Hypoxic stress granules trigger immunogenic dormancy in lung cancer

This study reveals that physiologic hypoxia in lung cancer induces translational arrest and sequestration of MHC class I antigen processing pathway mRNAs into stress granules, thereby blocking immunopeptide presentation and establishing a state of "immunogenic dormancy" that facilitates immune escape.

The Gist

Imagine your body's immune system as a highly trained security force (the T-cells) patrolling a city (your body). Their job is to spot criminals (cancer cells) and take them down. To do this, the security force needs to see "wanted posters" displayed on the walls of every building. These posters are made of tiny protein fragments called immunopeptides.

Normally, when the immune system sounds the alarm (releasing a signal called IFN-γ), the cancer cells are supposed to panic, rush to their "printing press" (the Immunoproteasome), and print thousands of new, urgent wanted posters to show the security force exactly what they look like.

The Problem: The "Hypoxic" Blackout
This study discovered that cancer cells have a clever trick up their sleeve, specifically when they are in low-oxygen zones (hypoxia). Tumors often grow so fast that they choke off their own oxygen supply, creating dark, airless pockets inside the tumor.

The researchers found that when cancer cells find themselves in these low-oxygen pockets, they don't just slow down; they hit a "pause button" on their printing press. Even if the immune system screams, "Print the posters! Show us your face!" the cancer cells in the low-oxygen zone simply refuse to print them.

The Mechanism: The "Stress Granule" Lockdown
How do they do this without breaking the machinery? It's not because the instructions (DNA/RNA) are missing. The instructions are there, loud and clear.

Instead, the low oxygen triggers a "stress response" in the cell. Think of this like a construction site during a storm. The workers (ribosomes) stop building, and the blueprints (mRNA) get shoved into a locked storage container called a Stress Granule.

• The Blueprints: The instructions to make the "wanted posters" are still written down.
• The Lock: The stress granules act like a padlock, keeping the blueprints in the storage bin so the workers can't read them or build the posters.

The result? The cancer cell becomes invisible. It's in a state of "Immunogenic Dormancy"—it's alive, but it's hiding from the immune system by refusing to show its ID.

The Twist: It's Not About Oxygen Sensors
Usually, when cells sense low oxygen, they use a specific sensor (HIF) to react. But the researchers found this trick works independently of that sensor. It's a direct reaction to the lack of air that shuts down the translation process.

The Solution: The "Unlocking" Key
The team tested a drug called 5-azacytidine (5-AZA). Think of this drug as a master key or a solvent that dissolves the padlock on the stress granules.

• When they added 5-AZA to the low-oxygen cancer cells, the stress granules dissolved.
• The blueprints were released back to the workers.
• The printing press started working again, and the "wanted posters" (immunopeptides) appeared on the surface of the cancer cells.
• Suddenly, the cancer cells were visible to the immune system again.

Interestingly, a very similar drug called Decitabine (which works on DNA) didn't work. This proved that the problem wasn't with the DNA instructions, but specifically with how the RNA instructions were being handled and locked away.

Real-World Evidence
The researchers didn't just see this in a lab dish. They looked at actual lung cancer tumors from human patients. They found that in the dark, low-oxygen centers of the tumors, the "printing presses" (immunoproteasomes) were completely missing. The cancer cells in those zones were effectively invisible to the immune system.

The Big Picture
This study reveals a new way cancer hides. It's not just about mutating to look different; it's about physically locking away its own ID cards when the environment gets tough (low oxygen).

Why this matters:
If we can find a way to keep those "stress granule locks" open (perhaps using drugs like 5-AZA), we could force the cancer cells to reveal themselves, even in the deepest, darkest parts of the tumor. This could help the immune system finish the job and make immunotherapy work better for more people.

In a Nutshell:

• The Villain: Cancer cells in low-oxygen zones.
• The Trick: They lock their "ID card" instructions in a stress granule storage bin, making them invisible to the immune system.
• The Hero: A drug (5-AZA) that unlocks the bin, forcing the cancer to show its face so the immune system can attack.

Technical Summary

1. Problem Statement

Effective anti-tumor immunity relies on the ability of cancer cells to process and present tumor-associated antigens (TAAs) and neoantigens via the Major Histocompatibility Complex class I (MHC-I) pathway. This process is heavily dependent on the Class I Antigen Processing and Presentation Pathway (C1APP), particularly the induction of the Immunoproteasome (IP) by inflammatory cytokines like IFN-γ.

• The Gap: While it is known that tumors often lose MHC-I expression, the specific mechanisms by which the tumor microenvironment (TME) suppresses the inducibility of the C1APP in response to immune pressure remain unclear.
• The Specific Question: Does tumor hypoxia (a common feature of solid tumors like Non-Small Cell Lung Cancer, NSCLC) impair the ability of cancer cells to upregulate the C1APP in response to IFN-γ, thereby creating a state of "immunogenic dormancy"?

2. Methodology

The study employed a multi-omics and mechanistic approach using human lung adenocarcinoma (A549) cells and patient-derived NSCLC tissues.

• Cell Culture Models: A549 cells and 8 other epithelial cancer lines were preconditioned in normoxia (20% O₂) or hypoxia (2% O₂) for 48 hours, followed by stimulation with IFN-γ, IFN-α, or TNF-α.
• Immunopeptidomics: HLA class I molecules were immunoprecipitated, and eluted peptides were analyzed using nanoLC–MS/MS (PASEF) to quantify the diversity, length, and source (TAA vs. neoantigen) of presented immunopeptides.
• Molecular Mechanism Dissection:
  • Protein/MRNA Analysis: Western blotting and RT-qPCR were used to assess IP subunit protein levels versus transcript abundance.
  • Signaling Pathways: The study tested the roles of HIF-1α/2α (using CRISPR/Cas9 knockouts and CoCl₂ mimetics), autophagy (bafilomycin A1), and the Ubiquitin-Proteasome System (bortezomib).
  • Translation Analysis: Polysome profiling was used to determine the translational status of C1APP mRNAs.
  • Stress Granules: Immunofluorescence for G3BP1 and RNA-FISH were used to visualize stress granules and localize specific C1APP transcripts (PSMB8, PSMB9, PSMB10).
  • Epitranscriptomics: Nanopore direct RNA sequencing and m5C MeRIP-seq were performed to investigate RNA methylation changes.
• Pharmacological Intervention: Cells were treated with 5-azacytidine (5-AZA) (incorporates into RNA and DNA) and decitabine (DAC) (incorporates only into DNA) to distinguish between epigenetic and epitranscriptomic mechanisms.
• Clinical Validation: Immunofluorescence staining of 9 stage II/III NSCLC patient tumors was performed to assess the spatial co-localization of IP subunits (β1i, β2i, β5i) and the hypoxia marker CA9.

3. Key Contributions

• Discovery of "Immunogenic Dormancy": The paper defines a reversible state where tumor cells retain the capacity to transcribe C1APP genes but fail to translate them into proteins under hypoxic conditions, rendering them invisible to T cells despite intact inflammatory signaling.
• Mechanistic Shift: It identifies translational arrest via stress granules as the primary mechanism, distinct from previously proposed mechanisms like HIF-mediated transcriptional repression or autophagy-dependent degradation.
• Epitranscriptomic Reversibility: It demonstrates that the hypoxia-induced blockade is reversible by 5-azacytidine, suggesting a role for RNA-associated mechanisms rather than DNA methylation.
• Clinical Correlation: It provides direct evidence from human NSCLC tumors that immunoproteasome expression is spatially excluded from hypoxic regions.

4. Key Results

• Hypoxia Blocks Immunopeptide Presentation:

  • Under hypoxia (2% O₂), IFN-γ treatment failed to increase the number or diversity of unique immunopeptides presented by A549 cells.
  • 73% of immunopeptides that were upregulated by IFN-γ in normoxia were absent or reduced in hypoxia.
  • This included both TAAs and a detected neoantigen (BTBD11-derived), indicating a broad suppression of tumor immunogenicity.
  • The specific amino acid anchor residues (P2 and C-terminal Ω) favored by IFN-γ in normoxia were lost in hypoxia.

• Translational Blockade, Not Transcriptional:

  • Protein vs. mRNA: While IFN-γ robustly induced IP protein expression (β1i, β2i, β5i) and downstream C1APP components in normoxia, this induction was blocked in hypoxia.
  • However, mRNA levels for these genes remained high and inducible in hypoxia, ruling out transcriptional repression.
  • Polysome Profiling: C1APP mRNAs shifted from heavy polysome fractions (actively translating) to pre-monosome fractions (40S/60S subunits) under hypoxia, indicating translational arrest.

• Mechanism: Stress Granules and ISR:

  • Hypoxia induced the Integrated Stress Response (ISR), evidenced by eIF2α phosphorylation.
  • C1APP mRNAs were sequestered into hypoxia-associated stress granules (marked by G3BP1), preventing their translation.
  • This mechanism was independent of HIF-1α/2α (CRISPR knockouts did not restore expression) and independent of autophagy or proteasomal degradation.

• Reversibility via 5-AZA:

  • Treatment with 5-azacytidine (but not decitabine) prevented stress granule formation, restored eIF2α phosphorylation balance, and rescued IFN-γ-induced IP protein expression and immunopeptide diversity under hypoxia.
  • This suggests the mechanism relies on RNA modification or RNA-protein interactions rather than DNA methylation.

• Clinical Evidence:

  • In 9 NSCLC patient tumors, IP subunits (β1i, β2i, β5i) showed significant spatial exclusion from CA9-positive (hypoxic) regions.
  • Spearman correlation analysis confirmed a negative correlation between IP subunits and CA9 in 6 of 7 quantifiable patients.

5. Significance

• Therapeutic Implications: The findings suggest that hypoxic tumor regions are "immunologically silent" not because they lack antigens, but because they cannot present them. This explains why some tumors with high mutational burdens still resist immunotherapy.
• Novel Target: The study identifies stress granules and the translational control of C1APP as a critical checkpoint for immune evasion.
• Combination Strategies: The ability of 5-azacytidine to reverse this blockade suggests that combining hypoxia-targeting agents or epitranscriptomic modulators with immune checkpoint inhibitors could "wake up" dormant tumor cells, making them visible to the immune system.
• Paradigm Shift: This work moves beyond the view of hypoxia as merely a driver of transcriptional changes (via HIFs) or protein degradation, highlighting a rapid, reversible, post-transcriptional mechanism of immune escape.