N-Acylated Ciprofloxacin Derivatives: Synthesis and In Vitro Biological Evaluation as Antibacterial and Anticancer Agents

A novel series of N-acylated ciprofloxacin (CP) conjugates 1–21 were synthesized and screened as potential antimicrobial agents. Conjugates 1 and 2 were 1.25–10-fold more potent than CP toward all Staphylococci (minimal inhibitory concentration 0.05–0.4 μg/mL). Most of the chloro- (3–7), bromo- (8–11), and CF3-alkanoyl (14–16) derivatives expressed higher or comparable activity to CP against selected Gram-positive strains. A few CP analogues (5, 10, and 11) were also more effective toward the chosen clinical Gram-negative rods. Conjugates 5, 10, and 11 considerably influenced the phases of the bacterial growth cycle over 18 h. Additionally, compounds 2, 4–7, 9–12, and 21 exerted stronger tuberculostatic action against three Mycobacterium tuberculosis isolates than the first-line antitubercular drugs. Amides 1, 2, 5, 6, 10, and 11 targeted gyrase and topoisomerase IV at 2.7–10.0 μg/mL, which suggests a mechanism of antibacterial action related to CP. These findings were confirmed by molecular docking studies. In addition, compounds 3 and 15 showed high antiproliferative activities against prostate PC3 cells (IC50 2.02–4.8 μM), up to 6.5–2.75 stronger than cisplatin. They almost completely reduced the growth and proliferation rates in these cells, without a cytotoxic action against normal HaCaT cell lines. Furthermore, derivatives 3 and 21 induced apoptosis/necrosis in PC3 cells, probably by increasing the intracellular ROS amount, as well as they diminished the IL-6 level in tumor cells.


INTRODUCTION
Bacterial inflammations, evoked mainly by Gram-positive and Gram-negative isolates, represent a serious health threat, being responsible for the majority of healthcare-associated infections that lead to health service overload and high mortality. The latest is caused by the growing fluoroquinolone resistance of pathogens to the current therapeutic options, involving methicillin-resistant or vancomycin-resistant Staphylococci, as well as Streptococci. 1 The resistance develops mostly by amino acid alterations at various locations on subunit A of bacterial gyrase, located at the binding site of the enzyme, that diminishes the drug affinity to the gyrase−DNA complex. 1 This mechanism of mutation was observed for Streptococcus pneumoniae isolates. 2,3 The second reason of the drug resistance is the mutation in gene coding membrane proteins, responsible for the transport of a drug into the bacterial cell. 4 That alteration reduces the level of transmembrane protein molecules or stimulates quinolone removal through efflux pumps. The above types of resistance were observed for some Gram-negative rods and Staphylococcus aureus strains, 1,3 as well as for cancer cells. 5 Fluoroquinolones, including ciprofloxacin (CP), moxifloxacin (MXF), and gatifloxacin, are a group of synthetic antibiotics that either inhibit the action of bacterial DNA gyrase, adenosine triphosphate-hydrolyzing topoisomerase II, and/or prevent from the detachment of gyrase from DNA. 6,7 Topoisomerase II allows the relaxation of supercoiled DNA, by breaking, crossing over, and finally resealing both strands of the DNA chain. 8 Quinolone chemotherapeutics are also able to inhibit bacterial topoisomerase IV, which participates in chromosomal DNA partition during cell division. 9,10 The structural manipulations of the second-generation quinolones are a successful way of searching for new and more effective antibacterial drug candidates. A key advance in the development of CP-derived compounds is the enhanced activity against the isolates of S. aureus, S. pneumoniae, and group A streptococci. 11,12 Several essential positions of the quinolone skeleton are subjects for modifications. Among them, fixed substituents at C-3 and C-6 are necessary for enhanced inhibition of DNA gyrase. The pharmacophore of CP necessary for powerful antimicrobial properties is the 4pyridone nucleus with a carboxylic group attached to the C-3 position. Modification of this area, e.g., reduction to aldehyde or esterification, gave compounds with decreased antimicrobial potency, although an alteration of the 3-carboxylic functionality can be used for switching its activity to anticancer. 13 On the other hand, when the carboxylic group is replaced with an isothiazole moiety, active antibacterial agents with weak solubility or improved toxicity were obtained. 14 Interestingly, copper(II)−CP complexes, derived from hydroxamic acid, enhanced the permeability of S. aureus cells and affected proteins involved in virulence, the synthesis of nucleotides, and DNA repair mechanisms. 15 The highly electronegative fluorine atom at C-6 seems to be optimal and crucial for the binding with bacterial DNA topoisomerase. Moreover, it was shown that the insubstantial aliphatic (e.g., ethyl) or alicyclic (cyclopropyl) functionality at the N-1 position of the quinolone is responsible for improved growth inhibitory properties. 16 The same beneficial effect against Gram-positive isolates in comparison with CP was achieved by introduction of difluorophenyl or an oxetane ring, as well as the closure of both N-1 and C-8 atoms by an alkoxyamino moiety. 9,17 Alkylation of the C-8 location favors in vivo parameters of the drug and raises its activity toward anaerobic rods. 18 Addition of a methoxy group here, with concurrent alteration of the C-7 cyclic substituent, resulted in fourth-generation antibiotics, such as MXF. However, the newer generation of fluoroquinolones is not as potent as CP against Pseudomonas aeruginosa isolates. 12 Similarly, the examples of third-generation quinolones (grepafloxacin and sparfloxacin) show that small substituents at C-5, such as methyl or amino functionalities, resulted in markedly increased activity. 17,19 However, the most typical and reasonable strategy for the modification of the CP structure is the substitution of its C-7 position. The presence of an amino cyclic group at this location is the main feature responsible for improved action toward Gram-negative organisms and favorable pharmacokinetics. 20,21 The chemical adjustments of unsubstituted piperazine in the CP skeleton included attachments of at least one methyl group (thirdgeneration grepafloxacin and sparfloxacin), amino pyrrolidines (fourth-generation trovafloxacin), or a bicyclic ring system (MXF) 21 but also N-acylation of the ring. Just among the amide analogues of CP, isatin, 22 and sulfonamide, 23 the C-7 derivatives exhibited excellent activity against standard and clinical Gram-positive and Gram-negative isolates. It was proved that the mentioned sulfonamide-based fluoroquinolones targeted both bacterial DNA gyrase and topoisomerase IV, exerting low or even no side effects on the central nervous system in mice. 23 The modification of the piperazine nitrogen, in order to increase the lipophilicity and change the biological potency, was also described. The positive effects of the substitution of the piperazinyl ring by (hetero)aryl-triazole 24−27 and alkylaryl 28 fragments, its alkylation or acylation with alicyclic, alkyl, or halogenoalkyl moieties, 29−31 on the overall antibacterial profile of CP were investigated. Moreover, conjugates of CP with cell-penetrating peptides were found to be inhibitors of the yeast type II DNA topoisomerase. 32 Additionally, CP has been recommended as the second-line antitubercular agent, especially when resistance or intolerance to the first-line drugs appeared. 10,33 However, when used as monotherapy or as the only effective tuberculostatic in a multidrug series, the risk of fluoroquinolone resistance rapidly develops. 34 Recently, it was described that amide-tethered CP−isatin hybrids are severalfold more effective against multidrug-resistant Mycobacterium tuberculosis 35 or standard H 37 Rv 22 pathogens when compared with the parent drug. The modification of the piperazine moiety via a benzodiazepine moiety gave a new nontoxic agent, exerting antitubercular activity comparable to the quinolone antibiotic, with gyrase inhibition as a possible mechanism of action. 36 What is more, the anticancer activities of fluoroquinolones are an effect of restraining of eukaryotic topoisomerase IIα, the analogue of DNA gyrase. It results in the formation of doublestrand breaks in nucleic acids and incomplete DNA synthesis, which lead to S-phase arrest in a cancer cell. 37 Since the affinity of CP to the eukaryotic enzyme is weaker than to the prokaryotic homologue, an anticancer treatment requires higher drug dosages in comparison with quantities used in bacterial infections. Therefore, the apoptosis-inducing dose of this quinolone, effective in antitumor therapies, equals at least 200−300 μg/mL. 38 The anticancer properties of CP have been studied previously against various types of mammalian pathological cells, including bladder, colorectal, leukemia, and human prostate cancer cell lines. Cancerous cells exhibited different susceptibilities to this chemotherapeutic. 39−41 Hepatocellular cancer cells (Hep G2) remained quite resistant, whereas colon carcinomas (SW-403) were found to be more sensitive. 39 It has been estimated recently that the fluoroquinolone derivative induces oxidative stress in hepatoma-derived cells, increases activation of caspases involved in different pathways of cell death, and imposes double-strand breakages in the cellular DNA. 37 Moreover, CP triggers apoptosis of human triple-negative breast cancer (MDA-MB-231) cells by leading to a loss of the mitochondrial transmembrane potential and stimulation of the cell cycle arrest at the S-phase. 6 Interestingly, the effect of this antibiotic in prostate cancer (PC3) cells was mediated by the cell cycle arrest at the S-G2/M phase 41 and by inhibition of NF-κB binding to DNA. 42 It also gives a sensitizing effect and inhibits the efflux function of the ATP-binding cassette (ABC) transporter, decreasing the IC 50 values of known chemotherapeutics in ABCB-1 overexpressing cells. 5 Methylnaphthalenyl derivatives of CP induced the percentage of cells at G2/M phases, the percentage of cells in early apoptosis, and the level of the active caspase-3 in prostate cancer (PC-3) cells. 28  urea-linked CP−chalcone compounds. 44 An apoptosis-inducing effect and significant reduction in IL-6 levels were also observed after incubation of PC3 cells with CP fatty acid conjugates 45 or its alkanoyl connections. 46 We have initiated a screening program to search for new fluoroquinolone-derived compounds as potential antimicrobial agents with enhanced antitumor activity. 45,47 In this study, the homogeneous series based on the CP pharmacophore was synthesized, by single modification on its C-7 position, and tested for their antibacterial, antitubercular, and cytotoxic properties, with determination of possible mechanisms of action.

RESULTS AND DISCUSSION
2.1. Chemistry. The aim of the research was to synthesize a versatile series of amide derivatives of CP by coupling its C-7 piperazinyl group with halogenated acyl chlorides, differing in hydrocarbon moiety and the type of halogen (Scheme 1). The conjugated acyl residues were unsaturated (1), alicyclic (2), or saturated (3−21). Aliphatic derivatives possessed either short (3−12 and 14−20) or medium (13 and 21) hydrocarbon chains. The carbon linker ended with chlorine (3−7), bromine (8−13), or trifluoromethyl (14−16) residue. The synthesis of presented compounds was conducted in a one-step reaction under mild conditions. After addition of acyl chloride to CP and triethyl amine suspension (at 2−5°C) in dichloromethane, the resulting solution was stirred at room temperature for 3 h. The low yields of synthesis of short-chain amides 3 and 8 (28−34%) resulted from their poor solubility in the solvent and problems with separation and purification. Derivatives with a longer aliphatic chain (4−7 and 10−13) were obtained with higher yields (41−86%). The TLC analysis of reactions showed that side products were formed. After separation by column chromatography, the MS and NMR analyses confirmed that double-condensation products 17−21 were synthesized, with yields below 10%. On the other hand, the reaction of CP with 3-chloropropionyl chloride yielded the unsaturated derivative 1, instead of the dimeric product. Amide 2 was formed with a yield of 5% as a product of degradation of compound 10 during the purification process by column chromatography.
2.2. Biological Studies. 2.2.1. In Vitro Antibacterial Activity. The antibacterial properties of the synthesized fluoroquinolone conjugates 1−21 were investigated by employing standard Gram-positive bacteria (S. aureus NCTC 4163, ATCC 25923, ATCC 6538, and ATCC 29213 and Staphylococcus epidermidis ATCC 12228 and ATCC 35984) and Gram-negative rods (P. aeruginosa ATCC 15442 and Escherichia coli ATCC 25922). As presented in Table 1, derivatives 1 and 2 showed great antibacterial properties, being 1.25−10-fold more active than the reference CP toward all Staphylococci [minimal inhibitory concentration (MIC) 0.05− 0.4 μg/mL]. The acryloyl compound 1 exerted a 5−10 times higher potency than the fluoroquinolone antibiotic against two S. aureus pathogens (NCTC 4163 and ATCC 25923). In addition, majority of the halogenoalkyl derivatives (3−11 and 14−16) were equally or more potent than CP toward the studied Gram-positive cocci. For molecules with a linker of up to five carbons, no clear dependence between the properties of chloro-and bromoalkanoyl analogues was found; all of them were comparably highly active. Achieving MICs of 0.1−0.4 μg/ mL, they were 1.25−2.5 times more effective than unbound CP against most of the Staphylococcal pathogens.
The antimicrobial potential of substituted chlorohydrocarbon moieties toward Gram-positive rods is decreasing as follows: acetyl > pentanoyl > propanoyl > hexanoyl ≫ butanoyl, however without sharp differences in their overall biological strength. The chloroacetyl conjugate (3) was the most powerful; however, the growth inhibitory properties of longer alkanoyls (4, 6, and 7) were similarly strong against all strains (MIC 0.1−0.8 μg/mL). The chlorobutanoyl derivative (5) seemed to be a stronger inhibitor than CP, but only toward two selected pathogens. Within the bromine-containing analogues, the pentanoyl linker (11) was the limiting value of the strong antimicrobial activity. That compound was 1.25− 2.5 more active than CP against all tested strains (MIC 0.1−04 μg/mL). Other bromoalkanoyls (8−10) assigned the same level of biological potency as the reference drug toward four Staphylococci. In contrast to the chlorohexanoyl (7) derivative, more effective toward half of the Gram-positive rods (MIC 0.2−0.4 μg/mL) as compared to CP, the bromohexanoyl analogue (12) possessed moderate activities (MIC 0.8−1.6 μg/mL). Thus, the group of compounds with bromoalkanoyl linkers can be ranked as follows (in decreasing order): pentanoyl > butanoyl > propanoyl > acetyl ≫ hexanoyl > undecanoyl. The 3-trifluoromethylalkanoyl series (14−16) was a little less potent than the above-mentioned monohalogenoalkanoyl counterparts. Whereas both butanoyl (15) and pentanoyl (16) derivatives were equally active (MIC 0.2 μg/ mL) toward all Gram-positive strains, the propanoyl compound (14) assigned half of the activity of its close analogues, being still more potent against two Staphylococci, as compared to CP. The dimeric compounds (17−21) constituted a group with the weakest antimicrobial properties, and the longer the spacer, the greater decrease in antimicrobial properties was observed. The most potent acetyl derivative 17, with MIC equaled 0.2−1.6 μg/mL against four strains, was less active than the reference drug. Considering the susceptibility of standard Gram-negative pathogens to the studied conjugates, two bromo derivatives, butanoyl (10) and pentanoyl (11), at a MIC of 0.025−0.1 μg/mL maintained 60% of CP potency against both studied standard rods. Less powerful chlorobutanoyl (5) and cyclopropanecarbonyl (2) analogues kept 30− 60% of the drug efficiency (MIC 0.025−0.2 μg/mL). Furthermore, other chloro-(1, 3, 6, and 7), bromo-(8 and 9), trifluoromethylalkanoyl (14−16), and dimeric (19) derivatives, applied at a MIC of ≤ 1 μg/mL, exhibited considerable antimicrobial properties, mainly against the more susceptible strain of E. coli.
The activity of all derivatives was next tested against an expanded panel of clinical isolates ( Table 2). The MICs of the most potent amides toward Staphylococcal pathogens were in the range from 0.1 to 0.5 μg/mL. Studied S. aureus strains revealed to be more sensitive as compared to S. epidermidis. The strongest growth inhibitory potency toward S. aureus 180 and 5595 isolates was observed for the chloropentanoyl (6) derivative, which at 0.1 μg/mL acted 5-fold more effective than the reference. Similarly, its chloropropanoyl analogue (4) was 2.5 times more active than CP. Amides bearing cyclopropanecarbonyl (2), chloroacetyl (3), and chlorobutanoyl (5) fragments inhibited bacterial growth at 0.4 μg/mL, stronger than the reference antibiotic. The same level of activity against the S. aureus 5595 isolate was observed for acetyl conjugates 8 and 17. The hexanoyl (7 and 12), propanoyl (9 and 14), and pentanoyl (15) derivatives were as potent as CP against two mentioned Staphylococcal strains (180 and 5595). Moreover, compounds 1, 10, 11, and 19 gained 62.5% of CP inhibitory activity. Concerning other S. aureus isolates (T5595 and T5591), amides 2, 3, 6, 9, and 14 acted effectively at 0.1−0.2 μg/mL, being 1.25−2.5 times more powerful as CP. Compounds 4, 5, 7, 15, and 16 with MICs of 0.4 μg/mL had 62.5% of the reference activity. Among S. epidermidis rods, the KR4243/1 strain was the most susceptible to the presence of the studied amides. Conjugates 2−7, 9, 14, 15, and 17 inhibited its growth at a concentration of 0.1−0.2 μg/mL, which gave 1.25−2.5 of the strength of CP alone. Similarly, it was observed that derivatives 2−6 and 8 exerted at 6.4−12.8 μg/mL a stronger activity than the standard drug toward the more resistant S. epidermidis 4341 strain. Several amides (2, 4, 6, 8, and 17) were as potent as the reference against the S. epidermidis 5253 isolate (MICs 12.8−25.6 μg/ mL). Additionally, conjugates 2−11, 14, and 15 expressed 65% of CP activity against the Staphylococcus pasteuri strain. Gramnegative isolates of clinical origin appeared to be considerably less vulnerable to the presence of synthesized quinolone derivatives. The highest activity was observed for the 3chlorobutanoyl compound (5), followed by bromoalkanoyl substances with four-carbon (10) and five-carbon (11) atoms in their amide fragment. The growth of two E. coli rods was inhibited by derivative 5 at 0.025 μg/mL, which was 2.4−1.2 times lower than the inhibitory concentration observed for CP. Moreover, the mentioned compounds applied at 0.05 and 0.8 μg/mL were 1.2 times more potent than the standard antibiotic against Escherichia cloacae and P. aeruginosa 37 strains, respectively. The same doses of these compounds were effective toward other selected E. coli and P. aeruginosa pathogens, gaining at least 60% of the reference drug activity. Among other tested substances, the cyclopropanecarbonyl conjugate 2 was highly potent against the E. cloacae isolate, at MIC 1.2 times lower than CP alone. Only amides 5, 10, and 11 at 6.4 μg/mL exhibited the growth inhibitory potency against Klebsiella pneumoniae rods, achieving 62.5% of the reference activity. Moreover, compounds 2−4, 6, 8, 9, 12, 14, and 18 acted against selected E. coli and E. cloacae strains at low concentrations of 0.1−0.8 μg/mL.
To sum up, all chloroalkanoyl derivatives were highly active against Gram-positive hospital isolates, independently of the type of bacterial strain. Their growth inhibitory strength slightly and gradually got lower from that of chloropentanoyl (6), cyclopropanecarbonyl (2), acetyl (3), propanoyl (4), butanoyl (5), and hexanoyl derivative (7). Within the group of CF 3 -containing compounds, the longer was the alkanoyl linker, the weaker activity was assigned. Only CF 3 -propanoyl amide (14) was as potent as the most active monohalogen conjugates (3 and 9). A similar tendency was noticed for bromoalkanoyl derivatives: as the hydrocarbon chain got longer, the biological properties have decreased. Bromopropanoyl (9), acetyl (8), but also inexpertly hexanoyl (12) amides were the most effective in this group. Among dimeric derivatives (17−21), the most active was that with the shortest linker (17), however severalfold weaker in comparison with CP-derived monomers. As compared with the susceptibility of standard strains, the group of compounds active toward hospital isolates was restricted. Spontaneous mutation and horizontal gene transfer observed for clinically isolated bacteria mostly reduced their sensitivity to antimicrobial agents. Derivatives highly effective against both collections of Gram-positive bacteria included mainly the cyclic analogue (2), chloroalkanoyls (3−6), and shorter CF 3 -anlanoyl conjugates (14 and 15). The same compounds (5, 10, and 11) were active against Gram-negative isolates of different origin.

Bacterial Growth Curve
Assay. The growth curves of P. aeruginosa ATCC 15442, E. coli ATCC 25922, and S. aureus ATCC 6538 strains, representing changes in the number of bacterial populations over 18 h in culture, were generated (Figures 1, 2, and S1). Conjugates 5, 10, and 11 in the concentration range between 0.8 and 0.0016 μg/mL were selected for these tests, as they exerted both the strongest potency toward Gram-negative isolates and low MIC values against Staphylococci. The obtained results supplemented MICs presented previously in Table 1, as well as illustrated the impact of compounds on the three distinct phases of the bacterial growth cycle, lag, exponential (log), and stationary, compared with controls and CP alone.
As shown in Figure 1, the clearest concentration�activity dependence was found for P. aeruginosa isolates. All tested compounds visibly influenced the phases of bacterial cell growth ( Figure 1A−C). CP conjugates at doses ≥0.2 μg/mL fully inhibited bacterial growth. Their concentration of 0.1 μg/ mL caused an extension of the initial (lag) phase; thus, the cells did not enter the exponential phase. It was also noticed that the studied amides dosed at 0.03 and 0.05 μg/mL still limited bacterial cell metabolism. In the presence of derivative 5, the exponential phase was shifted to 7−12 and 8−15 h, respectively, of the experiment, as compared with the control ( Figure 1A). The reference chemotherapeutic CP, applied at ≥0.1 μg/mL, completely inhibited P. aeruginosa cell growth ( Figure 1D). At doses of 0.03−0.006 μg/mL, it influenced cell metabolism but did not suppress cell divisions. An elongation of the exponential phase to 9 h was noticed then. When used at 0.05 μg/mL, it elongated the lag phase up to 14 h.
Moreover, the studied derivatives influenced E. coli strains in a different way ( Figure 2). Unbound CP inhibited bacterial cell growth at ≥0.006 μg/mL, while derivatives 5 and 10 applied at ≥0.013 μg/mL. The inhibitory concentration of compound 11 was the highest (≥0.025 μg/mL). A sharp increase in the absorbance within 6 h of the experiment, following by an acute decrease after 7 h, was observed in the presence of CP alone or their conjugates 10 and 11, used at the lowest doses of 0.0016 μg/mL. It corresponded with a reduction of the live bacteria population. CP at 0.003 μg/mL disturbed cell metabolism, as the absorbance values of the experimental phase were lower in comparison with the control. The lag phases in the presence of conjugates 5 and 10 applied at 0.006 μg/mL were elongated to 6−9 h, respectively. In contrast to the control, the growth rate of the bacteria co-cultured with 10 was extended between 9 and 14 h of the test, whereas the short exponential phase in the presence of compound 5 passed into the prolonged stationary phase. Compound 11 strongly altered cell metabolism already at a concentration of 0.013 μg/mL, as the bacterial cells did not reach its exponential phase.
The impact of CP conjugates on the growth of S. aureus ATCC 6538 isolate was also remarkable ( Figure S1), which correlated with the observed MIC values. All tested fluoroquinolones fully suppressed the cell growth at ≥0.2 μg/mL. CP, when applied at 0.1 μg/mL, limited the bacterial growth up to 12 h of the experiment, after which the cells doubled the number. The same concentrations of all CP derivatives shifted the doubling time and elongated the exponential phases to 8−9 h, as compared to controls. What is more, for compound doses ranging between 0.05 and 0.006 μg/mL, the maximal growth rate of bacteria reached lower values than for control experiments.

Biofilm Eradication Activity.
Preliminary antibacterial studies revealed that derivatives 5, 10, and 11 are the most potent against free-floating planktonic cells of standard Gramnegative strains of P. aeruginosa ATCC 15442 and E. coli ATCC 25922 and strongly active toward S. aureus isolates; thus, they were further studied for their ability to eradicate the mature biofilm of the same pathogens. They were tested at doses ranging from 1/2 MIC to 8 MIC (according to Table  S1), compared to the untreated biofilm biomass as the positive control. None of the tested compounds, even CP, have achieved minimum biofilm eradication concentration (MBEC), which eradicates ≥50% of the biofilm structure ( Figure 3). A moderate activity was observed for thiourea derivatives 5 and 11 and unbound CP, which at a concentration of 4 MIC eradicated the biofilm formed by the E. coli strain by approx. 20%. Moreover, at doses of 4−2 MIC, conjugate 11 and CP alone eliminated the mature biofilm of S. aureus ATCC 6538 isolates by 19−25%.
To sum up, the longer was the alkanoyl chain, the higher antimycobacterial activity was observed. Short-chain chloroalkanoyls represented the most potent group and were followed by bromoalkanoyl derivatives, then moderately effective dimers, and finally weakly active CF 3 -derived amides. Chlorine-containing compounds with four to six carbon linkers (5−7), as well as the bromopentanoyl derivative (11), were comparably active, more than also similarly potent cyclopropanecarbonyl (2) and bromobutanoyl (10) conjugates. Equally good antituberculostatic activities were found for bromohexanoyl (12), chloropropanoyl (4), undecanoyl (21), and bromopropanoyl (9) compounds. In contrast, the antitubercular action of chloroalkanoyls of the shortest chains (1 and 3) was only moderate.
2.2.5. Inhibition of Bacterial DNA Topoisomerases. The antibacterial properties of fluoroquinolones are the result of the inhibition of type II topoisomerases�first of the subunits gyrA and gyrB of DNA gyrase, which is responsible for replication, recombination, and repair of DNA. The second target is topoisomerase IV, accounted for the decatenation of the linked daughter chromosomes at the terminal stages of DNA replication. Quinolone drugs bind to the complex formed between DNA and the topoisomerase enzyme, forming a three-component complex, which inhibits bacterial cell growth. 23,24 To compare the enzyme inhibitory properties of the most active antimicrobial agents (1−3, 5, 6, 10, and 11) with parental quinolones, their impact on S. aureus DNA gyrase and topoisomerase IV was evaluated (Table 4 and Figures 4 and 5). The amide 5 inhibited DNA supercoiling by gyrase and topo IV, with IC 50 values of the same order of activity as the references (3.2 and 2.7 μg/mL, respectively), which suggests a related mechanism of action. Compounds 1, 2, 6, 10, and 11 showed almost no difference in activities (IC 50 ≤ 10 μg/mL), and the results obtained for them indicate that the synthesized CP-derived amides may act similarly by suppression of bacterial DNA topoisomerases.
Decreasing amounts of compounds were incubated with 500 ng of relaxed pBR322 and run on agarose gel. Lanes 1 and 18: 2.2.6. Molecular Docking Studies. The binding modes of the most active antibacterial series 1−13 to both DNA gyrase and topoisomerase IV was investigated using molecular docking. The docking result analyses indicate that, for all the studied compounds, the binding of the CP scaffold is essential and very similar to that in the known CP−protein complexes. 48,49 There are differences between particular compounds in the binding energy (and cluster sizes) but within the Autodock4 standard error (which is around 2.5 kcal/mol), see Table 5 for compound 5 and CP and Table S2 for all derivatives. Therefore, for the visualization purposes, we selected the chlorobutanoyl derivative (5), presenting the most potent inhibitory properties as compared to CP (see Table 4 showing IC 50 values). Figure 6 shows the binding mode of compound 5 in comparison to that of CP. The interaction details are presented in Figures 7 and 8. As demonstrated in the figures, the chlorobutanoyl group of compound 5 forms several van der Waals interactions. Altogether, considering the general docking scores (Table 5) and interaction details (Figures 3−5), the effect of chlorobutanoyl group is weak (or negligible in comparison to CP) in light of the docking results.
2.2.7. Cytotoxic Activity. To evaluate the chemotherapeutic profile of the synthesized series 1−21, the in vitro antitumor activity has been determined for human prostate (PC3) cancer and normal HaCaT cell lines by using the MTT colorimetric assay (Table 6), including CP, doxorubicin, and cisplatin as reference drugs. While the unconjugated fluoroquinolone showed no activity toward PC3 cells as compared with almost all the synthesized compounds (IC 50 > 100 μM), three derivatives exhibited high (2.02 μM) to moderate (15.7 μM) cytotoxic action. The most potent 2-chloroacetyl derivative 3 and the dimer 21, bearing an undecanoyl linker, exhibited growth inhibitory activity at concentrations 2.02 and 4.8 μM, respectively. Their cytotoxic properties were then 6.5−2.75fold stronger than those of cisplatin. Moreover, their selectivity indexes (SI) equaled 17.6 and 15.2, respectively. The cytotoxic activity of the butanoyl conjugate 15 was comparable to that of this chemotherapeutic (IC 50 = 15.7 μM).
All microbiologically active conjugates, particularly 1−2, 4, 5, 7−11, and 14−16, did not influence the growth of normal HaCaT cells. The results proved that the promising derivative 3 shares both strong antibacterial and cytotoxic properties, with the highest selectivity vs tumor cell lines.

Antiproliferative Activity.
To monitor cell proliferation in response to long-term conjugate treatment, the trypanblue exclusion assay was further performed. The results  Figure 9). Thus, they exerted a cytostatic effect on them, suppressing their growth and proliferation. However, the cytotoxic influence of derivative 3 on tested cancer cells was also observed, as it decreased the PC3 cell viability by 48% ( Figure 10). Other CP-derived conjugates did not affect cell viability. In addition, while higher concentrations ( Table S4, after 72 h of treatment with conjugates 3 and 21, the number of PC3 cells in late apoptosis or necrosis increased to 78.22 and 69.29%, respectively. In addition, the incubation with derivative 21 showed a significantly higher percentage of tumor cells in early apoptosis (28.22%), as compared to the control. The noticeable pro-apoptotic effect of the derivative 15 was also observed, with 37.64% of cells in late apoptosis or being necrotic. As expected, the test performed with HaCaT cells cultured with the studied compounds 3 and 21 revealed a late apoptosis/necrosis at the level of 47.23−34.95% as in untreated controls. The conjugate 15 did not indicate appreciable apoptosis-inducing action in normal cells.

Inhibition of IL-6
Release. The predominant role of interleukin-6 (IL-6) in cancer is its key promotion of tumor growth. Since it was proved that CP and its derivatives can reduce the level of IL-6, 45 both PC3 and HaCaT cells were treated with IC 50 concentrations of the most cytotoxic derivatives of a series (3, 15, and 21). The strongest effect in PC3 cells was observed for the acetyl derivative 3, which inhibited interleukin release 3.5 times, as compared to the control ( Figure 13). That response was greater than the influence of CP alone, which diminished IL-6 secretion in these cells 2.4 times. Moreover, compound 15 decreased interleukin concentration 1.9 times. Similarly, as CP, conjugates 3 and 21 stimulated human keratinocytes to  produce IL-6. No significant pro-or anti-inflammatory properties of derivative 15 was found.
2.2.11. ROS Inducing Activity. Many chemotherapeutics increase the intracellular level of reactive oxygen species (ROS) and as a result disturb redox homeostasis and induce oxidative stress and ROS-mediated apoptosis in cancer cells. Doxorubicin and platinum complexes, such as cisplatin, target tumor mitochondria and elevate cellular ROS (particularly H 2 O 2 ) production, which could lead to apoptosis and autophagy in cancer cells. The second reason of increased ROS generation is the inhibition of the cellular antioxidant system, including superoxide dismutase (SOD), peroxidases, and catalases. 50 Therefore, we examined the ability of apoptosis-inducing agents (3, 15, and 21) to influence H 2 O 2 production in both PC3 and HaCaT cell cultures.
As noticed, the conjugates applied in their IC 50 concentrations affected differently ROS synthesis in both types of cells (Figures 14 and S3). After 2 h of compound treatment, the amount of ROS in PC3 cells increased 7.7−9.5 times and then measured after 12 h (or 72 h, data not shown), which was comparable to the control level. In contrast, normal HaCaT cells were less vulnerable to the presence of CP derivatives, and the intensity of ROS generation in these cells, independently of treatment time, was nearly equivalent to the control probe. One can conclude that the increased level of H 2 O 2 may lead to activation of apoptosis or necrosis in PC3 cells. However, compounds 3 and 21 induced HaCaT cell death probably by a different, ROS-independent mechanism.

CONCLUSIONS
The current study describes the synthesis of novel CP derivatives (1−21), obtained by modification of the C-7 piperazinyl moiety with halogenated acyl chlorides. Conjugates were evaluated for their in vitro growth inhibitory activity against a panel of Gram-positive and Gram-negative bacteria. Many compounds (1−11 and 14−16) exhibited a potent antistaphylococcal action against standard strains, with IC 50 values between 0.05 and 0.4 μM, which were 1.25−10 folds more effective than unconjugated CP. In addition, the MIC values of the most potent amides (2−9, 14, and 15) toward selected clinical Staphylococci were in the range from 0.1 to 0.5 μg/mL, being 1.25−5-fold lower than for the parental drug. Alkanoyl derivatives 5, 10, and 11 acted more effectively toward hospital E. coli and E. cloacae strains, as compared to  CP alone. During 18 h of observation, conjugates 5, 10, and 11 fully inhibited P. aeruginosa and E. coli cell growth at doses of ≥0.2 and ≥0.025 μg/mL, respectively. At lower concentrations (0.1−0.003 μg/mL), the compounds significantly disturbed the bacterial cell metabolism, as they extended the lag and elongated/shifted the exponential phases or prolonged the stationary phase. Similarly as CP, compounds 5 and 11 at doses of 2−4 MIC moderately eradicated the biofilm formed by selected Gram-negative pathogens. The other advantage of the synthesized compounds (2, 5−7, and 10−12) is their antitubercular activity, being at least equipotent to CP and 64− 128 higher than the first-line antimycobacterial agents, such as INH, RMP, SM, and EMB, especially against the multidrugresistant M. tuberculosis Spec. 210. As shown, the antibacterial profile of compounds 1, 2, 5, 6, 10, and 11 appears to correlate with their ability to inhibit both S. aureus DNA gyrase and topoisomerase IV. The binding mode of target compounds 1− 13 to both topoisomerases is comparable to that of the known CP−protein complexes. Moreover, the MTT assay against PC3 cells indicated that derivatives 3 and 21 have exceptional cytotoxic activity (IC 50 2.02−4.8 μM), stronger than cisplatin, with high selectivity over normal HaCaT cells. These amides almost completely blocked the overall proliferation rate in PC3 cells (by 98.03−93.78%), significantly increasing their death rate, without reduction of HaCaT cell viability. The flow cytometry analysis demonstrated that both CP conjugates greatly induced apoptosis in prostate cancer cells, giving 78.22−69.29% cells in late apoptosis or necrosis. Conjugates 3 and 15 affected the PC3 cell growth by a 3.5−1.9-fold reduction of the interleukin level. As estimated, compounds 3, 15, and 21 also elevated intracellular ROS (particularly H 2 O 2 ) generation, which could explain their pro-apoptotic influence on cancer cells. Modifications of the CP structure have increased its antibacterial spectrum and cytotoxic potency against cancer cells, retaining the relative safety against the normal cell line. The performed studies allowed to select Figure 10. Trypan blue assay. The effect of compounds 3, 15, and 21 on viability in PC3 and HaCaT cells. Cells were incubated for 72 h with tested compounds used in their IC 50 concentrations; then, cells were harvested, stained with trypan blue, and analyzed using the cell counter. Data are expressed as the mean ± SD. ***p ≤ 0.001, **p ≤ 0.01, and *p ≤ 0.05, as compared to the control. Figure 11. Effect of compounds 3, 15, and 21 on early and late apoptosis in PC3 and HaCaT cells. Cells were incubated for 72 h with the tested compounds applied in their IC 50 concentrations; then, cells were harvested, stained with annexin V-FITC and PI, and analyzed using flow cytometry. Data are expressed as % of cells in the early stage of apoptosis, and as % of cells in the late stage of apoptosis or necrosis. Data are expressed as the mean ± SD. ***p ≤ 0.0001, **p ≤ 0.001, and *p ≤ 0.01 as compared to the control. derivative 3, which shares both strong antibacterial and antitumor properties, with an in-depth mechanism of its cytotoxic activity.

General Procedure for the Synthesis of CP Conjugates 1−21.
To a magnetically stirred suspension of CP (1.00 g; 3.02 mmol) and triethylamine (0.84 mL; 6.04 mmol) in dry CH 2 Cl 2 (120 mL), a solution of halogenated carboxylic acid chloride (3.02 mmol) in dry CH 2 Cl 2 (5 mL) was dropped over 5 min at 2−5°C. After 30 min, the cooling bath was removed, and the reaction mixture was stirred at room temperature for 3 h. Next, water (50 mL) and 3% HCl (aq) solution were added to get pH equaled 3−4. After separation of the phases, the water layer was extracted with CH 2 Cl 2 (50 mL). The combined organic layers were washed with water (50 mL) and dried over anhydrous Na 2 SO 4 . Then, the solvent was evaporated under reduced pressure, and the product was isolated using column chromatography on silica gel, with the CH 2 Cl 2 /MeOH mixture (0−8% MeOH) as an eluent system. The monoamides were eluted using 1−4% of methanol, but the elution of dimers required a stronger mobile phase (5−10% of methanol in CH 2 Cl 2 ). For the synthesis of   (14). Pale beige solid, 0.32 g (24%). mp 292. 8 (8), and K. pneumoniae (28 and 510). Antibiotic susceptibility testing including the resistance phenotypes of hospital strains were determined using VITEK 2 Compact and VITEK 2 AES.

7-[4-(3,3,3-Trifluoropropionyl)-piperazin-1-yl]-1cyclopropyl-6-fluoro-4-oxo-1,4-dihydro-quinoline-3-carboxylic Acid
The MIC was determined by the twofold microdilution method according to the CLSI reference procedure with some modifications. 51 The bacteria were cultured in brain heart infusion agar (BHI) and incubated at 37°C for 24−48 h. Bacterial inoculum were prepared in a sterile saline solution and diluted in MH II liquid medium to a final concentration of 10 6 colony-forming units per mL (cfu/mL). The reference CP was tested at the range of 0.03−32 μg/mL, whereas the concentrations of conjugates varied from 0.025 to 25.6 μg/mL. The final concentration of DMSO in working solutions was less than 1%. Bacteria were grown overnight in the presence of different concentrations of the tested compounds. After a 18 h period of incubation, the lowest concentration of drugs that inhibited the visible growth of bacteria was considered as the MIC value. Tests were repeated independently three times.
The VITEK 2 Compact (BioMeŕieux) automated system for the antimicrobial susceptibility testing of microorganisms was used in accordance with the manufacturer's directions.

Bacterial Growth Curve
Assay. The growth rate of P. aeruginosa ATCC 15442, E. coli ATCC 25922, and S. aureus ATCC 6538 strains was observed by inoculating the microtiter plates with BHI broth, containing 5 × 10 5 colony-forming units (cfu) per mL of bacteria, loaded with varying concentrations (0.8−0.0016 μg/mL) of compounds 5, 10, 11, and CP as the reference drug. The plates were incubated at 37°C and rotated at 180 rpm. After inoculation, the optical density (OD) at 600 nm was monitored every 5 min interval for 18 h. Assays were repeated three times on different days. The results were expressed as the mean of three experiments.

Biofilm Eradication Assay.
Three bacterial strains E. coli ATCC 25922, P. aeruginosa ATCC 15442, and S. aureus ATCC 6538 were used to assign biofilm eradication properties of compounds 5, 10, 11, and CP as the reference drug. To quantify the biomass of the bacteria biofilm, the crystal violet method was used with some modifications. 52 The bacteria were incubated to the mid-log phase, diluted to 103 cfu/mL in brain−heart infusion medium, and seeded in a 96-well plate. Bacterial biofilms were formed by overnight incubation at 37°C with rotation at 140 rpm. After biofilm formation, all wells were washed with PBS to remove nonadherent bacteria cells. Next, the biofilm was treated with compounds at concentrations ranging from 1/2 MIC to 8 MIC (Table S1). The biofilms with tested substances were incubated for 20 h at 37°C with shaking at 140 rpm. The bacterial biofilm was stained with 0.1% (v/v) CV for 10 min. Afterward, the biofilm was rinsed with PBS and dried. To solubilize the adsorbed crystal violet, wells with stained biofilms were incubated in 30% acetic acid. To estimate the total biofilm biomass, optical density (OD) of the resulting solution was measured at 595 nm. Biofilm assays were repeated at least three times on different days, with two technical replicates assessed each time. A minimum eradication concentration (MBEC) was determined as the lowest concentration of compound, which could damage ≥50% of the formed biofilm structure.
4.3.4. In Vitro Antimycobacterial Activity. The synthesized compounds 1−21 were tested in vitro for their tuberculostatic activity against typical strains [M. tuberculosis H 37 Rv strain (ATCC 25618), M. tuberculosis Spec. 210, and M. tuberculosis Spec. 192] using the MABA method (Microplate Alamar Blue Assay method). 53,54 Investigations were performed by the twofold serial microdilution method (in 96-well microliter plates) using the Middlebrook 7H9 Broth medium (Beckton Dickinson) containing 10% of OADC (Beckton Dickinson). The inoculum was prepared from fresh LJ culture in the Middlebrook 7H9 Broth medium with OADC, adjusted to a no. 1 McFarland tube, and diluted 1:20. The stock solution of a tested agent was prepared in DMSO. Each stock solution of a tested compound was diluted in the Middlebrook 7H9 Broth medium with OADC by 4-fold the final highest concentration to be tested. Compounds were diluted serially in a sterile 96well microtiter plates using 100 μL Middlebrook 7H9 Broth medium with OADC. The concentrations of tested agents ranged from 0.125 to 512 μg/mL. A growth control containing no antibiotic and a sterile control without inoculation were also prepared on each plate. The plates were incubated at 37°C for a week. After the incubation period, 30 μL of Alamar blue solution was added to each well, and the plate was reincubated for 24 h. The growth was indicated by the color change from blue to pink. The lowest concentration of a 4.3.7.6. Statistical Analysis. The statistical calculation was performed using the Statistica 13.1 (StatSoft, Inc, USA) program. The quantitative comparisons were made using Student's t-test. The IC 50 values were estimated by CompuSyn version 1.0. The results of all presented experiments were expressed as the mean ± SD and considered statistically significant at p < 0.05.