How Spectroscopy and Computational Techniques Illuminated Glycerol’s Impact on ADO Catalysis

In a recent study published in the Journal of Biological Chemistry, the researchers integrated multiple techniques, such as continuous-wave and pulsed EPR, Mössbauer spectroscopy, circular dichroism (CD), and density functional theory (DFT) modeling, to build a detailed, complementary picture of how glycerol alters cysteamine dioxygenase (ADO)’s active site and substrate binding.¹

The lead authors of this study are Brad Pierce, a Professor at the University of Alabama², and Joshua Helms, a graduate student at the University of Alabama. Part 1 of our conversation covered how ADO can be a uniquely dual-function enzyme that links metabolism and oxygen sensing.³ In Part 2 of our conversation with Brad S. Pierce and Joshua R. Helms, the lead authors of this study, they discuss how all the abovementioned techniques were used in their study and the important physiological questions that their study raised.

How did the multiple spectroscopic and computational techniques work together?

Each spectroscopic technique used addresses a specific aspect of the problem at hand; only by combining each method were we able to construct a full picture of how glycerol was influencing ADO catalysis.

Continuous-wave (CW) electron paramagnetic resonance (EPR) spectroscopy gave us a "fingerprint" of the electronic environment around the iron center. By comparing the signals observed from enzyme samples prepared with and without glycerol, we could distinguish between different iron coordination states and track changes in how cysteamine was bound to the iron center. Crucially, CW EPR is a quantitative technique which allowed us to analytically determine the fraction of bidentate coordinated cysteamine relative to the total iron present within the enzyme.

Pulsed EPR, specifically the electron spin echo envelope modulation (ESEEM) methodology used here, provides a vast enhancement in spectral resolution, allowing us to observe weak dipolar interactions between the electrons on the iron center and nuclear spins on any nearby molecules of glycerol. By using a deuterium-labeled form of glycerol, we could detect its proximity to the iron site (<7 Å) with remarkable sensitivity. This observation effectively demonstrated that glycerol doesn't just change the solution environment in a general fashion, but rather that its presence in the substrate binding channel directly increases the fraction of cysteamine bound to the iron site in a bidentate fashion.

Mössbauer spectroscopy was used as an independent check to confirm that glycerol did not directly coordinate to the iron site or alter its geometry or coordination number. It also provided an essential control experiment, confirming that all iron present in the sample was present in the ADO active site and not adventitiously bound to the exterior of the enzyme.

Given the amount of glycerol used in these experiments, it is not unreasonable that the overall tertiary structure for ADO could be altered resulting in differential behavior. Far-UV CD spectroscopy reports on the secondary structural content of the protein backbone and can serve as a sensitive indicator of gross conformational changes. These experiments confirmed that glycerol did not globally alter the folded structure of ADO. This allowed us to narrow our explanation of observed ADO rate enhancement to local changes within the ADO active site rather than a global protein conformational change.

Finally, DFT computational modeling of ADO iron center bound to cysteamine allowed us to test all six possible bidentate binding configurations and rank them by energy. The lowest-energy conformation placed the cysteamine thiol group near a stabilizing tyrosine residue and its amino group near the catalytically essential aspartate residue. The computations also helped explain why certain binding orientations are unfavorable, particularly those that place negatively charged groups near the anionic aspartate residue. These same computations were also used to calculate the expected EPR spectroscopic parameters for the lowest energy conformation. A close match between calculated and observed spectroscopic parameters offers an additional degree of computational benchmarking.

What are the broader Implications for understanding non-heme iron enzymes?

Non-heme iron enzymes are found across virtually all living organisms, from bacteria to plants to humans, these enzymes carry out an impressive range of chemical reactions. By harnessing the reactivity of a single Fe(II) site and molecular oxygen, these enzymes perform transformations that are essential to life, including adding oxygen to carbon-containing molecules, breaking chemical bonds, and introducing halogen atoms into organic molecules. They are relevant to antibiotic biosynthesis, DNA repair, oxygen sensing, and many other processes critical to health and disease.

Understanding how these enzymes control their reactivity requires a closer look at the two layers of molecular architecture surrounding the mononuclear iron site. The innermost layer, which is referred to as the first coordination sphere, consists of protein derived amino acids that are directly bound to the iron atom. Surrounding this is the outer coordination sphere, a shell of nearby residues and solvent molecules that are not directly bound to the iron but exert powerful influences through weaker non-covalent interactions such as hydrogen bonding. These outer-sphere interactions help stabilize fleeting chemical intermediates, determine which substrates the enzyme will accept, and steer product outcomes throughout the reaction. Indeed, among the analogous smTDOs, a conserved hydrogen bonding network plays a well-documented role in controlling affinity for molecular oxygen and when it is allowed to bind to the iron site.

What our study reveals goes beyond what has previously been appreciated in this field. To our knowledge, this is the first time that molecular crowding within the outer coordination sphere, in this case caused by glycerol physically occupying space near the iron center, has been shown to directly shift how a substrate binds. This directly influences whether the enzyme can react with molecular oxygen. The enzyme-substrate complex exists as a mixture of two forms: one where cysteamine attaches to the iron through a single contact point (monodentate), and one where it attaches through two (bidentate). Only the bidentate form can activate molecular oxygen to initiate product formation. This kind of equilibrium mixture, where active and inactive forms coexist, may also explain why previous attempts to determine the structure of the ADO-substrate complex by X-ray crystallography were unsuccessful.

This work highlights the utility of spectroscopic techniques that are sensitive to the chemical environment of a metal center, particularly quantitative methods such as EPR and Mössbauer spectroscopy. These approaches can detect subtle differences in how a substrate is coordinated to the metal which would be invisible or ambiguous in a crystal structure. As shown in our recent study, these subtle shifts in substrate-binding conformation can be the difference between an active and an inactive enzyme. As we will discuss, these findings may carry implications that extend well beyond the laboratory.

Could similar effects occur in living cells, and what are the physiological implications?

The results obtained from controlled laboratory experiments should always be interpreted with appropriate caution before extrapolating them to a complex environment such as a living cell. With this caveat in mind, this is perhaps the most intriguing open question raised by our study.

Unlike the dilute buffer solutions used in laboratory experiments, the interior of a cell is extraordinarily crowded. Roughly 30–40% of the cell's interior volume is occupied by proteins, nucleic acids, sugars, and other biological molecules. This phenomenon, referred to as “macromolecular crowding”, has long been recognized as an important feature of cellular biochemistry, though its specific effects on individual enzymes remain poorly understood.

Our results suggest that ADO is particularly sensitive to molecular crowding effects within its substrate binding channel. This raises an intriguing question: what happens when small metabolites or other cellular components occupy the same position within the ADO active site that glycerol occupies in our experiments? In principle, this site could function as a cryptic allosteric pocket, one capable of modulating ADO activity up or down depending on which molecules happen to occupy it under a given set of cellular conditions.

One important qualification deserves mention. We observe that binding of polypeptide substrates with N-terminal cysteine residues is considerably less sensitive to added glycerol as compared to the small molecule substrate, cysteamine. This is likely because the trailing polypeptide chain itself provides an intrinsic steric crowding effect, driving a greater fraction of the substrate into the bidentate binding configuration independent of external crowding agents. Consequently, the degree to which macromolecular crowding influences ADO activity in a cellular context is likely substrate specific.

More broadly, if the equilibrium between the inactive monodentate and active bidentate enzyme-substrate complex can be shifted by molecular crowding, then fluctuations in the concentrations of small molecules, or changes in overall cellular crowding, could tune the sensitivity of ADO toward molecular oxygen. This would give the cell a surprisingly elegant mechanism to adjust how actively ADO degrades its target proteins in response to changing oxygen availability, adding a new layer of nuance to how oxygen sensing and protein stability are coordinated in mammalian biology.

The broader implications are significant. ADO sits at the intersection of sulfur metabolism and oxygen-dependent protein degradation, and fine-tuned regulation of its activity could influence taurine biosynthesis, the stability of G-protein signaling regulators, and the cell's overall adaptive response to fluctuating oxygen levels. Whether endogenous molecules that genuinely mimic glycerol's crowding effect on ADO exist and are functionally relevant in vivo is an exciting open question, one we hope future studies, both in our laboratory and others, will examine more directly moving forward.

References

1. Helms, J. R.; Probst, M.; Paris, J.; et al. EPR Spectroscopy Reveals Glycerol-dependent Activation of Cysteamine Dioxygenase (ADO) Enables Bidentate Substrate Coordination. J. Biol. Chem. 2026, 111438. DOI: 10.1016/j.jbc.2026.111438
2. University of Alabama, Brad S. Pierce. UA.edu. Available at: https://chemistry.ua.edu/people/bradley-s-pierce/ (accessed 2026-04-27).
3. Pierce, B. S.; Helms, J. R.; Wetzel, W. How Glycerol Boosts Catalytic Efficiency in Thiol Dioxygenase (ADO). Spectroscopy. Available at: [Add link when ready] (accessed 2026-04-27).