CRISPR screens establish regulatory maps of immunosuppressive surface molecules in cancer

This study utilizes a time-controllable CRISPR screening system to map the regulatory networks of key immunosuppressive surface molecules in cancer, identifying the membrane-trafficking factor DNAJC13 as a critical regulator whose suppression enhances T-cell-mediated tumor destruction and improves survival in pancreatic cancer models.

The Gist

Imagine your body's immune system is a highly trained security force (the T-cells) patrolling a city, looking for criminals (cancer cells). Normally, these security guards have a "stop and check" protocol. However, cancer cells are clever; they wear special "Do Not Disturb" badges on their surface. When the security guards see these badges, they think, "Oh, this is a VIP, I shouldn't bother them," and they walk away. This allows the cancer to grow unchecked.

Some of the most famous "Do Not Disturb" badges are proteins like PD-L1. Doctors have developed drugs that act like a pair of scissors, cutting these badges off so the security guards can finally do their job. But there's a problem: we don't fully understand how the cancer cells make these badges in the first place. If we knew the factory assembly line, we could shut it down completely, rather than just cutting the badges off one by one.

This paper is about finding the "foremen" and "machines" inside the cancer cell factory that are responsible for building and placing these "Do Not Disturb" badges on the surface.

The Problem with Previous Searches

In the past, scientists tried to find these factory foremen by breaking genes randomly and waiting to see what happened. But there was a catch: if they broke a gene that was essential for the cancer cell to stay alive, the cell would die before the scientists could even check the badges. It's like trying to find out who runs the bakery by firing everyone; if you fire the baker, the shop closes, and you never get to see if they were actually making the bread. This meant scientists only found the "nice-to-have" workers, missing the critical ones that kept the cell alive and made the badges.

The New Trick: A "Pause Button" for the Factory

The researchers in this paper came up with a clever solution. They built a cancer cell line with a remote-controlled "pause button" (a Tet-inducible Cas9 system).

1. The Setup: They loaded the cells with a library of instructions to break every single gene in the human genome, but they kept the "break" switch turned OFF.
2. The Freeze: They froze these cells in a deep freeze. Because the switch was off, the cells were healthy and happy, even though they carried the instructions to break themselves.
3. The Trigger: When they were ready, they thawed the cells and flipped the switch ON. Suddenly, the genes started breaking.
4. The Sort: Because they could control exactly when the genes broke, they could look at the cells very quickly (before the "essential" workers died) and use a high-tech sorter (FACS) to pick out the cells that had lost their "Do Not Disturb" badges.

This allowed them to find 154 different genes that control the badges, many of which they would have missed with old methods.

The Big Discovery: The "Traffic Cop" (DNAJC13)

Among the hundreds of genes they found, one stood out like a lighthouse: DNAJC13.

Think of the cancer cell as a busy airport. The "Do Not Disturb" badges (PD-L1, CD276, etc.) are planes that need to be parked at the gate to be seen by the security guards.

• DNAJC13 is the Traffic Cop at the airport.
• Its job is to make sure these planes (badges) are loaded onto trucks and driven out to the tarmac (the cell surface).
• If you fire the Traffic Cop (knock out DNAJC13), the planes get stuck in the hangar. They never reach the surface.

The researchers found that DNAJC13 doesn't just manage one type of plane; it manages a whole fleet of "Do Not Disturb" badges at once (PD-L1, CD276, CD73, and others).

Why This Matters

When the researchers removed DNAJC13 from cancer cells:

1. The Badges Disappeared: The cancer cells lost almost all their "Do Not Disturb" signals.
2. The Security Guards Attacked: When they mixed these "naked" cancer cells with human T-cells in a dish, the T-cells tore the cancer cells apart much faster than usual.
3. It Worked in Mice: In mice with pancreatic cancer, removing DNAJC13 didn't kill the mice (the cells could still grow), but it made the cancer much more vulnerable to the immune system, and the mice lived much longer.

The Takeaway

This study is like finding the master switch for the cancer's camouflage. Instead of just trying to cut off one badge (like current PD-L1 drugs do), scientists might be able to target DNAJC13.

If we can develop a drug that temporarily "fires" the DNAJC13 Traffic Cop, the cancer cell loses its entire fleet of camouflage badges at once. This would make the cancer cells visible to the immune system, turning a stealthy criminal into an obvious target, allowing the body's natural defenses to wipe them out. It's a promising new strategy to help immunotherapy work better for more people.

Technical Summary

1. Problem Statement

Cancer cells evade immune surveillance by expressing immunosuppressive surface molecules (e.g., PD-L1, CD47, CD276, HLA-E) that inhibit T-cell and NK-cell activity. While therapeutic blockade of these checkpoints has revolutionized cancer treatment, the tumor cell-intrinsic mechanisms controlling the expression and trafficking of these molecules remain incompletely understood.

Previous attempts to identify these regulators using CRISPR/Cas9 or haploid gene-trap screens faced a critical limitation: they relied on stable gene knockouts and prolonged culture periods. This approach inadvertently selected against essential genes; cells with knockouts that compromised proliferation or survival were lost before the flow-cytometry readout could be performed. Consequently, these screens were biased toward non-essential regulators and missed critical factors that might be essential for cell fitness but also regulate immune checkpoints.

2. Methodology

To overcome the bias of conventional screens, the authors developed a time-controllable, FACS-based CRISPR screening platform:

• Inducible Cas9 System (iCas9): They engineered a tetracycline/doxycycline (Dox)-inducible Cas9 system in the RKO colorectal cancer cell line (which constitutively expresses high levels of PD-L1). This allowed them to maintain a library-transduced cell population without Cas9 activity, preserving cells with essential gene knockouts until the screen began.
• Screening Workflow:
  1. Library Transduction: Cells were transduced with a genome-wide sgRNA library (Vienna library) and expanded/cryopreserved before Cas9 induction.
  2. Induction & Sorting: Upon Dox induction, Cas9 was activated. Cells were sorted via Fluorescence-Activated Cell Sorting (FACS) at two distinct time points:
     • Day 4: Sorted for the lowest 1% of surface protein expression.
     • Day 10: Sorted for the lowest 0.1% (stricter gating) to capture regulators with slower turnover or delayed effects.
  3. Drop-out Control: A parallel culture was maintained for 18 days to identify "core essential" genes (those causing cell death regardless of surface protein levels).
• Targets: The study performed comparative screens for four major immunosuppressive molecules (PD-L1/CD274, CD47, CD276, HLA-E) and a control molecule (CD151) associated with tumor invasion.
• Validation & Omics: Hits were validated in independent cell lines (RKO and MIA-PaCa2) using single sgRNA assays. The authors also performed mass spectrometry-based surface proteomics to identify the full spectrum of proteins regulated by top hits.
• Functional Assays: Co-culture assays with primary human T-cells and in vivo orthotopic pancreatic tumor models (in immunocompetent and immunodeficient mice) were used to assess therapeutic potential.

3. Key Contributions

• Methodological Innovation: The study establishes a robust, renewable resource for genome-wide screens that can identify regulators of surface proteins independent of their effect on cell proliferation. By decoupling mutagenesis from the selection period, the authors captured essential genes that were previously missed.
• Regulatory Maps: The authors generated comprehensive regulatory maps for PD-L1, CD47, CD276, and HLA-E, identifying 154 significant regulators for PD-L1 alone (compared to the ~1-2 found in previous screens).
• Discovery of DNAJC13: A major finding is the identification of DNAJC13 (also known as RME-8) as a critical, context-specific regulator. Unlike general trafficking factors, DNAJC13 specifically coordinates the surface expression of a subset of immunosuppressive molecules without globally disrupting the surface proteome.

4. Key Results

• Expanded Regulator Lists: The screens identified 154 PD-L1 regulators, including components of the ESCRT-I complex, HOPS complex, and various trafficking factors. Crucially, many of these hits (e.g., NAPA, CDC73) were essential genes that would have been filtered out in traditional screens.
• DNAJC13 as a Master Regulator:
  • Specificity: DNAJC13 knockout significantly reduced surface levels of PD-L1 and CD276, as well as other immune checkpoints (PVR, PVRL2, NT5E/CD73), but had minimal effect on other surface proteins like CD151 or MHC Class I.
  • Mechanism: DNAJC13 acts as a post-translational regulator. It does not affect mRNA levels but controls the endosomal recycling of these proteins. Loss of DNAJC13 leads to the accelerated internalization and degradation of PD-L1 and CD276.
  • Context Specificity: While DNAJC13 is dispensable for the proliferation of most cancer cell lines (unlike CMTM6 or TAF7), its loss consistently downregulates immune checkpoints across diverse cell types.
• Proteomic Validation: Surface proteomics confirmed that DNAJC13 depletion leads to a coordinated downregulation of multiple immune-modulatory surface molecules, whereas CMTM6 depletion primarily affects PD-L1.
• Therapeutic Efficacy:
  • In Vitro: Knockout of DNAJC13 rendered cancer cells significantly more susceptible to T-cell-mediated killing. This effect was stronger than PD-L1 knockout alone and was mimicked by the combinatorial deletion of five DNAJC13-regulated molecules.
  • In Vivo: In an orthotopic pancreatic cancer model, mice bearing Dnajc13-deleted tumors showed delayed tumor progression and significantly improved survival compared to controls, specifically in immunocompetent hosts.

5. Significance

• New Therapeutic Target: The study identifies DNAJC13 as a promising target for cancer immunotherapy. Suppressing DNAJC13 offers a strategy to simultaneously downregulate multiple immunosuppressive checkpoints (PD-L1, CD276, etc.), potentially overcoming resistance to single-agent checkpoint blockade.
• Overcoming Screen Limitations: The work demonstrates that time-controlled CRISPR screens are essential for uncovering the full regulatory landscape of surface proteins, revealing that many critical regulators are also essential for cell survival.
• Mechanistic Insight: It elucidates the role of endosomal trafficking (specifically the Retromer/WASH interaction mediated by DNAJC13) in maintaining the "immune shield" of cancer cells, suggesting that disrupting this trafficking pathway is a viable therapeutic avenue.

In conclusion, this paper provides a comprehensive map of the genetic regulators of immune evasion molecules and highlights DNAJC13 as a potent, context-specific target to sensitize tumors to T-cell attack.