🦠 microbiology

Defective Viral Genomes Reveal History-dependent Self-regulation of Type I Interferon Dynamics

This study combines experimental data with a Bayesian computational model to demonstrate that defective viral genomes (DVGs) modulate Type I interferon dynamics through a history-dependent negative feedback mechanism, enabling pandemic viruses to replicate successfully by locally suppressing innate immune responses in neighboring cells despite high immunogenicity.

Original authors: Li, Y., Lourens, C., Nielsen, B. F., Levin, S., Perelson, A. S., Grenfell, B., te Velthuis, A. J. W.

Published 2026-10-31

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Li, Y., Lourens, C., Nielsen, B. F., Levin, S., Perelson, A. S., Grenfell, B., te Velthuis, A. J. W.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

When a virus invades the body, the immune system sounds an alarm. Specialized cells detect the invader and release a chemical signal called type I interferon. This signal acts like a general warning, telling nearby cells to lock down their doors and stop the virus from copying itself. It is a powerful defense, but it is also a double-edged sword. If the alarm rings too loudly or for too long, it can damage the body's own tissues, causing severe illness or even death. To survive an infection, the body must eventually turn the alarm down, allowing the immune system to reset and prevent self-inflicted harm. Yet, a puzzling mystery has long lingered in the study of respiratory viruses like influenza: how does the immune system know when to stop shouting, especially when the virus is still clearly present and multiplying?

For decades, scientists assumed that the level of the alarm signal was tied directly to the number of infected cells. The logic seemed simple: more infected cells meant more alarm, and as the number of infected cells dropped, the alarm would naturally fade. However, recent observations in both cell cultures and living mice showed something strange. In some infections, the alarm signal began to drop sharply while the virus was still spreading and the number of infected cells remained high. This disconnect suggested that the immune system was not just reacting to the current threat, but was somehow remembering the past. A new study by researchers at Princeton University and other institutions has now uncovered the mechanism behind this phenomenon, revealing that the immune system keeps a running tally of its own exposure to the alarm signal, and that this history determines when the signal is turned off.

The key to this discovery lies in a specific type of viral genetic material known as defective viral genomes. These are broken pieces of the virus that arise naturally during replication. They cannot reproduce on their own and often compete with the healthy virus for resources, but they are exceptionally good at triggering the immune alarm. In fact, infections containing high amounts of these defective pieces tend to produce a massive initial surge of the alarm signal. The researchers wanted to understand why, in these high-defect scenarios, the alarm signal would suddenly crash even while the virus and the broken pieces continued to rise. To solve this, they first turned to a controlled laboratory experiment using human cells. They exposed the cells to varying amounts of synthetic viral RNA and, separately, to different levels of the alarm signal itself. By measuring how the cells responded to new viral threats under these conditions, they found a clear pattern: the more alarm signal a cell had been exposed to in the past, the less responsive it became to new threats. The cells were not ignoring the virus; they were simply exhausted by the history of the alarm.

Building on this finding, the team developed a computer model to simulate what happens inside a living mouse during an influenza infection. They fed the model data from real mouse experiments where the virus was mixed with different amounts of these defective genomes. The model confirmed that the immune system does not just react to the current level of the alarm; it reacts to the total amount of alarm the system has experienced over time. When the body accumulates a high dose of the signal, it triggers a negative feedback loop that actively suppresses further production. This explains the mystery of the dropping alarm: in infections with many defective genomes, the initial surge is so intense that it quickly builds up a "history" of exposure, causing the system to shut down the alarm early, even while the virus is still active. The researchers found that this mechanism is crucial for the virus's survival. By forcing the immune system to turn off its own defenses prematurely, the virus creates a local environment where it can continue to spread to neighboring cells that have not yet been exposed to the alarm.

The study also explored how this dynamic affects potential treatments. The researchers simulated giving the mice extra doses of the alarm signal or introducing more defective genomes as a therapy. The results showed that the outcome depended entirely on the timing and the existing state of the infection. If the treatment was given too early, it could preserve too many healthy cells, allowing the virus to spread later. If given too late, the immune system might already be too exhausted to respond. The simulations suggested that there is no single perfect dose or moment for these interventions; instead, the success of a treatment relies on the complex, shifting balance between the virus, the broken pieces, and the immune system's memory of the battle. This work provides a clearer picture of how the immune system regulates itself, showing that the history of an infection is just as important as the current threat. It suggests that the virus does not just hide from the immune system, but actively manipulates the system's own rules to keep the doors open for its own spread.

Technical Summary: Defective Viral Genomes Reveal History-dependent Self-regulation of Type I Interferon Dynamics

Problem Statement
Type I interferons (IFN) are critical for clearing viral infections but must eventually decline to prevent immunopathology. While pandemic and highly pathogenic RNA viruses (e.g., H5N1, SARS-CoV-2) induce strong innate immune responses, they often replicate and spread successfully. A paradox exists in observed infection dynamics: in both in vitro and in vivo systems, IFN levels can decline while viral loads and defective viral genome (DVG) levels continue to rise. Previous within-host models typically attribute IFN decline to the depletion of infected cells or the clearance of extracellular IFN. However, these mechanisms fail to explain early IFN declines when viral stimuli remain high. The specific mechanisms by which DVGs—known strong inducers of IFN—modulate the negative feedback loops required to shut down IFN production remain poorly understood.

Methodology
The authors employed a multi-faceted approach combining in vitro experiments, computational modeling, and Bayesian inference:

In Vitro Reporter Assays: The authors utilized HEK293-Luc cells transfected with a synthetic 5′ppp hairpin RNA (a RIG-I ligand) to mimic viral RNA sensing without infectious replication. They titrated RNA doses against varying concentrations of exogenous IFN-β. Luciferase activity served as a readout for IFN-β promoter activation, isolating the effects of RNA stimulation and IFN history on subsequent induction from confounding factors like viral protein antagonism or target-cell depletion.
Computational Modeling: A system of ordinary differential equations (ODEs) was developed to describe within-host dynamics. The model tracks target cells ( $T$ $T$ ), infected cells ( $I$ $I$ ), full-length viral RNA ( $V$ $V$ ), DVGs ( $D$ $D$ ), Type I IFN ( $F$ $F$ ), and a novel state variable: cumulative IFN exposure ($cumF$).
- Feedback Mechanism: The model introduces a negative feedback term, $g(cumF) = 1 / (1 + cumF/cumF_{50})$ , which reduces the source of IFN induction based on the accumulated history of IFN exposure rather than instantaneous levels.
- DVG Integration: DVGs modulate IFN induction via a saturation function $s(D)$ and compete with standard virus for replication resources.
Model Calibration: The model was fitted using Bayesian inference (Stan, NUTS sampler) to published in vivo data from Penn et al. [20], which included longitudinal measurements of virus, DVGs, and IFN-α in BALB/c mice infected with influenza stocks containing low, medium, or high DVG loads.
Spatial and Intervention Simulations: A stochastic spatial model (building on Li et al. [45]) was used to simulate local IFN gradients and viral spread. Additionally, the model was perturbed to simulate DVG-based and IFN-based therapeutic interventions at various time points.

Key Results

Joint Regulation of IFN Induction: In vitro experiments demonstrated that IFN induction is non-linearly regulated by both the viral RNA stimulus and the history of IFN exposure. High levels of exogenous IFN suppressed subsequent RNA-induced IFN promoter activity at lower RNA doses, confirming that cumulative IFN exposure reduces the capacity for further induction.
Cumulative vs. Instantaneous Feedback: The in vivo model fitting revealed that a feedback mechanism driven by cumulative IFN exposure ($cumF$) was necessary to reproduce the observed early decline in IFN levels in high-DVG conditions. Models relying solely on instantaneous IFN levels or faster IFN clearance rates failed to capture this dynamic. The data suggests that the "memory" of IFN exposure, likely mediated by negative regulators like USP18 and SOCS proteins, drives the shutdown of IFN production even while the viral stimulus (DVGs) persists.
Dual Role of DVGs: DVGs exhibit a dual role: they act as potent inducers of IFN but also accelerate the onset of negative feedback. In high-DVG scenarios, the rapid accumulation of IFN triggers strong negative feedback, suppressing further IFN production. Paradoxically, this suppression can allow viral spread to continue locally.
Spatial Dynamics: Spatial simulations indicated that negative feedback creates a "microenvironment" where cells nearest the infection focus (receiving the highest IFN) become refractory to further IFN induction. Cells slightly further away, receiving lower IFN doses, remain responsive. This non-monotonic response allows the virus to spread in a wave-like manner through neighboring cells that are less protected, despite high systemic IFN levels.
Intervention Trade-offs: Simulations of DVG and IFN interventions showed that outcomes are highly dependent on the pre-existing infection state (DVG abundance and IFN history). Early DVG treatment in low-DVG infections could lower the viral peak but extend the infection duration by preserving susceptible target cells. Similarly, IFN treatment efficacy varied based on whether the host was already in a state of strong negative feedback.

Key Contributions

Mechanistic Insight: The paper identifies "history-dependent" negative feedback (cumulative IFN exposure) as a primary driver of early IFN decline during influenza infection, challenging models that rely solely on target-cell depletion or instantaneous IFN levels.
DVG Modulation: It elucidates how DVGs modulate the timing of the IFN response, acting as both triggers for activation and accelerators of self-regulation.
Spatial Explanation: The study provides a theoretical mechanism for how viruses can spread locally despite strong systemic immune responses, attributing this to the spatial heterogeneity of IFN refractoriness.
Model Framework: The authors provide a calibrated ODE framework that integrates DVG dynamics with cumulative IFN feedback, capable of reproducing complex in vivo trajectories where virus and IFN decouple.

Significance and Claims
The authors claim that their work explains two critical biological observations: (1) why IFN levels can decline while viral and DVG loads continue to rise, and (2) how pandemic viruses can successfully replicate despite inducing high levels of immunogenic RNA.

The paper posits that viral heterogeneity, specifically the presence of DVGs, modulates positive and negative feedback loops to shape infection outcomes. This suggests that the therapeutic use of DVGs (e.g., as therapeutic interfering particles) or IFN must be carefully calibrated, as the efficacy of such interventions depends heavily on the host's pre-existing IFN history and DVG burden. The authors conclude that understanding these feedback loops is essential for predicting infection dynamics and designing effective antiviral strategies, noting that their findings support further study into the therapeutic potential of viral RNA populations.

The authors remain modest regarding specific molecular mechanisms, noting that their "cumulative IFN" term is a phenomenological representation of processes like USP18 desensitization and receptor internalization, rather than a direct measurement of a single molecular species. They also acknowledge that their model focuses on the innate phase and does not explicitly model adaptive immunity or specific inflammatory cytokines like TNF, which may play roles in severe disease.

Technical Summary: Defective Viral Genomes Reveal History-dependent Self-regulation of Type I Interferon Dynamics

More like this