Characterization of Antimicrobial, Antioxidant, and Leishmanicidal Activities of Schiff Base Derivatives of 4-Aminoantipyrine

Our main interest is the characterization of compounds to support the development of alternatives to currently marketed drugs that are losing effectiveness due to the development of resistance. Schiff bases are promising biologically interesting compounds having a wide range of pharmaceutical properties, including anti-inflammatory, antipyretic, and antimicrobial activities, among others. In this work, we have synthesized 12 Schiff base derivatives of 4-aminoantipyrine. In vitro antimicrobial, antioxidant, and cytotoxicity properties are analyzed, as well as in silico predictive adsorption, distribution, metabolism, and excretion (ADME) and bioactivity scores. Results identify two potential Schiff bases: one effective against E. faecalis and the other with antioxidant activity. Both have reasonable ADME scores and provides a scaffold for developing more effective compounds in the future. Initial studies are usually limited to laboratory in vitro approaches, and following these initial studies, much research is needed before a drug can reach the clinic. Nevertheless, these laboratory approaches are mandatory and constitute a first filter to discriminate among potential drug candidates and chemical compounds that should be discarded.


Introduction
In the last decades, antibiotic resistances have been described in several microorganisms, including those having multiresistance phenotypes [1][2][3][4]. This is an alarming situation, as noted by the World Health Organization, and many researchers have focused on the development of new therapeutic alternatives. Oldfield and Feng have proposed different strategies to develop a new generation of antibiotics (resistance-resistant antibiotics) that should have the characteristic of acting against multiple targets simultaneously [5]. Alternatively, other authors have suggested the development of new antibiotic classes with novel chemical properties not present in existing classes of antibiotics [6]. In this context, Schiff bases offer an alternative that has not been adequately explored.
Schiff bases have received much attention due to their wide applications in several fields, ranging from industrial uses such as advanced nanomaterials [7] to chemotherapeutics and new drug development [8][9][10][11]. However, even regarding the chemotherapeutic applications, the reported roles are very divergent, sometimes with proposed roles that could seem contradictory. For example, antioxidant activity has been ascribed to Schiff bases, acting as cell-protective agents [12][13][14][15], but antimicrobial [16,17], antitumor [18,19], analgesic [12], anti-inflammatory [9], and antidiabetic activities [8,18] have also been reported. A detailed review covering the patented therapeutics applications of Schiff bases has been published by Khan and collaborators [18]. These diversities of functions could rely on the wide variety of Schiff base compounds that can interfere with biological activities. The open question is what is the mechanism(s) of action by which they operate. Recent publications suggest that different groups of Schiff bases can exert actions through different mechanisms. For example, some compounds have a potential application as antitumor drugs [18], but different molecular modes of action could explain this property. While some Schiff bases are able to intercalate into the DNA helix [6], others display DNA cleavage activity [20,21]. It is not possible to rule out the possibility that in some compounds, both activities are linked. Also, this DNA-damaging activity could explain antimicrobial activities in some cases.
Other Schiff bases seem to act by inhibiting specific targets. Some of the most prominent examples are the patented inhibitors specific for β-secretase, an enzyme present in aberrantly high levels in Alzheimer's patients [22]; Schiff base copper complexes that target and inhibit reversibly caspase enzymes involved in apoptosis [23]; or Schiff bases of thiazoles specific for urease, an enzyme present in bacteria, fungi, and plants [24] (all of them reviewed by Khan and collaborators [18]).
However, the broad range of described effects of these compounds make it reasonable to think that a more generalized mechanism may be involved. The role of Schiff bases in a generalized process, the in vivo nonenzymatic glycation of macromolecules, is very interesting. These reactions take place between a reducing sugar and a macromolecule-DNA, lipid, or protein-and generate advanced glycation end products (AGEs) [8,25] through a Schiff base intermediary. The accumulation in blood of AGE products, closely related to oxidative stress, triggers many health complications associated with diseases such as ageing, diabetes, Alzheimer's, and cancer, among others [8,25,26]. A class of Schiff bases has been described as agents with antiglycation activity, being very promising for prevention of late diabetic complications [27]. This suggests that some Schiff bases can interfere with AGE production, possibly indicating a generalized mode of action that can explain, at least partially, the wide roles and activities of these compounds. Protein glycation leads to crosslinking and protein aggregation, which indeed constitute a source of oxidative species. Some proteins suffering massive glycation are immunoglobulins, apolipoprotein, fibrinogen, and transferrin, which have been associated with several disorders [25]. It is possible that the antioxidant activity described for some Schiff bases [13,18] is related to the prevention of the protein glycation processes, but to our knowledge, nothing has been published regarding this; however, this correlation has been reported for some flavonoids inhibiting the formation of both initial and advanced stages of Maillard reaction in tissue protein sources [28]. Other mechanisms could operate by trapping of carbonyl compounds, similar to that reported with OPB-9195, a known inhibitor of in vitro formation of AGEs [29].
The diversified Schiff base structures and their different mechanisms of action can explain the wide variety of biological activities described for these compounds that deserve more attention as promising candidates for therapies are being more and more delineated. Some of the Schiff bases reported with interesting bioactivities are those derived from 4-aminoantipyrine, and early reports have shown that these compounds have also important antioxidant properties [14], as well as antifungal and antibacterial activity against several microorganisms [30][31][32][33]. However, most of these studies were performed by antibiograms. Here, we report the synthesis of a series of Schiff base derivatives of 4-aminoantipyrine [14,[30][31][32][34][35][36][37] and their evaluation of antifungal, antibacterial, (inhibitory concentration of 50% of population (IC 50 ), minimum inhibitory concentration (MIC), and noninhibitory concentration (NIC)), leishmanicidal, and antioxidant activities, which will provide useful information to explain the potential chemotherapeutic use of these compounds.

Synthesis of Schiff Base of 4-Aminoantipyrine
The Schiff bases 3 were synthesized following the scheme depicted in Figure 1. Compound 1 was reacted with the respective benzaldehydes 2a-j to obtain 3a-j, while 4 and 5 were obtained by reaction with furfural (6) and cinnamaldehyde (7), respectively. The reactions were performed using ethanol as solvent to give good to excellent yields ( Table 1). All compounds were characterized, and all the data obtained agreed with the proposed structures. The 1 H NMR spectra for 3a-j show a singlet between 9.41 and 9.79 ppm corresponding to the azomethine -CH=N proton, while 5 shows a doublet at 9.40 ppm. These signals are like those reported in the literature, and as expected, the azomethine proton for all compounds with electron-donating groups are shifted to a higher field relative to 3a, except for 3b, which shows a signal at 9.70 ppm, possibly due to an intramolecular hydrogen bond with the hydroxyl group.

Antimicrobial Evaluation
The in vitro antifungal and antimicrobial activities were determined by testing the Schiff bases against the fungi Saccharomyces cerevisiae, Candida albicans, and Aspergillus niger and the bacteria Staphylococcus aureus, Enterococcus faecalis (Gram-positive), Pseudomonas aeruginosa, and Escherichia coli (Gram-negative). Additionally, the values of minimum inhibitory concentration (MIC), noninhibitory concentration (NIC), and inhibitory concentration of 50% of population (IC 50 ) were determined as described by Lambert et al. [38,39].
Since the Schiff bases have a limited solubility in dimethyl sulfoxide (DMSO), first, we tested the toxicity of DMSO against the selected microorganism. According to published data [40], concentrations over 2% DMSO are significantly toxic for S. cerevisiae. Drop test assay ( Figure 2) revealed that wild-type cells from the BY4741 strain grow with no major difference at DMSO concentrations below 2%, but at concentrations of 3% DMSO or higher, wild-type cells begin to show diminished viability, and the effect is more acute in sod1 cells. Wild-type cells from W303 background seem to be more resistant, and sensitivity is observed at around 4% DMSO. Figure 2. DMSO toxicity. Drop test of serial dilution of yeast cell cultures (wild type from genetic background W303, wild type from genetic background BY4741, and sod1 mutant from genetic background BY4741) exposed to different DMSO concentrations ranging from 1 to 6%. A control without DMSO (−DMSO) is included.
More accurate determinations of noninhibitory concentration (NIC) [38] revealed that toxicity ranges from a minimum of 1.55% DMSO (E. faecalis) to a maximum of 15.86% (E. coli) ( Table 2). Considering the limited solubility of Schiff bases and the DMSO toxicity, we set up our protocol at a maximal DMSO working concentration of 2%. Values of NIC, MIC, and IC 50 are summarized in Table 3, and all parameters were compared with that of the commercial drugs geneticin (GN) for S. cerevisiae, voriconazole (VR) for A. niger and C. albicans, and ciprofloxacine (CP) for all bacteria studied. It is important to note that although in the literature there appear several reports regarding the activities of some of Schiff bases reported here with these microorganisms [14,[30][31][32][33][34][41][42][43][44][45][46][47][48], those tests were done using different methods of evaluation in different strains, when it was reported. Overall, A. niger and S. aureus are inhibited by Schiff base 3f; C. albicans by 3b and 3f; S. cerevisiae by 5; E. faecalis by 3a; and E. coli by 3b, 3d, 3f, and 5. Only P. aeruginosa was not sensitive to any of the assayed compounds. Interestingly, different microorganisms are inhibited by different compounds, revealing specific activities against specific microorganisms. The specificity is a property required in drug candidates for applications in pharmacology. Compounds 3c, 3e, 3g, 3h, 3i, 3j, and 4 did not show any activity against the tested microorganisms, and the activity of 3e, 3f, and 5 had not been reported previously. Schiff base 3f is the only compound that showed activity with all bacteria, except E. faecalis, and 3d had the lowest activity (MIC of 10.2 ± 0.1 µg/mL). When the observed MIC concentration was >200 µg/mL or out of the assay range concentration, the drug was considered ineffective (reported as NE, no effect). Table 3. Growing inhibitory parameters (MIC, NIC, and IC 50 , expressed in µg/mL), determined by serial microdilution in 96-well plates, for Schiff bases 3a-j, 4, and 5. The sensitivity was tested for the microorganism indicated in the first column. Geneticin and voriconazole were used as control against fungi, ciprofloxacin for bacteria, and amphotericin and pentamidine for leishmania. Next, we evaluated qualitatively if the inhibitory effect is exerted by killing the microorganism (bactericidal/fungicidal effect) or by preventing growth (bacteriostatic/fungistatic effect). After incubating cells in the presence of different concentration of Schiff bases for 24 h, cells were plated on media free of drug. Results indicated that only compound 5 has a fungicidal mechanism on S. cerevisiae, since cells are not able to recover after exposure (see Supplementary Figure S16A). The rest of the Schiff bases assayed showed a bacteriostatic behavior, allowing the growth of microorganism on drug-free media after 24 h of exposure (Supplementary Figure S16B).
Finally, Schiff bases were evaluated against L. mexicana parasites in promastigote form by (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. The results are shown in Table 3. Some compounds were inactive or too weak (a cutoff value of IC 50 > 200 µM was considered), but others showed leishmanicidal activity to different extents, particularly 5 and 3f, with the latter being the most powerful. Pentamidine and amphotericin were included as controls.
In conclusion, Schiff bases 3f and, to a lesser extent, 5 show a wide range of activities against several bacteria. Also, 3a, 3b, and 3d are interesting candidates, with narrow but specific activity. Compounds 3f and 5 showed the best activity against parasites. So, we will focus our next efforts in evaluation of these compounds.

Evaluation of Antioxidant/Oxidative Activity
All the synthesized compounds were evaluated for antioxidant activity by the N,N-diphenyl-N -picrylhydrazyl (DPPH) assay (Table 4). All compounds evaluated showed lower radical scavenging activity than the standards ascorbic acid (entry 13), quercetin (entry 14), and caffeic acid (entry 15). In general, the presence of hydroxyl groups in the para position of the aromatic ring (3d, 3i, and 3j) confers a better radical scavenging activity compared to those with hydroxyl groups in other positions or with other electron-donating groups, with the exception of 3h, which contains one dimethylamino group (entry 8) that also can stabilize the radical by resonance. Compounds 3d, 3h, 3i, and 3j possess the best antioxidant activities, comparable to the standards, but only 3d presents activity against bacteria (E. coli), while 3f and 5 are effective against leishmania promastigotes. Next, we wanted to evaluate the oxidative activity. To do this, drop test assays were performed with wild-type (W303 and BY4741 backgrounds) and sod1 (BY4741 background) yeast cells in the presence of 200 µg/mL of the analyzed Schiff bases ( Figure 3). The SOD1 gene encodes superoxide dismutase 1, an enzyme with a major role in detoxifying ROS species generated during metabolism or because of environmental exposure. sod1 mutant cells are unprotected against oxidative agents, decreasing survival in their presence. Figure 3A shows the relevant results obtained with drugs 3b and 5, where sod1 cells are sensitive to these compounds compared to the control (no drug, DMSO), after 48 h of incubation at 25 • C. Also, wild-type cells are shown as control, unaffected by the Schiff bases. Figure 3B and Table 5 show the graphic representation of optical density (OD600) of wild and sod1 cells exposed to increasing concentrations of compounds 3b and 5, and the MIC, IC 50 , and NIC values obtained for these compounds, respectively. Drop test of serial dilution of yeast cell cultures (wild type from genetic background W303, wild type from genetic background BY4741, and sod1 mutant from genetic background BY4741) exposed to Schiff bases at 200 µg/mL. A control without drug and with 2% DMSO are included. (B) Graphic representation of optical density (OD600) of wild-type (black line) and sod1 (grey line) cells (BY4741) exposed to increasing concentrations of compounds 3b and 5. These results reveal that 3b and 5 act by an oxidative mechanism, in accordance with the results obtained in the DPPH assay, which detected no significant antioxidant activity of these compounds.

Toxicity in Mammalian Cell Lines
The specificity and selectivity for the pathogen target is a key point for any potential drug. To evaluate these, we determined the cytotoxicity (% of living cells) of mammal macrophage cells (murine RAW 264.7, ATCC TIB 71) after exposure to Schiff bases; see Table 6. Cytotoxicity was determined at 400 µM of each Schiff base, except for those marked with an asterisk (*), which were assayed at 300 µM because of the limited solubility inherent to these compounds. Compounds 3d, 3f, 3h*, 3j, and 5 showed high cytotoxicity with ≤20% of cells surviving, with 3f and 5 being the most cytotoxic, thus making them useless for pharmacological purposes, at least in this simple formulation (chemical compound directly in solution). Cytotoxicity data restrain the potential application to 3a and 3i, the only two compounds producing viability of >25% at 0.1165 and 0.1293 mg/mL, respectively. A control with saponine was included, giving an IC 50 value of 0.1400 mg/mL. 3a and 3i are effective against E. faecalis and as an antioxidant, respectively. Overall, these results are an early filter indicating a potential application for compounds 3a and 3i, which deserve to be analyzed more deeply.

Evaluation of Pharmacokinetic and Drug-Like Properties
To evaluate the drug-likeness of a compound, Lipiniski's rule of five (ROF) is usually used. This rule states that good absorption or permeation are more likely when: (a) there are no more than five hydrogen bond donors (HBD); (b) no more than 10 hydrogen bond acceptors (HBA); (c) formula weight less than 500; and (d) n-octanol/water partition coefficient (log P) is less than 5. Two or more violations of ROF suggest the probability of problems in bioavailability [49]. One of the big problems with the development of new drugs is that they usually fail before reaching the clinics due to poor pharmacokinetics, and these physicochemical drug descriptors provide a useful tool for evaluating drug activity. These descriptors were calculated for all the synthesized compounds 3a-j, 4, and 5 using the Osiris DataWarrior software (Table 7) [50]. Additionally, the bioactivity scores of all compounds were calculated using the Molinspiration cheminformatics software (Table 8), and all parameters were compared with those of the standard drugs voriconazole (VR) and ciprofloxacin (CP). Table 7. Theoretical prediction of adsorption, distribution, metabolism, and excretion (ADME) properties of Schiff bases 3a-j, 4, and 5, calculated using Osiris DataWarrior software. Voriconazol and ciprofloxacin were included as controls.  Calculations of the physicochemical drug descriptors (see Table 7) show that molecular weights of all synthesized compounds are below 500; HBD and HBA values are 4-7 and 0-2, respectively; logP and logS values were below 2.2 and above −3.70, respectively; and all compounds have a low number of nrotb (3)(4)(5); so, a very low conformational flexibility and good oral bioavailability are expected. According to this, all Schiff bases meet Lipinski´s rule of five. Additionally, all compounds show permeability to cell membranes, and half of them (3a-c, 3h, 4, and 5) show PSA values from 35.91 to 56.14 Å 2 , suggesting good permeability at the blood-brain barrier.
The evaluation of toxicity risk assessment is also important prior to structural drug design, and many potential drugs fail to reach the clinical stage because of ADME toxicity issues. The toxicity risk predictor locates fragments within a molecule that constitute a potential toxicity risk for one of four major toxicity classes: mutagenicity, tumorigenicity, reproductive effect, and irritant effect. The results are displayed, regarding the category specified, as high, low, or none as risk for a compound to possess a toxic effect. The result show that all the compounds have a high risk to be mutagenic and only 3d and 3h show a low and high risk, respectively, to be tumorigenic. In the other categories, none of the compounds showed effect.
Drug-likeness (DL) of a compound may be defined as a complex balance of various molecular properties and structural features which determine whether a particular molecule is similar to known drugs [33]. Considering that most drugs have drug-likeness with a positive value, it is desirable that new candidate stay in the positive range. Osiris DataWarrior calculations show that all nitro compounds, 3f, and 3g have negative values, while other compounds have values higher than 4.08, a value similar to that observed for voriconazole. The drug score index (DS) is a parameter that combines drug-likeness, cLogP, logS, molecular weight, and toxicity risks in a single descriptor that is very useful for determining whether a compound can be qualified as a potential drug. According to established criteria, values closer to 1 indicate a higher probability of a compound being a future drug candidate. All the evaluated compounds exhibited poor to moderate scores (0.30 to 0.54), 4 being the compound with the highest value. Finally, the bioactivity scores of all compounds for being a G protein-coupled receptor (GPCR) ligand, ion channel modulator, kinase inhibitor, nuclear receptor ligand, protease inhibitor, and enzyme inhibitor were calculated and are shown in Table 8. A high score implies a higher probability for a molecule to be biologically active. For many organic molecules, the prediction is as follows: if the bioactivity score is more than 0.00, it is active; if −0.50 to 0.00, moderately active; and if less than −0.50, inactive. According to this, all compounds in this study are predicted to be inactive as GPCR ligands, ion channel modulators, nuclear receptor ligands, and protease inhibitors, and only 3j shows a moderate activity as a kinase inhibitor, while 3b, 3d, 3e, 3h, 3i, and 3j show similar activity as enzyme inhibitors. Both standard drugs show moderate to high activity against all bioactivities evaluated. These results suggest that the mechanism of action would be unspecific and/or generalized.

Discussion
In this study, we synthesized a series of Schiff base derivatives of 4-aminoantipyrine and tested their cytotoxic properties against selected bacteria, fungi, and leishmania parasites, as well their antioxidant/oxidative activities. Finally, we analyzed their properties regarding their suitability for therapeutic uses.
Schiff bases are an emerging alternative for disease treatments. In the last decades, many types of resistance have been described in bacteria, generating the need for new drugs for dealing with the diseases caused by microbes [3,4]. In addition, the prevalence of cancer and metabolic diseases such as diabetes are predicted to rapidly grow in the next years [51,52]. For many of these diseases, there are still no effective therapies. Schiff bases have shown a wide range of properties that make them promising candidates as alternative drugs, but they have not been adequately characterized. There have been described-and some patented-Schiff bases that are effective against microorganisms (bacteria, virus, fungi, parasites) or against cancer, with antiglycation, antioxidant, or anti-inflammatory activities, among others [15,18]. A major question is how these compounds act against several targets and display different properties. The most reliable answer is that due to the vast variety of compounds, there should be many modes of action.
Our results indicate that Schiff bases 3a-b, 3d, 3f, and 5 show antimicrobial activities with moderate inhibitory values ( Table 3). Evaluation of scavenging activity reveals antioxidant properties for 3d and 3h-j, while compounds 3b, 3f, and 5 possess oxidative activity. Overall, according to the in silico calculations, Schiff bases present good scores for the parameters evaluated, such as bioavailability, and have membrane permeability similar to commercial drugs (DS/DL), except 3d and 3f-h, which have poor scores for DS (<0.5), DL, and risk of tumorigenicity. The main problem with the low DL score of 3f is the presence of the nitro group in the structure, with is related to the toxicity of some drugs [53]. Actually, drug discovery pipelines for the 'neglected diseases' are now heavily populated with nitro compounds [54,55], so this compound should not be discarded. Also, cytotoxicity is significant in mouse macrophages. These analyses limit compounds 3a and 3i to be the ones with the most promising potential applications (summarized in Figure 4), with some considerations discussed below. Compounds 3c, 3e, 3g, and 4 do not present any kind of activity and were discarded from further analysis. It deserves to be discussed that compound 3g is equivalent to compound 9 previously described by Thirunarayanan and collaborators [30]. The authors describe microorganism sensitivity by the Kirby-Bauer method with comparison to the commercial drugs used as controls, namely ciprofloxacin for bacteria and miconazol for fungi. They reported significant activity against B. subtilis, M. luteus, S. aureus, E. coli, P. aeruginosa, and the fungi A. niger and T. viride. In our hands, only S. cerevisiae presents a relative sensitivity to Schiff base 3g, but for the rest of the tested microorganisms, the MIC values obtained were too high (>200 µg/mL). These two techniques are not directly comparable, since the Kirby-Bauer method is qualitative and MIC is quantitative, but two possible explanations for these differences are as follows: First, Thirunarayanan and collaborators did not report the disk potency (µg) used in the assay; and second, the origin of the strains used in their work is not indicated and apparently are not from the American Type Culture Collection (ATCC). Together, this could explain the differences obtained with the same compound. Nevertheless, we should omit compound 3g from next analyses, since we are not able to describe any significant activity. Compounds 3b, 3f, and 5 show both antimicrobial and oxidative activity, opening the possibility that the antimicrobial effect is a consequence of its oxidative activity. Since the effect is specific (3b for C. albicans and E. coli; 3f for C. albicans, A. niger, S. cerevisiae, E. coli, S. aureus, and L. mexicana; 5 for S. cerevisiae, E. coli, and L. mexicana), we cannot rule out that these compounds are oxidizing specific targets. This is consistent with the bioactivity analyses, at least for 3b, indicating that it could be acting as an enzyme inhibitor (Table 8). However, due to the difference observed in the time-response experiments, the action mechanism of both compounds could be different. Additionally, 3a and 3d showed antimicrobial activity against E. faecalis and E. coli, respectively.
The analysis of the structure-activity of the Schiff bases shows that 5 presents an extension in the imine conjugation and the aromatic ring does not possess any substituent, while 3f has an ortho hydroxyl group (electron-donating group) and a meta nitro group (electron-withdrawing group), and it is evident there is a synergistic effect between both substituents, because when only the hydroxyl group (3b) or only the nitro group (3g) is present, the activity is limited to C. albicans and E. coli in the first case, and no activity observed in 3g, while no activity against leishmania was observed for either compound. The hydroxyl group in the ortho position can form an intramolecular hydrogen bond with the imine nitrogen [56], which, due to the inductive and resonance effect of the nitro group, favors the keto-enol tautomerism to generate the keto tautomer which could be, in part, responsible for the enhancement of the observed antimicrobial and leishmanicidal activities.
Toxicity was assayed in RAW cells (Table 6). Results indicate that compounds 3a and 3i are the least toxic in our system. Although it is true that other systems should be tested, such as primary macrophages isolated from mice or human macrophages cell lines, the results with RAW cells are a sensitive indication of toxicity that should be considered before continuing any new experimental design. Indeed, 3a and 3i are effective against E. faecalis and as antioxidant, respectively. E. faecalis is an opportunistic pathogen highly associated with resistance development in the last decades. This bacterial infection is common in hospital environments, presenting such clinical manifestations as urinary tract infections, bacteremia, and endocarditis, among others [57].
Regarding compounds 3d and 3h-j, their potential application is related to their antioxidant activity, and thus as cell-protective agents. In accordance with our results, antioxidant activity for compounds 3i-j has been previously described [14], along with a very low inhibition of RAW cell viability. These compounds have similar activities as the controls: ascorbic acid, quercetin, or caffeic acid (Table 4). Nevertheless, potential applications of 3d and 3h are not very promising, as the ADME test gave poor scores for several of the parameters (Table 7). Neither is 3j a good candidate, as the percentage of viable RAW cells was compromised in our experiments (Table 6). So, only 3a seems to be a reliable option for an antioxidant activity approach.
The biggest limitation of these compounds is that all of them seem to present mutagenicity characteristics, according to our predictive analyses for ADME toxicity issues (Table 7). If so, this prediction should be tested experimentally; it would be interesting to exploit these compounds to identify chemical varieties that keep or improve their useful properties, while diminishing the associated mutagenicity. Nevertheless, a possible explanation is that compounds 3a, 3b, 3d, 3f, and 5 exert their antimicrobial and/or leishmanicidal activities through a mutagenic mechanism, but this could not be the main mechanism as none of the other compounds show this activity, despite reported mutagenicity. Although this is not a desired property for antimicrobial compounds, this observation opens the question as to whether these compounds could be used in cancer treatment.
It is recognized that most anticancer drugs are potentially hazardous substances, since they are mutagenic, teratogenic, and/or carcinogenic [58] and as they exert their activities probably due to these properties. Some of them have been classified as carcinogenic by the International Agency for Research on Cancer (IARC), and the in silico prediction of their ADME properties show poor scores in some categories (Supplementary Table S1). For example, cyclophosphamide, a drug used in the treatment of chronic lymphocytic leukemia, lymphomas, soft tissue and osteogenic sarcomas, and solid tumors, is in group 1 (carcinogenic to humans) according to IARC, and according to the calculations, presents a high risk to be mutagenic, tumorigenic, and have reproductive effects. Considering this, it cannot be ruled out that Schiff bases could have anticancer properties. Indeed, a review of the literature describes some Schiff bases with associated anticancer activity [32]. However, more studies should be done to determine if these compounds are able to discriminate between normal mammalian cells and certain kinds of pathogens or cancer cells [18,19], or if, in contrast, they are generalized mutagens useless for therapeutics.
The toxicity of drugs can be greatly controlled using nanoformulations (nanoparticles, nanotubes, liposomes, etc.) [41,42]. These systems enable specific selection of the desired target and delivery of the drug locally and in a controlled way, reducing the side effects generated by conventional drug delivery systems that usually require high dosage. So, the choice of the system will depend on the final purpose of the formulation. Nanomedicine offers new approaches to drug design, improving the specificity and limiting the undesired toxicity. Major efforts have been focusing on developing nanocarriers to fight against cancer. Main limitations of chemotherapy are multidrug resistance and nonspecific tissue targeting, generating serious side effects that compromise the patient's health. Nanoformulations of chemotherapeutic agents enhance the selective delivery, diminishing the toxicity. Despite the associated benefits, only a few nanoformulations for chemotherapy have been approved by the FDA and are currently marketed. Nevertheless, many research projects and clinical trials are studying different nanoparticles, disclosing them as an emerging therapeutic approach [43,52]. Nanoparticles are not only useful in chemotherapy; many potential applications against infectious microorganisms and brain or metabolic diseases are emerging [41,44].
The work here presented is a first approach to discriminate among compounds to determine those which have application potential and those which are useless (have no biological activity or intrinsic toxicity, etc.). Additional efforts to characterize and improve Schiff base properties and toxicity should be done by the scientific community to uncover the real uses and applications of these compounds in therapy as antimicrobial agents or otherwise, such as anti-inflammatory, antipyretic, antitumor, or antidiabetic agents.

General Procedures
All solvents and reagents were from Sigma Aldrich (St. Louis, MO, USA) and used without further purification. All melting points are uncorrected and were determined on a Büchi Melting Point M-560 (Büchi Labortechnik AG, Flawil, Switzerland). Infrared spectra (KBr discs) were recorded on a Perkin Elmer RX1 infrared spectrophotometer (PerkinElmer, Waltham, MA, US); wave numbers are reported in cm −1 . NMR spectra were recorded at 298 K on a Bruker Advance 500 MHz spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany) equipped with a z-gradient, triple-resonance ( 1 H, 13

Synthesis of Schiff Base Derivatives (3a-j, 4, 5)
The synthesis of the Schiff base derivatives 3a-j was adapted from a procedure reported by Alam and coworkers [14]. To an ethanol solution (10 mL) of 4-amino-1,5-dimethyl-2-phenylpyrazol-3-one (1) (1.0 g, 4.92 mmol) was added an ethanol solution (10 mL) of substituted benzaldehyde (4.92 mmol), and the mixture was refluxed for 6 h. The progress of the reaction was monitored by TLC. The precipitates formed were collected by filtration, purified by recrystallization with ethanol, and then dried under vacuum to produce the pure compounds. Spectra analyses can be consulted in the Supplementary  Materials (Figures S1-S15).

Evaluation of Antimicrobial Activity
The antimicrobial activities of the synthesized compounds were tested against the Gram-positive bacteria Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 29212, the Gram-negative bacteria Escherichia coli ATCC 25922 and P. aeruginosa ATCC 27853, and the fungal strains Candida albicans ATCC 10231, Aspergillus niger ATCC 6275, and S. cerevisiae (backgrounds W303 and BY4741), using the microdilution method [46,47].
The bacterial inoculum was prepared by emulsifying overnight colonies from an agar medium in broth (Mueller-Hinton broth). The suspensions were adjusted to a final organism density of 5 × 10 5 colony former units (cfu)/mL. The suspensions of C. albicans were made in saline solution and adjusted to an optical density of 0.155 at 530 nm. The emulsifying of A. niger cultures was done by pouring 5 mL of saline and 0.1% of polysorbate 80 over the Petri dishes. Then, 100 µL of this suspension were diluted in 5 mL of saline and adjusted to an optical density of 0.1 at 530 nm. For S. cerevisiae, fresh exponential YPAD (yeast extract 1%, peptone 2%, adenine 0.004%, dextran 2%) cultures were adjusted to an optical density of 0.5 at 600 nm, and then diluted to the equivalent of 1 × 10 4 cells/mL (10 −3 dilution). Stock solutions of Schiff bases were prepared by dissolving them in DMSO at 10 mg/mL. For controls, ciprofloxacin was used for all bacteria tested, voriconazol for A. niger and C. albicans, and geneticin for S. cerevisiae. The range of concentrations used was empirically selected for each organisms and the Schiff base. Drug sensitivities were assayed by serial dilutions in 96-well plates as previously described [39,46,47]. A row with media and cells but without drug was the positive control. The blank was done by preparing a row with media and drugs, but without cells. Plates were fitted with a tight lid before incubation to prevent desiccation. Plates were incubated at 37 • C in air for 24 h for bacterial cultures and 35 • C for 48 h for fungi. The turbidity was recorded at 600 nm for bacteria and 530 nm for fungi in a plate reader (Synergy H1, Biotek), and calculations of MIC (minimal inhibitory concentration), NIC (noninhibitory concentration), and IC 50 (inhibitory concentration for 50% of the population) were done with GraphPad Prism [39]. The minimum bactericidal concentration (MBC) and fungicidal concentration (MFC) were determined by spreading 100 µL of culture from each well onto a plate (Mueller-Hinton agar for bacteria, Sabouraud dextrose agar for fungi except S. cerevisiae, and YPAD for S. cerevisiae). MBC and MFC values represented the lowest concentration of a compound that completely inhibited growth.

Evaluation of Leishmanicidal Activity
Leishmanicidal activities of all Schiff bases were evaluated by measuring promastigote mitochondrial activity using MTT colorimetric assay as described previously [57]. In this study, promastigotes from L. mexicana (M379), donated from the Instituto de Medicina Tropical Alexander von Humboldt (Universidad Cayetano Heredia, Peru) were cultured at 24 • C in Schneider's Drosophila Medium supplemented with 20% fetal bovine serum and 100 IU/mL penicillin + 100 mg/mL streptomycin. The medium was renewed every four days. Parasite density was determined in a Neubauer chamber.
Into each well of a 96-well cell culture dish was dispensed 1 × 10 6 parasites. A stock solution of each of the Schiff Base was prepared in DMSO, and serial dilutions were added to the parasite suspension, keeping the solvent concentration at 1%. The final volume was 100 µL for each well and triplicate conditions were carried out. Pentamidine treatment (100 µg/mL) and untreated parasites were used as positive control and negative control, respectively. After exposure to the compounds for 48 h in culture medium, 10 µL of a solution of 5 mg/mL MTT dissolved in PBS was added to each well. The plate was incubated at 24 • C for 4 h in darkness. Later, the plate was centrifuged at 4400 rpm for 10 min and the culture medium was then aspirated. Next, 100 µL of DMSO were added into each well to solubilize the formazan and the plate was shaken for 5 min. The colored formazan salt was measured by recording changes in absorbance at 570 nm using a microplate reader, a BioTek Synergy HT spectrophotometer. A reference wavelength of 630 nm was used to subtract background. Optical densities were analyzed as the quantity of formazan is directly proportional to the number of viable parasites. Data were analyzed with the statistical software GraphPad Prism.

Cell Viability
RAW 264.7 cells were maintained in Dulbecco's Modified Eagle Medium (DMEM) (Gibco, Invitrogen, NY, USA) supplemented with 10% fetal bovine serum (FBS) (Eurobio) and 100 IU/mL penicillin + 100 µg/mL streptomycin (Gibco, Invitrogen) at 37 • C in an atmosphere containing 5% CO 2 . The medium was renewed once a week. The viability was determined using MTT (thiazolyl blue tetrazolium bromide) dye assay as described previously for leishmanicidal activity assessment, with some variations. Here, 5 × 10 4 cells/well in a final volume of 100 µL were deposited into a 96-well plate in triplicate. Saponin (2 and 4 mg/mL) and untreated cells were used as the positive control and negative control, respectively. Schiff bases were dissolved in DMSO to obtain different serial concentrations (0.01-2000 µM). After 48 h exposure to the compounds, 10 µL/well of MTT (5 mg/mL MTT in PBS) was added, and the plate was incubated at 37 • C for 2 h in darkness. Cells were pelleted by centrifugation at 4400 rpm for 10 minutes and the media was removed. Next, 100 µL/well of DMSO were added and the absorbance at 570 nm was recorded. A reference wavelength of 630 nm was used for background subtraction.

Drop Test
For the drop test with S. cerevisiae, wild-type strain (background W303), and wild type and sod1 mutant cells (background BY4741) were grown overnight in YPAD media until an optical density of 0.5 at 600 nm was reached (equivalent to 1 × 10 7 cells/mL). Three serial dilutions were prepared and 5 µL of each was spotted onto a plate (approximately 1 × 10 5 , 1 × 10 4 , 1 × 10 3 , and 1 × 10 2 cells/spot). YPAD plates were supplemented with 200 µg/mL of each Schiff base, and one plate supplemented with 2% DMSO and one plate with YPAD without any drug were used as controls. Plates were incubated at 25 • C for 24, 48, and 72 h.

DPPH Radical Scavenging Assay
The stock solutions of the compounds were prepared by dissolving 3 in DMSO to a concentration of 4 mg/mL, then diluted with methanol to give a concentration of 400 µg/mL. This was used immediately. The experimental procedure was adapted from the literature [59]. Briefly, in a 96-well plate, 100 µL of solution of DPPH (2, 2-diphenyl-1-picrylhydrazyl) radical (0.2 mM in methanol) was added to 100 µL of methanolic solutions of 3a-j, 4, or 5, prepared as serial two-fold dilutions from the stock solution. Standards were also prepared in the same concentrations. The mixture was incubated in the dark at room temperature for 30 min, and the absorbance was read at 515 nm on a microplate reader, a BioTek Synergy H1 spectrophotometer.
The percentage of DPPH scavenging was then calculated by using the following formula: The antioxidant activity of the compound was expressed as IC 50 , which is defined as the concentration that could scavenge 50% of the DPPH free radicals.

Theoretical Prediction of ADME Properties and Bioactivity Scores
The predictions of ADME toxicity parameters and bioactivity scores were calculated as described previously [49,50,60,61]. The Osiris DataWarrior version 4.4.4 software on a Windows 10 operating system [50] was used to determine the drug-like properties of the synthesized compounds. These properties included the molecular polar surface area (PSA), octanol-water partition coefficient (clogP), aqueous solubility (clogS), number of rotatable bonds (nrotb), number of hydrogen donors (HBD), number of hydrogen acceptors (HBA), toxicities (mutagenic, tumorigenic, irritant, and reproductive), drug scores (DS), and drug likeness (DL). The bioactivity scores, including activities as a GPCR ligand, ion channel modulator, kinase inhibitor, nuclear receptor ligand, protease inhibitor, and enzyme inhibitor, were calculated using the Molinspiration Cheminformatics software.
Evaluation of biocidal/biostatic effect, Table S1. Theoretical prediction of ADME properties of some drugs used in chemotherapy.