Ferredoxin Reductase (FdxR, also referred to as Adrenodoxin Reductase) is a multidomain flavoprotein located in the inner mitochondrial membrane. FdxR is responsible for reducing Fdx1 (also known as Adrenodoxin), which, in turn, is the sole electron donor that supports the cytochrome P450 (CYP) function in mitochondria. Therefore, the FdxR:Fdx1 complex is a necessary component of the electron relay system that supports the function of all seven mitochondrial CYP enzymes (Fig. 1A & B) [1]. These include CYPs that are necessary for the production of steroids and corticosteroids (CYP11A1, CYP11B1, CYP11B2), vitamin A (CYP27C1), and the metabolism of vitamin D (CYP27A1, CYP27B1, CYP24A1) [[1], [2], [3], [4], [5], [6], [7], [8], [9]]. More recently, FdxR was also discovered to play a central role in the FeS biosynthetic pathway by way of its interaction and reduction of the second known mitochondrial ferredoxin, Fdx2, which in turn binds to cysteine desulfurase (Fig. 1A) [[10], [11], [12], [13], [14]]. However, while a structure of the FdxR:Fdx1 complex has been reported and the interaction has been extensively characterized [[15], [16], [17], [18]], there is no structure of the FdxR:Fdx2 complex. Despite this, there appear to be some similarities in the FdxR recognition of Fdx1 and Fdx2. A recently published NMR study by Grifagni and colleagues confirmed that the FdxR interaction with Fdx2 involves contributions from the anionic ⍺ helix-3 of Fdx2, similar to the interaction with Fdx1 [19]. However, this same study also reported that the interaction requires contributions from the C-terminal tail of Fdx2. Notably, we and others have reported that the C-terminal tail of Fdx1 is not necessary for the interaction with FdxR, but is instead required to regulate Fdx1 dimerization [20].Given its critical role in mitochondrial health, it isn't surprising that genetic variants of FdxR can result in a spectrum of mitochondriopathies [21,22], including loss of visual and auditory acuity, developmental delays, and ataxia. Peng and colleagues described a signature of FdxR-related mitochondrial dysfunction as general oxidative stress resulting from decreased oxygen consumption and increased reactive oxygen species that together cause atrophy of optic nerves [[22], [23], [24], [25], [26], [27], [28]]. This is consistent with findings from Slone et al. in 2020, who reported that FdxR mutations can result in loss of iron regulation in mitochondria, as would be expected by disruption of the FeS cluster biosynthetic pathway [29]. Specifically, such cases are likely to involve changes in the FdxR:Fdx2 redox complex. However, until recently, there had been few, if any, reports of FdxR-associated diseases that correlated with steroid or vitamin D imbalances, as would be expected from disruption of the FdxR:Fdx1 complex that may affect mitochondrial CYP function [30,31]. Pignatti and colleagues describe adrenal insufficiency and ambiguous genitalia for two siblings carrying a homozygous FdxR mutation [32]. Cells cultured from these patients confirmed imbalances in sex steroids that implicate compromised CYP11B1 and CYP11B2 function.In this study, three pathogenic variants of FdxR (R211Q, R275C and R355Q; equivalent to human variants R242W, R306C and R386W) were recombinantly expressed and purified from E. coli. These variants of FdxR in the bovine protein are representative of mutations in distinct domains of the reductase (Fig. 1B). The mutant R355Q is a spontaneous mutation documented in mice (this mutation correlates with the severely damaging R386W in humans) and is located on the FAD binding domain of FdxR [29]. The mutant R275C has been found in humans and is located on the NADPH binding domain (Fig. 1C) [28]. Lastly, the engineered mutant R211Q is designed to mimic the human mutation R242W and is found on the putative binding surface for Fdx1 and Fdx2 [28]. It should be noted here that we have kept the numbering for each of these mutations consistent with the numbering from the crystal structure of bovine FdxR (PDB ID – 1CJC) [33], and with the numbering from the mature protein that is missing the mitochondrial targeting sequence. Therefore, these mutations correlate with R242Q, R306C, and R389Q in the genetic sequence [24,29,34,35], as shown in Fig. 1C.

To capture potential impacts along both of the primary FdxR-related pathways, we investigated the effects of each variant on the FdxR:Fdx1 and the FdxR:Fdx2 complexes. Although Fdx1 and Fdx2 are structurally homologous proteins, they share only a 30 % sequence identity. We theorized that minor differences in their respective complexes with FdxR may result in differential responses to each variant.

Key findings from this study include the discovery that FdxR binds Fdx2 in a distinct manner than Fdx1, as informed by chemical cross-linking, cytochrome c assays, and 2D NMR spectroscopy. This difference explains the variant-specific effects that are observed. For example, R275C and R355Q were found to disrupt electron delivery to Fdx1 but not Fdx2, suggesting that disruption of CYP-related function is an under-reported feature of FdxR-associated mitochondriopathies. These and other findings are discussed in the context of clinical descriptions of FdxR-related disease.