New Pyrazolyl Thioureas Active against the Staphylococcus Genus

To meet the urgent need for new antibacterial molecules, a small library of pyrazolyl thioureas (PTUs) was designed, synthesized and tested against difficult-to-treat human pathogens. The prepared derivatives are characterized by a carboxyethyl functionality on C4 and different hydroxyalkyl chains on N1. Compounds 1a–o were first evaluated against a large panel of Gram-positive and Gram-negative pathogens. In particular, the majority of PTUs proved to be active against different species of the Staphylococcus genus, with MIC values ranging from 32 to 128 µg/mL on methicillin-resistant Staphylococcus strains, often responsible for severe pulmonary disease in cystic fibrosis patients. Time-killing experiments were also performed for the most active compounds, evidencing a bacteriostatic mechanism of action. For most active derivatives, cytotoxicity was evaluated in Vero cells, and at the tested concentrations and at the experimental exposure time of 24 h, none of the compounds analysed showed significant toxicity. In addition, favourable drug-like, pharmacokinetic and toxicity properties were predicted for all new synthesized derivatives. Overall, the collected data confirmed the PTU scaffold as a promising chemotype for the development of novel antibacterial agents active against Gram-positive multi-resistant strains frequently isolated from cystic fibrosis patients.


Introduction
The World Health Organization (WHO) estimates that there are 700,000 casualties per year worldwide due to drug-resistant infections with a projection of 10 million deaths by 2050 and a general cost for the global economy up to USD 100 trillion.Unfortunately, the WHO also noted that in 2020, none of the 43 antibiotics in clinical use had fully solved the problem of drug resistance [1].In this regard, the bacteria included in the ESKAPE group (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumonia, Acinetobacter baumannii, Pseudomonas aeruginosa, Enterobacter species) have been recognized by the Infectious Diseases Society of America (IDSA) as the most dangerous pathogens, due to their remarkable resistance to the most common conventional antibiotics.Additional concerns derived from the isolation of strains resistant to vancomycin and linezolid, considered the last line of defence against Gram-positive bacterial infections [2][3][4].The clinical management of infections caused by these exceptional pathogens is often complex and problematic, especially in patients hospitalized or suffering from concomitant diseases.
In this regard, cystic fibrosis (CF) is a genetic disorder associated with the production of sticky and thick mucus that accumulates in different organs, including the lungs.This condition facilitates the adhesion and growth of pathogens in the respiratory epithelium and exposes CF patients to respiratory infections, mainly caused by methicillin-resistant Staphylococcus aureus (MRSA).Noteworthy, chronic lung infections with this Gram-positive multi-resistant pathogen have been associated with more severe pulmonary disease and increase the decline in lung function in CF patients [5,6].Clinical studies on CF patients have shown that the incidence of methicillin-sensitive S. aureus (MSSA) is significantly lower than that of MRSA [7][8][9].
The pyrazole scaffold represents a privileged substructure in medicinal chemistry research and a number of pyrazole derivatives have been evaluated as effective compounds in different therapeutic areas [10][11][12][13][14][15][16], including infectious diseases [17,18] and also against MRSA [19].In previous studies, pyrazole derivatives I (Figure 1) showed significative antibacterial activity against Gram-positive antibiotic-resistant strains [20].The prepared molecules were characterized by a N1-hydroxy-2-phenylethyl chain, a carboxyethyl or tert-butyl substituents on C3 or C4 and a (thio)ureido moiety at position 5 of the pyrazole ring.The structure activity relationships (SARs) evidenced that a carboxyethyl group on C4 and a substituted thioureido function on C5 are key structural determinants for the antimicrobial activity of this series (derivatives II, Figure 1).Pyrazolyl thioureas (PTUs) II resulted inactive against all tested Gram-negative species but showed a good antibacterial potency against strains of the Staphylococcus genus resistant to methicillin and linezolid (MIC values between 32 and 64 µg/mL) and also against vancomycin-resistant Enterococcus strains (MIC values between 32 and 64 µg/mL).
In this regard, cystic fibrosis (CF) is a genetic disorder associated with the production of sticky and thick mucus that accumulates in different organs, including the lungs.This condition facilitates the adhesion and growth of pathogens in the respiratory epithelium and exposes CF patients to respiratory infections, mainly caused by methicillin-resistant Staphylococcus aureus (MRSA).Noteworthy, chronic lung infections with this Gram-positive multi-resistant pathogen have been associated with more severe pulmonary disease and increase the decline in lung function in CF patients [5,6].Clinical studies on CF patients have shown that the incidence of methicillin-sensitive S. aureus (MSSA) is significantly lower than that of MRSA [7][8][9].
The pyrazole scaffold represents a privileged substructure in medicinal chemistry research and a number of pyrazole derivatives have been evaluated as effective compounds in different therapeutic areas [10][11][12][13][14][15][16], including infectious diseases [17,18] and also against MRSA [19].In previous studies, pyrazole derivatives I (Figure 1) showed significative antibacterial activity against Gram-positive antibiotic-resistant strains [20].The prepared molecules were characterized by a N1-hydroxy-2-phenylethyl chain, a carboxyethyl or tert-butyl substituents on C3 or C4 and a (thio)ureido moiety at position 5 of the pyrazole ring.The structure activity relationships (SARs) evidenced that a carboxyethyl group on C4 and a substituted thioureido function on C5 are key structural determinants for the antimicrobial activity of this series (derivatives II, Figure 1).Pyrazolyl thioureas (PTUs) II resulted inactive against all tested Gram-negative species but showed a good antibacterial potency against strains of the Staphylococcus genus resistant to methicillin and linezolid (MIC values between 32 and 64 µg/mL) and also against vancomycin-resistant Enterococcus strains (MIC values between 32 and 64 µg/mL).The subject of this work is the design and synthesis of a novel small library of PTU molecules (compounds 1a-o) to confirm if this scaffold is a pharmaceutically relevant chemotype for the development of novel antibacterial agents, particularly active against Gram-positive species.The subject of this work is the design and synthesis of a novel small library of PTU molecules (compounds 1a-o) to confirm if this scaffold is a pharmaceutically relevant chemotype for the development of novel antibacterial agents, particularly active against Gram-positive species.

Results and Discussion
The novel pyrazole small library of PTUs (compounds 1a-o, Table 1) are characterized by: (1) a carboxyethyl function on C4 as previous derivatives II, (2)

Results and Discussion
The novel pyrazole small library of PTUs (compounds 1a-o, Table 1) are characterized by: (1) a carboxyethyl function on C4 as previous derivatives II, (2)   Novel PTU library 1 was evaluated for their antibacterial activity against several Gram-positive and Gram-negative species, using oxacillin as the reference compound.Time-killing experiments on MRSA strains and cytotoxicity evaluation on Vero cells were also performed for the most active compounds.Additionally, in silico prediction of pharmacokinetic properties, drug-likeness and toxicity (ADMET) of all novel pyrazole compounds was performed.

Chemistry
Novel pyrazole library was obtained following a consolidate procedure [20], following Scheme 1.Briefly, the condensation of the commercially available oxiranes 2a-d with hydrazine monohydrate led to the corresponding hydrazino-ethanols 3a-d, that were reacted with ethyl ethoxymethylene cyanoacetate in anhydrous toluene (3a) or absolute ethanol (3b-d) at 70-80°C to give the suitable pyrazole intermediates 4a-d as yellow solids [21].Finally, the thiourea moiety was introduced via a one-pot reaction in anhydrous THF for 12 h between the 5-amino-pyrazoles 4a-d and the proper benzoyl isothiocyanate 5ad, commercially available or prepared according to the literature method [22].Derivatives 1 were obtained in yields ranging from 35% to 94% ( Novel PTU library 1 was evaluated for their antibacterial activity against several Gram-positive and Gram-negative species, using oxacillin as the reference compound.Time-killing experiments on MRSA strains and cytotoxicity evaluation on Vero cells were also performed for the most active compounds.Additionally, in silico prediction of pharmacokinetic properties, drug-likeness and toxicity (ADMET) of all novel pyrazole compounds was performed.

Chemistry
Novel pyrazole library was obtained following a consolidate procedure [20], following Scheme 1.Briefly, the condensation of the commercially available oxiranes 2a-d with hydrazine monohydrate led to the corresponding hydrazino-ethanols 3a-d, that were reacted with ethyl ethoxymethylene cyanoacetate in anhydrous toluene (3a) or absolute ethanol (3b-d) at 70-80 • C to give the suitable pyrazole intermediates 4a-d as yellow solids [21].Finally, the thiourea moiety was introduced via a one-pot reaction in anhydrous THF for 12 h between the 5-amino-pyrazoles 4a-d and the proper benzoyl isothiocyanate 5a-d, commercially available or prepared according to the literature method [22].Derivatives 1 were obtained in yields ranging from 35% to 94% (Table 1) as yellow oils or crystalline white solids.

Antibacterial Activity
The antibacterial potency of PTU 1a-o was evaluated against a panel of fifteen bacterial isolates (Table 2), representative of clinically relevant Gram-positive (eleven strains) and Gram-negative (four strains) species including four S. aureus strains (MRSA), three S. epidermidis isolates (two MRSE and one resistant to methicillin and linezolid), two E. faecalis strains (one vancomycin-sensitive and one vancomycin resistant, VRE), two E. faecium isolates (one vancomycin-sensitive and one VRE), two E. coli isolates resistant to carbapenem (one was a New Delhi metallo-β-lactamase (NDM)-producing isolate) and two P. aeruginosa (multidrug-resistant isolates, MDR) strains.
Table 2. MIC values (expressed as µg/mL) of PTU 1 and reference compound oxacillin (oxa) against bacteria of the Gram-positive and Gram-negative species, obtained from experiments carried out at least in triplicate.* denotes resistant to methicillin; ** denotes resistant to methicillin and linezolid; # denotes resistance to vancomycin; NDM: New Delhi metallo-beta-lactamase producer; MDR: multidrug-resistant strain; CR carbapenem resistant.In bold are evidenced the most active compounds.
All tested compounds resulted inactive (MIC > 128 µg/mL) against the Gram-negative and Enterococcus genus.Conversely, the majority of derivatives (10 out of 15) showed a widespread activity against the most clinically relevant Staphylococcus species (i.e., S. aureus MRSA and S. epidermidis MRSE), with MIC values lower in some cases than those

Antibacterial Activity
The antibacterial potency of PTU 1a-o was evaluated against a panel of fifteen bacterial isolates (Table 2), representative of clinically relevant Gram-positive (eleven strains) and Gram-negative (four strains) species including four S. aureus strains (MRSA), three S. epidermidis isolates (two MRSE and one resistant to methicillin and linezolid), two E. faecalis strains (one vancomycin-sensitive and one vancomycin resistant, VRE), two E. faecium isolates (one vancomycin-sensitive and one VRE), two E. coli isolates resistant to carbapenem (one was a New Delhi metallo-β-lactamase (NDM)-producing isolate) and two P. aeruginosa (multidrug-resistant isolates, MDR) strains.
Table 2. MIC values (expressed as µg/mL) of PTU 1 and reference compound oxacillin (oxa) against bacteria of the Gram-positive and Gram-negative species, obtained from experiments carried out at least in triplicate.* denotes resistant to methicillin; ** denotes resistant to methicillin and linezolid; # denotes resistance to vancomycin; NDM: New Delhi metallo-beta-lactamase producer; MDR: multidrug-resistant strain; CR carbapenem resistant.In bold are evidenced the most active compounds.

Strain
All tested compounds resulted inactive (MIC > 128 µg/mL) against the Gram-negative and Enterococcus genus.Conversely, the majority of derivatives (10 out of 15) showed a widespread activity against the most clinically relevant Staphylococcus species (i.e., S. aureus MRSA and S. epidermidis MRSE), with MIC values lower in some cases than those of oxacillin, used as a reference compound (32-128 mg/mL against 128-512 mg/mL).Interestingly, PTUs 1i and 1m were identified as the most active derivatives of the series because they showed MIC values in a close range (32-64 µg/mL) against seven of the considered Gram-positive isolates.Also compounds 1d, 1e, 1k and 1n resulted active against six Gram-positive strains.
To further define the antibacterial properties of the series, PTUs 1a,d,e,h,i,m were selected as representative examples of differently N1 substituted pyrazoles (1a and 1d, R = Me; 1e and 1h, R = Et; 1i, R = nPr; 1m, R = nBu) and tested against additional Staphylococcus species for a total of 14 isolates (Table 3).All analysed compounds proved to be ineffective against S. saprophyticus, S. warneri and S. simulans, but showed relevant antibacterial activity against S. lugdunensis and S. auricularis species (MIC value range = 16-32 µg/mL).Moreover, derivative 1d specifically inhibited the proliferation of methicillin-resistant S. capitis 71 strain (MIC = 64 µg/mL), whereas compound 1h affected the growth of S. hominis 124, without influencing the S. hominis 125.Finally, derivatives 1d,e,h,i showed similar activity against S. haemoliticus 115 isolate, resulting ineffective against the other two considered S. haemoliticus strains.Regarding S. auricularis, compounds 1e,i,m resulted the most potent, displaying MIC values of 16 µg/mL.To investigate whether PTUs act as bacteriostatic or bactericidal, time-killing experiments were carried out on MRSA isolates, very relevant in the daily clinical practice of CF patients.Compounds 1c,d,n,h were selected as representative examples of the chemical diversity of the series and tested against four different MRSA strains (i.e., S. aureus 17, 18, 187 and 195).The experiments were carried out at concentrations four times the MIC values.As exemplified in Figure 2, the four tested compounds proved to act as a bacteriostatic agent, because they were all able to maintain virtually unchanged (10 5 CFU/mL) the concentration of the initial bacterial inocula for all 24 h of this study.Similar trends were obtained for all the analysed MRSA strains.

Cytotoxicity Evaluation
To verify if PTUs here reported are characterized by a cytotoxicity activity, selected compounds 1c (R =methyl) and 1n (R = butyl), chosen among the most active ones and as representative examples of the chemical diversity of the series, were tested on Vero cells at the most representative MIC values obtained (32 µg/mL and 64 µg/mL for both compounds, Figure 3). .At a concentration of 32 µg/mL, tested compounds showed the highest viability rates.According to ECVAM guidelines, compounds 1c and 1n can therefore be considered low toxic (at 32 µg/mL) and moderately toxic (at 64 µg/mL).

Pharmacokinetic Properties, Drug-Likeness and Toxicity (ADMET) Prediction
To further characterize the pharmaceutical potentials of PTUs 1, the drug-likeness and pharmacokinetic properties of the series were calculated using the SwissADME program [23].Derivatives IIa,b (Figure 1) were used as reference molecules (Table 4).

Cytotoxicity Evaluation
To verify if PTUs here reported are characterized by a cytotoxicity activity, selected compounds 1c (R = methyl) and 1n (R = butyl), chosen among the most active ones and as representative examples of the chemical diversity of the series, were tested on Vero cells at the most representative MIC values obtained (32 µg/mL and 64 µg/mL for both compounds, Figure 3).

Cytotoxicity Evaluation
To verify if PTUs here reported are characterized by a cytotoxicity activity, sele compounds 1c (R =methyl) and 1n (R = butyl), chosen among the most active ones an representative examples of the chemical diversity of the series, were tested on Vero c at the most representative MIC values obtained (32 µg/mL and 64 µg/mL for both c pounds, Figure 3).

Pharmacokinetic Properties, Drug-Likeness and Toxicity (ADMET) Prediction
To further characterize the pharmaceutical potentials of PTUs 1, the drug-liken and pharmacokinetic properties of the series were calculated using the SwissADME gram [23].Derivatives IIa,b (Figure 1) were used as reference molecules (Table .At a concentration of 32 µg/mL, tested compounds showed the highest viability rates.According to ECVAM guidelines, compounds 1c and 1n can therefore be considered low toxic (at 32 µg/mL) and moderately toxic (at 64 µg/mL).

Pharmacokinetic Properties, Drug-Likeness and Toxicity (ADMET) Prediction
To further characterize the pharmaceutical potentials of PTUs 1, the drug-likeness and pharmacokinetic properties of the series were calculated using the SwissADME program [23].Derivatives IIa,b (Figure 1) were used as reference molecules (Table 4).Collectively, this in silico evaluation predicted for PTUs 1 favourable physiochemical and DMPK properties that, in some case, would result better than those of previously described compounds II.
In detail, the replacement of the II phenyl group with aliphatic, linear chains (namely, methyl, ethyl, propyl or butyl R substituents, Table 1) led to an increase in the Csp 3 fraction with an improvement of the predicted bioavailability.Respect to previous II, in most cases the number of rotatable bonds of H bond acceptors and H bond donors are the same of previous II.Moreover, 1 and II would display the similar polarity, as indicated by the topological polar surface area (TPSA) descriptor (137.57Å 2 ).This descriptor has proved to be indispensable for predicting the permeability of a molecule towards biological membranes.It has in fact been demonstrated that when the TPSA value is greater than 140 Å 2 , the molecules have difficulty permeating the barriers; on the contrary, when it is less than 140 Å 2 , the passage through the lipophilic barriers is easier.
The different N1 substituents of the pyrazole ring would also affect the lipophilicity of the compounds, being the logPvalues within the desired range (logP between −0.7 and +5.0) for all analysed PTUs.Except for 1m-o (N1 hydroxyhexyl derivatives), all compounds were predicted to be water soluble rather than moderately soluble as compounds II.
Compounds 1a-i,1m would be highly absorbed in the gastrointestinal tract, whereas derivatives 1j-l and 1n-o, characterized by a fluoro-substituted phenyl ring on thioureido function and a more embedded chain on N1 (hydroxypentyl and hydroxyexyl) would be poorly absorbed.As derivatives II, the novel compounds would not be able to pass the blood-brain barrier (BBB) and enter in the central nervous system.Furthermore, unlike the previous II, derivatives 1 would be substrates of the P-gp efflux pump.
The CYP inhibition properties of compounds 1 would be different from that predicted for derivatives II.Thus, 1A2 and 2D6 isoforms would not be affected by PTUs 1, while 2C19, 2C9 and 3A4 enzymes would be inhibited by derivatives 1e-o.
As derivatives II, no violations of the Lipinski rules have been identified for PTUs 1 that would not display any pan-assay interference compounds (PAINS) alerts.According to the Brenk filters, the thiourea thiocarbonyl group on the C5 position was spotted as a problematic fragment [24].Collectively, this in silico evaluation predicted for this new PTU library good physicochemical, lipophilicity and water solubility properties; in some cases, better than previous II.
In addition, also predicted acute toxicity (lethal dose, LD 50 ) for rats after oral administration (Table 5) was calculated using the Advanced Chemistry Development (ACD) Percepta platform (ACD/Percepta Platform.Advanced Chemistry Development, Inc.; Toronto, ON, Canada: 2015).Reliability index values are shown as R.I. (values higher than 0.30 are ranked as reliable by the software).The software prediction is performed based on the software implemented training libraries, which include experimentally determined pharmacokinetic and safety properties for different series of compounds.
Notably, the newly developed compounds 1a-o were predicted to have LD 50 values in the 1600-3300 mg/kg range.
Elemental analysis was determined with an elemental analyser EA 1110 (Fison-Instruments, Milan, Italy); compounds have been considered pure when the difference between calculated and found values is ± 0.4 (Table S1 in Supporting Material).Hydrazines 3 and pyrazole intermediates 4a-c were prepared according to the already published procedures [21,26].Benzoyl isothiocyanates 5b-d were prepared according to the literature method [20,22].

General Synthesis of 5-Thioureido Pyrazoles 1a-o
The proper 5-amino-pyrazoles 4a-d (1 mmol) and the suitable benzoyl isothiocyanate 5a-d (1 mmol), commercially available or previously prepared modifying the literature method [22], in anhydrous THF (10 mL) was refluxed for 12 h.After cooling to room temperature, the solution was concentrated under reduced pressure and the crude was dissolved in ethyl acetate (20 mL); the organic phase was washed with 6N HCl (10 mL), then with NaHCO 3 saturated solution (10 mL) and water (10 mL), dried (MgSO 4 ) and concentrated under reduced pressure.The crude was purified by flash chromatography using a mixture of diethyl ether/ligroin (b.p. 40-60 • C) 3/1 as the eluent.

Bacterial Species Considered in This Study
A total of 29 clinical strains were used in this study, all belonging to a collection obtained from the School of Medicine and Pharmacy of University of Genoa (Italy), identified by VITEK ® 2 (Biomerieux, Firenze, Italy) or the matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometric technique (Biomerieux, Firenze, Italy).
Particularly, fifteen strains were used to preliminary test the 15 new PTU derivatives, which were eleven Gram-positive strains (four MRSA Staphylococcus aureus strains), threeStaphylococcus epidermidis isolates (two MRSE and one resistant to methicillin and linezolid), two Enterococcus faecalis strains (one vancomycin-sensitive and one vancomycinresistant (VRE)), two Enterococcus faecium isolates (one vancomycin-sensitive and one (VRE)) and four Gram-negative isolates (two Escherichia coli isolates, one was a New Delhi metallo-β-lactamase (NDM)-producing isolate and both carbapenem resistant, and two Pseudomonas aeruginosa MDR isolates).
In addition, fourteen isolates of the Staphylococcus genus were also evaluated, including one S. saprophyticus, two S. capitis (one was resistant to methicillin), one S. warneri (resistant to methicillin), two S. simulans (both resistant to methicillin), two S. lugdunensis (one was resistant to methicillin), three S. haemoliticus (two were resistant to methicillin), two S. hominis (both resistant to methicillin) and one S. auricularis (resistant to methicillin).

Determination of the Minimal Inhibitory Concentrations (MICs)
MIC values were determined following the microdilution procedures detailed by the European Committee on Antimicrobial Susceptibility Testing EUCAST [27], as also reported in our previous studies [20,28].Briefly, serial two-fold dilutions in Mueller-Hinton (MH) broth (Merck, Darmstadt, Germany) of all the fifteen samples (dissolved in DMSO), ranging from 128 to 1 µg/mL, were used.DMSO was also tested as a control to verify the absence of antibacterial activity of the solvent used for the experiments.Cultures of all the selected bacteria, after overnight incubation, were diluted to yield a standardized inoculum of 1.5 × 10 8 CFU/mL.Appropriate aliquots of each suspension were added to 96-well microplates containing dilutions of compounds to be tested to yield a final concentration of about 5 × 10 5 cells/mL.After 24 h of incubation at 37 • C, the lowest concentration of sample that prevented visible growth was recorded as the MIC.All MICs were obtained at least in triplicate and results were expressed reporting the modal value; that is, the value that has been observed most frequently.In case of equivocal or not clear results, more than three determinations of MICs were carried out.

Killing Curves
Killing curve assays for most interesting compounds (1a, 1e, 1i and 1m) were performed on the four MRSA isolates selected for this study as previously reported [20,28].
A mid logarithmic phase bacterial culture was diluted in MH broth (10 mL) containing 4× MIC of all the compounds to give a final inoculum of 3.0 × 10 5 CFU/mL.The same inoculum was added to MH broth, as a growth control.Tubes were incubated at 37 • C with constant shaking for 24 h.Samples of 0.20 mL from each tube were removed at 0, 2, 4, 6 and 24 h, diluted appropriately with a 0.9% sodium chloride solution to avoid carryover of compounds being tested, plated onto MH plates and incubated for 24 h at 37 • C. Growth controls were run in parallel.The percentage of surviving bacterial cells was determined for each sampling time by comparing colony counts with those of standard dilutions of the growth control.Results have been expressed as log 10 of viable cell numbers (CFU/mL) of of surviving bacterial cells over a 24 h period.All time-kill curve experiments were performed in triplicate.

Maintenance of Cell Cultures
Vero cell line, isolated from kidney epithelia cells extracted from Cercopithecus aethiops, was certified by STR DNA profile analysis by Biological Bank, a Core Facility of the IRCCS San Martino University Hospital-IST National Institute for Cancer Research (Genoa, Italy).Vero cells were routinely cultured al 37 • C under 5% CO 2 in a DMEM medium (Euroclone, Milan, Italy) plus 10% heat inactivated FBS serum (Euroclone, Milan, Italy).No antibiotic or anti-fungine solutions were added to the standard or experimental medium in order to avoid any potential interference of these drugs with the experimental conditions.The medium was changed every 2 to 3 days and cells were sub-cultured by TripLE TM Express (Invitrogen, Life Technologies, Carlsbad, CA, USA) treatment when the original flask was approximately 75% confluent.All cell cultures were found to be mycoplasma-free during regular checks with the Reagent Set Mycoplasma Euroclone (Euroclone, Milan, Italy).

Cell Viability Index MTT Test
At the end of each experimental treatment, the cell viability was assessed by thiazolyl blue tetrazolium dye reduction assay (MTT) (Euroclone, Milan, Italy) [29].The optical densities (OD) of the dissolved formazan crystals (for the MTT test) were determined spectrophotometrically at 570 nm.
A chemical compound was considered toxic if the cell viability was reduced by 15% compared to untreated cultures, according to the ECVAM's guidelines testing any cytotoxic effects of the compounds and in parallel for measuring cell proliferation, according to manufacturer's instructions.

Conclusions
The obtained results pointed at the PTU scaffold as a pharmaceutically relevant chemotype for the development of novel antibacterial agents active against Gram-positive species.In fact, novel PTUs here reported proved to be equally or more active that previous derivatives I. Compounds 1a-o evidenced interesting activity towards Gram-positive resistant pathogen, often associated with severe pulmonary disease in CF patients [30][31][32][33][34]; in fact, several novel PTUs proved to be effective against different species of the Staphylococcus genus, with MIC values ranging from 32 to 128 µg/mL on MRSA and MRSE strains.In addition, time-killing experiments confirm the bacteriostatic actions of this class of compounds.
For all the synthesized compounds, favourable pharmacokinetic properties were calculated, evidencing for a major part of PTUs' good ADMET properties.Finally, considering the preliminary cytotoxicity results obtained on Vero cells, it is reasonable to assume that the PTU library here reported have low toxicity.
Collected data allow us to draw the following SAR considerations regarding N1 and thiourea moiety substitution (Figure 4): (1) the replacement of hydroxy-2-phenylethyl chain with hydroxyalkyl substituent at N1 improves the antimicrobial activity of the compounds, but make PTUs here reported P-gp efflux pump substrates; to manufacturer's instructions.

Conclusions
The obtained results pointed at the PTU scaffold as a pharmaceutically relevant chemotype for the development of novel antibacterial agents active against Gram-positive species.In fact, novel PTUs here reported proved to be equally or more active that previous derivatives I. Compounds 1a-o evidenced interesting activity towards Gram-positive resistant pathogen, often associated with severe pulmonary disease in CF patients [30][31][32][33][34]; in fact, several novel PTUs proved to be effective against different species of the Staphylococcus genus, with MIC values ranging from 32 to 128 µg/mL on MRSA and MRSE strains.In addition, time-killing experiments confirm the bacteriostatic actions of this class of compounds.
For all the synthesized compounds, favourable pharmacokinetic properties were calculated, evidencing for a major part of PTUs' good ADMET properties.Finally, considering the preliminary cytotoxicity results obtained on Vero cells, it is reasonable to assume that the PTU library here reported have low toxicity.
Collected data allow us to draw the following SAR considerations regarding N1 and thiourea moiety substitution (Figure 4): (1) the replacement of hydroxy-2-phenylethyl chain with hydroxyalkyl substituent at N1 improves the antimicrobial activity of the compounds, but make PTUs here reported P-gp efflux pump substrates; (2) the presence of the N1 hydroxyexyl chain at N1 reduced solubility (compounds 1m-o); (3) the ortho-fluoro substitution of the thiourea benzoyl ring is detrimental for activity (as in derivatives 1b,f,j), whereas para and meta fluoro-substituted compounds evidenced a potency comparable to their non-substituted analogues (1a,e,i,m); (4) the simultaneous presence of the fluorine atom and longer hydroxyalkyl chain (hydroxypentyl 1j-l or hydroxyexyl 1n,1o) increases the Csp3 fraction with an improvement of the predicted bioavailability; (5) the carboxyethyl function at C4 position resulted relevant for biological activity, as previously evidenced for derivatives IIa,b.
Additional chemical modifications on the C4 and C3 pyrazole nucleus (as in previous derivatives I [20]) will be performed to extend the SAR consideration about the PTU chemotype.As in our previous studies on pyrazole derivatives I and II [20], these collected results supported that, upon nano-formulation with proper polymer matrices, the new Tiourea moiety substitution: • ortho-fluoro substitution detrimental for activity; • Non substituted derivatives and para and meta fluoro-substituted compounds showed comparable activity; • the simultaneous presence of fluorine atom and longer hydroxyalkyl improve predicted bioavailability.

N1-substitution:
• hydroxyalkyl substituent at N1 improves the antimicrobial activity of the compounds; • substitution of phenyl group with alkyl chain could make PTUs Pgp efflux pump substrate; • hydroxyexyl chain at N1 reduced predicted solubility; C4 substitution: • carboxyethyl function fundamental for microbiological activity Additional chemical modifications on the C4 and C3 pyrazole nucleus (as in previous derivatives I [20]) will be performed to extend the SAR consideration about the PTU chemotype.
As in our previous studies on pyrazole derivatives I and II [20], these collected results supported that, upon nano-formulation with proper polymer matrices, the new synthesized compounds could provide novel pyrazole-based drug delivery systems with enhanced and enlarged-spectrum of antibacterial activity, particularly against Gram-positive MRSA and MRSE species.

Figure 1 .
Figure 1.Structure of previous pyrazoles I and II and workflow of the applied strategies leading to the design of a novel small PTU library (compounds 1a-o).

Figure 1 .
Figure 1.Structure of previous pyrazoles I and II and workflow of the applied strategies leading to the design of a novel small PTU library (compounds 1a-o).

Figure 3 .
Figure 3. Cytotoxicity of compounds 1c (in light blue) and 1n (in blue) performed at concentrations 32 and 64 g/mL on Vero cells.Data, expressed as percentage of viability versus untreated cultures and extrapolated by MTT assay, are the means ± SD of three separate experiments performed in triplicate.* = p < 0.01 versus untreated cultures (ANOVA and Dunnett test).At a concentration of 32 µg/mL, tested compounds showed the highest viability rates.According to ECVAM guidelines, compounds 1c and 1n can therefore be considered low toxic (at 32 µg/mL) and moderately toxic (at 64 µg/mL).

Figure 3 .
Figure 3. Cytotoxicity of compounds 1c (in light blue) and 1n (in blue) performed at concentrat 32 and 64 g/mL on Vero cells.Data, expressed as percentage of viability versus untreated cult and extrapolated by MTT assay, are the means ± SD of three separate experiments performe triplicate.* = p < 0.01 versus untreated cultures (ANOVA and Dunnett test).At a concentration o µg/mL, tested compounds showed the highest viability rates.According to ECVAM guidel compounds 1c and 1n can therefore be considered low toxic (at 32 µg/mL) and moderately toxi 64 µg/mL).

Figure 3 .
Figure 3. Cytotoxicity of compounds 1c (in light blue) and 1n (in blue) performed at concentrations 32 and 64 µg/mL on Vero cells.Data, expressed as percentage of viability versus untreated cultures and extrapolated by MTT assay, are the means ± SD of three separate experiments performed in triplicate.* = p < 0.01 versus untreated cultures (ANOVA and Dunnett test).At a concentration of 32 µg/mL, tested compounds showed the highest viability rates.According to ECVAM guidelines, compounds 1c and 1n can therefore be considered low toxic (at 32 µg/mL) and moderately toxic (at 64 µg/mL).
(2) the presence of the N1 hydroxyexyl chain at N1 reduced solubility (compounds 1m-o); (3) the ortho-fluoro substitution of the thiourea benzoyl ring is detrimental for activity (as in derivatives 1b,f,j), whereas para and meta fluoro-substituted compounds evidenced a potency comparable to their non-substituted analogues (1a,e,i,m); (4) the simultaneous presence of the fluorine atom and longer hydroxyalkyl chain (hydroxypentyl 1j-l or hydroxyexyl 1n,1o) increases the Csp3 fraction with an improvement of the predicted bioavailability; (5) the carboxyethyl function at C4 position resulted relevant for biological activity, as previously evidenced for derivatives IIa,b.

Table 1 .
Chemical structure, melting point and reaction yield of novel PTU derivatives 1a-o.

Table 1 .
Chemical structure, melting point and reaction yield of novel PTU derivatives 1a-o.

Table 1 )
as yellow oils or crystalline white solids.

Table 3 .
MIC values of some PTUs 1 and reference compound oxacillin (oxa) against other Staphylococcus species, obtained from experiments carried out at least in triplicate, expressed as µg/mL.* denotes resistant to methicillin.In bold are evidenced the most active compounds.

Table 4 .
Predicted pharmacokinetics and drug-like properties of compounds