Structure–activity relationships of human AKR-type oxidoreductases involved in bile acid synthesis: AKR1D1 and AKR1C4

Wen‐Hwa Lee, Petra Lukacik, Kunde Guo, E. Ugochukwu, K.l. Kavanagh, Brian D. Marsden, Udo OppermannMolecular 2008 · DOI

TL;DR

This paper investigates the structure-activity relationships of two human enzymes, AKR1D1 and AKR1C4, crucial for bile acid synthesis. By combining X-ray crystallography, homology modeling, and in silico docking, the researchers identified a key tryptophan residue ("switch tryptophan") that significantly influences substrate specificity, offering insights into enzyme evolution and potential drug design.

What Problem Does This Paper Address?

Bile acid synthesis is a vital metabolic pathway for cholesterol elimination and digestion. This complex process involves numerous enzymatic steps, with aldo-keto reductases (AKRs) playing a critical role in modifying steroid intermediates. Specifically, AKR1D1 and AKR1C4 are involved in reducing double bonds and ketone groups, respectively, in the A-ring of these intermediates. However, the AKR superfamily is known for its broad substrate specificity and plasticity, making it challenging to understand how subtle structural differences lead to distinct catalytic activities and specificities. This paper aims to elucidate the structural basis for the substrate specificity of AKR1D1 and AKR1C4, focusing on how their active sites accommodate different steroid substrates and products.

Methods

The study employed a multi-pronged approach:

  • X-ray Crystallography: The crystal structure of AKR1C4 was determined, providing a high-resolution snapshot of the enzyme's active site.
  • Homology Modelling: For AKR1D1, which lacked a crystal structure at the time, homology models were built using existing AKR structures as templates. This technique leverages evolutionary relationships to predict the 3D structure of a protein based on sequences of known structures.
  • In Silico Docking: This computational technique was used to predict how substrates and products bind within the active sites of the modelled enzymes. By simulating the interaction between ligands and receptors, docking helps identify favorable binding poses and understand the forces driving these interactions.
  • Loop Modelling and Grafting: To address specific active site features, such as an occluded entrance in the AKR1C4 crystal structure, researchers used loop modelling to create alternative conformations and grafted these modified loops onto the existing structure.

Datasets

The study utilized existing protein data bank (PDB) entries for various AKR enzymes, including:

  • AKR1C1 (1MRQ)
  • AKR1C2 (1IHI, 1J96)
  • AKR1C3 (1RY0, 1S1P)
  • AKR1C4 (2FVL)
  • Rat AKR1C9 (1AFS)
  • Human AKR1D1 (3BUV, 3BUR, 3COT, 3BV7, 3CMF)

These structures served as templates for homology modeling and as benchmarks for validating docking results.

Key Insights and Ideas

  • The "Switch Tryptophan": A key finding is the identification of a conserved tryptophan residue (Trp227/230) in the active site, termed the "switch tryptophan." This residue can adopt two distinct rotameric states: "across" (facing the active site entrance) and "inward" (pointing into the active site core).
  • Rotameric State Dictates Binding: The orientation of this "switch tryptophan" is strongly correlated with the binding mode of steroid substrates and dictates the enzyme's catalytic activity. The "inward" rotamer appears to be crucial for productive catalysis in AKR1D1 and AKR1C4.
  • Flexibility and Specificity: The study highlights how subtle changes in the active site, particularly the conformation of the "switch tryptophan" and surrounding residues, can fine-tune substrate specificity without requiring major loop rearrangements. This explains the observed plasticity of the AKR superfamily.
  • Validation of Docking Protocols: The researchers successfully validated their in silico docking and homology modeling approaches by comparing the predicted poses with high-resolution crystal structures of AKR1D1 in complex with steroids.

Key Results

  • Docking experiments using models incorporating an "inward" rotamer of the "switch tryptophan" successfully predicted catalytically competent poses for AKR1D1 and AKR1C4.
  • The study identified Trp227/230 as a critical determinant of substrate specificity in family 1 AKRs.
  • The "switch tryptophan" in its "inward" conformation positions the steroid substrate for efficient hydride transfer from the NADP(H) cofactor.
  • The "across" rotamer of the "switch tryptophan" was found to be less productive for catalysis, often leading to non-productive poses.
  • The study demonstrated that small side-chain rearrangements, rather than major loop conformational changes, are sufficient to achieve differential substrate specificity and reaction outcomes.

Limitations and Future Work

  • Homology Model Limitations: While homology modeling was essential for AKR1D1, it relies on the accuracy of the template structures and the evolutionary conservation between the target and template proteins. The paper acknowledges that loop positions might differ to some extent.
  • In Silico Nature: Docking experiments are predictive and require experimental validation. The study's strength lies in its validation against experimental structures.
  • Future Directions: The authors suggest extending these modeling and docking approaches to other members of the AKR family to understand their molecular mechanisms, predict new ligands, and design novel enzymes with specific catalytic functions. They also propose using these insights for experimental designs like site-directed mutagenesis to further validate the proposed binding modes.

Summary updated

BibTeX
@inproceedings{lee2008structureactivity,
  author = {Wen‐Hwa Lee and Petra Lukacik and Kunde Guo and E. Ugochukwu and K.L. Kavanagh and Brian D. Marsden and Udo Oppermann},
  title = {{Structure–activity relationships of human AKR-type oxidoreductases involved in bile acid synthesis: AKR1D1 and AKR1C4}},
  booktitle = {Molecular},
  year = {2008},
  doi = {10.1016/j.mce.2008.09.042},
}
Figure