Effects of Fe2+/Fe3+ Binding to Human Frataxin and Its D122Y Variant, as Revealed by Site-Directed Spin Labeling (SDSL) EPR Complemented by Fluorescence and Circular Dichroism Spectroscopies

Frataxin is a highly conserved protein whose deficiency results in the neurodegenerative disease Friederich’s ataxia. Frataxin’s actual physiological function has been debated for a long time without reaching a general agreement; however, it is commonly accepted that the protein is involved in the biosynthetic iron-sulphur cluster (ISC) machinery, and several authors have pointed out that it also participates in iron homeostasis. In this work, we use site-directed spin labeling coupled to electron paramagnetic resonance (SDSL EPR) to add new information on the effects of ferric and ferrous iron binding on the properties of human frataxin in vitro. Using SDSL EPR and relating the results to fluorescence experiments commonly performed to study iron binding to FXN, we produced evidence that ferric iron causes reversible aggregation without preferred interfaces in a concentration-dependent fashion, starting at relatively low concentrations (micromolar range), whereas ferrous iron binds without inducing aggregation. Moreover, our experiments show that the ferrous binding does not lead to changes of protein conformation. The data reported in this study reveal that the currently reported binding stoichiometries should be taken with caution. The use of a spin label resistant to reduction, as well as the comparison of the binding effect of Fe2+ in wild type and in the pathological D122Y variant of frataxin, allowed us to characterize the Fe2+ binding properties of different protein sites and highlight the effect of the D122Y substitution on the surrounding residues. We suggest that both Fe2+ and Fe3+ might play a relevant role in the context of the proposed FXN physiological functions.


Analysis of immobilized FXN
The gradual spectral changes upon Fe 3+ addition have been analyzed as combination of the two "pure" contributions shown in Figure S1 left: 1) the mobile nitroxide lineshape obtained from the A193C in the absence of Fe 3+ , red spectrum; 2) the immobilized lineshape obtained from the A193C at 100 M with a FXN:Fe 3+ 1:50 ratio, pink spectrum.
A weighted sum of the two components (previously normalized to the same number of spins by dividing each spectrum by its double integral) has been performed to reproduce the spectra at various FXN concentrations and FXN:Fe 3+ molar ratios. A full example is reported in the right panel of Figure S1.

Reversibility of Fe 3+ aggregation via chelation by EDTA
The reversibility of FXN aggregation was checked by incubating the sample with A193C at a 1:20 FXN:Fe 3+ molar ratio, taking the EPR spectrum before (red) and after (pink) incubation of the sample with a three-fold molar excess of EDTA relative to iron, as can be seen from Figure S3 left. The lineshape partially reverts to the one in the absence of Fe 3+ , showing that aggregation is at least partially reversible. As a control that EDTA does not perturb the protein, we report the spectra of the protein with/without EDTA in the absence of Fe 3+ , Figure S3 right.

M-TETPO-labelled FXN with Fe 3+
To check that the protein labelled with M-TETPO behaves like the one labelled with MTSSL, we analyzed three mutants A114C, A114C/D122Y, and A193C in the presence of Fe 3+ . As can be seen from the spectra reported in FigureS4, all mutants show a progressive immobilization upon increasing Fe 3+ additions, like their MTSSL-labelled counterparts.      Figure S9. Briefly, from the fractional saturation ( =

Fe 2+ fluorescence quenching analysis
, where F is the fluorescence at a given Fe 2+ concentration, Ff the one without Fe 2+ and Fb the one at saturating Fe 2+ concentrations) the number of binding sites on the acceptor (p) is determined at the crossing of the two blue lines as shown in the top part of Figure S9; the result is p=1 for both proteins.
Once the stoichiometry has been determined, the binding function (   Procedure to prepare M-TETPO from alcohol 1:

Synthesis of M-TETPO
Methanesulfonyl chloride (50 L, 0.65 mmol) was added to a solution of alcohol 1 (134 mg, 0.59 mmol) and NEt3 (90 L, 0.65 mmol) in CH2Cl2 (10 mL) at 0°C and the solution was stirred at ambient temperature. After 2h, the reaction was quenched with 1M HCl solution (5 mL) and extracted with CH2Cl2 (3 x 10 mL). The combined organic layers were washed with brine (10 mL), dried over MgSO4, filtered and the solvent was removed in vacuum. The residue oil was dissolved in DMF (10 mL), K2CO3 (245 mg, 1.77 mmol) and protected maleimide (195 mg, 1.18 mmol) were added. After stirring at 60 °C for 16 h, the reaction was diluted with H2O (10 mL) and extracted with EtOAc (3 x 10 mL). The combined organic layers were washed with brine (2 x 10 mL), dried over MgSO4, filtered and the solvent was removed in vacuum. The residue was purified by column chromatography (SiO2, petroleum ether-EtOAc 2:1) to yield unreacted mesylate 2 (55 mg,