Mutant induced neurons and humanized mice enable identification of Niemann-Pick type C1 proteostatic therapies

Ruth D. Azaria, Adele B. Correia, Kylie J. Schache, M. Camila Fernández Zapata, Koralege C. Pathmasiri, Varshasnata Mohanty, Dharma Theja Nannapaneni, Brandon L. Ashfeld, Paul Helquist, Olaf Wiest, Kenji Ohgane, Qingqing Li, Ross A. Fredenburg, Brian S. J. Blagg, Stephanie M. Cologna, Mark L. Schultz, Andrew P. LiebermanJCI 2024 · DOI

TL;DR

Researchers have developed novel human induced neurons and a humanized mouse model to better test potential therapies for Niemann-Pick type C1 disease, a severe lysosomal disorder. They identified a promising compound, mo56-hydroxycholesterol (mo56Hc), that can help correct the protein misfolding and trafficking defects characteristic of the disease.

What Problem Does This Paper Address?

Niemann-Pick type C1 (NPC1) disease is a devastating genetic disorder characterized by the buildup of cholesterol in lysosomes, leading to progressive neurodegeneration and early death. The disease is caused by mutations in the NPC1 gene, which encodes a crucial protein responsible for cholesterol export from late endosomes and lysosomes. Many of these mutations lead to misfolded NPC1 proteins that are retained in the endoplasmic reticulum (ER) and degraded, preventing them from reaching their functional location.

Developing effective treatments for NPC1 disease has been challenging, partly due to limitations in existing model systems. Previous studies have shown that mouse models don't always accurately reflect human disease phenotypes, especially concerning how mutant NPC1 proteins are handled and how they respond to potential therapies. Furthermore, testing therapies in non-neuronal cells might not translate well to the brain, a primary target of NPC1 pathology. This paper aims to address these limitations by creating more accurate human-relevant models to identify and validate proteostatic therapeutics – drugs that help proteins fold correctly and traffic to their intended destinations.

Methods

The researchers employed a multi-pronged approach:

  • Human Induced Neurons (iNeurons): They generated isogenic human induced pluripotent stem cell-derived neurons (iNeurons) carrying specific NPC1 missense mutations. This allowed them to study the disease in a human neuronal context, which is critical given the neurodegenerative nature of NPC1 disease. They used CRISPR/Cas9 gene editing to introduce these mutations, ensuring a precise genetic background.
  • Compound Rescreening: They rescreened a panel of small molecules previously reported to correct NPC1 folding and trafficking. This involved treating the engineered iNeurons with these compounds and assessing NPC1's ability to escape ER quality control and reach the Golgi apparatus. A key assay used was the Endo H (Endoglycosidase H) resistance assay, which distinguishes between high-mannose glycans (added in the ER and sensitive to Endo H) and complex glycans (added in the Golgi and resistant to Endo H). Increased Endo H resistance indicates successful trafficking beyond the ER.
  • Humanized Mouse Model: Recognizing the limitations of existing mouse models, they engineered a novel "humanized" mouse. They used CRISPR/Cas9 to insert the full-length human I1061T NPC1 cDNA into the mouse Npc1 gene. The I1061T mutation is the most common cause of NPC1 disease in individuals of Western European ancestry. This humanized model was designed to better recapitulate human NPC1 protein defects and responses to therapies.
  • In Vivo Validation: They characterized the humanized mice for disease phenotypes, including motor impairment, neurodegeneration, and lipid accumulation. They also tested the efficacy of the identified therapeutic compound (mo56Hc) in this new mouse model.
  • Thermal Stability Assays: To understand the mechanism of action of the identified therapeutic, they used nano-differential scanning fluorimetry to assess the thermal stability of purified NPC1 protein in the presence of different compounds. This technique helps determine how well a compound stabilizes the protein's structure.

Datasets

The paper does not explicitly mention the use of large, pre-existing datasets for training or evaluation in the machine learning sense. Instead, it focuses on generating and utilizing experimental data from:

  • Isogenic human iNeurons: These cell lines, engineered with specific NPC1 mutations, served as a primary platform for compound screening.
  • Humanized I1061T NPC1 mice: This newly generated mouse model provided in vivo data for disease characterization and therapeutic validation.
  • Human and mouse fibroblasts: These cell lines were used for comparative studies and validation of findings from iNeurons and mice.

The study introduces a new benchmark for testing proteostatic therapies for NPC1 disease by providing well-characterized human iNeurons and a humanized mouse model that more accurately reflect human disease mechanisms.

Key Insights and Ideas

  • Human iNeurons as a Superior Model: The study highlights that human iNeurons are a more relevant model for testing NPC1 proteostatic therapies than previously used cell types, especially for understanding neuronal pathology.
  • Species-Specific Differences Matter: The critical finding that human I1061T NPC1 protein behaves differently from mouse I1061T NPC1 protein in terms of trafficking and response to therapies underscores the importance of using human-relevant models. This explains why previous mouse models were less effective for testing certain drugs.
  • mo56Hc as a Potent Chaperone: mo56-hydroxycholesterol (mo56Hc) emerged as a highly effective pharmacological chaperone, significantly improving the folding and trafficking of multiple NPC1 mutants, including the common I1061T variant.
  • Mechanism of mo56Hc Action: Thermal stability assays suggest mo56Hc stabilizes NPC1 protein by acting on its core structure, likely facilitating its proper conformation for trafficking, rather than competing with cholesterol binding.
  • Humanized Mouse Model's Predictive Power: The newly developed humanized I1061T NPC1 mouse model successfully recapitulates key disease phenotypes and, crucially, responds to mo56Hc treatment, validating its utility for in vivo drug testing.

Key Results

  • Identification of mo56Hc: Out of several tested compounds, mo56Hc was identified as a potent pharmacological chaperone that significantly improved NPC1 trafficking in human iNeurons and fibroblasts.
  • Rescue of Multiple Mutants: mo56Hc demonstrated efficacy in rescuing the trafficking defects of several ER-retained NPC1 mutants (I1061T, R934L, R1186H) in iNeurons.
  • Humanized Mouse Phenotypes: The humanized I1061T NPC1 mice exhibited age-dependent motor impairment, diminished body weight, reduced survival, and cerebellar degeneration, mirroring aspects of NPC1 disease.
  • Therapeutic Response in Humanized Mice: The humanized mice and their derived fibroblasts showed a positive response to mo56Hc treatment, with improved NPC1 trafficking and reduced cholesterol accumulation, similar to observations in human cells.
  • Species-Specific Drug Response: The study confirmed that human I1061T NPC1 protein responds to mo56Hc, while mouse I1061T NPC1 protein does not, reinforcing the need for human-relevant models.

Limitations and Future Work

  • Limited Compound Efficacy: While mo56Hc showed promise, the paper notes that other tested compounds did not exhibit significant efficacy in their iNeuron model, suggesting that more research is needed to identify a broader range of effective therapeutics.
  • Mechanism of mo56Hc: While thermal stability assays provided insights, the precise binding site and detailed molecular interactions of mo56Hc with NPC1 are still areas for further investigation.
  • Disease Progression in Humanized Mice: Although the humanized mice recapitulated key phenotypes, the disease progression in these models might still occur on a different timeline compared to human patients, requiring careful long-term studies.
  • Broader Therapeutic Landscape: The authors suggest that their developed models (iNeurons and humanized mice) provide a robust platform for the continued testing and validation of proteostatic therapeutics, paving the way for the development of disease-modifying therapies for NPC1 disease. Future work will likely focus on optimizing mo56Hc or identifying other novel compounds using these advanced models.

Summary updated

Abstract

Therapeutics that rescue folding, trafficking, and function of disease-causing missense mutants are sought for a host of human diseases, but efforts to leverage model systems to test emerging strategies have met with limited success. Such is the case for Niemann-Pick type C1 disease, a lysosomal disorder characterized by impaired intracellular cholesterol trafficking, progressive neurodegeneration, and early death. NPC1, a multipass transmembrane glycoprotein, is synthesized in the endoplasmic reticulum and traffics to late endosomes/lysosomes, but this process is often disrupted in disease. We sought to identify small molecules that promote folding and enable lysosomal localization and functional recovery of mutant NPC1. We leveraged a panel of isogenic human induced neurons expressing distinct NPC1 missense mutations. We used this panel to rescreen compounds that were reported previously to correct NPC1 folding and trafficking. We established mo56-hydroxycholesterol (mo56Hc) as a potent pharmacological chaperone for several NPC1 mutants. Furthermore, we generated mice expressing human I1061T NPC1, a common mutation in patients. We demonstrated that this model exhibited disease phenotypes and recapitulated the protein trafficking defects, lipid storage, and response to mo56Hc exhibited by human cells expressing I1061T NPC1. These tools established a paradigm for testing and validation of proteostatic therapeutics as an important step toward the development of disease-modifying therapies.
BibTeX
@inproceedings{azaria2024mutant,
  author = {Ruth D. Azaria and Adele B. Correia and Kylie J. Schache and M. Camila Fernández Zapata and Koralege C. Pathmasiri and Varshasnata Mohanty and Dharma Theja Nannapaneni and Brandon L. Ashfeld and Paul Helquist and Olaf Wiest and Kenji Ohgane and Qingqing Li and Ross A. Fredenburg and Brian S. J. Blagg and Stephanie M. Cologna and Mark L. Schultz and Andrew P. Lieberman},
  title = {{Mutant induced neurons and humanized mice enable identification of Niemann-Pick type C1 proteostatic therapies}},
  booktitle = {JCI},
  year = {2024},
  doi = {10.1172/jci.insight.179525},
}
Figure