Antimicrobial and Antioxidant Activity of Some Nitrogen-Containing Heterocycles and Their Acyclic Analogues

Abstract

We investigated antimicrobial and antioxidant activity of nitrogen-containing heterocycles and their acyclic analogues, some of which can be considered as promising in terms of biological activity. Based on structure, 26 tested compounds were divided into 4 groups. In the test with 2,2-diphenyl-1-picrylhydrazyl (DPPH), the compounds of the group 2 had the highest radical-binding activity (RBA) (53–78%), while those of group 3 had the lowest values (1.5–5.2%). In oxygen radical absorbance capacity assay, all compounds from groups 1, 2 and 3 showed high RBA: 44–94% at 50 µM. The highest bacteriostatic activity against Escherichia coli was found for four compounds in group 2 (MIC = 0.25–1 mM) and low bacteriostatic activity for group 3 (MIC > 4 mM). Some relationships between the structure of compounds and the values of the MIC are revealed. It was also found that four substances from different groups had the ability to inhibit the formation of colonies in E. coli from 1.3 to 5.7 times. Four compounds reduced specific biofilm formation by 40–60%. The tested substances did not induce the expression of the sulA gene controlled by the SOS system, which indicates the lack of genotoxic activity. None of the tested compounds had pro-oxidant activity. This was shown by both the absence of production hydrogen peroxide in a bacteria-free medium and inability to induce expression of the katG gene encoding HPI catalase in growing E. coli. The data obtained could be useful in the development of new drugs.

Introduction

The data obtained in recent decades point out that a microbiome may be considered as an additional organ that carries out important functions and significantly affects the health of the host organism. It has been shown that the state of the microbial community dwelling in the human gut affects the immune [1] and the central nervous system [2] of mammalians and along with gastrointestinal diseases [3] may contribute to the development of metabolic disturbances including diabetes [4] and obesity [5]. The use of antibiotics and other classes of drugs may change the status and activity of commensal microflora, thus affecting the whole organism [6]. This explains the growing interest in assessing the impact of potential drugs on the gastrointestinal microbiome.

Escherichia coli is a convenient object for the primary screening of chemical compounds for biological activity towards microorganisms of the human gastrointestinal tract. These bacteria, that are constant inhabitants of the intestine, have been studied in detail at the genetic and biochemical level and rapidly grow in laboratory conditions both on rich and minimal media. Due to wide accessibility of genetically engineered strains, test systems based on E. coli make it possible to investigate not only the antimicrobial activity of chemical compounds, but also other types of biological effects on living system, such as mutagenic (antimutagenic) and redox activity. The latter includes both oxidative (prooxidative) and antioxidant properties. Therefore, E. coli test systems enable a quite complete assessment of the biological effects of studied compounds.

The formation of oxygen and nitrogen active forms is a part of some normal physiological processes. Their overproduction exceeding the capacity of the antioxidant system represents a risk factor for the development of various pathologies among which neurodegenerative diseases make up a significant part [7–9].

In recent years, the role of ROS in the bactericidal effect of antibiotics of different classes has been actively studied [10–12]. This is especially interesting in connection with the radical-based hypothesis of antibiotic killing proposed by the Collins group [13–16]. For example, the mechanism of antituberculosis action of promising compounds—substituted triazines includes intracellular formation of nitrogen oxide (II) [17]. From this point of view, factors that alter the production of radicals may significantly affect the efficacy of antibiotics. The search for substances that initiate the formation of ROS in bacteria is considered as one of the promising strategies for creating effective antibacterial drugs [18, 19]. Similar approach is believed to be productive for anticancer therapy as well. Overall, a controlled generation of intracellular ROS using organic small molecules could alleviate the off-target effects in cells [20].

Enaminone fragment (N–C=C–C=O) containing compounds known for diverse biological effects [21–23]. Enaminones attract constant attention as a promising pharmacophore fragment in medicinal chemistry and a versatile tool with valuable redox properties in organic synthesis [24]. The aim of this work was to study the antimicrobial and antioxidant effects for the set of enaminone moiety containing compounds applying chemical and microbial test systems. Additionally, we expected to find some relationships between antimicrobial and antioxidant activities as well as to define structure features determining these properties. The selected set includes 24 oxo-derivatives of nitrogen-containing heterocycles and their acyclic analogues (Fig. 1) united by the similarity of the central enaminone fragment which is extended by an additional carbonyl group.

Fig. 1.

Groups of studied compounds: R¹ = Alk or Ar (1–4). X = OMe (1a, 1b, 4); Nar or NAlk (1b, 2, 3); O (2, 3). Y = O (4). X + Y = (4)

The type of substituents around the core structural fragment, as well as its rigidity/flexibility and stereoisomerism around the double bond, determines the difference in structures (Fig. 1, Table 1). Additionally, two compounds were included which do not contain enamine moiety (1b subset).

Table 1.

Radical-binding activity and MIC of test substances

Group 1a	R1	R2	X	AAPH (50 µM), %	DPPH (1 mM), %	MIC, mM
CBR-368	Ph	n- COOEt	OMe	NdR	26.5 ± 6.3	> 2
CBR-350	n-C2H5OC6H4	H	OMe	79.9 ± 1.2	27.1 ± 2.1	> 4
CBR-76	n-C2H5OC6H4	2,4-Me2	OMe	79.7 ± 4.6	21.3 ± 2	> 5
CBR-367	n-BrC6H4	2,4-Me2	OMe	72.5 ± 5.5	20.7 ± 1.9	> 4
CBR-366	t-Bu	n-Br	OMe	84.2 ± 4	19 ± 2.1	> 10

Group 1b	R1	X	AAPH (50 µM), %	DPPH (1 mM), %	MIC, mM
CBR-371	n-MeC6H4	OMe	60.6 ± 3.6	46.9 ± 1.5	1
CBR-352	Ph	n-NO2C6H4NH	56.2 ± 1.9	59.2 ± 1.2	0.5

Group 2	X	R1	AAPH (50 µM), %	DPPH (1 mM), %	MIC, mM
CBR-288	O	n-FC6H4	50.2 ± 1.7*	74.8 ± 0.8**	1
CBR-384	O	t-Bu	87.5 ± 1.3	69 ± 0.5**	0.5
CBR-385	O	C6H5	81.2 ± 6.2	71.4 ± 1.9**	NdM (> 0.125)
CBR-386	O	Me	88.5 ± 1.8	78 ± 0.1**	0.25
CBR-64	NC6H5	p-Cl-C6H4	44.3 ± 9.9	53 ± 1**	NdM (> 0.5)

Group 3	X	R1	AAPH (50 µM), %	DPPH (1 mM), %	MIC, mM
CBR-123	N-Me	n-MeC6H4	94 ± 5.6	2.3 ± 0.3	> 4
CBR-124	N-Me	n-ClC6H4	73 ± 1.9	1.5 ± 0.3	> 4
CBR-125	NH	n-ClC6H4	82.3 ± 1.7	5.2 ± 0.7	> 4
CBR-382	O	n-ClC6H4	81.7 ± 1.9	2.4 ± 0.4	> 4
CBR-324	O	n-MeC6H4	80.1 ± 2.8	1.5 ± 0.3	> 4

Group 4	R1	X	Y	R2	AAPH (50 µM), %	DPPH (1 mM), %	MIC, mM
CBR-160A	t-Bu	COOMe	O	6-Br	NdR	1.6 ± 0.6	NdM (> 1)
CBR-11	Ph	COOMe	O	H	NdR	0.7 ± 0.5**	NdM (> 0.25)
CBR-41	Ph	COOMe	O	6-Me	NdR	0.7 ± 0.3	> 4
CBR-383	Ph	COOMe	O	6-Br	NdR	1.1 ± 0.3	NdM (> 0.125)
CBR-266	Ph	H	O	H	NdR	0.9 ± 0.5	> 4
CBR-272	Ph	COOMe	NOMe	H	NdR	0.9 ± 0.3	> 4
CBR-375	Ph	COOMe	NOMe	8-Me	NdR	3.1 ± 1.3	4
CBR-379	Ph	H	NCH2Ph	6-Me	NdR	9.2 ± 0.7	> 2
CBR-376	Ph			H	16.3 ± 1.2	76.7 ± 3.5	0.25

Nd^M—MIC has not been determined due to the low solubility of substances

Nd^R—no radical-binding activity detected

*The studied concentration was 10 µM

**The studied concentration was 0.5 mM

It was expected, that overall these structural variations around the main core allow to reveal some structure–activity relationships.

Materials and Methods

Materials

DPPH (1,1-diphenyl-2-picrylhydrazyl), AAPH (2,2-azobis(2-amidinopropane) dihydrochloride), DMSO (dimethyl sulfoxide), agar, Luria–Bertani broth, 2-nitrophenyl-β-d-galactopyranoside (ONPG), Trolox, Amplex Red, Horseradish peroxidase (HRP) were obtained from Sigma-Aldrich Chemical Co. Other reagents were of analytical grade and acquired from Reachim.

All tested compounds were synthesized in Scientific and educational center for applied chemical and biological research at PNRPU, Russia. For all experiments, test compounds were dissolved in DMSO.

Evaluation of Radical-Scavenging and Prooxidant Activity In Vitro

ORAC Assay

The ORAC assay was performed according to Ou et al. [25] with some modifications. Fluorescein was used to measure fluorescent decay caused by the presence of peroxyl radicals, induced by AAPH. Fluorescence had being measured using Tecan Infinite M1000Pro (λ_ex 485 nm and λ_em 520 nm) for 120 min. Results are expressed as the degree of decrease in the fluorescence reduction caused by radicals.

DPPH-Assay

The ability of compound to bind to a stable DPPH radical was evaluated according to [26]. The inhibiting effect of DPPH was calculated according to the equation:

$$\text{Inhibition}=100\left(\text{absorbance of control}-\text{absorbance of sample}\right)/\text{absorbance of control},\%.$$

Measurement of H₂O₂ Production

Rate of hydrogen peroxide production was assayed by Amplex Red—horseradish peroxidase detecting system (AR/HRP) [27]. H₂O₂ concentration in samples was measured using spectrofluorimeter Shimadzu RF-1501 (λ_ex 563 nm and λ_em 587 nm).

Bacterial Strains

The parental strain of E. coli BW25113 (wild type) was obtained from the Keio collection (Escherichia coli Genetic Stock Center). The strain NM3012 was constructed by the introduction of sulA(sfiA)::lacZ fusion from E. coli DM4000 (a gift from M. Volkert) into the strain BW25113 via PI transduction. The strain NM3021 (katG::lacZ) was created by transformation of BW25113 with the plasmid pKT1033 [28].

Action of Chemicals on Bacterial Growth and Biofilm Formation in Plates

Minimum inhibitory concentration (MIC) values were determined using the microdilution method as outlined in [29]. Escherichia coli BW25113 cultures were inoculated to 96-well plates, containing dilution series of twofold substances concentrations and fresh M9 medium with glucose (0.2%). After 22 h of incubation at 37 °C, the cell density (OD₆₀₀) in each well of a 96-well plate was measured using the microplate spectrophotometer.

For bacterial growth rate studies, bacterial cells from an overnight culture after centrifugation were resuspended in fresh M9 medium with glucose (0.2%) to the initial OD₆₀₀ = 0.1. The culture suspension was added to 96-well plates (5 μl per well) containing 5 μl of the tested substance dissolved in DMSO and 190 μl of medium. Bacteria were cultivated in plates with shaking at 150 rpm at 37 °C. Cell growth was monitored by measuring OD₆₀₀ on microplate spectrophotometer Bio-Rad xMark every 15 min. The specific growth rate (µ) was calculated by equation µ = ΔlnOD₆₀₀/Δt, where t is the time in hours.

For colony-forming studies, 10 μl of the samples were tenfold diluted with 0.9% NaCl in separate plates. Then 10 μl drops (three drops of each dilution, which produced 5–50 colonies) were moved onto LB-agar plates. Samples for determination of the number of colony-forming units (CFU) were taken 30, 60 and 120 min after treatment of bacteria with compounds tested. Colonies were counted after 24-h incubation at 37 °C.

Biofilm formation was assessed using a crystal violet staining method [30, 31]. Bacteria were grown overnight in 96-well polystyrene microtitre plates statically at 37 °C for 22 h to obtain biofilms. Background wells contained cell-free medium with an equivalent content. Then, the OD₆₀₀ was measured and the broth was removed. The wells were washed twice with 200 μl of saline. The wells were air dried and 150 µl per well of 0.1% crystal violet solution was added. Staining was carried out for 30 min. Then the dye was removed and washed five times with distilled water. The wells were air dried for 1 h. Biofilms were quantified using 200 µl of 96% ethanol pipetted into each well. After 5 min, 125 µl of solution was transferred in a separate plate and OD₅₄₀ was measured using an xMark™ Bio-Rad spectrophotometer. Specific biofilm formation (SBF) was calculated by the formula:

$$\text{SBF}=\left(\text{A}-\text{B}\right)/\text{C}$$

where A is the OD₅₄₀ of the stained biofilms, B is the OD₅₄₀ of the stained background and C is the OD₆₀₀ value in planktonic culture, measured after 22 h of incubation.

Peroxide Induced Stress in Flasks

Cells grown overnight were precultured in 50 ml of fresh medium M9 with glucose (0.2%) at 37 °C in 250-ml flasks with shaking at 150 rpm to OD₆₀₀ of 0.4 and solution of substances were added. 15 min after 2 mM of H₂O₂ was added to culture and the optical density was measurement every 15 min for 105 min.

Determination of β-Galactosidase Activity

β-Galactosidase activity was determined as described by [29], using microplate spectrophotometer. β-Galactosidase activity was expressed in Miller units, calculated by the equation:

$$\text{Activity}=\left[\left({\text{OD}}_{420}-1.75\ast {\text{OD}}_{550}\right)/\text{t}\ast {\text{OD}}_{600}\right]\text{V}$$

where OD₄₂₀ and OD₅₅₀ represent the optical densities of the samples, OD₆₀₀—the optical density of the bacterial culture, t—the duration of exposition with 2-nitrophenyl-β-d-galactopyranoside (ONPG) in min, V—the dilution coefficient.

Statistical Analysis of the Data

Each result was indicated as the mean value of at least three independent experiments ± the standard error of the mean (SEM). Significant difference was analyzed by Student’s t-test. A p value of 0.05 was used as the cut-off for statistical significance. Results were analyzed by means of Statistica 6 (ver. 6, 2001; StatSoft Inc.) and GraphPad Prism 6.

Results

MICs

The MICs of all tested substances against E. coli bacteria were significantly higher than those of known antibiotics [32]. There was a definite relationship between the structure of the compounds and the MIC values. The largest number of compounds with relatively high bacteriostatic activity (0.25–1 mM) was in the group 2. In contrast, the group 3 contained compounds with low bacteriostatic activity (MIC > 4 mM). The groups 1 and 4 included compounds with both high MICs (> 2 and even > 10 mM) and low values (from 0.25 to 1 mM). It is noteworthy that, in the group 1, lower MICs values were exhibited by compounds of group 1b, containing 2-hydroxyl group instead of 2-arylamine group. The lowest MIC value in the group 4 was shown by the compound CBR-376, which differs from the rest of the group by containing a planar tricyclic system.

Radical-Binding Activity

The data for 26 compounds were obtained in two test systems evaluating antiradical activity of compounds—DPPH and ORAC tests.

The radical-binding activity (RBA) of compound determined by their ability to bind DPPH radical was very different (Table 1). In this test, the compounds of the group 2 had the highest activity, with the RBA values varying from 53% (CBR-64) to 78% (CBR-386) for 0.5 mM concentration. Group 3 compounds showed low activity with RBA values ranging from 1.5 (CBR-124) to 5.2 (CBR-125). Seven compounds of the group 4 had similar activity: from low (3.1%) to none activity level (0.5%). The only sample in this group has shown descent radical-binding activity (76.7%) was tricyclic CBR-376. The compounds of the group 1 occupied an intermediate position, showing values of radical- binding activity from 19 to 59.2%.

The ability of compounds to preclude fluorescein degradation by peroxyl radicals in ORAC test was high for all compounds from groups 1, 2 and 3 (Table 1), except CBR-368 from group 1 which contains an electron withdrawing COOC₂H₅ group in enamine fragment and turned out to be inactive. The compound CBR-123 of the group 3 had the highest radical-binding activity—94%. The activity of compounds from group 4 were not determined except CBR-376.

Action of Chemicals on Bacterial Growth in Plates

The data obtained from MIC test, ORAC and DPPH radical-binding assay were used to define a set of 10 compounds for further study of biological activity in a microbial test system. The focus was on representing structural diversity in this set as well as the wide range of values found in these three test systems. The final list includes CBR-366, CBR-371, CBR-124, CBR-125, CBR-382, CBR-384, CBR-386, CBR-288, CBR-11 and CBR-376.

Low solubility in water of some substances was a significant problem. In order to determine an effect on bacterial growth of these substances, they were added to the cell culture at the maximum possible concentration without precipitation. Therefore, the tested concentrations differ. For compounds with an established MIC value, a concentration equal to the MIC was chosen.

Tested compounds were added to the growing culture in the exponential growth phase, when the specific growth rate was about 0.66 h⁻¹. 15 min after addition of only DMSO, the growth rate of bacteria decreased to 0.24 h⁻¹ and remained close to this value during further cultivation (Fig. 2). The cultivation conditions in the plates do not allow to achieve the same mixing of the culture as, for example, in flasks. It results in a rapid decrease of oxygen in the wells of the plate and a transition from aerobic to microaerobic conditions. This effect is clearly visible in Fig. 2: the decrease in the growth rate in the first 15 min is due to the adaptation of bacteria to the cultivation conditions. A further change in the growth rate is probably associated with the depletion of oxygen in the environment.

Fig. 2.

Effect of test substances on the growth rate of E. coli in plates. *The arrow indicates the time when substances were added

The tested compounds had different effects on the rate of bacterial growth.

In group 1, treatment of bacteria with 0.5 mM CBR-366 caused a biphasic decrease in growth rate to almost zero at 15 and 60 min after the addition of the test compound. After 120 min, the growth rate (µ) increased to the value observed in the control (Fig. 2a). A two-phase drop in µ was also observed under the action of CBR-371, however, after the second drop, the growth rate increased only to 0.1 h⁻¹ (Fig. 2a). In group 2, reversible growth inhibition was observed with 0.5 mM CBR-384 and 0.1 mM CBR-288, respectively, 15 and 60 min after the addition of these compounds. No effect of 0.25 mM CBR-386 on the growth rate was found (Fig. 2b). In group 3, reversible inhibition of growth rate was observed only with 0.5 mM CBR-382 (Fig. 2c). In group 4, a statistically significant and reversible decrease in the growth rate was observed under the action of 0.25 mM CBR-376 (Fig. 2d).

Colony-Forming Ability and Biofilm Formation

It was found that CBR-371 (group 1), CBR-384, CBR-386 (group 2), and CBR-376 (group 4) reduced colony formation (CFU). Compared to cells treated with only DMSO, in bacteria treated with these compounds, the number of CFUs decreased by 1.8, 5.7, 1.3 and 2.5 times, respectively (Fig. 3).

Fig. 3.

The number of colony-forming units (CFU) under the action of the test substances. *Indicates statistically significant difference between the compound and DMSO only, p < 0.05

Three of these compounds also had the ability to reduce specific biofilm formation. Compared to cells treated with DMSO alone, the SBF value in cells growing in the presence of CBR-371, CBR-384, and CBR-386 were 58.0, 37.3 and 62.7% respectively (Fig. 4). In addition, a decrease in SBF was observed under the action of CBR-124 (39.9%).

Fig. 4.

Specific biofilm formation (SBF) under the action of the studied substances. *Indicates statistically significant difference from the compound and DMSO only, p < 0.05

Prooxidant and Genotoxic Activity

CBR-288, CBR-384 and CBR-386 from the group 2 and CBR-124, CBR-125 and CBR-382 from the group 3 were tested. At first, it was determined if the compounds are able to produce hydrogen peroxide in a bacteria-free medium. The AmplexRed fluorophore was used and no peroxide was detected after 30-min incubation.

The second method for prooxidant activity evaluation was based on the use of the genetically engineered strain E. coli NM3021 (katG::lacZ), carrying fusion between the promoter of katG gene and of the structural β-galactosidase gene. The katG gene encodes HPI catalase, which catalyzes the decomposition of H₂O₂ in growing E. coli. It was shown that the studied substances do not affect the expression of the katG gene. Thus, both methods did not reveal the prooxidant activity of the tested compounds.

To assess the ability of substances to damage DNA, a genetically engineered strain of E. coli NM3012 carrying the sulA::lacZ gene fusion was used. The sulA gene is part of the SOS system that regulates the response of bacteria to DNA damage. It was found that none of the 10 studied substances increases the expression of the sulA gene, which indicates the absence of a genotoxic effect.

Peroxide Induced Stress

The antioxidant activity of the tested compounds was evaluated by their effect on the growth rate of E. coli treated with 2 mM hydrogen peroxide. CBR-288, CBR-384 and CBR-386 from the group 2 and CBR-124, CBR-125 and CBR-382 from the group 3 were tested. The first three compounds showed high radical-binding activity in the DPPH-test, the other three showed low activity. In the bacterial test, these substances differed significantly in terms of MICs and survival (Table 1, Fig. 3).

Preliminary studies have shown that the effect of the tested compounds on bacterial growth under peroxide stress is more pronounced when E. coli bacteria are grown under aerobic conditions achieved in flasks placed in a rotary shaker.

In the presence of DMSO alone, treatment of bacteria with 2 mM H₂O₂ led to an initial decrease in the specific growth rate by about 5 times. The second phase occurred 15 min after the addition of H₂O₂. During this phase, growth resumed and by the end of cultivation reached the level observed in the untreated culture (Fig. 5). Under the described conditions, in the absence of H₂O₂, the addition of 0.5 mM CBR-384 to the medium led to an irreversible decrease in the growth rate by about 10 times (µ = 0.05 h⁻¹).

Fig. 5.

Effect of 0.5 mM CBR-384 (a), 0.25 mM CBR-386 (b) and 0.25 mM CBR-124 (c) on the growth rate of E. coli treated with hydrogen peroxide. The action of CBR-124 is typical for compounds such as CBR-288, CBR-125 and CBR-382

The addition of H₂O₂ to the culture treated with CBR-384 increased the growth rate to 0.35 h⁻¹, which was approximately 30% lower than in the control and peroxide treatment alone (Fig. 5a). Thus, when cultured in flasks, CBR-384 did not demonstrate a protective effect against peroxide stress, rather on the contrary, peroxide reduced the bacteriostatic effect of CBR-384.

Under conditions described above, significant growth inhibition of E. coli was also found under the action of 0.25 mM CBR-386. By the end of cultivation, the growth rate of cells treated with this compound was approximately three times lower than in the control. In the initial phase, the addition of peroxide alone and peroxide in the presence of CBR-386 had the same effect on bacteria. However, the growth rate was restored to normal levels by the end of the culture if the cells were treated with peroxide alone. Under the combined action of CBR-386 and H₂O₂, the growth rate remained as low as under the action of CBR-386 alone (Fig. 5b). In total, CBR-384 and CBR-386 had a pronounced bacteriostatic effect and did not protect bacteria growing in flasks from peroxide stress. It should be noted that in the plates CBR-386 did not affect the growth rate, and CBR-384 inhibited growth only in the first 15 min after addition.

In the absence of H₂O₂, compounds CBR-124, CBR-125, CBR-288, and CBR-382 had no effect on the growth rate of E. coli. The presence of these compounds in the medium did not change the response of bacteria to peroxide stress (Fig. 5c).

As a result of a two-stage screening of nitrogen-containing heterocyclic compounds and their acyclic analogs, four compounds were identified with the highest activity against E. coli bacteria (Table 2).

Table 2.

Activities of four substances in growing cultures of E. coli and in vitro RBA tests

Substances	AAPH (50 µM), %	DPPH (1 mM1/ 0.5 mM2), %	MICs, mM	CFUa	Growth rate in platesb	Growth rate in flasksc	H2O2 actiond	Biofilm formation (SBF)a
CBR-371 (1b group)	60.6 ± 3.6	46.9 ± 1.51	1	1.9	++	NT	NT	1.7
CBR-384 (2 group)	87.5 ± 1.3	69.0 ± 0.52	0.5	3.9	+	Inhibition	+	2.7
CBR-386 (2 group)	88.5 ± 0.1	78.0 ± 0.12	0.25	1.3	0	Inhibition	+	1.6
CBR-376 (4 group)	16.3 ± 1.2	76.7 ± 3.51	0.25	2.9	0	NT	NT	1.1

^aFold decrease in CFU or SBF compared to treated by DMSO cells

^b(++)—Irreversible and (+) reversible growth rate inhibition. 0—no effect

^cGrowth rate in flacks. NT not tested

^d(+)—substance increases growth inhibition caused by H₂O₂

Discussion

It is assumed that the test, based on the ability to bind DPPH radical, reflects a mixed mode of action: hydrogen transfer (HAT) and single electron transfer (SET) mechanisms [33]. The more pronounced radical-binding activity of group 2 compounds, along with lack of activity of group 3 compounds may indicate a contribution to the efficiency of DPPH radical binding by the presence of a long conjugated system. Group 1a compounds, that are open-chain analogues of group 2, demonstrated intermediate values. Apparently free rotation of N-aryl ring in 1a structures hinders more efficient for conjugation planar geometry which is common for structures 2a [34]. The assumption of the importance of the long conjugated system is also consistent with the fact that the only compound in group 4 that showed activity in the test with DPPH is CBR-376, which, unlike the other representatives of group 4, contains an additional annelated cycle, i.e. more efficient conjugated system.

When AAPH is used as an initiator of the formation of radicals, the mechanism of hydrogen atom transfer (HAT) is realized [35, 36]. In this test, the differences in the values of radical scavenging activity of the compounds of the first three groups were significantly less pronounced than in the DPPH-test (Table 1). With the exception of CBR-368, all of these compounds had RBA ranging from 44.3 to 94.0. Inactivity of CBR-368 possibly can be attributed to the presence of an electron withdrawing COOC₂H₅ group. The high activity should be noted for compounds of group 3, which was not effective in the test based on the interception of the DPPH radical. Similarly to DPPH based test, out of nine compounds in group 4, only one substance (CBR-376) demonstrated RBA.

Since different mechanisms are involved in two test systems applied for radical-binding evaluation it could be expected that the relationship between antibacterial action and RBA may differ for these two types of data. In group 1 higher RBA values in the DPPH test for CBR-352 and CBR-371 corresponded to their stronger bacteriostatic effect (MIC). The same combination of these parameters was observed for all compounds of the group 2 and for CBR-376 from the group 4. Compounds of group 3 had a weaker bacteriostatic effect and showed a lower RBA in the test with DPPH. Thus, in all the above cases, an increase in the radical-binding activity in the test with DPPH corresponded to an increase in the bacteriostatic effect. In other cases, high RBA in the ORAC test was combined with a wide range of MIC values. It is noteworthy that in the group 1, lower MICs values were shown by 1b compounds—CBR-371 and CBR-352, containing a hydroxyl group instead of an arylamine group in position 2. The lowest MIC value in the group 4 was shown by the CBR-376 compound, which differs from the rest of the group 4 in that it contains a planar tricyclic system.

In the compounds of both groups 2 and 3 diketoenamine moiety is involved in the formation of a heterocyclic system (morpholine or piperazine) which is fused with aromatic ring in group 2 and not annulated in group 3. Apparently, this structural feature—the presence of the benzene ring—makes a significant contribution to the action of substances on bacteria: CBR-384 and CBR-386 demonstrated bacteriostatic effect while their counterparts CBR-124 and CBR-125 are not active. Bacteriostatic effect depends on oxygen availability. Interestingly, compound CBR-124, unlike compound CBR-125, inhibited biofilm formation and did not affect growth parameters. Perhaps this effect is associated with a greater lipophilicity of the CBR-124 since an important first step in biofilm formation—attachment and adhesion bacteria to surfaces—is governed by physicochemical factors of the media and bacteria prefer to colonize more hydrophobic surfaces [37].

Correlation analysis of the data shown in Table 2 revealed that with an increase in the radical-binding activity of the compounds (in the test with DPPH), their bacteriostatic activity increased, as was determined by the decrease in MIC (r =  − 0.997). A direct relationship was also found between the ability of these compounds to reduce the formation of colonies (CFU) and inhibit the formation of biofilms (SBF) (r = 0.8).

Conclusions

Studied here representatives of nitrogen-containing heterocyclic compounds and their acyclic analogs—all containing diketoenamine moiety—have various combinations of antimicrobial and radical-binding activity which depends both on the structural features and type of the test system. Most of studied compounds don’t affect E. coli growth and may be considered as prospective radical-binding structures.

The MIC is the most commonly used characteristic of the antimicrobial activity of promising antibiotics. This work shows that MIC determination alone is not sufficient for evaluation antimicrobial activity of the compounds. A more complete characterization can be given by a combination of several complementary methods, including the determination of the growth parameters, biofilm formation, and pro- and antioxidant activity.

Author Contributions

GT performed experiments and wrote the paper. OK and AB synthesized the studied substances and described the structural features for article. GS and OO developed a research plan, were responsible for interpreting the results and editing the article. All authors read and approved the final manuscript.

Funding

The reported study was funded by RFBR, project number 20-34-90016 and state assignment AAAA-A19-119112290009-1.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.