Despite recent advances in the structural biology of membrane proteins, structural and functional studies of wild-type, membrane-bound cytochrome P450 and its redox partners remain a significant challenge. A key bottleneck lies in the preparation of stable, natively folded protein samples that retain full functionality. Research in our lab has led to the development of synthetic amphipathic polymers capable of directly isolating membrane proteins along with their native lipids from cellular membranes, enabling functional reconstitution into lipid nanodiscs. These polymer-based nanodiscs enable the application of both solution and solid-state NMR experiments facilitating high-resolution structural studies. In my talk, I will present experimental findings on cytochrome P450, cytochrome b5, and cytochrome P450 reductase, demonstrating successful reconstitution and analysis of protein-protein and protein-lipid interactions. Furthermore, we exploit the intrinsic lipid exchange properties of nanodiscs to identify lipid preferences of cytochrome proteins, offering new insight into lipid-protein interactions.

CYP17A1 is a critical component of steroid hormone biosynthesis, converting progesterone to androstenedione and pregnenolone to dehydroepiandrosterone via two sequential reactions. The first hydroxylation step at the C17 position is catalyzed by Compound 1, while the mechanism of the second C17-C20 bond scission continues to be debated for many years. Using CYP17A1 incorporated in lipid Nanodisc membranes we combined spectroscopic results with functional studies to probe for the alternative productive pathways via Compound 1 or peroxo-ferric driven catalysis. The former pathway requires two electrons from a protein redox partner and two protons from water. The protonation steps are at least partially rate-limiting, therefore the steady-state rates of P450 hydroxylation are usually slower in deuterated solvent (D₂O) by a factor of 1.5 - 3. However, in CYP17A1 a pronounced inverse kinetic solvent isotope effect (KSIE ~ 0.4 – 0.8) is observed for the C-C lyase step, with the product forming rate faster in D₂O. Using numerical modeling of the P450 steady-state kinetics we demonstrate that such significant inverse KSIE cannot be obtained for a pure Compound 1 driven catalytic cycle of P450. An alternative catalytic intermediate, which does not require protons from water, the unprotonated ferric-peroxo complex explains the observed KSIE.

The peroxo- intermediate, as well as the transition state hemi-ketal, have been trapped and characterized by resonance Raman and UV-VIS spectroscopy with both 17OH-pregnenolone and 17OH-progesterone, and functionally important positional differences of these substrates were documented. With several mutations differently affecting hydroxylation and C-C lyase steps, the involvement of a heme peroxo-anion intermediate was confirmed for the latter reaction. Comparison of proton inventory measurements for the wild-type CYP17A1 and E305G mutant for hydroxylation and lyase reaction validates a different reaction mechanisms and predominant product formation via a peroxo-ferric intermediate in C-C bond scission. Supported by NIH GM118145.

Nikos S Hatzakis

^a Department of Chemistry & Nanoscience center, University of Copenhagen, Denmark,

^b Novo Nordisk Foundation Centre for optimized oligo escape and control of disease , University of Copenhagen Denmark

^*Presenting author: Hatzakis@chem.ku.dk

Abstract

Eukaryotic cytochrome P450 (CYP) activation relies on electron donations from NADPH-dependent cytochrome P450 reductase (POR). POR, through a complex network of structural dynamics orchestrates, the electron donation selectively activating downstream P450s. We recently recorded the efficient electron relay to downstream CYPs to be facilitated by the transient formation of a dynamic CYP-POR complex, “metabolon”(1). This metabolon facilitates production of dhurrin minimizing metabolic cross talk.

We recently introduced the model of biased metabolism(2), a mechanism akin to biased signaling of GPCRs, where small ligands bind to POR bias its conformational dynamics stabilizing different conformational states that are linked to distinct metabolic outcomes. We combined computational modeling and functional assays, Single molecule(3) and FRET studies(4, 5) of POR reconstituted in nanodiscs (6), and whole cell assays to evaluate the effect of ligands on POR mediate metabolism . The results of our medium throughput screening assays were implemented in a macheni learning framework that designed the next generation of POR ligands that showed strong effect on POR. In summary we have designed and evaluated pathway-specific ligands suppressing undesired, disease-related, metabolic pathways while stimulating beneficial downstream processes.

Relevant publications from my group

1 Laursen, T. et al. Science 2016, 354, 890.

2 Jensen, S. B. et al. Nat. Commun. 2021, 12, 2260.

3 Hansen J K. et al. Nature methods -2025.

4 Stella, S. et al. Cell 2018, 175, 1856..

5 Laursen, T. et al.. ACS Chem. Biol. 2014, 9, 630.