Effects of intersegmental transfers on target location by proteins

We study a model for a protein searching for a target, using facilitated diffusion, on a DNA molecule confined in a finite volume. The model includes three distinct pathways for facilitated diffusion: (a) sliding - in which the protein diffuses along the contour of the DNA (b) jumping - where the protein travels between two sites along the DNA by three-dimensional diffusion, and finally (c) intersegmental transfer - which allows the protein to move from one site to another by transiently binding both at the same time. The typical search time is calculated using scaling arguments which are verified numerically. Our results suggest that the inclusion of intersegmental transfer (i) decreases the search time considerably (ii) makes the search time much more robust to variations in the parameters of the model and (iii) that the optimal search time occurs in a regime very different than that found for models which ignore intersegmental transfers. The behavior we find is rich and shows surprising dependencies, for example, on the DNA length.

💡 Research Summary

The paper presents a quantitative model for how DNA‑binding proteins locate specific target sites when the DNA is confined within a finite cellular volume. Building on the classic facilitated‑diffusion framework, which traditionally includes only one‑dimensional sliding along the DNA contour and three‑dimensional “jumps” through the solution, the authors introduce a third pathway: intersegmental transfer (IT). In an IT event a protein simultaneously contacts two distinct DNA segments that are brought into spatial proximity by DNA looping or crowding, and then moves from one segment to the other without fully dissociating into the bulk.

The authors define three key kinetic parameters: the one‑dimensional diffusion coefficient D₁ governing sliding, the three‑dimensional diffusion coefficient D₃ governing jumps, and the probability p_IT that an encounter with a nearby DNA segment results in an intersegmental transfer. Characteristic lengths for each mode are introduced: λ_s = √(2D₁τ_s) for sliding, λ_j = √(6D₃τ_j) for jumps, and λ_IT, which scales with the average distance between contacting DNA segments and therefore with the DNA density ρ = L/V (L is the contour length, V the confining volume). The typical search time T is estimated as the product of the number of “search cycles” needed to scan the whole genome and the average duration of a single cycle.

A scaling analysis shows that when p_IT = 0 the model reduces to the well‑known result T ∝ L/(λ_s)·τ_s + L/(λ_j)·τ_j, i.e. the search time is dominated by the balance between sliding and jumping. When IT is allowed, the effective step length becomes λ_eff ≈ λ_s + λ_j + p_IT·λ_IT. Because λ_IT can be much larger than the product λ_s·λ_j, even a modest p_IT dramatically reduces the number of cycles required. In the IT‑dominated regime the search time scales as T ∝ (1/p_IT ρ a³)·(L⁻¹), where a is the characteristic DNA segment size. This inverse dependence on L is a striking departure from the linear dependence in the classic model.

To validate the scaling predictions, the authors perform extensive Monte‑Carlo simulations. DNA is represented as a random coil of length L confined in a cubic box of volume V; proteins undergo stochastic transitions among sliding, jumping, and IT according to the prescribed rates. Parameters are chosen to reflect typical bacterial conditions (D₁ ≈ 10⁶ bp² s⁻¹, D₃ ≈ 10 µm² s⁻¹, τ_s ≈ 10⁻⁴ s, τ_j ≈ 10⁻⁶ s). By varying L (10⁴–10⁶ bp), ρ, and p_IT (0–0.5), the simulations confirm that (i) the mean search time drops by an order of magnitude once p_IT exceeds ≈0.05, (ii) the dependence on D₁ becomes weak in the IT‑rich regime, and (iii) the optimal combination of D₁, D₃, and p_IT occurs at a point where the effective step length from IT matches the geometric mean of λ_s and λ_j, rather than the classic balance between sliding and jumping alone.

The discussion connects these theoretical findings to biological reality. In vivo DNA is highly compacted into chromatin or nucleoid structures, leading to substantial segment‑segment contacts and therefore large ρ a³ values. Under such conditions, intersegmental transfers are expected to be frequent, especially for small transcription factors that can bind two DNA segments simultaneously. The model predicts that cells can achieve rapid and robust target location without fine‑tuning the sliding diffusion coefficient, simply by exploiting the geometry of the DNA polymer. The authors suggest experimental tests using single‑molecule tracking or FRET‑based assays that can directly observe IT events.

In conclusion, the inclusion of intersegmental transfer fundamentally reshapes the landscape of facilitated diffusion. It not only accelerates the search process but also makes it far less sensitive to variations in microscopic parameters, and it shifts the optimal search regime to one where DNA geometry, rather than protein diffusion constants, is the dominant factor. This work provides a more complete theoretical framework for protein‑DNA target search and offers new perspectives for designing synthetic DNA‑binding systems or interpreting the kinetics of gene regulation in crowded cellular environments.