New Glycosalen–Manganese(III) Complexes and RCA₁₂₀ Hybrid Systems as Superoxide Dismutase/Catalase Mimetics

Abstract

Reactive oxygen species are implicated in several human diseases, including neurodegenerative disorders, cardiovascular dysfunction, inflammation, hereditary diseases, and ageing. Mn^III–salen complexes are superoxide dismutase (SOD) and catalase (CAT) mimetics, which have shown beneficial effects in various models for oxidative stress. These properties make them well-suited as potential therapeutic agents for oxidative stress diseases. Here, we report the synthesis of the novel glycoconjugates of salen complex, EUK-108, with glucose and galactose. We found that the complexes showed a SOD-like activity higher than EUK-108, as well as peroxidase and catalase activities. We also investigated the conjugate activities in the presence of Ricinus communis agglutinin (RCA₁₂₀) lectin. The hybrid protein–galactose–EUK-108 system showed an increased SOD-like activity similar to the native SOD1.

1. Introduction

Oxidative stress occurs from the imbalance between radical species and antioxidant defences. It has been associated with various adverse health conditions, such as chronic inflammation, diabetes, cancer, neurodegenerative diseases (Parkinson’s disease, Alzheimer’s disease, ALS), and ageing [1,2].

Oxidative injury in cells is controlled and preserved through enzymatic and nonenzymatic antioxidant systems [3]. The superoxide dismutase (SOD), catalase (CAT), and peroxidase enzymes form the front line of defence against oxidative stress products. The main nonenzymatic antioxidants are vitamins (C and E), β-carotene, uric acid, and GSH.

In light of their impact on health, there has been an interest in targeting reactive oxygen species (ROS) in the development of redox medicine [4]. Many antioxidants and redox drugs have been administrated as therapeutic strategies for a variety of diseases [5]. The therapeutic potential of SOD enzymes has also been explored, and bovine CuZn–SOD preparations (Orgotein^®) were administrated in the late 1970s to treat inflammatory diseases. Orgotein is only used as an anti-inflammatory drug in non-human animals [6]. MnSOD has also been investigated [7]. The therapeutical administration of SOD enzymes has several limitations, including the immunogenicity in the case of the SOD enzyme from non-human sources, the low intracellular uptake, short half-lives, administration route, and costs [8].

Low molecular weight complexes of redox metals (Fe, Cu, Mn) mimicking natural SOD enzymes have been widely reported [7,9,10]. SOD mimetics (SODm) would be potential therapeutics and useful probes to elucidate the physiopathologic role of intracellular superoxide radicals [10,11]. SODm have all been shown to positively affect the inflammatory state in lung epithelial cells in models of chronic obstructive pulmonary disease [12]. Mn^II or Mn^III complexes have been successfully studied as catalytic drugs among a variety of metal complexes, due to the low Fenton-based toxicity of manganese [13,14,15]. The Mn^II complexes of cyclic polyamines [15] aneN5 (1,4,7,10,13-pentaaza cyclopentadecane)-type ligands have been studied as SODm, and Imisopasem Manganese (M40403) and Avasopasem Manganese (M40419 or GC-4419) are the best systems in this family for diseases related to ROS dyshomeostasis. The GC-4419 is in a Phase 3 clinical trial to reduce side effects from anticancer therapy [16,17]. In 2020 Galera Therapeutics Inc. announced the phase II clinical trial with GC4419 for critical illness due to COVID-19 [18].

The Mn^II complexes of polyamine (salan-type) ligands have also been investigated as SOD mimetics [19]. The salan ligands have also been functionalized with cell-penetrating peptides to improve their cell uptake and antioxidant activity [19]. Other families of SODm include the Mn^III complexes of porphyrin– and Mn^III–salen (H₂salen = N,N′ bis(salicylidene)ethylenediamine)-type complexes [20,21]. The Mn^III complexes of a variety of porphyrins have been widely studied [14,15,22,23].

Mn^III–salen derivatives have been developing as combined SOD/CAT mimetics mainly by Eukarion Company [24,25,26]. Eukarion (EUK)-8 (or EUK-108 when the ligand is CH₃COO^- instead of Cl⁻) and EUK-134 (or EUK-113 when the ligand is CH₃COO^- instead of Cl^-) are the prototype molecules of this family of complexes and have been tested in several pathological conditions [5,27,28]. Currently, EUK-134 is a component of numerous anti-ageing skincare products for its antioxidant properties. Various salen-type complexes have been studied recently in in vitro and in vivo animal models of diseases closely related to oxidative stress, such as neurodegenerative diseases and cancer [25,29,30,31,32].

Based on the interest in salen derivatives [7,24,28,29,33,34], herein, we report the synthesis of two new Mn^III salen glycoconjugates. In particular, we conjugated an N,N′ bis(salicylidene) ethylenediamine derivative with glucose or galactose to synthesize EUK-108-analogous molecules (Figure 1).

We exploited the new features introduced by sugar moiety, such as the binding with specific lectins [35]. Ricinus communis agglutinin I (RCA₁₂₀, RCA I) is a glycoprotein composed of two A and two B units linked by a disulfide bond. Subunit A is the catalytic unit and confers toxicity to the protein, while chain B binds carbohydrate chains with non-reducing terminal galactose residues [36]. RCA₁₂₀ has low toxicity compared to the homolog ricin protein.

In the last few years, metal complexes have been caged in protein scaffolds to form artificial metalloenzymes, also called hybrids, and engineer new reactions [37,38,39,40]. Many anchoring strategies have been explored to build new artificial metalloenzymes. The properties of the scaffold protein can modulate the catalytic activity of the metal complex that acts as an artificial prosthetic group. The protein moiety provides a unique microenvironment (chirality, dielectric coefficient) and can improve solubility and substrate accessibility [37,38,39,40]. Some examples of the non-covalent strategy exploit the affinity of the metal complex for the protein, such as the streptavidine–biotin technology, which is applied in vivo for catalyzing abiotic reactions [41].

We synthesized non-covalent conjugates of RCA₁₂₀ with galactose–salen conjugate 4a. We investigated the SOD-like, peroxidase, and catalase activities of the metal complexes and the RCA₁₂₀–glycosalen complexes.

The hybrid protein–galactose–salen system showed an increased SOD-like activity.

2. Materials and Methods

2.1. Material

Anhydrous DMF, anhydrous methanol, acetobromo-α-D-galactose, acetobromo-α-D-glucose, Nα, Nβ-Di-Boc-L-2,3-diaminopropionic acid (dicyclohexylamine salt) (DAPBoc), cysteamine (2-mercaptoethylamine, mea), o-(benzotriazole-1-yl)-N,N,N′,N′-bis(tetramethylene)O-(1H-benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate (TBTU), 1-hydroxybenzotriazole (HOBT), and lectin from Ricinus communis (RCA₁₂₀) were purchased from Sigma-Aldrich and were used without any purification. CM Sephadex C-25 (Sigma) (NH₄⁺ form) and DEAE-Sephadex A-25 (HCO₃⁻ form) were used for ion exchange chromatography. Ultra-pure water (Milli-Q Element, Millipore, Italy) was used for all the experiments. Thin layer chromatography (TLC) was carried out on silica gel plates (Polygram SIL G/UV254 0.2 mm Macherey-negel). Glycosidic derivatives were detected on TLC using UV or the anisaldehyde test.

EUK-108 was synthesized as reported elsewhere [42].

2.2. Synthesis of Salen Ligands and Mn^III Complexes

2.2.1. Synthesis of 1-deoxy-1[(S-cysteamine)]-ß-galactose (1a) and of 1-deoxy-1[(S-cysteamine)]-ß-glucose (1b)

Mea (1.9 g, 24.6 mmol) and sodium methanolate (24.6 mmol) in water (2 mL) were added to a solution of α-D-acetobromo-galactose (2.42 g, 5.9 mmol) in DMF (2 mL). The reaction was carried out at 70 °C under stirring and nitrogen. After 3h, the solvent was evaporated. The product was obtained through the hydrolysis of acetyl groups in NaOH (1%) for 2h. The product was purified through reverse-phase RP8 column chromatography, and eluted using a NH₄Cl solution (0.1 M). The appropriate fractions were collected and evaporated (Rf = 0.2, PrOH/H₂O/AcOEt/NH₃, 4:3:2:1). The solid residue was further purified by a CM Sephadex C-25 column (20 × 60 mm, NH₄⁺ form) eluted using water and then a linear gradient of NH₄HCO₃ solution (400 mL) 0 → 0.3 M. Product 1a was obtained, with a yield of 30%. Rf = 0.42 (PrOH/H₂O/AcOEt/NH₃ 4:3:2:1).

ESI-MS: m/z = 240.1 (1a + H)⁺

¹H NMR (D₂O, 500 MHz) δ(ppm): 4.37 (1H, d, J_H1,H2 = 10.0 Hz, H-1 of Gal); 3.86 (1H, d, J_H3,H4 = 4,0 Hz, H-4 of Gal); 3.67–3.56 (3H, m, H-5, H-6_a, H-6_b of Gal); 3.54 (1H, dd, J_H3,H4 = 4.0 Hz, J_H2,H3 = 10.0 Hz, H-3 of Gal); 3.46 (1H, t, J_H1,H2 = 10,0 Hz, H-2 of Gal); 2.80 (2H, m, CH₂ in α to NH₂); and 2.75 (m, 2H, CH₂ in β to NH₂).

The same procedure given for 1a was followed to synthesize 1b ), starting from α-D-acetobromo-glucose. The yield was 35%. Rf = 0.42 (PrOH/H₂O/AcOEt/NH₃ 4:3:2:1).

ESI-MS m/z = 240.0 (1b + H)⁺

1H NMR (D₂O, 500 MHz) δ(ppm): 4.44 (1H, d, J_H1,H2 = 10.0 Hz, H-1 of Glc); 3,80 (1H, dd, J_H5,H6 = 9.0 Hz, J_H6,H6′ = 12.0 Hz, H-6_a of Glc); 3,60 (1H, dd, J_H6a,H6b = 12.0 Hz, H-6_b of Glc); 3.38 (2H, m, H-3, H-5 of Glc); 3.31 (1H, t, J_H4,H5 = 9.0 Hz, H-4 of Glc); 3.23(1H, t, J_H1,H2 = 10.0 Hz, H-2 of Glc); 2,83 (2H, m, CH₂ in α to NH₂); and 2.73 (2H, m, CH₂ in β to NH₂).

2.2.2. Synthesis of 1-deoxy-1[(S-cysteamidopropyl(1,2-diamino)]-ß-galactose] 2a and 1-deoxy-1[(S-cysteamidopropyl(1,2-diamino)]-ß-glucose] 2b

DAPBoc (250 mg, 0.82 mmol), TBTU (315mg, 0.82 mmol), and HOBT (111 mg, 0.82 mmol) in dry DMF (30 mL) were stirred under nitrogen at 25 °C. After 15 min, 1a was added (200 mg, 0.82 mmol) to the solution. The reaction was stopped when the reagents disappeared in TLC (PrOH/H₂O/AcOEt/NH₃, 5:2:1:1). DMF was evaporated and the solid was purified through a reverse-phase C-8 column with a gradient of H₂O/acetone (0→50%). The Boc group was removed with CF₃COOH for 1h. CF₃COOH was evaporated and the product was passed through a DEAE-Sephadex A-25, using water as the eluent. Rf = 0.31 (PrOH/H₂O/AcOEt/NH₃,5:2:2:2); and the yield was 40%.

ESI-MS m/z = 326.4 (2a +H)⁺

¹H NMR (D₂O, 500 MHz) δ(ppm): 4.40 (1H, d, J_H1,H2 = 10.0 Hz H-1 of Gal); 3.88 (1H, d, J_H3-H4 = 3.0 Hz, H-4 of Gal); 3.70–3.58 (3H, m, H-5, H-6_a, H-6_b of Gal); 3.55 (1H, dd, J_H2-H3 = 10.0 Hz, J_H3,H4 = 3.0 Hz, H-3 of Gal); 3.46 (1H, t, J_H1,H2 = 10.0 Hz, H-2 of Gal); 3.40 (3H, m, CH₂ in b to S, CH of DAP); 2.85 (1H, m, CH₂ of DAP); and 2.78 (4H, m,CH₂ in a to S and CH₂ of DAP).

2b was synthesized as reported for 2a

ESI-MS m/z = 326.6 (2b + H)⁺

¹H NMR (D₂O, 500 MHz) δ (ppm): 4.36 (1H, d, J_H1,H2 = 10.0 Hz H-1 of Glu); 3.84 (1H, d, J_H3,H4 = 10.0 Hz H-4 of Glc); 3.68–3.56 (3H, m, H-5, H-6_a, H-6_b of Glc); 3.53 (1H, dd, J_H2,H3, = 4.0 Hz, J_H3-H4, = 10.0 Hz, H-3 of Glc); 3.46 (H, t, J_{H,-H 2}= 10.0 Hz, H-2 of Glc); 3.38 (3H, m, CH₂ in β to S, CH of DAP); 2.82 (2H, m, CH₂ of DAP); and 2.75 (4H, m, CH₂ in α to S).

2.2.3. Synthesis of 1-deoxy-1[(S-cysteamidopropyl(1,2-diamino)N,N′-bis(salicylidene))]-ß-galactose] 3a

2a (50 mg, 1.5 mmol) was dissolved in anhydrous methanol (5 mL) and a stoichiometric amount of salicylaldehyde (12.5 mL, 3.0 mmol) was added under stirring at 25 °C. A yellow solid was formed. After 6 h, the solvent was evaporated under vacuum and the solid obtained was washed with hexane/acetone (1:1). The yield was 88%.

ESI-MS m/z = 534.5 (3a + H)⁺

¹H NMR (CD₃OD, 500 MHz) δ (ppm): 8.54 (s; 1H, m, imine H); 8.45(s; 1H, s, imine H); 7.32 (4H, m, H-6 and H-3); 6.82–6.83 (4H, m, H-4 and H-5); 4.33 (1H, d, J_XY = 6.0 Hz CH in α to O=C); 4.31 (1H, d, J_1,2 = 9.70 Hz Gal H-1and CH); 4.15 (2H, m, CHa in β to O=C-NH); 4.05 (2H, m, CHb in β to O=C-NH); 3.84 (1H, d, J_H3,H4 = 5.0 Hz, Gal H-4); 3.75 (1H, m, Gal H-5); 3.46 (1H, dd, J_H2,H3 = 11.0 J_H3,H4 = 5.0 Hz, Gal H-3); 3.57–3.48 (3H, m, Gal H-6, H-6′, H-2); 3.42 (2H, m, CH₂ in α to-NH-C=O); 2.89 (2H, m, CH_a in β to -NH-C=O); and 2.77 (2H, m, CHb in β to O=C-NH). The glucose–salen conjugate 3b was synthesized as reported for 3a.

ESI-MS m/z = 534.0 (3b + H)⁺

¹H NMR (CD₃OD, 500 MHz) δ(ppm): 8.53 (1H, s, imine H); 8.45 (1H, s, imine H); 7.43–7.20 (4H, m, H-6 and H-3); 6.97–6.82 (4H, m, H-4 and H-5); 4.36 (1H, d, J_H1,H2 = 9.7 Hz, Glc H-1); 4.33 (1H, m, CH in α O=C-NH); 4,14 (2H, m, CHa in β to O=C-NH); 4.05 (2H, m, CHb in β to O=C-NH); 3.84 (1H, d, J_H3,H4 = 5.0 Hz, Glc H-4); 3.84 (1H, d, J_3,4 = 12.0 Hz, Glc H-6); 3.61–3.45 (5H, m, Glc H-5, H-4, H-6′, CH₂ in α to -NH-C=O); 3.26 (1H, m, Glc H-3); 3,19 (1H, t, J_H1-H2 = 10,0 Hz, Glc H-2); 2.88 (2H, m, CHa in β to -NH-C=O); and 2.72 (2H, m, 2H, CHb in β to O=C-NH).

2.2.4. Synthesis of Manganese(III) Complexes of 3a and 3b

3a (115 mg, 0.086 mmol) and NaOH (0.172 mmol, methanolic solution) were dissolved in anhydrous methanol (5 mL). Mn(CH₃COO)₂ (0.086 mmol) was added to the solution under stirring. The yellow solution turned brown and the reaction mixture was refluxed. After 3 h, the precipitate was filtered off and washed with acetone. The solid was further purified through precipitation from a water and acetone solution.

The yield was 90%.

4a: ESI-MS m/z = 586.2 [4a-Ac]⁺

4b was synthesized as reported for 4a.

4b: ESI-MS m/z = 586.0 [4b-Ac]⁺

2.2.5. Preparation of Hybrid Mn^III and RCA₁₂₀ System

4a- or 4b-RCA₁₂₀ adducts were prepared by incubating equimolar amounts of RCA₁₂₀ lectin and Mn complex for 10 min in the buffer solution, at 25 °C.

2.3. Instrumentation

The NMR spectra were recorded at 25 °C with a Varian Inova 500 spectrometer. The ¹H NMR spectra were acquired using standard Varian library pulse programs. The 2D spectra (COSY, TOCSY, HSQC) were acquired using 1K data points, 256 increments, and a relaxation delay of 1.5 s.

The mass spectra were recorded with a Mariner Perseptive Biosystem ESI-MS.

The UV-visible spectra were recorded with an Agilent 8452A diode array spectrophotometer. The CD spectra were performed on a JASCO model J-810 spectropolarimeter.

The 4a, 4b, and EUK-108 stock solutions were prepared in EtOH/water (1:1). The final EtOH concentration did not exceed 5% and did not affect the measurements. The spectra of freshly prepared solutions were recorded at 25 °C.

2.4. Superoxide Dismutase Assay

The SOD-like activity was determined using the indirect method [43]. Superoxide anion was generated by the xanthine/xanthine oxidase system and spectrophotometrically detected by monitoring the nitro blue tetrazolium (NBT) reduction to formazan at 560 nm, for 600 s.

The amount of xanthine oxidase required to produce a ΔA₅₆₀ nm/min = 0.024 was added to 2 mL of the reaction mixture (NBT 250 μM, xanthine 50 μM, phosphate buffer 0.010 M, pH 7.4), as reported elsewhere [44]. The ΔA₅₆₀ = 0.024 nm/min corresponds to the 1.1 mM/min^· O₂ production rate. The NBT reduction rate was also measured in the presence of the Mn^III complexes (concentration ranging from 10⁻⁵ to 10⁻⁸ M). All measurements were carried out at 25 °C under stirring. In separate experiments, urate production was spectrophotometrically monitored at 295 nm to exclude the inhibition of xanthine oxidase activity in the presence of Mn^III complexes or RCA₁₂₀.

The I₅₀ value (the concentration of metal complex required to inhibit 50% of NBT reduction) was determined.

2.5. Catalase Activity Assay

The catalase activity was determined as reported elsewhere [34]. An H₂O₂ solution (30 mM) was incubated at 25 °C with theMn^III complex (concentration ranging from 10⁻⁵ to 10⁻⁶ M) in 0.010 M phosphate buffer (pH 7.4).

ABTS (2,2-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (50 mM) and horseradish peroxidase (1.3 mg/mL) were added after 50 min. The absorbance at 735 nm was measured after 5 min, and the amount of H₂O₂ consumed per min per mmole from the compound was determined.

2.6. Peroxidase Activity Assay

The peroxidase activity of 4a, 4b, and EUK-108 compounds was assayed by monitoring spectrophotometrically the H₂O₂-dependent oxidation of ABTS [45]. The ABTS (0.2 mM), Mn^III complex (10 mM), and H₂O₂ (0.5 mM) in phosphate buffer (50 mM, pH 7.4) were incubated at 25 °C.

ABTS oxidation was monitored at 735 nm (ε_ABTS⁺ = 15,000) where the metal complexes did not show any absorption. No oxidation reaction of ABTS with the complexes was observed in the assay condition.

3. Results and Discussion

3.1. The Ligands

4a and 4b were synthesized from sugar amino-derivatives via the multistep procedure reported in Figure 2. The anomerically pure 1a and 1b were obtained in good yield through a nucleophilic substitution reaction between α-D-acetobromo-glycoside and cysteamine in water at basic pH. The amino group of 1a and 1b was linked to DAPBoc through a condensation reaction, and the products were purified using reverse-phase chromatography.

After removing the Boc-protecting groups through acid hydrolysis, the derivatives 2a and 2b were obtained and isolated through ionic-exchange chromatography. 3a and 3b were synthetized from 2a and 2b and sal-aldehyde at 25 °C in anhydrous methanol.

The NMR and ESI-MS spectra confirmed the identity of the products and intermediates (Figure S1–S13). In the ¹H NMR spectrum of 1a and 1b, the signals of the sugar unit and the ethylenic chain were present. The chemical shift of the H-1 signal and the coupling constant values (10.0 Hz) confirmed the β-configuration of the anomeric carbon of the sugar residue. In the ¹H NMR spectra of 2a, the signals of the galactose protons were assigned using the COSY and TOCSY spectra. The protons of the cysteamine moiety resonated at 3.50 ppm and at 2.78 ppm, while the protons of the ABX system of the propionyl chain resonated at 3.40 ppm (X), 2.83 and 2.80 ppm (A, B). A similar trend was found in the 2b spectrum.

In the ¹H NMR spectra of 3a, the imino protons appeared at 8.60 and 8.49 ppm. The signal at 8.60 ppm was assigned to the imino group α to the amido group. The chemical shifts of the protons of the two benzene rings also resonated differently. The signals of the ABX system of the diamino–propionyl chain were evident in the spectra. The spectra of 3b were similar to the spectra of 3a.

3.2. The Manganese(III) Complexes

The Mn^III complexes 4a and 4b were synthesized, as reported for salen complexes [46], at 25 °C and were isolated by adding acetone to the reaction mixture. The UV–Vis spectra of 4a and 4b were not significantly different from that of the parent salen complex EUK-108. In the spectra, there were the typical transitions at 323 nm, due to π → π* C=N groups and at 388 nm, due to LMCT, suggesting the same coordination environment of the Mn^III compared to EUK-108. In the ESI mass spectra (Figure S13), a leading peak corresponded to the complex species [Mn^III complex]⁺ (4a or 4b without CH₃COO⁻ ligand).

3.3. SOD-like Activity

The reaction of 4a and 4b with the superoxide anion was determined by competition kinetic experiments with NBT being used as the target molecule [47,48]. Plotting Vo/Vcat-1 versus the complex concentration produced a straight line with a slope k_cat/k_NBTx[NBT] [49] (Figure S14), where Vo was the NBT reduction rate Vcat was the NBT reduction rate in the presence of the investigated complex, and k_NBT was 5.88 × 10⁴ M⁻¹s⁻¹ [43,50]. The I₅₀ and k_cat values are reported in Table I. 4a and 4b showed good SOD-like activity, and their I₅₀ values were about ten times lower than that of EUK-108. The I₅₀ values were similar to those reported for the EUK-108–cyclodextrin conjugate [44]. The effect of the modification of the salen–diamine bridge on catalytic activity has been explored in a few cases [26,45,51]. In the case of the cyclodextrin derivatives previously studied, we hypothesized the role of the hydrophobic cavity and/or OH groups in order to explain the improved SOD-like activity. The results on monosaccharide conjugates may suggest the prominent role of OH groups in improving the SOD-like activity of the catalytic center. Molecular modelling suggests that, in 4a and 4b, the sugar side-chain may bend toward the catalytic center (Figure 3), which could improve the SOD-like activity.

4a and 4b activities were also investigated in the presence of RCA₁₂₀. The 4a/ or 4b/RCA₁₂₀ adducts were synthesized by pre-incubating equimolar amounts of RCA₁₂₀ lectin and 4a or 4b for 15 min in the buffer solution at 25 °C.

The RCA₁₂₀ alone did not show any interference with the SOD assay. The I₅₀ value of the 4a–RCA₁₂₀ adduct was 5.0 × 10⁻⁸ M and the k_cat was 2.8 × 10⁸ M⁻¹ s⁻¹ (

Table 1. SOD-like activity in the Fridovich assay (NBT = 250 μM) of 4a, 4b, and their RCA₁₂₀ adducts.

Complex	I50 (μM)	kcat (M−1 s−1)
EUK-108	2.05 (±0.03)	(7.2 ± 0.9) × 106
4a	0.20 (±0.04)	(7.4 ± 0.7) × 107
4b	0.25 (±0.05)	(5.9 ± 0.4) × 107
4a/RCA120	0.05 (±0.01)	(2.9 ± 0.3) × 108
4b/RCA120	0.20 (±0.03)	(8.0 ± 0.4) × 107

, Figure S15). These values were significantly higher than those of the free 4b and they were of the same order of magnitude as that of the native SOD1 enzyme (1.4 × 10⁸ M) [52]. No activity improvement was observed when 4b was pre-incubated with RCA120 [8].

We hypothesized that RCA₁₂₀ could bind 4a, as reported for other galactose conjugates [35,53], and that the protein environment could modulate the activity of the catalytic center, as found for other SOD-mimetic hybrid systems [39,54,55,56]. The interaction between the RCA₁₂₀ and 4a complex was investigated using CD spectroscopy. In the UV region of the CD spectra (Figure S16), RCA₁₂₀ showed a positive peak at 190 nm and negative peaks at 208 nm and 220 nm, in keeping with the partial α-helix structure of RCA₁₂₀ (15%) [53]. The intensity of the signals was slightly modified after the addition of the 4a complex to RCA₁₂₀ (Figure S16), suggesting the interaction of 4a with the protein. No modification of the CD spectrum of RCA₁₂₀ was found when 4b was added to RCA₁₂₀ (Figure S17). These data can explain the effect on the SOD activity in the presence of the RCA₁₂₀ protein, which can only bind the 4a complex.

3.4. Catalase Activity

Both salen complexes 4a and 4b exhibited a similar catalase activity, and their activities (

Table 2. Catalase and peroxidase activity of 4a, 4b, and their RCA₁₂₀ adducts.

Complex	CAT Activity	Peroxidase Activity
	µmol H2O2/min/mmol	µM ABTS/min
EUK108	32.2 (±1.7)	18.3 (±0.7)
4a	40.8 (±1.5)	21.6 (±0.6)
4b	40.6 (±1.2)	23.3 (±0.5)
4a/RCA120	31.5 (±1.5)	35.7 (±0.7)
4b/RCA120	28.8 (±1.2)	25.3 (±1.0)

) were slightly lower than EUK-108’s [45].

The reaction mechanism by which the Mn complexes act as catalase mimetics has been investigated [45]. It involves the oxidation of Mn^III by H₂O₂ to oxo–manganese and the formation of H₂O. The oxo–manganese complex is then reduced to Mn^III by another H₂O₂ molecule to form O₂.

The presence of RCA₁₂₀ slightly changed the CAT activity of both complexes.

3.5. Peroxidase Activity

The peroxidase activities of 4a, 4b, and EUK-108 were determined using the ABTS assay [29,30,34]. ABTS was the model substrate which reacted with H₂O₂

2 ABTS + H₂O₂ + 2H+ = 2 ABTS·+ + H₂O

Mn complexes can catalyze the oxidation of ABTS with H₂O₂ (peroxidase activity). The mechanism for salen–Mn^III complexes consisted in the formation of the oxo–manganese complex intermediate, which could oxidate the ABTS generating ABTS⁺·[57,58,59,60].

The UV-vis spectrum of the green ABTS^·+ radical cation showed characteristic bands at 415, 650, 735, and 815 nm. The solution of ABTS and H₂O₂ without the Mn complex was stable for several hours at 25 °C [33].

The amount of the ABTS⁺ formed per minute (µM ABTS/min) at the concentration of the complex 1.0 × 10⁻⁵ M is reported in Table 2.

The two Mn^III glycosalen complexes showed similar peroxidase activity compared to EUK-108.

The RCA₁₂₀–4a adduct slightly improved the peroxidase activity compared to 4a (Table 2). No effect of RCA120 was found for 4b.

4. Conclusions

Antioxidant enzymes play a crucial role in physiological and pathological states. SOD mimetics may provide therapeutic strategies in redox medicine to reduce the severity of diseases related to oxidative stress. A family of SOD/Catalase mimetics are Mn^III complexes of salen-type ligands.

We synthesized new glycoconjugates of EUK-108 and their Mn^III complexes. The sugar moiety increased the solubility in the water of the system compared to that of the free EUK-108. The glycoconjugation improved the efficiency of SOD mimetics by about ten times compared to that of the free EUK-108. Instead, the glycoconjugates showed a similar catalase and peroxidase activity to the EUK-108. The sugar moiety conferred new properties to the conjugates, such as the binding with specific lectins. We found that the galactose derivative interacted with RCA₁₂₀ lectin, which can selectively recognize galactose units. The hybrid system galactose–EUK-108–RCA₁₂₀ lectin showed five times better SOD-like activity compared to that of the free galactose–EUK-108 conjugate. The hybrid system showed a SOD activity similar to that of the native SOD1 enzyme. The RCA₁₂₀ protein environment could modulate the activity of the catalytic center.

These results disclose the potential of supramolecular mimetics. Based on the interest in antioxidant enzyme mimetics to counteract oxidative stress, this approach could be helpful for further improvement of the SOD mimetics.