Fiscalin Derivatives as Potential Neuroprotective Agents

Neurodegenerative diseases (ND) share common molecular/cellular mechanisms that contribute to their progression and pathogenesis. In this sense, we are here proposing new neuroprotection strategies by using marine-derived compounds as fiscalins. This work aims to evaluate the protective effects of fiscalin derivatives towards 1-methyl-4-phenylpyridinium (MPP+)- and iron (III)-induced cytotoxicity in differentiated SH-SY5Y cells, an in vitro disease model to study ND; and on P-glycoprotein (P-gp) transport activity, an efflux pump of drugs and neurotoxins. SH-SY5Y cells were simultaneously exposed to MPP+ or iron (III), and noncytotoxic concentrations of 18 fiscalin derivatives (0–25 μM), being the cytotoxic effect of both MPP+ and iron (III) evaluated 24 and 48 h after exposure. Fiscalins 1a and 1b showed a significant protective effect against MPP+-induced cytotoxicity and fiscalins 1b, 2b, 4 and 5 showed a protective effect against iron (III)-induced cytotoxicity. Fiscalins 4 and 5 caused a significant P-gp inhibition, while fiscalins 1c, 2a, 2b, 6 and 11 caused a modest increase in P-gp transport activity, thus suggesting a promising source of new P-gp inhibitors and activators, respectively. The obtained results highlight fiscalins with promising neuroprotective effects and with relevance for the synthesis of new derivatives for the treatment/prevention of ND.


Introduction
Neurodegenerative diseases (ND) are debilitating conditions that affect the central nervous system (CNS), mostly related to the aging of the world population. ND are characterized by the dysfunction and loss of the neuronal structure and function, resulting in uncontrolled neuronal death that leads to a progressive decline in brain functions [1][2][3]. These diseases are correlated with a wide spectrum of clinical symptoms, involving cognitive decline, memory loss and/or the impairment of motor functions [1,2,4,5]. Furthermore, the severity of the ND symptoms gradually progresses alongside the disease development, leading to a reduced capability for independent living. With the increase in the average lifespan, the prevalence of CNS diseases tends to grow, alongside an increased impact of the serious effects of such diseases on the quality of life as well as an increased burden on healthcare systems worldwide [1][2][3].
ND diseases, including Alzheimer's (AD) and Parkinson's diseases (PD), have a complex etiology and pathogenesis, which explains why the existing drugs available for their treatment have limited efficacy [1,3]. The study of ND, and particularly PD, have Therefore, the obtained results highlight several fiscalins with promising neuroprotective effects as well as with relevance to the synthesis of new derivatives to be potentially used in the treatment and/or prevention of ND, since fiscalins 1a and 1b showed a significant protective effect against MPP + -induced cytotoxicity and fiscalins 1b, 2b, 4 and 5 showed a significant protective effect against iron (III)-induced cytotoxicity. Moreover, fiscalins 4 and 5 lead to P-gp inhibition, while fiscalins 1c, 2a, 2b, 6 and 11 caused a modest but significant increase in P-gp transport activity, thus suggesting a promising source of new P-gp inhibitors and activators, respectively.
A 50 mM stock solution of each fiscalin derivative ( Figure 1) was prepared in DMSO and all stock solutions were stored at −20 • C and freshly diluted on the day of the experiment, in fresh cell culture medium (ensuring that DMSO concentration did not exceed 0.1% of the exposure media).

SH-SY5Y Cell Culture and Differentiation
SH-SY5Y cells (ATCC, United States of America) were routinely cultured in 25 cm 2 flasks using DMEM with 4.5 g/L glucose, supplemented with 10% heat-inactivated FBS, 1% antibiotic mixture (100 U/mL of penicillin and 100 µg/mL of streptomycin) and 1% NEAA. Cells were maintained in a 5% CO 2 -95% air atmosphere, at 37 • C, and the medium was changed every 2-3 days. When 80-90% confluence was reached, the cell cultures were passaged by trypsinization (0.25% trypsin/1 mM EDTA). In all experiments, the cells were seeded in 96-or 24-well plates at a density of 25,000 cells/cm 2 and used 6 days after seeding. In order to obtain cells with a dopaminergic neuronal phenotype, SH-SY5Y cells were differentiated as previously described [41,42]. Briefly, SH-SY5Y cells were seeded in complete DMEM medium containing 10 µM RA, and cultured for 3 days at 37 • C. After 3 days, 80 nM of TPA was added to the cultures, and cells were cultured for another 3 days at 37 • C. The cells used in all experiments were taken between the 21st and 27th passages.

Fiscalin Cytotoxicity
Fiscalin cytotoxicity was evaluated in differentiated SH-SY5Y cells, by the neutral red (NR) uptake, resazurin (REZ) reduction and sulforhodamine B (SRB) binding assays, 24 and 48 h after exposure. For that purpose, SH-SY5Y cells were seeded in 96-well plates at a density of 25,000 cells/cm 2 and differentiated through the RA and TPA differentiation protocol. Six days after seeding, differentiated SH-SY5Y cells were exposed to fiscalins (0-50 µM), prepared in fresh cell culture medium. Triton™ X-100 (0.1%) was used as positive control for all the cytotoxicity experiments. At least four independent experiments were performed, in triplicate.

Neutral Red Uptake Assay
The NR uptake assay is based on the ability of viable cells to incorporate and bind the NR dye in the lysosomes, providing a quantitative analysis of the number of viable cells in culture [43,44]. At 24 or 48 h after exposure to the tested fiscalins (1a, 1b, 1c, 2a, 2b and 3-15), the cell culture medium was removed and the cells were washed with HBSS (+/+), followed by the addition of fresh cell culture medium containing 50 µg/mL NR, and incubation, at 37 • C, in a humidified 5% CO 2 -95% air atmosphere, for 60 min. After 60 min incubation with NR, the cell culture medium was removed, and the dye that was only absorbed by viable cells extracted with a lysis buffer [absolute ethyl alcohol/distilled water (1:1) with 5% acetic acid]. The absorbance was then measured at 540 nm, in a multiwell plate reader (PowerWaveX BioTek Instruments, Winooski, VT, USA). The results were expressed as percentage of NR uptake relatively to that of the control cells (0 µM). At least four independent experiments were performed, in triplicate.

Resazurin Reduction Assay
The REZ reduction assay, also known as alamarBlue assay, is based on the conversion of the blue nonfluorescent dye resazurin to the pink fluorescent resorufin by mitochon-drial reductases and other cytoplasmic enzymes, such as diaphorases. Consequently, the quantity of the produced resorufin is related to the number of metabolically viable cells in culture [43,45]. After exposure to the tested fiscalins (1a, 1b, 1c, 2a, 2b and 3-15) for 24 or 48 h, the cell culture medium was removed and the cells were washed with HBSS (+/+), followed by the addition of fresh cell culture medium containing 10 µg/mL REZ. The cells were then incubated, at 37 • C, in a humidified 5% CO 2 -95% air atmosphere, for 60 min. The fluorescence was read afterwards in a multiwell plate reader (PowerWaveX BioTek Instruments, Winooski, VT, USA), using an excitation and emission wavelengths of 560 nm and 590 nm, respectively. The results were expressed as percentage of REZ reduction relative to that of the control cells (0 µM). At least four independent experiments were performed, in triplicate.

Sulforhodamine B Binding Assay
The sulforhodamine B (SRB) protein-staining assay is an in vitro cytotoxicity assay developed to measure all the cellular protein content. This assay is based on the dye's ability to bind to basic amino acids of cellular proteins under mild acidic conditions. Consequently, the total protein mass, which is related to the number of cells in culture, can be determined by spectrophotometry [43,46]. After exposure to the tested fiscalins (1a, 1b, 1c, 2a, 2b and 3-15) for 24 or 48 h, the cell culture medium was removed, the cells were washed with HBSS (+/+) and fixed overnight, at −20 • C, with a methanolic solution of 1% acetic acid (v/v). The fixing medium was then removed, replaced by 0.05% SRB solution (prepared in 1% acetic acid), followed by incubation, at 37 • C, for 60 min. After incubation, the SRB solution was removed, and the cells were washed with 1% acetic acid (v/v) to remove the unbound dye. The plates were left to air-dry at room temperature and the bound SRB was subsequently extracted with a Tris base solution (10 mM, pH 10.5). The absorbance was measured, at 540 nm, in a multiwell plate reader (PowerWaveX BioTek Instruments, Winooski, VT, USA). The results were expressed as percentage of SRB binding relative to that of the control cells (0 µM). At least four independent experiments were performed, in triplicate.

P-Glycoprotein Modulation Studies
Changes in the expression and/or activity of ABC transporters at the blood-brain barrier (BBB) have been associated with many neurological diseases, as well as with the pharmacoresistance to central-nervous-system (CNS) acting drugs [22]. The modulation of the activity of ABC transporters, including P-gp, can be identified through the intracellular accumulation of a probe substrate in the presence of a test compound [47]. Consequently, an increased intracellular accumulation of a probe substrate in the presence of a given compound indicates that the test compound is an inhibitor of the efflux transporter [48,49]. On the opposite, a decreased intracellular accumulation of a probe substrate in the presence of a test compound suggests its potential for the efflux transporter activation [50][51][52]. A fluorescent P-gp substrate widely used in accumulation assays is rhodamine 123 (RHO 123), which allows for the direct evaluation of P-gp activity by measuring RHO 123 intracellular fluorescence [47,[51][52][53][54][55][56]. Accordingly, in the present work, the fiscalins' effect on P-gp transport activity was evaluated in a multiwell plate reader, by the RHO 123 accumulation assay, using RHO 123 (10 µM) as a P-gp fluorescent substrate, and ZOS (5 µM), as a specific third-generation P-gp inhibitor. Although the RHO 123 and ZOS concentrations were selected based in previous studies performed by our group, those studies were performed in different cell lines (Caco-2, SW480 and RBE4 cells [51][52][53][54]56,57]). Therefore, preceding the evaluation of the effects of fiscalin derivatives on P-gp activity, the cytotoxicity of both RHO 123 (0-10 µM) and ZOS (0-5 µM) was evaluated in differentiated SH-SY5Y cells, by the REZ reduction, NR uptake, and SRB binding assays, upon 24 h of exposure. Only the 24 h timepoint was selected, since in most of the accumulation assays performed to assess the potential modulatory effect of new compounds on P-gp activity, shorter periods of incubation with both the P-gp substrate and inhibitor are required [51][52][53][54][56][57][58]. These cytotoxicity results are present in the Supplementary Materials, available on the journal page ( Figure S1). In addition, and although the experimental conditions used to evaluated the fiscalins' effects on P-gp activity were based on previous studies [51,[53][54][55][56][57], the RHO 123 concentration, as well as the duration of the incubation with the P-gp fluorescent substrate, were optimized using differentiated SH-SY5Y cells. Furthermore, these optimization results are available in the Supplementary Materials ( Figure S2).

RHO 123 Accumulation Assay in the Presence of Fiscalins
For the in vitro evaluation of the effects of fiscalins on P-gp transport activity, SH-SY5Y cells were seeded in 24-well plates at a density of 25,000 cells/cm 2 , following the differentiation protocol with RA and TPA, as previously mentioned. Six days after seeding, the cell culture medium was removed, and the cells were exposed to noncytotoxic concentrations of fiscalins 1a, 1b, 1c, 2a, 2b and 3-11 (0-25 µM), in the presence and in the absence of the P-gp inhibitor (ZOS). Briefly, after differentiation, SH-SY5Y cells were submitted to the following procedures:

2.
RHO 123 accumulation in the presence of the P-gp inhibitor (inhibited accumulation, IA): differentiated SH-SY5Y cells were simultaneously exposed to the tested fiscalins and to the specific P-gp inhibitor, ZOS (5 µM), both prepared in HBSS (+/+). After 30 min, RHO 123 (10 µM), prepared in HBSS (+/+), was added, and the cells were further incubated for 90 min, at 37 • C, in a humidified 5% CO 2 -95% air atmosphere. Control cells were only exposed to the P-gp fluorescent substrate and to the P-gp inhibitor.
Each of the previously mentioned exposure conditions (NA and IA, in the presence or in the absence of the tested compounds) were performed in triplicate. After the incubation period, the cells were washed twice with HBSS (+/+) and lysed with 0.1% Triton™ X-100 for 30 min at room temperature and in the absence of light. After lysis, the intracellular RHO 123 fluorescence was measured in a multiwell plate reader (PowerWave-X, BioTek Instruments, Winooski, VT, USA) at excitation/emission wavelengths of 485/528 nm, respectively, and expressed as fluorescence intensity (FI). Four independent experiments were performed, in triplicate. P-gp activity was then evaluated through the ratio between the FI after accumulation of RHO 123 in the presence of the P-gp inhibitor (IA) and the FI of the accumulation of RHO 123 in the absence of the P-gp inhibitor (NA) (Equation (1)): The results were expressed as percentage of control cells (0 µM fiscalins). When the P-gp activity increases, the amount of RHO 123 that is pumped out of the cell is higher, and consequently the fluorescence intensity decreases as a consequence of the decrease in the RHO 123 intracellular content. Therefore, as the fluorescent substrate is pumped out of the cells during the accumulation period due to the pump activation, the FI NA will decrease, and the FI IA /FI NA ratio will be higher. Conversely, when P-gp is inhibited, its activity is reduced, leading to an increased intracellular RHO 123 accumulation, and therefore, to a higher fluorescence intensity. Consequently, the ratio FI IA /FI NA will be lower, since a higher fluorescence intensity under normal conditions (FI NA ) will be detected, because of the decreased RHO 123 efflux.

Evaluation of Fiscalins' Neuroprotective Effects
To evaluate the potential neuroprotective effects of the fiscalin derivatives under study, differentiated SH-SY5Y cells were exposed to two different stressors: MPP + -iodide, the active metabolite of the dopaminergic neurotoxin MPTP and iron (III), through ferric nitrilotriacetate reagent (FeNTA).
2.6.1. Fiscalins' Protective Effects against MPP + -Induced Cytotoxicity SH-SY5Y cells were seeded at a density of 25,000 cells/cm 2 into 96-well plates, following the RA and TPA differentiation protocol previously mentioned. Six days after seeding, the cells were exposed to fiscalins (0-25 µM) prepared in fresh cell culture medium. After 30 min of preincubation, MPP + (0-1500 µM) was added and the cells were then incubated at 37 • C in a humidified 5% CO 2 -95% air atmosphere, for 24 or 48 h. Following the incubation period, MPP + cytotoxicity was evaluated by the NR uptake assay, as described above, since it was previously shown to be the most sensitive method to detect MPP + -induced cytotoxicity (data not shown). At least four independent experiments were performed, in triplicate. For each experiment, a fresh stock solution of MPP + (10 mM) was prepared (protected from light) in PBS (+/+), diluted afterwards in fresh cell culture medium and used immediately.

Fiscalins' Protective Effects against Iron (III)-Induced Cytotoxicity
To evaluate the potential neuroprotective effects of the fiscalin derivatives under study against iron (III)-induced cytotoxicity, differentiated SH-SY5Y were exposed to ferric nitrilotriacetate (FeNTA), a ferric (Fe 3+ ) iron aggressor. Initially, the NTA (250 mM, pH 7.4) solution was prepared by combining a nitrilotriacetic acid disodium salt (Na 2 NTA) solution (100 mM) and a nitrilotriacetic acid trisodium salt (Na 3 NTA) solution (100 mM) until pH = 7 was achieved. Iron (III) chloride was then added to the NTA to obtain the FeNTA solution.
A fresh stock solution of FeNTA (100 mM) was prepared for each experiment (protected from light) and was firstly incubated for 15 min before use to ensure the complete formation of the ferric oxidation form before addition to the cell cultures, and then diluted in fresh cell culture medium [17,59]. SH-SY5Y cells were seeded at a density of 25,000 cells/cm 2 into 96-well plates, following the RA and TPA differentiation protocol previously mentioned. Six days after seeding, the cells were exposed to fiscalins (0-25 µM) prepared in fresh cell culture medium. After 30 min of pre-incubation, FeNTA (0-1500 µM) was added, and the cells were then incubated, at 37 • C, in a humidified 5% CO 2 -95% air atmosphere, for 24 or 48 h. Following the incubation period, FeNTA cytotoxicity was evaluated by the NR uptake assay, as described in the previous sections. At least four independent experiments were performed, in triplicate.

Statistical Analysis
GraphPad Prism 8 for Windows (GraphPad Software, San Diego, CA, USA) was used to perform all statistical calculations. Three tests were performed to evaluate the normality of the data distribution: the KS normality test, D'Agostino and Pearson omnibus normality test and Shapiro-Wilk normality test. For data with parametric distribution, a one-way ANOVA was used to perform the statistical comparisons, followed by Dunnett's multiple comparisons test. Statistical comparisons between groups in experiments with two variables were made using two-way ANOVA, followed by Holm-Šídák's multiple comparisons test. Details of the performed statistical analysis are described in the figure legend. In all cases, differences were considered to be significant for p values lower than 0.05.

Fiscalin Cytotoxicity
The cytotoxicity of 18 fiscalins, 1a, 1b, 1c, 2a, 2b, 3 to 15 (0-50 µM), were evaluated in differentiated SH-SY5Y cells, by the NR uptake, REZ reduction and SRB binding assays, 24 and 48 h after exposure, in order to select the nonharmful fiscalins and their respective noncytotoxic concentrations to be further used in the subsequent neuroprotection and P-gp modulation studies.
Similar results were detected in the REZ reduction (Supplementary Materials, Figure S3) and in the SRB binding (Supplementary Materials, Figure S4) assays, with a concentrationand time-dependent cytotoxic effect being observed for several of the tested compounds, comparatively to the control cells (0 µM).
By the overall analysis of the cytotoxicity assays' results, the SRB binding assay proved to be the less sensitive method, while the NR uptake assay was shown to be the most sensitive, with this one being further used in the neuroprotection studies. Importantly, compounds 12, 13, 14 and 15 were the most cytotoxic among the tested fiscalins, with a substantial and significant decrease in cell viability, as detected by all three cytotoxicity assays. Considering their high cytotoxicity, these four compounds were excluded from the subsequent studies.
The remaining compounds showed no significant cytotoxicity for concentrations below or equal to 25 µM, being the concentrations 10 and 25 µM the selected ones to be further used in the neuroprotection and P-gp modulation studies. However, compounds 4, 5, 10 and 11 demonstrated to be cytotoxic for concentrations equal or above 25 µM, particularly for the 48 h exposure timepoint. In this sense, for fiscalins 4 and 5, upon 24 h of exposure, and for compounds 4, 5, 10 and 11 upon 48 h of exposure, the concentration selected was only 10 µM. Noteworthy, the noncytotoxic concentrations selected (10 and 25 µM) have already been reported for in vitro studies with similar fiscalin derivatives [30,32,34,38,60].

P-Glycoprotein Modulation Studies
P-gp activity was evaluated in differentiated SH-SY5Y cells after incubation of the tested fiscalin derivatives, 1a, 1b, 1c, 2a, 2b, 3-11 (0-25 µM) in the presence and absence of ZOS (5 µM), by performing a RHO 123 accumulation for a period of 90 min, aiming to evaluate the potential immediate effects of the fiscalin derivatives on P-gp transport activity as a result of a direct inhibition or activation of this pump. Results are presented as mean ± SD from at least four independent experiments, performed in triplicate. Statistical comparisons were made using two-way ANOVA, followed by Tukey's multiple comparisons test (at each timepoint, for comparisons between concentrations) or by Sidak's multiple comparisons test (for each concentration, for comparisons between timepoints). In all cases, p values < 0.05 were considered significant [* p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001 versus control (0 µM), at each timepoint; $ p < 0.05; $$ p < 0.01; $$$ p < 0.001; $$$$ p < 0. According to the obtained results, no significant effect on P-gp activity was observed for fiscalins 1a, 1b, 9 and 10, at any of the tested concentration and when compared to control cells (0 µM) (Supplementary Materials, Figure S5). Fiscalins 2a and 11 were shown to slightly but significantly increase P-gp activity at the 10 µM concentration (110.2%, p < 0.05, and 115.5%, p < 0.01, after exposure to 10 µM of compounds 2a and 11, respectively), while fiscalin 6 caused a slight but significant (p < 0.05) increase in P-gp activity at 5 µM (114.3%, when compared to control cells). In addition, fiscalins 1c and 2b slightly but significantly increased P-gp activity at 5 and 10 µM (P-gp transport activity significantly increased to 109.4%, p < 0.05, and 115.3%, p < 0.001, and to 117.0%, p < 0.01, and 112.3%, p < 0.05, after exposure to 5 and 10 µM of fiscalins 1c and 2b, respectively). These results, as shown in Figure 3a, although not significant for all the tested concentrations (0-25 µM), indicate a P-gp activation effect.

MPP + as an Agent for Chemical-Induced Cytotoxicity-Evaluation of Fiscalins' Protective Effects
To evaluate the potential protective effect of fiscalins towards MPP + -induced cytotoxicity, differentiated SH-SY5Y cells were simultaneously exposed to MPP + (0-1500 µM) and to fiscalins 1a, 1b, 1c, 2a, 2b, 3-11. The cytotoxic effect of MPP + , in the presence and absence of the fiscalins, was further evaluated 24 and 48 h after exposure by the NR uptake assay. According to the cytotoxicity assays, fiscalins 1a, 1b, 1c, 2a, 2b, 3, 6, 7, 8 and 9 were tested at 10 and 25 µM for 24 and 48 h of exposure. For both timepoints, fiscalins 4 and 5 were only tested at 10 µM, given the observed cytotoxicity towards differentiated SH-SY5Y cells. For the same reason, fiscalins 10 and 11, due to their cytotoxicity profile, were tested at 10 and 25 µM in the studies involving 24 h of exposure to MPP + , and only at 10 µM for studies with 48 h of exposure.
For fiscalin 1b, at 10 µM, a minimal but significant (p < 0.05) increase in the NR uptake was observed for the 1000 µM concentration of MPP + , when compared to MPP + alone, indicating the capacity for this fiscalin to also reduce MPP + -induced cytotoxicity. However, fiscalin 1b did not retain the significant protective effect 48 h after exposure to MPP + .
For fiscalin 2a, at 25 µM, a significant decrease in the NR uptake was observed for concentrations equal or above 500 µM of MPP + (NR uptake significantly decreased to 77.6%, p < 0.05, 71.1%, p < 0.05, and 59.3%, p < 0.0001, 24 h after exposure to 500, 1000 and 1500 µM MPP + in the presence of 25 µM compound 2a, when compared to 85.1%, 78.4% and 75.3% for 500, 100 and 1500 µM MPP + alone). A significant reduction (p < 0.01) in the NR uptake was also detected 24 h after exposure for 1500 µM of MPP + in the presence of compound 2a at 10 µM and when compared to MPP + alone, as shown in Figure 4a. In a similar way, 24 h after exposure, the MPP + cytotoxic effect was increased in the presence of fiscalin 6 (NR uptake significantly decreased to 75.2%, p < 0.01, 66.9%, p < 0.001, and 57.1%, p < 0.0001, 24 h after exposure to 500, 1000 and 1500 µM of MPP + in the presence of 25 µM of compound 6, when compared to 85.3%, 79.0% and 75.5% for 500, 100 and 1500 µM of MPP + alone). Furthermore, for the highest concentration tested of MPP + , a significant reduction (p < 0.01) in the NR uptake was also detected in the presence of compound 6 at 10 µM (65.0%), and when compared to MPP + alone (75.5%) (Figure 4a). Nevertheless, fiscalins 2a and 6 showed a conflicting behavior towards MPP + -induced cytotoxicity in differentiated SH-SY5Y cells, after 48 h of exposure (Figure 4b). In fact, a significant reduction in the NR uptake was observed for the 1500 µM of MPP + in the presence of fiscalins 2a and 6 at 10 and 25 µM. However, in the presence of fiscalins 2a and 6 at 25 µM, a slight but significant increase in the NR uptake was observed for the 250 µM MPP + concentration (NR uptake significantly increased to 84.7%, p < 0.05, 48 h after exposure to 250 µM of MPP + in the presence of 25 µM fiscalin 2a, when compared to 78.7% for MPP + alone; and to 80.2%, p < 0.01, 48 h after exposure to 250 µM MPP + in the presence of 25 µM of fiscalin 6, when compared to 72.1% for MPP + alone).
On the other hand, for fiscalins 1c and 7, at 10 and 25 µM, no significant differences on the NR uptake were detected for any of the tested concentrations of MPP + at either timepoint (Supplementary Materials, Figures S6 and S7). These results suggest that these compounds neither have a protective effect towards MPP + -induced cytotoxicity nor further increase its cytotoxic effect in differentiated SH-SY5Y cells. For fiscalins 2b, 4 and 5, at both tested concentrations, no significant differences on the NR uptake were detected for any of the MPP + concentrations at 24 h of exposure (Supplementary Materials, Figure S6), but after 48 h, these fiscalins were found to significantly increase MPP + -induced cytotoxicity towards differentiated SH-SY5Y cells for concentrations equal or above 1000 µM of MPP + (Supplementary Materials, Figure S7). Moreover, fiscalins 3, 8, 9, 10 and 11 were found to increase MPP + -induced cytotoxicity towards differentiated SH-SY5Y cells, 24 (Supplementary Materials, Figure S6) and 48 h (Supplementary Materials, Figure S7) after exposure, particularly for fiscalins 10 and 11, whose MPP + -induced cytotoxicity was significantly higher in their presence, at noncytotoxic concentrations, with the cytotoxic effect being significantly higher for 25 µM of these two compounds, and when compared to 10 µM, showing a concentration-dependent worsening of the MPP + cytotoxic effect.  fiscalins (1a, 1b, 2a and 6) (10 and 25 µM), evaluated by the neutral red uptake assay, 24 (a) and 48 h (b) after exposure. Results are presented as mean ± SD from 4 independent experiments, performed in triplicate. Statistical comparisons were made using two-way ANOVA, followed by Holm-Šídák's multiple comparisons test. In all cases, p values < 0.05 were considered significant [* p < 0.05; ** p < 0.001; *** p < 0.001; **** p < 0.0001 versus MPP + alone; # p < 0.05; ## p < 0.001 at each MPP + concentration for comparisons between the

FeNTA as an Agent for Iron-Induced Cytotoxicity-Evaluation of Fiscalins' Protective Effects
Iron cytotoxic effect was evaluated in differentiated SH-SY5Y cells by the NR uptake assay, upon 24 and 48 h of exposure to FeNTA (0-1500 µM), in the presence of fiscalins 1a,  1b, 1c, 2a, 2b, 3-11. Again, and accordingly to the previous cytotoxicity assays performed in this work, fiscalins 1a, 1b, 1c, 2a, 2b, 3, 6, 7, 8, and 9 were tested at 10 and 25 µM, for 24 and 48 h of exposure, given the lack of significant cytotoxicity towards differentiated SH-SY5Y cells. However, for both timepoints, fiscalins 4 and 5 were only tested at 10 µM, given the observed cytotoxicity towards differentiated SH-SY5Y cells observed at higher concentrations. For the same reason, fiscalins 10 and 11 were tested at 10 and 25 µM in the studies involving 24 h of exposure to FeNTA, and only at 10 µM for studies with 48 h of exposure.
As observed along with MPP + -induced cytotoxicity, three different patterns of fiscalins' effects were observed after 24 h of exposure to FeNTA: fiscalins 1b, 2b, 4 and 5 had a protective effect for some of the concentrations tested ( Figure 5a); fiscalins 1a, 1c, 2a, 3, 6, 7, 9, 10 and 11 did not affect the NR uptake at any tested concentration (data showed in Supplementary Materials, Figure S8); and compound 8 was found to significantly increase FeNTA-induced cytotoxicity, at both tested concentrations ( Figure S8).
In the presence of fiscalin 2b, at 10 µM, a significant increase (p < 0.01) in the NR uptake was observed only for 1500 µM FeNTA, and when compared to FeNTA alone. Additionally, a significant increase in the NR uptake was observed for all tested FeNTA concentrations in the presence of fiscalin 2b at 25 µM, and when compared to FeNTA alone, 24 h after exposure, as shown in Figure 5a (101.9%, p < 0.05, 99.5%, p < 0.01, 89.0%, p < 0.001, and 72.6%, p < 0.001, for 250, 500, 1000 and 1500 µM of FeNTA in the presence of 25 µM of fiscalin 2b, when compared to 93.8%, 91.0%, 78.8% and 62.4% observed for 250, 500, 1000 and 1500 µM of FeNTA alone). However, 48 h after exposure to FeNTA, in the presence of fiscalin 2b, the protective effect completely disappeared, and the FeNTA cytotoxic effect was slightly increased in the presence of 25 µM of fiscalin 2b, as shown in Figure 5b.
In the presence of fiscalin 4, at 10 µM, a significant increase (p < 0.01) in the NR uptake was observed for the 1500 µM of FeNTA (65.0%), as shown in Figure 5a, and when compared to FeNTA alone (55.4%). Moreover, upon 48 h of exposure to FeNTA in the presence of fiscalin 4 at 10 µM, a significant increase in the NR uptake was also observed for 250, 500 and 1000 µM of FeNTA, when compared to FeNTA alone. In fact, NR uptake slightly but significantly increased to 90.9% (p < 0.05), 80.5% (p < 0.05) and 65.6% (p < 0.01), 48 h after exposure to 250, 500 and 1000 µM of FeNTA in the presence of 10 µM of fiscalin 4, and when compared to 83.3%, 71.8% and 56.6% observed for 250, 500 and 1000 µM of FeNTA alone (Figure 5b). Additionally, 24 h after exposure (Figure 5a), a significant increase in the NR uptake (p < 0.05) was observed for 1000 µM of FeNTA in the presence of fiscalin 5 at 10 µM, and when compared to FeNTA alone. However, 48 h after exposure, no protective or increased cytotoxic effect was observed for FeNTA in the presence of fiscalin 5 (Figure 5b).

Discussion
Marine-derived compounds, including fiscalins, present important biological activities that might provide valuable tools in drug discovery and development, and are also considered promising alternatives, particularly for the treatment of neurodegenerative diseases [15,30]. Therefore, with this work we aimed at discovering new fiscalin derivatives with promising neuroprotective properties. To select noncytotoxic working concentrations, differentiated SH-SY5Y cells were initially exposed to the 18 fiscalin derivatives and their cytotoxicity were assessed by the NR uptake, REZ reduction and SRB binding assays, 24 and 48 h after exposure. According to the obtained results, and given the lack of significant effects, concentrations equal to or below 25 µM were selected as noncytotoxic working concentrations to be used in the neuroprotection and P-gp modulation studies. Of all the cell viability methods used, the SRB binding assay proved to be the less sensitive method, while the NR uptake assay showed to be the most sensitive, with this one being further used in neuroprotection studies. Given their cytotoxicity profile, fiscalins 12, 13, 14 and 15 were excluded from further experiments. Considering structure-cytotoxicity relationships, it was possible to assess that the presence of halogens in the fiscalin scaffold ( Figure 1) clearly generates derivatives with higher cytotoxicity (which is the case of these compounds).
One of the known hallmarks of AD is the accumulation of Aβ peptides and subsequent formation of senile plaques, leading to neuronal degeneration. An imbalance between the production of Aβ peptides and their elimination leads to Aβ peptide accumulation. Within the BBB, P-gp, as one of the transporters responsible for the Aβ peptides' elimination, has an important role in Aβ detoxification, and consequently in the pathogenesis of AD [61,62]. In vivo studies have suggested an increased Aβ accumulation, in the presence of P-gp inhibitors, a consequence of the decrease in P-gp activity [63]. Additionally, Aβ accumulation was shown to be reduced in the presence of P-gp inducers [64]. In the present work, some fiscalin derivatives, namely fiscalins 1c, 2a, 2b, 6, and 11, showed a modest but significant increase in P-gp activity. Despite the short incubation period, the activation of this pump mediated by these compounds demonstrates their potential for modulating P-gp transport activity, and thus for increasing Aβ efflux, suggesting a promising source of new P-gp activators, with potential to increase Aβ detoxification, in the scope of AD treatment and/or prevention.
Despite the role of P-gp in Aβ detoxification, this transporter is also known to prevent the accumulation of therapeutic drugs within the brain, by actively pumping these drugs back into the blood stream, limiting their therapeutic efficacy [54,65]. For instance, studies have shown that P-gp inhibitors may increase the therapeutic efficacy of anticancer drugs. In fact, in vitro anticancer activity of doxorubicin was improved when in the presence of inhibitors of this efflux pump, some of which were marine-derived compounds, including fiscalin derivatives [37]. Moreover, some commonly used PD drugs, such as levodopa or dopamine agonists, are known to be P-gp substrates [66,67]. Therefore, P-gp inhibitors may ultimately have potential to improve pharmacological treatments for CNS diseases. Overall, regarding the present work, the results obtained for fiscalins 3, 7, 8, and, particularly, 4 and 5 clearly demonstrate the inhibitory ability of these compounds towards the P-gp transport activity, therefore suggesting a promising source of new inhibitors, with potential to be used in CNS diseases, including ND, as adjuvants for drugs that are substrates of this efflux pump, possibly increasing their therapeutic efficacy.
Given the structure-activity relationship of these compounds, although compounds 1a, 1b and 1c present the same structure, the different stereochemistry revealed to be important for the modulation of P-gp activity. For instance, compound 1c with R stereochemistry at C-4 (ring C, Figure 1), was able to activate the P-gp, even though compounds 1a and 1b, both with S stereochemistry at C-4, showed no activity on this efflux pump. Additionally, the insertion of a chloro substituent in the anthranilic portion in fiscalin 1c (fiscalin 10) led to an absence of P-gp modulatory activity. Differently, compounds 2a and 2b, also stereoisomers, were both shown to activate P-gp, probably due to the presence of an isobutyl group in position 1, instead of an isopropyl group, which is the case of fiscalins 1a, 1b and 1c. Moreover, and taking into account fiscalin 2b, by adding one hydroxyl group (fiscalin 6) or two chloro substituents (fiscalin 11) in the aromatic ring of the anthranilic portion of its structure (ring A, Figure 1), no modification on P-gp activation was observed. Furthermore, fiscalins 3-5, which have larger substituents at C-1, showed to inhibit P-gp. Particularly for fiscalins 4 and 5, both with a phenolic ring in this position, which presented a significant concentration-dependent P-gp inhibition. In fact, given the structure-activity relationships, the presence of a phenolic ring attached to the fiscalin scaffold, adding to their long carbonated chain, could be related to the P-gp inhibitory effect of fiscalins 4 and 5, both with R stereochemistry at C-4. Additionally, compounds 7 and 8, compounds with larger substituents in the aromatic ring of the anthranilic portion of the fiscalin scaffold (ring A, Figure 1), showed to also inhibit P-gp, although compounds with smaller substituents in this position (-OH, compound 6) were able to activate this efflux pump.
The loss of dopaminergic neurons within the substantia nigra is one of the pathological hallmarks of PD. This disease pathogenesis has been experimentally replicated, in several in vivo and in vitro models, by using the neurotoxin 1-methyl-4-phenylpyridinium (MPP + ), the active metabolite of 1-methyl-4-phenyl-2,3,6-tetrahydropyridine (MPTP). This toxic compound is capable of severely damaging neuronal cells, inducing cell death, leading to a syndrome that closely resembles PD [6][7][8]68]. MPP + is responsible for the inhibition of complex I of the mitochondrial electron transport chain, leading to reactive oxygen species (ROS) production and ATP depletion, altering the mitochondrial membrane potential, resulting in apoptosis and dopaminergic neuron death [7,8,[68][69][70]. Furthermore, MPP + -induced dopaminergic cytotoxicity has been commonly used to investigate potential neuroprotective agents, in vitro and in vivo [69].
Several natural compounds, including marine-derived compounds, have shown promising neuroprotective properties towards MPP + -induced cytotoxicity with potential effectiveness for ND treatments, particularly for PD [7,60,71,72]. For instance, marine fungal metabolites such as neoechinulin A and xyloketal B have shown neuroprotective effects against MPP + -induced cytotoxicity in PC12 cells [73,74]. Additionally, neoechinulin A has also demonstrated to ameliorate the cytotoxic effect caused by rotenone, another chemical agent able to reproduce PD features [75]. According to the reported data, the indole alkaloid from marine fungi, neoechinulin A, significantly protected PC12 cells against the cytotoxicity induced by the Parkinson's disease-inducing neurotoxin, MPP + , by ameliorating the downstream events that result from mitochondrial complex I inactivation. However, the blocking of rotenone-induced cell death was only observed upon cotreatment with neoechinulin A, and not upon neoechinulin A pretreatment before rotenone. Nevertheless, the mechanism(s) underlying the cytoprotective effects remains unclear [75]. Moreover, a study has already demonstrated fiscalins' potential protective effects upon rotenone-induced cytotoxicity [30]. In fact, fiscalin derivatives, including the synthetic derivatives of the secondary metabolites fumiquinazoline G and fiscalin B (compound 1b), and epi-3 showed more than 25% protection against rotenone-induced cytotoxicity towards SH-SY5Y cells, as evaluated by the MTT assay, suggesting their neuroprotection capacity [30]. Adding the results obtained in the present work, fiscalin derivatives, although with limited studies available, seem to be promising neuroprotective agents. Indeed, the discovery of fiscalins A-C [28] as substance P inhibitors was also previously described as a promising novel neuroprotective therapy on early stages of PD, in an in vivo intrastriatal 6-hydroxydopamine model [29], anticipating this scaffold with potential neuroprotective properties that could be relevant for the treatment of ND, namely PD.
In our study, upon 24 h of exposure, two of the tested fiscalin derivatives (1a and 1b) caused a significant increase in the NR uptake, suggesting a potential protective effect against MPP + -induced cytotoxicity. However, after 48 h of exposure, only compound 1a maintained a significant protective effect. Moreover, 48 h after exposure, fiscalins 2a and 6 did also show a significant protective effect against MPP + -induced cytotoxicity, although only for the smallest concentration of MPP + tested. These fiscalins could be important ROS scavengers and inhibit their production, therefore having a beneficial effect on the OS induced by the MPP + .
So far, the data obtained allowed us to suggest that stereochemistry also affected the protective profile of the tested compounds, which was verified by comparing the results obtained for fiscalins 1a, 1b and 1c (Figure 1). In this case, it was observed that the S stereochemistry at C-4 (ring C, Figure 1) favoured the protective effects against MPP +induced cytotoxicity after 24 and 48 h of cell treatment. By altering the stereochemistry from S to R, the protective capacity was lost, with no significant effects being observed for fiscalin 1c (R stereochemistry at C-4) when compared with fiscalins 1a and 1b (S stereochemistry at C-4), which were shown to significantly protect differentiated SH-SY5Y cells against MPP +induced cytotoxicity. In addition, the R stereochemistry in the C-1 position could have also led to the protective effects observed for both fiscalins 1b and 2a. Similar results were obtained for fiscalins 2a and 2b, since fiscalin 2a, with S stereochemistry at C-4, presented a protective effect, and fiscalin 2b, with R stereochemistry at C-4, had no protective effect observed. However, this trend was shown to be inverted by the insertion of the hydroxyl group in the anthranilic portion of the fiscalin scaffold (ring A, Figure 1), as observed for fiscalin 6, with R stereochemistry at C-4, probably due to the known antioxidant activity of phenolic groups [76,77]. Surprisingly, the protective effect of fiscalins 2a and 6 was only observed 48 h after exposure, and for the smallest MPP + concentration tested (250 µM), revealing a late protective effect. Overall, the obtained results suggest that, amongst all fiscalin derivatives tested, fiscalin 1a may have a promising neuroprotective effect, being potentially promising for the treatment of PD, or for the development of new drugs targeting this disease.
Iron has several important roles in healthy brain functions. However, when its homeostasis is disrupted and iron levels begin to increase, oxidative stress and cell death can be triggered. Iron may induce oxidative stress due to its central role in ROS generation, and because of it can also oxidise dopamine (DA) to a highly reactive quinone, which ultimately leads to cellular dysfunction or cell death [19,78]. Additionally, ferric iron represents an important pro-oxidant that triggers oxidative stress by inducing ROS generation, for example, in neuroblastoma cells [59,79]. The importance of iron-induced toxicity has been supported by the protective effects demonstrated by iron chelators observed in cell culture models used to study ND, such as the SH-SY5Y cell line [16,19,59,80,81]. Importantly, iron has already been implicated in the pathology of several ND, including AD and PD [16,18,82]. Moreover, metal ions, particularly iron, provide a promising pharmacological target for the treatment of ND [17,83].
For experimental studies, the iron (III)-NTA complex prevents the hydrolysis of iron (III) at physiological pH, and has been widely used as an iron supply for numerous biological studies, since NTA moderate affinity for iron makes FeNTA an excellent model for mimicking small ligands involved in metal-chaperone proteins [84]. Therefore, to evaluate the protective effect of fiscalins towards ferric iron-induced cytotoxicity, ferric nitrilotriacetic acid (FeNTA) was the selected agent. Iron's cytotoxic effect was, therefore, evaluated in differentiated SH-SY5Y cells by the NR uptake assay, upon 24 and 48 h of exposure to FeNTA (0-1500 µM). The FeNTA concentration range was selected based in previous works, which demonstrated the FeNTA cytotoxic effect for concentrations of 250 µM, 1000 µM, and up to 2500 µM, and also in SH-SY5Y cells [59,85,86].
Fiscalin derivatives 1b, 2b, 4 and 5 have demonstrated significant protective effects towards iron (III)-induced cytotoxicity. They may play a role in ROS scavenging or have iron-chelating properties that possibly explain these protective effects. Therefore, these fiscalins could have promising neuroprotective effects or be useful for the synthesis of new derivatives to be further used in ND pathogenesis. Nevertheless, fiscalin 4 was able to maintain its significant protective effect, by reducing FeNTA cytotoxicity towards differentiated SH-SY5Y cells, upon 24 as well as 48 h of exposure, being the only tested fiscalin that showed a potentially sustained protective effect against iron (III)-induced cytotoxicity. Therefore, the obtained results suggest that this fiscalin may have a neuroprotective effect against metal-induced neurotoxicity, and adding this effect to its inhibitory effect on P-gp, it could be potentially promising for the treatment or for the development of new drugs targeting ND. Despite this, some of the tested fiscalins significantly increased FeNTA's cytotoxic effect towards differentiated SH-SY5Y cells, including fiscalin 2b, which after 48 h lost its protective effect seen right after 24 h of exposure. Accordingly, the pro-oxidant behaviour of several compounds has been suggested over the years. For instance, compounds that have proven antioxidant properties have demonstrated to also have a pro-oxidant behaviour [87]. An example of that is the widely accepted antioxidant vitamin C, which can also become a pro-oxidant when combined with iron, by reducing Fe 3+ to Fe 2+ , which consequently leads to hydroxyl radicals and hydrogen peroxide generation [87][88][89]. In addition, α-Tocopherol, which is another powerful antioxidant, in high concentrations can become a pro-oxidant due to its antioxidant mechanism. When α-Tocopherol reacts with a free radical it can become a radical itself, becoming highly reactive in the absence of sufficient antioxidant defences [87]. Moreover, myricetin, a flavonoid with antioxidant activity, has the ability to scavenge ROS as well as to chelate iron. However, it gains pro-oxidant properties due to the reduction of molecular oxygen and Fe 3+ to Fe 2+ [90]. Regarding fiscalins 2b, 4 and 5, their protective effect towards FeNTA-induced cytotoxicity could be mostly related to a direct effect on iron, given the lack of any protective effect against MPP + -induced cytotoxicity. Additionally, fiscalin 1b could be interfering in the OS imbalance by scavenging ROS, despite any iron-chelating properties, due to the protective effect verified towards both FeNTA and MPP + cytotoxicity.
Regarding our work, the analysis of the data obtained in the FeNTA experiments showed that overall, the insertion of a phenolic group substituent in the C-1 position led to an increased protective effect against iron-induced cytotoxicity for fiscalins 4 and 5. In fact, the same stereochemistry (1S,4S) in the positions C-1 and C-4 (ring C, Figure 1) led to a long-term protection effect, verified for fiscalin 4, which did not occur for the rest of the fiscalins, namely fiscalin 5, with different stereochemistry (1R,4S). Furthermore, and attending to the stereochemistry, it was possible to assess that the presence of an isopropyl group at C-1, with a R configuration (fiscalin 1b), and an isobutyl group at C-1, with a S configuration (fiscalin 2b), were crucial to the protective effect against the ferric iron-induced cytotoxicity.
Lastly, taking the obtained results together, it is possible to extract some overall considerations concerning structure-protective activity amongst the library of fiscalin derivatives tested, as shown in Figure 6. First of all, the results obtained indicated that differences in stereochemistry were critical for the activity of these compounds. For instance, compounds 1a, 1b and 1c, which are stereoisomers, presented different behaviours among the performed assays, highlighting the importance of the tridimensionality structure of compounds in their activity. Second, the substituent at position 1 seems to influence fiscalin activity, with short side chains, such as in fiscalins 1 and 2, possibly prompting their neuroprotective effects. Moreover, adding an hydroxyl group on the aromatic substituent at C-1, such as in fiscalin 5, was shown to be possibly beneficial for the protective effect, which is in accordance with the study on SH-SY5Y cells with cytotoxicity induced by rotenone [30]. Third, different substitutions in the aromatic ring of the anthranilic portion, besides halogens, did not seem to have helped the fiscalins' protective activity. Overall, it is important to highlight that amongst all the fiscalin derivatives tested in this work (Figures 6 and 7), fiscalin 1a has proven to be a promising compound for the treatment of ND such as PD, given its sustained protective effect on differentiated SH-SY5Y cells towards MPP + -induced cytotoxicity, verified for both timepoints tested; moreover, fiscalin 1b could be particularly important in the treatment or prevention of ND, given the protective effect observed in both disease models used upon this work (MPP + and FeNTA). Additionally, fiscalin 2b, due to its protective effect towards iron-induced cytotoxicity and the P-gp activation capacity, could be promising for the Aβ peptides detoxification, and therefore, for the treatment or prevention of AD. Furthermore, fiscalins 4 and 5 also revealed important biological activities with particular benefit for the treatment and/or prevention of ND, as they accumulate a concentration-dependent P-gp inhibitory capacity alongside with a sustained protective effect against FeNTA-induced cytotoxicity, particularly for fiscalin 4. Overall, although further studies are needed, from the obtained data it is possible to highlight that the fiscalin scaffold seems to be a promising new source of neuroprotective agents that may open new perspectives in the treatment of ND, such as AD and PD.  (1a, 1b, 1c, 2a, 2b, 3-15) under investigation in the present study.

Conclusions
In the present work, some fiscalin derivatives have shown relevant biological activities regarding the modulation of the P-gp, as well as protective effects towards cytotoxicityinducing agents (MPP + and Iron (III)). Fiscalins 1c, 2a, 2b, 6 and 11 caused an increase in P-gp transport activity, while fiscalins 3, 4, 5, 7 and 8 caused significant P-gp inhibition, thus suggesting a promising source of new P-gp activators and inhibitors, respectively. Moreover, fiscalins 1a, 1b, 2a and 6, particularly fiscalins 1a and 1b, showed a significant protective effect against MPP + -induced cytotoxicity, and fiscalins 1b, 2b, 4 and 5 showed a protective effect against iron (III)-induced cytotoxicity.
Despite the important biological properties presented in this work, the mechanisms underlying the observed neuroprotective effects are yet unknown, persisting the need for more mechanistic and in vivo studies to further elucidate their mechanisms of action and their potential as new disease-modifying drugs. Indeed, future perspectives of the present work aim at the evaluation, in vivo, of an MPTP model of PD, of the pharmacokinetics and pharmacodynamics profiles of the most promising fiscalin derivatives, as well as the assessment of potential side effects. Furthermore, and knowing that ferroptosis is involved in the etiology of several ND apart from PD and AD (such as amyotrophic lateral sclerosis or Huntington's disease), these derivatives may also in the future represent new therapeutic strategies to be tested in the scope of the treatment and/or prevention of such diseases.
In conclusion, fiscalin derivatives, although with limited studies available, seem to be promising neuroprotective agents. Noteworthy, the fiscalin scaffold may provide a widespread library of new chemical structures with potential neuroprotective properties that could be relevant for the treatment of ND, namely AD and PD, and targeting different pathophysiological mechanisms of such diseases.