Oncogenes and tumor suppressor genes are enriched in stop-loss mutations generating protein extensions

This study analyzes mutation data from over 20,000 cancer patients to reveal that stop-loss mutations, which generate C-terminal protein extensions, are significantly enriched in oncogenes and tumor suppressor genes and can functionally alter protein processing, as demonstrated by the impaired cleavage of the immunostimulatory peptide thymosin alpha 1 in PTMA.

The Gist

Imagine your DNA is a massive instruction manual for building a human body. Each gene is a specific recipe, and at the very end of every recipe, there is a giant red "STOP" sign (a stop codon). This sign tells the cellular machinery, "Okay, the protein is done! Put down the tools and move to the next recipe."

This paper is about what happens when that "STOP" sign gets erased or changed into a "GO" sign.

The "Run-On Sentence" Problem

In the world of genetics, a mutation that turns a "STOP" sign into a "GO" sign is called a Stop-Loss mutation.

Think of it like a sentence in a book:

• Normal: "The cat sat on the mat. STOP."
• Stop-Loss Mutation: "The cat sat on the mat. GO. The cat then ate a sandwich, danced a jig, and tried to fly to the moon..."

Because the "STOP" sign is gone, the cell keeps reading the instructions even after it should have finished. It starts reading the "garbage text" that usually sits behind the scenes (the 3' UTR), which wasn't meant to be part of the recipe. This results in a protein that has a weird, extra tail attached to the end of it.

What Did the Scientists Find?

The researchers looked at the DNA of 20,801 cancer patients to see how often these "run-on sentences" happened. Here is what they discovered, using some simple analogies:

1. It's a Common Mistake in Cancer
They found thousands of these mutations. It turns out that cancer cells are messy editors; they frequently delete the "STOP" signs. About 12% of the patients had at least one of these mutations.

2. The "Bad Guys" Are Involved
You might think these mutations are just random noise, but the scientists found they aren't. These "run-on" mutations happen much more often in the genes that control cancer (the "villains" like oncogenes and the "heroes" like tumor suppressors).

• Analogy: It's like finding that in a city where the police and the criminals are both breaking traffic laws, it's not just random. The people who are supposed to be in charge (or causing trouble) are the ones most likely to have their stop signs removed.

3. The "Extra Tail" Makes the Protein Weird
When the protein gets that extra tail, it changes its personality.

• The Sticky Tail: The extra bits are often "sticky" (hydrophobic) and "electric" (positively charged).
• The ID Badge: Because these tails are so weird and electric, they act like a bright, flashing ID badge. The body's immune system (the security guards) can easily spot these "foreign" tails and recognize them as intruders.
• The Double-Edged Sword: This is interesting because while the mutation might help the cancer grow, it also makes the cancer cell easier for the immune system to see. This might explain why patients with lots of these mutations sometimes respond better to immunotherapy (treatments that wake up the immune system).

4. The Case of the "PTMA" Protein
The researchers zoomed in on one specific gene called PTMA, which was the most frequently mutated gene in their study.

• The Normal Job: PTMA is a protein that helps cells grow. It also has a special "chop-off" mechanism. The cell can cut off the front part of PTMA to make a tiny peptide called Thymosin Alpha 1. This tiny peptide is a superhero for the immune system; it wakes up the T-cells to fight cancer.
• The Mutation's Effect: When the "STOP" sign is lost in PTMA, the protein gets that weird extra tail at the back. The scientists found that this extra tail messes up the "chop-off" mechanism.
• The Result: The cell makes the full, weird PTMA protein, but it fails to cut off the immune-boosting Thymosin Alpha 1.
• The Analogy: Imagine a factory that makes a machine (PTMA) which also produces a special fuel (Thymosin) that powers the fire department (immune system). The mutation breaks the machine so it keeps running, but it jams the nozzle that sprays the fuel. The fire department never gets the fuel, so the fire (cancer) burns out of control.

Why Does This Matter?

This paper is a big deal because:

1. It's a New Clue: For a long time, scientists mostly looked at "typos" in the middle of recipes (missense mutations). They ignored the "missing stop signs." This study shows those missing signs are actually very important in cancer.
2. It Explains Immunity: It gives us a reason why some tumors are easier for the immune system to spot (because of those weird, electric tails).
3. It Offers a Target: By understanding how the PTMA mutation stops the production of the immune-boosting peptide, doctors might one day find a way to fix that "jam" and help the immune system fight the cancer naturally.

In short: Cancer cells often delete the "STOP" signs in their DNA, creating weird, long proteins. These weird proteins can help the cancer grow, but they also wear a bright neon sign that the immune system can see. However, in some cases (like with PTMA), the mutation also accidentally turns off the body's natural immune-boosting fuel, giving the cancer an unfair advantage.

Technical Summary

1. Problem Statement

Cancer genomes accumulate numerous mutations, yet rare mutation types like stop-loss mutations (converting a stop codon to a sense codon) remain poorly characterized. Unlike missense mutations, stop-loss mutations cause ribosomal translation to continue into the 3' Untranslated Region (3' UTR), generating C-terminal protein extensions.

• Knowledge Gap: While some stop-loss mutations are known to cause protein degradation (via hydrophobic degrons) or translation inhibition, their prevalence across the pan-cancer landscape, their specific impact on oncogenes vs. tumor suppressors, and their potential to generate immunogenic neoantigens are not well understood.
• Hypothesis: The authors hypothesize that stop-loss mutations are enriched in cancer-associated genes and that the resulting C-terminal extensions possess biophysical properties (hydrophobicity, charge) that may influence protein stability, localization, and immune recognition.

2. Methodology

The study employed a multi-step approach combining large-scale bioinformatics analysis with experimental validation.

A. Data Compilation and Bioinformatics

• Cohort: Aggregated somatic mutation data from 20,801 patients across 20 different cancer cohorts (including TCGA, ICGC, Hartwig Medical Foundation, CPTAC, and specific studies on bladder, lung, and liver cancers).
• Scope: Analyzed 23,231 protein-coding genes.
• Filtering: Identified stop-loss mutations (TAA/TAG/TGA $\to$ sense codon) with strict criteria: population allele frequency < 5% (gnomAD), sequencing depth $\ge$ 30X, and alternative allele depth $\ge$ 3X.
• Extension Prediction: Used Ensembl GRCh38 annotations to predict C-terminal extensions until the next in-frame stop codon. Discarded cases with no downstream stop codon.
• Biophysical Analysis: Calculated hydrophobicity (Kyte-Doolittle scale) and isoelectric point (pI) for extensions, comparing them to canonical proteins and in silico translated intronic open reading frames (iORFs).
• Immunogenicity Prediction: Used MHCflurry (v2.0.6) to predict strong MHC Class I binders (9-mer peptides, IC50 < 50 nM) within the extensions.
• Enrichment Analysis: Compared stop-loss frequency in Network of Cancer Genes (NCG), COSMIC tumor suppressor genes (TSGs), and oncogenes against non-cancer genes using Fisher's exact tests.

B. Experimental Validation (Focus on PTMA)

• Target: PTMA (Prothymosin alpha), the gene with the highest recurrence of stop-loss mutations (14 patients).
• Cloning: Synthesized Wild-Type (WT) and Stop-Loss Extended (Ext) PTMA ORFs (9 amino acid extension: QTAKKEKLN) cloned into a Sleeping Beauty backbone.
• Cell Lines: JHH6, HuH7 (Hepatocellular carcinoma), and HeLa (Cervical carcinoma).
• Assays:
  • Proliferation: xCELLigence RTCA (real-time impedance) and competitive co-culture flow cytometry.
  • Drug Resistance: Cisplatin treatment with MTS assay to determine IC50.
  • Protein Analysis: Western blotting to detect full-length protein size and N-terminal cleavage (Thymosin alpha 1 generation).
  • Localization: Subcellular fractionation (cytoplasmic vs. nuclear) followed by immunoblotting.
  • Expression: RT-qPCR to verify transcript levels.

3. Key Results

A. Pan-Cancer Landscape of Stop-Loss Mutations

• Prevalence: Identified 3,757 stop-loss mutations affecting 3,249 genes (14% of all coding genes) in 12.18% of patients.
• Mutation Types: 99% were single nucleotide variants (G>C, G>T, T>C). The most common resulting amino acids were Tyrosine (Y) and Serine (S).
• Recurrence: ~11% of mutated genes showed recurrent stop-loss mutations (occurring in >1 patient). The top recurrent genes were PTMA (14 patients), PCDH9 (8 patients), and SOX9 (6 patients).

B. Biophysical Properties of Extensions

• Composition: Extensions are significantly more hydrophobic and have a higher isoelectric point (positive charge) than canonical proteins, though less hydrophobic than random iORFs.
• Length: Median extension length is 19 amino acids (95% < 82 aa).
• Immunogenicity: Extensions containing predicted strong MHC I binders (pMHC) were significantly longer, more hydrophobic, and more positively charged than those without. This suggests a mechanism for increased tumor immunogenicity.

C. Enrichment in Cancer Genes

• Global Enrichment: Cancer-associated genes (NCGs) have 37% more stop-loss mutations than non-cancer genes.
• Oncogenes vs. TSGs: Both oncogenes and tumor suppressor genes show similar enrichment frequencies (~21-22%), challenging the notion that stop-loss mutations primarily inactivate tumor suppressors via degradation.

D. Functional Analysis of PTMA (Prothymosin alpha)

• Expression: PTMA is overexpressed in various cancers (e.g., TGCT, LIHC, COAD).
• Protein Stability & Localization:
  • The stop-loss mutant (9 aa extension) accumulates in the cytoplasm, whereas WT PTMA is predominantly nuclear.
  • The extension likely interferes with the C-terminal nuclear localization signal (TKKQKT).
• Cleavage Defect: The mutant protein shows impaired cleavage of the N-terminal domain. Consequently, the production of Thymosin alpha 1 (a 28-aa immunostimulatory peptide derived from the N-terminus) is reduced.
• Phenotype: While cell proliferation and cisplatin resistance were similar between WT and mutant cells, the reduction in Thymosin alpha 1 suggests a mechanism for immune evasion (loss of immunostimulatory peptide) rather than direct oncogenic hyperactivity.

4. Significance and Contributions

1. Comprehensive Catalog: This study provides the largest curated database of stop-loss mutations in cancer to date, moving beyond isolated case studies to a genome-wide perspective.
2. Dual Role in Tumorigenesis: Demonstrates that stop-loss mutations are not limited to inactivating tumor suppressors (via degradation) but are also enriched in oncogenes, potentially altering protein function, localization, and processing.
3. Immunological Mechanism:
   • Neoantigen Generation: The hydrophobic and positively charged nature of extensions favors MHC I presentation, potentially making tumors more visible to the immune system (explaining the link to immunotherapy response found in previous bladder cancer studies).
   • Immune Evasion: The PTMA case study reveals a novel mechanism where stop-loss mutations reduce immunogenicity by preventing the generation of the immunostimulatory Thymosin alpha 1 peptide.
4. Therapeutic Implications: The findings suggest that stop-loss mutations could serve as biomarkers for immunotherapy response or targets for vaccines (via the generated neoantigens). Conversely, understanding the loss of immunostimulatory peptides (like Thymosin alpha 1) could inform combination therapies.

Conclusion

The paper establishes that stop-loss mutations are a significant, recurrent driver of cancer biology. They generate C-terminal extensions with distinct biophysical properties that alter protein localization and processing. While these mutations can generate immunogenic neoantigens, they can also disrupt the production of endogenous immunostimulatory peptides, highlighting a complex dual role in shaping the tumor-immune microenvironment.