Time-resolved cryo-EM reveals conformational trajectory of allosteric activation in isocitrate lyase

Time-resolved cryo-EM studies of *Mycobacterium tuberculosis* isocitrate lyase 2 reveal that acetyl-CoA binding drives asymmetric, half-of-site activation through a conformational selection mechanism that shifts the enzyme's pre-existing equilibrium toward an active state.

The Gist

Imagine a tiny, four-legged robot inside a bacterium called Mycobacterium tuberculosis. This robot is an enzyme named Isocitrate Lyase 2 (ICL2). Its job is to act as a gatekeeper, deciding when the bacterium can eat and grow.

Normally, this robot is "asleep" or inactive. It's like a car with the engine off and the doors locked. But when the bacterium needs to survive a tough infection, it sends out a specific chemical key: Acetyl-CoA. When this key fits into the robot's lock, the robot wakes up and starts working 50 times faster.

Scientists have always known that this happens, but they didn't know how the robot moved from "asleep" to "awake." It happened too fast to see with normal cameras.

The "Time-Travel" Camera

To solve this mystery, the researchers used a special high-speed camera called Time-Resolved Cryo-EM. Think of this as a super-fast strobe light that can take pictures of a moving object in slow motion, freezing it at different stages of a split-second movement.

They mixed the robot with the chemical key and took snapshots at three moments:

1. 0.15 seconds (The Instant of Awakening): Almost all the robots were still in the "asleep" position. Interestingly, one part of the robot had already fallen off (dissociated), creating a gap. This gap was actually the "open door" that allowed the key to enter.
2. 1 second (The Transition): The robots were in the middle of a dance. Some were still asleep, some were fully awake, and most were in a weird, half-way state where they were trying to figure out how to lock their new position.
3. 30 minutes (The Final Pose): All the robots were now fully awake and working.

The "Seesaw" Mechanism

The most exciting discovery was how the robot moved. It wasn't a simple "on/off" switch. Instead, the researchers found a "Seesaw" mechanism.

Imagine the robot has two arms (a dimer) that work together. When the key turns the lock, these arms don't just move up and down together. They pivot around a central point like a playground seesaw:

• Arm A swings down and locks into a "closed" position, ready to do the work (catalysis).
• Arm B swings up and stays "open," waiting for the next piece of raw material to arrive or for the finished product to leave.

Then, they switch! Arm B closes to work, and Arm A opens to rest. This "half-of-site" activity means the robot is incredibly efficient, never letting both sides get in each other's way. It's like a two-lane highway where one lane is always open for traffic while the other is being repaved, ensuring the flow never stops.

The "Stuck" Robot Experiment

To prove this theory, the scientists built a "stuck" version of the robot using tiny molecular glue (disulfide bonds) to lock the arms in the "open" position permanently.

• Result: This glued robot didn't need the chemical key to wake up. It was always ready to work.
• Bonus: Because both arms could work at the same time (instead of taking turns on the seesaw), this glued robot was actually twice as fast as the normal, activated robot! However, it was a bit pickier about what it ate (it needed a higher concentration of food to get started).

Why This Matters

This study is a big deal because it shows us the movie of how a protein changes shape, rather than just two still photos (before and after).

It proves that enzymes don't just wait for a key to force them open. Instead, they are constantly wiggling and dancing in many different shapes. The chemical key just happens to grab the ones that are already in the "right" shape and says, "Stay like that!" This is called Conformational Selection.

In short: The bacterium's survival machine is a dynamic, seesawing robot. By using a high-speed camera, scientists finally saw the robot wake up, figured out its unique two-step dance, and even built a super-fast version of it. This helps us understand how bacteria survive infections and could lead to new ways to shut them down.

Technical Summary

1. Problem Statement

Isocitrate lyase 2 (ICL2) from Mycobacterium tuberculosis (Mtb) is a critical enzyme for bacterial survival during persistent infection, acting as a gatekeeper for central carbon metabolism. While it is known that ICL2 undergoes a dramatic conformational rearrangement upon binding the allosteric effector acetyl-CoA, increasing catalytic efficiency by 50-fold, the mechanistic link between effector binding and these structural/functional changes remained unclear. Specifically, the temporal sequence of conformational states and the dynamic mechanism driving the transition from an inactive tetramer to an active state were not fully understood.

2. Methodology

The authors employed a multi-disciplinary approach to capture the dynamic activation trajectory of Mtb-ICL2:

• Time-Resolved Cryo-Electron Microscopy (Cryo-EM): This was the primary technique used to visualize conformational states at specific time points after mixing the enzyme with its substrate (isocitrate/Mg²⁺) and the allosteric effector (acetyl-CoA).
  • 0.15 seconds: Achieved using a custom microfluidics mixer-sprayer-plunger apparatus for rapid mixing and immediate vitrification.
  • 1 second and 30 minutes: Achieved via manual mixing on grids followed by flash-freezing.
• Complementary Techniques:
  • Small-Angle X-ray Scattering (SAXS): To validate global conformational changes in solution.
  • Isothermal Titration Calorimetry (ITC): To measure binding thermodynamics and cooperativity.
  • Kinetic Analysis: To determine turnover numbers ($k_{cat}$) and Michaelis constants ($K_M$).
  • Protein Engineering: Creation of specific mutants (e.g., disulfide-locked variants and linker charge mutants) to test mechanistic hypotheses.
• Data Analysis: 3D variability analysis was performed on cryo-EM data to identify continuous motions within the active states.

3. Key Results

A. Conformational Trajectory of Activation

Time-resolved cryo-EM captured a distinct temporal progression of ICL2 states:

• 0.15 sec (Early State): The enzyme is predominantly (99%) in an inactive state. Crucially, one pair of C-terminal dimers has dissociated, with one C-terminal domain lacking density. This suggests that dissociation of the C-terminal dimer is a prerequisite for acetyl-CoA binding.
• 1 sec (Intermediate State): The population is heterogeneous. While inactive (16%) and fully active (32%) forms exist, the majority (52%) are in intermediate stages where monomers form transient contacts with the N-terminal core before stabilizing as active dimers.
• 30 min (Equilibrium State): Only active conformations remain. The population consists of:
  • Symmetric Active State (68%): Resembles the known acetyl-CoA-bound crystal structure (RMSD 1.7 Å).
  • Asymmetric Active State (32%): A novel state not seen in crystal structures, showing density for only one active C-terminal dimer per tetramer.

B. Mechanism of Allostery: Conformational Selection

• Cooperativity: Acetyl-CoA binding exhibits positive cooperativity (Hill number = 1.6), where initial binding enhances affinity for subsequent binding.
• Model Validation: The data supports a conformational selection model. Acetyl-CoA does not induce a new fold but shifts a pre-existing, weakly populated equilibrium toward the active state. The dissociation of the C-terminal dimer creates a high-affinity state for the effector.

C. The "Seesaw" Mechanism and Half-of-Site Reactivity

• Dynamic Motion: 3D variability analysis revealed a 5° pivot of the active C-terminal dimer around a fulcrum (Q635-E640 loop), causing a 4 Å displacement.
• Asymmetric Regulation: This motion acts as a "seesaw":
  • In one monomer, the flexible linker becomes ordered and stabilizes the active site loop in a "closed" conformation (catalytically active).
  • In the opposing monomer, the linker remains disordered, keeping the active site loop "open" (allowing substrate entry/product release).
• Result: This mechanism enforces half-of-site reactivity, where only one active site per dimer is engaged at a time.

D. Mutational Validation

• Linker Mutation (Q597E/H598E): Enhancing charge interactions between the linker and active site loop increased activity to wild-type levels even without acetyl-CoA.
• Disulfide-Locked Variant (E640C/N731C/E733C): Locking the C-terminal domains in the active conformation resulted in:
  • No change in acetyl-CoA binding affinity.
  • Adoption of the active conformation without acetyl-CoA (confirmed by SAXS).
  • Double the turnover number ($k_{cat}$) compared to the activated wild-type (2.5 s⁻¹ vs. 1.2 s⁻¹), suggesting that in the locked variant, all four active sites are catalytically proficient simultaneously, unlike the wild-type which is limited by the seesaw mechanism.

4. Significance and Contributions

• Methodological Advancement: This study demonstrates the power of time-resolved cryo-EM to resolve transient intermediates and dynamic trajectories in metabolite-driven enzyme allostery, moving beyond static snapshots.
• Mechanistic Insight: It provides the first structural visualization of the conformational selection process in a metabolic enzyme, detailing how effector binding shifts equilibrium via dimer dissociation and re-association.
• Regulatory Mechanism: The discovery of the "seesaw" mechanism explains how Mtb-ICL2 achieves half-of-site reactivity, a regulatory feature critical for fine-tuning metabolic flux.
• Therapeutic Implications: Understanding the precise activation trajectory and the essential role of C-terminal dissociation offers new targets for drug design aimed at inhibiting Mtb survival during persistent infection.

In summary, the paper elucidates the structural trajectory of Mtb-ICL2 activation, revealing that acetyl-CoA binding selects for a dissociated intermediate that reorganizes into an active state governed by a dynamic, asymmetric "seesaw" motion.