TABLE 3.
The effect of developmental timing of double-strand breaks on choice of DSB repair pathways
| Endonuclease source | % UIE-72Ca (maternal) | % UIE-72Cb (zygotic onward) | % βIE-3c (meiotic prophase) |
|---|---|---|---|
| SSA (1) | 81.0 ± 2.4 | 68.4 ± 1.7 | 8.0 ± 0.5 |
| NHEJ (1) | 10.4 ± 2.7 | 31.1 ± 1.3 | 44.8 ± 1.5 |
| Total (1) | 91.4 | 99.5 | 52.8 |
| SSA (2) | 82.5 ± 2.1 | 66.1 ± 1.6 | 16.7 ± 2.7 |
| NHEJ (2) | 12.1 ± 1.9 | 18.7 ± 1.7 | 66.3 ± 2.3 |
| Conversion | 0.4 ± 1.7 | 11.3 ± 1.5 | 1.4 ± 1.7 |
| Total (2) | 95.1 | 96.1 | 84.4 |
Estimation of pathway usage was performed with the timing of breakage being regulated by the expression of I-SceI endonuclease. When endonuclease was expressed in the G0 male only through maternal effect (second column) DSB formation was restricted to early stages of development. In the third column (also in Table 1) endonuclease is expressed throughout development. The fourth column represents endonuclease expressed late in germ-line development of the G0 male by the use of a β2-tubulin promoter. As in the previous experiments, standard errors were computed from replicate information without bias from premeiotic clustering (see the appendix).
Crosses 1 and 2 were performed by a variation of the scheme in Figure 4. Parental G0 males do not carry the endonuclease transgene themselves, but obtain endonuclease through maternal effect. Some of these males were crossed to females lacking endonuclease to yield G1 progeny for SSA measurements, and others were crossed to endonuclease-bearing females. All progeny from the latter crosses can be scored for NHEJ in cross 1 or for NHEJ + conversion in cross 2 since endonuclease will be expressed through maternal effect. For cross 1, the number of replicates and the counts for SSA were 68 and 2782/3435, and for NHEJ they were 77 and 303/2923. For cross 2, the SSA frequency was estimated from 85 replicates and counts of 2840/3441, while NHEJ + conversion estimation used 144 replicates for counts of 788/6272. A total of 212 G1 offspring were tested by PCR, of which 205 were NHEJ and 7 were conversions. The crossover frequency of 0.3% (38/12,428) had a standard error of 0.1%.
See Table 1.
Endonuclease source, βIE-3 (Bellaiche et al. 1999) ensured DSB formation only near the end of germ-line development in the G0 males. As was the case for maternal-effect expression, it was necessary to perform separate crosses for SSA measurements vs. NHEJ or NHEJ + conversion, since the βIE-3 transgene did not produce the early endonuclease required for scoring mosaic G1 progeny. For cross 1, the SSA measurement came from 79 single-G0 male replicates yielding counts of 264/3305, and the NHEJ estimate came from 50 replicates and counts of 919/2051. In cross 2, we estimated SSA from 62 replicates and counts of 320/1912. The NHEJ + conversion estimates came from 35 replicates and counts of 810/1197. We tested 337 G1 offspring by PCR, with 330 of them proving to be NHEJ and 7 being conversions. (The crossover frequency was not measured for this group.)