DNA end tethering through break-induced DNA--protein condensation

Technical Summary: DNA End Tethering through Break-Induced DNA–Protein Condensation

Problem Statement
Cells rely on robust mechanisms to repair DNA double-strand breaks (DSBs) to maintain genomic stability. A critical prerequisite for successful repair is the physical tethering of the two broken DNA ends to prevent their dispersion. Recent experimental work has demonstrated that DNA–protein co-condensation, specifically involving PARP1, can tether broken ends in vitro. However, a fundamental biophysical puzzle remains: how does such condensation reliably achieve spatiotemporal localization specifically at the break site and capture both broken ends, given the intrinsic stochasticity of the system and the high abundance of repair proteins that could theoretically condense anywhere? While the physics of DNA–protein condensation (driven by capillary forces and polymer elasticity) is understood, the mechanism ensuring rapid, localized nucleation at the break to achieve tethering is unclear.

Methodology
The authors employ a combination of Brownian dynamics simulations and theoretical modeling to investigate the minimal physical requirements for break-induced tethering.

1. Brownian Dynamics Simulations:

   • System: A bead-and-spring polymer model embedded in a viscous medium. The polymer is initially intact and pinned at its termini.
   • Break Event: At $t=0$, the polymer is symmetrically broken, creating two broken ends labeled as species $E$ (formerly $D$).
   • Catalytic Conversion: The model posits that broken ends ($E$) act as catalysts, converting soluble proteins (state $A$, treated as a background field) into a condensation-competent state ($B$) at a rate $k$.
   • Interactions: Beads interact via a truncated Lennard-Jones 8-4 potential. Parameters are tuned such that the polymer and $B$ beads do not condense spontaneously ($\epsilon_1, \epsilon_2$), but co-condense when interacting ($\epsilon > 0$).
   • Dynamics: Bead positions evolve according to an overdamped Langevin equation. The system tracks the separation of broken ends and the formation of condensates.

2. Theoretical Modeling:

   • Free Energy Landscape: The authors construct an effective free energy landscape $F(s_{out}, Z)$, where $s_{out}$ is the contour length of the polymer outside the condensate and $Z$ is the extension of the chain.
   • Components: The free energy includes terms for polymer elasticity (using the Marko-Siggia force-extension relation for semi-flexible chains), condensation free energy density ($\nu$), and interfacial energy ($\gamma$) accounting for surface tension and capillary forces during condensate fusion.
   • Dynamics: Linear response equations describe the evolution of $s_{out}$ and $Z$, governed by mobilities $\Gamma_s$ (condensation) and $\Gamma_Z$ (polymer retraction).

Key Contributions and Results

• Break-Induced Catalysis as a Nucleation Mechanism: Simulations confirm that if broken ends catalyze the conversion of soluble proteins to a condensation-competent state, condensates form locally at the break site. Without this localized catalysis (or without phase separation interactions), tethering fails.
• Kinetic Competition Governs Outcome: The study identifies a kinetic competition between two processes:
  1. Condensation Dynamics: The growth of polymer–protein condensates at the broken ends.
  2. Polymer Relaxation: The retraction of the broken chain segments due to entropic elasticity.
  • Success: If condensation is fast relative to polymer retraction (high $k$, low initial extension $L$), the condensates grow large enough to fuse, tethering the ends.
  • Failure: If polymer retraction is faster than condensation (low $k$, high $L$), the broken ends move apart before the condensates can fuse, leading to tethering failure.
• Stochastic vs. Reliable Regimes: The simulations reveal that for intermediate parameters, the outcome is stochastic (partial success). However, specific parameter regimes exist where tethering is reliable even in the presence of noise.
• Free Energy Landscape Analysis: The theoretical model maps the system's evolution onto a free energy landscape with distinct basins of attraction:
  • A success state (global or metastable minimum) where $Z > 0$ and $s_{out} > 0$.
  • A failure state where $Z \to 0$ and $s_{out} \to 0$.
  • A saddle point separating these states. The stability of the success state depends on the initial extension $L$; as $L$ increases, the energy barrier protecting the success state diminishes, eventually causing the success state to disappear.
• Quantitative Agreement with Experiments: By applying the model parameters to experimental data from Chappidi et al. (2024) regarding PARP1-DNA co-condensation, the authors calculate a stationary state consistent with observed tethering distances and predict a tethering force of approximately 2.3 pN.

Significance and Claims
The paper claims to provide a physical mechanism explaining how DNA end tethering is achieved reliably despite stochasticity. The central argument is that DNA end-dependent conversion of proteins (from soluble to condensation-competent) coupled with DNA–protein condensation is sufficient to tether broken ends.

The authors emphasize that:

1. Localized condensation at the broken ends is both necessary and sufficient for sustained tethering.
2. The mechanism relies on the broken ends acting as catalysts to trigger a conformational change or cooperative interaction in proteins (e.g., PARP1), ensuring nucleation occurs precisely at the damage site.
3. The reliability of the repair process is governed by the kinetic ratio between condensation speed and polymer relaxation.
4. While the molecular details of the trigger remain unclear, the coarse-grained physical model successfully rationalizes experimental observations of PARP1-mediated tethering.

The work does not propose new experimental protocols or claim to solve all aspects of in vivo repair but establishes a minimal physical framework for understanding the biophysics of DNA end capture. Future work is noted as required to incorporate noise into the theoretical equations to fully explore stochastic regimes and to elucidate the specific molecular triggers of the protein conversion.