POLD3 Is Haploinsufficient for DNA Replication in Mice

Abstract

The Pold3 gene encodes a subunit of the Polδ DNA polymerase complex. Pold3 orthologs are not essential in Saccharomyces cerevisiae or chicken DT40 cells, but the Schizosaccharomyces pombe ortholog is essential. POLD3 also has a specialized role in the repair of broken replication forks, suggesting that POLD3 activity could be particularly relevant for cancer cells enduring high levels of DNA replication stress. We report here that POLD3 is essential for mouse development and is also required for viability in adult animals. Strikingly, even Pold3(+/-) mice were born at sub-Mendelian ratios, and, of those born, some presented hydrocephaly and had a reduced lifespan. In cells, POLD3 deficiency led to replication stress and cell death, which were aggravated by the expression of activated oncogenes. Finally, we show that Pold3 deletion destabilizes all members of the Polδ complex, explaining its major role in DNA replication and the severe impact of its deficiency.

Figure 1. POLD3 is essential for mouse development.

(A) Scheme illustrating the targeting construct used for the generation of Pold3 mutant mice (top), and how the Pold3^lox and Pold3^- alleles are obtained upon recombination with Flp or Cre recombinases, respectively. (B) Southern blot illustrating the presence of ES clones harboring the correct integration of the Pold3^GT:lox allele that were subsequently used for the generation of mutant mice. The 10.2 Kb band corresponds to the endogenous Pold3 gene, and the 6.5 Kb band to the targeted allele. (C) Table representing the genotypes observed from inter-crossing Pold3 heterozygous mice. The expected values are those estimated from the offspring of a homozygous-lethal cross. Only 62% (185/298) of the expected Pold3^+/- mice are born. (D) Kaplan-Meyer curves of Pold3^+/+ (n = 149) and Pold3^+/- (n = 185) mice. The p value was calculated with the Mantel-Cox log rank test. ***P<0.001. (E) Representative pictures of 2 month-old Pold3^+/+ and Pold3^+/- littermate mice illustrating the enlarged head morphology (red arrow) and overall dwarfism found in a subset of Pold3 heterozygous mice. (F) Brain sections of Pold3^+/+ and Pold3^+/- mice. Note that the forebrain (red arrow) is almost inexistent on Pold3 heterozygous brain (dashed lines). The skull cavity of the mutant mouse was found full of liquid upon necropsy.

Figure 2. Pold3 deletion is severely toxic in adult mice.

(A) Kaplan-Meyer curves of mice from the indicated genotypes that were exposed to a 4-OHT containing diet that started at week 5. The p value was calculated with the Mantel-Cox log rank test. ***P<0.001. (B) WB illustrating the depletion of POLD3 levels that is observed on organs from Ub^Cre/Pold3^lox/lox mice fed the 4-OHT diet for 1 month. Note that no reduction in POLD3 levels is observed in the spleen. (C,D) γH2AX immunohistochemistry from the liver (C) and lung (D) of Pold3^lox/lox and Ub^Cre/Pold3^lox/lox mice fed the 4-OHT diet for 1 month. H/E: Hematoxylin/Eosin. Scale bar (black) indicates 200 μm.

Figure 3. POLD3 sustains DNA replication and prevents RS in B cells.

(A) Number of cells after stimulating 2.5x10⁵ freshly isolated splenic B-lymphocytes for 72 hrs with LPS. Dashed lines indicate mean values. **P<0.01; ***P<0.001. (B) FACS analysis of DNA content (propidium iodide, PI) versus 5’-bromo-2’-deoxyuridine (BrdU) incorporation in Pold3^lox/lox and CD19^Cre/Pold3^lox/lox B cells, 48 hrs after stimulation with LPS (25 μg/ml). BrdU was used at 10μM for 30 min. (C) Chromosomal abnormalities found per metaphase in B cells of the indicated genotypes 2 days after stimulation with LPS (25 μg/ml). Dashed lines indicate mean values. ***P<0.001. (D) Distribution of the different types of chromosomal rearrangements observed in metaphases from CD19^Cre/Pold3^lox/lox B cells. (E) Fork rates were measured in stretched DNA fibers (see Methods) prepared from B cells of the indicated genotypes. Data are representative of three indicative experiments. Approximately 900 tracks were measured to estimate fork rates. ***P<0.001. (F) Fork asymmetry (represented as left vs right fork lengths) in DNA fibers prepared from B cells of the indicated genotypes. The determination coefficient is indicated in blue (R2). See also Figure S1.

Figure 4. POLD3 maintains the stability of the POLD complex.

(A) WB illustrating the levels of POLD1, POLD2, POLD3 and POLE in B cell cultures from the indicated genotypes, 48 hrs after stimulation with LPS (25 μg/ml). CDK2 was used as a loading control. (B) WB of POLD1, POLD3 and POLE levels from Pold3^+/+ and Pold3^+/- B cells. TUBULIN is shown as a loading control. (C) Representative FACS analysis of DNA content (PI) versus BrdU incorporation in Pold3^+/+ and Pold3^+/- B cells. 48 hrs after stimulation with LPS (25 μg/ml) cells were exposed (or not) to 0.5 μM aphidicolin (APH) for 4 hrs, and then to BrdU (10μM) for another 30 min. Note that the APH treatment had a bigger impact on in limiting DNA replication on Pold3 heterozygous cells. (D) Fork rates were measured in stretched DNA fibers prepared from B cells of the indicated genotypes treated as in (C). Data from two independent B cell cultures of each genotype are shown. Approximately 300 tracks were measured to estimate fork rates in each condition. ***P<0.001. (E) WB illustrating the levels of POLD1, POLD2 and POLD3 in Ub^Cre/Pold3^lox/lox MEF exposed (or not) to 4-OHT (1 μM) and that were serum starved (0.1% serum) for 72 hrs and subsequently released in 15% serum-containing media for 24 hrs. (F) FACS analyses of DNA content (PI) from Ub^Cre/Pold3^lox/lox MEF exposed (or not) to 4-OHT (1 μM) and that were serum starved (0.1% serum) for 72 hrs and subsequently released in 15% serum-containing media for the indicated times. In (E) and (F) 4-OHT was present throughout the entire experiment where indicated. See also Figures S2, S3 and Table S1.