Comment on "The [4Fe4S] cluster of human DNA primase functions as a redox switch using DNA charge transport"

Abstract

O'Brien et al (Research Article, 24 February 2017, eaag1789) proposed a novel mechanism of primase function based on redox activity of the iron-sulfur cluster buried inside the C-terminal domain of the large primase subunit (p58C). Serious problems in the experimental design and data interpretation raise concerns about the validity of the conclusions.

Fig. 1. Comparison of functionally relevant primase substrate and correctly folded p58C (4) with the substrate and misfolded p58C used by O’Brien et al. (1)

(A) Side-by-side comparison of p58C structures with correctly folded (PDB code 5F0Q) and misfolded (PDB code 3L9Q) substrate-binding regions highlighted in different colors. These regions are overlapped in the left bottom quarter. The positions of Y309, Y345 and Y347 relative to [4Fe-4S]²⁺ in two structures are shown in the right bottom quarter. The reason for p58C misfolding might be explained by substitution of structurally important I271 by serine in erroneously named “wild-type” p58C (PDB code 3L9Q) or by asparagine in its mutants Y345F and Y347F (PDB codes 5I7M and 5DQO, respectively). (B) Crystal structure of p58C in complex with a duplex containing 5′-triphosphate RNA and 3′-overhang DNA (4). (C) Primase substrate (6) with the 5′-triphosphate of a primer and the 3′-overhang of a template that are required for p58C binding are shown in blue. (D) Substrate used by O’Brien et al. (1) for electrochemistry experiments. In (C) and (D) the red-colored 5′-overhangs do not participate in binding of correctly folded p58C. The figure was prepared using the PyMOL Molecular Graphics System (version 1.8, Schrödinger, LLC).

Fig. 1. Comparison of functionally relevant primase substrate and correctly folded p58C (4) with the substrate and misfolded p58C used by O’Brien et al. (1)

(A) Side-by-side comparison of p58C structures with correctly folded (PDB code 5F0Q) and misfolded (PDB code 3L9Q) substrate-binding regions highlighted in different colors. These regions are overlapped in the left bottom quarter. The positions of Y309, Y345 and Y347 relative to [4Fe-4S]²⁺ in two structures are shown in the right bottom quarter. The reason for p58C misfolding might be explained by substitution of structurally important I271 by serine in erroneously named “wild-type” p58C (PDB code 3L9Q) or by asparagine in its mutants Y345F and Y347F (PDB codes 5I7M and 5DQO, respectively). (B) Crystal structure of p58C in complex with a duplex containing 5′-triphosphate RNA and 3′-overhang DNA (4). (C) Primase substrate (6) with the 5′-triphosphate of a primer and the 3′-overhang of a template that are required for p58C binding are shown in blue. (D) Substrate used by O’Brien et al. (1) for electrochemistry experiments. In (C) and (D) the red-colored 5′-overhangs do not participate in binding of correctly folded p58C. The figure was prepared using the PyMOL Molecular Graphics System (version 1.8, Schrödinger, LLC).

Fig. 2. Side-by-side comparison of experimental setups for examination of RNA synthesis termination by human primase

(A) Example of primer elongation reactions showing the effect of template:primer structure on the efficiency of RNA synthesis termination by human primase as described by Baranovskiy et al. (6). When using the correct substrate containing 5′-triphosphate and 3′-overhang, there is pronounced termination of reaction mainly at 9-mer primers (lanes 3 and 4), which are optimal for extension by Polα (4). However, in the case of template:primer without 5′-triphosphate and 3′-overhang, primase loses the ability to terminate synthesis at 9-mer primers (lanes 6 and 7) and has dramatically reduced activity, which requires a higher load of the enzyme and longer reaction time. Primase activity reactions were reproduced as in (6). The products were labeled by incorporation of [α-³³P]GTP at the seventh position of primer. Note that the negatively charged triphosphate moiety increases the mobility of RNA primers. Lanes 1 and 5, control incubations in absence of enzyme or primer, respectively. Lane 2, reaction was not supplied with CTP and UTP. (B) Example of incorrectly designed primer elongation reactions, which are not capable of capturing physiologically relevant primer termination. The image was adopted from Fig. 5A (1). Note high enzyme concentrations and long reaction time. The products with a length of 29 nucleotides or less are the result of RNA synthesis initiated from T₂₉ in the template (primase initiates RNA synthesis only from a template pyrimidine in the presence of ATP or GTP (13)). Appearance of primers longer than 10 nucleotides is due to the absence of Polα or its catalytic core, which allows primase to rebind the 9-mer primer without involvement of p58C and extend further (6, 14). Unfortunately, the gel provided does not allow visualization of the products of de novo synthesis from 2-mer to 9 to 12-mer. Counting of bands below 29-mer product indicates that the lowest visible band corresponds to 18-mer primer, not to 10-mer as shown in Fig. S14 in (1).