Treslin, DUE-B, and GEMC1 cannot complement Sld3 mutants in fission yeast

Abstract

Initiation of DNA replication in eukaryotes is an evolutiontionarily conserved process that involves two distinct steps: the formation of pre-replication complexes at replication origins in G1 and the assembly of pre-initiation complexes in S phase, which leads to activation of the replication helicase. For the assembly of pre-initiation complexes in yeast, formation of Sld2-Dpb11-Sld3 complex is a critical event that requires phosphorylation of Sld2 and Sld3 by CDK. In mammals, RecQL4 and TopBP1 are excellent ortholog candidates for Sld2 and Dpb11, respectively. In this past year, three TopBP1-interacting proteins Treslin/Ticrr, GEMC1 and DUE-B have been identified in metazoans as possible functional orthologs of the yeast Sld3. To test this hypothesis, we carried out several complementation tests in fission yeast. The proteins were expressed at various levels in the temperature sensitive sld3-10 mutant and in cells that lack endogenous Sld3. Our result showed that none of these metazoan proteins could rescue growth defect of the sld3 mutants. Although the result may have several interpretations, it is possible that the helicase activation in mammals has diverged in complexity during evolution from that in yeasts and may involve multiple players that interact with TopBP1.

Initiation of DNA replication in eukaryotes is a dynamic process that is highly regulated during the cell cycle (Masai, et al., 2010). It is achieved through assembly and disassembly of a number of replication factors onto specific chromosomal loci, called replication origins, to ensure that the genomic DNA is replicated only once in each cell cycle. The origin recognition complex (ORC) binds to replication origins throughout the cell cycle. When cell exists mitosis, the mini-chromosome maintenance (MCM) complex binds to replication origins, which is facilitated by ORC, Cdt1 and Cdc6/Cdc18, to form the pre-replicative complexes (pre-RCs). When the cell enters S phase, several factors are further recruited to form the pre-initiation complexes (pre-ICs) for activation of the replication helicase Cdc45-MCM-GINS (CMG) complex (Ilves et al., 2010). Formation of pre-ICs also requires the cooperation of at least two protein kinases, the cyclin-dependent kinase (CDK) and Dbf4-dependent Cdc7 protein kinase (DDK). The activated CMG helicase unwinds origin DNA and promotes assembly of RPA and DNA polymerases for initiation of DNA synthesis.

However, the exact mechanism involved in helicase activation at replication origins remains not very clear. It has also been shown recently in yeasts that the CDK dependent interaction between Dpb11, Sld2 and Sld3 is a critical event for the assembly of the pre-ICs (Tanaka, et al., 2007, Zegerman & Diffley, 2007). Phosphorylation of Sld2 and Sld3 by CDK on multiples sites allows binding of Dpb11 by its carboxy- and amino-terminal pairs of BRCT repeats, respectively. It is believed that formation of the Sld2-Dpb11-Sld3 complex is involved in the stable loading of Cdc45 and GINS and subsequent activation of the CMG helicase (Heller, et al., 2011). However, how conserved is this event in higher eukaryotes remains unclear.

Most of the replication factors involved in pre-RCs and pre-ICs formation such as ORC, MCM, Cdc45, and the two protein kinases CDK and DDK are highly conserved in all eukaryotes. Based on sequence similarity, RecQL4 and TopBP1 are possible vertebrate homologs of Sld2 and Dpb11/Cut5, respectively (Yamane & Tsuruo, 1999, Sangrithi, et al., 2005). In this past year, three proteins, DUE-B, Treslin/Ticrr, and GEMC1, have been discovered as possible candidates for the mammalian Sld3 orthologs. DUB-E (DNA Unwinding Element Binding) was isolated as a c-myc origin binding protein (Casper, et al., 2005). GEMC1 contains a partial homology with the coiled-coil domain of the Cdt1 inhibitor Geminin (Balestrini, et al., 2010). Treslin (TopBP1-interacting, replication-stimulating protein) was found in Xenopus egg extract (Kumagai, et al., 2010). A closely similar protein was also identified in zebrafish called Ticrr (TopBP1-interacting, checkpoint, and replication regulator) with a checkpoint defect (Sansam, et al., 2010). Like Sld3, all three proteins can form a complex with TopBP1 and they all function in the loading of Cdc45 (Mueller, et al., 2011). Among the three proteins, only Treslin/Ticrr shares some sequence similarity with yeast Sld3 and its interaction with TopBP1 appears to be dependent on site-specific phosphorylation by CDK (Sanchez-Pulido, et al., 2010, Boos, et al., 2011, Kumagai, et al., 2011).

Proteins of higher eukaryotes can sometimes complement the mutations of similar gene products in yeasts, which provide the strongest evidence of homology. One classic example is that expression of human CDC2 or S. cerevisiae CDC28 in the fission yeast S. pombe can complement the Cdc2 mutant (Lee & Nurse, 1987, Langan, et al., 1989). To see whether DUE-B, GEMC1 or Treslin/Ticrr are functional orthologs of yeast Sld3, we performed the complementation tests in fission yeast. For this purpose, full-length genes of human DUE-B, GEMC1 and Ticrr and Xenopus Treslin were individually cloned into S. pombe expression vectors and expressed at different levels in fission yeast under the control of thiamine repressive nmt1⁺, nmt41 or nmt81 promoters. To monitor protein expression, an HA epitope was fused to the N-terminus of GEMC1 and Ticrr and the C-terminus of Treslin. Expression of the three proteins was then confirmed by Western blotting using mouse anti-HA monoclonal antibody. Expression of untagged DUE-B was confirmed by immunoblotting using polyclonal antibody.

Having confirmed the protein expression, we performed the complementation test in the temperature-sensitive S. pombe mutant sld3-10 cells (Nakajima & Masukata, 2002). As shown in Fig.1a, while the sld3-10 cells grew well at the permissive temperature 30°C, they could not grow at the nonpermissive temperature 37°C. When wild-type Sld3 was exogenously expressed, the cells grew well under the nonpermissive temperature similar to wild type cells, suggesting that the growth defect was caused by the mutation in sld3⁺ gene. However, when each of the candidate homologous proteins was expressed in the mutant cells, no cell growth could be observed at 37°C, indicating that the expressed proteins could not rescue the growth defect of the sld3-10 mutant cells. Over-expression of GEMC1 and Ticrr, but not DUE-B or Treslin, suppressed the cell growth even under the permissive temperature, suggesting that they are toxic. Consistent with the observed cytotoxicity, the cells with overexpressed GEMC1 or Ticrr were dramatically elongated (Fig. 1b), unlike the cells with expressed Sld3, DUE-B, Treslin that showed cell length almost like the cells with an empty vector.

Fig. 1. Expression of human DUE-B, GEMC1, Ticrr and Xenopus Treslin at various levels cannot complement the sld3 mutations in S. pombe.

(a) Expression of the candidate proteins cannot rescue the temperature sensitive phenotype of sld3-10 mutant. The candidate Sld3 homologous human proteins and Xenopus Treslin (noted on the left) were expressed individually in sld3-10 mutant cells under the thiamine repressive promoters nmt81, nmt41 and nmt1 (marked on the right) with increasing promoter strength and confirmed by Western blotting. Cells were diluted in five-fold steps and spotted on EMM6S plates containing no (−) or 20 μg/ml (+) of thiamine. Plates were incubated at 30°C (permissive) or 37°C (non-permissive) for three days. A wild-type strain and the sld3-10 cells containing an empty vector or a vector expressing Sld3 under the medium-strength nmt41 promoter were used as the controls. Unlike Sld3, which fully rescued the grow defect of sld3-10 at 37°C, expression of DUE-B, GEMC-1, Ticrr or Treslin at various levels failed to complement the mutant under the non-permissive temperature. Due to leakage of the nmt41 promoter, a partial rescuing effect was observed with exogenously expressed Sld3 even in the presence of thiamine. (b) While expression of DUE-B and Treslin had no or little effect, over-expression of GEMC1 or Ticrr was toxic, which dramatically suppressed the cell growth. The sld3-10 cells with the indicated proteins expressed under the control of nmt1 promoter were incubated at 30°C for 3 days on plates without thiamine and examined under a microscope. Black bar indicates 20 μm. (c) In the presence of the candidate homologous proteins, the exogenously expressed Sld3 remains essential for cells lacking endogenous Sld3. Two vectors were introduced into Δsld3::kanR cells. The first one carries an ura4⁺ marker for expression of Sld3 under the control of its own promoter while the second vector with a LEU2 marker expresses the indicated candidate proteins. The N-terminal (1-1029 aa) and middle (484-1029 aa) regions of human Ticrr that contain some sequence similarity to yeast Sld3 were expressed under the nmt1 promoter and shown as Ticrr(N) and Ticrr(M), respectively. The expression of DUE-B and Treslin was under the nmt1 promoter while the expression of GEMC1 and the full-length Ticrr was controlled by nmt41 promoter. Cells were cultured in nonselective EMM6S[leu-] medium containing uracil for about 20 generations and then spread on three plates to form single colonies. Colonies were replicated onto 5-FOA plates to select against ura4⁺ cells. The ura4⁻ colonies were counted and presented as the percentages of the total colonies. Columns are the averages of three independent results with standard deviations. A few ura4⁻ colonies appeared in cells expressing DUE-B, Treslin or the N-terminus of Ticrr. However, as confirmed by colony PCR and subsequent restriction digestion of the PCR products, they all contained the full-length sld3⁺ open-reading frame, indicating that sld3⁺ gene was integrated into their genome. The negative control was an empty vector with the LEU2 marker.

Since the sld3-10 mutant protein may interfere with the complementation of the expressed candidate proteins by blocking their access to the targets of action, we then performed the test in a mutant cell that lacks endogenous Sld3 protein. The sld3⁺ gene was deleted by replacing it with a kanamycin resistant marker in a haploid S. pombe cell. The essential replication function of Sld3 was covered by Sld3 expressed from a vector that carries an ura4⁺ marker. If the candidate proteins can fully or partially replace the essential function of Sld3, it would be expected that the cells could survive without the need of the plasmid for expression of exogenous Sld3. However, we found that the cells with expressed DUE-B, GEMC1, Ticrr or Xenopus Treslin could not lose the plasmid. In contrast, the same cells carrying a second Sld3-expression vector could lose the plasmid without affecting its survival. The sequence of a small region of human Ticrr between 700 to 800 amino acids is conserved among the Sld3 family proteins (Sanchez-Pulido, et al., 2010). We expressed two fragments of Ticrr (1-1029 aa) and (484-1029 aa) that contain the conserved region in S. pombe. However, neither of the fragments could allow the cells to lose the Sld3 expressing vector. Together, our data showed that expression of human DUE-B, GEMC1, Ticrr or Xenopus Treslin in fission yeast could not rescue the growth defect caused by the Sld3 mutations.

Sld3 plays an important role in the loading of Cdc45, a key step during the initiation of eukaryotic DNA replication. Since most of the mechanisms involved in DNA replication are highly conserved among eukaryotes, it is believed that a functional homolog of Sld3 must exist in mammals. However, our result shows that all three recently identified candidate metazoan proteins failed to complement the Sld3 mutations in fission yeast. Several possibilities could explain the result: first, these candidate proteins are inactive or form aggregates that prevent them from binding to Cut5 when expressed in fission yeast. However, the observed cytotoxicity with overexpression of Ticrr and GEMC1 suggests that they may be “functional” proteins when expressed in fission yeast. In support of this notion, DUE-B and Treslin purified from insect cells are also known active proteins in loading to chromatin. Second, these proteins cannot be properly localized to the site of action in nucleus. Third, these proteins are not true functional homologues of Sld3 and a perfect match of Sld3 remains to be uncovered. Although the three human proteins shared some similarities with yeast Sld3, differences do exist. For example, phosphorylation of budding yeast Sld3 by CDK is essential for their interactions with Dpb11 (Tanaka, et al., 2007, Tanaka, et al., 2007). Although DUE-B can be phosphorylated by casein kinase 2 (CK2), its interaction with TopBP1 seems to be independent of the phosphorylation (Chowdhury, et al., 2010). While the loading of GEMC1 requires TopBP1, it can be loaded to chromatin in the presence of Geminin (Balestrini, et al., 2010), suggesting that the loading may begins even before the formation of pre-RCs. It has been shown in both yeasts that the loading of Dpb11/Cut5 requires the pre-loading of Sld3 (Kamimura, et al., 2001, Nakajima & Masukata, 2002). However, the loading of Treslin is not required for the association of TopBP1 with the chromatin in Xenopus egg extract, although the loading of Cdc45 is affected (Kumagai, et al., 2010). Recently, two CDK specific phosphorylation sites have been identified in Ticrr/Treslin whose phosphorylation is required for interaction with TopBP1 (Boos, et al., 2011, Kumagai, et al., 2011). However, these are just two out of 70 potential CDK phosphorylation sites. The function of additional phosphorylation sites on Ticrr and how Ticrr-TopBP1 interaction promotes the loading of GINS and Cdc45 remains unknown. Finally, the function of Sld3 in helicase activation in mammals may have diverged dramatically from that in yeasts during evolution and the functions of Sld3 in mammals may involve multiple proteins that interact with TopBP1 for Cdc45 loading. Interestingly, most of the mechanisms involved in the complex formation at the replication origins are highly conserved among eukaryotes. It is unclear why the mechanism involved in Cdc45 loading and helicase activation is less conserved in higher eukaryotes. One possibility is that more regulations are needed for this step in metazoans. Although evidence has been provided in support of the notion that Treslin/Ticrr is the functional ortholog of Sld3 (Kumagai, et al., 2010, Sanchez-Pulido, et al., 2010, Sansam, et al., 2010, Boos, et al., 2011, Kumagai, et al., 2011), it remains possible that Treslin/Ticrr works with additional TopBP1-interacting factors to fulfill the exact function of Sld3.