Hinderin, a five-domains protein including coiled-coil motifs that binds to SMC3

Abstract

Background: The structural maintenance of chromosome proteins SMC1 and SMC3 play an important role in the maintenance of chromosomal integrity by preventing the premature separation of the sister chromatids at the onset of anaphase. The two proteins are constitutive components of the multimeric complex cohesin and form dimers by interacting at their central globular regions.

Results: In order to identify proteins that by binding to SMC3 may interfere with the protein dimerization process, a human cDNA library was screened by the yeast two-hybrid system by using the hinge region of SMC3 as bait. This has lead to the identification of Hinderin, a novel five domains protein including two coiled-coil motifs and sharing a strikingly structural similarity to the SMC family of proteins. Hinderin is ubiquitously expressed in human tissues. Orthologue forms of the protein are present in other vertebrates but not in lower organisms. A mapping of the interaction sites revealed that the N- and C-terminal globular domains mediate the binding of Hinderin to SMC3. Hinderin/SMC3 complexes could be recovered by immunoprecipitation from cell lysates using an anti-SMC3 antibody, thus demonstrating that the two proteins interact in vivo. On the contrary, Hinderin did not interact with SMC1. In vivo the rate of SMC1/SMC3 interaction was decreased by the ectopic expression of Hinderin.

Conclusions: Hinderin is a novel binding partner of SMC3. Based on its ability to modulate SMC1/SMC3 interaction we postulate that Hinderin affects the availability of SMC3 to engage in the formation of multimeric protein complexes.

Figure 1

Hinderin: Genomic organization, structural domains, and expression pattern in human tissues. A) The sequence of the Hinderin ORF was used to BLAST the human genome database. The numeration of the exon boundaries is relative to the transcriptional start site. The size of the intronic sequences is based on the numeration of the human contiguous sequences. B) Northern blot hybridization of Hinderin in HeLa and HCT116 cells. A single main transcript of ~ 4.5 kb is identifiable. C) Prediction of coiled-coil domains in Hinderin. The numbers on the abscissa corresponds to the amino acid residues. The probability of the polypeptide to assume a coiled coil conformation is plotted on the ordinate axis. The contribution of the different exons to the five-domain structural organization of Hinderin is also illustrated. The sequence encoded by exons 3 through 6, bears a 36% homology to the SMC protein consensus sequence. A bipartite nuclear localization signal is harbored in exon 7. D) Expression of Hinderin in different human tissues. PCR amplification was stopped after 30 cycles and the product analyzed on 2% agarose staining with ethidium bromide. The G3PDH transcript was amplified in 20 cycles and used to show uniformity of the source cDNA in the different specimens.

Figure 2

Yeast two-hybrid assay of the interaction of different regions of SMC3 and SMC1 with Hinderin. A) SMC3 and Hinderin constructs in pGBKT7 are schematized on the left-hand side. All numerals refer to the amino acid sequence. The SMC3-465/807 construct harbors the protein hinge domain and was utilized as bait for the screening of the Clontech Matchmaker human cDNA library. SMC1, SMC3 and Hinderin constructs cloned in pACT2 or pGADT7 are illustrated on the right-hand side of the panel. H-47/577 represents the pACT2 clone retrieved from the screening. We scored the strength of interaction based on the rate of appearance of Blue colonies and the intensity of the color developed. The null score (-) was assigned when no blue colonies were visible after ten days. To obtain a quantitative measure of the rate of protein interaction, yeast colonies were grown in selective media and after lysis the β-galactosidase activity released in the supernatant assayed using a chromogenic substrate as detailed in the Methods. The results shown are the mean ± SD of three independent determinations.

Figure 3

Protein coimmunoprecipitation and competitive binding of Hinderin to SMC3. A) To monitor the Hinderin interaction, 293 cells were transfected with 1 μg/ml of Hinderin-V5 expression vector and incubated for 48 h. Mock transfected cells served as control. For SMC1 and SMC3 coimmunoprecipitation experiments, cell lysates (500 μl) were incubated with either 25 μg of anti-SMC1 or anti-SMC3 antibody for 2 h at RT. The immunocomplexes were then captured on agarose-protein G and analyzed by electrophoresis on 12% SDS-PAGE. After transblotting to a nitrocellulose membrane, Hinderin-V5 was identified using an anti-V5 monoclonal antibody and HRP-conjugated secondary antibody. The migration position of the Hinderin-V5 fusion protein present in the input cell lysate (200 μl) and in the immunoprecipitate is indicated by an arrow. Goat immunoglobulins are identified by an asterix. The presence of SMC1 and SMC3 in the Hinderin-V5 immunoprecipitate was assessed by incubation of cell lysates (500 μl) with 10 μg murine anti-V5 antibody. The immunocomplexes absorbed on agarose-protein G were analyzed on 8% SDS-PAGE and immunoblotted with either anti-SMC1 or anti-SMC3 antibody. SMC1 and SMC3 immunoblots of the input material (200 μl) are also illustrated. B) The effect of Hinderin on the interaction between the SMC3 and SMC1 hinge domains was investigated in 293 cells expressing different level of Hinderin by using a mammalian two-hybrid assay system. SMC3-474/702 and the SMC1-474/663 hinge domains in pBIND and pACT vectors respectively, were cotransfected in 293 cells together with Hinderin-V5 expression vector (0.3 or 1 μg/ml) or alternatively 1 μg/ml of the empty pcDNA3.1 vector (mock-transfected) and the reporter pG5/luc vector using Lipofectamine. Cells transfected with pBIND-Id and pACT-MyoD fusion protein expression vectors were used as control to assess the specificity of the Hinderin effect on protein-protein interaction. Luciferase activity was assayed after 48 h. The bars represent the mean ± SD of the values (n = 3). Semiquantitative RT-PCR was employed to assess the transcript level of GAL4:SMC3 and VP16:SMC1 fusion proteins and of G3PDH in cells ectopically expressing different levels of Hinderin.

Figure 4

Proposed model of interaction of Hinderin with SMC3. A) The five-domain structure of Hinderin is compared to that of SMC3. The different structural domains are drawn in scale to allow a direct comparison of the size of the terminal globular regions, the two coiled-coil domains and the central globular domain in the two proteins. B) Mode of interaction of the SMC1-SMC3 dimer. The juxtaposition of the cohesins hinge domains interacting through sites located at the globular-coiled coil domain boundaries (ref. 6) is illustrated. C) Postulated mechanism for the competitive binding of Hinderin to the hinge domain of SMC3. The N- and C-terminal globular domains of Hinderin are shown to interact with binding sites located on the SMC3 hinge. The Hinderin central globular domain is not involved in the binding to SMC3 but may play a role by orienting the N- and C-terminal globular domains toward their targets.