The phosphoproteomic landscape of the neurological manifestations in tuberous sclerosis complex

This study utilizes proteomic and phosphoproteomic analyses to reveal that while cortical tubers show limited mTORC1 activation, subependymal giant cell astrocytomas (SEGAs) exhibit strong mTORC1-driven dysregulation of the proteome and extensive alterations in mRNA splicing, thereby expanding the known repertoire of mTORC1 targets in the human brain.

The Gist

The Big Picture: A Broken "Master Switch" in the Brain

Imagine your brain is a bustling, high-tech city. In this city, there is a master control room called mTORC1. Its job is to manage the city's growth, energy, and maintenance.

In a healthy brain, there are two security guards, TSC1 and TSC2, who stand in front of this control room. Their job is to make sure the control room doesn't get too excited. They keep the "growth switch" turned down to a safe, steady level.

Tuberous Sclerosis Complex (TSC) is a genetic condition where these security guards are broken. Because the guards are missing or broken, the master control room (mTORC1) goes haywire. It screams "GROW! BUILD! EAT!" at full volume. This causes two types of problems in the brain:

1. Cortical Tubers: These are like "scars" or "bumps" on the brain's surface. They cause seizures and learning difficulties.
2. SEGAs: These are larger, tumor-like growths that can block fluid flow in the brain.

For a long time, scientists knew the "switch" was broken, but they didn't know exactly what the city was doing wrong because of it. This paper is like sending a team of detectives into the city to take a snapshot of everything happening at the molecular level.

---

Detective Work 1: Investigating the "Scars" (Tubers)

The researchers took samples of the Tubers (the scars) and compared them to healthy brain tissue.

What they found:

• The Energy Crisis: The "power plants" (mitochondria) in the tuber cells were running on low battery. The cells were tired and struggling to breathe.
• The Construction Crew was Confused: The internal scaffolding (cytoskeleton) that holds the cells together was messy and disorganized.
• The Surprising Twist: Even though we know the security guards (TSC1/2) were broken, the researchers could not find strong evidence that the master control room (mTORC1) was screaming "GROW!" in these specific scars.

The Analogy:
Imagine a construction site where the foreman (mTORC1) is supposed to be shouting orders. In the Tubers, the foreman seems quiet. Why? Because the "broken guard" cells are actually very rare in these scars. The scar is mostly made of confused, tired workers, with only a tiny few actually screaming orders. The noise of the few "screaming" cells gets drowned out by the silence of the rest.

---

Detective Work 2: Investigating the "Tumors" (SEGAs)

Next, the team investigated the SEGAs (the tumors). This was a different story.

What they found:

• The Alarm is Blaring: Here, the security guards were completely gone. The master control room (mTORC1) was screaming at maximum volume.
• The Factory is Overproducing: The cells were churning out massive amounts of "machinery" (ribosomes) to build proteins. It was like a factory running 24/7, producing parts it didn't need.
• The City is on Fire (Inflammation): The tissue was flooded with "firefighters" (immune cells) and warning signs (inflammatory proteins). The brain was treating the tumor like an invader.
• The "Spelling" Chaos: This is the most exciting discovery. The researchers found that the master control room wasn't just telling cells to grow; it was messing with the instruction manuals (RNA).

The Analogy:
Think of the instruction manual for building a house. Usually, the manual says: "Put the window here, the door there."
In the SEGAs, the master control room (mTORC1) started phosphorylating (tagging) the editors of these manuals. It was like a chaotic editor who, instead of just writing the manual, started cutting and pasting pages randomly.

• Sometimes a page was skipped.
• Sometimes a page was duplicated.
• Sometimes the wrong chapter was pasted in.

This is called mis-splicing. The result? The cells were trying to build houses using instructions that made no sense. They were building walls where doors should be, or windows where floors should be. This chaos likely makes the tumor grow uncontrollably and behave strangely.

---

The "Aha!" Moment: The Splicing Connection

The researchers realized that the "broken switch" (mTORC1) was directly attacking the splicing machinery.

• The Metaphor: Imagine a train station where trains (mRNA) arrive with extra cargo (introns) that needs to be removed before the train can leave. Special workers (spliceosomes) remove the cargo.
• The Problem: In SEGAs, the master control room (mTORC1) started tagging these workers with "high priority" stickers. This made the workers hyperactive and confused. They started removing the wrong cargo or leaving the right cargo on.
• The Result: The trains left the station with the wrong cargo, leading to a chaotic city.

Even more interestingly, the researchers found that the workers themselves were having their instruction manuals messed up. The genes that tell the workers how to do their job were being "spliced" incorrectly. It was a double whammy: the workers were confused, and the instructions on how to fix the confusion were also broken.

---

Why Does This Matter?

1. Why Tubers are hard to treat: Since the "screaming" cells are rare in tubers, drugs that block the "screaming" (like mTOR inhibitors) might not work well on the whole scar. The problem might be the tired, confused energy state of the cells, not just the growth signal.
2. Why SEGAs are dangerous: The tumors aren't just growing fast; they are fundamentally broken at the level of their genetic instructions.
3. New Targets for Medicine: This study gives scientists a massive new list of "suspects." Instead of just looking at the master switch, doctors might need to look at the splicing machinery. If we can fix the "editing" of the instruction manuals, we might be able to stop the tumors from growing or fix the brain's confusion.

Summary in One Sentence

This study discovered that while the "growth switch" in TSC brain tumors (SEGAs) is wildly overactive and causes chaos in how genetic instructions are edited (splicing), the "scars" (Tubers) are more like tired, confused cells where the switch isn't as loud, suggesting we need different strategies to treat each part of the disease.

Technical Summary

1. Problem Statement

Tuberous Sclerosis Complex (TSC) is a genetic disorder caused by mutations in TSC1 or TSC2, leading to the hyperactivation of the mechanistic target of rapamycin complex 1 (mTORC1). This results in severe neurological manifestations, including epilepsy, autism, and intellectual disability, primarily driven by two types of brain lesions: cortical tubers and subependymal giant cell astrocytomas (SEGAs).

Despite the known role of mTORC1, the specific molecular mechanisms driving the pathology in human brain tissue remain poorly understood. Previous studies have relied heavily on transcriptomics or animal models, which may not accurately reflect the post-translational modifications (phosphorylation) and protein abundance changes in human lesions. Furthermore, there is a lack of systematic identification of mTORC1 substrates in the human brain, and the molecular differences between tubers and SEGAs are not fully characterized.

2. Methodology

The authors employed an unbiased, quantitative proteomics and phosphoproteomics approach using fresh-frozen surgical resection tissues from TSC patients.

• Sample Cohort:
  • Tubers: 6 patients (ages 7–20).
  • SEGAs: 5 patients (ages 11–25).
  • Controls: Temporal lobe epilepsy (TLE) resections (n=6) and age-matched postmortem (PM) cortical tissue (n=8).
• Genetic Analysis: Whole-exome sequencing and targeted deep sequencing were performed to identify germline mutations and Loss of Heterozygosity (LOH) in TSC1/TSC2.
• Proteomic Workflow:
  • Digestion: Samples were digested with Trypsin (standard) and Chymotrypsin (to increase coverage of non-tryptic peptides).
  • Labeling: Tandem Mass Tag (TMT) 18-plex labeling was used for multiplexed quantitative analysis.
  • Fractionation/Enrichment: Sequential Metal Oxide Affinity Chromatography (SMOAC) using TiO2 and Fe-NTA was used to enrich phosphopeptides. Flow-through was used for total proteome analysis.
  • Mass Spectrometry: Analysis was performed on an Orbitrap Eclipse Tribrid mass spectrometer using data-dependent acquisition (DDA) with HCD MS2 and SPS MS3 methods.
• Bioinformatics:
  • Proteomics: MaxQuant for identification; Limma for statistical analysis (p < 0.005).
  • Phosphoproteomics: Motif analysis and Gene Ontology (GO) enrichment via STRING.
  • Splicing Analysis: Re-analysis of previously published RNA-seq data from SEGAs using IsoformSwitchAnalyzeR, rMATS, and DEXSeq to detect isoform switching, splicing events, and differential exon usage.

3. Key Contributions

• First Deep Phosphoproteomic Landscape: This study provides the first comprehensive, unbiased map of the proteome and phosphoproteome in human TSC brain lesions (tubers and SEGAs).
• Discovery of Novel mTORC1 Targets: Identification of over 2,000 novel proteins whose phosphorylation is significantly increased in SEGAs, expanding the known repertoire of mTORC1 substrates in the human brain.
• Mechanistic Insight into Splicing: Uncovered a direct link between mTORC1 activation and the dysregulation of mRNA splicing via the phosphorylation of splicing factors and hnRNPs.
• Differentiation of Lesion Types: Demonstrated distinct molecular profiles between tubers (where mTORC1 activation is difficult to detect) and SEGAs (where mTORC1 is robustly activated), explaining their different clinical behaviors.

4. Key Results

A. Cortical Tubers: Subtle Changes, No Detectable mTORC1 Activation

• Genetics: Tubers showed germline mutations but lacked consistent somatic "second hits" (LOH) in bulk tissue, consistent with the "two-hit" hypothesis where only a small fraction of cells are biallelically inactivated.
• Proteomics: Only ~300 proteins were significantly altered. The most prominent change was a decrease in mitochondrial respiration proteins (oxidative phosphorylation, electron transport chain).
• Phosphoproteomics:
  • No significant increase in canonical mTORC1 targets (4E-BP1/2, RPS6) was detected.
  • Changes were observed in cytoskeleton organization (increased) and neuronal function (decreased).
  • Conclusion: The low abundance of cells with complete TSC1/TSC2 loss in tubers prevents the detection of global mTORC1 activation in bulk tissue.

B. SEGAs: Robust mTORC1 Activation and Inflammation

• Genetics: 50% of analyzed SEGAs showed LOH of TSC2, and protein levels of TSC1/TSC2 were drastically reduced (45–84% decrease).
• Proteomics:
  • Ribosome Biogenesis: Massive upregulation of ribosomal proteins (33 large, 27 small subunits), consistent with mTORC1-driven translation.
  • Neuroinflammation: Strong upregulation of MHC Class I/II proteins (e.g., HLA-G, HLA-DRB5) and inflammatory markers (ANXA1, GPNMB, S100A11), indicating an active immune response.
  • Neuronal Loss: Downregulation of synaptic and neuronal development genes (e.g., NPTX1).
• Phosphoproteomics:
  • mTORC1 Activation: Canonical targets (4E-BP1/2, RPS6) were strongly hyperphosphorylated.
  • Novel Targets: Identified 6,060 phosphosites across 2,154 proteins with increased phosphorylation.
  • RNA Metabolism: The most significant enrichment was in RNA metabolism and mRNA splicing. Phosphorylation of spliceosome components (e.g., SF3B1, PUF60) and 16 heterogeneous nuclear ribonucleoproteins (hnRNPs) was significantly increased. Notably, 5 of these hnRNPs were phosphorylated at residues known to be mTORC1 targets.

C. Widespread Dysregulation of mRNA Splicing

• Re-analysis of RNA-seq data confirmed that SEGAs exhibit large-scale alternative splicing dysregulation.
• Isoform Switching: ~15% of genes showed significant isoform switches.
• Splicing Events: 1,572 significant splicing events (skipped exons, retained introns, etc.) were identified.
• Feedback Loop: The genes encoding the splicing machinery itself (including hnRNPs) showed altered splicing patterns, suggesting a self-reinforcing cycle of splicing dysregulation driven by mTORC1.

5. Significance

• Therapeutic Implications: The study highlights that mTORC1 inhibition (e.g., via everolimus) may need to target not just protein synthesis but also the RNA splicing machinery. The dysregulation of splicing factors could contribute to the refractory nature of TSC-associated epilepsy and cognitive deficits.
• Biomarker Development: Proteins like ANXA1, GPNMB, and S100A11, along with specific phospho-signatures, serve as potential biomarkers for SEGA progression and treatment response.
• Mechanistic Paradigm: The paper establishes a new model where mTORC1 activation in the brain leads to a "splicing catastrophe," disrupting the proteome at the transcriptional and post-transcriptional levels. This explains why SEGAs are highly proliferative and pathogenic compared to tubers.
• Resource: The dataset (PXD069405) provides a foundational resource for the TSC research community to explore novel drug targets beyond the canonical mTORC1 pathway.

In summary, this study moves beyond the "mTORC1 = protein synthesis" dogma to reveal that in human brain tumors (SEGAs), mTORC1 drives a complex rewiring of the phosphoproteome that specifically targets RNA splicing, leading to widespread transcriptomic instability and neuroinflammation.