Membrane clustering and the role of rebinding in biochemical signaling

In many cellular signaling pathways, key components form clusters at the cell membrane. Although much work has focused on the mechanisms behind such cluster formation, the implications for downstream signaling remain poorly understood. Here, motivated by recent experiments, we study via particle-based simulation a covalent modification network in which the activating component is either clustered or randomly distributed on the membrane. We find that while clustering reduces the response of a single-modification network, clustering can enhance the response of a double-modification network. The reduction is a bulk effect: a cluster presents a smaller effective target to a substrate molecule in the bulk. The enhancement, on the other hand, is a local effect: a cluster promotes the rapid rebinding and second activation of singly active substrate molecules. As such, the enhancement relies upon frequent collisions on a short timescale, which leads to a diffusion coefficient at which the enhancement is optimal. We complement simulation with analytic results at both the mean-field and first-passage distribution levels. Our results emphasize the importance of spatially resolved models, showing that significant effects of spatial correlations persist even in spatially averaged quantities such as response curves.

💡 Research Summary

The paper investigates how the spatial organization of membrane‑bound activating enzymes influences the behavior of push‑pull signaling networks, using both particle‑based lattice simulations and analytical theory. The authors focus on two canonical network motifs: a single‑modification circuit, in which a substrate S is phosphorylated to S* by an activating enzyme Ea and dephosphorylated back by a deactivating enzyme Ed, and a double‑modification circuit, where S undergoes two successive modifications (S → S* → S**) with the same kinetic parameters for each step. Ea molecules are immobilized on the membrane either as a random distribution or as clusters of size N, while Ed and the substrate diffuse freely in a three‑dimensional cytoplasmic volume.

Key dimensionless parameters are introduced: χ = k3/k6 (relative catalytic activity of Ea versus Ed) as the input, φ as the steady‑state fraction of active substrate (output), and α, β, γ, η to describe network symmetry and sensitivity in the well‑mixed limit. Spatial parameters include the cluster size N, surface density μ, the ratio δ = k1/(4πaD) that compares intrinsic binding to diffusion‑limited binding, and ζ describing the depth of the cytoplasmic slab.

Simulation results reveal a striking dichotomy. In the single‑modification network, clustering always reduces the maximal response φ compared with a random arrangement. This “bulk effect” arises because a cluster presents a smaller effective target area to substrate molecules arriving from the bulk; consequently the encounter rate between substrate and Ea is lowered. In contrast, in the double‑modification network clustering can increase φ, sometimes dramatically. The authors attribute this “local effect” to rapid rebinding: after the first phosphorylation a partially active substrate (S*) remains near the cluster and is more likely to encounter another Ea molecule before diffusing away, thereby completing the second modification.

The enhancement depends sensitively on diffusion. When diffusion is very slow (small δ) the substrate cannot explore the cluster efficiently, so rebinding is rare and clustering offers no benefit. When diffusion is very fast (δ → ∞) spatial memory is lost and the system behaves as well‑mixed, again eliminating the advantage. At intermediate diffusion rates a maximal rebinding probability is achieved, producing an optimal diffusion coefficient at which the response enhancement peaks. This non‑monotonic dependence is captured analytically by first‑passage time distributions for a particle returning to a reactive patch.

The effect is also contingent on network parameters. The enhancement is strongest in linear‑sensitivity regimes (γ ≈ 1) where neither enzyme is saturated, and when the network is biased toward deactivation (α > β). In ultrasensitive regimes where enzymes become saturated, the substrate is already bound for long periods and the extra benefit of clustering disappears.

Analytical mean‑field calculations, extended to include spatial corrections, reproduce the simulation trends and provide explicit formulas linking φ to the dimensionless groups (α, β, γ, η, δ, μ, N). The authors show that increasing cluster size N amplifies both the reduction in the single‑modification case and the enhancement in the double‑modification case, but only up to a point; beyond a certain size the marginal gain diminishes because the cluster surface becomes saturated with substrate.

Biologically, the findings suggest that membrane clusters of proteins such as Ras or bacterial CheA can modulate signaling not merely by concentrating enzymes but by shaping the rebinding kinetics of substrates. In pathways that rely on distributive dual phosphorylation (e.g., MAPK cascades), clustering could provide a mechanism for tuning ultrasensitivity and response speed without altering protein expression levels. The work underscores that spatial correlations, often ignored in deterministic well‑mixed models, can have measurable impacts on average output curves, highlighting the necessity of spatially resolved modeling in cellular biophysics.

Overall, the study delivers a quantitative framework for understanding how membrane clustering and diffusion jointly regulate push‑pull signaling, revealing conditions under which clustering is detrimental, neutral, or beneficial, and offering predictions that can be tested experimentally in both prokaryotic and eukaryotic systems.