Reducing DNA polymerase alpha in the absence of Drosophila ATR leads to P53-dependent apoptosis and developmental defects

Abstract

The ability to respond to DNA damage and incomplete replication ensures proper duplication and stability of the genome. Two checkpoint kinases, ATM and ATR, are required for DNA damage and replication checkpoint responses. In Drosophila, the ATR ortholog (MEI-41) is essential for preventing entry into mitosis in the presence of DNA damage. In the absence of MEI-41, heterozygosity for the E(mus304) mutation causes rough eyes. We found that E(mus304) is a mutation in DNApol-alpha180, which encodes the catalytic subunit of DNA polymerase alpha. We did not find any defects resulting from reducing Polalpha by itself. However, reducing Polalpha in the absence of MEI-41 resulted in elevated P53-dependent apoptosis, rough eyes, and increased genomic instability. Reducing Polalpha in mutants that lack downstream components of the DNA damage checkpoint (DmChk1 and DmChk2) results in the same defects. Furthermore, reducing levels of mitotic cyclins rescues both phenotypes. We suggest that reducing Polalpha slows replication, imposing an essential requirement for the MEI-41-dependent checkpoint for maintenance of genome stability, cell survival, and proper development. This work demonstrates a critical contribution of the checkpoint function of MEI-41 in responding to endogenous damage.

Figure 1.—

Enhancer of mus304 is an allele of DNApol-α180. E(mus304) was roughly mapped and predicted to be a mutation in DNApol-α180 (see materials and methods). Sequencing of this region confirmed a loss of an “A” in the third exon at codon 301, resulting in a frameshift and a premature stop 29 codons downstream. An EMS mutagenesis and screen for mutations conferring rough eyes to mei-41 mutants resulted in two new alleles (see materials and methods); both were nonsense mutations in glutamine codons. Shaded boxes are coding exons. Mutations are marked with asterisks.

Figure 2.—

Reducing Polα in mei-41 mutants results in a variety of phenotypes. (A) Wing discs of third instar larvae were dissected, fixed, and stained with an antibody to cleaved human caspase 3, marking apoptotic cells. (B) As shown previously (Brodsky et al. 2000), mei-41; polα/+ mutants have a rough-eye phenotype that includes fused ommatidia and tissue loss. mei-41 mutants are indistinguishable from wild type and are used for comparison. This rough-eye phenotype of mei-41; polα/+ mutants was rescued by eliminating P53. (C) Quantification of apoptosis phenotype demonstrated in A. mei-41 mutants had an increase in apoptosis compared to wild type (P < 10⁻⁶), and this was quantitatively more severe when Polα was reduced (P < 10⁻⁴ when compared to mei-41). Mutations in p53 restored apoptosis to the levels seen in mei-41 single mutants (P = 0.19 compared to mei-41).

Figure 3.—

Reducing Polα in mei-41 mutants results in an increase of LOH. mwh mutant flies have multiple hairs from each hair cell of the adult wing, and mwh/+ flies are phenotypically normal. LOH at mwh will result in clones of cells with multiple hairs per cell (circled). LOH can occur through spontaneous mutation, gene conversion, deletion, or mitotic crossing over. Individual adult wings were scored for mwh clones. Bars represent the average number of clones per wing, and lines are the standard deviation based on 10–12 wings/genotype. Significance was determined by an unpaired t-test with Welch's correction.

Figure 4.—

Reducing mitotic cyclins rescues the rough-eye phenotype and apoptosis of mei-41; polα/+ mutants. (A) The rough-eye phenotype of mei-41; polα/+ mutants is partially rescued by reducing CycB and completely rescued when CycA is reduced. (B and C) The apoptosis phenotype observed in mei-41; polα/+ mutants is rescued by reducing CycA. Samples were prepared and scored as described in Figure 2. Bars represent averages of 7–10 imaginal wing discs/genotype, and lines represent standard deviations. Significance was determined by an unpaired t-test with Welch's correction.

Figure 5.—

Analysis of grp and lok mutants when Polα is reduced. (A) While eyes of lok; polα/+ mutants are indistinguishable from those of wild-type flies, grp single mutants and grp lok double mutants have rough eyes when Polα is reduced. (B and C) lok; polα/+ have only a slight increase in apoptosis compared to wild type (P < 0.05) whereas grp; polα/+ and grp lok; polα/+ mutants have a greater increase in apoptosis (P < 10⁻⁴) but are not significantly different from each other (P = 0.16). Samples were prepared as described in Figure 2. Bars represent averages of 7–14 imaginal wing discs/genotype, and lines represent standard deviations. Significance was determined by an unpaired t-test with Welch's correction.