A Persistent Random-Walk Model of Molecular Transport in Neuronal Dendritic Trees

Technical Summary: A Persistent Random-Walk Model of Molecular Transport in Neuronal Dendritic Trees

Problem Statement
The transport of messenger RNA (mRNA) from the neuronal soma to synaptic targets is critical for synaptic plasticity, learning, and memory. While experimental observations confirm that mRNA cargo motion is stochastic, characterized by intermittent "run-and-pause" dynamics and directional reversals, existing theoretical frameworks have not fully integrated these microscopic dynamics with target-specific first-passage analysis on reconstructed, branched dendritic morphologies. Classical diffusion models are insufficient for long-range transport, and previous intermittent search models often rely on idealized geometries rather than experimentally reconstructed neuronal structures. The core problem addressed is how local run-pause dynamics and directional persistence translate into the probability and time required for individual mRNA molecules to reach specific synaptic regions within a complex dendritic tree.

Methodology
The author develops a two-level analytical framework to bridge microscopic cargo dynamics and macroscopic transport on the dendritic tree:

1. Microscopic Level (Segment Dynamics):

   • Model: The motion of mRNA along an individual dendritic segment is modeled using a Persistent Telegraph Process with Pauses (PTPP). This continuous-time Markov process operates in a combined state space of continuous position ($x$) and four discrete internal states: anterograde running ($r+$), retrograde running ($r-$), anterograde paused ($p+$), and retrograde paused ($p-$).
   • Parameters: The model incorporates constant velocity ($v$), exponentially distributed run durations (rate $\lambda$) and pause durations (rate $\mu$), and directional persistence probabilities ($\alpha$ for anterograde, $\beta$ for retrograde) governing the direction chosen after a pause.
   • Derivation: Using infinitesimal generator operators, the author derives analytical expressions for:
     • Absorption Probabilities: The probability of traversing a segment of length $l$ to the distal end versus returning to the entry point.
     • Conditional Mean First-Passage Times (CMFPTs): The average time to reach a specific boundary, conditioned on successful arrival at that boundary (excluding trajectories absorbed at the other end).
   • Key Analytical Result: The author derives asymptotic expressions for CMFPTs, showing that under asymmetric persistence, the CMFPT scales linearly with segment length, contrasting with the quadratic scaling of diffusive transport.

2. Macroscopic Level (Tree Dynamics):

   • Model: The transport across the entire dendritic tree is described by a coarse-grained semi-Markov model.
   • Structure: Nodes represent the soma, branching points, and synaptic targets (modeled as absorbing states). Dendritic segments act as links.
   • Transition Mechanics: Transition probabilities between nodes are derived from the segment-level traversal and return probabilities. Sojourn times (waiting times) are determined by the segment-level CMFPTs. Unlike continuous-time Markov chains, this semi-Markov formulation allows the waiting time at a node to vary depending on where the process arrived from and where it goes next.
   • Calculation: The author utilizes the theory of absorbing Markov chains, constructing a fundamental matrix and a weighted sojourn time matrix to compute target-specific absorption probabilities and CMFPTs from the soma to any synaptic target.

Key Contributions

• Unified Framework: The paper presents the first framework combining experimentally characterized run-pause dynamics and directional persistence at the segment level with target-specific first-passage analysis on reconstructed dendritic morphologies.
• Analytical Expressions: Closed-form analytical expressions are derived for segment traversal probabilities and conditional mean first-passage times under both symmetric and asymmetric directional persistence.
• Counterintuitive Finding on Persistence: The model reveals that stronger directional persistence in a given direction does not necessarily lead to faster transport in that direction. Specifically, under certain parameter regimes (particularly when retrograde persistence dominates), the conditional mean travel time to the opposite end can be shorter than in the opposite direction, despite the probability of successful arrival being exponentially smaller.
• Identification of Transport Regimes: The framework identifies two distinct asymptotic transport regimes for long dendritic segments based on the dominance of anterograde ($\alpha$) versus retrograde ($\beta$) persistence.

Results
Numerical simulations using parameters derived from experimental data (e.g., $v \approx 1.0 \mu m/s$, $\lambda = 4.4 s^{-1}$, $\mu = 12.5 s^{-1}$) and reconstructed neuronal morphologies (human Purkinje and mouse pyramidal neurons) yield the following results:

1. Retrograde Dominance Regime ($\beta > \alpha$):

   • When retrograde persistence dominates, the conditional mean first-passage time (CMFPT) to successfully reached synaptic targets becomes weakly dependent on the distance from the soma.
   • The system enters a quasi-stationary state within the transient dendritic network for a prolonged period before absorption.
   • While the time to reach a target is relatively constant regardless of distance, the probability of successfully reaching distal targets decreases exponentially with distance. Thus, the primary limitation in this regime is the low probability of realization, not the travel time.

2. Anterograde Dominance Regime ($\alpha > \beta$):

   • When anterograde persistence dominates, transport resembles motion along a single extended path.
   • The CMFPT increases approximately linearly with the soma-to-synapse distance.
   • The slope of this linear growth is determined by the persistence parameters and the velocity.

3. Validation:

   • Monte Carlo simulations of the PTPP on single segments confirm the theoretical predictions for both symmetric and asymmetric cases.
   • The model successfully reproduces the linear scaling of CMFPT with distance for anterograde dominance and the distance-independence for retrograde dominance across different neuronal morphologies.

Significance and Claims
The paper claims that the relative strength of anterograde and retrograde persistence is a critical regulatory parameter for dendritic mRNA localization. The findings suggest that:

• In the retrograde-dominant regime (which aligns with experimentally estimated parameters), the "cost" of reaching distal targets is not the time required but the low probability of successful delivery. This raises questions about the efficiency of mRNA localization at remote synaptic regions in vivo and suggests potential biological mechanisms (e.g., activity-dependent changes in mobility or repeated transport attempts) to overcome this limitation.
• The framework provides a theoretical basis for understanding how local stochastic dynamics scale up to determine the spatial distribution of transcripts in complex neuronal geometries.
• The model serves as a foundation for future extensions, such as incorporating mRNA degradation, non-uniform branching probabilities, and target-specific capture probabilities, to better link stochastic transport dynamics with experimentally measured spatial profiles of dendritic mRNA.

The author emphasizes that while the model treats terminal dendritic tips as synaptic targets for numerical implementation, the framework is flexible enough to accommodate synaptic targets located within dendritic branches. The approach is also noted as potentially applicable to other stochastic transport problems on filamentous and network-like structures.