specific and modified nick in S. pombe constitutes a novel type of chromosomal imprint. Kaykov, A. and Arcangioli, B. Curr. Biol. 2004; 14: R915–R917Abstract | Full Text | Full Text PDF | PubMed | Scopus (30)See all References2] to become compatible with a ribonucleotide on the 3’ end but not the 5’ end of the nick [1xRNase-sensitive DNA modification(s) initiates S. pombe mating-type switching. Vengrova, S. and Dalgaard, J.Z. Genes Dev. 2004; 18: 794–804Crossref | PubMed | Scopus (66)See all References1].We agree that a unifying explanation would be ideal. To achieve this, we should not make more assumptions before answering the following questions. What are the methods used by Dalgaard and colleagues to prepare intact/nicked DNA? How is their PCR approach able to synthesize across the intact heat-labile imprinted strand? What is the direct evidence for a mixed population of one and two ribonucleotides? Ultimately, identification of the machinery responsible for the imprint remains the major issue.