Ultrastructural Study of the Effects of Hybrid Compounds of Natural Monoterpene Carvacrol and Synthetic Cationic Amphiphile DL₄12 on S. aureus and E. faecalis Cells

Abstract

Ultrastructure changes in S. aureus and E. faecalis bacteria incubated with synthetic cationic amphiphile DL₄12 and its hybrids with the natural monoterpene carvacrol were studied. The hybrid compounds DL₄CAR-6, DL₅CAR-6, DLpCAR-6, and DLoCAR-6 contained two carvacrol molecules and differed in central linker structure. The study was conducted on ultrathin sections of bacteria fixed by the Ryter–Kellenberger method and on a Jem 1400 transmission electron microscope (Jeol, Tokyo, Japan). Ultrastructure changes in S. aureus and E. faecalis incubated with compound DL₄12 were species-specific. Destructive changes in S. aureus cells when exposed to DL₄12 compound and all DL₄12-carvacrol hybrids did not differ. DL₄12 and DL₄12-carvacrol hybrids in E. faecalis cells damaged all structures except the cell wall. Compound DL₄12 and its hybrids disrupted the ultrastructure of nucleoid and DNA strands in both bacterial species. Complete disorganization of ribosomes in cells of both bacteria occurred upon incubation with compound DL₄12 and its carvacrol-bearing analog DL₄CAR-6. Inclusions in bacterial cells exposed to all compounds had the same ultrastructure. The study showed that all compounds used possess multitarget properties; the structure of the central linker of hybrid compounds plays a significant role in the nature of their damaging effect on S. aureus and E. faecalis cells.

1. Introduction

Growing bacterial resistance to clinically used antibacterial compounds (ABCs) is a global health problem. The development and implementation of new antibiotics into practice is not keeping pace with the rate of resistance mechanisms development in bacteria, which is why there is an increasing demand for new antibiotics [1,2]. The World Health Organization (WHO) has established four criteria for innovation, namely: a new ABC (i) must be from a new class of chemicals, (ii) have a new target, (iii) have a new mechanism of action and (iv) be free from cross-resistance to existing antibiotics [3].

Heimann D. and colleagues published in 2025 an analysis of seventy ABCs, based on fundamentally new chemical compounds that are currently in various stages of clinical trials [4]. More than two-thirds of these compounds were aimed at disrupting the function of a single target, primarily a specific protein. However, point mutations in bacterial DNA quickly alter the structure of the target protein, which reduces the affinity of the antibiotic for it and, accordingly, reduces the effectiveness of drugs acting on a single target. Therefore, multitargeted ABCs directed at non-proteogenic targets, such as RNA and lipids, are coming to the fore [4]. Multitarget ABCs contain two or more pharmacophore groups, which are combined into a single molecule using a linker, which ensures action on several targets in the bacterial cell. The effectiveness of multitargeted compounds against antibiotic-resistant bacteria is undeniable, but their development is not active. According to a review [4], only one in 70 compounds has two targets, and another is a mixture of two compounds directed at different targets.

Previously, we proposed compounds based on quaternary salts of 1,4-diazabicyclo [2.2.2]octane (DABCO) with the general structure shown in Figure 1A as multitarget ABCs. These compounds, according to their chemical structure, belong to cationic amphiphiles, one example of which is antimicrobial peptides, the antibacterial properties of which have been described in the published literature [5,6,7].

Cationic amphiphiles are believed to be capable of causing disruption of the transmembrane potential, leakage of cytoplasmic contents, and, ultimately, cell death [8]. We previously showed that DABCO-based cationic amphiphiles possess RNA hydrolyzing activity in vitro [9]. It could be assumed that after penetrating the cell membrane, these compounds would affect bacterial RNA.

The structure of the central linker can potentially influence the antibacterial properties of cationic amphiphiles, and we have previously studied this influence. Thus, it was shown that the length of the central linker of 4–6 carbon atoms ensures maximum activity of RNA hydrolysis by cationic amphiphilic compounds based on DABCO [10,11]. Antibacterial activity also showed a bell-shaped dependence on the length of the hydrophobic residue (alkyl chain), with the highest activity observed in compounds containing alkyl chains with 10–12 carbon atoms [12].

A study of the antibacterial activity of the synthesized cationic amphiphiles showed that the most effective compound against a number of Gram-positive and Gram-negative microorganisms was DL₄12 [12]. The name of the compound reflects its structure: D—DABCO, L_4—tetramethylene linker, 12—dodecyl residue, Figure 1A. We revealed a clear multitarget effect of compound DL₄12 on Salmonella enterica and Staphylococcus aureus using transmission electron microscopy (TEM): the ultrastructure of the cell envelope, ribosomes and nucleoid of bacteria was disrupted [13,14].

The cationic amphiphile DL₄12 has two hydrophobic tails that may be replaced by compounds with known antibacterial activity, which may lead to a stronger final hybrid. We implemented this approach by incorporating the antibiotic ciprofloxacin into the structure of a DABCO-based cationic amphiphile. (Figure 1B) [15]. To understand the mechanisms of hybrid structure action, we conducted a comparative study of the effects of the antibiotic ciprofloxacin, a cationic amphiphile DL₄12, and the hybrid structure on S. aureus using TEM. The effect of the hybrid structure on S. aureus, although combining the effects of ciprofloxacin and the cationic amphiphile, was not their simple sum [13].

The next step in the development of hybrid compounds is this study, which examined the damaging effects of new cationic amphiphile compounds based on DABCO and the natural monoterpene carvacrol (Figure 1C).

Carvacrol, a component of many natural essential oils, has demonstrated potent damaging activity against a variety of bacteria and fungi. Physicochemical studies of carvacrol’s action have revealed membrane depolarization, potassium and phosphate leakage, increased membrane permeability, and inhibition of cellular respiration [16,17,18,19,20]. Obviously, changes in physiological parameters inevitably lead to disruption of bacterial cell structure. However, only two TEM studies of S. aureus have been published, describing cytoplasmic coagulation, the formation of mesosome-like structures, membrane damage, and the “release” of cellular contents [20,21,22].

It was clear to us that the described effects of carvacrol were due to its action as a single molecule, whereas incorporation into a compound could “neutralize” its effects. Nevertheless, carvacrol’s impressive array of activities prompted us to select it as a component of the DABCO-based hybrid compound.

In this work, we investigated the ultrastructural changes induced by the hybrid compounds DABCO and carvacrol in the cells of two Gram-positive bacteria. The Staphylococcus aureus and Enterococcus faecalis used in our work are known as causative agents of many human diseases [23,24], and S. aureus is included in the WHO List of Priority Bacterial Pathogens [25].

2. Results

2.1. Ultrastructure of S. aureus and E. faecalis Intact Cells

Intact cells of S. aureus and E. faecalis were round cocci measuring 500–600 nm in size with a fairly high electron density (Figure 2A,D). The cells were actively dividing; in sections, almost half of the cells had a division septum. Dividing cells of E. faecalis had a characteristic ovoid shape, and the cell wall formed angles near the division septum (Figure 2D).

The cell envelope of both bacterial species consisted of a cell wall, which was a thick layer of peptide glycans in the form of a thin fibrous material of medium electron density, and an underlying cytoplasmic membrane. The ultrastructure of the membranes of both species was visible as a thin, low-density band only in some areas; it was generally indistinguishable in sections (Figure 2C,F). The intermediate layer between the cell wall and cytoplasmic membrane was composed of electron-dense “grains” (1–2 nm).

The fixation technique we used allowed us to identify differences in the cell wall structure of the bacteria studied. The cell wall of S. aureus (25–30 nm) had a smooth surface or was covered with a “fur” of short filaments; three layers were recognized within the wall: the outer layer and the membrane-adjacent layer had a higher electron density than the layer located between them (Figure 2C). The cell wall of E. faecalis (20–25) nm had a smooth surface and was composed of homogeneous material of medium or low electron density (Figure 2F).

The cell cytoplasm was granular with medium to high electron density in both bacterial species. In the cytoplasm of S. aureus cells, rare small (5–10 nm) particles of high electron density were observed, which might be individual ribosome subunits (Figure 2C). The cytoplasm ultrastructure of S. aureus and its electron density were the same throughout the cell. In E. faecalis cells, the cytoplasm was composed of two zones (layers): a wide outer zone filled with spherical particles (10–12 nm) with medium electron density, and a narrow zone around the nucleoid composed of homogeneous material of lower electron density than the rest of the cytoplasm (Figure 2D).

The nucleoid in S. aureus and E. faecalis cells was represented by a single fragment of irregular shape, usually located in the center of the cell (never touching the membrane) (Figure 2A,D). The nucleoid in S. aureus cells was seen as a zone of granular material of the average electron density, in which DNA strands were sometimes visible, while in E. faecalis, nucleoid DNA strands were seen more clearly. In the S. aureus nucleoid, the “blurred” DNA strands were uniformly distributed without any visible sign of organization (Figure 2B). In contrast, in the E. faecalis nucleoid, the DNA strands were more distinct, had a higher electron density, and were organized into bundles of varying morphology (Figure 2E).

Thus, intact cells of S. aureus and E. faecalis had the same set of cellular structures, but differed in subtle details, and this led to a different visual appearance of the cells.

2.2. S. aureus and E. faecalis Ultrastructure Alteration by DL₄12

Exposure of S. aureus cells to DL₄12 resulted in damage to the ultrastructure of the cell wall, cytoplasmic membrane, cytoplasm and nucleoid. In contrast, the cell wall of E. faecalis cells retained the structure observed in intact cells; however, the cytoplasmic membrane, cytoplasm, and nucleoid of the bacterium cells were damaged (Figure 3).

Interaction of DL₄12 with the S. aureus envelope led to significant loosening of the cell wall and its detachment from the cytoplasmic membrane surface; the electron density of the intermediate layer decreased (Figure 3A,C). A distortion of the division septum was noted in most dividing S. aureus cells. Penetration of the DL₄12 compound into E. faecalis cells did not cause visible changes in the cell wall; however, the intermediate layer was altered; it was not visible in sections or was seen as fragmented.

The cytoplasmic membrane of both bacterial species was observed over the entire surface of cells incubated with the DL₄12 compound (Figure 3C,F), whereas in intact cells, the membrane was poorly visualized. In some areas, the cytoplasmic membrane was replaced by layered electron-dense inclusions with clear boundaries (Figure 3), in the thickness of which layers (5–6 nm) were distinguished. The inclusions gradually increased in size and penetrated deeper into the cell, their shape approached roundness, and the material became homogeneous, starting from the center; finally, spherical homogeneous inclusions were visible on sections (Figure 3G–J). Apparently, the formation of the inclusions was the result of the interaction of the DL₄12 compound that had penetrated into the cells with cellular components. The structure of inclusions did not differ in cells of S. aureus and E. faecalis.

Incubation of S. aureus cells with DL₄12 caused “disappearance” of ribosomes; the cytoplasm was uniformly fine-grained (Figure 3C). In E. faecalis cells, visible ribosomes were deformed and partially aggregated; they acquired an irregular shape and stuck together to form particles 20–25 nm in size (Figure 3F). The cells of E. faecalis lost the ribosome-free zone around the nucleoid, and ribosome aggregates were found right up to the borders of the nucleoid (Figure 3D,F).

The nucleoid structure in S. aureus cells varied when incubated with DL₄12, apparently reflecting its disorganization. Instead of a single nucleoid, several fragments were observed in sections, some of which were displaced toward the cytoplasmic membrane (Figure 3A,B). The nucleoid’s “filling” material became electron-transparent, and DNA strands were better seen against this background; signs of structural organization of DNA strands were not detected (Figure 3B).

The nucleoid of E. faecalis cells acquired a roundish shape when cells were incubated with DL₄12 (Figure 3D–F). The DNA strands were organized into bundles and had very clear outlines (Figure 3E). In some areas, homogeneous material with average electron density was detected between the strands, which may be the result of DNA strand fraying.

2.3. S. aureus Ultrastructure Alteration by DL₄CAR-6, DL₅CAR-6, DLpCAR-6 and DLoCAR-6 Compounds

The ultrastructure of S. aureus cells incubated with DL₄CAR-6, DL₅CAR-6, DLpCAR-6, and DLoCAR-6 showed changes indistinguishable from each other and from those observed with DL₄12 (Figure 3 and Figure 4). Cells of all samples demonstrated significant damage to the ultrastructure of the cell wall, the presence of inclusions; the cytoplasm became homogeneous, visible ribosomes were not observed, and the ultrastructure of the nucleoid changed (Figure 4).

Under the influence of all the above compounds, the nucleoid of S. aureus cells was fragmented and, in some cells, could touch the cytoplasmic membrane. DNA strands had an average electron density (Figure 4E–H), and the pattern of changes completely corresponded to that observed under treatment of S. aureus with DL₄12 (Figure 3). In dividing cells, a curvature of the division septum was observed.

The results obtained indicate that all carvacrol-containing compounds tested in this study, regardless of the structure of the central linker, effectively damaged S. aureus cells and demonstrated a multitarget effect. All compounds caused complete ribosome disorganization and changes in the nucleoid structure in S. aureus cells.

2.4. E. faecalis Ultrastructure Alteration by DL₄CAR-6

The DL₄CAR-6 compound significantly altered the ultrastructure of all E. faecalis cells, with the exception of the cell wall. In most cells (95%), the cytoplasmic membrane was visible throughout the entire cell wall (Figure 5A,C), whereas in intact cells, it was seen only in small areas. Membrane-associated inclusions gradually penetrated into the cytoplasm and increased in size. Electron-dense particles of irregular shape (15–22 nm) appeared in the cytoplasm (Figure 5A–C), which, judging by their size, could be the result of ribosome aggregation. The cytoplasmic layer around the nucleoid disappeared, and the peripheral layer containing ribosomes spread to the borders of the nucleoid.

The nucleoid in E. faecalis cells was severely damaged, the bundles of DNA strands were disrupted, and some strands were replaced by granular, electron-dense material (Figure 5A–C).

Images of E. faecalis cells treated with DL₄CAR-6 were different from those of the bacterium in control and DL₄12-treated samples. Cells looked unusually dense, the cytoplasm was densely filled with electron-dense granules, and we designated such cells as “dense”.

A small number of E. faecalis cells (5%) incubated with DL₄CAR-6 underwent changes similar to those observed in cells exposed to DL₄12. The inclusions appeared in the cells, the cytoplasm became homogeneous and had average electron density, ribosomes were not visible and DNA strands became more distinct.

2.5. E. faecalis Ultrastructure Alteration by DL₅CAR-6

Compound DL₅CAR-6 differed from DL₄CAR-6 by only one additional methylene group in the central linker, and this small feature caused significant differences in the compound’s effect on E. faecalis cell ultrastructure.

Incubation of E. faecalis cells with the DL₅CAR-6 compound altered the ultrastructure of all cells, but in different ways. In most of the cells (95%) incubated with compound DL₅CAR-6, the cytoplasmic membrane became more visible, and membrane-associated inclusions appeared (Figure 5F). The cytoplasm had average electron density and looked homogeneous; the layer surrounding the nucleoid was absent. This state of cytoplasm allowed us to see ribosome uniform distribution throughout the cytoplasm (Figure 5D–F). The ribosomes did not form electron-dense aggregations, as in the case of DL₄CAR-6. Compared to the “dark” cells described above, these cells looked much “lighter”, and from now on, we will designate them as “light”. The nucleoid contained granular material of average electron density, in which DNA strands were poorly visible; the bundles were absent (Figure 5D–F).

The changes in the remaining 5% of E. faecalis cells incubated with DL₅CAR-6 were similar to the changes observed in cells incubated with the DL₄CAR-6 compound: cells had a “dark” cell phenotype.

2.6. E. faecalis Ultrastructure Alteration by DLpCAR-6

Incubation of E. faecalis cells with the DLpCAR-6 compound differently affected the cells. The ultrastructure of 70% of cells (Figure 5G–I) was very similar to that described above in the “light” cells under the influence of the DL₅CAR-6 compound.

The cytoplasmic membrane was hardly visible, intermediate layer was prominent. Inclusions associated with the membrane appeared in the cells, the cytoplasm had average electron density and looked homogeneous, the ribosomes were clearly visible and were evenly distributed throughout the cytoplasm (Figure 5G–I). The ribosomes did not form electron-dense aggregations, as in the case of DL₄CAR-6. The nucleoid was severely affected; it was composed of an amorphous substance of the average electron density, in which DNA strands were hardly visible (Figure 5G–I).

The morphology of remaining E. faecalis cells (30%) was similar to the “dark” cells observed after incubation with the DL₄CAR-6 compound: membrane-associated inclusions were observed in the cells, ribosome aggregates were detected in the cytoplasm, and the structure of DNA strand bundles was altered.

2.7. E. faecalis Ultrastructure Alteration by DLoCAR-6

Incubation of E. faecalis cells with DLoCAR-6, as with other carvacrol-containing compounds, affected the ultrastructure of all cells differently. The majority of the cells (80%) corresponded to “dark” cells in morphology (Figure 5J–L): contained inclusions, ribosome aggregates were located in the cytoplasm, and damage to the structure of DNA strand bundles was observed in the nucleoid (Figure 5J–L). The remaining cells (20%) of the sample were “light,” similar to those observed during incubation with DL₅CAR-6 and DLpCAR-6.

3. Discussion

The development of multitarget drugs that attack several targets in bacterial cells is considered one of the promising approaches to combating bacteria with multiple antibiotic resistance [26,27]. Such compounds can be obtained by synthesizing molecules that combine groups with already known antibacterial activity [27,28].

In this work, we studied the effect of hybrids of carvacrol (a monoterpene with antibacterial activity [17]) and DL₄12, a compound we previously synthesized that exhibits multitarget activity and high antimicrobial efficacy against various bacteria in vitro [11].

Using TEM, we demonstrated that the pronounced destructive effect of DL₄12 is primarily due to damage to the protein-synthesizing apparatus of Gram-positive and Gram-negative bacteria [13,14].

The aim of this work was to study the ultrastructure of S. aureus and E. faecalis cells exposed to DL₄12-carvacrol hybrids (Figure 1). The chemical synthesis and characteristics of hybrids are presented in the Supplementary File. Each hybrid molecule included two carvacrol molecules that replaced the hydrophobic tails of DL₄12; the molecules differed in the structure of the central linker.

(1)

DL₄CAR-6: the central linker was identical to DL₄12 (tetramethylene);

(2)

DL₅CAR-6: the central linker was longer (pentamethylene);

(3)

DLoCAR-6: a benzene group formed the central linker (aromatic linker replaced the aliphatic one);

(4)

DLpCAR-6: the central linker was formed by a p-xylene molecule (aromatic).

All four hybrid compounds caused damage to the ultrastructure in S. aureus and E. faecalis cells; however, the ultrastructural manifestations of the effects varied both between bacterial species and between the compounds. Interspecies differences were undoubtedly related to differences in the morphology of intact bacteria (Figure 2), which reflected the features of their molecular organization.

To our knowledge, no studies have been published on the relationship between the ultrastructure and molecular organization of bacteria, so we are unable to include the results of other authors in the discussion of the results obtained in this work.

All carvacrol-containing compounds introduced into the culture medium with Gram-positive bacteria came into contact with the cell wall. This contact resulted in disorganization of the wall structure for S. aureus cells (Figure 2 and Figure 3), whereas the cell wall of E. faecalis retained the structure observed in intact cells (Figure 2 and Figure 3). It should be noted that the destruction of the S. aureus cell wall and the stability of the E. faecalis wall were observed during incubation of bacteria with all the compounds used. The changes occurring inside the cells allow us to conclude that the cell wall of both bacterial species was not a barrier for DL₄12 and all its derivatives.

The cell wall of E. faecalis apparently has pores suitable for the migration of all five compounds, allowing them to move into the cell. The cell wall of S. aureus, which has a different molecular composition [29], becomes disorganized and loosened, forming “pathways” for the penetration of compounds. We believe that this effect may be due to the structure of the S. aureus cell wall, which contains pentapeptide glycine cross-links in the structure of cell wall murein, in contrast to the more complex and diverse cross-links in the murein composition of the cell walls of other bacteria [29,30]. Severe damage to the cell wall of S. aureus by cationic amphiphiles may be critical for its viability: the MIC values for all cationic amphiphiles for this bacterium are incomparably lower than for E. faecalis (

Table 1. Minimum inhibitory concentration (MIC) values of the synthesized compounds.

Compounds	MIC (µM)
	E. faecalis	S. aureus
DL412	63	4
DL4CAR-6	63	4
DL5CAR-6	250	8
DLoCAR-6	250	8
DLpCAR-6	500	8
Ciprofloxacin *	400	2

* Reference compound. DL₄12, where D is DABCO (1,4-diazabicyclo [2.2.2]octane), L₄ is tetramethylene linker, 12 is a dodecyl residue, and its conjugates with two carvacrol molecules—DL₄CAR-6, DL₅CAR-6, DLoCAR-6 and DLpCAR-6 (Figure 1)—where CAR is carvacrol, L₅ is pentamethylene linker, Lp is 1,4-bis(methyl)benzene, Lo is 1,2-bis(methyl)benzene, and 6 is a pentamethylene residue.

).

Exposure to the compound DL₄12 resulted in ribosome disorganization, and the cytoplasm of S. aureus and E. faecalis cells appeared homogeneous and less electron-dense (Figure 3). We observed a similar effect in Salmonella Gram-negative cells [14] and S. aureus cells when studying the DL₄12-ciprofloxacin hybrid [13]. Interestingly, when incubated with DL₄12, S. aureus ribosomes “disintegrated,” while E. faecalis ribosomes deformed and clumped together (Figure 3).

Compound DL₄12 exhibited RNAase activity in vitro, which was discovered shortly after its synthesis [9]. This activity logically explains the compound effect on bacterial ribosomes, and we use this hypothesis to analyze our results. RNAase activity was fully manifested in the DL₄CAR-6 compound, while in other carvacrol-bearing compounds, its manifestations were similar to those of DL₄12 in S. aureus, but significantly reduced in E. faecalis cells.

Replacement of the hydrophobic “tails” of DL₄12 with carvacrol while maintaining the tetramethylene central linker (compound DL₄CAR-6) led to ribosome aggregation and the formation of electron-dense particles in E. faecalis cells. In S. aureus cells, exposure to DL₄CAR-6 resulted in complete ribosome disorganization, as did DL₄12. Complete ribosome disorganization in S. aureus cells also occurred upon exposure to all compounds containing carvacrol.

Increasing the length of the central linker by one methylene group (compound DL₅CAR-6) did not cause changes in the integrity of ribosomes in E. faecalis cells; however, it altered their distribution in the cell: ribosomes were evenly distributed in the cytoplasm, not concentrated at the periphery, as in intact cells. Replacing the central methylene linker with aromatic ones (DLoCAR-6 and DLpCAR-6) resulted in the same change in ribosome localization in E. faecalis cells. It is likely that intact E. faecalis cells possess mechanisms that ensure the concentration of ribosomes in the cell’s periphery, and the compounds DL₅CAR-6, DLoCAR-6, and DLpCAR-6 disrupt the functioning of these mechanisms.

Changes in nucleoid morphology were a striking change in the cell structure of S. aureus and E. faecalis under the influence of DL₄12 and its carvacrol-modified derivatives. In this work, we used the fixation of bacteria with the Ryter–Kellenberger method [31], which ensures fixation of the nucleoid in accordance with modern concepts: DNA is immersed in a “filling” substance that maintains its structure and functions. The substance contains many different types of RNA [31], which could be a target for the RNAse activity of the compounds studied.

TEM of ultrathin sections is not the best method for studying the nucleoid, but we nevertheless identified several changes in its structure as far as possible. When exposed to DL₄12, the nucleoids of S. aureus and E. faecalis bacteria changed shape, the DNA strands became thicker and straighter, and the electron density of the “filling” substance decreased. When incubated with all carvacrol-containing compounds, nucleoid fragmentation was observed in all S. aureus cells, and the DNA strands became less distinct.

On the contrary, in E. faecalis cells, when incubated with DL₄12, the nucleoid became rounded and acquired a clear boundary. The structure of the “filling” substance became granular, possibly due to the appearance of “grains” associated with the disorganization of DNA strands. The described nucleoid changes persisted when exposed to all carvacrol-containing compounds; however, the extent of the changes varied significantly.

Compound DL₅CAR-6 induced less pronounced changes in E. faecalis cells than DL₄CAR-6, which could be related to the addition of one methylene group to the central linker. All E. faecalis cells exhibited destructive changes in varying morphology when exposed to compounds containing a central aromatic linker. The cells were divided into two large groups based on the nature of their ultrastructural changes (“dark” and “light” cells). We believe these differences may be due to the stage of the bacterial life cycle.

Our results showed that replacing two hydrophobic tails in compound DL₄12 with two carvacrol molecules, while maintaining the tetramethylene linker, did not affect the nature and level of damage to S. aureus and E. faecalis cells, and the changes were the greatest among carvacrol-containing compounds.

Studies of the effects of other compounds have convincingly demonstrated (using E. faecalis as an example) that the structure of the central linker in cationic amphiphiles plays an important role in the development of cell damage detected at the ultrastructural level. Even a small increase in the length of the aliphatic linker in the DL₅CAR-6 compound significantly altered the pattern of cell damage. The replacement of the tetramethylene linker with aromatic ones not only affected the pattern and level of cell damage but also revealed the heterogeneity of the E. faecalis cell population in terms of the character of cells’ ultrastructural changes.

The results of our study confirm the informative and effective use of transmission electron microscopy of ultrathin sections for studying bacteria and their changes during incubation with various antibiotics. This method allows, in particular, for the direct determination of the multitarget properties of developed compounds and their cellular targets. The results we obtained will undoubtedly be useful to developers of multitarget antibacterial compounds.

4. Materials and Methods

4.1. Antibacterial Compounds

The following compounds were used in the work: DL₄12, where D is DABCO (1,4-diazabicyclo [2.2.2]octane), L₄ is tetramethylene linker, 12 is a dodecyl residue, and its conjugates with two carvacrol molecules—DL₄CAR-6, DL₅CAR-6, DLoCAR-6 and DLpCAR-6 (Figure 1)—where CAR is carvacrol, L₅ is pentamethylene linker, Lp is 1,4-bis(methyl)benzene, Lo is 1,2-bis(methyl)benzene, 6 is a pentamethylene residue (Figure 1C).

The compound DL₄12 was first synthesized in the Laboratory of Organic Synthesis of the Institute of Chemical Biology and Fundamental Medicine, Siberian Branch of the Russian Academy of Sciences (Novosibirsk, Russia) [11,12]. The design and synthesis of carvacrol-containing compounds was carried out for the first time at the above-mentioned laboratory and is presented in Supplementary File S1, and ¹H and ¹³C NMR Spectra of the synthesized carvacrol-containing compounds are presented in Supplementary File S2. Stock solutions of ABCs were prepared in dimethyl sulfoxide (DMSO) (Serva, Heidelberg, Germany) at a concentration of 20 mM and diluted with Mueller-Hinton broth (MHB, Thermo Fisher Scientific, Waltham, MA, USA) for incubation with bacteria. We used all ABCs at a concentration of 100 µM (0.01 mM), as in previous studies [14].

4.2. Microorganisms and Their Incubation with ABCs

The work used bacteria strains obtained from the Collection of Extremophilic Microorganisms and Type Cultures of the Institute of Chemical Biology and Fundamental Medicine of the Siberian Branch of the Russian Academy of Sciences (Novosibirsk, Russia): Gram-positive bacteria Staphylococcus aureus ATCC 25923 and Enterococcus faecalis ATCC 51299.

Both bacterial strains were grown on Luria–Bertani agar medium (LB, Difco Laboratories, Baltimore, MD, USA), then inoculated into MHB broth and cultured overnight in a shaker-incubator at 37 °C and 180 rpm (ES-20, SIA Biosan, Riga, Latvia). To prepare the samples for TEM, the bacterial cell concentration was adjusted with MHB to 1 × 10⁷ CFU/mL (OD595 = 0.5). Optical density (OD) was determined using a densitometer DEN-1 (Biosan, Riga, Latvia). The resulting suspension (25 mL) was added to 50 mL plastic tubes, followed by 25 mL of a solution of one of the ABCs in MHB. The final ABC concentration was 100 µM (0.01 mM). Samples were incubated at 37 °C and 180 rpm (ES-20, SIA Biosan, Riga, Latvia) for exactly 15 min. Intact bacterial suspensions were used as a control. After incubation with ABCs, 500 μL of 25% glutaraldehyde solution (SPI supplies, West Chester, PA, USA) was added to the bacterial suspensions to stop the interaction of ABCs with bacterial cells.

4.3. Determination of Minimum Inhibitory Concentrations (MICs)

Antimicrobial susceptibility testing was performed using the broth microdilution method in 96-well plates (TPP, Trasadingen, Switzerland), according to the guidelines of the European Committee on Antimicrobial Susceptibility Testing (EUCAST) [32]. A culture grown on LB agar (BD, Franklin Lakes, NJ, USA) was inoculated into MHB and incubated overnight. Stock solutions of the compounds in DMSO were prepared at a concentration of 20 mM. The final cell concentration of the overnight broth cultures was adjusted in MHB to approximately 5 × 10⁵ CFU/mL (OD595 = 0.1). Bacterial growth was assessed after 24 h by measuring the optical density at 595 nm for each well using a microplate reader (Uniplan, PICON, Moscow, Russia). MIC values were defined as the lowest compound concentration that completely inhibited visible bacterial growth. All experiments were performed in triplicate, and the highest MIC value among the three replicates was selected for further analysis. MIC values are given in Table 1.

4.4. Samples Processing for TEM Study

The bacterial suspensions (50 mL) containing 500 μL of 25% glutaraldehyde solution were centrifuged at 8600× g for 10 min (2-16PK, SIGMA Laborzentrifugen GmbH, Osterode am Harz, Germany), and the supernatant was removed. The pellets were transferred to 2 mL plastic tubes and fixed using a modified Ryter–Kellenberger method [31] in 1.5 mL of 1% OsO₄ in 0.2 M citrate buffer, pH 6.0, and left overnight at room temperature. Then the pellets were washed with citrate buffer, and 1 mL aqueous solution of uranyl acetate (0.5%, pH 3.0) was added; the mixture was incubated for 2 h.

Fixed samples were dehydrated in solutions of ethanol and acetone of increasing concentration and embedded in an Epon-Araldite mixture (EMS, Hatfield, PA, USA), forming two blocks from each sample. A series of ultrathin sections were prepared on a Leica EM UC7 ultramicrotome (Leica, Wetzlar, Germany), using a diamond knife (Diatome, Nidau, Switzerland). The ultrathin sections were contrasted with lead citrate and uranyl acetate (both from EMS, Hatfield, PA, USA) solutions and examined in a Jem 1400 TEM (JEOL, Tokyo, Japan) at an accelerating voltage of 80 kV. Digital images were obtained using a Veleta side-entry TEM camera (EM SIS, Muenster, Germany). The measurements were made using the iTEM software version 5.2 (EM SIS, Muenster, Germany).

4.5. Ultrastructure Analysis of Damaged Cells

For TEM studies, 20–25 ultrathin sections were prepared from each block. Each ultrathin section contained approximately 1000 bacterial cell fragments, through which the section passed randomly. We analyzed only those cell fragments that the section passed through near the equator. To obtain a clear image, it is important that the section be perpendicular to the cells. If the section plane is tilted, the linear structure of the membranes is not visible, and the ribosomes become “blurred”. These features of electron microscopy necessitate viewing a large number of cells to select those suitable for analysis. We analyzed at least 200 matched bacterial cell sections for each chemical compound.

To determine the proportion of damaged cells, direct counting was performed on ultrathin sections on a TEM screen. At 6000–8000× magnification, grid units completely filled with bacteria were selected. Then, at a magnification of 20,000–30,000×, cross-sections of 100 S. aureus or E. faecalis cells were selected and counted. The percentage of cells with destructive changes was determined for compounds that led to different changes in bacteria. (data are presented in the relevant section).