Molecular design principles for Photosystem I-based biohybrid solar fuel catalysts

This study presents the first molecular structures of active Photosystem I-platinum nanoparticle biohybrids to define the interface topology and electron transfer pathways, thereby establishing design principles for optimizing protein-nanomaterial architectures for solar fuel production.

The Gist

Imagine you have a tiny, ancient solar-powered factory inside a plant cell. This factory is called Photosystem I (PSI). Its job is to catch sunlight and turn it into electricity (in the form of moving electrons). Usually, this electricity is used to make food for the plant. But scientists want to hijack this factory to make something else: hydrogen fuel, which is a clean energy source we can store and use later.

To do this, the scientists attached tiny specks of platinum (a metal catalyst) to the factory. Think of the platinum specks as tiny buckets waiting to catch the electricity and turn it into hydrogen gas.

This paper is about figuring out exactly how to attach those buckets so they catch the electricity as efficiently as possible. The researchers built two different versions of these "bio-hybrid" factories and used a super-powerful microscope (Cryo-EM) to take 3D pictures of them.

Here is the breakdown of their discovery using simple analogies:

1. The Two Factory Designs

The scientists tested two different ways to build their solar-fuel system:

• Design A: The Full Factory (Trimeric PSI)

  • This is the complete, natural version of the plant's solar factory. It has all its parts, including a "loading dock" on the bottom where the electricity usually exits.
  • The Problem: When they attached the platinum buckets here, the buckets couldn't get very close to the exit door. It was like trying to park a large truck in a driveway that was blocked by a fence. The buckets were too far away from the electricity source, so the transfer was slow and inefficient.
  • The Result: It worked, but not great.

• Design B: The Stripped-Down Factory (PSI Core)

  • The scientists took the "loading dock" (a group of proteins called PsaC, PsaD, and PsaE) off the factory. They removed the fence!
  • The Surprise: Without the fence, the platinum buckets could park right up against the exit door. You'd think this would be much faster.
  • The Result: Surprisingly, it was slower.

2. Why Did the "Better" Design Fail?

This is the most interesting part of the story. Even though the buckets were closer to the electricity in the stripped-down factory, the system produced less fuel. Why?

• The "Battery Leak" Analogy:
  In the full factory, the electricity travels a long path through the plant's internal wiring. This path acts like a buffer or a shock absorber. It keeps the electricity stable for a long time (about 65 milliseconds), giving it plenty of time to find the bucket.

  In the stripped-down factory, the scientists cut the path short. The electricity is now much closer to the bucket, but it's also much closer to the "hole" (the starting point) where it can leak back out before it gets captured. It's like trying to catch a ball thrown from 1 foot away versus 10 feet away. If you are too close, the ball might bounce back to the thrower before you can grab it.

  Because the electricity "leaked" back too quickly (recombination), the platinum buckets never got enough charge to make hydrogen.

3. The Secret of the "Glue"

The paper also explains how the platinum sticks to the plant protein.

• Imagine the plant protein is a magnet with a positive side (like the North pole).
• The platinum buckets are coated in a chemical that makes them negative (like the South pole).
• They stick together because opposite charges attract.
• The researchers found that the "glue" is strongest at specific spots on the plant protein. By understanding exactly which amino acids (the building blocks of the protein) act as the glue, they can design better systems in the future.

4. The Big Lesson for the Future

The main takeaway is that getting closer isn't always better.

To make a great solar-fuel system, you need a balance:

1. The Bucket: It needs to be close enough to catch the electricity.
2. The Buffer: The factory needs to keep the electricity stable long enough so it doesn't leak back out before the bucket catches it.

The Blueprint:
The scientists say that to build the ultimate solar-fuel machine, we shouldn't just strip away parts to get closer. Instead, we need to:

• Keep the "buffer" (the long path) to prevent leaks.
• But engineer the "loading dock" so the buckets can get closer without blocking the path.
• Maybe even build a "conveyor belt" (using other molecules) to bring the electricity to the bucket faster than it can leak back.

Summary

This paper is like a mechanic taking apart a car engine to see how to make it run on a new type of fuel. They found that simply removing parts to get the fuel injector closer to the engine didn't work because the engine lost its stability. Now, they have a detailed map (the 3D structures) showing exactly where to put the fuel injector and how to keep the engine running smoothly, paving the way for clean, solar-powered hydrogen fuel in the future.

Technical Summary

1. Problem Statement

Direct solar-to-chemical conversion using Photosystem I (PSI) coupled with abiotic catalysts (specifically Platinum Nanoparticles, PtNPs) offers a promising route for sustainable hydrogen ($H_2$) production. However, the rational design of these "biohybrid" systems has been hindered by a lack of high-resolution structural data. Key challenges include:

• Unclear Binding Interfaces: The specific molecular interactions and binding sites of PtNPs on the PSI surface were not fully understood.
• Electron Transfer Efficiency: It was unclear how the geometry of the protein-nanoparticle interface affects electron transfer rates from PSI cofactors to the catalyst.
• Scaffold Limitations: The role of stromal subunits (PsaC, PsaD, PsaE) and terminal iron-sulfur clusters ($F_A$, $F_B$) in modulating catalyst access and electron transfer pathways was not structurally defined.
• Performance Gaps: While biohybrids can produce $H_2$, turnover frequencies are often lower than biological electron transfer rates, suggesting kinetic bottlenecks or suboptimal catalyst positioning.

2. Methodology

The authors employed a multidisciplinary approach combining structural biology, biochemistry, and computational modeling:

• Sample Preparation:
  • Trimeric PSI: Isolated from the thermophilic cyanobacterium Thermosynechococcus vestitus ($Pt\text{-}PSI_{tri}^{T.v.}$).
  • Core PSI: Generated from the mesophilic Synechococcus leopoliensis by treating trimeric PSI with 6.8 M urea to remove stromal subunits (PsaC, PsaD, PsaE) and terminal $[4Fe-4S]$ clusters ($F_A$, $F_B$), leaving the core with only the $F_X$ cluster ($Pt\text{-}PSI_{cor}^{S.leo.}$).
  • Biohybrid Formation: Negatively charged mercaptosuccinic acid-stabilized PtNPs (~2 nm diameter) were electrostatically self-assembled onto the stromal face of both PSI variants.
• Structural Determination (Cryo-EM):
  • Single-particle cryo-electron microscopy (cryo-EM) was used to determine structures at 3.39 Å (trimeric) and 3.57 Å (core) resolution.
  • PtNP binding sites were identified as high-density regions in the electron density maps.
• Computational Modeling (MD Simulations):
  • Multiple 1 µs Molecular Dynamics (MD) simulations were performed on the resolved structures (and a previously solved S. lividus trimer) using NAMD and CHARMM36 force fields.
  • Simulations analyzed protein dynamics, side-chain rotamer states, and the stability of PtNP-PSI interactions over time.
• Characterization:
  • Photocatalytic $H_2$ production was measured via gas chromatography under illumination.
  • Size-exclusion chromatography (SEC) and Transmission Electron Microscopy (TEM) confirmed oligomeric states and PtNP sizes.

3. Key Contributions

• First High-Resolution Structures of Active Biohybrids: The paper presents the first atomic-level structures of active PSI-PtNP assemblies, comparing a full thermophilic trimer with a mesophilic core lacking terminal subunits.
• Mapping Binding Topology: The study definitively identifies two distinct PtNP binding sites on trimeric PSI (Site A and Site B) and a single dominant site on the core, revealing how steric and electrostatic factors dictate binding.
• Mechanistic Insight into Electron Transfer: By correlating structure with MD simulations and catalytic data, the authors establish a direct link between interface geometry, cofactor-to-catalyst distance, and electron transfer efficiency.
• Design Principles: The work provides a blueprint for engineering bio-nano interfaces, moving beyond trial-and-error to rational design based on electrostatics and steric constraints.

4. Key Results

A. Structural Findings

• Trimeric PSI ($Pt\text{-}PSI_{tri}^{T.v.}$):
  • Contains two PtNP binding sites per monomer.
  • Site A: Located near the native electron acceptor binding site (subunits PsaA, PsaE, PsaC, PsaF, PsaJ), ~14 Å from the terminal $F_B$ cluster. It overlaps with the native acceptor patch but is sterically hindered by the stromal subunits.
  • Site B: A distal site near the periphery (PsaB, PsaE, PsaF, PsaX), ~38 Å from $F_B$, likely non-productive for catalysis.
• Core PSI ($Pt\text{-}PSI_{cor}^{S.leo.}$):
  • Lacks PsaC, PsaD, PsaE, and $F_A/F_B$.
  • Shows a single, highly productive binding site located almost exactly where native acceptors bind, but now directly adjacent to the $F_X$ cluster (~10 Å distance).
  • Removal of stromal subunits exposes a positively charged patch on PsaA and PsaB, allowing the PtNP to bind closer to the electron transfer chain.

B. Molecular Dynamics & Interactions

• Electrostatics: Binding is driven primarily by electrostatic interactions between the negatively charged PtNP surface and positively charged residues (Lys, Arg) on the PSI stromal face.
• Key Residues:
  • Site A (Trimer): Dominated by PsaA residues (Lys27, Lys30, Arg36).
  • Site B (Trimer): Involves PsaB-Lys314 and PsaF-Lys155; dependent on the presence of subunit PsaX (absent in some species like S. leopoliensis).
  • Core: The PtNP interacts strongly with PsaA (Lys27, Lys30) and exposed PsaB residues (Lys545, Arg564).
• Binding Strength: MD simulations suggest strong binding free energies (~−10 to −20 kcal/mol), corresponding to sub-nM dissociation constants, which is significantly tighter than native acceptor binding.

C. Catalytic Performance & Kinetics

• Trimeric Biohybrids: Achieved a turnover frequency (TOF) of 5,360 mol $H_2$ mol PSI$^{-1}$ h$^{-1}$. However, electron transfer is limited because only a small fraction of PtNPs are positioned close enough to $F_B$ for efficient tunneling; most are sterically blocked by stromal subunits.
• Core Biohybrids: Despite the PtNP being closer to the electron source ($F_X$) and the electron being more reducing (higher driving force), the TOF dropped significantly to 1,300 mol $H_2$ mol PSI$^{-1}$ h$^{-1}$.
• The Bottleneck: The reduced activity in the core system is attributed to the shortened charge-separated state lifetime. Removing $F_A/F_B$ reduces the lifetime from ~65 ms to ~1 ms. This accelerates unproductive charge recombination ($F_X^-$ to $P700^+$) faster than the electron can transfer to the PtNP, and donor-side supply (reduction of $P700^+$) becomes the rate-limiting step.

5. Significance and Design Principles

The study establishes three critical design principles for optimizing PSI-based solar fuel systems:

1. Donor-Side Engineering is Crucial: Merely shortening the electron transfer chain or increasing driving force is insufficient. Strategies to rapidly re-reduce $P700^+$ (e.g., covalent tethering of redox partners, conductive electrodes) are required to prevent charge recombination.
2. Interface Geometry Optimization: To maximize efficiency, catalysts must be positioned within effective tunneling distance (<14 Å) of the terminal cofactors. The stromal subunits in native PSI act as a steric barrier; engineering the interface to expose the core or modifying catalyst size/shape can improve the fraction of productive binding events.
3. Electrostatic Tuning: Utilizing conserved positively charged patches on the PSI core (e.g., PsaA Lys/Arg residues) can guide catalyst binding to productive sites, even in the absence of native acceptor subunits.

Conclusion: This work provides a structural and mechanistic blueprint for engineering next-generation biohybrid catalysts. It demonstrates that while removing terminal subunits allows closer catalyst access, preserving the long-lived charge separation of the native system is equally vital for high-efficiency solar fuel production.