The HSP90-CDC37 Chaperone System Orchestrates RAF1 Kinase Activation Through a Pre-Dimerization Mechanism

⚕️

This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine your body is a bustling city, and inside every cell, there are millions of tiny machines called proteins that keep everything running. Some of these machines are "switches" that tell cells when to grow and divide. One of the most important switches is a protein called RAF1.

However, RAF1 is a bit of a troublemaker. When it's first made, it's floppy, unstable, and prone to falling apart. If it breaks, the cell can't function; if it gets stuck in the "on" position, it can cause cancer. To fix this, the cell employs a specialized construction crew: a team of molecular helpers called HSP90 and CDC37.

For a long time, scientists thought this crew just acted like a simple "safety blanket," wrapping around RAF1 to keep it warm until it was ready. But this new study reveals that they are actually much more like architects and sculptors, actively shaping RAF1 into its final form using a very specific, step-by-step process.

Here is the story of how they do it, broken down into simple analogies:

1. The "Casting Mold" (The Sculpting Phase)

Imagine you are a potter trying to make a perfect clay vase. You don't just let the clay sit there; you press it into a mold to give it shape.

In this study, the scientists discovered that HSP90 and CDC37 create a unique "casting mold" for RAF1.

The Problem: RAF1 has a critical part called the $\alpha$ C helix (think of it as the handle of the vase). If this handle isn't formed perfectly, the whole machine is useless.
The Solution: The chaperone team builds a special groove (the mold) using CDC37 and HSP90. They gently guide the floppy handle of RAF1 into this groove.
The "Anchor": There is a specific "hook" on RAF1 (a molecule called Histidine) that snaps into a pocket in the mold. This locks the handle in place, preventing it from flopping around, but also preventing it from growing too long too fast. It's like a safety brake that holds the handle in a "primed" position, waiting for the perfect moment to release.

2. The "Double-Client" Assembly Line

Usually, these construction crews work on one protein at a time. But RAF1 is special. The study found that HSP90 can actually hold two RAF1 proteins at once, side-by-side.

Think of this like a double-car garage.

One car (RAF1) is being worked on by the mechanics.
The second car (another RAF1) is parked right next to it, also being worked on.
This "pre-dimerization" (getting two cars ready together) is unique to RAF1. Other proteins (like CDK4, which the study used as a control) just get one car at a time.
By holding two RAF1s together, the chaperones are essentially pre-assembling a team. This is crucial because RAF1 needs to pair up with another RAF1 (or a different RAF family member like BRAF) to actually turn on the cell's growth signals.

3. The "Traffic Light" System (ATP Hydrolysis)

How does the crew know when to let go of RAF1? They use a chemical fuel called ATP.

Think of ATP as the green light for the construction crew.
When HSP90 has ATP, it holds the mold tight, keeping RAF1 stable.
When HSP90 burns the ATP (turns it into ADP), it's like the light turning red. This signals the crew to relax their grip, release the "handle" (the $\alpha$ C helix), and let RAF1 finish folding on its own.
The study also found a third helper, p23, which acts like a traffic controller. It can delay the light turning red, giving the crew more time to make sure the handle is perfectly shaped before letting RAF1 go.

4. The "Quality Control" Check

Before RAF1 is released, the crew checks its "ID card" (phosphorylation).

The study found that while RAF1 is in this mold, the crew adds specific "stamps" (chemical tags) to it.
They add "Go" stamps (activating tags) and remove "Stop" stamps (inhibiting tags).
This ensures that when RAF1 finally leaves the construction site, it is fully activated and ready to do its job.

Why Does This Matter?

This discovery changes how we view cancer treatment.

Old View: HSP90 is just a passive blanket. If we block it, the cancer protein falls apart.
New View: HSP90 is an active manager. It decides when and how the protein gets activated. It even helps cancer cells (like those with the BRAF mutation) form dangerous teams (dimers) that drive tumor growth.

The Takeaway:
This paper shows that the cell doesn't just "fix" broken proteins; it orchestrates their creation. The HSP90-CDC37 system is like a master conductor, using a mold, a double-garage, and a traffic light system to ensure that RAF1 is built perfectly, paired up correctly, and activated at the right time. Understanding this "conductor" gives scientists new ways to stop cancer cells from building their dangerous machines in the first place.

1. Problem Statement

RAF kinases (ARAF, BRAF, RAF1) are central regulators of the RAS-MAPK signaling pathway, controlling cell proliferation and survival. Dysregulation of RAF kinases is a primary driver of tumorigenesis. While it is established that the HSP90-CDC37 chaperone system is essential for the maturation and stabilization of RAF1 (and oncogenic BRAF/ARAF), the precise molecular mechanism by which this system facilitates RAF1 folding, activation, and dimerization remains poorly understood. Specifically, it is unclear how the chaperone system transitions RAF1 from an unstable, immature state to a catalytically competent, dimerization-ready enzyme, and whether this process involves unique structural intermediates distinct from other HSP90 kinase clients (e.g., CDK4).

2. Methodology

The authors employed a multi-disciplinary approach combining structural biology, proteomics, and cell biology:

Sample Preparation: RAF1-chaperone complexes were co-expressed in Expi293F cells with KRASG12V (to stabilize the GTP-bound state) and CDC37. Complexes were isolated via sequential affinity purification (Strep-tag on RAF1 and Halo-tag on KRAS) to capture transient intermediates.
Cryo-Electron Microscopy (Cryo-EM): Vitrified samples were imaged on a Titan Krios microscope. Extensive 3D classification, focused refinement, and 3D variability analysis were performed using CryoSPARC to resolve compositional and conformational heterogeneity within the sample.
Structural Modeling: Atomic models were built and refined using AlphaFold predictions, rigid-body fitting, and tools like Coot, Isolde, and PHENIX.
Phosphoproteomics: Mass spectrometry (DDA and DIA modes) was used to map phosphorylation sites on RAF1, HSP90, and CDC37 within the isolated complexes, comparing the RRHCC (2:2:2) complex with previously characterized RHC (1:2:1) complexes.
Biochemical Assays: Site-directed mutagenesis (targeting the "histidine anchor" and "acidic patch") was used to test functional importance. Pull-down assays with various RAF isoforms (wild-type, oncogenic mutants, kinase-dead) and CDK4 were performed to assess dimerization specificity.
Cellular Functional Assays: RAF-less Mouse Embryonic Fibroblasts (MEFs) were used to test the ability of different RAF homo- and heterodimers to restore MAPK signaling (pMEK levels) and cellular proliferation.

3. Key Contributions and Results

A. Structural Discovery: The RRHCC Complex and Asymmetric Folding

The study resolved previously uncharacterized intermediate assemblies, most notably the RRHCC complex (2 RAF1 : 2 CDC37 : 2 HSP90).

Asymmetric Architecture: Unlike the symmetric binding seen in other clients, the RRHCC complex is asymmetric. Two RAF1-CDC37 heterodimers bind to opposite faces of a closed HSP90 dimer.
Distinct Folding States: The two RAF1 protomers (RAF1.1 and RAF1.2) adopt radically different conformations:
- RAF1.1: Resembles the previously known RHC state; its N-lobe is largely disordered/unfolded, threading into the HSP90 lumen.
- RAF1.2: Adopts a unique "pre-dimerization" state. Its N-lobe is partially folded, with the critical $\alpha$ C helix stabilized in a specific groove formed by CDC37.1-MD and HSP90-A-MD.
The "Casting Mold": The authors identified a novel interface acting as a molecular "casting mold" that cradles the partially folded $\alpha$ $α$ C helix of RAF1.2.
- Histidine Anchor: Residue H402 of RAF1 inserts into a pocket on HSP90, acting as a nucleation point.
- Acidic Patch: An acidic patch on CDC37 interacts with the helix, acting as a "brake" to prevent premature full extension and folding, ensuring the helix remains in a primed, metastable state.

B. The RHCp23 Complex and ATPase Regulation

The authors resolved the RHCp23 complex (RAF1-HSP90-CDC37-p23).

Co-occupancy: p23 binds to the closed HSP90 clamp alongside CDC37, a state not previously seen with kinase clients.
Mechanism: p23 appears to stabilize the complex and regulate ATP hydrolysis. In the RHCp23 state, p23 may delay ATP hydrolysis at one site or facilitate the release of the client once hydrolysis occurs. The structure suggests p23's flexible tail interacts with CDC37-MD, potentially tilting it to release the captured RAF1 N-lobe.

C. Pre-Dimerization is RAF-Specific

Selective Dimerization: The RRHCC complex (2:2:2 stoichiometry) is unique to RAF kinases. While RAF1, BRAF, and BRAFV600E form homodimers within the chaperone complex, the canonical client CDK4 does not.
Role of RBD-CRD: Chimeric experiments showed that fusing the RAF-specific RAS-binding domain (RBD) and Cysteine-rich domain (CRD) to CDK4 was sufficient to induce dimerization within the chaperone complex. This suggests the RBD-CRD module, likely in conjunction with KRAS-GTP, drives the formation of these higher-order assemblies.
Heterodimerization: The chaperone system also supports the formation of RAF1-BRAF heterodimers, particularly with oncogenic BRAFV600E.

D. Phosphorylation Remodeling

Phosphoproteomics revealed that the RRHCC complex is enriched for activating phosphorylations (e.g., S338, S339, S471, T491) while lacking inhibitory phosphorylations (e.g., S29, S642) compared to the RHC complex. This indicates the chaperone environment actively primes RAF1 for signaling by modulating its phosphorylation landscape.

E. Functional Validation

Mutagenesis: Disrupting the H402 anchor or the acidic patch significantly reduced RAF1 stability or altered release kinetics, confirming the structural observations.
Signaling Output: In RAF-less MEFs, RAF1-BRAFV600E and RAF1-BRAFD594A (kinase-dead) heterodimers restored MAPK signaling and cell proliferation more effectively than monomers or certain homodimers. This demonstrates that the chaperone-mediated pre-dimerization is functionally critical for oncogenic signaling, including paradoxical transactivation.

4. Significance

Paradigm Shift: The study redefines HSP90-CDC37 from a passive "stabilizer" of unfolded proteins to an active regulator that orchestrates a stepwise, asymmetric folding pathway. It couples folding, nucleotide-state control, and post-translational modification to dimer readiness.
Mechanistic Insight: It reveals a unique "pre-dimerization" mechanism specific to RAF kinases, where the chaperone system facilitates the assembly of homo- and heterodimers before release from the membrane.
Therapeutic Implications: The identification of the "casting mold" and the "histidine anchor" provides new, specific structural targets for disrupting oncogenic RAF signaling. Unlike general HSP90 inhibitors, these findings suggest the possibility of designing isoform-specific inhibitors that block the unique pre-dimerization interface or the specific chaperone-client interactions required for RAF maturation, potentially sparing other essential HSP90 clients.

In summary, this paper provides a comprehensive structural and functional map of RAF1 maturation, revealing that the HSP90-CDC37 system actively shapes the conformational landscape of RAF kinases to ensure they are correctly folded, phosphorylated, and pre-assembled into dimers capable of driving oncogenic signaling.