The iron–thiol–oxygen nexus for iron flux from bare and ferritin-caged minerals and safeguarding DNA: the impact of the thiol structure and protein coat

Abstract

The interplay between iron, sulfur, and oxygen underpins the redox regulation of iron across biological and geochemical systems. Prior to the great oxygenation event (GOE), sulfur fostered a reducing environment essential for Fe²⁺ bioavailability. Post-GOE, the advent of the oxidative environment depleted iron-bioavailability and likely spurred the evolution of ferritin, a nanocage protein that detoxifies Fe²⁺ and catalytically synthesizes the ferrihydrite bio-mineral. Biological iron usage necessitates its reduction and mobilization from bio-minerals, where thiols can play a critical role as electron donors. This study probes the efficacy of various cellular and synthetic thiols in mediating Fe³⁺/Fe²⁺ redox-cycling, O₂ consumption, and the dissolution/mobilization of iron minerals from bare and ferritin protein-encapsulated ferrihydrites to correlate their structure–activity relationship. Furthermore, the antioxidative properties of thiols were assessed through DNA protection and radical scavenging assays. This work reports the formation of thiol-specific transient species upon interaction of thiols with Fe³⁺, which exhibit synergistic O₂ consumption, rapidly generating a hypoxic microenvironment. The thiol-mediated iron mobilization is influenced by the mineral accessibility/size (Na₂S/TG vs. GSH), O₂ consumption ability and iron chelating feature (–SH/–COO⁻vs. –NH₃⁺: TG/DHLA vs. Cys/GSH), highlighting entropic contributions (higher efficacy of dithiols over monothiol: DTT/DHLA vs. 2-ME) and restriction posed by protein encapsulation (bare vs. encapsulated ferrihydrites). Inclusion of ferritin cage-variants offers a perspective on evolutionary upgradation of the protein coat, showing how the stability of a mineral core is governed by the specific design of its inorganic–protein interface. These findings underscore the crucial role of cooperativity among iron–sulfur–oxygen interactions in cellular homeostasis, providing quintessential insights into therapeutic strategies for regulating iron metabolism and oxidative stress mitigation.

---

Introduction

Iron, sulfur, and oxygen are noteworthy among the known essential elements, and their redox interplay is crucial for sustaining life.^1–4 Iron–sulfur–oxygen chemistry has garnered significant research interest due to its involvement in vital biological processes, including electron transfer (bioenergetics), catalysis, gene regulation, and maintaining cellular redox balance.^5–7 The intricate relationship between iron and sulfur dates back to the pre-great oxygenation event (GOE) era, where Fe²⁺ was thought to be the favored oxidation state due to the reducing environment, rich in sulfur, that fostered Fe²⁺ solubility (∼10⁻² M at pH 7.0) and bioavailability.^8–12 Owing to their redox behavior, a hypoxic environment was prevalent on the early Earth, where H₂S was believed to be the primary electron source for cellular bioenergetics.^1,8 However, the sudden rise of atmospheric O₂ due to oxygenic photosynthesis by cyanobacteria (as a consequence of the GOE)^13,14 led to the generation of insoluble Fe³⁺ (solubility ∼10⁻¹⁸ M at pH 7.0), which drastically reduced its bioavailability to meet cellular requirements (from µM to mM).¹⁵

As an adaptive response, organisms orchestrated several mechanisms to acquire iron: namely, (1) reduction of Fe³⁺ to Fe²⁺ (reductive pathway),^16–19 (2) direct complexation, which enhances iron solubility (siderophore-based),^20,21 (3) acidification (enhances solubility/favors dissolution),^22–26 or (4) encapsulation to increase its bioavailability.^23,27 Nonetheless, reduction of Fe³⁺ leads to the generation of Fe²⁺, a Fenton substrate, which if free and excess can complicate the life processes by producing lethal reactive oxygen species (ROS) such as hydroxyl radicals (OH˙) in the presence of H₂O₂.^3,28–30 Similar to iron, molecular O₂ (the product of the GOE) is critical for aerobic organisms as it participates as the terminal electron acceptor during ATP synthesis (oxidative phosphorylation) in the mitochondrial electron transport chain, a primary site of ROS production.^8,31,32 This presents a dual challenge for both iron and O₂: indispensable yet paradoxically dangerous, their redox properties can lead to cellular toxicity if not tightly regulated. To address the challenges posed by the redox chemistry of iron in the advent of increased O₂ levels, organisms evolved sophisticated machinery, such as ferritin, a self-assembled protein nanocage that buffers free iron levels.³³ This evolutionary protein rapidly scavenges excess toxic Fe²⁺ and functions as a primary reserve, ensuring that the organism has a reliable source of iron when needed.^33–35 The ferritin protein cage holds and safeguards iron in a soluble ferrihydrite form while also regulating its release with remarkable precision, minimizing iron precipitation and toxicity.^33,36–40

Similar to iron, sulfur is one of the most abundant elements since the pre-GOE era, but needs to be incorporated into organic frameworks by the biosynthetic process, for utilization in various enzymatic/life processes. While sulfur and oxygen share a family, sulfur's larger atomic size (∼100 pm), high polarizability (3.298 × 10⁻⁴⁰ F m²), lower electronegativity (2.58), and diverse oxidation states grant it unique advantages,⁶ critical to various physiological processes. A sulfur-containing amino acid, methionine (Met), is essential for cellular functions, serving as a precursor in the transsulfuration pathway,⁴¹ which regulates sulfur metabolism and redox cycling.⁴² This pathway enables the conversion of Met into cysteine (Cys, E_m,7 = −0.22 V), a semi-essential amino acid that serves as a precursor for biosynthesis of glutathione (GSH, E_m,7 = −0.24 V), the primary antioxidant that neutralizes reactive oxygen/nitrogen species (ROS/RNS).^43,44 The GSH level may shoot up to ∼10 mM, particularly at the ROS generation site (mitochondria), to protect the organelle from oxidative stress.^45,46 The high concentrations of GSH are also reported in the cytosol, which contributes to its reducing environment where most of the ferritins are present.^44,47,48 Both Cys and GSH act as substrates for the biosynthesis of hydrogen sulfide (H₂S), a smaller antioxidant that also functions as a signaling molecule while facilitating electron transfer and detoxification of free radicals.^49,50 At neutral pH (∼7.0), H₂S predominantly exists as HS⁻ (pK_a ∼ 7.0) with cellular concentration varying from nM to µM.^51,52 Dihydrolipoic acid (DHLA), a natural thiol formed by the reduction of alpha-lipoic acid (ALA), is noteworthy for its ability to cross the blood–brain barrier and neutralize ROS/RNS, accounting for its medical relevance, often marketed as a cellular activator to reduce oxidative stress.⁵³ Additionally, some thiol moieties are involved in oxidative defense through thiol-specific proteins like peroxiredoxins, GSH-peroxidase (regulates peroxide levels) and Cys rich metallothioneins for detoxification of heavy metal ions.^6,54 A fraction of sulfide in the brain (∼160 µM) is reported to modulate iron flux from ferritin.⁵¹ This indicates the potential of thiol-based compounds, not only to relieve oxidative stress but also to mobilize iron from its natural source. This may have implications in neuroferritinopathy, a pathophysiological condition characterized by iron accumulation in the brain in the form of iron-oxide nanoparticles.^55,56 Moreover, iron minerals, including ferritin encapsulated ferrihydrite, are utilized for the synthesis of various iron-containing enzymes/proteins, including iron–sulfur clusters (using desulfurase enzymes, which catalytically derive sulfur from Cys).^57,58

Similarly, the synthetic thiols such as thioglycolic acid (TG), dithiothreitol (DTT), and 2-mercaptoethanol (2-ME) as reductants play a significant role in numerous chemical–biochemical studies.^37,59,60 For instance, TG is used for a limit test for iron,⁶¹ a qualitative assay specifically developed for alkaline conditions, whereas DTT and 2-ME are regularly used in proteomic analysis to minimize disulfide formation.⁶⁰

Despite the significance of these natural and synthetic thiols,⁶² comprehensive research comparing their interaction with oxidants such as Fe³⁺/O₂ and their ability to act as reductants (mineral dissolution from both bare and ferritin encapsulated ferrihydrites) and as antioxidants (relieving oxidative stress) is limited. Herein, this report evaluates iron–thiol interactions and Fe³⁺ reduction by a set of thiols (Fig. 1A), using stopped-flow rapid kinetics, and iron–thiol–O₂ interaction by amperometry. Moreover, this study aims to link these dynamics to the iron release trend from bare and ferritin encapsulated iron minerals under normoxic conditions to analyze the structure–reactivity relationship of thiols. This work also elucidates the comparative effectiveness of the protein cage and the cage type in retaining the iron mineral core, by including two ferritin variants: Mycobacterium tuberculosis BfrB and amphibian (frog M) ferritin, versus the bare ferrihydrite mineral. The incorporation of the bare ferrihydrite mineral serves a dual purpose: (1) to establish the role of the ferritin protein cage as a physical barrier to reductive iron dissolution and (2) to provide insight into natural iron acquisition strategies, where microbes and plants secrete thiols alongside chelators to ensure iron dissolution/acquisition from natural iron mineral reservoirs. Eukaryotic bullfrog M ferritin shares approximately 70% sequence identity with human H ferritin²³ and BfrB (a non-heme binding ferritin) serves as a primary iron repository in Mtb.⁶³ The rationale of using two different ferritin cages may further provide clues on the entry pathways of reductants/exit pathways of iron and the stability/reactivity of the encapsulated iron mineral (Fig. 1B). Lastly, the anti-oxidative ability of these thiols was evaluated using DNA cleavage and radical scavenging assays.

Fig. 1 Natural and synthetic thiols employed as electron donors (A) and bare and ferritin encapsulated ferrihydrite (Fh) minerals, Fe³⁺ solution, dissolved O₂ and free radicals (DPPH) employed as electron acceptors (B) are utilized in this study to analyze the intricate relationship between iron, thiol and O₂ and their impact on reductive iron dissolution/mobilization. The mid-point potentials of the thiols listed in panel A correspond to pH 7.0, represented as E_m,7.

The current report demonstrates that the thiol molecular architecture (i.e., the size, extent of branching, and functional groups, –NH₃⁺, –COO⁻, and –SH) significantly influenced electron transfer efficiency during O₂/Fe³⁺ reduction and mineral dissolution, revealing the superior efficacy of Na₂S. Moreover, thiol-mediated Fe³⁺ reduction led to synergistic behaviour in O₂ consumption, generating a hypoxic microenvironment. Additionally, the results established the superior reducing capabilities of dithiols over monothiols. Notably, bare ferrihydrite exhibited a greater release of iron compared to its ferritin-encapsulated counterpart, emphasizing the importance of the protein cage and reductant accessibility to the mineral core. Overall, these findings will advance our understanding of the interplay between iron, thiol, and O₂ towards iron homeostasis and iron chelation/mobilization during iron overload and cellular redox balance.

Experimental section

Materials

L-Cysteine (Cys), 2-mercapto ethanol (2-ME), Tris freebase, agarose, and ammonium sulfate were sourced from HiMedia. L-Glutathione reduced (GSH), thioacetic acid (TAA), sodium thioglycolate (TG), cysteamine hydrochloride (Cyst.), anhydrous FeCl₃, ferrozine (Fz), 4-4′-dithiopyridine (4-DPS), 3-(N-morpholino)propane sulfonic acid sodium salt (MOPS-NaCl) and 2,2-diphenyl-1,1-picryl hydrazyl (DPPH) were procured from Sigma-Aldrich. Dithiothreitol (DTT) and sodium sulfide (Na₂S) flakes extrapure were obtained from SRL and dihydrolipoic acid (DHLA) was obtained from TCI. Gadolinium acetate tetrahydrate and TEM grids were sourced from TED PELLA.

Thiol quantification by 4-DPS assay

The amount of reduced forms of sulfur-based reducing agents (thiol, –SH) was estimated using the 4-DPS assay, prior to using them in any experiments.^64,65 Initially, the 4-DPS assay was standardized using different concentrations of cysteine by monitoring the formation of thiopyridone at 324 nm using a UV-visible (Shimadzu UV-1900) spectrophotometer. Similarly, 100 µM 4-DPS was added to thiols of different concentrations (10, 20, and 30 µM) in 100 mM Tris buffer (pH = 7.0), and the absorbance at 324 nm was recorded. The concentration of the reduced form of thiol was determined using the estimated molar extinction value obtained from the standard curve (ε = 20 mM⁻¹ cm⁻¹).

Synthesis of recombinant frog M and Mtb BfrB ferritin proteins

The recombinant Mycobacterium tuberculosis (Mtb) bacterioferritin B (BfrB) and frog M ferritin were overexpressed in Escherichia coli strains, BL21 (λDE3) and BL21 (DE3) pLysS, respectively, and purified using earlier reports.^59,63,66 Briefly, ferritin protein purification was carried out using heat treatment (only for frog M ferritin), ammonium sulfate precipitation, and anion exchange chromatography. To assess the structural integrity and purity of the proteins, both native-PAGE and SDS-PAGE were employed.^23,37,67 The purified Mtb BfrB and frog M ferritin were concentrated, stored in a 100 mM MOPS-NaCl, pH = 7.0 buffer, and their concentrations were quantified by Bradford assay.

Synthesis of bare and ferritin-encapsulated ferrihydrite minerals

The bare ferrihydrite nanoparticle was synthesized using a “bottom up” approach as per earlier reports.⁶⁸ Briefly, concentrated NaOH was added to a freshly prepared 50 mM FeCl₃ (anhydrous) solution with continuous stirring to achieve a pH of 7.0. The resulting precipitate containing ferrihydrite nanoparticles was washed, dried, and stored in the dark to avoid any photochemical reactions.

The ferritin encapsulated ferrihydrite (mineralization; ∼480 Fe atoms/cage) was prepared by using purified protein samples following previously established protocols.^37,69 In short, a freshly prepared Fe²⁺ solution (FeSO₄ in 1 mM HCl) was added to a 2.08 µM ferritin protein cage in 100 mM MOPS-NaCl buffer (pH = 7.0). The iron-loaded ferritins were incubated at 25 °C for 2 hours, followed by additional incubation at 4 °C for 12 hours.¹⁶ The mineralized ferritin samples are denoted as ‘encapsulated Fh’ throughout this report.

Characterization of bare and ferritin-encapsulated ferrihydrite minerals

The ferritin-encapsulated ferrihydrite was analyzed using TEM, as per earlier reports.^37,67,69 Briefly, after mineralization (∼480 Fe atoms/cage), 5 µL of the ferritin sample (1 mg mL⁻¹) was drop-cast onto a carbon-coated copper grid for obtaining unstained TEM images, and for negative staining of the ferritin protein coat a 2% (w/v) solution of gadolinium acetate tetrahydrate was used to rinse the TEM grid. After proper drying of the grid, the samples were analyzed using a transmission electron microscope (FEI Tecnai G2 TF30-ST) with a LaB₆ electron gun operating at 300 keV, and distribution profiles of both the cage and the mineral core were obtained using the ImageJ software. The iron content in the mineralized ferritins (Fe atoms/cage) was determined using the ferrozine assay, as per previous reports^33,37 and was found to be 440 ± 20 Fe atoms/cage.

The synthesized bare ferrihydrite mineral was characterized by transmission electron microscopy (TEM), field emission scanning electron microscopy (FESEM), Fourier transform infrared (FT-IR) spectroscopy, and powder X-ray diffraction (P-XRD). TEM analysis of bare Fh was done by drop casting the samples onto a copper grid, after sonicating the nanoparticles in ethanol, and the samples were analyzed similarly to the ferritin encapsulated Fh.

Kinetics of dissolved O₂ consumption by thiols and thiol–iron mixtures

An oxygraph from Hansatech Instruments, having a Clark-type microelectrode, was employed to monitor the dissolved O₂ consumption kinetics. This electrode works under the amperometric principle involving the measurement of current produced upon the reduction of molecular oxygen as a function of time.^16,69,70 The current response of the oximeter was calibrated using aerated water and sodium dithionite, prior to the kinetic experiments.³³

The kinetics of dissolved O₂ consumption are studied in three segments: (1) control experiments (with buffer, mineralized ferritin, bare Fh, Fe³⁺, Fe²⁺, and Zn²⁺ salt), (2) by only thiols, and (3) by thiol–iron mixtures (bare/encapsulated Fh and Fe³⁺). The dissolved O₂ consumption by control reactions and by only thiols was performed by injecting respective reactants/thiols using a Hamilton syringe into the reaction chamber containing 100 mM MOPS-NaCl (pH = 7.0). For O₂ consumption by thiol–iron mixtures, experiments were performed by first injecting thiols into the sample chamber containing buffer (after 2 min of data acquisition), followed by the addition of iron (Fe³⁺/bare/encapsulated Fh) using a Hamilton syringe. The average rate of O₂ consumption by control, only thiols, and thiol–iron mixtures was calculated and compared, from 30 minutes of data acquisition. Fe³⁺/bare Fh were prepared in 1 mM HCl, and encapsulated Fh (∼480 Fe atoms/cage) was prepared in 100 mM MOPS-NaCl (pH = 7.0).

Reductive iron dissolution from bare and encapsulated ferrihydrite minerals

For reductive iron dissolution from bare Fh, 1 mg of the sample was added to a solution of 100 mM MOPS-NaCl (pH = 7.0) and ferrozine (1 mM). To initiate the iron release, 2.5 mM thiols were added, and the final reaction volume was 1 mL. The samples were kept on the shaker (shaking speed = 40–50 rpm) and centrifuged (3500 rpm) at a different time point, to collect 10 µL of the supernatant, which was diluted to 200 µL (20X dilution). The absorbance at 562 nm was recorded, and the concentration of liberated iron, i.e., the Fe(II)–ferrozine complex, [Fe(Fz)₃]⁴⁻, was estimated using the reported molar extinction coefficient (ε = 25.4 mM⁻¹ cm⁻¹).^16,17

Similarly, the reductive iron release/dissolution kinetics from encapsulated Fh (∼480 Fe atoms/cage) was tracked by monitoring absorbance at 562 nm for the formation of the [Fe(Fz)₃]⁴⁻ complex, by using a UV-visible spectrophotometer (Shimadzu UV-1900) following reported protocols.^69,71 In brief, the kinetic experiment was initiated by the addition of 2.5 mM thiols to a buffer solution (100 mM MOPS-NaCl, pH = 7.0) containing an ∼0.21 µM mineralized ferritin cage (∼100 µM of iron) and 1 mM ferrozine at 25 °C.

Fe³⁺ reduction by sulfur-based reducing agents: stopped-flow rapid kinetic analysis

To assess the reducing capability of various sulfur-based reducing agents, the reduction of Fe³⁺ was monitored using a rapid mixing stopped-flow system. The reduction kinetics of Fe³⁺ was investigated by mixing equal volumes of freshly prepared Fe³⁺ (100 µM FeCl₃ prepared in 1 mM HCl) and a solution mixture comprising thiol (2.5 mM) and ferrozine (1 mM) in 10 mM MOPS-NaCl buffer (pH = 7.0) using a stopped-flow rapid mixing unit (Hi-Tech SFA-20) integrated with an Agilent Cary-3500 UV-VIS spectrophotometer. Following rapid mixing, both the spectral kinetics and the time-course of formation of the [Fe(Fz)₃]⁴⁻ complex were monitored at 562 nm to estimate the extent of Fe³⁺ reduction by thiols. The Fe³⁺ reduction kinetic parameters were obtained from the plot of ΔA_{562 nm}vs. time for the formation of [Fe(Fz)₃]⁴⁻ by non-linear fitting with the double exponential equation.

A_(562)t = A₁·e^k₁t + A₂·e^k₂t + A_offset

Here, A_t represents the absorbance values measured at 562 nm at different time points. The parameters A₁, A₂, and k₁, k₂ correspond to the amplitudes for the rapid and slow phases and observed rate constants of the iron reduction process, respectively. The term A_offset denotes the offset value at equilibrium.³³

Iron–thiol interaction studies using stopped-flow rapid kinetics

The study of kinetics of thiol–Fe³⁺ interaction/complexation was done using a stopped-flow rapid mixing unit, as mentioned above, except that the ferrozine was not added. Reactant solutions: Fe³⁺ (100 µM) and thiol (2.5 mM) were mixed, and the complexation kinetics of thiol–Fe³⁺ interaction were monitored by recording spectra at different time points. Further time courses (ΔA vs. time) for the formation of corresponding thiol–Fe³⁺ complexes were recorded by monitoring the rate of change in absorbance at their peak maxima (λ_max), which was different for each thiol.

DNA protection assay

The effectiveness of thiol molecules towards inhibition of DNA damage was assessed in vitro, to evaluate their antioxidative effect against OH˙ generated by Fenton substrates (Fe²⁺/H₂O₂), following earlier reports.³³ Briefly, the plasmid DNA (isolated plasmid, about 4361 base pairs, ∼100 ng) was incubated separately with different thiols (2.5 mM, 15 min), followed by the addition of freshly prepared FeSO₄ (100 µM) and incubation for 10 min. For initiating Fenton's reaction (OH˙ formation), 4.5 mM H₂O₂ was added to the reaction mixture and was further incubated at room temperature for 15 minutes. To assess the effect of thiols in averting oxidative cleavage of DNA, electrophoresis was performed using 0.8% (w/v) agarose gel stained with ethidium bromide (EtBr), at 70 V for 1 hour in 1X TAE buffer. Finally, the DNA bands were visualized using a ChemiDoc MP imaging system (Bio-Rad).

Free radical scavenging assay

The antioxidative ability of thiols was assessed using the DPPH free radical scavenging (FRS) assay, as per earlier reports.⁷² In brief, stock solutions of DPPH and thiols were prepared in methanol and 10 mM MOPS-NaCl buffer (pH = 7.0), respectively, prior to the assay. The assay was performed by mixing DPPH (60 µM, ∼990 µL) with thiols (30 and 60 µM, ∼10 µL) and then incubating at room temperature, in the dark, for about 30 minutes. Control reaction samples, i.e., only DPPH (∼60 µM) and DPPH with ascorbate (AscH⁻), were also incubated separately to account for maximum and minimum absorbance changes for the DPPH radical. The spectra of the resulting solution were taken, and absorbance at 517 nm (ε = 12.3 mM–1 cm⁻¹) was noted for each thiol using a UV-vis spectrophotometer. The extent of DPPH reduction by thiols (radical scavenging ability) was calculated using the following equation:
Here, A_max is the absorbance of the DPPH radical, A_obs is the absorbance of the DPPH radical after mixing with thiols, and A_min is the absorbance obtained by mixing AscH⁻ with DPPH solution to quench the radical.

Results and discussion

Sulfur, with its versatile redox chemistry spanning oxidation states from −2 to +6, serves as a linchpin in biological/cellular redox networks. Assimilated primarily through sulfur-containing amino acids (cysteine and methionine), it forms bioactive molecules like glutathione (GSH) and hydrogen sulfide (H₂S) and is part of Fe–S cluster biogenesis, which is associated with antioxidant defence (GSH-peroxidase), electron transfer, and metal ion buffering (e.g., metallothioneins). In the pre-GOE era, sulfur-enriched reducing conditions ensured the bioavailability of soluble Fe²⁺, obviating the biological iron storage. Post-GOE, atmospheric oxygenation elevated oxidative stress and oxidized Fe²⁺ to insoluble Fe³⁺, driving mineral precipitation and iron scarcity, likely spurring the evolution of ferritin protein. The ferritins play a major role in cellular detoxification and redox maintenance for which thiols also contribute an important part. However, the investigation of their interaction and structural impact (the influence of structure–reactivity) on bare iron and iron mineral (bare and encapsulated ferrihydrite) under aerobic conditions are still a miss and are evaluated here through various strategic experiments.

Thiol quantification and determination of apparent redox potential

To understand the structure–activity relationship of thiol-based reducing agents, a series of experiments were designed and performed with identical concentrations of their reduced form. As thiols are prone to oxidation under aerobic conditions, the thiol quantification assay (4-DPS assay) was performed prior to the other experiments to determine the amount of reduced form (thiol, –SH) of sulfur-based reducing agents in their stock solution.^64,65 This also aided in calculating/estimating the apparent redox potential (E_app). This assay utilizes 4,4′-dipyridyl disulfide as a substrate, which reacts with –SH, thereby forming 4-thiopyridone (with a stoichiometry of 1 : 1), and absorbs at 324 nm. The molar absorptivity value (ε) was calculated from the standard curve (ε = 20 010 M⁻¹ cm⁻¹ at 324 nm, Fig. S1A, SI), which matches closely with the reported value of 21 400 M⁻¹ cm⁻¹.^64,65 The estimated molar absorptivity value was used to calculate the reduced form of thiol by implementing the Beer–Lambert law (Fig. S1B, SI). The percentage of reduced forms of thiols prior to the experiments (freshly prepared) was found to be between 89 and 98% (Table 1). By using the amount of reduced and oxidized species, the apparent redox potential was calculated from the balanced redox equation (see redox species in Fig. S2, SI) using the Nernst equation (Table 1) at pH 7.0. The total change in apparent redox potential (E_app) was negative, 27–50 mV from the reported mid-point potential (E_m,7) values.

Table 1 Reported pK_a values and apparent reduction potentials (E_app) of sulfur-based reducing agents at pH 7.0. E_app obtained from thiol quantification assay and the appropriate Nernst equation

Thiols	Reported pKa of –SH group	Reported Em,7 (mV)	f Red	f Ox		E app (mV)
Note: (1) The fractions of the reduced forms (-SH) were estimated by 4-DPS assay in pH 7.0 buffer. (2). The apparent redox potentials (Eapp) of first three reducing agents (Na2S/H2S, DHLA and DTT) were obtained using the Nernst equation (from balanced redox reaction) of the form: . Similarly, the Eapp values for other reducing agents (Na2S/H2S, TG, GSH, 2-ME, Cyst., and Cys) were obtained using the Nernst equation (from balanced redox reaction) of the form: . (3) All the thiols undergo 2e^− transfer, and as ΔpH is zero (in buffer), the H^+ term is omitted from the Eapp calculation. (4) Balanced equations for each thiol molecules can be found in Fig. S2, SI.
Na2S	7^6	+170^45	0.89	0.11	8.10	+143
DHLA	9.2, 11.4^74	−320^74	0.91	0.10	9.10	−349
DTT	9.2, 10.1^75	−327^75	0.95	0.06	15.83	−363

---

Thiols		Reported Em,7 (mV)	f Red	f Ox		E app (mV)
Na2S	7^6	−230^6	0.89	0.11	7.21	−272
GSH	8.7^6	−240^6	0.95	0.05	18.05	−277
TAA	3.4	—	0.95	0.05	18.05	—
Cys	8.4^6	−220^6	0.97	0.03	31.36	−264
TG	9.3^76	−140	0.92	0.08	10.58	−170
2-ME	9.6^77	−196^77	0.92	0.08	10.58	−226
Cyst.	8.4^77	−203^77	0.98	0.02	48.02	−253

---

Characterization of bare and protein-encapsulated ferrihydrite minerals

The synthesized bare ferrihydrite mineral was characterized by various techniques. The powder-X-ray diffraction (P-XRD) pattern of bare ferrihydrite showed two characteristic broad peaks for “d” values around ∼2.5 Å and 1.5 Å, which is a typical identification of two-line ferrihydrite.^68,73 The FT-IR pattern displayed major peaks for Fe–O stretching vibration (620 cm⁻¹) and –OH stretching (3400 cm⁻¹) (Fig. S3, SI) of bare ferrihydrite. The TEM analysis showed a fringe pattern and crystallinity in its SAED pattern. The d-spacing value for bare ferrihydrite, obtained from TEM (∼0.25 nm), was found to be comparable with that obtained from P-XRD (∼0.25 nm) and matched with the earlier reports (Fig. 2A).^68,73 The TEM analysis also provided the size range of 6–20 nm for freshly prepared bare ferrihydrite nanoparticles having a cluster like appearance, further supported by FESEM analysis (Fig. S3C and D, SI).

Fig. 2 TEM analysis of bare/ferritin encapsulated ferrihydrite. TEM images of (A) bare ferrihydrite and (B) ferritin encapsulated ferrihydrite (unstained and negatively stained ferritin coat of mineralized samples, ∼480 Fe atoms/cage) with their respective SAED pattern. The TEM images of frog M ferritin are presented here for protein-encapsulated ferrihydrite.

For ferritin protein-encapsulated ferrihydrite, TEM analysis was performed for both stained (negative staining of the protein coat using 2% gadolinium acetate tetrahydrate) and unstained mineralized ferritin samples. Unstained images showed distinct small-sized (∼4 nm) iron minerals, whereas stained images show core–shell architectures for the encapsulated iron mineral inside the nanocavity of the self-assembled ferritin protein cage. The SAED pattern further established the crystallinity of the ferritin mineral core (Fig. 2B).

The thiol–iron combination synergistically consumes dissolved O₂

The reactivity of different thiols towards O₂ and iron may be different, due to differences in their structure, and is not investigated in detail. To study the relative efficiency and preference of thiols towards scavenging molecular O₂, dissolved O₂ consumption kinetics were investigated, both in the presence and in the absence of iron (Fig. 3). The importance of such customization is to comprehend the favored electron transfer route i.e., from thiols to dissolved O₂vs. iron (oxidants) and may provide insights towards understanding the differential efficiency of thiols in iron release kinetics from bare and ferritin encapsulated ferrihydrite (discussed in the section on iron release).

Fig. 3 Consumption of dissolved O₂ by thiols and thiol–iron mixtures. The kinetic experiments were performed in 100 mM MOPS-NaCl buffer (pH = 7.0) with 2.5 mM thiol and 100 µM iron (Fe²⁺/Fe³⁺/bare/encapsulated iron mineral). The arrow indicates the time point of injection of the respective reactants. Kinetic traces of dissolved O₂ consumption for (A) control reactions: Fe²⁺ (auto-oxidation), Fe³⁺ and Fe³⁺ minerals (bare/encapsulated Fh) and Zn²⁺, (B) only thiols and (C) and (D) thiol–iron combinations; only representative time courses with Na₂S and Cys are presented here (see the details for all the thiols in Fig. S4, SI). (E) Comparison of the average rate of dissolved O₂ consumption (determined from B–D and Fig. S4) for thiols exhibiting faster kinetics (category I). The experiments were repeated thrice, and the averages of the data sets are presented here along with the error bar (standard deviation).

The control experiments (in the absence of thiols) were performed to evaluate the rate of O₂ consumption by buffer, Fe³⁺, Fe²⁺, Zn²⁺, bare ferrihydrite (Fh) and ferritin (encapsulated Fh) and it was found to be extremely slow (buffer – 0.08 µM min⁻¹, Fe³⁺ and Zn²⁺ similar to buffer – 0.05 µM min⁻¹, and bare/encapsulated Fh – 0.7–1 µM min⁻¹), except for Fe²⁺ (Fig. 3A). The substantial dissolved O₂ consumption (2.1 µM min⁻¹) was observed only for auto-oxidation of Fe²⁺ in a pH 7.0 buffer.

Among nine thiols subjected to dissolved O₂ consumption, in the absence of iron, only Na₂S/HS⁻ and DHLA were found to be efficient in creating the oxygen-free reaction mixture in a span of ∼20 min (Fig. 3B). Based on two-electron reduction potential (E_m,7 (HS₂⁻, H⁺/2HS⁻) = −230 mV), it will be difficult to explain the differential reactivity of thiols towards dissolved O₂. For instance, Na₂S/HS⁻ consumed O₂ at a higher rate in comparison to GSH, Cys, and 2-ME, despite having a similar two-electron reduction potential (E_m,7 ∼ −220 ± 20 mV). Moreover, positive reduction potential (E_m,7 (S⁰, H⁺/HS⁻) = +0.17 V) for Na₂S/HS⁻ is also reported,⁴⁵ which suggests that it is a weaker reducing agent than Cys and GSH. Also the one-electron reduction of O₂ by HS⁻ is thermodynamically not favoured (ΔE_m,7 = −1.26 V: E_m,7 (O₂/O₂˙⁻) = −0.35 V and E_m,7 (S˙⁻, H⁺/HS⁻) = +0.91 V);⁶ however, the radical intermediates (S˙⁻ and SO₂˙⁻) react with O₂ rapidly at a diffusion controlled rate (k ∼ 10⁹ M⁻¹ s⁻¹), producing species that kinetically drive the equilibrium in the forward direction.⁶ Also, for DHLA, the formation of an entropically favored five-membered intramolecular ring, (Fig. S2, SI) after oxidation possibly drives the O₂ consumption kinetics.

Similarly, dissolved O₂ consumption kinetics were investigated by adding iron (Fe³⁺, bare Fh, and ferritin encapsulated Fh) to the thiol solution i.e., thiol–iron mixtures. For some thiols, the dissolved O₂ kinetics significantly enhanced in the presence of iron, creating a hypoxic reaction chamber, exhibiting synergistic behavior. Based on the O₂ consumption rate exhibited by the thiol–iron mixture, thiol molecules are segregated into two categories, i.e., (I) thiols consuming O₂ at a faster rate (Na₂S/HS⁻, DHLA, TG, Cys, and DTT) and (II) thiols consuming a minimal amount of O₂ (2-ME, Cyst., GSH and TAA) (Fig. S4, SI).

Among category I, for DHLA, DTT, and Cys, the rate of O₂ consumption was colossal in the presence of Fe³⁺, followed by bare Fh, and then mineralized ferritin. Such a trend may explain that these thiols can easily access Fe³⁺ and bare Fh in comparison to their encapsulated counterparts (Fig. S4, SI). However, for Na₂S/TG, faster O₂ consumption kinetics were observed for encapsulated Fh. In this case, Na₂S completely outplays the accessibility issue towards the caged mineral core, owing to the smaller size, thereby promptly generating Fe²⁺. Owing to the inherent catalytic nature of ferritin (due to the presence of the ferroxidase (F_ox) center) the generated Fe²⁺ rapidly get oxidized, kick-starting the redox cycling process (oxidation/reduction) until the entire chamber became anaerobic (Fig. 3C). In the case of TG, a similar explanation may be used for faster O₂ consumption observed in the presence of mineralized ferritin owing to its smaller size. The above observation suggests that protein encapsulation not only modulates/tunes the stability/reactivity of the mineral core but also serves as a barrier for reductants.

In category II, the O₂ consumption kinetics were slow, almost similar to the buffer (see Fig. S4, SI), which may be linked to lower reducing capability (TAA), bigger size, and branching (GSH), which limit iron mineral accessibility and thus overall O₂ consumption efficacy. 2-ME, despite having a smaller size and structural similarity to DTT, consumed a lower amount of O₂ even in the presence of Fe³⁺, which can be linked to its inferior reducing nature, due to the lower entropic change (see Scheme 1, further discussed in the later section).

Scheme 1 Entropic facilitation of dithiols over monothiols for O₂ consumption. (n) denotes the number of molecules/moles of reactants and products.

Reductive iron mobilisation from bare and ferritin caged ferrihydrite minerals: the impact of protein encapsulation and thiol concentration/structure

The reductive dissolution of iron minerals is one of the most efficient methods to clean the rust and one of the iron acquisition strategies used by plants/microbes.^16,17,68 Biological iron utilization requires reduction of Fe³⁺ to Fe²⁺,^5,47 a process driven by electron transfer, and can be facilitated by thiol-based agents.^61,62 The role of these thiols as reducing agents is of particular interest as these can flux iron out not only from bare iron minerals (for meeting the iron requirement by plants and microbes), but also from a ubiquitous, highly conserved intracellular iron storage protein, ferritin. The ferritin cage sequesters free Fe²⁺ and concentrates it in the form of ferrihydrite bio-mineral inside the ∼8 nm protein cage/cavity, rendering it inert yet accessible upon physiological demand.^78,79 However, the 2 nm thick protein coat may restrict/regulate the reductant access owing to the narrow pore, which would selectively allow only smaller reductants/thiols to traverse through the protein coat to reach the iron mineral core. Thus, the efficiency of the thiol molecules to dissolve and mobilize iron may not truly reflect their reducing capability. Therefore, to analyze the effect of protein encapsulation on reductive iron mobilization and to compare the reducing ability of thiols, bare ferrihydrite was utilized. In addition, iron dissolution studies were carried out with two types of ferritin, i.e., amphibian ferritin (frog M) and bacterial ferritin (Mtb BfrB-non-heme binding ferritin), to understand the impact of the cage type in stabilizing/safeguarding the iron mineral against the reducing environment.

(i) Impact of the protein coat and its type

The iron mobilization kinetics from ferrihydrite mineral via thiol-based reductants were investigated, and the released iron was tracked by monitoring the formation of the [Fe(Fz)₃]⁴⁻ complex (λ_max = 562 nm, ε =25.4 mM⁻¹ cm⁻¹). The results demonstrate that iron dissolution from bare Fh exceeds that of protein-encapsulated Fh (by ∼3 fold, depending on the thiol employed), owing to the absence of 2 nm protein barrier, (keeps easy access of reductants at bay). This hierarchy of iron release from bare Fh mirrors the inherent reducing capacity of each thiol in the absence of macromolecular confinement, which rules out the limiting role of ferritin pores towards access of thiols to the protein caged iron mineral core.

Among all thiols, Na₂S/HS⁻ indiscriminately released the highest amount of iron from both encapsulated and bare ferrihydrite (Fig. 4 and 5). Although iron release was easier/faster for the bare ferrihydrite mineral, the trend and the difference/fold of increment (the ratio of iron release by each thiol with respect to the lowest) were different. The unrestricted access of thiols to iron in the bare ferrihydrite mineral exhibited a distinct pattern in reductive iron efflux (Na₂S > TG ∼ DTT > DHLA > Cys > TAA > Cyst. > GSH > 2-ME; (Fig. 4C) from the trend displayed by encapsulated ferrihydrite (Na₂S > TG > DHLA ≫ DTT > Cys > Cyst. > 2-ME > GSH > TAA) (Fig. 4A, B and 5). The factors like size, branching, delocalization of electron cloud, the number of –SH groups per molecule, iron chelation ability, and the presence of –COO⁻/–NH₃⁺ possibly dictate the overall iron release profile. For instance, the branched thiols like DTT are relatively less efficient during iron release from encapsulated ferrihydrite, when compared with bare ferrihydrite (Fig. 4C and 5).

Fig. 4 Impact of thiol structure, protein cage, and cage type on reductive iron mobilization. The iron dissolution kinetics were compared between frog M ferritin (A), Mtb BfrB (B), and bare ferrihydrite (C). Here, 1 mg of bare Fh mineral and 0.2 µM ferritin (containing 100 µM caged mineral ∼500 Fe atoms/cage) were separately mixed with 2.5 mM reducing agent and 1 mM ferrozine, in 100 mM MOPS-NaCl (pH = 7.0). The percentage of iron release was calculated after 2 hours of reaction, and the fold of increment (the ratio of iron release by each thiol with respect to the lowest-TAA). Unlike for the encapsulated mineral, a discontinuous assay was carried out for bare Fh to monitor iron mobilization, and the final absorbance was reported by multiplying the dilution factor (× 20). (D) Superior efficiency of dithiols over monothiols towards mineral dissolution and iron mobilization. The reaction conditions were the same as mentioned above, except that the concentrations of –SH groups were maintained constant (2.5 mM monothiol and 1.25 mM dithiols).

Fig. 5 Impact of thiol concentration and the ferritin cage type on reductive iron mobilization from encapsulated ferrihydrite. Kinetic traces of reductive iron mobilization from mineralized frog M (A) and Mtb BfrB (B) with variation in thiol concentration. The iron dissolution reaction from 0.2 µM ferritin (containing 100 µM caged mineral ∼500 Fe atoms/cage) was initiated by mixing 2.5 mM/1 mM/0.25 mM reducing agent separately along with 1 mM ferrozine, in 100 mM MOPS-NaCl buffer (pH = 7.0).

Iron release kinetics also revealed that the type of protein shell can influence the process by acting as a redox gatekeeper, i.e., in comparison to frog M ferritin, the Mtb BfrB ferritin exhibits 2-fold greater iron release (except for HS⁻) (Fig. 4A, B and 5). This observation suggested that either the mineral core or the protein cage of frog M ferritin is more stable and less susceptible to iron mineral dissolution. However, for Na₂S/HS⁻ the type of protein coat had no impact, as it indiscriminately released ∼80–90% of stored iron from both the ferritin cages within ∼5 min and reached saturation. The influence of the type of ferritin coat was clearly observed in the case of TG (the second-highest iron-releasing thiol), where the amount of released iron was drastically different (∼25% and ∼60% from frog M ferritin and Mtb BfrB, respectively) (Fig. 4A and B). This divergence reflects intrinsic differences in protein cage design. The interconnected structural variations in the protein cage possibly underpin this phenomenon, i.e., (1) electrostatic gating at the pore/channels, (2) long-range electron transfer through redox-active residues in the protein matrix, and (3) protein-induced stability of the mineral core and their redox properties.

Electrostatic surface analysis suggests that both ferritins have negatively charged residues lining the 3-fold pores/channels (mostly glutamates and aspartates), which may inhibit the access of anionic reductants (Fig. S5). However, the 4-fold pores containing mixed residues may possibly facilitate partial reductant access (Fig. S5). Despite these minor structural differences, electrostatic exclusion appears to be a substantial barrier to direct reductant entry, which may dictate the reactivity of thiols. Furthermore, long-range electron transfer through redox-active amino acids cannot be ruled out, especially for larger thiols. Paradoxically, frog M ferritin, despite housing more redox-active residues (8 Tyr and 1 Trp for frog M vs. 4 Tyr and 1 Trp for BfrB),^23,80 exhibited slower iron release. This inverse relationship implies that residue spatial orientation and electronic coupling in frog M ferritin sub-optimally support ET, suggesting a structurally more resistant shell to reductive iron core solubilization. Furthermore, if protein-induced constraints elevate the reorganization energy (λ) (i.e., by restricting solvent reorganization) or shift the effective reduction potential (E°′) of the ferrihydrite mineral core, electron transfer kinetics become slow despite favourable ΔG.^32,81 This “redox gating” mechanism ensures regulated iron release, preventing uncontrolled Fenton chemistry. However, exceptions exist, i.e., H₂S/HS⁻ outsmarting these protein barriers, showing exceptional performance even in restrictive cages like frog M.

(ii) Impact of thiol concentration

The physiological concentration of these thiols varies from µM (H₂S in the brain) to mM (GSH in the cytosol). Carrying out iron mobilization experiments with lower concentrations may also help in differentiating the reactivity/efficiency of these thiols towards the dissolution of iron minerals. So, the iron mobilization kinetics from ferritin encapsulated ferrihydrite via thiol-based reductants were studied at three different concentrations, i.e., low (250 µM/0.25 mM), medium (1 mM), and high (2.5 mM) (Fig. 5) to understand the impact of thiol concentration on iron mobilization.

The iron release from intact ferritin nanocages was the highest for Na₂S/HS⁻, both for frog M and BfrB, which get enhanced with increment in thiol concentration (Fig. 5). This revealed that an even lower amount of Na₂S/HS⁻ (0.25 mM) was sufficient to flux out ∼50% of the iron irrespective of the type of ferritin cage, which further increased to ∼80–90% by increasing the thiol concentration to 10 fold (Fig. 5). The kinetic profile indicated a distinct initial iron release rate, i.e., ∼50 µM min⁻¹ for 2.5 mM and reached saturation in ∼5 min, whereas at 0.25 mM concentration of Na₂S/HS⁻ the rate was slower (∼3 µM min⁻¹) and reached saturation after ∼50 min.

Similar to Na₂S/HS⁻, increasing the concentration of the rest of the thiols also increased the rate as well as the amount of released iron (Fig. 5). This concentration-dependent iron release result is possibly linked with the dissolved O₂ consumption kinetics. By increasing the thiol concentration (the dissolved O₂ concentration decreases, and higher iron release is observed under anaerobic conditions, Fig. S6, SI), the competitive pathway is minimized, which may lead to unidirectional electron flow to the iron mineral for the reduction and dissolution processes. Moreover, this mineral dissolution process is a surface phenomenon, where fraction saturation of the bare mineral with thiols can influence the iron release kinetics. Therefore, a higher concentration of thiol can be expected to saturate a greater fraction of the iron mineral, leading to an increased iron release, as observed in Fig. 5. This explanation may hold true at least for smaller thiols, even in the case of encapsulated ferrihydrite.

(iii) Structural influence of thiol-based reducing agents

To comprehend the impact of thiol structures on their reactivity, this set of thiol molecules were employed as reductants during the reductive iron mobilization from both bare and encapsulated ferrihydrite minerals. The iron release trend from encapsulated ferrihydrite for all thiols is Na₂S > TG > DHLA ≫ DTT > Cys > Cyst. > 2-ME > GSH > TAA, revealing the supremacy of Na₂S in releasing iron from encapsulated ferrihydrite. The smaller size of Na₂S/HS⁻ may allow efficient mobility through the ferritin protein shell (irrespective of the protein cage/coat) and facile access to the iron core, thereby enhancing the reduction of ferric iron (Fe³⁺). Among organic thiols, the second highest reductive iron dissolution was shown by TG (pK_a = 9.3, Table 1), despite its relatively higher reduction potential (E_m,7 = −140 mV). Its high efficacy may be attributed to its relatively smaller size and the presence of iron chelating functionalities (–COO⁻/–SH group) that give rise to synergistic (–SH/–COO⁻) cooperation towards Fe³⁺ chelation followed by reduction to promote iron release. Upon coordination with Fe³⁺, the drop in the pK_a value of the –SH group is expected, leading to generation of thiolate (–S⁻), a stronger reducing moiety that may explain faster iron release by TG. When compared between Cys and TG, having a comparable size and –the SH/–COO⁻ group, the amount of iron released by Cys was much lower than that released by TG. Moreover, unlike TG, Cys, having a lower reduction potential (E_m,7 = −220 mV), released less iron, likely due to the presence of the –NH₃⁺ group, which may have affected Fe³⁺ complexation and thus reduction, thereby inhibiting the overall iron release process.

The third place for iron release is taken by DHLA; this molecule contains two adjacent –SH groups, which upon oxidation form a stable five membered ring (Scheme 2). While comparing iron release based on the number of –SH groups, both DHLA and DTT having two reducing moieties (–SH) may chelate and enhance the iron mobilization (Fig. 5). The better efficacy of DHLA may be ascribed to the presence of two reducing moieties (–SH), iron chelating –SH/–COO⁻ groups and the hydrophobic chain with less branching. These structural attributes might provide an easy passage through the protein coat to the core, thereby facilitating chelation cum reduction of the iron mineral.

Scheme 2 Entropic facilitation of dithiols over monothiols for Fe³⁺ reduction. (n) denotes the number of molecules/moles of reactants and products.

An astounding difference in iron release was observed between mono- and dithiols, despite having similar structural moieties (Fig. 1 and 4). Dithiols like DHLA and DTT released significantly higher amounts of iron in comparison to the monothiol (2-ME) at equal (2.5 mM) concentration (Fig. 4 and 5). Furthermore, to comprehend the superiority of molecular features of dithiol, the iron release experiments were also carried out at half concentration, i.e., 1.25 mM DTT and DHLA and compared with 2.5 mM 2-ME (Fig. 4D). The results (Fig. 4D) clearly demonstrated the superior reducing capabilities of dithiols even at half-concentration. This highlights the role of entropic factors and the possible formation of a stable oxidized product, i.e., intramolecular disulfide rings upon oxidation (Scheme 2), a configuration that thermodynamically drives Fe³⁺ reduction and synergistically consumes O₂ (Fig. 3 and Fig. S4, SI), which explains their redox behavior towards O₂/iron. Therefore, dithiols (DTT/DHLA) outperform monothiols (2-ME) even at stoichiometrically equivalent concentrations (Fig. 4D), providing compelling evidence that entropic optimization via intramolecular cyclization dramatically accelerates the reaction kinetics of O₂ consumption followed by Fe³⁺ reduction or vice versa, thereby promoting iron mobilization (Scheme 2).

GSH, despite having lower redox potentials (E_m,7 = −240 mV), showed minimal Fe³⁺ reduction/release activity, emphasizing the necessity of both mineral accessibility and iron chelation efficiency. The presence of –NH₃⁺ and branching possibly inhibits the entry of GSH through the ferritin pores, resulting in lower iron release (Fig. 4 and 5). At neutral pH, TAA exists as CH₃–COS⁻ (pK_a = 3.4), but the electron density on sulfur probably gets diminished (by the carbonyl group), which may affect mineral dissolution and iron release efficacy (Fig. 4 and 5). These findings underscore the importance of the size, molecular architecture (the presence of functional groups –COO⁻ and –NH₃⁺), and number of –SH groups in modulating thiol-mediated iron mobilization from protein-bound mineral cores.

Overall, this study dissects the impact of sulfur-based reductants (spanning inorganic H₂S/HS⁻ to organic thiols) on mobilization of iron from both bare and ferritin-encapsulated iron mineral cores, revealing how the molecular architecture of these thiols and protein cage stability orchestrate electron transfer efficiency towards O₂ and iron mineral. Thus, the efficacy of thiol-based reducing agents facilitating iron mobilization to a different extent stems from several interrelated factors, such as molecular determinants, entropic gains, and the presence of a protein cage (and cage stability), which are summarized in Scheme 3.

Scheme 3 The molecular determinants of thiols and other factors that play a key role in dictating the iron release from bare and protein-encapsulated ferrihydrite minerals.

Kinetics of Fe³⁺ reduction is thiol specific and correlates with reductive Fe²⁺ release from bare ferrihydrite

The reduction kinetics of FeCl₃ by the selected set of thiols were monitored using a stopped-flow rapid mixing system integrated with a UV-vis spectrophotometer. This Fe³⁺ reduction kinetics would help in evaluating the structure–activity correlation among thiols. Fe³⁺ reduction kinetics can be compared with iron release kinetics (bare vs. encapsulated ferrihydrite) to explain the influence of protein cages and pores in dictating the accessibility of thiols to the mineral core. Therefore, the mixture of thiol with ferrozine (Fe²⁺ chelator) was rapidly mixed with the Fe³⁺ solution (FeCl₃), and the reduction kinetics were monitored by tracking the formation of the [Fe(Fz)₃]⁴⁻ complex at 562 nm (Fig. 6 and Fig. S7).

Fig. 6 Stopped-flow rapid kinetics of Fe³⁺ reduction by thiols in the presence of ferrozine. The Fe³⁺ reduction kinetics of thiols was performed by mixing equal volumes (1 : 1) of freshly prepared FeCl₃ solution (100 µM in 1 mM HCl) with the thiol–Fz mixture (2.5 mM thiol and 1 mM Fz in 10 mM MOPS-NaCl, pH = 7.0) using a rapid mixing unit. (A) and (B) The spectral kinetics for Fe³⁺ reduction by representative thiols, i.e., Na₂S/HS⁻ and 2-ME, respectively, indicated the formation of the [Fe(Fz)₃]⁴⁻ complex at 562 nm. Time courses for Fe³⁺ reduction for (C) a longer and (D) a shorter time window, and (E) nonlinear fitting with a double exponential equation (see all thiols in Fig. S8, SI). (F) Time courses of Fe³⁺ reduction by Na₂S/HS⁻ (from D) were compared with Fe²⁺–Fz complexation.

Similar to iron release and O₂ consumption kinetics (Fig. 3 and 4), iron reduction kinetics were faster for Na₂S/HS⁻, and the overall trend for the thiols was found to be in this order: Na₂S > TG ∼ DTT > Cys > DHLA > TAA > Cyst. > GSH > 2-ME (Fig. 6C). This chronological order is observed from ΔA (Table 2) after 5 min of reaction and is relatively similar to the iron mobilization trend from bare ferrihydrite (Fig. 5C). This indicates that the presence of a 2 nm ferritin protein coat serves as a barrier for interaction of reductants with the mineral core and is responsible for the different trends observed (except for Na₂S/TG) (Fig. 4 and 5). Moreover, the observed rate constants for Fe³⁺ reduction by different thiols were obtained and are compared in Table 2.

Table 2 The observed rate constants for Fe³⁺ reduction by thiols; obtained by fitting the kinetic profile of: ΔA_562nmvs. time (s) (shown in Fig. 6E and Fig. S8, SI) using a double exponential equation

Thiols	Rapid phase k1 (s^−1)	Slow phase k2 (s^−1)	ΔA562nm (after 5 min)
Note: the kinetic parameters were obtained from three independent data sets.
Sodium sulfide (Na2S)	24.5 ± 1.1	1.19 ± 0.36	0.12 ± 0.013
Dihydrolipoic acid (DHLA)	19.6 ± 4.1	0.45 ± 0.11	0.04 ± 0.007
Cysteamine hydrochloride (Cyst.)	17.9 ± 1.9	0.95 ± 0.25	0.03 ± 0.002
Dithiothreitol (DTT)	12.9 ± 1.8	0.32 ± 0.07	0.05 ± 0.008
Sodium thioglycolate (TG)	9.8 ± 3.1	0.23 ± 0.06	0.06 ± 0.005
L-Cysteine (Cys)	9.2 ± 1.9	0.11 ± 0.02	0.05 ± 0.006
Thio acetic acid (TAA)	6.9 ± 1.4	0.81 ± 0.05	0.02 ± 0.004
2-Mercapto ethanol (2-ME)	1.4 ± 0.6	0.79 ± 0.02	0.006 ± 0.001
Glutathione (GSH)	1.7 ± 0.3	0.31 ± 0.08	0.005 ± 0.001

---

For iron reduction, possibly either the Fe³⁺–thiol interaction or Fe³⁺–Fz complexation occurs as the first step, followed by electron transfer (Scheme 4). Fe²⁺–Fz complexation is much faster than the Fe³⁺ reduction by the most efficient thiol (Na₂S/HS⁻) (Fig. 6F), suggesting that the electron transfer from thiol to Fe³⁺ may be the rate-limiting step. Depending on the type of thiol, iron samples (Fe³⁺/bare Fh/encapsulated Fh) and the sequence of events (Scheme 4), the electron transfer (ET) from thiol to iron may occur either via an outer-sphere (long range ET) or an inner-sphere ET mechanism. This double exponential behavior/nature of the iron reduction kinetic profile (Fig. 6E, Table 2 and Fig. S8, SI) probably indicates the occurrence of different mechanisms (Scheme 4).

Scheme 4 Proposed mechanism for Fe³⁺ reduction by thiols. This scheme highlights the possibility of both the inner/outer sphere electron transfer processes. As the complexation of Fe²⁺ by ferrozine is very fast, the electron transfer process may dictate the kinetics of Fe³⁺ reduction.

Among all thiols, Na₂S/HS⁻ demonstrated the most efficient O₂ and iron-reducing capability, which is true for the reduction of all types of iron (Fe³⁺, bare Fh, and ferritin caged Fh) used in this study. Its two 2-electron reduction potentials [E_m,7 (S⁰, H⁺/HS⁻) = +0.17 V, E_m,7 (HS₂⁻, H⁺/2HS⁻) = −230 mV] cannot explain Na₂S/HS⁻ as a stronger reductant/antioxidant. The superiority of HS⁻ stems from its unique redox chemistry. While organic thiols form disulfides (RSSR) upon oxidation, HS⁻ is known to undergo one-electron oxidation despite unfavorable redox potential [E°′ (S˙⁻, H⁺/HS⁻) = +0.91 V], generating inorganic radicals (HS˙) that can rapidly dimerize (k = 9 × 10⁹ M⁻¹ s⁻¹) to H₂S₂, further reported to decompose to elemental sulfur and polysulfides (S_n).⁶ Removal of the products formed by one-electron transfer pathways kinetically drives the equilibrium in the forward direction, i.e., reduction of Fe³⁺ and molecular O₂via the following equations:

HS⁻ + O₂ → S˙⁻ + HO₂˙ (i)

HS⁻ + Fe³⁺ → S˙⁻ + Fe²⁺ + H⁺ (k > 10⁸ M⁻¹ s⁻¹) (ii)

S˙⁻ + HS˙ → HS₂⁻ (k = 9 × 10⁹ M⁻¹ s⁻¹) (iii)

HS₂⁻ + H⁺ → 1/nS_n(s) + H₂S (g) (iv)

Consequently, HS⁻ achieves unparalleled O₂ and Fe³⁺ reduction kinetics (the observed rate constants for TG, k₁ = 9.8 ± 3.1 s⁻¹ and for HS⁻, k₁ = 24.5 ± 1.1 s⁻¹, 2.5 times faster than TG), underscoring the possibility of direct electron transfer. This type of reaction may also be expected to occur in cellular compartments such as mitochondria, where this radical-driven autoxidation cascade of HS⁻ may mobilize Fe²⁺ from ferritin mineral for iron–sulfur cluster biogenesis.^82,83

Fe³⁺–thiol interaction: stopped flow rapid kinetic analysis revealed the formation of thiol-specific transient species

The thiol–Fe³⁺ interaction/complexation study was done using a stopped-flow rapid mixing unit, as mentioned above, except that the ferrozine was not added. The rationale behind this customization is to understand the sequence of events (chelation/reduction) during the iron reduction phenomena, which would further escalate our understanding of the reductive iron mobilization from ferrihydrite bio-minerals. The UV-vis absorption profile of only thiols in the buffer did not change even after 30 min of incubation (see Fig. S9, SI). However, when Fe³⁺ was rapidly mixed with the set of thiols, a transient spectral signature was observed only for three thiols (TG, Cys, and DTT) (Fig. 7A–C and Fig. S10, SI). The peak maxima (λ_max) for the corresponding Fe³⁺–thiol complexes were found to be at 535 nm, 500 nm, and 491 nm for TG, DTT, and Cys, respectively (Fig. 7A–C). Based on the λ_max, further single-wavelength kinetic experiments were carried out, which revealed the differential stability of Fe³⁺–thiol transient species, exhibiting the following trend: TG > Cys > DTT (Fig. 7D–I).

Fig. 7 Kinetic analysis of Fe³⁺–thiol interactions by stopped-flow rapid mixing. The Fe³⁺–thiol interaction kinetics was studied by mixing equal volumes (1 : 1) of freshly prepared Fe³⁺ (200 µM in 1 mM HCl) with 5 mM of respective thiols (in 100 mM MOPS-NaCl, pH = 7.0 buffer) to maintain the Fe³⁺–thiol ratio at 1 : 25. Spectral kinetics (A)–(C). The progress curves of Fe³⁺–thiol interaction of selective thiols for shorter (D)–(F) and longer (G)–(I) time windows revealed the differential stability of transient species.

The higher reduction efficiency of TG may be attributed to the formation of a purple coloured, transient, Fe³⁺–thiolate charge-transfer complex, which decayed within 40 ms, whereas Cys–Fe³⁺ kinetics were relatively slow (see Fig. 7). The accumulation of these unknown transient species depends on the relative rates of their formation and decay, which may be linked to the structure and redox behavior of thiols and may be related to the iron release trend. Interestingly, for Na₂S, formation of black colored colloidal solution was observed, which impeded further kinetic analysis.

Different thiols safeguard DNA differently against the Fe²⁺/H₂O₂ based Fenton reaction

Lethality of OH˙, generated by Fenton's reaction (Fe²⁺ and H₂O₂), is well established/documented.^33,46,84 Due to its high oxidative (E_m,7 ∼ +2.3 V) nature, it has damaging effects on DNA, protein, membranes, and various cellular organelles.⁸⁵ As a counterbalance to this oxidative stress and to protect cellular integrity, living organisms synthesize/secrete antioxidants/antioxidative proteins, such as thiols/thiol-based proteins (e.g., H₂S, GSH/GSH-peroxidase, peroxiredoxins, metallothioneins, etc.).^6,84 Therefore, DNA protection assay was performed to analyze whether all these thiols can effectively protect DNA against ROS equally or have differential/relative efficacy.

The oxidative damage inflicted by OH˙, generated by Fe²⁺/H₂O₂, on plasmid DNA was well evidenced in lane 4 of Fig. 8A, with the complete disappearance of the DNA band. The control experiments executed in the presence of thiol and in the absence of either H₂O₂ or Fe²⁺ revealed that the DNA band intensity was unaltered. However, in the absence of H₂O₂ (in lane 3, Fig. 8A), Fe²⁺ affected DNA migration, highlighting the possible consequences of electrostatic interactions (charge neutralization or conformational changes) between negatively charged DNA and Fe²⁺. Among the nine thiols, Na₂S/HS⁻ and Cyst. exhibited complete DNA protection (lane 9 in panel A for Na₂S and lane 4 and 9 for Na₂S/HS⁻ and Cyst. respectively in panel B in Fig. 8), whereas thiols such as Cys and 2-ME provided moderate amounts of protection (lane 7 and 12, respectively in panel B, Fig. 8). Surprisingly, DHLA, DTT, TG, GSH and TAA were ineffective against oxidative damage of DNA. This finding elucidates that protection of DNA against Fenton's reaction is thiol specific. The inefficiency of thiols like DTT and DHLA, despite containing two –SH groups and exhibiting higher O₂ consumption/iron release, in protecting the DNA against the damaging effects of OH˙ also poses a question. The PC₅₀ values for thiols offering maximum DNA protection, i.e., Na₂S/HS⁻ and Cyst., were found to be ∼1.04 and ∼0.22 mM respectively (Fig. 8C and Fig. S11, SI), which suggests that the lower PC₅₀ value of Cyst. accounts for the higher potency of the thiol for safeguarding DNA. The relative efficiency contributing to the mechanism of DNA protection possessed by certain thiols (specificity) is mystifying at this juncture.

Fig. 8 DNA protection and radical scavenging abilities of thiols. Comparison of the DNA protection (antioxidative) activity of sulfur-based reducing agents is performed using agarose gel (0.8% w/v) electrophoresis. The “±” signs indicate the “with/without” addition of respective constituents. DNA cleavage activity is assessed for: (A) control reactions and Na₂S, (B) all thiols, in the presence of Fenton's substrates (Fe²⁺ and H₂O₂). DNA protection was observed only for Na₂S, Cys, Cyst., and 2-ME (lane 9 in panel A and lanes 4, 7, 9, and 12 respectively in panel B). Lane 1 corresponds to the ladder of known sizes. (C) The concentration dependent DNA protection activity values of Na₂S and Cyst. respectively (see gel images in Fig. S11, SI). The thiol samples were incubated in the dark for 30 minutes to have a complete reaction with DPPH. (D) The image depicts the amount of DPPH reduction by 30 and 60 µM thiols. The corresponding spectra of DPPH + thiol samples after incubation (E) and the % FRS (free radical scavenging assay) were calculated by using the control reactions (i.e., initial absorbance –DPPH only and complete reduction of DPPH by sodium ascorbate) (F).

As an antioxidative feature, thiols differentially protect DNA. However, this dichotomy in DNA protection arises possibly from the coaction of the following factors: i.e., (1) Fe²⁺ chelation/complexation (decreasing the availability of free Fe²⁺), (2) electrostatic compatibility (binding interaction of specific thiols with DNA), (3) kinetic competency (competitive quenching of H₂O₂/OH˙ by thiols), (4) steric/chemical constraints, and (5) formation of Fe²⁺–thiol–DNA ternary complexes. The Fe²⁺–thiol–DNA ternary complexes may sterically shield DNA from the OH˙ radical. These small thiols may react directly with H₂O₂ or rapidly scavenge OH˙ (k ∼ 10¹⁰ M⁻¹ s⁻¹;⁸⁴ competing with DNA damage (k ∼ 10⁹ M⁻¹ s⁻¹).⁸⁴ Specific thiol–iron interactions may modulate the redox cycling of Fe²⁺/Fe³⁺ by chelation, ultimately suppressing localized OH˙ formation and preserving the DNA. Cyst. and Cys having the –NH₃⁺ group enable electrostatic association with the DNA phosphate backbone (negatively charged) and may position the thiol moiety to intercept OH˙ generated in proximity. Under steric/chemical constraints, the bigger and branched structure (GSH and DHLA) impedes DNA access, while TAA contains an electron-withdrawing C=O carboxyl group that reduces thiol reactivity. The bulky thiols (GSH and DHLA) fail completely despite intrinsic antioxidant capacity. These results validate that effective DNA protection requires optimized radical scavenging kinetics coupled with steric and charge compatibility for DNA proximity. This insight explains the inefficacy of abundant cellular thiols like GSH, despite their general antioxidant role, and fundamentally resolves why nature relies on enzymatic detoxification (e.g., peroxiredoxins and GPx) utilizing these thiols as substrates to prevent OH˙ formation, rather than solely on direct scavenging.

Antioxidant activity of thiols determined by free radical scavenging assay

DPPH (2,2-diphenyl-1-picrylhydrazyl) is a stable free radical with a deep violet color, which undergoes a color change to pale yellow upon reduction by an antioxidant/reducing agent. This distinct colorimetric shift allows for easy spectrophotometric measurement, which enables DPPH to be used widely for assessing antioxidant activity/radical scavenging activity. The inclusion of this assay is significant for evaluating thiols, as the thiol-based compounds serve as vital components of the oxidative defense system in biological machineries. Thiols, with their sulfhydryl (–SH) groups, play a crucial role in neutralizing reactive oxygen species (ROS) and maintaining redox homeostasis. Assessing the ability of thiols to quench DPPH radicals will provide insight into their potential antioxidant activity and structural influence on radical interactions.

Among nine thiols tested at 30 µM, radical scavenging efficiency varied significantly (Fig. 8D). DHLA demonstrated the highest activity (62% neutralization), while glutathione (GSH) showed the lowest (∼14%). Increasing the concentrations of all thiols to 60 µM showed that 2-mercaptoethanol (2-ME) and thioacetic acid (TAA) quenched near-complete DPPH reduction (>95%), whereas DHLA, dithiothreitol (DTT), and cysteamine (Cyst.) achieved ≥75% quenching (Fig. 8D and F). TG, cysteine (Cys), and Na₂S/HS⁻ reduced ∼60–70% of radicals at 60 µM concentration (Fig. 8F). Notably, GSH remained ineffective even at 60 µM (<20% neutralization; Fig. 8D–F). The thiols such as 2-ME and TAA achieve near-complete radical reduction at elevated concentrations (60 µM), whereas the superior efficacy of DHLA and DTT may be attributed to their dithiol (–SH) groups, which possibly enhance their electron-donating capacity for radical scavenging even at 30 µM (Fig. 8D and F). Moderate radical quenching observed for TG (thioglycolic acid), Cys (cysteine), and Na₂S aligns with their lower reducing power relative to the DPPH radical (E_1/2 +320 mV in methanol), where the thermodynamic driving force (ΔG°) for reduction is diminished. Conversely, GSH (glutathione) demonstrates markedly low activity (<20% even at 60 µM), primarily due to steric constraints imposed by its branched backbone (Fig. 8D–F). Collectively, these results underscore that DPPH scavenging kinetics are dictated by thiol reactivity, molecular topology (e.g., denticity and steric effects), and electron-donating capacity, highlighting how structural features fine-tune antioxidant function. This trend was found to be completely different from the reactivity of thiols with other oxidants (O₂ and Fe³⁺), which possibly suggests the involvement of different mechanisms.

Summary

By linking sulfur-rich pre-GOE and oxygen-rich post-GOE chemistries to Fe redox evolution, the study provides a unique mechanistic continuum from abiotic iron–sulfur chemistry to biomineral regulation, extending inorganic redox principles. The interplay between iron, sulfur, and oxygen is pivotal for the cellular redox balance. This study defines how the molecular architecture of thiols dictates their efficiency as reductants during reduction of Fe³⁺/iron mineral towards iron mobilization and O₂ consumption and as antioxidants. We demonstrate that their efficacy is governed by a confluence of factors beyond redox potential, including the molecular size, chelation ability, and type/number of functional groups. For instance, dithiols (DTT/DHLA) exceed monothiols (2-ME) even at stoichiometrically equivalent concentrations via entropic advantages that significantly accelerate O₂ consumption and Fe³⁺ reduction kinetics, thereby promoting iron mobilization from both bare and caged ferrihydrites. Similarly, the presence of –COO⁻ in TG and –NH₃⁺ in Cys/GSH modulates the iron flux, revealing their opposing effects.

This study uses the ferrihydrite core of ferritin as a well-defined, biologically relevant inorganic mineral. Our direct comparison of iron mobilization from the bare mineral vs. the protein-encapsulated mineral provides fundamental insights into how a protein nanocage itself acts as a critical regulator, thereby altering the mineral reactivity and stability of an inorganic solid, under spatial constraints, akin to host–guest inorganic systems. The bare ferrihydrite releases substantially more iron than its encapsulated counterpart, establishing that the physical access of reductants to the mineral core can surpass the long-range electron transfers via the protein cage. By including different ferritin cage variants, our work offers a perspective on evolution, i.e., iron mobilization is protein-cage specific, controlled by cage stability (the bacterial ferritin core was more susceptible to the reductive iron mobilization than the amphibian ferritin core), suggesting an evolutionary refinement in the stability of the ferritin coat.

Notably, Na₂S/HS⁻ was found to be uniquely potent, rapidly mobilizing iron from both ferritin nanocages and bare ferrihydrite, effectively nullifying the protein cage barrier, while exhibiting superior DNA protection and O₂ consumption. In conclusion, the overall findings underscore the importance of sulfur–iron interactions in mitigating oxidative stress. This work highlights the diverse role of thiols in regulating iron metabolism and offers valuable insights for the potential development of therapeutic agents aimed at enhancing iron reduction and mobilization across various physiological and pathophysiological conditions (for iron overload)⁸⁶ and for catalysis/nano-biotechnology (for preparing apo-ferritin nanocages).^87,88

Author contributions

Rabindra Kumar Behera and Narmada Behera conceptualized this work. Tanaya Subudhi performed the experiments and drafted the manuscript. Rabindra Kumar Behera and Tanaya Subudhi revised the document. All authors have read, discussed, and agreed to the final version of the manuscript.

Conflicts of interest

The authors declare no conflict of interest.

Abbreviations

GOE	Great oxygenation event
ROS	Reactive oxygen species
RNS	Reactive nitrogen species
Mtb	Mycobacterium tuberculosis
BfrB	Bacterioferritin B
Fh	Ferrihydrite
MOPS	3-(N-Morpholino)propane sulfonic acid
TG	Sodium thioglycolate
GSH	Reduced L-glutathione
Na2S	Sodium sulfide
2-ME	2-Mercapto ethanol
DHLA	Dihydrolipoic acid
DTT	Dithiothreitol
Cys	L-Cysteine
Cyst.	Cysteamine hydrochloride
TAA	Thioacetic acid
4-DPS	4-4′-Dithiopyridine
DPPH	1,1-Diphenyl-2-picrylhydrazyl
MLCT	Metal to ligand charge transfer
ET	Electron transfer
E 1/2	Midpoint potential
E m,7	Midpoint potential at pH 7.0
PC	Protection concentration
Tyr	Tyrosine
Trp	Tryptophan
FRS	Free radical scavenging
Fz	Ferrozine
Fox	Ferroxidase center
TEM	Transmission electron microscopy
PAGE	Poly-acrylamide gel electrophoresis

Data availability

The data supporting this article have been included as part of the supplementary information (SI). The supplementary information includes thiol estimation by 4-DPS assay, balanced redox equations for thiols, characterization of bare ferrihydrite, impact of iron on dissolved O₂ consumption, pore electrostatics of ferritin variants, reductive iron release under anaerobic conditions, stopped-flow rapid kinetics of iron thiol interactions and iron reduction by thiols alongside agarose gel images of control experiments and PC₅₀ determination. Supplementary information is available. See DOI: https://doi.org/10.1039/d5tb02365c.

Acknowledgements

R. K. B. would like to thank the Science and Engineering Research Board (SERB), India (CRG/2020/005332) and the Science and Technology Department, Odisha, India (ST-SCSTMISC-0036-2023) for the research grants. We are thankful to Dr Elizabeth C. Theil (C.H.O.R.I., USA), Dr Anil K. Tyagi and Dr Garima Khare (University of Delhi, South Campus) for their generous support in providing the ferritin clones. The authors are also thankful to Dr Akankshika Parida for experimental guidance and incessant support.

References

1. Biological Inorganic Chemistry: Structure and Reactivity, ed. H. Gray, E. I. Stiefel, J. S. Valentine and I. Bertini, University Science Books, 2007.
2. R. Crichton, Iron Metabolism, John Wiley & Sons, Ltd, 2009, pp. 17–58.
3. Z. Lu and J. A. Imlay, Nat. Rev. Microbiol., 2021, 19, 774–785.
4. R. Lill, Nature, 2009, 460, 831–838.
5. E. C. Theil, T. Tosha and R. K. Behera, Acc. Chem. Res., 2016, 49, 784–791.
6. M. R. Filipovic, J. Zivanovic, B. Alvarez and R. Banerjee, Chem. Rev., 2018, 118, 1253–1337.
7. J. L. Casey, M. W. Hentze, D. M. Koeller, S. W. Caughman, T. A. Rouault, R. D. Klausner and J. B. Harford, Science, 1988, 240, 924–928.
8. B. Alberts, A. Johnson, J. Lewis, M. Raff, K. Roberts and P. Walter, Molecular Biology of the Cell, Garland Science, 4th edn, 2002.
9. A. Bekker, H. D. Holland, P. L. Wang, D. Rumble, 3rd, H. J. Stein, J. L. Hannah, L. L. Coetzee and N. J. Beukes, Nature, 2004, 427, 117–120.
10. P. G. Falkowski and Y. Isozaki, Science, 2008, 322, 540–542.
11. L. R. Kump, Nature, 2008, 451, 277–278.
12. A. D. Sheftel, A. B. Mason and P. Ponka, Biochim. Biophys. Acta, Gen. Subj., 2012, 1820, 161–187.
13. J. F. Kasting, Science, 2001, 293, 819–820.
14. R. A. Kerr, Science, 2005, 308, 1730–1732.
15. T. Douglas, D. P. E. Dickson, S. Betteridge, J. Charnock, C. D. Garner and S. Mann, Science, 1995, 269, 54–57.
16. N. Behera, G. Bhattacharyya, S. Behera and R. K. Behera, JBIC, J. Biol. Inorg. Chem., 2024, 29, 455–475.
17. P. K. Koochana, A. Mohanty, S. Das, B. Subhadarshanee, S. Satpati, A. Dixit, S. C. Sabat and R. K. Behera, Biochim. Biophys. Acta, Gen. Subj., 2018, 1862, 1190–1198.
18. J. Pullin, J. M. Bradley, G. R. Moore, N. E. Le Brun, M. T. Wilson and D. A. Svistunenko, Angew. Chem., Int. Ed., 2021, 60, 8376–8379.
19. H. Yao, Y. Wang, S. Lovell, R. Kumar, A. M. Ruvinsky, K. P. Battaile, I. A. Vakser and M. Rivera, J. Am. Chem. Soc., 2012, 134, 13470–13481.
20. J. M. Bradley, D. A. Svistunenko, M. T. Wilson, A. M. Hemmings, G. R. Moore and N. E. Le Brun, J. Biol. Chem., 2020, 295, 17602–17623.
21. A. Chao, P. J. Sieminski, C. P. Owens and C. W. Goulding, Chem. Rev., 2019, 119, 1193–1220.
22. S. L. Byrne, N. D. Chasteen, A. N. Steere and A. B. Mason, J. Mol. Biol., 2010, 396, 130–140.
23. R. K. Raut, G. Bhattacharyya and R. K. Behera, Dalton Trans., 2024, 53, 13815–13830.
24. D. Sala, S. Ciambellotti, A. Giachetti, P. Turano and A. Rosato, J. Chem. Inf. Model., 2017, 57, 2112–2118.
25. K. N. Raymond, B. E. Allred and A. K. Sia, Acc. Chem. Res., 2015, 48, 2496–2505.
26. S. Dhungana and A. L. Crumbliss, Geomicrobiol. J., 2005, 22, 87–98.
27. E. C. Theil, Ann. N. Y. Acad. Sci., 2010, 1202, 197–204.
28. C. C. Winterbourn, Biological Chemistry of Reactive Oxygen Species, in Encyclopedia of Radicals in Chemistry, Biology and Materials, ed. C. Chatgilialoglu and A. Studer, 2012 DOI:10.1002/9781119953678.rad077.
29. S. J. Dixon and B. R. Stockwell, Nat. Chem. Biol., 2014, 10, 9–17.
30. H. Sies, V. V. Belousov, N. S. Chandel, M. J. Davies, D. P. Jones, G. E. Mann, M. P. Murphy, M. Yamamoto and C. Winterbourn, Nat. Rev. Mol. Cell Biol., 2022, 23, 499–515.
31. M. Wang, J. Gao, P. Muller and B. Giese, Angew. Chem., 2009, 48, 4232–4234.
32. H. B. Gray and J. R. Winkler, Chem. Sci., 2021, 12, 13988–14003.
33. A. Parida, G. Bhattacharyya, S. Mallik and R. K. Behera, Chem. Sci., 2025, 16, 3978–3997.
34. E. C. Theil, R. K. Behera and T. Tosha, Coord. Chem. Rev., 2013, 257, 579–586.
35. E. C. Theil and R. K. Behera, The Chemistry of Nature's Iron Biominerals in Ferritin Protein Nanocages, in Coordination Chemistry in Protein Cages, ed. T. Ueno and Y. Watanabe, John Wiley & Sons, Inc., 2013 DOI:10.1002/9781118571811.ch1.
36. R. K. Watt, R. J. Hilton and D. M. Graff, Biochim. Biophys. Acta, Gen. Subj., 2010, 1800, 745–759.
37. A. Parida and R. K. Behera, in Protein Cages: Design, Structure, and Applications, ed. T. Ueno, S. Lim and K. Xia, Springer US, New York, NY, 2023, pp. 121–134 DOI:10.1007/978-1-0716-3222-2_7.
38. F. Bou-Abdallah, Biochem. Biophys. Acta, 2010, 1800, 719–731.
39. R. K. Raut, G. Bhattacharyya, S. Mallik and R. K. Behera, ChemBioChem, 2025, e202500620.
40. K. Honarmand Ebrahimi, P.-L. Hagedoorn and W. R. Hagen, Chem. Rev., 2015, 115, 295–326.
41. G. J. McBean, Antioxidants, 2017, 6(3), 62.
42. A. W. Hansen and K. V. Venkatachalam, Biochem. Biophys. Rep., 2023, 35, 101529.
43. M. Y. Hamed and J. Silver, Inorg. Chim. Acta, 1983, 80, 115–122.
44. C. Kumar, A. Igbaria, B. D'Autreaux, A. G. Planson, C. Junot, E. Godat, A. K. Bachhawat, A. Delaunay-Moisan and M. B. Toledano, EMBO J., 2011, 30, 2044–2056.
45. O. Kabil, N. Motl and R. Banerjee, Biochem. Biophys. Acta, 2014, 1844, 1355–1366.
46. C. C. Winterbourn, Nat. Chem. Biol., 2008, 4, 278–286.
47. E. C. Theil, Inorg. Chem., 2013, 52, 12223–12233.
48. H. Shi, K. Z. Bencze, T. L. Stemmler and C. C. Philpott, Science, 2008, 320, 1207–1210.
49. D. R. Linden, M. D. Levitt, G. Farrugia and J. H. Szurszewski, Antioxid. Redox Signaling, 2009, 12, 1135–1146.
50. O. Kabil and R. Banerjee, J. Biol. Chem., 2010, 285, 21903–21907.
51. S. Cassanelli and J.-M. Moulis, Biochim. Biophys. Acta, Protein Struct. Mol. Enzymol., 2001, 1547, 174–182.
52. G. Cirino, C. Szabo and A. Papapetropoulos, Physiol. Rev., 2022, 103, 31–276.
53. P. R. Mendes, S. Felix Ddos, P. C. Silva, G. H. Pereira and M. O. Simoes, Nutr. Hosp., 2014, 31, 883–889.
54. S. Mishra and J. Imlay, Arch. Biochem. Biophys., 2012, 525, 145–160.
55. A. Giuffre and J. B. Vicente, Oxid. Med. Cell. Longevity, 2018, 2018, 6290931.
56. F. Carmona, Ò. Palacios, N. Gálvez, R. Cuesta, S. Atrian, M. Capdevila and J. M. Domínguez-Vera, Coord. Chem. Rev., 2013, 257, 2752–2764.
57. C. E. Outten, J. Biol. Chem., 2017, 292, 15990–15991.
58. L. Loiseau, S. Ollagnier-de-Choudens, L. Nachin, M. Fontecave and F. Barras, J. Biol. Chem., 2003, 278, 38352–38359.
59. B. Subhadarshanee, A. Mohanty, M. K. Jagdev, D. Vasudevan and R. K. Behera, Biochim. Biophys. Acta, Proteins Proteomics, 2017, 1865, 1267–1273.
60. S. N. Mthembu, A. Sharma, F. Albericio and B. G. de la Torre, ChemBioChem, 2020, 21, 1947–1954.
61. M. S. Joo, G. Tourillon, D. E. Sayers and E. C. Theil, Biol. Met., 1990, 3, 171–175.
62. R. C. Hider and X. L. Kong, Biometals, 2011, 24, 1179–1187.
63. G. Khare, P. Nangpal and A. K. Tyagi, Biochemistry, 2013, 52, 1694–1704.
64. R. Schmidt, M. G. Logan, S. Patty, J. L. Ferracane, C. S. Pfeifer and A. J. Kendall, ACS Omega, 2023, 8, 9356–9363.
65. R. E. Hansen, H. Østergaard, P. Nørgaard and J. R. Winther, Anal. Biochem., 2007, 363, 77–82.
66. R. K. Behera and E. C. Theil, Proc. Natl. Acad. Sci. U. S. A., 2014, 111, 7925–7930.
67. A. Mohanty, M. K, S. S. Jena and R. K. Behera, Biomacromolecules, 2021, 22, 1389–1398.
68. C. Dehner, N. Morales-Soto, R. K. Behera, J. Shrout, E. C. Theil, P. A. Maurice and J. L. Dubois, JBIC, J. Biol. Inorg. Chem., 2013, 18, 371–381.
69. A. Parida, A. Mohanty, B. T. Kansara and R. K. Behera, Inorg. Chem., 2020, 59, 629–641.
70. A. Mohanty, B. Subhadarshanee, P. Barman, C. Mahapatra, B. Aishwarya and R. K. Behera, Inorg. Chem., 2019, 58, 4741–4752.
71. A. Mohanty, A. Parida, B. Subhadarshanee, N. Behera, T. Subudhi, P. K. Koochana and R. K. Behera, Inorg. Chem., 2021, 60, 16937–16952.
72. W. Brand-Williams, M. E. Cuvelier and C. Berset, LWT – Food Sci. Technol., 1995, 28, 25–30.
73. F. M. Michel, H.-A. Hosein, D. B. Hausner, S. Debnath, J. B. Parise and D. R. Strongin, Biochim. Biophys. Acta, Gen. Subj., 2010, 1800, 871–885.
74. E. Sharifi-Zahabi, H. Abdollahzad, S. Mostafa Nachvak, J. Moloudi, M. R. Golpayegani, S. Asiaei, L. Rezavand, Z. Iraji and K. Jamshidi, Int. J. Clin. Pract., 2021, 75, e14062.
75. F. Bartoccini, M. Retini, R. Crinelli, M. Menotta, A. Fraternale and G. Piersanti, J. Org. Chem., 2022, 87, 10073–10079.
76. G. I. Giles and C. Jacob, Biol. Chem., 2002, 383, 375–388.
77. J. C. Lukesh, III, M. J. Palte and R. T. Raines, J. Am. Chem. Soc., 2012, 134, 4057–4059.
78. F. Bou-Abdallah, J. J. Paliakkara, G. Melman and A. Melman, Pharmaceuticals, 2018, 11(4), 120.
79. P. K. Koochana, A. Mohanty, B. Subhadarshanee, S. Satpati, R. Naskar, A. Dixit and R. K. Behera, Dalton Trans., 2019, 48, 3314–3326.
80. P. K. Koochana, A. Mohanty, A. Parida, N. Behera, P. M. Behera, A. Dixit and R. K. Behera, JBIC, J. Biol. Inorg. Chem., 2021, 26, 265–281.
81. H. B. Gray and J. R. Winkler, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 3534.
82. P. Arosio and S. Levi, Biochim. Biophys. Acta, Gen. Subj., 2010, 1800, 783–792.
83. A. Pandey, J. Pain, A. K. Ghosh, A. Dancis and D. Pain, J. Biol. Chem., 2015, 290, 640–657.
84. C. C. Winterbourn, in Methods in Enzymology, ed. E. Cadenas and L. Packer, Academic Press, 2013, vol. 528, pp. 3–25.
85. B. Halliwell, Free Radical Biol. Med., 2009, 46, 531–542.
86. N. Behera, G. Bhattacharyya and R. K. Behera, Chem. – Asian J., 2025, 20(24), e00803.
87. M. Taher, B. Maity, T. Nakane, S. Abe, T. Ueno and S. Mazumdar, Angew. Chem., Int. Ed., 2022, 61, e202116623.
88. A. Mohanty, A. Parida, R. K. Raut and R. K. Behera, ACS Bio Med. Chem. Au, 2022, 2, 258–281.

---

Footnote

† This paper is dedicated to Professor Anadi Charan Dash (FASc, FRSC) on the occasion of his 85th birthday, for his outstanding contribution and mentorship.

---