Characterization of the Type III restriction endonuclease PstII from Providencia stuartii

Abstract

Type III Restriction-Modification systems characteristically methylate a base in only one strand of their recognition sequence. In an earlier paper (1), we reported that PstII recognizes the asymmetric sequence 5′-CTGATG-3′ and methylates one of the adenines in the bottom strand of this sequence, forming 5′-Cm6ATCAG-3′. We have recently learned that this assignment is incorrect; PstII in fact methylates the adenine in the top strand, forming 5′-CTGm6ATG-3′. Prompted by data from Single Molecule Real Time (SMRT)-sequencing of the Providencia stuartii genome that unambiguously shows that PstII forms 5′-CTGm6ATG-3′ (Iain A. Murray, Geoffrey G. Wilson and Richard D. Morgan, personal communication), we revisited the methylation data (Figure 4 in (1)). The original synthetic DNA oligonucleotides used in our study were annealed and the radioactive methylation assay was repeated using a purified recombinant PstII preparation from 2005. The results obtained agree with the SMRT-sequencing data (Figure 1): synthetic pre-methylation of the adenine in the CTGATG-strand (168F/168RM3) reduced incorporation of radioactivity from S-[methyl-3H]-adenosyl-L-methionine approximately 100-fold compared to pre-methylation of either of the adenines in the CATCAG-strand (168FM1/168R and 168FM2/168R). The oligonucleotide numbering system (premethylated at position 1, 2 or 3) in the bar graph in Figure 4 of (1) is consistent with the naming of the original methylated DNA strands as listed in Figure 1a here. In reviewing our earlier experiment, we found the data were correct and that the error arose during manuscript preparation due to mislabelling of the adenines in the DNA cartoon of Figure 4 of (1). A corrected version of Figure 4 is given here (Figure 2). This error was then carried through to the text and to Figure 2A of (1), a corrected version of which is shown here (Figure 3).