Study of Biological Activities and ADMET-Related Properties of Novel Chlorinated N-arylcinnamamides

A series of eighteen 4-chlorocinnamanilides and eighteen 3,4-dichlorocinnamanilides were designed, prepared and characterized. All compounds were evaluated for their activity against gram-positive bacteria and against two mycobacterial strains. Viability on both cancer and primary mammalian cell lines was also assessed. The lipophilicity of the compounds was experimentally determined and correlated together with other physicochemical properties of the prepared derivatives with biological activity. 3,4-Dichlorocinnamanilides showed a broader spectrum of action and higher antibacterial efficacy than 4-chlorocinnamanilides; however, all compounds were more effective or comparable to clinically used drugs (ampicillin, isoniazid, rifampicin). Of the thirty-six compounds, six derivatives showed submicromolar activity against Staphylococcus aureus and clinical isolates of methicillin-resistant S. aureus (MRSA). (2E)-N-[3,5-bis(trifluoromethyl)phenyl]- 3-(4-chlorophenyl)prop-2-enamide was the most potent in series 1. (2E)-N-[3,5-bis(Trifluoromethyl)phenyl]-3-(3,4-dichlorophenyl)prop-2-enamide, (2E)-3-(3,4-dichlorophenyl)-N-[3-(trifluoromethyl)phenyl]prop-2-enamide, (2E)-3-(3,4-dichloro- phenyl)-N-[4-(trifluoromethyl)phenyl]prop-2-enamide and (2E)-3-(3,4-dichlorophenyl)- N-[4-(trifluoromethoxy)phenyl]prop-2-enamide were the most active in series 2 and in addition to activity against S. aureus and MRSA were highly active against Enterococcus faecalis and vancomycin-resistant E. faecalis isolates and against fast-growing Mycobacterium smegmatis and against slow-growing M. marinum, M. tuberculosis non-hazardous test models. In addition, the last three compounds of the above-mentioned showed insignificant cytotoxicity to primary porcine monocyte-derived macrophages.


Introduction
One of the basic strategies in finding new bioactive compounds is inspiration from natural substances [1][2][3]. This includes a variety of phenolic acid derivatives and analogs [4].
These acids, including cinnamic acid, are vital components of plants that are often found in fruits, vegetables, cereals, and legumes. They act in plants in a wide range of processes, such as pigmentation, growth, reproduction and resistance to pathogens or predators. Cinnamic acid is a naturally occurring aromatic acid with a long history of use as an additive in plant fragrances and spices. Its derivatives are important promising substances with great potential in the search for new pharmacologically active substances [5,6]. Cinnamic acid and its derivatives have attracted the attention of scientists in recent decades due to low toxicity and a wide range of biological activities such as antibacterial, antiviral, anti-inflammatory, cytotoxic, antidiabetic, hepatoprotective, antioxidant, neuroprotective, anxiolytic, and antituberculotic [7][8][9][10][11][12][13].
Various infections are an increasing worldwide threat. The growing drug resistance of bacterial strains is also a major danger. In addition, the development of cross-resistant and multidrug-resistant strains is a significant problem [14][15][16]. The most common resistant bacterial strains include methicillin-resistant Staphylococcus aureus, vancomycin-resistant S. aureus, vancomycin-resistant enterococci, penicillin-, and macrolides-resistant Streptococcus pneumoniae, cotrimoxazol-resistant Escherichia coli, the third generation of cephalosporinresistant E. coli and Klebsiella pneumoniae, and carbapenem-resistant E. coli, K. pneumoniae, and Pseudomonas aeruginosa [16]. Tuberculosis caused by Mycobacterium tuberculosis is still one of the most lethal communicable diseases in the world. The spread of multidrug-, extensively drug-and totally drug-resistant tubercular strains is a great problem worldwide [17]. Therefore, it is important to approach these real threats responsibly and find new molecules with the potential to fight resistant microorganisms.
Recent studies investigating the antimicrobial activities of an extensive series of variously substituted cinnamic anilides have revealed compounds with promising anti-infective potential [10,12]. As the introduction of halogen into the molecule structure is known to increase antibacterial activity, e.g., [18][19][20][21][22] two series of anilides with presumed antibacterial activity from 4-chlorocinnamic and 3,4-dichlorocinnamic acids were designed. All newly synthesized compounds were evaluated for their activity against gram-positive bacteria and mycobacteria, and their cytotoxic profiles on cancer cell lines and the primary cells were studied in detail to identify a suitable candidate with optimal physicochemical and biological properties for further investigation, because in addition to the above-mentioned antimicrobial derivatives, various cinnamic acid derivatives are described as promising anticancer agents [7,8], e.g., derivatives of sinapic acid [23], caffeic acid [24], cinnamic acid-quinolone hybrids [25], oleanolic acid-cinnamic acid ester derivatives, glycyrrhetinic acid-cinnamic acid ester derivatives [26], and other variously substituted cinnamic acids [27].

Synthesis and Physicochemical Properties
All the studied 4-chlorocinnamanilides (series 1) and 3,4-dichlorocinnamanilides (series 2) were prepared according to Scheme 1 as described previously [10,12]. The condensation of (di)chlorocinnamic acid with appropriate substituted anilines using phosphorus trichloride in dry chlorobenzene under microwave conditions yielded a series of target mono-or dechlorinated N-aryl-(di)chlorocinnamamides 1a-1r and 2a-2r, see Table 1.  Table 1. Structures of ring-substituted (2E)-3-(4-chlorophenyl)-N-arylprop-2-enanilides 1a-1r, (2E)-3-(3,4-dichlorophenyl)-N-arylprop-2-enanilides 2a-2r, calculated lipophilicities (log P/Clog P), and experimentally determined log k, log D 6.5 and log D 7.4 values of investigated compounds. It can be stated that the pharmacokinetic (absorption, distribution, metabolism, elimination; ADME) and pharmacodynamic properties of bioactive agents are most significantly affected by lipophilicity, which is influenced by the chemical constitution and structure of individual agents. In addition to the mentioned ADME profile, it is responsible for interactions with biological targets. Lipophilicity is recognized as one of the most important parameters influencing ADME and bioactivity [28,29] and is also part, e.g., of Lipinski's Rule of Five (Ro5) [30]. This thermodynamic parameter describing the molecule's affinity for an aqueous or relatively lipophilic environment can be characterized by the logarithm of the partition coefficient (log P). In this study, lipophilicity was measured by reversedphase high-performance liquid chromatography (RP-HPLC) under isocratic conditions with methanol as the organic modifier in the aqueous mobile phase and the logarithm of the capacity factor k was calculated, which is used as the lipophilicity index converted to the log P scale [28,31]. In addition to the log k values, the logarithm of the distribution coefficient D pH at pH 6.5 and 7.4 was determined to clarify the behavior of the compounds under physiological conditions and possible ionization [28,29]. Instead of water, a buffer of suitable pH was used as part of the mobile phase. In addition to the above-mentioned experimental methods, commercially available ChemBioDraw Ultra 13.0 and ACD/Percepta ver. 2012 programs were used to calculate the lipophilicity values of all prepared compounds. All the results are shown in Table 1.
The log P values predicted by ChemDraw were identical for all three positional isomers. The Clog P values also calculated by ChemDraw were identical for at least two positional isomers, so this software is not applicable to these studied derivatives, unlike ADC/Percepta, which provided unique log P values for each derivative/isomer. The similarities/differences between the predicted and calculated values are shown in the graphs in Figure 1. It is logical and evident that the lipophilicity of the compounds differs within both series; in this case, by about 0.2. It can be seen from all the graphs that the matches between the predicted log P and the experimentally determined log k and log D, respectively, are very good with a correlation coefficient r in the range of 0.90-0.99 (n = 36). Most substituents contain fluorine (either as a single atom or CF 3 or OCF 3 moieties). It is known that fluorine substituents easily form various inter-and intramolecular interactions, which can significantly affect the resulting behavior of the compounds in an aqueous environment [32][33][34][35]. If all fluorine-free derivatives were eliminated from the graphs, and the effects of the fluorine-based substituents in both series were evaluated separately, a slightly stronger agreement would be found in series 1 (average r = 0.91, n = 11) than in series 2 (average r = 0.89, n = 11) among fluoro-based substituents for all dependences of experimental log k and log D values on predicted lipophilicity log P. In the graphs of Figure 1A-C, significantly deflected points below the trend line can be observed for compounds 1h and 2h (R = 2-CF 3 ). If these compounds are removed from the dependences, an average correlation coefficient r > 0.97 (n = 10) is obtained. After the elimination of the derivatives with these ortho-substituted CF 3 moieties, a slight deviation can be observed for compound 2k (R = 4-OCF 3 ), after the removal of which r increases from 0.97 to 0.99 (n = 9). Within series 1, the elimination of 1k will increase r in the order of ten-thousandths. Thus, based on all these facts, it can be stated that within the monochlorinated series 1, mutual interactions caused by the presence of fluoro-based substituents are much more significant.

In Vitro Antimicrobial Activity
The investigated compounds were screened for in vitro antibacterial activity against the reference and quality control strains Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 29212 and representatives of multidrug-resistant bacteria and clinical isolates of methicillin-resistant S. aureus (MRSA) 63718, SA 630, SA 3202 [36] that were obtained from the National Institute of Public Health (Prague, Czech Republic), and three isolates from American crows of vanA-carrying vancomycin-resistant E. faecalis (VRE) 342B, 368, and 725B [37]. In addition, all the compounds were evaluated in vitro against Mycobacterium smegmatis ATCC 700084 and M. marinum CAMP 5644 as a safe alternative to M. tuberculosis. The genus Mycobacterium is a closely related group of fast and slow-growing species. M. tuberculosis causes one of the most serious human infections, tuberculosis [17]. Alternative model pathogens for M. tuberculosis can be used in laboratory studies to reduce risks and facilitate laboratory handling. M. smegmatis is an ideal representative of a fast-growing non-pathogenic microorganism particularly useful in the study of basic cellular processes of particular importance for pathogenic mycobacteria [38]. M. marinum is very closely related to M. tuberculosis and is the cause of TB-like infections in poikilothermic organisms, especially frogs and fish. M. marinum is a good model for study mainly due to the lower risk for laboratory workers, genetic relatedness and similar pathology as for human TB [39]. Activities are expressed as the minimum inhibitory concentrations (MICs). Surprisingly, the screening showed antimicrobial activity for only a few compounds, so only derivatives with antimicrobial activity are listed in Table 2. All other non-listed compounds had MICs of 256 µg/mL and greater.  Of the prepared 36 compounds, only three compounds from series 1 and 4 from series 2 were antibacterial-effective. The compounds of series 1 were completely inactive against E. faecalis and VRE and against M. marinum. On the other hand, when the compounds showed activity, the efficacy was comparable to or better than that of clinically used drugs. Compounds 1i (R 2 = 3-CF 3 ), 1p (R 2 = 3,5-Cl 3 ) and 1q (R 2 = 3,5-CF 3 ) showed activity within the monochlorinated series 1. In dichlorinated series 2, compounds 2i (R 2 = 3-CF 3 ), 2j (R 2 = 4-CF 3 ), 2k (R 2 = 4-OCF 3 ) and 2q (R 2 = 3,5-CF 3 ) were active. The isolated activity of the unsubstituted (R 2 = H) derivative 2a against M. smegmatis was unexpected based on previous experience with analogues [40][41][42]. As MICs against MRSA isolates were, in fact, comparable with the MIC values observed against methicillin-susceptible S. aureus ATCC 29213, it could be assumed that the presence of mecA gene [43] did not affect the activity of these compounds. Thus, it can be speculated concerning the specific activity against Staphylococcus sp. Similarly, the close activity of the compounds against both VRE and E. faecalis indicates a mechanism of action unrelated to vancomycin resistance [37].

In Vitro Cell Viability
The effect of the compounds on viability of eukaryotic cells was evaluated on human leukemia cell line THP-1, human synovial cell line SW982, and isolated primary porcine monocytes-delivered macrophages (MDM). The general cytotoxic effect was studied on THP-1 cells in the serum-free medium (only THP-1 as the only cells used in this study are able to survive in serum-free medium). Most of the compounds had no cytotoxic effect on THP-1 cells. However, 1i, 1p, 1q, 2a, 2i, 2j, 2k, and 2q decreased cell viability with IC 50 concentrations ranging from 0.5-7.5 µM (Table 2). On the other hand, the results showed that no cytotoxic effects were observed in the culture medium containing 10% FBS on any of the cell lines used to test the cytotoxicity of the compounds except derivative 2q, which had a mild cytotoxic effect on MDM (IC 50 5-10 µM). This could be caused by the direct interaction of such compounds with serum proteins or by the effect of mitogens, cytokines, and hormones present in the serum. Obviously, the full medium is closer to physiological conditions in living organisms, so it can be assumed that these compounds could be non-toxic there.
Subsequently, compounds 2k and 2q were selected for analysis by flow cytometry and cytotoxic effects were detected after treatment of cells by 2k and 2q in serum-free medium at two selected concentrations, 1.0 µM and 0.5 µM, respectively. Apoptosis was not observed but necrosis was significantly induced after 2q treatment. Moreover, 2k increased the number of dead cells (20%) (Figure 2). CCK-8 assay, which is based on cell viability detection, showed a 50% reduction. Thus, it can be assumed that 2k attenuates cell metabolism.

Structure-Activity Relationships
As mentioned above, Ro5 is one of the accepted recommendations concerning the physicochemical parameters of biologically active agents. Ro5 contains the limits of specific molecular descriptors (Table 3) determined on the basis of experimentally and statistically obtained results such that a compound meeting this recommendation is druglike. However, a good druglike score does not make a molecule a drug and vice versa [44]. It is clear that bio-friendly properties are important in the context of specific interactions; therefore, Table 3 shows the profile of mainly Ro5 parameters characterizing the set of active compounds. In addition, based on previous experience, when we were convinced of the importance of electronic σ properties of substituents, i.e., the ability to affect the electron density of a compound, which results in the strength of interactions with biological targets [10][11][12]34,45,46], Table 3 shows the predicted electronic σ parameters of the whole substituted anilide ring characterizing the ability to withdraw or donate electrons to molecule system. For the same reason, i.e., the influence of the biological properties by the surface tension (ST) of the compounds [36,47], the calculated values of this parameter are given in Table 3. Based on the data presented in Table 3, it can be stated that in general, the investigated compounds meet the Ro5 requirements. However, it should be mentioned that compound 2q has the highest molecular weight within the effective agents (but in the limit) and all compounds except derivatives 1i and 2a possess slightly higher lipophilicity than recommended by Ro5. Thus, it can be stated that individual anilide fragments are characterized by electron-withdrawing properties and relatively high lipophilicity.
As approximately one-fifth of the prepared compounds demonstrated antimicrobial activity, a comprehensive structure-activity relationships (SAR) study cannot be performed. Based on the comparison with compounds unsubstituted on the cinnamic acid skeleton, it can be stated that the introduction of a chlorine atom in position C (4) significantly increased the antimicrobial activity. The introduction of the second chlorine at position C (3) resulted in a further significant extension and increase of activity against E. faecalis and VRE, which was not observed in the previous study. In previous studies, the activity of non-chlorinated N-cinnamanilides was found only for 3-CF 3 and 3,5-CF 3 substitutions against S. aureus, MRSA and M. tuberculosis, and only antistaphylococcal activity was determined in the case of 3-F-4-CF 3 substituted derivative [10,12].
The activity of the 3-CF 3 and 3,5-CF 3 substituted derivatives (compounds 1i, 2i, 1q, 2q) could be expected [10]. The activity of 3,5-Cl substituted compound in series 1 and the activity of 4-CF 3 and 4-OCF 3 substituted derivatives in series 2 are gratifying. On the other hand, the total loss of any effect of derivatives 1r, 2r (R 2 = 2-Br-4-OCF 3 ) is surprising. In most cases, these are mono-/di-substitutions at the meta positions of the anilide ring by lipophilic electron-withdrawing substituents capable of forming hydrogen bonds.
However, graphs were created to better understand the relationships. The dependence of the activity against S. aureus, expressed as log (1/ Table S1 in Supplementary Materials.) Increasing activity can be observed with increasing log k value. For the most effective compounds (1p, 1q and 2q), the effect of lipophilicity becomes less pronounced and the curve becomes a certain plateau. The influence of electronic effects on activity against S. aureus is shown in Figure 3b. There is a noticeable trend of increasing activity with increasing electron-withdrawing properties of the anilide part of the molecule. The deviated position in Figure 3b for 2k (R 2 = 4-OCF 3 ) is probably caused by the specific characteristics of this moiety, e.g., [48][49][50][51], are therein similar to the unpredictable behavior of methoxy-substituted compounds observed in other studies, e.g., [45,[52][53][54][55]. The last Figure 3c shows the dependence of the surface tension on the activity. The antistaphylococcal activity within compounds of series 2 decreases with increasing surface tension, i.e., with decreasing ability of the compounds to act as surfactants. A similar course of trends between structure and activities of active compounds of both series can be traced for other antimicrobial activities against E. faecalis, MRSA, VRE, M. smegmatis and M. marinum, and are therefore are not listed.
The relationships between the cell viability, expressed as log(1/IC 50 [M]), on THP-1 cells in the serum-free medium on the physicochemical properties are illustrated in the graphs in Figure 4. The toxic effect increases with increasing lipophilicity, electronwithdrawing properties (within series 2) and increasing surface activity (ability to reduce the surface tension of series 2 compounds). This observation supports the above-presented theory that the observed cytotoxic effect is mainly caused by cell necrosis, see Figure 2.

General Methods
All reagents were purchased from Merck (Sigma-Aldrich, St. Louis, MO, USA) and Alfa (Alfa-Aesar, Ward Hill, MA, USA). Reactions were performed using an Anton-Paar Monowave 50 microwave reactor (Graz, Austria). The melting points were determined on a Kofler hot-plate apparatus HMK (Franz Kustner Nacht KG, Dresden, Germany) and are uncorrected. Infrared (IR) spectra were recorded on Nicolet iS5 IR spectrometer (Thermo Scientific, West Palm Beach, FL, USA). The spectra were obtained by the accumulation of 256 scans with 2 cm −1 resolution in the region of 4000-450 cm −1 . All 1 H-and 13 C-NMR spectra were recorded on a JEOL JNM-ECA 600II device (600 MHz for 1 H and 150 MHz for 13 C, JEOL, Tokyo, Japan) in dimethyl sulfoxide-d 6 (DMSO-d 6 ). 1 H and 13 C chemical shifts (δ) are reported in ppm. High-resolution mass spectra were measured using a high-performance liquid chromatograph Dionex UltiMate ® 3000 (Thermo Scientific, West Palm Beach, FL, USA) coupled with an LTQ Orbitrap XL TM Hybrid Ion Trap-Orbitrap Fourier Transform Mass Spectrometer (Thermo Scientific) equipped with a HESI II (heated electrospray ionization) source in the positive and negative mode.

In Vitro Antibacterial Evaluation
The minimum inhibitory concentrations (MICs) were evaluated by the microtitration broth method according to the CLSI [60] with some modifications. The compounds were dissolved in DMSO (Sigma, St. Louis, MO, USA) to get concentration 10 µg/mL and diluted in a microtitration plate in an appropriate medium, i.e., Cation Adjusted Mueller-Hinton (CaMH, Oxoid, Basingstoke, UK) for staphylococci, and brain heart infusion (BHI, Oxoid) for enterococci to reach the final concentration of 256-0.125 µg/mL. The plate was inoculated by tested microorganisms. The final concentration of bacterial cells was 10 5 for bacteria. Ampicillin (Sigma) was used as reference drugs. A drug-free control and a sterility control were included. The plates were incubated for 24 h at 37 • C for staphylococci and enterococci. After static incubation in the darkness in an aerobic atmosphere, the MIC was visually evaluated as the lowest concentration of the tested compound, which completely inhibited the growth of the microorganism. The experiments were repeated three times. The results are summarized in Table 2.

In Vitro Antimycobacterial Evaluation
The evaluation of in vitro antimycobacterial activity of the compounds was performed against Mycobacterium marinum CAMP 5644 and M. smegmatis ATCC 700084. The broth dilution micro-method in Middlebrook 7H9 medium (Difco, Lawrence, KS, USA) supplemented with ADC Enrichment (Difco) was used to determine the minimum inhibitory concentration (MIC) as previously described [61]. The compounds were dissolved in DMSO (Sigma), and the final concentration of DMSO did not exceed 2.5% of the total solution composition. The final concentrations of the evaluated compounds ranging from 256 µg/mL to 0.125 µg/mL were obtained by twofold serial dilution of the stock solution in a microtiter plate with sterile medium. Isoniazid and rifampicin (Sigma) were used as reference antibacterial drugs. Bacterial inocula were prepared by transferring colonies from culture to sterile water. The cell density was adjusted to 0.5 McFarland units using a densitometer (Densi-La-Meter, LIAP, Riga, Latvia). The final inoculum was made by 1:1000 dilution of the suspension with sterile water. Drug-free controls, sterility controls and controls consisting of medium and DMSO alone were included. The determination of results was performed visually after three days of static incubation in the darkness at 37 • C in an aerobic atmosphere for M. smegmatis and after 21 days of static incubation in the darkness at 28 • C in an aerobic atmosphere for M. marinum. The minimum inhibitory concentrations (MICs) were defined as the lowest concentration of the compound at which no visible bacterial growth was observed. The MIC value is routinely and widely used in bacterial assays and is a standard detection limit according to the CLSI [60]. The results are summarized in Table 2.

In Vitro Cell Viability Analysis
Human monocytic leukemia cell line THP-1 and human synovial cell line SW982 were purchased from the European Collection of Cell Cultures (ECACC, Salisbury, UK) and were used for determination of the influence of test compounds on cell viability, as described previously [52]. These cell lines represent typical suspension and adherent cells, respectively. They were cultivated with test substances in the presence or absence of fetal bovine serum (FBS) in the cultivation medium, and the relative cell viability (the ratio between cells treated with compounds and cells treated only with DMSO) was determined by CCK-8 kit (Sigma) after 24 h.
Porcine monocyte-derived macrophages (MDMs) were prepared, as described previously [62]. Briefly, blood was collected from three-month-old pigs and peripheral blood mononuclear cells (PBMC) were isolated by gradient centrifugation (830× g, 40 min) using Histopaque-1077 (Sigma). CD14+ cells were isolated by positive magnetic bead selection with LS separation columns (Miltenyi Biotec, Bergisch Gladbach, Germany). Isolated monocytes were cultured in 96-well plates at a concentration of 1 × 10 5 cells per well in 0.2 mL of DMEM supplemented with 10% FBS and 1% antibiotics (Antibiotic Antimycotic Solution 100 × 10,000 units penicillin, 10 mg streptomycin, and 25 µg amphotericin B per mL; Sigma) and incubated for five days at 37 • C in 5% CO 2 to differentiate into macrophages. Then, MDMs were cultured for 24 h with tested substances and the relative cell viability was determined by CCK-8 kit (Sigma) following manufacturer instructions.

Flow Cytometry Analysis
Annexin V eFluor450 (Invitrogene, Waltham, MA, USA) with LIVE/DEAD Fixable Far Red Dead Cell Stain Kit (Invitrogene) were used for the analysis of cell death. THP-1 cells resuspended in serum-free medium and treated by two selected drugs, such as 2q and 2k, by IC 50

Conclusions
A series of thirty-six 4-chlorocinnamanilides and 3,4-dichlorocinnamanilides were designed, prepared and tested against S. aureus ATCC 29213 and E. faecalis ATCC 29212 as reference strains and against MRSA and VRE isolates. Furthermore, the derivatives were tested against M. smegmatis ATCC 700084 and M. marinum CAMP 5644 as nonhazardous models of M. tuberculosis. Cytotoxicity was assessed against two human cancer cell lines (THP-1 and SW982) and against primary porcine monocyte-derived macrophages.  1q, 1p, 2j and 2q showed submicromolar activity against S. aureus and MRSA isolates. In addition, agents from series 2 were effective against E. faecalis and VRE isolates and against both mycobacterial strains. By comparison with on cinnamon scaffold non-chlorinated derivatives, it can be stated that chlorination resulted in a significant extension and increase of antimicrobial activity. The lipophilicity (log k and log D) of the compounds was determined experimentally and other physicochemical properties of the prepared compounds were predicted. Based on a rational choice of parameters, general trends between structure and biological activity were outlined. It was found that the higher lipophilicity and surface properties of compounds and electron-withdrawing properties of anilide substituents have a positive effect on antibacterial activity and unfortunately cytotoxicity. However, when eukaryotic cells were cultured in fetal bovine serum, which mimics real in vivo conditions, most derivatives demonstrated cytotoxicity, expressed as IC 50 values, above 10 µM; only compound 2q showed cytotoxicity in the range of IC 50 5-10 µM. These compounds induced cell death (necrosis) of THP-1 cells in a serum-free medium. Based on all of the above-mentioned facts, the most effective agent 2q cannot be considered as suitable for further investigation, in contrast to agents 2j, 2k and 2i, which have wide antibacterial potential. Compounds 1p and 1q can be considered as specifically antistaphylococcal compounds.