A human lysosomal storage disorder toolkit for decoding proteome landscapes in cortical and dopaminergic-like induced neurons

This study presents a human embryonic stem cell toolkit lacking 23 lysosomal storage disorder genes, coupled with multi-omics analyses and cryo-electron tomography, to define distinct proteomic signatures, mitochondrial vulnerabilities, and synaptic defects in cortical and dopaminergic neurons.

The Gist

Imagine your body is a bustling city. Inside every cell in that city, there is a specialized recycling plant called the lysosome. Its job is to take out the trash, break down old proteins and fats, and recycle the good parts to build new things.

Now, imagine a Lysosomal Storage Disorder (LSD). This is like a glitch in the recycling plant's machinery. Because a specific tool (a gene) is missing or broken, certain types of trash (like specific fats or sugars) pile up instead of being recycled. Over time, this trash builds up, clogs the plant, and eventually causes the whole city (the cell, and eventually the brain) to malfunction.

There are about 70 different types of these disorders, each caused by a different broken tool. But scientists have struggled to understand exactly how these different broken tools cause such different problems, especially in the brain.

This paper is like a massive new toolkit designed to solve that mystery. Here is what the researchers did, explained simply:

1. Building a "Genetic Lego Set"

The researchers created a library of human stem cells (the "blank slates" that can become any type of cell). They used gene-editing scissors (CRISPR) to break 23 different "recycling tools" (genes) in these cells.

• The Analogy: Think of this as a set of 23 different Lego models where they intentionally removed one specific brick from each model. This allows them to see exactly what happens when that one specific brick is missing.

2. Turning Cells into "Brain Neurons"

They took these broken Lego models and turned them into two specific types of brain cells:

• Cortical Neurons: The general "thinking" cells of the brain.
• Dopaminergic Neurons: The "mood and movement" cells (the ones that die in Parkinson's disease).
• The Goal: They wanted to see if breaking the same tool caused different problems in the "thinking" cells versus the "movement" cells.

3. Taking a "Proteomic Snapshot"

They didn't just look at the cells; they took a high-resolution photo of every single protein inside them (the "proteome").

• The Analogy: Imagine walking into a factory and counting every single screw, gear, and conveyor belt. They found that when a specific gene was broken, the factory didn't just stop working; it started rearranging its entire layout. Some machines were built in excess, while others disappeared completely.

4. The Big Discoveries

By comparing the "snapshots" of all 23 broken models, they found some fascinating patterns:

• The "Cell Type" Surprise: Breaking the same gene caused different messes in different brain cells. For example, breaking the GBA1 gene messed up the "power plants" (mitochondria) in the movement cells, but not as much in the thinking cells.
• The "Synapse" Problem: In cells missing the ASAH1 gene, the "communication stations" (synapses) where neurons talk to each other were falling apart.
  • Real-world proof: When they tested these cells, they fired electrical signals much slower than normal cells. It's like a phone line that is full of static and can't get a clear signal through.
• The "Swollen Trash Can": Using a super-powerful microscope (Cryo-Electron Tomography), they looked inside the recycling plants of the ASAH1 broken cells.
  • The Visual: Instead of a tight, organized recycling center, the trash cans were swollen, bloated, and empty of the usual dense layers. They were filled with weird, bubbly pockets. It looked like a garbage truck that had been overfilled with air and was about to burst.

5. The "Connection Map"

The researchers also built a computer program to map how proteins stick together to form teams (complexes).

• The Analogy: If proteins are workers, they usually work in teams (like a construction crew). The researchers found that when a gene was broken, it didn't just remove one worker; it often caused the whole construction crew to fall apart because the workers couldn't find each other anymore.

Why Does This Matter?

This paper is a roadmap.

1. For Scientists: It gives them a ready-made library of broken cells to test new drugs on. Instead of guessing which cells to use, they can pick the exact model that matches a patient's specific disease.
2. For Patients: It helps explain why some people with these disorders get Parkinson's-like symptoms while others get different problems. It suggests that the "trash" isn't just sitting there; it's actively breaking the power plants and communication lines of the brain.
3. For the Future: By understanding exactly which "teams" of proteins are falling apart, doctors might be able to design therapies that fix the specific broken team rather than just trying to clean up the trash.

In short: The researchers built a giant library of broken brain cells, took a deep look inside them, and discovered that different broken genes cause unique, specific disasters in the brain's power grid and communication network. This map will help scientists navigate toward better treatments for these difficult diseases.

Technical Summary

1. Problem Statement

Lysosomal Storage Disorders (LSDs) are a heterogeneous group of ~70 diseases caused by defects in lysosomal catabolic functions, leading to the accumulation of undegraded substrates (e.g., glycosphingolipids, cholesterol). While the genetic basis of many LSDs is known, the molecular mechanisms linking specific gene deficiencies to distinct multisystem disease phenotypes—particularly in cell-type-specific contexts like neurons—remain poorly understood. Previous studies were limited to non-neuronal models (e.g., HeLa cells) or focused on single genes. There is a critical need for a systematic, scalable resource to map how the loss of diverse LSD genes alters proteomic landscapes, organelle function, and protein complexes in relevant neuronal cell types.

2. Methodology

The authors developed a comprehensive toolkit and analytical pipeline combining stem cell engineering, deep proteomics, and structural biology:

• Stem Cell Toolkit Generation:

  • Utilized an H9 human embryonic stem cell (hESC) line engineered with an inducible AAVS1-NGN2 cassette for neuronal differentiation and an endogenously tagged TMEM192-3xHA locus for lysosome immunoprecipitation (LysoIP).
  • Generated a library of 23 CRISPR-Cas9 edited hESC lines, each harboring homozygous frameshift indels in specific LSD genes. The library covers sphingolipidoses (e.g., GBA1, ASAH1, SMPD1), neuronal ceroid lipofuscinoses (NCL), mucolipidosis, Niemann-Pick, and others.
  • Differentiated these lines into two neuronal subtypes: cortical-like induced neurons (iN) and midbrain dopaminergic-like neurons (iDA) at Day 30 and Day 50.

• Proteomic Profiling:

  • Performed global mass spectrometry (nDIA-hrMS2) on total cell lysates from biological triplicates of all 23 mutant lines and controls.
  • Detected approximately 10,000 proteins per sample.
  • Validated data quality via replicate correlation (>0.98), principal component analysis (PCA), and marker induction checks.

• Informatic Pipeline for Protein Complex Analysis:

  • Developed a novel pipeline to decode protein-protein interaction (PPI) networks.
  • Integrated data from CORUM, Complex Portal, PDB, BioGrid, STRING, and AlphaFold Multimer predictions.
  • Created a "Tier 1" confidence neuronal PPI network and calculated "disruption scores" for protein complexes based on the abundance reduction of constituent subunits in mutant lines.

• Structural and Functional Validation:

  • Cryo-Electron Tomography (Cryo-ET): Performed on ASAH1-deficient HeLa cells (as a tractable model) to visualize endolysosomal ultrastructure.
  • Electrophysiology/Imaging: Used Fluo-4 calcium imaging and machine learning pipelines to assess neuronal firing rates in ASAH1-/- iDA cells.
  • Lipidomics: Analyzed lysosomal lipid composition in ASAH1-/- HeLa cells.

3. Key Contributions

• First-Generation LSD Toolkit: Creation of a versatile hESC library with 23 distinct LSD gene knockouts, enabling direct comparison across disease classes in a human neuronal context.
• Cell-Type Specificity: The first deep proteomic comparison of iN vs. iDA cells across multiple LSD genotypes, revealing that organelle vulnerabilities are highly dependent on both the specific gene mutation and the neuronal subtype.
• PPI Network Decoding: A new informatic framework that moves beyond organelle-level abundance changes to identify specific disrupted protein complexes and trimers/tetramers using structural prediction data.
• Multi-Modal Validation: Integration of proteomics, lipidomics, cryo-ET, and functional electrophysiology to provide a holistic view of LSD pathology.

4. Key Results

• Organelle Landscapes:

  • Different LSD genotypes caused distinct patterns of protein abundance changes across organelles (lysosomes, mitochondria, synaptic vesicles).
  • Mitochondria: GBA1 and ASAH1 deficiencies in iDA cells (but not iN) showed significant negative correlations with mitochondrial proteins, specifically affecting Oxidative Phosphorylation (OXPHOS) complexes.
  • Synapses: ASAH1-/- iDA cells displayed reduced abundance of synaptic vesicle (SV) fusion proteins, correlating with functional deficits.

• Protein Complex Disruption:

  • The pipeline identified specific "lost" or "disrupted" nodes in protein complexes.
  • GBA1-/- cells showed disruption in Mitochondrial Complex I (OXPHOS).
  • MCOLN1-/- cells showed disruption in the BLOC-1 complex.
  • ASAH1-/- cells showed disruption in the GATOR complex, Cytochrome C oxidase, and the RAB7 GEF complex (CCZ1-MON1-RMC1).

• Functional Consequences of ASAH1 Deficiency:

  • Synaptic Structure: ASAH1-/- iDA cells exhibited disorganized presynaptic structures with fewer, non-uniform synaptic vesicles and mislocalized synaptophysin (SYP) relative to Bassoon.
  • Neuronal Firing: ASAH1-/- iDA cells showed significantly reduced neuronal firing rates compared to controls, which was partially rescued by KCl stimulation but not to wild-type levels.
  • Ultrastructure (Cryo-ET): ASAH1-/- endolysosomes were swollen and lacked the dense internal lamellar membranes seen in wild-type cells. Instead, they contained numerous intraluminal vesicles (ILVs) studded with glycosylated proteins.
  • Lipidomics: Confirmed a ~3-fold increase in ceramide and an increase in bis(monoacylglycerol)phosphate (BMP) lipids in ASAH1-/- lysosomes, suggesting a compensatory mechanism or altered membrane dynamics.

5. Significance

• Mechanistic Insight into PD: The study reinforces the link between sphingolipid metabolism (GBA1, ASAH1) and Parkinson's Disease (PD) by demonstrating specific vulnerabilities in mitochondrial OXPHOS and synaptic function in dopaminergic neurons, mirroring PD pathophysiology.
• Resource for the Field: The toolkit and associated proteomic datasets (available via MASSIVE and Zenodo) provide a foundational resource for the community to define molecular signatures of LSDs, screen for therapeutic targets, and understand cell-type-specific disease mechanisms.
• Methodological Advance: The integration of AlphaFold-based structural predictions with quantitative proteomics offers a powerful new approach to mapping protein complex stability and vulnerability in disease states, adaptable to other genetic screens (e.g., CRISPR, Perturb-seq).
• Therapeutic Implications: By identifying specific organelle and complex vulnerabilities (e.g., mitochondrial dysfunction in GBA1 mutants, synaptic defects in ASAH1 mutants), the study highlights potential targets for precision medicine interventions in LSDs and related neurodegenerative disorders.