DNA-PK interacts with cyclic dinucleotides and inhibits type I Interferon responses

This study reveals that DNA-PKcs directly binds to cyclic dinucleotides to inhibit STING-mediated type I interferon responses, establishing a novel regulatory mechanism with significant implications for developing STING-targeting therapeutics.

The Gist

The Big Picture: The Body's Fire Alarm and the "Firefighter"

Imagine your body is a massive city. Inside every cell, there is a highly sensitive fire alarm system called STING. Its job is to detect when something dangerous, like a virus or a piece of broken DNA, has sneaked into the wrong place (the cytoplasm).

When the alarm is triggered, it releases a chemical "siren" called cGAMP (a cyclic dinucleotide). This siren tells the cell's defense team to start building walls and launching attacks (producing Interferons) to fight off the invader. This is great when there is a real fire (an infection).

But here is the problem: If the alarm keeps ringing non-stop, or if the siren is too loud, the city goes into a panic. This leads to chronic inflammation, which damages the city's own buildings (tissues) and causes diseases like auto-immune disorders or cancer.

For a long time, scientists knew how to turn the alarm on, but they didn't know who was responsible for turning it off once the danger passed. They were looking for a "signal terminator."

The Discovery: DNA-PKcs is the "Siren Trap"

This paper reveals that DNA-PKcs (a protein usually known for fixing broken DNA) has a secret second job: it acts as a siren trap.

Here is how it works, step-by-step:

1. The Alarm Rings: A virus enters, and the cell produces the siren (cGAMP).
2. The Trap Springs: The DNA-PK protein sees the siren floating around. Instead of just ignoring it, DNA-PK grabs onto it physically.
3. The Muzzle: By grabbing the siren, DNA-PK effectively "muzzles" it. It stops the siren from reaching the main alarm panel (STING).
4. The Result: The alarm stops ringing. The immune response calms down, and the cell returns to a peaceful, homeostatic state.

The Analogy: Think of DNA-PKcs as a bouncer at a club. The siren (cGAMP) is a rowdy guest trying to get into the VIP section (STING) to start a party (immune response). The bouncer (DNA-PK) grabs the guest by the collar and says, "You've had enough; you can't go in there right now." This prevents the party from getting out of hand and destroying the club.

The Twist: The Siren Fights Back

Interestingly, the paper also found that the siren (cGAMP) fights back. When DNA-PK grabs the siren, the siren actually paralyzes the bouncer.

• DNA-PK's Normal Job: It's a mechanic that fixes broken DNA (like a repair crew fixing a pothole).
• The Conflict: When the siren binds to DNA-PK, it stops the mechanic from fixing the pothole.

This creates a fascinating loop: The immune system (the siren) can temporarily pause the body's ability to repair DNA. This suggests that when your body is fighting a virus, it might prioritize the fight over maintenance, even if it means leaving a few potholes (DNA damage) unfixed for a while.

Why This Matters for Medicine

The researchers tested this with drugs designed to wake up the immune system (STING agonists), which are being used in cancer therapy to help the body fight tumors.

• The Problem: Some of these drugs are like a "super-siren." They are so loud that the body's natural bouncer (DNA-PK) gets overwhelmed, or conversely, the bouncer catches the drug and neutralizes it before it can do its job.
• The Solution: The paper shows that if you temporarily "tie up" the bouncer (using a drug called NU7441 to inhibit DNA-PK), the super-siren works much better. The immune system wakes up fully and attacks the cancer more effectively.

The Takeaway:
This discovery is like finding a missing piece of the puzzle for how our immune system knows when to stop fighting. It explains why some people get stuck in a state of constant inflammation (the bouncer isn't working) and offers a new strategy for cancer treatment: If you want the immune system to attack a tumor, you might need to temporarily disable the "brakes" (DNA-PK) so the "gas pedal" (STING agonists) can work at full speed.

Summary in One Sentence

The body has a protein called DNA-PK that acts like a safety valve, grabbing and neutralizing immune alarm signals to prevent panic, but this same protein can also be tricked to help cancer drugs work better by temporarily removing that safety valve.

Technical Summary

1. Problem Statement

The cGAS-STING pathway is a critical component of the innate immune system, detecting cytosolic double-stranded DNA (dsDNA) and producing cyclic dinucleotides (CDNs), specifically 2'3'-cGAMP, to activate the STING adaptor protein. This triggers a signaling cascade leading to Type I Interferon (IFN) production.

• The Gap: While regulatory mechanisms for cGAS and STING proteins are well-documented, there is a lack of identified direct mammalian regulators of intracellular CDNs (the second messengers themselves).
• The Conflict: Previous literature presents conflicting data regarding the role of DNA-dependent protein kinase catalytic subunit (DNA-PKcs). Some studies suggest it acts as a co-activator of cGAS or a dsDNA sensor, while others indicate that its absence leads to potentiated Type I IFN responses, suggesting an inhibitory role. The mechanistic link between DNA-PKcs and the regulation of CDN signaling remains unclear.
• Therapeutic Context: STING agonists are being developed as immunotherapeutics (e.g., for cancer), but their efficacy can be dampened by endogenous regulatory mechanisms. Understanding how CDNs are regulated is crucial for optimizing these therapies.

2. Methodology

The authors employed a multidisciplinary approach combining in vitro biochemistry, cell biology, structural modeling, and in vivo mouse models.

• Binding Assays:
  • Immunoprecipitation (IP) & Pull-downs: Used FLAG-tagged, endogenous, and recombinant DNA-PKcs to test direct binding with 2'3'-cGAMP, 3'3'-cGAMP, c-di-AMP, and c-di-GMP.
  • ELISA: Quantified the amount of CDNs bound to DNA-PKcs after IP.
  • Thermal Shift Assays (TSA): Measured protein stabilization upon ligand binding to confirm direct interaction and identify binding domains.
• Kinase Activity Assays:
  • Used an ADP-Glo assay to measure DNA-PKcs ATP hydrolysis activity in the presence of various CDNs and STING agonists (E7766, ADU-S100, diABZI).
• Cellular Models:
  • Cell Lines: T98G (cGAS-null), THP-1 (myeloid), and HEK293T cells.
  • Genetic Manipulation: Used siRNA knockdown, CRISPR-Cas9 knockout (DNA-PKcs, STING, cGAS, IFNAR), and rescue experiments with human STING haplotypes.
  • Pharmacological Inhibition: Utilized NU7441 (a specific DNA-PKcs inhibitor) and other kinase inhibitors.
• Structural Biology:
  • Molecular Docking & Dynamics: Used ZDOCK and Gromacs to model the interaction of CDNs and STING agonists within the catalytic pocket of DNA-PKcs (PDB: 5LUQ).
• In Vivo Models:
  • Wild-type, Sting⁻/⁻, and Ifnar⁻/⁻ mice treated with NU7441 followed by STING agonists (diABZI) to assess systemic inflammatory responses.
• Virology:
  • Infection assays with HSV-1, Monkeypox virus (MPXV), and VSV to measure the establishment of an antiviral state following combined STING activation and DNA-PKcs inhibition.

3. Key Contributions & Results

A. DNA-PKcs Directly Binds and Traps CDNs

• Direct Interaction: DNA-PKcs physically interacts with intracellular 2'3'-cGAMP and bacterial 3'3'-cGAMP. This interaction was confirmed via IP, pull-downs, and TSA.
• Binding Site: Structural modeling and mutagenesis (kinase domain deletion) revealed that CDNs bind specifically to the catalytic pocket of DNA-PKcs.
• Specificity: DNA-PKcs binds 2'3'-cGAMP and 3'3'-cGAMP but does not bind c-di-AMP or c-di-GMP. It also interacts with the synthetic agonist E7766 but not ADU-S100.
• Inhibition of Kinase Activity: Binding of CDNs (and E7766) to the catalytic pocket inhibits DNA-PKcs kinase activity in a dose-dependent manner, suggesting a reciprocal regulation where CDNs act as inhibitors of DNA-PKcs.

B. DNA-PKcs Acts as a Signal Terminator

• Enhanced Signaling upon Inhibition: Inhibiting DNA-PKcs (via NU7441 or genetic knockout) significantly boosts Type I IFN responses (IFNβ, ISG15, CXCL10, CCL5) triggered by 2'3'-cGAMP, 3'3'-cGAMP, and E7766.
• STING Dependence: The enhanced response requires STING but is independent of cGAS (demonstrated in cGAS-null cells) and IFNAR (autocrine loop).
• Mechanism: DNA-PKcs acts downstream of cGAS activation. It "traps" the produced CDNs, preventing them from activating STING, thereby terminating the signal and preventing chronic inflammation.
• Kinetics: In dsDNA stimulation models, DNA-PKcs inhibition initially suppresses early IFN responses (likely due to its role in cGAS co-activation) but leads to a massive boost in late-phase inflammatory cytokine production, confirming its role in signal termination.

C. Impact on STING Agonists and Therapeutics

• Differential Regulation: DNA-PKcs selectively inhibits the bioactivity of specific STING agonists.
  • E7766 & diABZI: Their activity is suppressed by DNA-PKcs (likely due to interaction with the catalytic pocket). Inhibiting DNA-PKcs potentiates their efficacy.
  • ADU-S100: Its activity is not affected by DNA-PKcs, consistent with docking data showing poor interaction.
• In Vivo Validation: In mice, co-administration of NU7441 with diABZI or 3'3'-cGAMP significantly increased inflammatory gene expression and cytokine production in the peritoneum and spleen compared to agonist alone. This effect was absent in Sting⁻/⁻ mice.

D. Functional Antiviral Consequences

• Antiviral State: Combining STING agonists with DNA-PKcs inhibition resulted in a significantly stronger antiviral state, reducing infection rates by HSV-1, MPXV, and VSV in both cell lines and primary human macrophages.
• Dual Role: The study clarifies the "dual" nature of DNA-PKcs: it aids in the initiation of the response (via cGAS co-activation) but is essential for termination (via CDN trapping) to prevent immunopathology.

4. Significance

• Novel Regulatory Mechanism: This is the first study to identify a direct mammalian regulator of intracellular CDNs. It establishes DNA-PKcs as a "molecular sink" or trap that limits the amplitude and duration of STING signaling.
• Resolution of Controversy: The findings reconcile conflicting literature by proposing a temporal model: DNA-PKcs promotes early signaling (via cGAS phosphorylation) but inhibits sustained signaling (via CDN sequestration).
• Therapeutic Implications:
  • Cancer Immunotherapy: The study suggests that the efficacy of STING agonists (particularly non-nucleotide macrocycles like E7766) may be limited by the status of DNA-PKcs in the tumor microenvironment.
  • Stratification: Patients with DNA-PKcs deficiencies or mutations might respond differently to STING-based therapies.
  • Drug Design: Future STING agonists could be engineered to avoid the DNA-PKcs catalytic pocket to prevent this "trapping" mechanism, or conversely, DNA-PKcs inhibitors could be used as adjuvants to boost STING agonist potency in immunosuppressed contexts.

In summary, the paper redefines DNA-PKcs not just as a DNA repair enzyme, but as a critical rheostat for innate immunity that balances protective antiviral responses against the risk of chronic inflammation by directly sequestering cyclic dinucleotide second messengers.