An autoinhibited state of 53BP1 revealed by small molecule antagonists and protein engineering

Abstract

The recruitment of 53BP1 to chromatin, mediated by its recognition of histone H4 dimethylated at lysine 20 (H4K20me2), is important for DNA double-strand break repair. Using a series of small molecule antagonists, we demonstrate a conformational equilibrium between an open and a pre-existing lowly populated closed state of 53BP1 in which the H4K20me2 binding surface is buried at the interface between two interacting 53BP1 molecules. In cells, these antagonists inhibit the chromatin recruitment of wild type 53BP1, but do not affect 53BP1 variants unable to access the closed conformation despite preservation of the H4K20me2 binding site. Thus, this inhibition operates by shifting the conformational equilibrium toward the closed state. Our work therefore identifies an auto-associated form of 53BP1-autoinhibited for chromatin binding-that can be stabilized by small molecule ligands encapsulated between two 53BP1 protomers. Such ligands are valuable research tools to study the function of 53BP1 and have the potential to facilitate the development of new drugs for cancer therapy.

Fig. 1. Interaction of 53BP1^TT with small molecules probed using X-ray crystallography.

a Domain structure of 53BP1. OD stands for oligomerization domain and UDR for ubiquitin-dependent recognition motif. 53BP1^TT is the tandem Tudor domain of 53BP1. 53BP1^FFR is the minimal foci-forming region of 53BP1. b Crystal structure showing the encapsulation of UNC3351 by two 53BP1^TT molecules. c Binding interfaces of UNC2991, UNC3351 and UNC3474 with 53BP1^TT in the crystal structures of their complexes. The 2mFo-DFc electron density maps contoured at 1σ level are shown as gray mesh around each compound. d Binding interfaces in the structure of 53BP1^TT bound to UNC3351 (left and middle panels) and to an H4K20me2 peptide (right panel). The side chains in the binding cage of 53BP1^TT are shown.

Fig. 2. Oligomerization of 53BP1^TT-UNC3474 probed using analytical ultracentrifugation (AUC) and NMR spectroscopy.

a Sedimentation velocity analysis of 53BP1^TT (20 μM) at different concentrations of UNC3474 using AUC. The sedimentation coefficient distributions are shown for free 53BP1^TT (black) and for 53BP1^TT at increasing concentrations of UNC3474; 10 μM (blue), 20 μM (light blue), 40 μM (yellow), 80 μM (orange) and 200 μM (red). b Top: Overlay of ¹H-¹⁵N HSQC NMR spectra of ¹⁵N-labeled 53BP1^TT, free (black) and bound to UNC3474 (red). Bottom: Examples of doubling of resonances (L1517 and F1519 signals) at different 53BP1^TT:UNC3474 molar ratios. c ¹H-¹⁵N-heteronuclear ZZ-exchange NMR spectroscopy of the interaction of 53BP1^TT with UNC2170 at 25 °C using 1 mM ¹⁵N-labeled 53BP1^TT and 9 mM UNC2170. Fitting was done for the signals of E1524 (inset). The black and red curves show the decay of the auto peaks (AA and BB) and blue and green curves show the buildup and decay of the exchange peaks (AB and BA). The interconversion rates are indicated.

Fig. 3. Small molecules stabilize a pre-existing lowly populated 53BP1^TT homodimer.

a Key inter-53BP1^TT contacts in the structure of 53BP1^TT-UNC3474 complex. Hydrogen bonds and salt bridges are represented by yellow dashed lines. b Structural overlay of wild-type 53BP1^TT and mutant 53BP1^TT-PN in which E1549 and D1550 are replaced by a proline and an asparagine, respectively. c NMR spectroscopy-monitored titration of ¹⁵N-labeled 53BP1^TT-PN with non-labeled H4K_C20me2 and p53K382me2 peptides and small molecule UNC1078. Shown in different colors are the overlaid ¹H-¹⁵N HSQC spectra of 53BP1^TT-PN recorded without (black) and with increasing amounts of added compounds, up to fourfold molar excess (red). d Same as c but for the titration of 53BP1^TT-PN with small molecules UNC2170, UNC2991, and UNC3474.

Fig. 4. Engineering of a disulfide-crosslinked 53BP1^TT homodimer that binds UNC3474.

a Sedimentation velocity analysis of 53BP1^TT-CC (20 μM) in the presence and absence of 10 mM DTT. b Crystal structure of 53BP1^TT-CC (blue and orange) showing the two disulfide bridges overlaid to the crystal structure of 53BP1^TT-UNC3474 (UNC3474 omitted) in gray. c Effects of adding 2 mM DTT and UNC3474 on the ¹H-¹⁵N HSQC spectrum of ¹⁵N-labeled 53BP1^TT-CC (1:1 UNC3474:53BP1^TT-CC molar ratio).

Fig. 5. Demonstration of an autoinhibited state for 53BP1 chromatin recruitment in cells using UNC3474 as a chemical probe.

a Immunoblotting showing the expression of the indicated 53BP1^FFR construct in stably transfected U2OS cells. β-actin staining was used as control. Source data are provided as a Source Data file. b Representative data for the effect of UNC3474 on WT 53BP1^FFR, 53BP1^FFR -PN and 53BP1^FFR-CC IRIF formation after cell exposure to 1 Gy of X-ray radiation. The scale bar represents 10 μm. The experiments were repeated three times independently with similar results. c Effect of UNC3474 on WT 53BP1^FFR, 53BP1^FFR-PN and 53BP1^FFR-CC IRIF number. IRIF from more than 100 cells were counted in each condition. Data represent mean values ± s.d. A representative plot of n = 3 independent experiments is shown. P values were calculated using a two-tailed t-test. For (c) and (d), the labels *, ** and *** indicate P < 0.05, P < 0.01 and P < 0.001, respectively, and ns means not significant. Source data are provided as a Source Data file. d Number of γH2A.X IRIF for the cells analyzed in (c). A representative plot of n = 3 independent experiments is shown. Source data are provided as a Source Data file.