The interaction between living systems and magnetic fields has recently witnessed a renewed interest due to the importance of possible quantum processes harnessed by living systems (Lambert et al., 2013; Kim et al., 2021). For utility in biological applications, a better understanding of the quantum mechanisms at the biomolecular level is needed to direct desired outcomes in cell physiology (Usselman et al., 2014; Usselman et al., 2016; Franco-Obregón, 2023). Among the proposed mechanisms for weak magnetic field sensing in biology (Qin et al., 2016; Nordmann et al., 2017; Lin et al., 2019; Gao et al., 2021), the leading quantum process is the radical pair mechanism (RPM). Living systems are replete with forming and breaking chemical bonds, with many reactions creating radical pair (RP) intermediates. However, biological RP reactions must satisfy specific physical and chemical requirements to accomplish magnetic sensing (Timmel et al., 1998; Player and Hore, 2019). The flavoprotein cryptochrome has been proposed to be a biological magnetic receptor (Ritz et al., 2000), where a RP is initialized by either photoexcitation or during the redox cycle of the flavin cofactor (Maeda et al., 2008; Hogben et al., 2009; Hore and Mouritsen, 2016). Other protein systems have been suggested to sense weak magnetic fields (Jones, 2016). Here, we demonstrate a general method, based on product yield detected magnetic resonance (PYDMR) ¹ , to investigate the RP-based magnetic sensing in reduced flavoenzymes that produce reactive oxygen species (ROS). This method is complementary to photoexcitation measurements, such as (auto)fluorescence in RP reactions (Evans et al., 2015; Ikeya and Woodward, 2021), and provides additional information via quantitative measurements on ROS product yields.

ROS are products of oxygen-dependent life in aerobic metabolism and are generally derived from molecular oxygen (O₂) in redox active processes (Jones and Sies, 2015). The main initial ROS products in metabolism are superoxide (O₂ ^•−) and hydrogen peroxide (H₂O₂) (Schieber and Chandel, 2014), which have inherent chemical properties that coincide with their reactivity and regulation within biological pathways. Under normal physiological levels, ROS serve as oxidative signaling molecules that affect biological and physiological processes, where excessive ROS levels lead to oxidative stress (Sies, 2017). Excessive oxidative stress can result in damage to lipids, proteins, and DNA within cells and has been linked the onset of several diseases (Cross et al., 1987). Cells utilize ROS in key signal transduction mechanisms and mitochondria bioenergetics that are crucial for adaptation to a changing oxidative environment (Wood et al., 2003; Brandes et al., 2009).

Mitochondria are the major source of ROS, with topological assays that show ROS production and contributions from different metabolic sites (Brand, 2010; Dröse and Brandt, 2012). β-oxidation is a primary catabolic pathway that involves the degradation of saturated fatty acids and has been shown as a source of ROS formation (Bartlett and Eaton, 2004; Rosca et al., 2012). Electron transfer flavoenzyme (ETF) is the main electron acceptor in mammalian β-oxidation and serves as an electron funnel from at least 11 unique flavoprotein dehydrogenases and some amino acid catabolism (Roberts et al., 1996). The electrons are then transferred to the ubiquinone pool (Q-pool) via the inter-membrane bound electron flavoprotein ubiquinone oxidoreductase (ETF-QO) (Watmough and Frerman, 2010). ETF shuttles electrons by a single flavin adenine dinucleotide (FAD) cofactor. In addition to electron transfer, the ETF FAD site can serve as a secondary role for a ROS oxidative signaling terminal point, which involves the partitioning of O₂ ^•− and H₂O₂. ROS are produced through the interaction of the reduced flavin cofactor with molecular oxygen, presumably because of a disruption of electron flow to the Q-pool (Olsen et al., 2007; Burke, 2023).

The local flavin protein environment tunes the relative thermodynamic midpoint potentials for the three flavin redox states of oxidized quinone (0e⁻), radical semiquinone (1e⁻), and fully reduced hydroquinone (2e⁻) (Romero et al., 2018). Therefore, flavoenzymes produce exclusively O₂ ^•- (1e⁻) or H₂O₂ (2e⁻) or populations of both ROS, depending on the local flavin environment. For example, flavodoxins are 1e⁻ transferases because of the high flavin redox couple, and alternatively, dehydrogenases form mainly H₂O₂ due to the low flavin redox couple. For a magnetic field sensitive flavoenzyme, the redox couple must be sufficiently low, but not too high, to produce both O₂ ^•− and H₂O₂. Human ETF midpoint potentials have been determined for the Fl_hydroquinone/Fl_semiquinone (−75 mV) and for the Fl_semiquinone/Fl_quinone (+15 mV), with human ETF shown to produce both O₂ ^•− and H₂O₂ (Rodrigues and Gomes, 2012; Henriques et al., 2021). Some of the local protein environment factors the modulate redox potentials include solvent accessibility, hydrogen bonding, backbone amide dipoles, and local charge (Swanson et al., 2008; Usselman et al., 2008).

Redox active flavoproteins can undergo a proton coupled electron transfer (1e⁻) to activate O₂ to create a caged RP between the flavin semiquinone (FADH^•) and O₂ ^•− anion, Scheme 1 (Bruice, 1984; Massey, 1994; Reece et al., 2006; Chaiyen et al., 2012; Gadda, 2012; Imlay, 2013). Because the ground state of O₂ is a triplet state, FADH^•:O₂ ^•− is initially created in the triplet state. The FADH^•:O₂ ^•− presents a spin selective divergent point to release specific ROS products, where the reaction can either release O₂ ^•− through the triplet product channel or with an additional electron transfer can release H₂O₂ through the singlet channel, Figure 1.

At the RP formation, applied magnetic fields and local hyperfine interactions affect spin coherences that mix between the triplet and singlet states (Schulten and Wolynes, 1978). Therefore, internal and external magnetic fields can impact ROS products, redistributing the relative product ratios. Manipulating ROS levels using magnetic fields can potentially function as a cellular “redox switch,” which could have significant biological effects. Further understanding of quantum processes in this RP redox system could elucidate fundamental knowledge in ROS quantum biology.

To better understand the role of the RPM in ROS production at the biomolecular level in flavoenzymes, we selectively measured O₂ ^•− and H₂O₂ products with the static magnetic field artificially set to 50 µT (Earth’s magnetic field) and 20 nT static magnetic fields for recombinant human ETF. The methodology presented here can be used to study magnetic field effects in flavoproteins that are potential candidates for magnetic biosensors.

Under normal physiological conditions, ROS are oxidative signaling molecules that regulate a cellular redox network (Sies et al., 2022). Overproduction of ROS can lead to oxidative damage and a host of physiological or pathological outcomes. To better understand the phenotypic boundary between oxidative signaling or stress, biomolecular ROS quantification is essential. We demonstrate an experimental approach that can be utilized for quantitative measurement of flavoenzyme ROS generation. ETF was chosen due to its central role in bioenergetics and electron transfer pathway that feeds electrons to the mitochondria Q-pool. ETF was enzymatically reduced by MCAD and then ROS was selectively measured upon the reoxidation of the flavin cofactor. Our goals were to measure the relative proportions of ROS and the impact of magnet sensing on ROS product distributions of ETF. The reoxidation of ETF produced ROS partitioning of approximately four-fold more O₂ ^•- than H₂O₂, different than other findings of ETF ROS production (Rodrigues and Gomes, 2012). Flavin thermodynamic redox couples contribute to the observed ROS partitioning, in which the local protein environment tunes the flavoprotein redox properties (Swanson et al., 2008; Usselman et al., 2008; McDonald et al., 2011; Gran-Scheuch et al., 2023).

In addition, protein-protein interactions, such as with MCAD, induce conformational changes that can also impact midpoint potentials, analogous to points mutations in the vicinity of the flavin cofactor (Swanson et al., 2008; Usselman et al., 2008; Rodrigues and Gomes, 2012). Given that the amino acid environment in proximity to the flavin primarily determines the thermodynamic midpoint potentials, the local environment ultimately dictates normal ROS products and distributions in flavoenzyme structure-function relationships. The peptide environment around the FAD cofactor is not only crucial for redox tuning, but also serves as a flexible site for electronic coupling during electron transfer. A highly conserved arginine residue near the FAD cofactor was proposed to be responsible for stabilizing the superoxide radical pair is (Figure 1 right-bottom) (Roberts et al., 1996). Distance measurements indicate the need for a conformational movement of this residue during electron transfer to stabilize the semiquinone state and allow for ROS partitioning.

The protein molecular determinants that give rise to magnetic sensing are not well-understood and are perhaps rare in biology (Messiha et al., 2015). We proposed ETF as a potential redox magnetic sensor (Austvold, 2019), where flavin semiquinone radical and O₂ ^•− form a spin correlated RP initialized in the triplet-state, leading to characteristic magnetic field dependence on ROS product yields. Here, experimental results were conducted to compare the Earth’s static magnetic field at 50 µT and a lower static magnetic field at 20 nT. ETF reoxidation assays measured a 13% decrease in H₂O₂ production and an increase of 10% O₂ ^•− from 50 µT to 20 nT, Figure 3. The experimental ROS partitioning demonstrates one of the hallmark quantum signatures of the RPM in operation for ETF. Of critical importance is the spatial arrangement of O₂ relative to the flavin group (Chaiyen et al., 2012), as well as the required binding time of O₂ ^•− in proximity to the semiquinone for sufficient spin correlation. Recent molecular dynamics simulations discovered several novel ETF oxygen binding sites in ETF (Nielsen et al., 2019; Salerno et al., 2022), suggesting that ETF can activate O₂ through perhaps an outer sphere electron transfer. The RP distance can affect spin relaxation and thus magnetic sensing in the radicals, whereas the problems with spin relaxation can be essentially removed by a radical scavenger by the quantum Zeno effect (Kattnig, 2017).

Using the RP theory avian magnetoreception (Ritz et al., 2000), simulations have been performed to quantify ROS products that are dictated by coherent dynamics of singlet and triplet RP spin states (Procopio and Ritz, 2016). We have determined singlet and triplet product yields, and thus relative distributions of O₂ ^•− and H₂O₂ as a function of the static magnetic field strength. A realistic model was used to calculate the ROS products yields for static magnetic fields ranging from 10 nT to 100 μT. The theoretical RP results correlate with the observed ROS yields of a decrease in H₂O₂ singlet products and an increase in O₂ ^•− triplet products from 50 μT to 20 nT for the reoxidation of ETF, Figure 2. Thus, controlling ROS product channeling can be accomplished by using specific magnetic fields and configurations (Franco-Obregón, 2023).

Over the past 2 decades, cryptochrome experiments have shown increasing evidence for magnetic sensing, and more generally, the involvement of ROS (Ritz et al., 2000; Solov’yov and Schulten, 2009; Martino and Castello, 2011; Muller and Ahmad, 2011; Arthaut et al., 2017; Pooam et al., 2020). We previously reported that flavin-superoxide RP could be a broader magnetic sensing system in redox cell biology (Usselman et al., 2014; Usselman et al., 2016). Our ROS cellular research, combined with flavin-superoxide RP theoretical models, supports biomolecular ROS distributions from the results obtained through the re-oxidation of ETF. However, recently the primary magnetic receptor was suggested to be O₂ ^•− itself and perhaps O₂ ^•− dismutation, with observations supported by cellular (Martino and Castello, 2011), mouse (Carter et al., 2020) and planarian models (Van Huizen et al., 2019; Kinsey et al., 2023). If flavin-superoxide RP magnetic sensing is occurring, the discrepancy among the reports could involve magnetic field conditions that target different flavoenzymes, i.e., oxidases or monooxygenases (Massey, 1994; Imlay, 2013; Usselman et al., 2014; Gran-Scheuch et al., 2023). In addition, less is known about the initial adaptive ROS cellular responses because of the intrinsic antioxidant regulatory systems (Sies et al., 2022). Nonetheless, targeting different ROS producing systems greatly offers an expanded approach for magnetic field intervention (Vecheck et al., 2024) to remotely hack the redox code (Jones and Sies, 2015) and impart select cellular physiological responses.

Many oxidative metabolic pathways occur within the mitochondria and involve redox intermediates that can interact with O₂ to produce ROS, including ETF/ETF-QO (Watmough and Frerman, 2010; Perevoshchikova et al., 2013). Therefore, mitochondria are a vital source of ROS production within eukaryotic cells (Jones and Sies, 2015) and throughout the microbial biosphere (Imlay, 2013). ROS signaling by mitochondrial enzymes, including ETF, play a fundamental role in oxidative signaling (Sies et al., 2022), where the progression to cellular dysfunction can ultimately lead to inflammation and disease. While the effects of different magnetic field environments can alter ROS production (Barnes and Greenebaum, 2018; Gurhan et al., 2021), the persistent changes of oxidative signaling can have longer term impacts on cell physiology (Thoni et al., 2022; Franco-Obregón, 2023). Here, we show that ETF should be considered a target for further RPM investigations due to the importance of mitochondria bioenergetics, especially for biomedical engineering and therapeutic potential. Particularly, the intersection of electric voltages and magnetic spins offers a novel approach to investigate the connection between energy and living systems (Lee et al., 2023). The low magnetic fields strengths studied here also illustrate the importance of understanding how spin mechanisms could impact space health and agriculture.