Saw1 localizes to repair sites but is not required for recruitment of Rad10 to repair intermediates bearing short non-homologous 3′ flaps during single-strand annealing in S. cerevisiae

Abstract

SAW1 is required for efficient removal by the Rad1-Rad10 nuclease of 3′ non-homologous DNA ends (flaps) formed as intermediates during two modes of double-strand break repair in S. cerevisiae, single-strand annealing (SSA) and synthesis-dependent strand annealing (SDSA). Saw1 was shown in vitro to bind flaps with high affinity, but displayed diminished affinity when flaps were short (< 30 deoxynucleotides [nt]), consistent with it not being required for short flap cleavage. Accordingly, this study, using in vivo fluorescence microscopy showed that SAW1 was not required for recruitment of Rad10-YFP to DNA double-strand breaks (DSBs) during their repair by SSA when the flaps were ~10 nt. In contrast, recruitment of Rad10-YFP to DSBs when flaps were ~500 nt did require SAW1 in G1 phase of cell cycle. Interestingly, we observed a substantial increase in colocalization of Saw1-CFP and Rad10-YFP at DSBs when short flaps were formed during repair, especially in G1, indicating significant recruitment of Saw1 despite there being no requirement for Saw1 to recruit Rad10. Saw1-CFP was seldom observed at DSBs without Rad10-YFP. Together these results support a model in which Saw1, and Rad1-Rad10 are recruited as a complex to short and long flaps in all phases of cell cycle, but that Saw1 is only required for recruitment of Rad1-Rad10 to DSBs when long flaps are formed in G1.

SAW1 is increasingly required to recruit Rad10 as SSA flap-length increases from 20 to 50 bases in single-strand annealing in S. cerevisiae

“…HIS3 (663 bp) genes were integrated so they flanked the I- Sce I site in YER186C in strain PF025-7A ( Table 1 ) using adaptamer-mediated PCR and gene transplacement [ 10 , 13 ]. Transformants were selected on Synthetic Complete agar lacking histidine (SC-his agar), screened by PCR, and sequenced for the HIS3::x-bp:I-SceI:x-bp::HIS3 cassette where “x” was either 10, 20 or 40 deoxynucleotide base pairs giving rise to substrates that would contain 20, 30 or 50 deoxynucleotide flaps, respectively.…”

“…In all cases, focal planes were offset by 0.3 μm intervals along the Z-axis (a Z-stack). Colocalization of foci were analyzed by inspecting images from each focal plane of the Z-stack contrast enhanced as described [ 13 ]. Cells were classified as “G1”, “S/G2” or “M” as previously described [ 13 ].…”

“…However, Rad1-Rad10 has long been known to bind 3′-flaps in vitro without Saw1 [ 12 ]. We previously showed that SAW1 is not required for recruitment of Rad1-Rad10 to SSA sites when flaps are only ∼10 nt but is required in G1 phase when flaps are ∼500 nt [ 13 ]. Together, these studies showed that Saw1 mediates binding of Rad1-Rad10 to 3′-flaps, except when flaps are short, and that Rad1-Rad10 may function in some contexts without Saw1.…”

“…Together, these studies showed that Saw1 mediates binding of Rad1-Rad10 to 3′-flaps, except when flaps are short, and that Rad1-Rad10 may function in some contexts without Saw1. Interestingly, our prior work revealed that despite not being required to recruit Rad1-Rad10 to SSA sites in ∼10 nt flaps, Saw1 nonetheless localizes to such substrates [ 13 ].…”

“…Since Saw1 does not bind to flaps shorter than ∼10 nt and appears to exhibit maximal binding at approximately 30–40 nt, we wanted to test whether the requirement for Saw1 in recruitment of Rad1-Rad10 to chromasomally-situated DSBs in vivo would also manifest in the ∼20–50 nt range (the ∼10 nt substrate having already been tested in our earlier work and shown not to require Saw1 [ 13 ]). We also wanted to determine if any requirement for Saw1 would extend to SSA repair, and whether Saw1 localization at repair sites would correlate with its requirement to recruit Rad1-Rad10.…”