Developing a peptide to disrupt cohesin head domain interactions

Abstract

Cohesin mediates the 3-D structure of chromatin and is involved in maintaining genome stability and function. The cohesin core comprises Smc1 and Smc3, elongated-shaped proteins that dimerize through globular domains at their edges, called head and hinge. ATP binding to the Smc heads induces their dimerization and the formation of two active sites, while ATP hydrolysis results in head disengagement. This ATPase cycle is essential for driving cohesin activity. We report on the development of the first cohesin-inhibiting peptide (CIP). The CIP binds Smc3 in vitro and inhibits the ATPase activity of the holocomplex. Treating yeast cells with the CIP prevents cohesin's tethering activity and, interestingly, leads to the accumulation of cohesin on chromatin. CIP3 also affects cohesin activity in human cells. Altogether, we demonstrate the power of peptides to inhibit cohesin in cells and discuss the potential application of CIPs as a therapeutic approach.

Keywords: Cell biology; Molecular biology; Protein.

Graphical abstract

Figure 1

Cohesin’s SMC head domain The atomic structure of Smc1 (green) and Smc3 (blue). ATP molecules are in magenta (PDB: 6YUF). The CIP3-related region in Smc1 is in red. (A) Side view. (B) Top view. (C) Zoom-in into the CIP3-related region. (D) Schematic of the head engagement- and disengagement-inducing ATPase cycle. (E) Protein sequence alignments (ClustalX) of Smc1 show the conservation of the regions corresponding to CIP3.

Figure 2

CIP3 causes cell growth delay Strains yIO1000 (control) and yME-031 (pGAL-CIP1), yME-016 (pGAL-CIP2), and yME-019 (pGAL-CIP3) cells were grown in a galactose-containing medium to induce peptide expression. Cell growth was monitored every 120 min to measure the optical density of the culture at 600 nm. The growth rate in the logarithmic growth phase was calculated. The results of a representative experiment are shown. Multiple linear regression revealed no significant decrease in growth rate for cells expressing CIP1 (B = 0.015, SE = 0.119, p = 0.898) or CIP2 (B = −0.290, SE = 0.119, p = 0.022) in comparison with the control. A significant decrease in growth rate was found in cells expressing CIP3 (B = −0.727, SE = 0.119, p < 0.001).

Figure 3

CIP3-TAT binds to Smc3 in vitro (A) A structural model generated by HPEPDOC 2.0 server of CIP3-Smc3 head domain docking. Smc3 is in blue, ATP is in magenta, and CIP3 is in green. (B) Zoom in into the Smc3-CIP3 docking region. (C) The kinetic binding of the peptide, CIP3-TAT, to Smc3 head domain protein was explored by FEB Agile R100. 500 nM peptide was immobilized on the sensor chip, and the analyte, Smc3 head domain, in various concentrations (0.02–4 mM) was applied in solution to the chip. A graph showing the I-response in each analyte concentration without ATP (triangles) and with ATP (10 mM, circles). (D) K_d (in nM) and R² values were calculated from the results shown in A. (E) Time course analysis of ATP. Hydrolysis by cohesin with or without CIP3.

Figure 4

CIP3 inhibits sister chromatid cohesion and chromosome condensation (A) Flowchart outlining the sister chromatid cohesion assay. (B) yME-961 strain cells were grown and treated with peptides, as shown in A. Premature sister chromatid separation was determined by counting the number of cells showing two GFP dots. At least 300 cells were counted in 3 independent experiments. ∗∗p < 0.001. (C) Cells were grown to mid-log phase. The culture was divided into two flasks, one being treated with CIP3-TAT. Condensation was determined in cells at the G2/M phase by 2-photon microscopy. ∗∗p < 0.001. (D) yKS-008 strain (Smc3-V5) cells were grown and treated with CIP3-TAT peptide, as shown in A. Cells were processed for ChIP with antibodies against V5.

Figure 5

CIP3 causes a mitotic delay in human cells U2OS cells were plated in a 96-well plate and grown for 20 h to allow adherence. CIP3 peptide was added to the growth medium, and cells were grown for additional 20 h. Images were taken every 5 min. Mitosis length was determined by counting the number of images in which mitotic chromosomes were visualized. (A) Representative image. (B) Quantitation of mitosis length in untreated and CIP3-TAT treated cells. At least 300 cells were counted for each condition. ∗∗∗p < 0.0001.