Design and Synthesis of Novel Antioxidant 2-Substituted-5,7,8-Trimethyl-1,4-Benzoxazine Hybrids: Effects on Young and Senescent Fibroblasts

The exponential growth of the aged population worldwide is followed by an increase in the prevalence of age-related disorders. Oxidative stress plays central role in damage accumulation during ageing and cell senescence. Thus, a major target of today’s anti-ageing research has been focused on antioxidants counteracting senescence. In the current work, six novel 5,7,8-trimethyl-1,4-benzoxazine/catechol or resorcinol hybrids were synthesized connected through a methoxymethyl-1,2,3-triazolyl or a 1,2,3-triazoly linker. The compounds were evaluated for their antioxidant capacity in a cell-free system and for their ability to reduce intracellular ROS levels in human skin fibroblasts, both young (early-passage) and senescent. The most efficient compounds were further tested in these cells for their ability to induce the expression of the gene heme oxygenase-1 (ho-1), known to regulate redox homeostasis, and cellular glutathione (GSH) levels. Overall, the two catechol derivatives were found to be more potent than the resorcinol analogues. Furthermore, these two derivatives were shown to act coordinately as radical scavengers, ROS inhibitors, ho-1 gene expression inducers, and GSH enhancers. Interestingly, one of the two catechol derivatives was also found to enhance human skin fibroblast viability. The properties of the synthesized compounds support their potential use in cosmetic applications, especially in products targeting skin ageing.


Introduction
The aged population worldwide is increasing exponentially, with estimations by the United Nations that one out of six people will be over 65 years old in 2050 [1].One of the main consequences is the amplified prevalence of age-related disorders, such as cardiovascular diseases, cancer, musculoskeletal syndromes, and neurodegenerative diseases [2].According to the "free radical theory of ageing", proposed already in 1956 [3], oxidative stress is the major factor leading to age-related accumulation of defects.Observations at the cellular level support the idea that free radicals and reactive oxygen species (ROS) may lead to additive damages of subcellular organelles, especially mitochondria, thus creating further generation of more free radicals and a positive feedback loop [4].Moreover, ROS and free radicals have been shown to induce cellular senescence to many different cell types [5][6][7].Senescent cells are characterized by their inability to proliferate, by their pro-inflammatory and catabolic phenotype, and by their involvement in the pathogenesis and/or aggravation of age-related diseases [8].Consequently, cellular senescence has been included among the "Hallmarks of Ageing" [9].Accordingly, a major target of today's anti-ageing research has been focused on antioxidants counteracting senescence [10].
Antioxidants 2024, 13, x FOR PEER REVIEW 2 of 24 pro-inflammatory and catabolic phenotype, and by their involvement in the pathogenesis and/or aggravation of age-related diseases [8].Consequently, cellular senescence has been included among the "Hallmarks of Ageing" [9].Accordingly, a major target of today's anti-ageing research has been focused on antioxidants counteracting senescence [10].1,2-, 1,3-, and 1,4-benzoxazines are considered as one of the key classes of organic molecules endowed with a broad spectrum of biological activities [11,12].More specifically, 1,4-benzoxazines represent a privileged scaffold in drug discovery due to their fascinating pharmacological profile [13,14].They have been investigated as potential antimicrobial/antifungal [15][16][17][18][19][20], antioxidant [21][22][23][24][25][26], anti-infective [27], antidiabetic [28], or anticancer agents [29][30][31][32][33][34][35], as well as against neurodegenerative [36] and cardiovascular disorders [37,38].In particular, the 5,7,8-trimethyl-1,4-benzoxazine moiety can be considered as a bioisostere of the 5,7,8-trimethyl-1,4-benzopyran nucleus [39,40] that is the key heterocycle moiety of the well-known chain-breaking antioxidant vitamin E. Calogeropoulou et al. were the first to synthesize a number of 5,7,8-trimethyl-1,4-benzoxazine derivatives endowed with an array of biological activities, specifically against arrhythmias associated with ischemia-reperfusion injury [41], against toxoplasmosis inhibiting the proliferation of Toxoplasma gondii tachyzoites [27], against prion diseases inhibiting the formation of PrPSc [42], and finally as modulators of the AtoSC two-component system mediated signaling [43].Thus, as a continuation of our previous work, herein we employed the wellestablished medicinal chemistry approach of molecular hybridization [44] and synthesized six compounds combining the 5,7,8-trimethyl-1,4-benzoxazine scaffold and catechol or resorcinol moieties.The two pharmacophores are connected through the amide bond bioisostere of the 1,2,3-triazole heterocycle (Figure 1) [45].1,2,3-triazoles have been employed as linkers in the synthesis of hybrid compounds due to their sturdiness against oxidation or reduction and their increased solubility and improved interactions with biological targets due to their ability to form hydrogen bonds.Moreover, phenolic compounds such as catechol or resorcinol derivatives are active against ageing and age-related diseases and disorders, mainly due to their potent antioxidant profile.For example, one of the most well-known catechol derivatives, hydroxytyrosol (isolated from extra virgin oil), possesses powerful antioxidant, anti-inflammatory [46][47][48], wound healing [49], neuroprotective [50,51], anti-ageing [52,53], and cardioprotective properties [54,55].In turn, resorcinol-substituted compounds possess activity against hyperpigmentation [56][57][58], reactive oxygen species (ROS) [59], or as cholinesterase inhibitors [60].The new compounds were tested for their antioxidant activities in cellfree and cell-based systems.In particular, they were initially evaluated for their antioxidant capacity in a cell-free system and their ability to reduce intracellular ROS levels using human young (early-passage) and senescent skin fibroblasts.Furthermore, compounds possessing intracellular antioxidant properties were tested for their ability to induce the expression of gene heme oxygenase-1 (ho-1), known to regulate redox homeostasis, in young and senescent human skin fibroblasts.Finally, their effect on cellular glutathione (GSH) levels in human skin fibroblasts was assessed.1,2,3-triazoles have been employed as linkers in the synthesis of hybrid compounds due to their sturdiness against oxidation or reduction and their increased solubility and improved interactions with biological targets due to their ability to form hydrogen bonds.Moreover, phenolic compounds such as catechol or resorcinol derivatives are active against ageing and age-related diseases and disorders, mainly due to their potent antioxidant profile.For example, one of the most well-known catechol derivatives, hydroxytyrosol (isolated from extra virgin oil), possesses powerful antioxidant, anti-inflammatory [46][47][48], wound healing [49], neuroprotective [50,51], anti-ageing [52,53], and cardioprotective properties [54,55].In turn, resorcinol-substituted compounds possess activity against hyperpigmentation [56][57][58], reactive oxygen species (ROS) [59], or as cholinesterase inhibitors [60].The new compounds were tested for their antioxidant activities in cell-free and cell-based systems.In particular, they were initially evaluated for their antioxidant capacity in a cell-free system and their ability to reduce intracellular ROS levels using human young (early-passage) and senescent skin fibroblasts.Furthermore, compounds possessing intracellular antioxidant properties were tested for their ability to induce the expression of gene heme oxygenase-1 (ho-1), known to regulate redox homeostasis, in young and senescent human skin fibroblasts.Finally, their effect on cellular glutathione (GSH) levels in human skin fibroblasts was assessed.

General
Melting points were determined with an Electrothermal Digital Melting Point Apparatus, Cole-Parmer ET0001/Version 1.0 and are uncorrected. 1H NMR spectra were recorded on Varian spectrometers (Varian Medical Systems, Inc., Palo Alto, CA, USA) operating at 300 or 600 MHz. 13 C NMR spectra were recorded at 75 or 150 MHz, using CDCl 3 , CD 3 OD or (CD 3 ) 2 CO as solvents.Chemical shifts are reported in δ units, parts per million (ppm) downfield from TMS. Electro-spray ionization (ESI) mass spectra were recorded on a LC-MSn Fleet Thermo spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) using MeOH as solvent.HRMS spectra were recorded in the ESI or APCI mode on a UPLC-MSn Orbitrap Velos-Thermo spectrometer (Thermo Fisher Scientific, Waltham, MA, USA).Reactions under microwave irradiation were performed in a CEM Discover Lab Mate reactor (CEM Corporation, Matthews, NC, USA).Flash column chromatography (FCC) was performed on Merck silica gel 60 (230-400 mesh) Merck KGaA, Darmstadt, Germany) and TLC on Merck 60 F254 films (0.2 mm) precoated glass plates Merck KGaA, Darmstadt, Germany).Spots were visualized with UV light at 254 nm and PMA stain (phosphomolybdic acid 10% in absolute ethanol).All solvents were dried and/or purified according to standard procedures prior to use.All reagents employed in the present work were purchased from commercial suppliers and used without further purification.Reactions were run in flame-dried glassware under an atmosphere of argon.

General Procedure for the Deprotection of Methoxy Groups of Compounds 20 and 25
To a solution of compound 20 or 25 (1 eq) in dry CH 2 Cl 2 (0.05 M), BF 3 S(Me) 2 (10 eq) was added at 0 • C and the reaction mixture was stirred at ambient temperature for 2-24 h.After completion of the reaction (checked by TLC), the solvent and excess reagent were evaporated under argon stream.The residue was extracted with ethyl acetate.The organic layer was washed with saturated aqueous NaCl, dried over Na 2 SO 4 , and the solvent was evaporated in vacuo.The desired product was afforded after purification by FCC (CH 2 Cl 2 /acetone 90:10 v/v).

Cells and Culture Conditions
FF95 fibroblast cultures, previously established by outgrowth from foreskin of a oneyear-old healthy donor [61], were kindly provided by Prof. Karin Scharffetter-Kochanek.Cells were routinely cultured in Dulbecco's Modified Eagle Medium (DMEM; PAN Biotech GmbH, Aidenbach, Germany) supplemented with antibiotics, 100 IU/mL of penicillin and 100 µg/mL of streptomycin (Biosera, Nuaillé, France), as well as 10% (v/v) fetal bovine serum (FBS; Life Technologies Europe BV, Thessaloniki, Greece), in a humidified environment of 5% CO 2 and 37 • C. Cells were routinely subcultured once a week at a 1:2 split ratio, using a trypsin (Life Technologies Europe BV)-citrate (0.25%-0.3% w/v) solution.Cell counting following trypsinization and suspension in IsoFlow Sheath Fluid (Beckman Coulter Diagnostics, Brea, CA, USA) was performed using a Coulter counter (Beckman Coulter Diagnostics).Cells were tested for mycoplasma contamination at regular intervals and found to be mycoplasma-free.
For obtaining senescent fibroblast cultures, FF95 cells were exposed to sublethal repeated doses of ionizing radiation, as previously reported [62].Briefly, cells received a cumulative dose of 50 Gy through their exposure to a 60 Co gamma source (Gamma Chamber 4000 A, Isotope Group, Bhadha Atomic Research Company, Trombay, Bombay, India) at a rate of 2.5 Gy/min.The cultures were then left to grow and passaged (2-3 times) until they exhibited less than 5% 5-bromo-2 ′ -deoxyuridine (BrdU) incorporation and more than 90% senescence-associated β-galactosidase (SA-β-Gal) staining, hence considered to be senescent [63].In particular for SA-β-Gal staining, cells were sparsely plated on glass coverslips, left to grow for 5 days, then fixed with 3% (v/v) formaldehyde in PBS.Staining was implemented through incubation with 40 mM citric acid/sodium phosphate (pH 6.0) supplemented with 150 mM NaCl, 2 mM MgCl 2 , 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, and 1 mg/mL X-Gal at 37 0 C. SA-β-Gal-positive or -negative cells were observed under an ECLIPSE Ts2 inverted microscope (Nikon, Tokyo, Japan) and photographs were captured through a Basler acA1920-155ucMIC camera (Basler AG, Ahrensburg, Germany).

Cytotoxicity Assay
Cytotoxicity was estimated by a modification of the 3-(4,5-Dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) assay [65].Briefly, FF95 fibroblasts were plated in 96-well flat-bottomed transparent microplates in DMEM 10% FBS at a density of 10,000 cells/well.After an overnight incubation to ensure cell attachment, serial dilutions of the test compounds in culture medium were added and left to act on the cells for 72 h.Subsequently, the medium was replaced by serum-free, phenol-red-free DMEM (PAN Biotech GmbH) containing 1 mg/mL of MTT (Sigma), incubated for four hours; the MTT-formazan crystals formed were solubilized in 2-propanol (Sigma) and the absorbance was measured using a Spark multimode microplate reader (Tecan Group Ltd., Männedorf, Switzerland) at a wavelength of 550 nm (reference wavelength 690 nm).Corresponding vehicle (DMSO) dilutions were used as control.

Intracellular Reactive Oxygen Species (ROS) Levels
FF95 fibroblasts were plated, in clear-bottomed black 96-well microplates in DMEM 10% FBS at a density corresponding to the 1:2 split ratio.When the cultures have reached confluency, the medium was replaced by a serum-free one, containing non-cytotoxic concentrations of the test compounds.Following an overnight incubation, 2 ′ ,7 ′ -dichlorodihydrofluorescein diacetate (DCFH-DA; Sigma) was added at a final concentration of 10 µM, and-at various time-points-fluorescence emission was monitored at 520 nm following excitation at 485 nm, using a FLUOstar Optima microplate reader.DMSO-treated cell cultures served as negative control and Trolox was used as a positive control at a final concentration of 20 µM.Beyond basal ROS levels estimation, in order to assess the capability of the test compounds to prevent intracellular ROS level-stimulation, 1.5 mM H 2 O 2 in serum free medium was added after the 30 min incubation with DCFH-DA, and the fluorescence emission was measured as above [64].

Cellular Glutathions (GSH) Levels
An adaptation of the microplate assay based on the fluorescent probe monochlorobimane (mCB) was used for the assessment of the glutathione (GSH) levels in human skin fibroblasts [66].Briefly, FF95 fibroblasts were plated in clear-bottomed black 96-well microplates in DMEM 10% FBS at a density corresponding to the 1:2 split ratio.When the cultures reached confluency, the medium was replaced by the phenol-red-free, serum-free one, containing non-cytotoxic concentrations of the test compounds, and the cells were incubated for 20 h.At the end of the incubation period, the medium was replaced with 5 µM mCB (MedChemExpress, New Jersey, NJ, USA) diluted in Hanks' Balanced Salt Solution (HBSS) for a further 4 h, and fluorescence was recorded in a Spark (Tecan) plate reader using an excitation wavelength of 380 nm and emission wavelength of 480 nm.Treatment of the cells with 8 mM N-ethylmaleimide (NEM; Sigma), previously reported to deplete cellular GSH (Hedley et al., 1994), was used to assess nonspecific background labelling.N-acetycysteine (NAC; Sigma) at 10 mM was used as positive control [67], and vehicle (DMSO) as negative control.

RNA Extraction and Gene Expression Analysis
Gene expression analysis was based on quantitative real-time polymerase chain reaction (qRT-PCR) [63].Total RNA was extracted using Trizol (Life Technologies Europe BV), following the manufacturer's instructions, from FF95 cell cultures growing in DMEM 10% FBS in the presence or absence of non-cytotoxic concentrations of the test compounds for 24 h.First-strand complementary DNA (cDNA) was synthesized from 500 ng of the total RNA using PrimeScript™ RT Reagent Kit according to the manufacturer's instructions (Takara Bio Inc, Tokyo, Japan).Five microliters of the cDNA (1:25) per sample was subjected to qRT-PCR using the KAPA SYBR FAST qPCR kit (KAPA Biosystems, Wilmington, MA, USA).The reaction was carried out in an MX 3000 P QPCR Systems Cycler (Stratagene, La Jolla, CA, USA), and data analysis was performed with MxPro version 4.1 QPCR software (Stratagene, La Jolla, CA, USA).Relative mRNA levels compared to untreated cells were estimated as described before [63] with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) serving as the reference gene.The ∆∆Ct method was used to evaluate the relative messenger RNA expression of each gene.The primers used for amplification were ho-1: forward 5 ′ -GCC CTT CAG CAT CCT CAG TTC C-3 ′ , reverse 5 ′ -AGT GGT CAT GGC CGT GTC AAC-3 ′ ; gapdh: forward 5 ′ -GAG TCC ACT GGG GTC TTC-3 ′ , reverse 5 ′ -GCA TTG CTG ATG ATC TTG GG-3 ′ .
in Schemes 2 and 3. Briefly, esterification of (3,4-dihydroxyphenyl)acetic acid with meth-anol, in the presence of a catalytic amount of H2SO4 to the methyl ester 9 was followed by protection of the catechol moiety as the acetonide 10.Reduction of the ester in compound 10 with LiAlH4 to alcohol 11 followed by formation of the corresponding tosylate 12 [73] and reaction with NaN3 in DMF gave azide 13.TFA-mediated removal of the acetonide protecting group afforded the final azide 14 (Scheme 2).Reduction of (3,5-dimethoxylphenyl)acetic acid with LiAlH4 gave the corresponding alcohol 15, which was transformed to the respective mesylate ester 16 [74].Reaction of 16 with NaN3 in DMF gave the corresponding azide 17 [74] in high yield (90%).Deprotection of the methoxy groups to obtain the desired analogue 18 was achieved by using BF3•S(CH3)2 complex in CH2Cl2 [73] (Scheme 3).Following the synthesis of the desired alkynes (7, 8) and azides (13, 14, 17, and 18) 1,2,3-triazole-substituted benzoxazines were prepared regioselectively through a Cu 1 -catalyzed click' cycloaddition reaction [75,76].The click reaction was carried out in the presence of CuSO4•5H2O and sodium ascorbate (NaAsc) in a mixture of CH2Cl2/H2O (1:1) under conventional conditions or tBuOH/H2O (1:1) under microwave-assisted irradiation, as shown in Schemes 4 and 5. Thus, the catecholic final product 19 (TC488) was obtained by the cycloaddition reaction of alkyne 7 and azide 14 at room temperature in 57% yield.In turn, click reaction between alkyne 7 and the protected azide 17 at room temperature afforded the 1,2,3-triazolyl derivative 20 in 67% yield.Deprotection of the methoxy groups in 20 using BF3S(Me)2 also resulted in the removal of the p-methoxybenzyl group affording compound 22 (TC490) in 35% yield.In addition, we obtained the byproduct derivative 22a (TC491) in 14% yield resulting from the migration of the p-hydroxybenzyl group from N-4 to C6 of the 5,7,8-trimethyl-1,4-benzoxazine system.Thus, in order to obtain the deprotected resorcinol derivative 21 (TC489) bearing the p-methoxybenzyl group at N-4 for protection of the catechol moiety as the acetonide 10.Reduction of the ester in compo 10 with LiAlH4 to alcohol 11 followed by formation of the corresponding tosylate 12 and reaction with NaN3 in DMF gave azide 13.TFA-mediated removal of the aceto protecting group afforded the final azide 14 (Scheme 2).Reduction of (3,5-dimet ylphenyl)acetic acid with LiAlH4 gave the corresponding alcohol 15, which was tr formed to the respective mesylate ester 16 [74].Reaction of 16 with NaN3 in DMF gav corresponding azide 17 [74] in high yield (90%).Deprotection of the methoxy group obtain the desired analogue 18 was achieved by using BF3•S(CH3)2 complex in CH2Cl2 (Scheme 3).Following the synthesis of the desired alkynes (7, 8) and azides (13, 14, 17, and 1,2,3-triazole-substituted benzoxazines were prepared regioselectively through a Cu 1 alyzed click' cycloaddition reaction [75,76].The click reaction was carried out in the p ence of CuSO4•5H2O and sodium ascorbate (NaAsc) in a mixture of CH2Cl2/H2O (1:1) der conventional conditions or tBuOH/H2O (1:1) under microwave-assisted irradiatio shown in Schemes 4 and 5. Thus, the catecholic final product 19 (TC488) was obtaine the cycloaddition reaction of alkyne 7 and azide 14 at room temperature in 57% yiel turn, click reaction between alkyne 7 and the protected azide 17 at room temperatur forded the 1,2,3-triazolyl derivative 20 in 67% yield.Deprotection of the methoxy gro in 20 using BF3S(Me)2 also resulted in the removal of the p-methoxybenzyl group affor compound 22 (TC490) in 35% yield.In addition, we obtained the byproduct derivative (TC491) in 14% yield resulting from the migration of the p-hydroxybenzyl group from 4 to C6 of the 5,7,8-trimethyl-1,4-benzoxazine system.Thus, in order to obtain the de  In turn, click reaction between alkyne 7 and the protected azide 17 at room temperature afforded the 1,2,3-triazolyl derivative 20 in 67% yield.Deprotection of the methoxy groups in 20 using BF 3 S(Me) 2 also resulted in the removal of the p-methoxybenzyl group affording compound 22 (TC490) in 35% yield.In addition, we obtained the byproduct derivative 22a (TC491) in 14% yield resulting from the migration of the p-hydroxybenzyl group from N-4 to C6 of the 5,7,8-trimethyl-1,4-benzoxazine system.Thus, in order to obtain the deprotected resorcinol derivative 21 (TC489) bearing the p-methoxybenzyl group at N-4 for structure activity studies, we successfully performed the cycloaddition reaction with alkyne 7 and the deprotected azide 18 and TC489 were isolated in 67% yield (Scheme 4).
The synthesis of derivatives bearing a methoxymethyl-1,2,3-triazolyl spacer 24 (TC483) and 26 (TC484) were prepared from the click reaction of alkyne 8 and azide 13 or 17 to afford derivatives 23 and 25, respectively.Deprotection of the acetonide group in 23 using TFA afforded derivative 24 (TC483) in 81% yield.In turn, deprotection of the methoxy groups in 25 using BF 3 S(CH 3 ) 2 afforded the resorcinol deprotected derivative 26 (TC484) in 29% yield accompanied by traces of compound 26a in which only one methoxy group of the resorcinol moiety was deprotected (Scheme 5).The structural characterization of all the synthesized compounds was accomplished by 1 H and 13 C NMR spectroscopy and (high resolution) mass spectrometry.

Free-Radical Scavenging Activity of the Test Compounds
The six final compounds TC483, TC484, TC488, TC489, TC490, and TC491 were ini tially tested for their antioxidant capacity in a cell-free system, i.e., using the method o the free radical DPPH.As shown in Figure 2, four out of the six compounds were able to scavenge DPPH in a time-and concentration-dependent manner.The most active compounds seemed to be TC483 and TC488, which exhibited scavenging activity similar to the positive control (Trolox, 100 µM) at the highest concentration tested (100 µM).TC490 and TC491 were also effective DPPH scavengers, especially at 100 µM.The remaining two The structural characterization of all the synthesized compounds was accomplished by 1 H and 13 C NMR spectroscopy and (high resolution) mass spectrometry.

Free-Radical Scavenging Activity of the Test Compounds
The six final compounds TC483, TC484, TC488, TC489, TC490, and TC491 were initially tested for their antioxidant capacity in a cell-free system, i.e., using the method of the free radical DPPH.As shown in Figure 2, four out of the six compounds were able to scavenge DPPH in a time-and concentration-dependent manner.The most active compounds seemed to be TC483 and TC488, which exhibited scavenging activity similar to the positive control (Trolox, 100 µM) at the highest concentration tested (100 µM).TC490 and TC491 were also effective DPPH scavengers, especially at 100 µM.The remaining two compounds, i.e., TC484 and TC489 were marginally active at 100 µM and only following incubation for 24 and 3 h, respectively.compounds, i.e., TC484 and TC489 were marginally active at 100 µM and only follo incubation for 24 and 3 h, respectively.

Cytotoxic Activity of the Test Compounds
Then, the effect of the six compounds on the viability of human skin fibroblast assessed based on the widely accepted MTT method.As shown in Figure 3, all the pounds were not cytotoxic at 1 or 10 µM concentration.However, TC483, TC484, T and TC489 were cytotoxic at the highest concentration tested, i.e., 100 µM.On the hand, TC488 and TC489 increased cellular viability at 10 µM.However, testing via with an alternative method (see Supplementary Materials, Figure S19) indicated n ference from the vehicle.TC490 was not cytotoxic at any concentration tested, and thermore it mildly increased cellular viability.Finally, TC491 did not exhibit any st cally significant effect at any concentration tested.

Cytotoxic Activity of the Test Compounds
Then, the effect of the six compounds on the viability of human skin fibroblasts was assessed based on the widely accepted MTT method.As shown in Figure 3, all the compounds were not cytotoxic at 1 or 10 µM concentration.However, TC483, TC484, TC488, and TC489 were cytotoxic at the highest concentration tested, i.e., 100 µM.On the other hand, TC488 and TC489 increased cellular viability at 10 µM.However, testing viability with an alternative method (see Supplementary Materials, Figure S19) indicated no difference from the vehicle.TC490 was not cytotoxic at any concentration tested, and furthermore it mildly increased cellular viability.Finally, TC491 did not exhibit any statistically significant effect at any concentration tested.
Antioxidants 2024, 13, x FOR PEER REVIEW 13 of 24 compounds, i.e., TC484 and TC489 were marginally active at 100 µM and only following incubation for 24 and 3 h, respectively.

Cytotoxic Activity of the Test Compounds
Then, the effect of the six compounds on the viability of human skin fibroblasts was assessed based on the widely accepted MTT method.As shown in Figure 3, all the compounds were not cytotoxic at 1 or 10 µM concentration.However, TC483, TC484, TC488, and TC489 were cytotoxic at the highest concentration tested, i.e., 100 µM.On the other hand, TC488 and TC489 increased cellular viability at 10 µM.However, testing viability with an alternative method (see Supplementary Materials, Figure S19) indicated no difference from the vehicle.TC490 was not cytotoxic at any concentration tested, and furthermore it mildly increased cellular viability.Finally, TC491 did not exhibit any statistically significant effect at any concentration tested.

Intracellular Reactive Oxygen Species (ROS) Levels
Based on the above results, TC488 was considered as the most promising, since it was one of the most active radical scavengers (Figure 2) and at 10 µM it stimulated cell viability (Figure 3).Hence, it was further evaluated for its intracellular antioxidant activity by means of the DCFH-DA probe, which is internalized by the cells and upon oxidation by the cells' reactive oxygen species (ROS) yields the highly fluorescent DCF.As shown in Figure 4, TC488 at both 10 µM and 25 µM and at early time-points (up to 6 h) suppressed the intracellular ROS levels of human skin fibroblasts at a level comparable to that of the potent antioxidants NAC and Trolox.At later time-points, ROS levels tend to recover to control levels, especially in the presence of 10 µM TC488.
Figure 3. Cytotoxicity of the test compounds.Human skin fibroblasts were exposed to the indi concentrations of the test compounds for 72 h, and their viability was assessed with the method, as described under Materials and Methods.Average of three independent experim error-bars represent standard deviation (* p < 0.05; ** p < 0.01).

Intracellular Reactive Oxygen Species (ROS) Levels
Based on the above results, TC488 was considered as the most promising, since i one of the most active radical scavengers (Figure 2) and at 10 µM it stimulated cell viab (Figure 3).Hence, it was further evaluated for its intracellular antioxidant activit means of the DCFH-DA probe, which is internalized by the cells and upon oxidatio the cells' reactive oxygen species (ROS) yields the highly fluorescent DCF.As show Figure 4, TC488 at both 10 µM and 25 µM and at early time-points (up to 6 h) suppre the intracellular ROS levels of human skin fibroblasts at a level comparable to that o potent antioxidants NAC and Trolox.At later time-points, ROS levels tend to recov control levels, especially in the presence of 10 µM TC488.Interestingly, TC488 had similar effects on human skin fibroblasts that have rendered senescent through repeated mild doses of ionizing radiation (stress-ind premature senescence; SIPS).In Figure 5A, the validation of cell senescence through β-Gal staining is depicted.As shown in Figure 5B, despite the higher ROS levels tha characterizing senescent cells, TC488 had a dose-dependent antioxidant effect, inte ingly comparable to those of the positive control Trolox.Interestingly, TC488 had similar effects on human skin fibroblasts that have been rendered senescent through repeated mild doses of ionizing radiation (stress-induced premature senescence; SIPS).In Figure 5A, the validation of cell senescence through SAβ-Gal staining is depicted.As shown in Figure 5B, despite the higher ROS levels that are characterizing senescent cells, TC488 had a dose-dependent antioxidant effect, interestingly comparable to those of the positive control Trolox.
Beyond the suppression of ROS basal levels, TC488 was also capable of reverting hydrogen peroxide induction of ROS in both young and senescent human skin fibroblasts (Figure 6).Beyond the suppression of ROS basal levels, TC488 was also capable of reverting hydrogen peroxide induction of ROS in both young and senescent human skin fibroblasts (Figure 6).Besides TC488, analogue TC483 was also studied for its effects on the intracellular ROS levels of both young and senescent human skin fibroblasts, and it was found to sup- Beyond the suppression of ROS basal levels, TC488 was also capable of reverting hydrogen peroxide induction of ROS in both young and senescent human skin fibroblasts (Figure 6).Besides TC488, analogue TC483 was also studied for its effects on the intracellular ROS levels of both young and senescent human skin fibroblasts, and it was found to sup- Besides TC488, analogue TC483 was also studied for its effects on the intracellular ROS levels of both young and senescent human skin fibroblasts, and it was found to suppress them at a level comparable to that of the positive control Trolox, with a slight tendency for recovery at 24 h (Figure 7).TC483 (like TC488) was also capable of reverting H 2 O 2 -induction of ROS in both young and senescent human skin fibroblasts (Figure 8).
Compounds TC490 and TC491, although less effective than TC483 and TC488 as free radical scavengers (Figure 2) were also tested for their effect on intracellular ROS levels at a broader concentration range (since they were not cytotoxic at any concentration tested, see Figure 3).They were not effective in inhibiting basal ROS levels; however, TC490 at 10 µM was capable of a mild-yet statistically significant-inhibition of H 2 O 2 -stimulated ROS in both young and senescent skin fibroblasts, while it increased ROS levels at 100 µM (Figure 9, upper panel).Moreover, TC491 exhibited a dose-dependent mild inhibition of H 2 O 2 -stimulated oxidative stress, especially in the case of senescent cells, where all tested concentrations were active, while in young cells only the highest concentration had a statistically significant antioxidant effect (Figure 9, lower panel).
press them at a level comparable to that of the positive control Trolox, with a slight tendency for recovery at 24 h (Figure 7).TC483 (like TC488) was also capable of reverting H2O2-induction of ROS in both young and senescent human skin fibroblasts (Figure 8).press them at a level comparable to that of the positive control Trolox, with a slight tendency for recovery at 24 h (Figure 7).TC483 (like TC488) was also capable of reverting H2O2-induction of ROS in both young and senescent human skin fibroblasts (Figure 8).µM was capable of a mild-yet statistically significant-inhibition of H2O2-stimulated ROS in both young and senescent skin fibroblasts, while it increased ROS levels at 100 µM (Figure 9, upper panel).Moreover, TC491 exhibited a dose-dependent mild inhibition of H2O2-stimulated oxidative stress, especially in the case of senescent cells, where all tested concentrations were active, while in young cells only the highest concentration had a statistically significant antioxidant effect (Figure 9, lower panel).

Effect of the Test Compounds on Gene Expression
Since TC483 and TC488 were found to attenuate ROS levels, we studied their effect on the expression of heme oxygenase-1 (ho-1), a known target of the transcription factor Nrf2, the latter regulating the cellular antioxidant defense systems [77][78][79].As shown in Figure 10, a 24 h incubation with both 10 µM and 25 µM TC483 led to statistically significant induction of ho-1 gene expression in both young and senescent human skin fibroblasts.The induction was higher in young than in senescent cells.
TC488 on the other hand, was not effective at the concentration of 10 µM, while at 25 µM induced ho-1 gene expression only in young fibroblasts (Figure 10).
Since TC483 and TC488 were found to attenuate ROS levels, we studied their effect on the expression of heme oxygenase-1 (ho-1), a known target of the transcription factor Nrf2, the latter regulating the cellular antioxidant defense systems [77][78][79].As shown in Figure 10, a 24 h incubation with both 10 µΜ and 25 µM TC483 led to statistically significant induction of ho-1 gene expression in both young and senescent human skin fibroblasts.The induction was higher in young than in senescent cells.TC488 on the other hand, was not effective at the concentration of 10 µΜ, while at 25 µM induced ho-1 gene expression only in young fibroblasts (Figure 10).

Effect of the Test Compounds on Glutathione (GSH) Levels
The effects of TC483 and TC488 on cellular GSH levels were assessed by using the probe mCB.As shown in Figure 11, an overnight incubation with both compounds led to a dose-dependent increase of GSH levels.Hence, these two promising compounds seem to confer further antioxidant protection to human skin fibroblasts through GSH induction.

Effect of the Test Compounds on Glutathione (GSH) Levels
The effects of TC483 and TC488 on cellular GSH levels were assessed by using the probe mCB.As shown in Figure 11, an overnight incubation with both compounds led to a dose-dependent increase of GSH levels.Hence, these two promising compounds seem to confer further antioxidant protection to human skin fibroblasts through GSH induction.

Discussion
In the current work we report the synthesis and evaluation of the antioxidant activity of six novel hybrid compounds bearing the 5,7,8-trimethyl-1,4-benzoxazine core and a catechol (TC483 and TC488), or resorcinol group (TC484, TC489, TC490, and TC491) connected through a heteroaromatic linker, namely a 1,2,3-triazole.In compounds TC488, TC489, TC490, and TC491 the triazole ring is directly attached at C2 of the benzoxazine moiety, while in TC483 and TC484 the attachment is through a methoxymethyl spacer.Compounds TC483 and TC488 substituted by the ethyl-3,4-dihydroxyphenyl group can

Discussion
In the current work we report the synthesis and evaluation of the antioxidant activity of six novel hybrid compounds bearing the 5,7,8-trimethyl-1,4-benzoxazine core and a catechol (TC483 and TC488), or resorcinol group (TC484, TC489, TC490, and TC491) connected through a heteroaromatic linker, namely a 1,2,3-triazole.In compounds TC488, TC489, TC490, and TC491 the triazole ring is directly attached at C2 of the benzoxazine moiety, while in TC483 and TC484 the attachment is through a methoxymethyl spacer.Compounds TC483 and TC488 substituted by the ethyl-3,4-dihydroxyphenyl group can be envisioned as hybrids with hydroxytyrosol, the latter being known for its broad-spectrum biological properties.
Concerning the cell-free antioxidant capacity of the hybrids the catechol bearing analogues, TC483 and TC488 exhibited the highest scavenging activity, similar to Trolox.Conversely, the resorcinol compounds TC484 and TC489 were marginally active following incubation for 24 and 3 h, respectively.Concerning the effect of the hybrids on the viability of human skin fibroblasts, the methoxybenzyl analogues TC484, TC488, and TC489 were cytotoxic at the highest concentration of 100 µM.In contrast, the catechol compound TC488 and its resorcinol congener TC489 at 10 µM enhanced human skin fibroblast viability by approx.50% over the control, as judged by the MTT assay, but they did not show any effect based on the alternative Neutral Red assay (Figure S19).This may indicate an effect of the compounds on the mitochondrial enzymes responsible for MTT reduction without true effect on viability or a direct interaction between the compounds and MTT [80,81].Regarding the resorcinol analogues obtained following the removal of the methoxy benzyl group, TC490 exhibited a different activity profile, mildly supporting cell viability at all concentrations tested, while TC491 did not exhibit any statistically significant effect at any concentration.Interestingly, in TC491, the C6 of the benzoxazine moiety bears a 4hydroxybenzyl group which suggests that a free C6 position plays an important role in enhancing cell viability.The most promising hybrid from the cell-free antioxidant and cell viability tests (TC488) was found to suppress reactive oxygen species (ROS) at both 10 µM and 25 µM and at early time-points (up to 6 h) in both young and senescent human skin fibroblasts, at a level comparable to that of the positive controls NAC and Trolox.The methoxymethyl spacer-bearing analogue of TC488, compound TC483, was also capable of reducing intracellular ROS levels in both young and senescent human skin fibroblasts, at a level comparable to that of the positive control Trolox, with a slight tendency for recovery at 24 h.Among the less active derivatives TC490 and TC491, TC490 at 10 µM was active against intracellular ROS in both young and senescent skin fibroblasts, while TC491 was potent mainly in the case of senescent cells.These results show that deprotection of the 1,4-benzoxazine nitrogen is not beneficial for activity.The intracellular antioxidant activity of TC483 and TC488 was further confirmed by their ability to increase GSH levels in a dose-dependent manner.Since Nrf2 is the main transcription factor that regulates the expression of the cellular antioxidant defense system, TC483 and TC488 were assessed for their action on Nrf2's downstream effector, ho-1.TC483 resulted to an induction of ho-1 gene expression in both young and senescent human skin fibroblasts (at 10 µM and 25 µM) with preference for young cells, while TC488 led to statistically significant induction of the ho-1 gene expression in young human skin fibroblasts only at 25 µM.It should be noted here that other redox-sensitive transcription factors, such as activator protein-1 (AP-1), nuclear factor-kappa B (NF-kB), and hypoxia inducible factor (HIF), may also regulate ho-1; however, Nrf2 is being considered as its main transcriptional regulator [82,83].
Generally, the synthesized compounds have been shown to possess antioxidant activity for human skin fibroblast cultures, while no quantitative differences were observed between the response of young and senescent cells.Of course, a variety of stimuli, such as oxidative stress, ionizing radiation, ultraviolet radiation, etc., may lead to senescent cells with different characteristics.Hence, a different behavior of cells rendered senescent through other methods than the one used in the present study cannot be ruled-out.The current study is limited to cell-free and in vitro assay systems.The results presented here are promising; however, in vivo studies may be necessary to assess the safety and tolerance profile of the synthesized compounds prior to their further applications.

Figure 1 .
Figure 1.Design of the compounds of the present study.

Figure 1 .
Figure 1.Design of the compounds of the present study.

Figure 2 .
Figure 2. DPPH-scavenging activity of the test compounds.The indicated concentrations of th compounds were co-incubated with DPPH for the indicated time-intervals, and the absorbanc measured as described under Materials and Methods.Average of two independent experimen ror-bars represent standard deviation).

Figure 2 .
Figure 2. DPPH-scavenging activity of the test compounds.The indicated concentrations of the test compounds were co-incubated with DPPH for the indicated time-intervals, and the absorbance was measured as described under Materials and Methods.Average of two independent experiments (error-bars represent standard deviation).

Figure 2 .
Figure 2. DPPH-scavenging activity of the test compounds.The indicated concentrations of the test compounds were co-incubated with DPPH for the indicated time-intervals, and the absorbance was measured as described under Materials and Methods.Average of two independent experiments (error-bars represent standard deviation).

Figure 3 .
Figure 3. Cytotoxicity of the test compounds.Human skin fibroblasts were exposed to the indicated concentrations of the test compounds for 72 h, and their viability was assessed with the MTT method, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation (* p < 0.05; ** p < 0.01).

Figure 4 .
Figure 4. Intracellular ROS levels of human skin fibroblasts in the presence of TC488.Human fibroblasts were treated overnight with the indicated concentrations of TC488 or 5 mM NAC µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitor the indicated time-intervals, as described under Materials and Methods.Average of three indep ent experiments; error-bars represent standard deviation.

Figure 4 .
Figure 4. Intracellular ROS levels of human skin fibroblasts in the presence of TC488.Human skin fibroblasts were treated overnight with the indicated concentrations of TC488 or 5 mM NAC or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 5 .
Figure 5. Intracellular ROS levels of senescent human skin fibroblasts in the presence of TC488.(A) Representative photographs of human skin fibroblasts rendered senescent (as described under Materials and Methods) vs. their young counterparts, following their staining with SA-β-Gal (magnification 10×; scale bar=100 µm).(B) Senescent human skin fibroblasts were treated overnight with the indicated concentrations of TC488 or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 6 .
Figure 6.Reversal of oxidative-stress-induced ROS levels by TC488 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC488 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H2O2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 5 . 24 Figure 5 .
Figure 5. Intracellular ROS levels of senescent human skin fibroblasts in the presence of TC488.(A) Representative photographs of human skin fibroblasts rendered senescent (as described under Materials and Methods) vs. their young counterparts, following their staining with SA-β-Gal (magnification 10×; scale bar=100 µm).(B) Senescent human skin fibroblasts were treated overnight with the indicated concentrations of TC488 or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 6 .
Figure 6.Reversal of oxidative-stress-induced ROS levels by TC488 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC488 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H2O2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 6 .
Figure 6.Reversal of oxidative-stress-induced ROS levels by TC488 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC488 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H 2 O 2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 7 .
Figure 7. Intracellular ROS levels of young and senescent human skin fibroblasts in the presence of TC483.Cells were treated overnight with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 8 .
Figure 8. Reversal of oxidative-stress-induced ROS levels by TC483 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H2O2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 7 .
Figure 7. Intracellular ROS levels of young and senescent human skin fibroblasts in the presence of TC483.Cells were treated overnight with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 7 .
Figure 7. Intracellular ROS levels of young and senescent human skin fibroblasts in the presence of TC483.Cells were treated overnight with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA was added and fluorescence was monitored at the indicated time-intervals, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation.

Figure 8 .
Figure 8. Reversal of oxidative-stress-induced ROS levels by TC483 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H2O2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 8 .
Figure 8. Reversal of oxidative-stress-induced ROS levels by TC483 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of TC483 or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H 2 O 2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation.

Figure 9 .
Figure 9. Reversal of oxidative-stress-induced ROS levels by TC490 and TC491 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of the two compounds or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H2O2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation; asterisks denote statistically significant differences in comparison to control cells (Student's ttest, * p< 0.05, ** p < 0.01).

Figure 9 .
Figure 9. Reversal of oxidative-stress-induced ROS levels by TC490 and TC491 in young and senescent human skin fibroblasts.Human skin fibroblasts were pre-treated for 3 h with the indicated concentrations of the two compounds or 20 µM Trolox in serum-free medium, then DCFH-DA along with 1.5 mM H 2 O 2 was added and fluorescence was monitored after 1 h, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation; asterisks denote statistically significant differences in comparison to control cells (Student's t-test, * p< 0.05, ** p < 0.01).

Figure 10 .
Figure 10.HO-1 gene expression in young and senescent human skin fibroblasts following treatment with TC483 and TC488.Human skin fibroblasts were treated for 24 h with the indicated concentrations of the two compounds, then total mRNA was collected and gene expression studied, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation (** p < 0.01).

Figure 10 .
Figure 10.HO-1 gene expression in young and senescent human skin fibroblasts following treatment with TC483 and TC488.Human skin fibroblasts were treated for 24 h with the indicated concentrations of the two compounds, then total mRNA was collected and gene expression studied, as described under Materials and Methods.Average of three independent experiments; error-bars represent standard deviation (** p < 0.01).

Figure 11 .
Figure 11.GSH levels in human skin fibroblasts following treatment with TC483 and TC488.Human skin fibroblasts were treated overnight with the indicated concentrations of the two compounds or 10 mM NAC, then GSH levels were assessed using 5 µM mCB, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation (* p < 0.05; ** p < 0.01).

Figure 11 .
Figure 11.GSH levels in human skin fibroblasts following treatment with TC483 and TC488.Human skin fibroblasts were treated overnight with the indicated concentrations of the two compounds or 10 mM NAC, then GSH levels were assessed using 5 µM mCB, as described under Materials and Methods.Average of two independent experiments; error-bars represent standard deviation (* p < 0.05; ** p < 0.01).