Unphosphorylatable mutants of Cdc6 disrupt its nuclear export but still support DNA replication once per cell cycle

Abstract

Cdc6 is essential for eukaryotic DNA replication. We have mutated highly conserved CDK phosphorylation sites in Cdc6. Contrary to their reported phenotypes in human cells, unphosphorylatable DeltaCDK mutants fully support DNA replication in Xenopus eggs. WtCdc6 is actively exported from the nucleus, which could explain why nuclear permeabilization is required for reinitiation within one cell cycle. However, DeltaCDK mutants are retained in the nucleus, yet surprisingly they still support only one round of replication. As these highly conserved CDK sites are unnecessary for replication once per cell cycle, an alternative checkpoint role for monitoring completion of the S phase is suggested.

Figure 1

Xenopus Cdc6 is displaced from nuclei during replication in egg extract, and its phosphorylation varies with cell cycle phase. Xenopus sperm nuclei were incubated in egg extract and harvested at various times. (A) Reactions were stopped by fixation after 15, 30, 60, and 120 min, and nuclear XCdc6 was detected by affinity purified anti-XCdc6 antibody. Sperm nuclei were stained with propidium iodide (red), and XCdc6 was revealed by an anti-rabbit fluorescein-linked antibody (green). Panels show merged images (red + green = yellow). (B) Chromatin-bound XCdc6 after 10, 40, and 70 min of incubation detected by immunoblotting (top panel). Chromatin-bound XOrc1 is in the bottom panel. The total amount of XCdc6 and XOrc1 in the extract is shown (tot). (C) XCdc6 is hypophosphorylated during interphase and hyperphosphorylated in mitotic extracts. Immunoblots of XCdc6 show different electrophoretic mobility forms of the protein in interphase (lane 1, I) and mitotic extracts (lane 2, M). The interphase and mitotic extracts were incubated with (lanes 3,4) or without (lanes 5,6) λ-phosphatase.

Figure 2

Mutation of conserved CDK consensus sites abolishes phosphorylation of Cdc6. (A) Schematic representation of XCdc6 and XCdc6 mutants used in this work. XCdc6 contains five full consensus sites for CDKs (Ser 54, Ser 74, Ser 108, Ser 120, and Ser 411). XCdc6M5 (Δ1–125) is an amino terminal deletion mutant lacking four CDK consensus sites. XCdc6M9 (S54A, S74A, S108A, S120A, S411A) was obtained by mutating serines in the CDK sites into alanines. (B) Recombinant wild-type XCdc6 and XCdc6M9 mutant proteins were tested for phosphorylation under replication assay conditions. They were incubated in interphase extracts with [γ-³²P]ATP and then immunoprecipitated by using an anti-His antibody. The reactions were split into two aliquots for autoradiography and western blotting.

Figure 3

CDK phosphorylation of XCdc6 is not required for initiation of DNA replication. (A) Unphosphorylatable XCdc6 mutants do not inhibit DNA replication when added to interphase extracts at 10 ng/μL or 100ng/μL. Replication was measured by [α-³²P]dATP incorporation after addition of buffer alone (+B), XCdc6 (+wt), either of the unphosphorylatable mutants (+M9, +M5), or interphase extract (+I). (B,C) Unphosphorylatable XCdc6 mutants rescue the replication of XCdc6-depleted extracts. Xenopus egg extract was depleted of endogenous XCdc6 and supplemented with buffer (+B), recombinant XCdc6 (+wt), either of the mutants (+M9, +M5), or interphase extract (+I). Proteins were added at a final concentration of 10 ng/μL. Replication was detected by either [α-³²P]dATP (B) or biotin-dUTP incorporation (C). For confocal microscopy, nuclei were stained by propidium-iodide (red), and incorporated biotin was detected by fluorescein-linked streptavidin (green). Panels show merged images (red + green = yellow).

Figure 3

CDK phosphorylation of XCdc6 is not required for initiation of DNA replication. (A) Unphosphorylatable XCdc6 mutants do not inhibit DNA replication when added to interphase extracts at 10 ng/μL or 100ng/μL. Replication was measured by [α-³²P]dATP incorporation after addition of buffer alone (+B), XCdc6 (+wt), either of the unphosphorylatable mutants (+M9, +M5), or interphase extract (+I). (B,C) Unphosphorylatable XCdc6 mutants rescue the replication of XCdc6-depleted extracts. Xenopus egg extract was depleted of endogenous XCdc6 and supplemented with buffer (+B), recombinant XCdc6 (+wt), either of the mutants (+M9, +M5), or interphase extract (+I). Proteins were added at a final concentration of 10 ng/μL. Replication was detected by either [α-³²P]dATP (B) or biotin-dUTP incorporation (C). For confocal microscopy, nuclei were stained by propidium-iodide (red), and incorporated biotin was detected by fluorescein-linked streptavidin (green). Panels show merged images (red + green = yellow).

Figure 4

CDK-dependent phosphorylation of XCdc6 is required for its export from the nucleus. (A) XCdc6 recombinant protein rescues replication of XCdc6-depleted extract and becomes undetectable in the nuclei after replication (90 and 120 min, left panels). Addition of leptomycin B (+LMB, right panels) prevents translocation of XCdc6 to the cytosol. (B) The unphosphorylatable XCdc6M9 mutant is not subject to export but remains in the nuclei throughout and after replication (90 and 120 min, left panels). LMB does not affect the localization of this mutant (+LMB, right panels). Replication was detected by Texas Red-streptavidin (red); antigen, by fluorescein-linked secondary antibody (green). Panels show merged images (red + green =yellow). Nuclei were stained with TOTO-3 iodide (not shown).

Figure 5

XCdc6 phosphorylation mutants do not cause DNA overreplication. Xenopus sperm chromatin was incubated in (A) XCdc6 depleted extract or depleted extract supplemented with either interphase extract or recombinant XCdc6, (B) depleted extract supplemented with XCdc6M9, and (C) depleted extract supplemented with XCdc6M5. All the recombinant proteins were used at a final concentration of 10 ng/μL. Reactions were incubated for 5 hr in the presence of BrdUTP and [α-³²P]dATP. Buoyant density of unreplicated (LL), once-replicated (HL), and overreplicated (HH) DNA are marked at the top of each gradient profile. (D–E) Sperm chromatin was incubated in XCdc6-depleted extract rescued by addition of either XCdc6 or XCdc6M9 (10 ng/μL). (D) After 10, 40, and 70 min, reactions were stopped and sperm chromatin was assayed for the presence of chromatin-bound XCdc6 (top panel) and XCdc6M9 (bottom panel) by immunoblotting. (E) In a parallel experiment, sperm chromatin was assayed for the presence of XMcm3 by immunofluorescence before (15 min) and after (90 min) replication. Replication, nuclei, and antigen were stained as described for Figure 4, with the difference that anti-XMcm3 antibody was used.

Figure 5

XCdc6 phosphorylation mutants do not cause DNA overreplication. Xenopus sperm chromatin was incubated in (A) XCdc6 depleted extract or depleted extract supplemented with either interphase extract or recombinant XCdc6, (B) depleted extract supplemented with XCdc6M9, and (C) depleted extract supplemented with XCdc6M5. All the recombinant proteins were used at a final concentration of 10 ng/μL. Reactions were incubated for 5 hr in the presence of BrdUTP and [α-³²P]dATP. Buoyant density of unreplicated (LL), once-replicated (HL), and overreplicated (HH) DNA are marked at the top of each gradient profile. (D–E) Sperm chromatin was incubated in XCdc6-depleted extract rescued by addition of either XCdc6 or XCdc6M9 (10 ng/μL). (D) After 10, 40, and 70 min, reactions were stopped and sperm chromatin was assayed for the presence of chromatin-bound XCdc6 (top panel) and XCdc6M9 (bottom panel) by immunoblotting. (E) In a parallel experiment, sperm chromatin was assayed for the presence of XMcm3 by immunofluorescence before (15 min) and after (90 min) replication. Replication, nuclei, and antigen were stained as described for Figure 4, with the difference that anti-XMcm3 antibody was used.