Data-driven model reveals increased stability of CAG-expanded huntingtin RNA due to MID1 binding

Y. Liu, Annika Reisbitzer, Domagoj Dorešić, Jan Hasenauer, Sybille Krauß, Tatjana TchumatchenkobioRxiv 2025 · DOI

Abstract

RNA-binding proteins (RBP) are important regulators of RNA metabolism. In neurodegenerative disorders such as Huntington’s Disease (HD), disrupted RBP-RNA interactions contribute to neuronal dysfunction. One such RBP, Midline 1 (MID1), has been shown to aberrantly associate with mutant huntingtin (Htt) RNA, enhancing its translation, yet the mechanism driving this effect remains unknown. Here, we develop a computational model to understand the role of MID1. Based on previously published data, our model predicts that MID1 increases the stability of the Htt RNA. We experimentally validate this prediction, showing that overexpression of MID1 significantly prolongs the half-life of mutant Htt RNA. Furthermore, we evaluate model refinements, including clustering of MID1-bound RNA, which allow capturing all key observations in the data. Together, we provide a data-driven framework that underlines the importance of RBP-RNA interaction in post-transcriptional regulation. This framework also shows how individual molecular reactions jointly determine RNA stability and protein levels in HD.

Introduction: The Role of MID1 in Huntington’s Disease

Huntington’s Disease (HD) is a neurodegenerative disorder caused by an expansion of CAG trinucleotide repeats in the Htt gene. These repeats translate into long glutamine chains that make the huntingtin protein toxic. However, the biology isn't just about the protein; the mutant Htt RNA itself folds into stable "hairpin" structures that attract various RNA-binding proteins (RBPs).

One specific RBP, Midline 1 (MID1), is known to bind these mutant hairpins and increase translation. Existing biological data showed "supralinear" patterns—where protein levels increased much faster than the number of mutant genes would suggest—that simple binding models couldn't explain. The authors sought to uncover the hidden mechanisms driving this disproportionate protein surge and to determine if MID1 influenced more than just the "speed" of the translation factory.

Known interactions between Htt RNA, MID1, and translation machinery.

Methods and Modeling Approach

The researchers employed a "sequence of models" approach, using Ordinary Differential Equations (ODEs) to simulate the biochemical reactions within a cell. They started with a Baseline Model (simple binding and translation) and progressively added complexity:

  1. Extended Model: Incorporated explicit RNA synthesis and degradation (turnover).
  2. Nonlinear Translation: Added Hill-type kinetics to account for threshold-dependent translation effects.
  3. Clustering: Modeled the physical grouping of MID1-bound RNA into localized subcellular domains (hubs) where translation efficiency is higher.

To calibrate these models, they used Maximum Likelihood Estimation (MLE) via the pyPESTO and AMICI frameworks. Finally, they validated their model's most surprising prediction—that MID1 increases RNA stability—using a "wet-lab" experiment. They treated HEK293 cells expressing mutant Htt with Actinomycin D (to stop new RNA production) and used qPCR to measure how fast the existing RNA decayed in the presence or absence of excess MID1.

Datasets

The models were primarily trained on in vitro and in vivo (mouse model) data from a landmark 2013 study by Krauß et al., which provided steady-state measurements of huntingtin protein across different CAG lengths and MID1 knockdown conditions. Additionally, the authors generated a new dataset measuring the half-life of Htt RNA with 83 CAG repeats (Q83) under control and MID1-overexpression conditions to validate the model's predictions.

Key Insights: The Protective Shield and Translation Hubs

The model's first major "aha" moment was the discovery that the degradation rate of MID1-bound RNA was significantly lower than that of free RNA. Binding to MID1 essentially "hides" the toxic RNA from the cell's recycling machinery, acting as a protective shield.

Furthermore, to explain why wild-type protein levels actually drop in the presence of mutant alleles (Observation 4 in Figure 4), the authors moved away from the "well-mixed soup" assumption. They proposed that mutant RNA and MID1 form clusters. These clusters hog the translation machinery (like S6 kinase), leaving the global pool (where wild-type RNA resides) starved of resources. This interaction between global resource depletion and cluster-based production creates a complex landscape where protein levels can dip and surge based on the "tipping point" of cluster formation.

Key Results

The Extended Model with Clustering and Nonlinear Translation was the only version capable of reproducing all six key experimental observations (see Figure 4), particularly the suppressed wild-type levels in heterozygous mice.

Experimental observations from Krauß et al. (2013) used for model training and validation.

Experimentally, the authors confirmed that without extra MID1, mutant Htt RNA has a half-life of roughly 2 hours. When MID1 was overexpressed, the RNA levels remained stable over the same period, implying a massive increase in stability. The model explains the "more than double" protein increase in homozygous mutants as a result of the combined effects of increased RNA stability and the high efficiency of translation clusters.

Conclusion and Future Directions

While the "clustering" hypothesis is mathematically robust and explains the observed data, it currently lacks direct visual evidence. The authors suggest that live-cell imaging and single-molecule RNA visualization are the necessary next steps to "see" these proposed translation hubs in action. Additionally, the model assumes a 1:1 binding ratio between MID1 and RNA, whereas expanded CAG repeats might recruit multiple MID1 molecules, potentially adding another layer of cooperative complexity. This data-driven framework underlines the importance of RBP-RNA interaction in post-transcriptional regulation and shows how individual molecular reactions jointly determine RNA stability and protein levels in HD.

Summary updated

Abstract

Abstract RNA-binding proteins (RBP) are important regulators of RNA metabolism. In neurode-generative disorders such as Huntington’s Disease (HD), disrupted RBP-RNA interactions contribute to neuronal dysfunction. One such RBP, Midline 1 (MID1), has been shown to aberrantly associate with mutant huntingtin ( Htt ) RNA, enhancing its translation, yet the mechanism driving this effect remains unknown. Here, we develop a computational model to understand the role of MID1. Based on previously published data, our model predicts that MID1 increases the stability of the Htt RNA. We experimentally validate this prediction, showing that overexpression of MID1 significantly prolongs the half-life of mutant Htt RNA. Furthermore, we evaluate model refinements, including clustering of MID1-bound RNA, which allow capturing all key observations in the data. Together, we provide a data-driven framework that underlines the importance of RBP-RNA interaction in post-transcriptional regulation. This framework also shows how individual molecular reactions jointly determine RNA stability and protein levels in HD.
BibTeX
@misc{liu2025data,
  author = {Y. Liu and Annika Reisbitzer and Domagoj Dorešić and Jan Hasenauer and Sybille Krauß and Tatjana Tchumatchenko},
  title = {{Data-driven model reveals increased stability of CAG-expanded <i>huntingtin</i> RNA due to MID1 binding}},
  year = {2025},
  doi = {10.1101/2025.10.31.685889},
  howpublished = {bioRxiv},
}
Figure