Figure 1.
KU70 and KU80 genes in P. tetraurelia.
(A) Diagram of the WGD relationships between KU70 or KU80 genes in P. tetraurelia. The % of identity between genes (nt) and proteins (aa) are indicated along the arrows connecting two genes. (B) Transcription profiles of KU70 and KU80 genes during an autogamy time-course of strain 51, as determined by high-throughput RNA-seq. V: vegetative cells; S: starved or meiotic cells with intact parental MAC; T0: 50% of cells with fragmented MAC; the following time-points refer to hours after T0 [51]. On the vertical axis, FPKM represents the number of fragments per gene kb per million of fragments that were uniquely mapped on the genome. (C) Detection of KU70 and KU80c mRNA during autogamy, through northern blot hybridization. Control RNAi: RNAi against the nonessential ND7 gene [43], which encodes an exocytosis protein. V: vegetative cells. The times refer to hours after T0, the time at which 50% of cells have a fragmented MAC (Figure S1A).
Figure 2.
Nuclear localization of GFP-Ku fusions during autogamy.
Cells were microinjected with fusion transgenes expressing GFP-Ku70a (panels a, b, c) and GFP-Ku80c (panels d, e, f) under the control of their respective transcription signals. In autogamous cells, developing MACs are indicated by yellow arrowheads, the other DAPI-stained nuclei are fragments from the old vegetative MAC. For each protein, the GFP fluorescence sometimes concentrated in nuclear foci of unclear biological significance (see cells in panels c and f, and enlarged inserts on the right). In this particular experiment, expression of the GFP fusions had no significant effect on the recovery of viable post-autogamous progeny (87% progeny with functional new MACs for GFP-Ku70a, 90% for GFP-Ku80c).
Figure 3.
RNAi screen for essential KU genes during autogamy.
(A) Survival of the post-autogamous progeny of cells submitted to different combinations of RNAi. Kp: autogamy in standard K. pneumoniae medium; ICL7: RNAi against ICL7, a nonessential gene that encodes an infraciliary lattice centrin [47]; XR: RNAi against XRCC4. RNAi experiments against KU80 genes were performed using gene-specific inserts KU80-a2, KU80-b2 and KU80-c2 (see Figure S2). For each condition, 30 to ∼140 post-autogamous cells were analyzed. Each bar represents the percentage of viable post-autogamous cells carrying a functional new MAC, for each condition. Error bars represent the Wilson score intervals (95% confidence level), which are appropriate for a small number of trials or for values close to an extreme probability. (B) DAPI-staining of developing MACs during RNAi against ND7 (control), KU70, KU80c, XRCC4/LIG4 and LIG4 + KU80c. Developing MACs are indicated by yellow arrowheads. Upon LIG4 or XRCC4 RNAi, developing MACs exhibit faint DAPI staining, which correlates with a defect in DNA amplification [16]. (C) Northern blot hybridization of total RNA during a control time course experiment (ND7 RNAi) and in a KU80c RNAi. V: vegetative cells; T0: 60% of cells with fragmented MAC; other time-points refer to hours following T0 (Figure S1B).
Figure 4.
Molecular analysis of genome rearrangements after Ku80c depletion.
(A) Detection of DNA fragments with excised or non-excised IES 51G4404 from the surface antigen G51 gene. Total genomic DNA was extracted during autogamy of 51ΔA cells subjected to RNAi against ICL7 (control) or KU80c. To compare the two time-courses, similar autogamy stages were numbered from 1 to 6, based on the observation of DAPI-stained cells (Figure S1C). PstI-hydrolyzed total genomic DNA was run on 1% agarose gels. Southern blots were hybridized with the Gmac probe (in grey), which hybridizes to the flanking MAC DNA downstream of the IES. (B) Detection of fragmented or non-fragmented DNA downstream of the G51 gene by Southern blot hybridization of PstI-digested total genomic DNA (same samples as in A) run on 0.8% agarose gels. The subtelomeric tel51G probe is shown in grey. The white box in the bottom right diagram represents telomeric repeats. (C) PCR detection of de novo IES excision junctions during autogamy of 51ΔA cells, in an ICL7 (control) or a KU80c RNAi. Note that a few autogamous cells were present at time-point 1 in the control RNAi (see Figure S1C). (D) PCR detection of IES circle junctions during autogamy in an ICL7 (control) or a KU80c RNAi. Triangles in C and D represent PCR primers (see Table S1).
Figure 5.
Detection of programmed DSBs at IES boundaries during autogamy.
(A) LMPCR detection of DSBs at MAC or IES ends during autogamy of 51ΔA cells subjected to RNAi against ICL7 (control) or KU80c (same samples as in Figure 4). A Sanger DNA sequencing ladder provides size markers. On the diagram, the LMPCR linker is drawn as grey boxes and Paramecium DNA as black lines, with a black dot representing the 3′ end generated by Pgm-dependent cleavage. In the KU80c RNAi, the LMPCR signals at 51G4404 MAC ends are likely due to background DNA breaks generated at the MAC G51 locus during DNA extraction. This background is not detected for IESs of the A51 gene, because this locus is absent from the old MAC. (B) TdT tailing of free 3′OH ends during autogamy of 51ΔA cells subjected to RNAi against ICL7 (control) or KU80c (same samples as in (A)). On the diagram, the potentially resected 5′ end is represented by a dotted line.
Figure 6.
Nuclear accumulation of a Pgm-GFP fusion in Ku80c-depleted cells.
Cells were microinjected with a PGM-GFP fusion transgene and one transformant was submitted to RNAi against ICL7 (control: panel A) or KU80c (panel B). The progression of autogamy was monitored over a four-day starvation period (a and e: day 1; b and f: day 2; c and g: day 3; d and h: day 4). To compare the intensities of GFP fluorescence, signals were acquired with the same exposure time, and identical window settings were applied to the image display using the ImageJ software (National Institute of Health). Developing MACs are indicated by yellow arrowheads. In the enlarged inserts shown on the right of each panel, the display settings were modified to highlight the Pgm-GFP nuclear foci. The white arrowheads in panel B point to the DAPI-free regions, in which overproduced Pgm-GFP accumulates following KU80c RNAi. In the control RNAi, 93% of post-autogamous progeny had a functional new MAC, while the KU80c RNAi yielded no progeny with a functional new MAC.
Figure 7.
Physical interaction between Pgm and Ku.
(A) Co-precipitation of HA-Ku80c and Ku70a-HA with MBP-Pgm from insect cell extracts (top panel), revealed on western blots using an anti-HA antibody. (B) Co-immunoprecipitation of Pgm with HA-Ku80c and/or Ku70a-HA from insect cell extracts, revealed on western blots using an anti-Pgm antibody. In A and B, the input proteins from each extract are displayed in the bottom panel. (C) The interaction between HA-Ku80c and Pgm is resistant to DNaseI treatment. The co-precipitation experiment was performed as described in A, using MBP-Pgm, HA-Ku80c and 6His-Ku70a recombinant proteins produced from baculovirus vectors. EDTA was removed from the lysis buffer and replaced by 10 mM MgCl2. Half of the sample was treated with 40 µg/mL of Dnase I during the 2-hr incubation with amylose-coupled magnetic beads. The presence of HA-Ku80c in the purified complexes was revealed on western blots using an anti-HA antibody.
Figure 8.
Models for the assembly of an active DNA cleavage complex.
(A) Ku associates with DNA-bound Pgm and activates DNA cleavage. Pgm and its putative partners (in grey) would recognize and bind the boundaries of eliminated sequences. The binding of Ku (in blue) activates Pgm for DNA cleavage (symbolized by the switch from a rectangular box to an oval), perhaps by assisting the formation of a synapse between both IES ends, in a transpososome-like intermediate. The DNA-PKcs catalytic subunit (in peach) may also be part of the active DNA cleavage complex. (B) Ku forms a complex with Pgm in the absence of DNA, activating the DNA binding and/or cleavage activities of Pgm. Following DNA cleavage, conformational remodeling of the complex would position Ku on broken DNA ends and allow it to perform its classical role in C-NHEJ-mediated DSB repair. MAC DNA is represented in black, IES DNA in red.