Allosteric disulfide control of ligand binding and endocytosis of the natural killer cell receptor for HLA-G

This study reveals that the natural killer cell receptor KIR2DL4 utilizes an allosteric disulfide switch between Cys10-Cys28 and Cys28-Cys74 configurations, regulated by protein disulfide isomerase, to control its transition from an inactive state to an HLA-G-binding form that triggers endocytosis and supports placental development.

The Gist

The Big Picture: A Diplomatic Dance at the Border

Imagine your body is a kingdom, and your immune system is the army of guards (Natural Killer cells) patrolling the borders. Usually, these guards are trained to spot and destroy "invaders" like viruses or cancer cells.

However, during pregnancy, there is a special situation. The baby (fetus) is technically "half-foreign" because it has half its DNA from the father. The mother's immune system needs to be careful not to attack the baby, but it also needs to stay alert against real threats.

To solve this, the baby's cells (trophoblasts) send out a diplomatic signal called HLA-G. This is like a "Peace Flag." When the mother's immune guards see this flag, they don't attack; instead, they switch into "construction mode," releasing chemicals that help build blood vessels to feed the growing baby.

The guard's receptor that reads this flag is called KIR2DL4. But here's the mystery: How does this receptor know when to grab the flag and pull it inside the cell to start the construction work?

The Discovery: A Molecular Switch

The scientists in this paper discovered that the KIR2DL4 receptor has a hidden molecular switch made of a chemical "knot" called a disulfide bond.

Think of the receptor as a folding chair.

• State A (The Folded Chair): In its resting state, the chair is folded up tight. It's sitting in the storage room (inside the cell). It looks like a normal chair, but it can't hold anything. It cannot grab the "Peace Flag" (HLA-G).
• State B (The Open Chair): When the chair needs to work, it unfolds. Now it has a flat seat and can hold the flag.

The paper reveals that the "knot" holding the chair together can change its shape. It's like a knot that can be tied in two different ways:

1. Knot Type 1 (Cys10-Cys28): This is the "Folded" state. The receptor is stuck inside the cell, waiting.
2. Knot Type 2 (Cys28-Cys74): This is the "Open" state. The receptor moves to the surface of the cell, ready to catch the flag.

The Mechanism: The "Unknotter" (PDI)

How does the chair know when to unfold? The paper found a specific enzyme called PDI (Protein Disulfide Isomerase).

Think of PDI as a specialized mechanic or a knot-undoer.

• When the receptor is in the "Folded" state (Knot Type 1), the PDI mechanic comes along and unties that specific knot.
• Once untied, the receptor instantly re-ties itself into the "Open" state (Knot Type 2).
• Now, the receptor is on the cell surface, looking like an open chair, ready to grab the HLA-G flag.

If you block the PDI mechanic (using inhibitors), the knot stays tied in the "Folded" state. The receptor stays stuck inside the cell, can't grab the flag, and the baby doesn't get the help it needs for blood vessel growth.

The Evidence: How They Proved It

The scientists didn't just guess; they played detective:

1. The Random Breakage Test: They broke the receptor in 293 different random ways (mutagenesis). They found that if they broke the "knot" at specific spots (Cysteine 10 or 74), the receptor got stuck in the wrong place. Some couldn't grab the flag; others couldn't move to the surface.
2. The X-Ray Vision: They looked at the receptor under a microscope (crystal structure) and saw the "Folded" knot. But when they used a computer program (AlphaFold) to predict what the "Open" version should look like, it showed a different knot shape that perfectly matched the "Open" chair.
3. The Chemical Test: They took the receptor out of the cell and gave it to the PDI mechanic. The mechanic successfully untied the knot, proving that this enzyme controls the switch.

Why This Matters

This discovery is like finding the ignition key for a car that only starts when you are in a specific neighborhood.

• For Pregnancy: It explains how the mother's immune system knows exactly when to help the baby grow. The "switch" ensures the receptor is only active when it needs to be, preventing the immune system from getting confused or attacking the baby.
• For Science: It shows that proteins can change their shape and function just by swapping a tiny chemical knot. This "allosteric disulfide switch" might be a secret weapon used by many other proteins in our bodies to control complex processes.

In a Nutshell

The mother's immune cells have a special receptor that acts like a shape-shifting lock.

• Locked (Inside): It's folded up and can't do anything.
• Unlocked (Outside): A chemical "key" (PDI) changes the lock's shape, allowing it to grab the baby's "Peace Flag" (HLA-G).
• Result: The immune cell grabs the flag, goes back inside, and starts building a highway for the baby to grow.

Without this tiny chemical switch, the immune system might not know how to support the pregnancy, highlighting a beautiful and precise molecular dance that happens every day in human reproduction.

Technical Summary

1. Problem Statement

Human Leukocyte Antigen-G (HLA-G) is selectively expressed by fetal trophoblast cells during early pregnancy. It interacts with the Natural Killer (NK) cell receptor KIR2DL4 to induce a secretory response that supports placental development and vascular remodeling. While it is known that KIR2DL4 resides in endosomes and binds soluble HLA-G to trigger signaling, the structural mechanisms governing this interaction and the receptor's endocytosis were unknown. Specifically, direct binding of HLA-G to purified KIR2DL4 had not been reported, and the structural features controlling ligand binding versus constitutive endocytosis remained elusive.

2. Methodology

The authors employed a multi-disciplinary approach combining random mutagenesis, structural biology, mass spectrometry, and biochemical assays:

• Random Mutagenesis Screen: A library of 293 HA-tagged KIR2DL4 mutants (single and double amino acid changes) was generated using low-fidelity PCR. These were transfected into 293T cells and incubated with soluble HLA-G conjugated to Alexa-647.
• Confocal Microscopy: Cells were imaged to determine the subcellular localization of KIR2DL4 (plasma membrane vs. intracellular vesicles) and its colocalization with HLA-G.
• Mass Spectrometry (MS) & Redox Analysis:
  • Disulfide Mapping: Recombinant KIR2DL4 (from insect cells) and KIR2DL4 from human 293T cells were analyzed using differential cysteine alkylation (12C-IPA and 13C-IPA) followed by LC-MS/MS to identify disulfide bond pairings.
  • Redox State Quantification: The fraction of reduced vs. oxidized cysteines was calculated based on ion abundance ratios.
• Enzymatic Assays: Recombinant KIR2DL4 was incubated with thiol isomerases (PDI, Erp57, Thioredoxin) to test for bond reduction.
• Pharmacological Inhibition: 293T cells expressing KIR2DL4 were treated with PDI inhibitors (DTNB, pCMPS, Rutin) and anti-PDI antibodies to assess effects on receptor localization and HLA-G uptake.
• Structural Modeling: AlphaFold2 and AlphaFold3 were used to predict the 3D structure of KIR2DL4 in different disulfide configurations and to model the KIR2DL4–HLA-G complex.

3. Key Contributions

• Identification of an Allosteric Disulfide Switch: The study identifies a unique regulatory mechanism in KIR2DL4 involving three cysteine residues (Cys10, Cys28, Cys74) in the D0 domain.
• Discovery of Two Isoforms: The authors demonstrate that KIR2DL4 exists in two distinct disulfide-bonded isoforms in human cells:
  1. Cys10–Cys28: The "inactive" state (found in crystal structures), which resides in endosomes but cannot bind HLA-G.
  2. Cys28–Cys74: The "active" state, which localizes to the plasma membrane and binds HLA-G.
• Role of Protein Disulfide Isomerase (PDI): The study establishes that PDI is required to reduce the Cys10–Cys28 bond, allowing the formation of the Cys28–Cys74 bond necessary for ligand binding.
• Structural Basis for Binding: Using AlphaFold, the authors elucidated how the disulfide switch induces an allosteric conformational change in a specific loop (Val45–Pro48), enabling HLA-G binding.

4. Key Results

• Mutagenesis Phenotypes:
  • Mutants retaining the Cys10–Cys28 bond (e.g., C74W, C74R) were found in intracellular vesicles but failed to bind HLA-G.
  • Mutants forcing the Cys28–Cys74 bond (e.g., C10R, C10S, C10A) were retained at the plasma membrane and bound HLA-G, but required HLA-G to trigger internalization (they did not internalize spontaneously).
  • Mutants disrupting the Cys28 partner (C28S) resulted in misfolded proteins.
• Mass Spectrometry Findings:
  • In both insect-cell recombinant and human-cell derived KIR2DL4, both the Cys10–Cys28 and Cys28–Cys74 bonds were detected.
  • In human cells, the Cys10–Cys28 form was ~48% and the Cys28–Cys74 form was ~35%, confirming the existence of both isoforms in a physiological context.
• Enzymatic Regulation:
  • PDI specifically reduced the Cys10–Cys28 bond in vitro, converting it to a state capable of forming Cys28–Cys74. Erp57 and Thioredoxin had no effect.
  • Inhibiting PDI (via DTNB, pCMPS, or antibodies) trapped KIR2DL4 at the plasma membrane and prevented HLA-G uptake, mimicking the C10R mutant phenotype.
• Structural Insights:
  • The Cys10–Cys28 bond has a –RHstaple conformation, a signature of labile, high-stress allosteric bonds prone to cleavage.
  • AlphaFold modeling revealed that switching to the Cys28–Cys74 bond reorients the Val45–Pro48 loop in the D0 domain. In the Cys10–Cys28 state, this loop is too distant to interact with HLA-G. In the Cys28–Cys74 state, the loop moves ~5.4 Å, positioning Val45 and Pro48 to contact Met76 and Glu19 of HLA-G.
  • The D2 domain contacts remain conserved regardless of the disulfide state.

5. Significance

This paper provides a mechanistic explanation for how NK cells distinguish between "self" and "fetal" signals during pregnancy. The allosteric disulfide switch acts as a functional gatekeeper:

1. Regulation of Localization: The Cys10–Cys28 bond keeps the receptor in endosomes (inactive for binding).
2. Activation: PDI activity at the cell surface reduces this bond, allowing the receptor to adopt the Cys28–Cys74 configuration, translocate to the plasma membrane, and bind soluble HLA-G.
3. Physiological Impact: This mechanism ensures that KIR2DL4 only triggers the pro-pregnancy secretory response (vascular remodeling) when it successfully encounters fetal HLA-G.

This discovery adds KIR2DL4 to the list of proteins regulated by extracellular PDI-mediated disulfide switching (like integrins and HIV gp120) and highlights a novel redox-based control mechanism critical for successful human reproduction.