How Retroactivity Affects the Behavior of Incoherent Feedforward Loops

Junmin Wang, Călin Belta, Samuel A. Isaacson

TL;DR

Incoherent Feed-Forward Loops (IFFLs) are crucial biological circuits for rapid responses. This paper reveals that "retroactivity" (where downstream elements affect upstream ones) can actually improve IFFL performance, leading to faster responses and larger pulses, unlike in negative autoregulatory loops where it only slows things down. This flexibility might explain why IFFLs are so common across life.

What Problem Does This Paper Address?

Biological systems rely on complex networks of genes and proteins to sense and respond to their environment. A key question in systems biology is how recurring network structures, known as "network motifs," contribute to these functions. The Incoherent Feed-Forward Loop (IFFL) is a well-studied motif known for accelerating responses and generating pulses. However, most previous studies used simplified models that ignored "retroactivity." Retroactivity occurs when downstream binding sites (e.g., where a transcription factor binds to DNA) can influence the behavior of the upstream transcription factor itself. This is particularly relevant in eukaryotic cells where transcription factors can bind to a vast number of non-functional "decoy" sites. The authors aim to understand how varying levels of retroactivity, especially in contexts with high retroactivity, affect the dynamics and performance of IFFLs.

Methods

The researchers employed computational modeling using Ordinary Differential Equations (ODEs) to simulate IFFL behavior. They specifically adapted a framework developed by Gyorgy and Del Vecchio (2014) to explicitly incorporate retroactivity. This involved creating non-dimensionalized ODE models for different IFFL topologies (I1-FFL and I4-FFL) and systematically varying the levels of retroactivity on different nodes within the network. They performed extensive simulations across a wide range of kinetic parameters to ensure their findings were robust and parameter-independent. Mathematical proofs were also used to rigorously demonstrate key observations. To compare IFFLs with other motifs, they also modeled negative autoregulatory circuits and two-input circuits.

Datasets

Not discussed in the paper. The study relies on computational simulations and mathematical analysis rather than experimental datasets for training or evaluation.

Key Insights and Ideas

  • Retroactivity as a Feature, Not Just a Bug: Traditionally viewed as an impediment to robust synthetic circuit design, retroactivity in IFFLs can be harnessed to improve performance, leading to faster responses and larger pulses. This is a significant departure from the common understanding.
  • IFFL Flexibility: Unlike negative autoregulation, where retroactivity consistently slows down the response, IFFLs exhibit a much more nuanced response. Increasing retroactivity can speed up, slow down, or have no effect on response time and pulse amplitude, depending on the specific node and parameters.
  • Evolutionary Advantage: The authors hypothesize that this functional versatility of IFFLs under varying retroactivity levels might have provided them a selective advantage, contributing to their conserved abundance across diverse organisms, from bacteria to humans.
  • Topology Matters: The specific type of IFFL (e.g., I1-FFL vs. I4-FFL) and the logic of gene regulation (AND vs. OR) significantly influence how retroactivity affects its behavior.

Key Results

  • Increasing retroactivity on the regulatory node B in an I1-FFL was shown to shorten response times and increase pulse amplitude.
  • In contrast, increasing retroactivity in a negative autoregulatory circuit only slowed down the response time.
  • The effects of retroactivity were found to be parameter-independent, supported by mathematical proofs.
  • Simulations exploring a wide parameter space confirmed that the impact of retroactivity on IFFL response times is significant across large regions.

Limitations and Future Work

  • Minimal Model: The study uses a minimal model of IFFLs and excludes certain binding logic types (e.g., competitive logic) when two transcription factors bind to the same promoter.
  • Stochasticity: The current modeling framework is based on ODEs, which represent deterministic dynamics. Future work could explore the effects of retroactivity in stochastic gene expression models, which account for transcriptional bursting and noise.
  • Network Context: The authors suggest exploring how IFFLs interact with other network motifs (like negative feedback loops) and how retroactivity influences these combined systems.
  • Experimental Validation: While the authors mention the feasibility of creating synthetic IFFLs, direct experimental validation of their findings on retroactivity's beneficial effects would be a crucial next step.

Summary updated

Abstract

An incoherent feedforward loop (IFFL) is a network motif known for its ability to accelerate responses and generate pulses. It remains an open question to understand the behavior of IFFLs in contexts with high levels of retroactivity, where an upstream transcription factor binds to numerous downstream binding sites. Here we study the behavior of IFFLs by simulating and comparing ODE models with different levels of retroactivity. We find that increasing retroactivity in an IFFL can increase, decrease, or keep the network's response time and pulse amplitude constant. This suggests that increasing retroactivity, traditionally considered an impediment to designing robust synthetic systems, could be exploited to improve the performance of IFFLs. In contrast, we find that increasing retroactivity in a negative autoregulated circuit can only slow the response. The ability of an IFFL to flexibly handle retroactivity may have contributed to its significant abundance in both bacterial and eukaryotic regulatory networks.
BibTeX
@inproceedings{wang2020how,
  author = {Junmin Wang and Călin Belta and Samuel A. Isaacson},
  title = {{How Retroactivity Affects the Behavior of Incoherent Feedforward Loops}},
  booktitle = {iScience},
  year = {2020},
  doi = {10.1016/j.isci.2020.101779},
}
Figure