Extended Data Fig. 2: Characterization of localized, synchronous nascent strand degradation.
From: Replication fork uncoupling causes nascent strand degradation and fork reversal
(A) Cartoon depicting the source of θ* structures in Fig. 1B. The XmnI site is located 1013 base pairs away from the closest edge of the lacO array. (B) DNA structures from Fig. 1A were treated with human topoisomerase II (hTop2) or buffer control, then separated on an agarose gel and visualized by autoradiography. hTop2 treatment converted θ* signal back to θs, demonstrating that θ* structures are topoisomers of θs. Note that in lane 4 an additional band appears below the θ, suggesting that replication forks undergo remodeling (see Fig. 3 and related discussion in the main text). (C) Quantification of Replication Intermediates (RIs) as a % of total lane signal from Fig. 1E. Mean ± S.D., n = 5 independent experiments. (D) Quantification of total lane signal from Fig. 1E. Mean ± S.D., n = 5 independent experiments. (E) Plasmid DNA harboring a 32xlacO array (p[lacO]) or a 50xlacO array (p[lacOx50]) was incubated with LacR then replicated in Xenopus egg extracts. Once forks were localized to the LacR barrier, NSD was induced by addition of IPTG and aphidicolin. Purified DNA was subjected to restriction digest so that replication fork structures (RIs) could be visualized. (F) Samples from (e) were separated on an agarose gel and visualized by autoradiography. Note that a mobility shift can readily be detected for p[lacOx50], suggesting that fork remodeling occurs (see Fig. 4 and related main text). (G) Quantification of the amount of DNA degraded in (f). Normalized degradation accounts for the different backbone sizes of the 50x and 32x lacO plasmids. Mean ± S.D., n = 3 independent experiments. The 50xlacO array is ~60% larger than the 32xlacO array so if NSD was influenced by the proximity of the replication forks on either side of the lacO array there would be a substantial difference in NSD between p[lacO] and p[lacOx50]. However, the difference in normalized degradation between the two plasmids is negligible. Thus, NSD does not appear to be influenced by the proximity of the two converging forks.
