Novel pyrazoles as potent growth inhibitors of staphylococci, enterococci and Acinetobacter baumannii bacteria

Abstract

Background:

Methicillin-resistant Staphylococcus aureus, vancomycin-resistant enterococci and Acinetobacter baumannii cause serious antibiotic-resistant infections. Finding new antibiotics to treat these infections is imperative for combating this worldwide menace.

Methods & Results:

In this study, the authors designed and synthesized potent antimicrobial agents using 4-trifluoromethylphenyl-substituted pyrazole derivatives. In addition to their potency against planktonic bacteria, potent compounds effectively eradicated S. aureus and Enterococcus faecalis biofilms. Human cells tolerated these compounds with good selectivity factors. Furthermore, the authors provide evidence for the mode of action of compounds based on time-kill kinetics, flow cytometry analysis of propidium iodide-treated bacteria and oxygen uptake studies.

Conclusion:

This study demonstrated 20 novel compounds with potent antibacterial activity that are tolerated by human cell lines.

Microbial resistance to antibiotics has become one of the ultimate concerns for human well-being. As per the CDC, at least 2.8 million people get infected every year with antibiotic-resistant infections, and more than 35,000 die as a result of these diseases in the US alone [1]. Therefore, discovering new antimicrobial agents to treat infections caused by antibiotic-resistant microbes has become an urgent necessity. Methicillin-resistant Staphylococcus aureus (MRSA), a type of Gram-positive bacterium, is resistant to many commonly used antibiotics. In 2017, approximately 20,000 people in the US died of bloodstream S. aureus infections, and over 119,000 were infected by this organism. Vancomycin-resistant enterococci (VRE) can cause serious infections in healthcare settings. In 2017, vancomycin-resistant enterococci caused an estimated 54,500 infections and 5400 deaths in the US. Acinetobacter baumannii, a Gram-negative bacterium, can cause different infections in humans, including blood infections, urinary tract infections, lung infections and wound infections. Individuals with weak immune systems, diabetes or chronic obstructive pulmonary disease are more prone to infection. A. baumannii and the Gram-positive vancomycin-resistant enterococci and methicillin-resistant S. aureus have been listed by the CDC as urgent and serious threat bacteria [1].

The pyrazole nucleus is one of the most studied heterocycles, as its derivatives display a wide range of bioactivities, including antimicrobial [2,3], anticancer [4], antifungal [5], anti-inflammatory [6] and antidiabetic properties [7]. The pyrazole nucleus is the cornerstone of several well-established drugs, including celecoxib, rimonabant and fomepizole [8]. Similarly, hydrazone derivatives represent an important class of compounds with a broad scope of pharmacological properties [9]. Recently, the trifluoromethyl group (-CF₃) has attracted the attention of researchers because of its ability to work as a bioisostere in which the steric and electronic properties of compounds can change or prevent metabolic degradation. The trifluoromethyl group increases the lipophilicity, membrane permeation and metabolic stability of molecules and can alter receptor binding [10]. In the authors' lab, the authors have designed several antimicrobial agents using the pyrazole scaffold as a cornerstone and have diversified these agents with different bioactive substituents [11–14].

Methods

General considerations

Glass flask materials, solvents and chemical reagents were acquired from Oakwood Chemical (SC, USA) and Fisher Scientific (IL, USA). Growth media, antibiotics and other materials for biological studies were purchased from Fisher Scientific, Sigma-Aldrich (MO, USA) and Hardy Diagnostics (CA, USA). Bacterial strains and human embryonic kidney 293 (HEK-293) cell line was purchased from American Type Culture Collection (VA, USA). The ¹H and ¹³C nuclear magnetic resonance (NMR) spectra were recorded with a Mercury (Varian, CA, USA) at frequencies of -300 and 75 MHz, respectively. DMSO-d₆ solvent with TMS internal standard was used to record nuclear magnetic resonance spectra.

Synthesis of PA

The starting material (PA) was synthesized using the authors' previously reported procedure for similar pyrazole aldehydes [11]. A solution of commercially available 4-(trifluoromethyl)phenylhydrazine (1.76 g, 10 mmol) and 4-acetylbenzoic acid (1.64 g, 10 mmol) in 50 ml dry ethanol was refluxed for 12 h and the solvent evaporated in vacuo. The solid hydrazone was further dried for 30 min under a vacuum to remove traces of solvent. The solid hydrazone was dissolved in 20 ml of DMF, and the round-bottomed flask was sealed with a rubber septum and cooled under ice. POCl₃ (50 mmol) was added dropwise through the septum, and the round-bottomed flask was vented with a needle. The reaction mixture was brought to room temperature for 30 min and further heated to 80°C for 6 h. The reaction mixture was poured onto ice and stirred for 12 h to hydrolyze the iminium salt. The solid precipitate was filtered and washed with water repeatedly followed by drying to get the pure product without further purification (PA 3.49 g, 97%).

General procedure for the synthesis of hydrazone derivatives (1–20)

The 4-trifluoromethylphenyl-substituted pyrazole-derived hydrazone derivatives were synthesized by reacting the aldehyde-containing pyrazole (1 mmol) with substituted hydrazines (1.1 mmol) in ethanol and refluxing for 8 h. Sodium acetate (1 mmol) and acetic acid were added in case hydrochloride salts of the hydrazine derivatives were present. The product of this reaction was filtered and washed with ethanol (∼10 ml) followed by water (∼10 ml) to obtain the pure products.

MIC assay

The MIC values of novel compounds were determined using the standard microdilution method recommended by the Clinical and Laboratory Standards Institute (CLSI) as reported by the authors previously [11]. To determine MIC values, the starting concentration of compounds (50 μg/ml) was diluted twofold across the 96-well honeycomb plate columns in cation-adjusted Mueller Hinton broth (CAMHB), and bacterial culture was added. Plates were incubated for 18–20 h at 35°C before observing the minimum concentrations at which wells displayed no apparent turbidity. The MIC values were determined from duplicate results obtained from three separate experiments using fresh bacterial cultures on different days. The diluent used for compounds (DMSO) was kept at a final concentration of 2.5% in the microplate well, below the cytotoxicity level for bacteria.

HEK-293 cytotoxicity assay

A resazurin cell viability assay was used to evaluate the cytotoxicity of compounds against HEK-293 cells in 96-well honeycomb black plates as described by the authors previously [14,15]. Cells (6000 per well) were plated in Eagle's minimum essential medium containing 10% fetal bovine serum and incubated for 24 h at 37°C in the presence of 5% CO₂. After incubation, different concentrations of compounds were added to cells and incubated for an additional 24 h. After the second incubation, 40 μl resazurin (0.15 mg/ml w/v in media) was pipetted in and gently mixed. After an additional 4 h of incubation, the plate was read using a Cytation™ 5 plate reader (BioTek, VT, USA) with excitation at 560 nm and emission at 590 nm.

Biofilm inhibition & eradication assay

As reported in the authors' previous study, biofilm inhibition and destruction activity were assessed [14]. To assess inhibition efficacy, an overnight culture of bacteria was suspended in sterile phosphate-buffered saline (PBS) to a McFarland turbidity level of 0.5 and diluted 1:100 in CAMHB containing 1% glucose. This bacterial suspension was mixed with various concentrations of compounds in triplicate in a 96-well clear flat-bottomed plate and incubated for 24 h at 35°C. After incubation, the wells were washed three times with 1× PBS to remove planktonic cells and dried in a 60°C oven for 15 min before being stained with crystal violet (0.1% w/v in water). After staining, the wells were washed three times with distilled water to remove the unstained dye and dried in the oven for an additional 15 min. The stained biofilm was solubilized by adding 33% acetic acid solution, and optical density was measured at 620 nm using a Cytation™ 5 plate reader.

For the biofilm eradication assay, bacterial suspension prepared as mentioned earlier was seeded in 96-well plates to grow and develop biofilm for 24 h at 35°C. After incubation, the wells were washed three times with sterile 1× PBS to remove planktonic cells. Different concentrations of compounds in fresh, sterile CAMHB were added to each well and incubated for an additional 24 h at 35°C. After this second incubation, the authors washed, stained with dye, solubilized with 33% acetic acid and measured optical density as described earlier. Wells containing sterile media with no treatment were used as a positive control, and wells with a bacterial suspension containing 2.5% DMSO (compound diluent) were used as a negative control for both biofilm inhibition and eradication assay.

Time-kill assay

The time-kill assay for pyrazole-derived hydrazone against Gram-positive and Gram-negative bacteria was conducted as described in the authors' previous study [12]. Briefly, exponential phase bacteria in CAMHB were exposed to 4× MIC of the compound and incubated at 35°C. Every 2 h, a 20-μl aliquot was serially diluted in sterile 1× PBS and inoculated onto a blood agar plate by the 6 × 6 drop plate method to determine viable colony-forming units/ml to demonstrate the killing capacity of the compounds. The assay for each compound treatment was conducted in triplicate.

Flow cytometry assay for membrane permeability

Flow cytometry analysis was used to assess the integrity of the bacterial cell membranes using the DNA intercalating dye propidium iodide (PI) as previously described [16]. S. aureus American Type Culture Collection 700699 cells were grown to mid-log phase in CAMHB media. The cells were harvested by centrifugation at 4000 rpm for 10 min and washed with 1× PBS twice before being diluted to around 10⁵ colony-forming units/ml in 1× PBS. The bacterial suspensions were incubated with different concentrations of compounds for 30 min at 35°C. The cells were then harvested by centrifugation at 4000 rpm for 10 min, washed twice with 1× PBS to remove excess compounds and then incubated for 30 min with 10 μg/ml PI at 4°C. After incubation, the cells were centrifuged and washed with excess PBS to remove unbound dye. Data were recorded with an excitation wavelength of 488 nm (phycoerythrin–Texas Red A filter) using a BD FACSAria™ cell sorter (BD Biosciences, NJ, USA). Nontreated cells, and vancomycin and 70% ethanol with PI were used as negative and positive controls, respectively.

Oxygen uptake experiments

A. baumannii was grown overnight in CAMHB with shaking at 35°C and then transferred to fresh CAMHB and grown for several more hours until exponential growth was reestablished with a concentration at around 8 × 10⁸ cells/ml. Cells were centrifuged, and cell pellets (2 × 10⁹) were kept on ice until use. Each cell pellet was resuspended in 10 ml of CAMHB in a small vessel and stirred continuously. The O₂ sensor probe was inserted into the suspension, 50 μl of reagent was added (FCCP = 5 μM; compound 11 = 10 μg/ml) and relative dissolved oxygen measurements were obtained every 15 s for 3 min [17].

Results & discussion

Recently, the authors reported the synthesis and antimicrobial studies of 1,3-dibenzoic acid-derived pyrazole derivatives (Figure 1A) [18]. These compounds showed narrow-spectrum activity against A. baumannii. The authors have also reported the synthesis and antimicrobial properties of different substituents on the 3-phenyl ring of the pyrazole nucleus [13,18–20]. In addition, the authors have reported the antibacterial properties of coumarin- and naphthalene-bearing pyrazole compounds [12,15]. Based on the medicinal importance of the CF₃ group, the authors hypothesized that replacing one of the carboxylic groups with CF₃ would result in compounds (Figure 1B) with better antimicrobial properties. To the authors' delight, these compounds were found to be potent and broad-spectrum antimicrobial agents.

Figure 1. . Modification of (A) lead compounds to design (B) 4-trifluoromethylphenyl-substituted pyrazoles as potent antibacterial agents.

By using the authors' established procedure [14], the reaction of 4-trifluoromethylphenyl hydrazine (TH) with 4-acetyl benzoic acid (AB) in ethanol afforded the hydrazone derivative, which on reaction with POCl₃/DMF formed the pyrazole-derived aldehyde (PA), as shown in Figure 2. The pure aldehyde starting material (PA) was synthesized on a multigram scale in an excellent overall yield without workup or column purification. The reaction of the aldehyde starting material with different hydrazine derivatives afforded the hydrazone products. Based on the authors' previous structure–activity relationship studies [14], the authors focused on halogens and small hydrophobic substituents on the phenyl of the hydrazine moiety. The reaction of N,N-disubstituted hydrazines with the aldehyde (PA) formed the products (1–6) efficiently. Similarly, sulfonyl (7–8), monosubstituted (9–14), disubstituted (15–19) and polysubstituted (20) phenyl-derived hydrazones formed smoothly.

Figure 2. . Synthesis of 4-trifluoromethylphenyl-substituted pyrazole-derived hydrazones.

Antimicrobial studies

The authors tested the synthesized novel compounds for their antimicrobial potency against 16 bacterial strains (Table 1). Some of the compounds showed potent activity against the tested Gram-positive strains and strains of the Gram-negative bacterium A. baumannii. These strains included both antibiotic-sensitive and antibiotic-resistant bacteria. The starting material (PA) was inactive against the tested strains, which was consistent with the authors' previous findings with regard to pyrazole-derived aldehydes. None of the compounds showed any activity against the Gram-negative bacteria: Pseudomonas aeruginosa, Klebsiella pneumoniae and Escherichia coli. Previously, the authors reported that N,N-disubstituted hydrazone derivatives were moderate or inactive antibacterial agents [12,18]. In this series of compounds, N,N-disubstituted derivatives were potent growth inhibitors of Gram-positive bacteria. The N-methyl-N-m-tolyl hydrazone derivative (1) inhibited the growth of S. aureus strains, with MIC values of 1.56–3.12 μg/ml. This compound (1) also effectively inhibited the growth of enterococci, Bacillus subtilis and Staphylococcus epidermidis strains, with an MIC value as low as 0.78 μg/ml. The N-ethyl-N-phenyl-derived hydrazone (2) was an effective antimicrobial agent against Gram-positive bacteria. The N-ethyl-N-tolyl (3) and N-butyl-N-phenyl (4) derivatives showed similar potency. The N,N-diphenyl compound (5) showed very potent activity against all tested Gram-positive bacteria, with MIC values of 0.78–1.56 μg/ml. The N,N-dibenzyl derivative (6) was a good antibacterial agent but less potent than the N,N-diphenyl compound (5). Thus, the N,N-diphenyl derivative (5) was the most potent compound in the N,N-disubstituted series (1–6). Nevertheless, this compound (5) did not show any activity against Gram-negative strains. Although a number of sulfonamide antibiotics are known [21], the sulfonamide-derived hydrazones (7 and 8) failed to show any significant activity against the tested strains.

Table 1. . Antimicrobial properties of novel pyrazole-derived hydrazones.

Number	MIC (μg/ml)
	Sa23	Sa99	Sa12	Sa91	Sa92	Ef12	Ef21	Efs99	Bs	Se	Ab6	Ab5	Ab7
1	1.56	3.12	1.56	3.12	3.12	6.25	3.12	3.12	0.78	3.12	NA	NA	NA
2	3.12	3.12	3.12	3.12	3.12	6.25	3.12	3.12	1.56	3.12	NA	NA	NA
3	3.12	3.12	3.12	1.56	1.56	3.12	1.56	1.56	0.78	1.56	NA	NA	NA
4	1.56	3.12	1.56	1.56	1.56	3.12	1.56	3.12	0.78	1.56	NA	NA	NA
5	0.78	1.56	0.78	1.56	1.56	1.56	0.78	1.56	0.78	1.56	NA	NA	NA
6	3.12	3.12	3.12	3.12	3.12	25	1.56	3.12	1.56	3.12	NA	NA	NA
7	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA	NA
8	NA	NA	NA	NA	NA	NA	25	NA	12.5	NA	NA	NA	NA
9	3.12	3.12	3.12	6.25	6.25	6.25	3.12	12.5	1.56	6.25	3.12	12.5	3.12
10	1.56	1.56	1.56	3.12	3.12	3.12	3.12	6.25	1.56	3.12	3.12	12.5	3.12
11	3.12	3.12	3.12	3.12	3.12	6.25	3.12	3.12	1.56	6.25	3.12	3.12	1.56
12	1.56	1.56	1.56	1.56	1.56	3.12	0.78	1.56	0.78	1.56	NA	NA	NA
13	NA	NA	NA	NA	NA	NA	1.56	NA	3.12	NA	NA	NA	NA
14	6.25	NA	1.56	6.25	3.12	NA	0.78	6.25	6.25	1.56	NA	NA	NA
15	6.25	6.25	6.25	6.25	6.25	12.5	3.12	12.5	3.12	3.12	3.12	6.25	3.12
16	3.12	3.12	3.12	3.12	3.12	12.5	3.12	3.12	1.56	6.25	12.5	NA	6.25
17	3.12	3.12	6.25	3.12	3.12	12.5	1.56	3.12	0.78	6.25	NA	NA	NA
18	1.56	1.56	1.56	1.56	1.56	NA	1.56	NA	0.78	3.12	NA	12.5	NA
19	6.25	6.25	6.25	6.25	6.25	12.5	6.25	6.25	3.12	6.25	12.5	NA	12.5
20	6.25	12.5	12.5	6.25	3.12	12.5	3.12	6.25	1.56	3.12	NA	NA	NA
V	0.78	3.12	1.56	3.12	1.56	3.12	>50	NA	0.19	3.12	–	–	–
C	–	–	–	–	–	–	–	–	–	–	3.12	3.12	1.56

Gram-positive: antibiotic-susceptible Staphylococcus aureus ATCC 25923 (Sa23), antibiotic-resistant S. aureus ATCC 700699 (Sa99), S. aureus BAA-2312 (Sa12), S. aureus ATCC 33591 (Sa91), S. aureus ATCC 33592 (Sa92), antibiotic-susceptible Enterococcus faecalis ATCC 29212 (Ef12), vancomycin-resistant Enterococcus faecium ATCC 700221 (Ef21), vancomycin-resistant E. faecalis ATCC 51299 (Efs99), Staphylococcus epidermidis ATCC 700296 (Se), Bacillus subtilis ATCC 6623 (Bs), Acinetobacter baumannii ATCC 19606 (Ab6), A. baumannii ATCC BAA-1605 (Ab5) and A. baumannii ATCC 747 (Ab7).

ATCC: American type culture collection; C: Colistin; NA: Not applicable; V: Vancomycin.

In previous structure–activity relationship studies, the authors found that halogen-substituted phenyl hydrazones were potent antimicrobial agents [11–13]. Therefore, the authors focused on the synthesis and antimicrobial studies of halogen-substituted hydrazone derivatives. The authors found several potent compounds in this series. The 3-fluorophenyl derivative (9) was a good growth inhibitor across the tested strains. The 3-bromo derivative (10) was a very potent antimicrobial agent against the tested bacteria. This compound inhibited the growth of the tested strains, with an MIC value as low as 1.56 μg/ml. The 4-bromophenyl hydrazone derivative (11) was one of the best compounds in the series, with an MIC value as low as 3.12 μg/ml against S. aureus strains. This potent compound (11) inhibited the growth of enterococcal strains, with MIC values of 3.12–6.25 μg/ml. This compound was potent against B. subtilis, with an MIC value of 1.56 μg/ml. Compound 11 was an effective growth inhibitor of A. baumannii strains, with MIC values of 1.56–3.12 μg/ml. With a very strong electron-withdrawing group, trifluoromethyl-substituted phenyl hydrazone (12) was very effective against Gram-positive strains, with an MIC value as low as 0.78 μg/ml, but it failed to show any activity against A. baumannii. The nonhalogenated, strong electron-withdrawing cyano (13) and nitro (14) groups failed to show consistent activity across the tested bacteria. Thus, similar to the authors' previous findings [13,14], halogens in the form of mild electron-donating or electron-withdrawing groups were the key to the potent antibacterial activity of the compounds.

After finding potent monosubstituted compounds, the authors synthesized some disubstituted halogen-derived hydrazones. The 3,4-difluorophenyl hydrazone (15) was an active compound against the tested Gram-positive strains, with an MIC value as low as 3.12 μg/ml. The 2,5-difluoro-substituted compound (16) inhibited the growth of S. aureus, with an MIC value of 3.12 μg/ml. The dichloro-substituted phenyl hydrazone (17) showed activity similar to that of the difluoro-substituted compounds. The 3-chloro-2-fluoro-substituted compound (18) showed very good activity against S. aureus strains, with an MIC value of 1.56 μg/ml. The 3-chloro-4-fluro (19) and tetrafluoro (20) derivatives were moderately active against the tested bacteria. Proteomics studies of bacterial strains treated with sublethal concentrations of potent compounds will help to determine the differential activity of compounds against different strains of bacteria.

Structure–activity relationship

Based on the activity of the compounds, the authors were able to deduce a clear structure–activity relationship. The N,N-disubstitution on the hydrazone nitrogen (1–6) showed potent activity against Gram-positive strains. The N,N-diaryl derivative (5) was the most potent compound in this panel of compounds. The sulfonyl group quenched the activity of the resultant compounds (7 and 8). Moderate electron-withdrawing groups resulted in compounds (9–11) that were active against both Gram-positive and A. baumannii strains. Larger halogen atom (bromo)-containing compounds (10 and 11) were more active than the smaller halogen (fluoro)-containing molecule (9). Compounds with very strong electron-withdrawing groups, such as compound 12, showed potent activity against Gram-positive strains but failed to inhibit the growth of A. baumannii. Other strong electron-withdrawing (-CN and NO2) containing hydrazones (13 and 14) showed moderate activity against Gram-positive strains. Disubstitution on the phenyl ring also resulted in good activity of compounds (15–19). Multi-substitution with fluorine reduced the activity of compound 20.

Dibenzoic acid-containing pyrazole-derived anilines did not show any activity against Gram-positive strains [14]. Nevertheless, dibenzoic acid-derived hydrazones showed significant activity against A. baumannii strains [18]. Modification of dibenzoic acid susbtituents to 3-(trifluoromethylphenyl) resulted in potent antibacterial agents with specific activity against Gram-positive strains [14]. In this study, the authors found that N-(trifluoromethylphenyl)-containing pyrazole-derived hydrazones were potent antibacterial agents with activity against both Gram-positive and A. baumannii strains. In addition to their potent activity against Gram-positive strains, the authors found that hydrazone derivatives were active against A. baumannii. Corresponding aniline derivatives were only active against Gram-positive strains. Switching the trifluoromethyphenyl/benzoic acid (1/3 vs 3/1) substituents did not change the activity of the corresponding compounds (Figure 1).

Cytotoxicity

After antibacterial studies, the authors tested the compounds for cytotoxicity against the HEK-293 cell line. Cytotoxicity data for compounds showing activity against Gram-positive bacteria and A. baumannii strains are shown in Figure 3. As can be seen, monosubstituted compounds were more toxic to HEK-293 cells than disubstituted compounds. Nevertheless, these compounds showed a good selectivity factor, in the range of 10–30. These compounds were also submitted to the National Cancer Institute to test against NCI-60 cancer cell lines (https://dtp.cancer.gov/compsub/). These potent antimicrobial agents did not show any noticeable activity against the 60 cell lines in the cancer panel, including leukemia cells, at a concentration of 10 μM, which suggests that these compounds are safe for rapidly dividing healthy cells in humans. The cytotoxicity of the compounds may not be directly comparable with the MIC values of the potent compounds, as the MIC values were not determined in the presence of fetal bovine serum.

Figure 3. . Cytotoxicity of potent antibacterial agents 9, 10, 11, 15, 16 and 19 against the HEK-293 cell line using resazurin assay.

Values represent an average of IC₅₀ values determined in triplicate for each compound. Error bars represent standard deviation values.

Antibiofilm activity

Bacteria in biofilms can be resistant to both the immune system and antibiotic treatment [22]. Potent compounds were tested for their ability to inhibit the formation of biofilm as well as their ability to destroy the preformed biofilms of Enterococcus faecalis and S. aureus strains (Figure 4). Compounds 11 and 15 efficiently inhibited the growth of E. faecalis biofilm at all three concentrations: 2 × MIC, MIC and 0.5 × MIC. The positive control showed good efficacy at 2 × MIC and MIC but was less effective at 0.5 × MIC. These compounds (4, 11 and 15) were comparatively less effective against S. aureus biofilm inhibition. For preformed biofilms, compound 4 showed moderate and weak eradication ability against E. faecalis and S. aureus, respectively. Compound 11 eradicated >70% of E. faecalis and S. aureus biofilms at 2 × MIC and MIC values but showed weak eradication ability at 0.5 × MIC. The difluoro derivative (15) eradicated approximately 70 and 60%, respectively, of the biofilms of E. faecalis and S. aureus. Activity against biofilms was highly dependent on inhibition of bacterial growth, but the tested compounds were more effective than the positive control (vancomycin).

Figure 4. . Biofilm inhibition and eradication studies for compounds 11 and 15 against Enterococcus faecalis and Staphylococcus aureus biofilms.

Each value in the graph represents the mean of the treatments performed in triplicate. Error bars represent standard deviation values.

Time-kill kinetics

A time-kill kinetics assay is a study of the bactericidal activity of an antimicrobial agent over time. The authors studied the time-kill kinetics of potent compounds (11 and 15) against an S. aureus strain. Treatment with compound 11 eliminated >99.9% of bacteria by 8 h, indicating that this compound was bactericidal for the S. aureus American Type Culture Collection 700699 strain. Compound 15 was bacteriostatic for at least 24 h of treatment. As can be seen in Figure 5, the positive controls (vancomycin and gentamicin) were bactericidal for S. aureus, as reported previously [23,24]. Compounds (11 and 15) active against A. baumannii were bacteriostatic in their antimicrobial properties. The positive control, colistin, was bactericidal, as reported previously [25].

Figure 5. . Time-kill kinetics studies.

Time-kill kinetics of compounds 11 and 15 against (A) Staphylococcus aureus and (B) Acinetobacter baumannii. Each data point represents the mean value of colony counts performed in triplicate. Error bars represent standard deviation values.

Membrane permeability assay

In previous studies, based on PI/SYTO9 experiments, the authors found that hydrazone derivatives are membrane-disrupting agents [15]. In this study, the authors used flow cytometry analysis to determine the membrane-disrupting ability of the potent compounds. PI is a red fluorescent dye that is nonpermeant to healthy bacterial membranes. If a bacterium is unable to exclude it, PI will enter the cell and bind to DNA by intercalation. The authors measured PI permeation caused by the potent compound 11 using flow cytometry analysis as reported previously [16]. As can be seen in Figure 6, only approximately 8.5% of cells were stained in the absence of a compound (Figure 6A & B). Treated bacterial cells showed higher PI intensity, with a good dose–response observation. Treatment with 0.5 × MIC and MIC concentrations showed 24.9 and 88.1% PI-permeant cells, respectively (Figure 6C & D). Increasing the concentration to 2 × MIC and 4 × MIC increased those numbers to 96.1 and >99%, respectively (Figure 6E & F). Vancomycin, which interferes with cell wall synthesis, caused much less cell membrane disruption at 2 × MIC and 4 × MIC, suggesting that loss of membrane integrity upon exposure to compound 11 may be a direct effect (Figure 6G & H). Ethanol, a substance known to damage lipid membranes, showed results comparable to 4 × MIC of compound 11 (Figure 6I). The compounds may have directly damaged membrane structure or interfered with membrane function, resulting in failure to exclude PI. It will be important to see the effect of these potent antibacterials on red blood cells in future in vivo studies to rule out a hemolytic effect. It will also be interesting to see the effect of the potent compounds on mitochondrial function.

Figure 6. . Membrane permeability assay of a potent compound (11) against Staphylococcus aureus American Type Culture Collection 700699 using flow cytometry assay.

Oxygen uptake assay

Pyrazole–hydrazone derivatives resemble known uncoupling molecules that disrupt the bacterial membrane proton gradient [26,27]. As uncoupling agents are known to increase rates of oxygen uptake, the effect of compound 11 was tested against A. baumannii, which is a strict aerobe and most likely to show an effect. Figure 7 shows representative data in which the effect of compound 11 matches that of the known uncoupler FCCP, with a greater initial oxygen uptake compared with the negative control, to which only DMSO (compound solvent) was added. These results are in line with the membrane dysfunction mechanism suggested by the PI flow cytometry results.

Figure 7. . Oxygen uptake inhibition studies of compound 11 against Acinetobacter baumannii.

Conclusion

We have reported the design and synthesis of 20 novel compounds. These compounds were designed based on our previous reports. Based on our hypothesis, most of these compounds were potent against the tested Gram-positive bacterial strains, and some showed broader activity, with MIC values at sub μg/ml concentrations. N,N-disubstituted compounds showed potent activity against Gram-positive bacterial strains but failed to inhibit the growth of A. baumannii. Several halogen-substituted hydrazones showed potent activity against the tested strains, including A. baumannii. Bactericidal activity of the potent compounds as well as flow cytometry and oxygen uptake assays indicated the cell membrane-disrupting ability of these novel molecules. Future studies will include further analysis of the mode of action and molecular targets of the potent compounds to find the effect of different substituents in their antibacterial activity. Potent compounds will also be tested for their in vivo safety and efficacy.

Summary points.

• Novel pyrazole compounds containing trifluoromethylphenyl have been synthesized.

• Novel compounds are potent growth inhibitors of Gram-positive bacterial strains, including methicillin-resistant Staphylococcus aureus and vancomycin-resistant enterococci.

• Some of the compounds are active against Acinetobacter baumannii strains.

• Hydrazone moiety is the key to the activity against A. baumannii.

• Potent antibacterials are tolerated by human cell lines.

• Potent compounds are active against bacterial biofilms.

• Time-kill assays demonstrate the bactericidal properties of the compounds.

• Membrane permeability effects are shown by flow cytometry assay.