Synthesis and In Vitro Antimicrobial SAR of Benzyl and Phenyl Guanidine and Aminoguanidine Hydrazone Derivatives

A series of benzyl, phenyl guanidine, and aminoguandine hydrazone derivatives was designed and in vitro antibacterial activities against two different bacterial strains (Staphylococcus aureus and Escherichia coli) were determined. Several compounds showed potent inhibitory activity against the bacterial strains evaluated, with minimal inhibitory concentration (MIC) values in the low µg/mL range. Of all guanidine derivatives, 3-[2-chloro-3-(trifluoromethyl)]-benzyloxy derivative 9m showed the best potency with MICs of 0.5 µg/mL (S. aureus) and 1 µg/mL (E. coli), respectively. Several aminoguanidine hydrazone derivatives also showed good overall activity. Compounds 10a, 10j, and 10r–s displayed MICs of 4 µg/mL against both S. aureus and E. coli. In the aminoguanidine hydrazone series, 3-(4-trifluoromethyl)-benzyloxy derivative 10d showed the best potency against S. aureus (MIC 1 µg/mL) but was far less active against E. coli (MIC 16 µg/mL). Compound 9m and the para-substituted derivative 9v also showed promising results against two strains of methicillin-resistant Staphylococcus aureus (MRSA). These results provide new and potent structural leads for further antibiotic optimisation strategies.


Introduction
Bacterial infections with multidrug-resistant pathogens, such as methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococci (VRE), and multidrugresistant Escherichia coli, pose an increasing threat to the global human population [1][2][3]. These drug-resistant bacteria can cause lethal infections, making the treatment of infected patients increasingly difficult. Therefore, the discovery of novel therapeutic agents that are active against drug-resistant microorganisms remains a fundamental challenge, especially for medicinal chemistry. Most antibiotics currently in clinical use target one of the metabolic pathways of DNA, RNA, protein, or cell wall synthesis [4]. Due to the emergence of pathogens with reduced susceptibility to currently available antibiotic therapies, there is an urgent need to discover new antibiotics with new targets and mechanisms of action.
In recent years, bacterial cell division has attracted considerable attention as a potential antibiotic target [5,6]. Cell division in bacteria is achieved through a highly dynamic macromolecular complex that is characterized by a time-dependent assembly of specific cell division proteins [7], formed in an orchestrated fashion by the essential tubulin homolog FtsZ (filamentous temperature-sensitive protein Z). Most bacteria depend on FtsZ as the main protein for efficient cell division [8,9]. Therefore, FtsZ has been validated as a highly promising target for antibacterial intervention [5].
Antibacterial compounds known to target FtsZ are eg Berberine 1 and Sanguinarine 2 ( Figure 1). Berberine 1 is a natural plant alkaloid that has been described to target E. coli FtsZ [10,11]. It binds to FtsZ with high affinity in a region that overlaps with the Recent examples of synthetic, small-molecule-type, investigational antibiotics with guanidine motifs are compounds 7 and TXA497 8 (Figure 1) [27][28][29][30]. Biphenyl derivative 8 especially, despite its structural simplicity, displays quite a few remarkable antimicrobial properties [29,30]. First, 8 shows low MICs (1 µ g/mL) against several variations of S. aureus including MRSA. Furthermore, 8 also displays a very low MBC value (minimum A class of compounds regularly reported in the antibiotic context are guanidine derivatives [24]. Guanidine functionalities are commonly found in many biologically relevant molecules that constitute a versatile class of molecules with a wide range of applications. Compounds, either natural or synthetic, containing guanidine as a core unit, either in open or in cyclic form, display an array of pharmacological properties, including antimicrobial, antiviral, antiparasitic, and antifungal activities [24]. The great appeal of the guanidine moiety can be attributed to its hydrogen-bonding capability and protonatability at physiological pH in the context of interaction with biological targets. Bacterial cell envelopes are negatively charged, which may attract the guanidinium cation via electrostatic interaction and favour the binding of these compounds, leading to the disruption of cell membranes and cell walls. Therefore, guanidine derivatives have been exploited as privileged structural motifs in designing novel drugs for the treatment of various infectious and non-infectious diseases. Over many years, a large variety of synthetic small molecules with one or several guanidine units has emerged [24]. There are also synthetic polymeric guanidine derivatives that display very potent antibiotic activities against MRSA in skin infections and against the growth of Aspergillus parasiticus [25,26]. Recent examples of synthetic, small-molecule-type, investigational antibiotics with guanidine motifs are compounds 7 and TXA497 8 ( Figure 1) [27][28][29][30]. Biphenyl derivative 8 especially, despite its structural simplicity, displays quite a few remarkable antimicrobial properties [29,30]. First, 8 shows low MICs (1 µg/mL) against several variations of S. aureus including MRSA. Furthermore, 8 also displays a very low MBC value (minimum bactericidal concentration) of 1 µg/mL, leading to a ratio of MBC/MIC = 1 which essentially means that 99.9% of all bacterial cells are killed at the same concentration of minimum inhibition (MIC). The same MBC/MIC ratio for 8 was observed against E. coli, but, at 16 µg/mL, the compound concentrations were much higher [30]. However, more remarkably, it was found that 8 exhibits only a minimal potential for inducing resistance in S. aureus [30]. Initially, 8 was proposed to act on FtsZ dynamics and it was shown to target the GTP binding site of recombinant FtsZ in vitro [30]. However, in bacterial cells, 8 targets the bacterial cell membrane in addition to FtsZ, with some cells predominantly showing the effects of FtsZ inhibition and others predominantly showing the effects of cell membrane disruption [31]. The guanidino/amidino functionality is a key contributor to the antibacterial activity of this class of compounds.
The structural simplicity of 8 was highly attractive from a medicinal chemistry standpoint. As a part of a program to design a library of novel antimicrobial compounds, only a small set of benzyl guanidine 9 and aminoguanidine hydrazone derivative 10 with a variety of benzyloxy groups was initially envisaged ( Figure 2). However, the compound series was later expanded, and new broadly-related, but also diverse, guanidine derivatives were additionally synthesised and their antimicrobial activities against S. aureus and E. coli were evaluated [32]. The most potent inhibitors were then tested against the drug-resistant strains MRSA 3 and MRSA 15.
Molecules 2023, 28,5 4 of 39 derivatives were additionally synthesised and their antimicrobial activities against S. aureus and E. coli were evaluated [32]. The most potent inhibitors were then tested against the drug-resistant strains MRSA 3 and MRSA 15.

Chemistry
The meta-substituted benzyl guanidine compounds 9a−q were constructed from the corresponding 3-aminomethylphenol derivatives 11a−c via a guanylation reaction using Boc-protected S-methylisothiourea [33], followed by the benzylation of the phenol group under basic conditions to give 13a−q. Finally, treatment with trifluoroacetic acid in dichloromethane led to 9a−q, obtained as their guanidinium trifluoroacetate or chloride salts (Scheme 1). Benzyl guanidine derivatives 9r−v were prepared under the same conditions using 4-aminomethylphenol 14 as the starting material.

Chemistry
The meta-substituted benzyl guanidine compounds 9a-q were constructed from the corresponding 3-aminomethylphenol derivatives 11a-c via a guanylation reaction using Boc-protected S-methylisothiourea [33], followed by the benzylation of the phenol Molecules 2023, 28, 5 4 of 39 group under basic conditions to give 13a-q. Finally, treatment with trifluoroacetic acid in dichloromethane led to 9a-q, obtained as their guanidinium trifluoroacetate or chloride salts (Scheme 1). Benzyl guanidine derivatives 9r-v were prepared under the same conditions using 4-aminomethylphenol 14 as the starting material.

Chemistry
The meta-substituted benzyl guanidine compounds 9a−q were constructed from the corresponding 3-aminomethylphenol derivatives 11a−c via a guanylation reaction using Boc-protected S-methylisothiourea [33], followed by the benzylation of the phenol group under basic conditions to give 13a−q. Finally, treatment with trifluoroacetic acid in dichloromethane led to 9a−q, obtained as their guanidinium trifluoroacetate or chloride salts (Scheme 1). Benzyl guanidine derivatives 9r−v were prepared under the same conditions using 4-aminomethylphenol 14 as the starting material. In a benzylguanidine-based structural subset, the meta-and para-substituted compounds 20a−e and 24a−e (Scheme 2) were constructed from 3-and 4-aminomethylaniline 17 and 21 via a guanylation reaction using Boc-protected S-methylisothiourea, followed by the treatment of the resulting 18 and 22, respectively, with the corresponding arylsulfonyl chloride or benzoyl chloride in the presence of a base to achieve Boc-protected derivatives 19a−e and 23a−e. Treatment with trifluoroacetic acid in dichloromethane led to the removal of the Boc groups, and the final compounds 20a−e and 24a−e were obtained as their guanidinium trifluoroacetate salts. In a benzylguanidine-based structural subset, the metaand para-substituted compounds 20a-e and 24a-e (Scheme 2) were constructed from 3-and 4-aminomethylaniline 17 and 21 via a guanylation reaction using Boc-protected S-methylisothiourea, followed by the treatment of the resulting 18 and 22, respectively, with the corresponding arylsulfonyl chloride or benzoyl chloride in the presence of a base to achieve Boc-protected derivatives 19a-e and 23a-e. Treatment with trifluoroacetic acid in dichloromethane led to the removal of the Boc groups, and the final compounds 20a-e and 24a-e were obtained as their guanidinium trifluoroacetate salts.
To explore the potential effects of conformational restriction of the guanidine moiety, the tetrahydroisoquinoline-based compounds 29a-b and 33 were prepared via the route shown in Scheme 3. First, compounds 29a-b were prepared from the corresponding hydroxy-substituted 1,2,3,4-tetrahydroisoquinolines 25a-b by N-Boc protection, benzylation, guanylation, and Boc deprotection. Similarly, guanylation of 7-bromo-1,2,3,4tetrahydroisoquinoline 30 gave the corresponding 2-carboximidamide derivative 31 that was converted to 33 through a route involving a palladium-catalysed Suzuki coupling [30], followed by the removal of the Boc groups with TFA.
To explore the potential effects of conformational restriction of the guanidine moiety, the tetrahydroisoquinoline-based compounds 29a−b and 33 were prepared via the route shown in Scheme 3. First, compounds 29a−b were prepared from the corresponding hydroxy-substituted 1,2,3,4-tetrahydroisoquinolines 25a−b by N-Boc protection, benzylation, guanylation, and Boc deprotection. Similarly, guanylation of 7-bromo-1,2,3,4-tetrahydroisoquinoline 30 gave the corresponding 2-carboximidamide derivative 31 that was converted to 33 through a route involving a palladium-catalysed Suzuki coupling [30], followed by the removal of the Boc groups with TFA.
A subset of aminoguanidino hydrazone derivatives 10a−t (Scheme 5) was prepared in two steps from the corresponding 3-hydroxybenzaldehyde derivatives 37a−c, by benzylation of the hydroxyl group and condensation of the corresponding aldehydes 38a−t with N-aminoguanidine bicarbonate [34]. Most of the target compounds were obtained as their chloride salts and a few as acetates. Imidazole aminoguanidine (41a−d) and pyrrole aminoguanidine derivatives (41e−h) were synthesised as chloride salts in the same way using HCl (0.5M in MeOH) at 80 °C.
For phenylguanidino derivatives (Scheme 6), a guanylation reaction of para-aminophenol 42 generated the intermediate 43, which was subsequently subjected to benzylation to afford the N,N'-di-Boc protected guanidine derivatives 44a−c. Successive treatment with trifluoroacetic acid in dichloromethane gave 45a−c. Phenyl guanidine derivatives 48a−b were achieved in four steps. In the first step, 4-aminobenzylamine 21 was treated with either benzoyl chloride or benzenesulphonyl chloride and triethylamine in DMF to give 46a−b. Guanylation with Boc-protected S-methylisothiourea in the presence of mercury (II) chloride [35] then achieved 47a−b. Subsequent treatment with TFA removed the Boc groups to give 48a−b. For phenylguanidino derivatives (Scheme 6), a guanylation reaction of para-aminophenol 42 generated the intermediate 43, which was subsequently subjected to benzylation to afford the N,N -di-Boc protected guanidine derivatives 44a-c. Successive treatment with trifluoroacetic acid in dichloromethane gave 45a-c. Phenyl guanidine derivatives 48a-b were achieved in four steps. In the first step, 4-aminobenzylamine 21 was treated with either benzoyl chloride or benzenesulphonyl chloride and triethylamine in DMF to give 46a-b. Guanylation with Boc-protected S-methylisothiourea in the presence of mercury (II) chloride [35] then achieved 47a-b. Subsequent treatment with TFA removed the Boc groups to give 48a-b.  For a subset of benzyl guanidine derivatives 51a-b and 56 (Scheme 7), the reductive amination reaction of the aldehyde 38a (Ar = 2,3-dichlorophenyl) generated the amino intermediates 49a-b, which underwent a guanylation reaction to form the N,N -di-Boc protected guanidine derivatives 50a-b. Deprotection of the Boc groups generated the guanidinium trifluoroacetate salts 51a-b. The intermediate 53 was obtained through the reduction of the aldehyde 38a, followed by halogenation of the resulting benzyl alcohol 52. Treatment of 53 with S-methyl-N,N -bis(tert-butoxycarbonyl)isothiourea under basic conditions gave 54. Nucleophilic substitution of 54 with methylamine afforded the N,N -di-Boc protected guanidine 55, which was then hydrolysed in TFA to give the final compound 56. For a subset of benzyl guanidine derivatives 51a−b and 56 (Scheme 7), the reductive amination reaction of the aldehyde 38a (Ar = 2,3-dichlorophenyl) generated the amino intermediates 49a−b, which underwent a guanylation reaction to form the N,N'-di-Boc protected guanidine derivatives 50a−b. Deprotection of the Boc groups generated the guanidinium trifluoroacetate salts 51a−b. The intermediate 53 was obtained through the reduction of the aldehyde 38a, followed by halogenation of the resulting benzyl alcohol 52. Treatment of 53 with S-methyl-N,N'-bis(tert-butoxycarbonyl)isothiourea under basic conditions gave 54. Nucleophilic substitution of 54 with methylamine afforded the N,N'di-Boc protected guanidine 55, which was then hydrolysed in TFA to give the final compound 56.    Table 1 reveals that a significant proportion of the synthesised benzyl guanidine derivatives is more potent against S. aureus than against E. coli. For 9a-m (R 1 = R 2 = R 3 = H), only 9d, 9h, and 9k are more potent against E. coli than against S. aureus, with 9h showing the best activity (MIC = 4 µg/mL) of them. However, the biggest difference in potency against the two strains was found for 9d with MICs of >256 µg/mL and 8 µg/mL, respectively. The most potent compound in this subset was the 2-Cl-3-CF 3 derivative 9m with MICs of 0.5 µg/mL and 1 µg/mL, respectively, but 2,3-dichloro derivative 9g showed very similar potency with MICs of 1 µg/mL against both microbial strains. For 9n-o (R 1 = MeO, R 2 = R 3 = H), reduced potency was found. Comparison of 9n with 9c showed a significantly reduced potency for an H to MeO substitution. A MIC of 128 µg/mL for 9n and MICs of 16 µg/mL and 32 µg/mL for 9c were found. 4-Chloro derivatives 9o and 9b showed a similar pattern against E. coli but appeared to be equipotent against S. aureus with MICs of 8 µg/mL for both compounds. Derivatives 9p-q (R 1 = H, R 2 = R 3 = F) proved very potent against S. aureus with MICs of 0.5 µg/mL and 1 µg/mL, respectively. However, in contrast to 9g and 9m, which were two compounds very active against both strains, 9p-q were significantly less potent against E. coli. A few para-substituted benzyl guanidine derivatives 9r-v were also evaluated. The monochlorobenzyl derivatives 9s-t proved both moderately active and did not show any difference in potency between the two microbial strains. The dichlorobenzyl derivatives 9u-v, on the other hand, proved significantly more potent against S. aureus than against E. coli, with 9v showing the best overall antimicrobial activities with MICs of 0.5 µg/mL and 4 µg/mL, respectively. Substitution of one hydrogen at the N atom of the guanidine unit in 9g, where the benzyl unit is attached with a methyl or a methoxyethyl group, led to a significant decrease in antimicrobial potency, from a MIC of 1 µg/mL for 9g to MICs of 32 µg/mL for 51a-b. A slightly better, but still very weak activity, against S. aureus was found for 56, a derivative where the methyl group was introduced at the terminal N atom (MIC of 16 µg/mL).
The antimicrobial activities of tetrahydroisoquinoline guanidine derivatives 29a-b and 33 are summarised in Table 3. A comparison of 29a and 29b reveals that substitution in the 5-position seems more favourable than in the 7-positon. However, even 5-substituted derivative 29b proved only moderately active with MICs of 8 µg/mL for both S. aureus and E. coli. Replacing the O-benzyl linkage in the 7-position with a directly attached aromatic ring system seems to further reduce antimicrobial activity. 4-tert-Butylphenyl derivative 33 showed only weak potency with MICs of 64 µg/mL and >128 µg/mL, respectively.
The antimicrobial activities of the three benzyloxy guanidine derivatives against S. aureus and E. coli are summarised in Table 4. All compounds of this class that we tested so far displayed no significant potency. Table 1. Antibacterial activities of benzyl guanidine derivatives 9a-v, 51a-b, and 56 against S. aureus and E. coli. potent against S. aureus than against E. coli, with 9v showing the best overall antimicrobial activities with MICs of 0.5 µg/mL and 4 µg/mL, respectively. Substitution of one hydrogen at the N atom of the guanidine unit in 9g, where the benzyl unit is attached with a methyl or a methoxyethyl group, led to a significant decrease in antimicrobial potency, from a MIC of 1 µg/mL for 9g to MICs of 32 µg/mL for 51a−b. A slightly better, but still very weak activity, against S. aureus was found for 56, a derivative where the methyl group was introduced at the terminal N atom (MIC of 16 µg/mL).  Benzyl guanidines with aminosulfonylaryl or aminobenzoyl motifs as substituents either in the meta-(20a−e) or para-position (24a−e) proved uniquely inactive for both sets of compounds (all MICs > 32 µ g/mL, Table 2).

Cpd
Ar The antimicrobial activities of tetrahydroisoquinoline guanidine derivatives 29a−b and 33 are summarised in Table 3. A comparison of 29a and 29b reveals that substitution in the 5-position seems more favourable than in the 7-positon. However, even 5-substituted derivative 29b proved only moderately active with MICs of 8 µ g/mL for both S. aureus and E. coli. Replacing the O-benzyl linkage in the 7-position with a directly attached aromatic ring system seems to further reduce antimicrobial activity. 4-tert-Butylphenyl derivative 33 showed only weak potency with MICs of 64 µ g/mL and >128 µ g/mL, respectively.   The antimicrobial activities of the three benzyloxy guanidine derivatives against S. aureus and E. coli are summarised in Table 4. All compounds of this class that we tested so far displayed no significant potency.
The antimicrobial activities of the three benzyloxy guanidine derivatives against S. aureus and E. coli are summarised in Table 4. All compounds of this class that we tested so far displayed no significant potency.
The MIC values against S. aureus and E. coli of aminoguanidine hydrazone derivatives 10a-t and 41a−h are summarised in Table 5. Compounds 10a−f (R 1 = R 2 = H) showed overall moderate to good antimicrobial activities with the majority of MICs between 4 µ g/mL and 16 µ g/mL. Compound 10d was significantly active against S. aureus (MIC of 1 The MIC values against S. aureus and E. coli of aminoguanidine hydrazone derivatives 10a-t and 41a-h are summarised in Table 5. Compounds 10a-f (R 1 = R 2 = H) showed overall moderate to good antimicrobial activities with the majority of MICs between 4 µg/mL and 16 µg/mL. Compound 10d was significantly active against S. aureus (MIC of 1 µg/mL) but less against E. coli (MIC of 16 µg/mL) and was the most potent compound against S. aureus of all aminoguanidine hydrazone derivatives. Methoxy-substituted derivative 10g (R 1 = MeO, R 2 = H) appeared to be less potent (MICs of 16 µg/mL and 8 µg/mL) when directly compared with 10a (both MICs of 4 µg/mL). Chloro-substituted compounds 10h-t (R 1 = H, R 2 = Cl) showed MICs between 4 µg/mL and 32 µg/mL. The most potent derivatives here were 10j and 10r-s, all of which are mono-substituted in the benzyloxy motif (4-Cl, 3-CF 3 , 4-CF 3 ). All three compounds showed MICs of 4 µg/mL against both S. aureus and E. coli. Heterocyclic derivatives, including benzimidazole aminoguanidine hydrazones 41a-d as well as pyrrole aminoguanidine hydrazones 41e-h, displayed only moderate antimicrobial activities, with 41d showing the best potency against S. aureus (MIC of 8 µg/mL).       16 16 * Acetate salt. Table 6 shows a summary of the MIC values for para-substituted phenyl guanidine derivatives 45a−c and 48a−b against S. aureus and E. coli. All derivatives showed only moderate antimicrobial potency with 45a displaying the best activity against S. aureus (MIC of 8 µ g/mL).

Cpd
Ar 16 16 * Acetate salt. Table 6 shows a summary of the MIC values for para-substituted phenyl guanidine derivatives 45a-c and 48a-b against S. aureus and E. coli. All derivatives showed only moderate antimicrobial potency with 45a displaying the best activity against S. aureus (MIC of 8 µg/mL). Table 6. Antibacterial activities of phenyl guanidine derivatives 45a-c and 48a-b against S. aureus and E. coli. Table 6. Antibacterial activities of phenyl guanidine derivatives 45a−c and 48a−b against S. aureus and E. coli.

Cpd
Ar

Antimicrobial Activity against MRSAs
The benzyl guanidine derivatives 9m and 9v were also tested against MRSA 3 and MRSA 15. Bacterial growth was recorded against time at various concentrations of 9m and 9v. The example data for the growth of the MRSAs when treated with differing doses of 9v are shown in Figure 3. The minimum inhibitory concentration (MIC) and the survival index (SI) were established for each experiment (Table 7). In order to determine the SI, the growth of the treated bacteria was compared to the growth (measured as an increase in optical density over time) of the control, untreated bacteria, and the MIC was determined as the concentration that allowed an SI reduction of greater than 50%. As can be seen in Figure 3, as representative plots, although the total optical density change during the control growth varies slightly between the experiments, the growth curve progress and, hence, the overall growth profile of the control bacteria are very consistent, and each experiment's individual control bacterial growth curve allows for minor variations in growth between the different compound treatments. The majority of the doses above the

Antimicrobial Activity against MRSAs
The benzyl guanidine derivatives 9m and 9v were also tested against MRSA 3 and MRSA 15. Bacterial growth was recorded against time at various concentrations of 9m and 9v. The example data for the growth of the MRSAs when treated with differing doses of 9v are shown in Figure 3. The minimum inhibitory concentration (MIC) and the survival index (SI) were established for each experiment (Table 7). In order to determine the SI, the growth of the treated bacteria was compared to the growth (measured as an increase in optical density over time) of the control, untreated bacteria, and the MIC was determined as the concentration that allowed an SI reduction of greater than 50%. As can be seen in Figure 3, as representative plots, although the total optical density change during the control growth varies slightly between the experiments, the growth curve progress and, hence, the overall growth profile of the control bacteria are very consistent, and each experiment's individual control bacterial growth curve allows for minor variations in growth between the different compound treatments. The majority of the doses above the determined MIC values demonstrate a complete absence of growth of the bacteria over the

Chemistry
Methods and Materials: All chemicals and anhydrous solvents were purchased from either Sigma-Aldrich (now Merck: Gillingham, UK) or Alfa Aesar (Heysham, UK). All organic solvents of AR grade were supplied by Fisher Scientific (Loughborough, UK). Melting points were determined using a Stanford Research Systems Optimelt MPA100 (Stanford Research Systems, Sunnyvale, CA, USA) and were uncorrected. Thin-layer chromatography (TLC) was performed on pre-coated aluminium plates (Merck, silica gel 60 F 254 ). The products were visualised either by UV irradiation at 254 nm or by staining with 5% w/v phosphomolybdic acid in ethanol, followed by heating. Flash column chromatography was performed on pre-packed silica gel columns (RediSep Rf) and gradient elution (solvents indicated in text) on the Combiflash Rf system (Teledyne Isco). 1 H NMR spectra were recorded with a Bruker 400 or 500 MHz spectrometer. The chemical shifts were reported in parts per million (ppm), either relative to the corresponding solvent residual peaks or tetramethylsilane (TMS) as an internal standard. High-resolution mass spectra (HRMS) were recorded on a Bruker MicroTOF with ESI. All compounds were ≥95% pure by 1 H NMR spectroscopy.

Methods Cell Culture
The bacteria were grown for 17 h in 20 mL of tryptic soy broth (TSB) at 30 • C with oscillation (90 oscillations per minute). A total of 1 mL of the overnight culture was taken and centrifuged at 880 g for 7 min. The supernatant was removed, and the pellet resuspended in PBS. The centrifugation step was then repeated, and the resulting pellet redissolved in an appropriate volume to adjust the concentration of the bacterial solution to a density of 1 × 10 5 cells/mL based on the haemocytometer calculation in a 30 mL universal test tube.

Treatment Protocol
The synthetic compounds were dissolved initially at a concentration of 1 mg/mL in a suitable solvent (ethanol, methanol, or DMSO) before dilution with water in order to generate the relevant concentrations used within the bioassay. Each solvent was tested separately for its own toxicity, and it was ensured that the dilution required to produce the working solutions for the assay was sufficient to remove any toxic effect of the initial solvent used. The bacterial solution (10 µL) was added to 50 µL of either the synthetic compound solution or water as a negative control in a 96-well plate. This was sealed with a polyethylene seal prior to being analysed in a Skanit platereader (Thermoscientific, Cambridge, UK). The optical density at 550 nm was then recorded once every hour for 24 h, with a shaking step immediately prior to each reading being taken.

Quantification Method
The acquired data were then used to determine the growth of each species with or without the addition of each synthetic compound solution. The MIC values were determined as the minimum concentration of compound required to reduce the survival index to less than 50%. The survival indices (Table 7) were determined from the absorbance data as the ratio of the optical density of the control bacteria at the mid-log point of growth to the comparative optical density of the treatment bacteria multiplied by 100, as published previously [36].

Conclusions
A range of benzyl, phenyl guanidine, and aminoguanidine hydrazone derivatives were synthesised and evaluated. Some derivatives displayed excellent antimicrobial activity. The best benzyl guanidine derivatives 9g, 9m, and 9v displayed antimicrobial MICs of 0.5-1.0 µg/mL against S. aureus and 1-4 µg/mL against E. coli. The best aminoguanidine hydrazone derivative 10d showed very good potency against S. aureus (MIC 1 µg/mL) but was significantly less potent against E. coli (16 µg/mL), as indeed was benzyl guanidine 9p (0.5 µg/mL and 16 µg/mL, respectively). Overall, 10a, 10j, and 10r-s were less potent against S. aureus (MIC of 4 µg/mL) and more potent against E. coli (MIC of 4 µg/mL) compared to 10d. The most potent benzyl guanidines 9m and 9v were tested against methicillin-resistant Staphylococcus aureus (MRSA 3 and MRSA 15) and showed very promising potencies with MICs of 0.5-2.0 µg/mL and low survival indices (1.76-12.76%). Further work is currently underway to establish the mechanism of action of these new guanidine derivatives, and it seems clear, as expected, that the guanidino/amidino functionality is a key contributor to the antibacterial activity. However, a preliminary further evaluation of a selection of the most potent compounds shows potentially complex profiles that may not include the most expected bacterial cell division machinery, such as FtsZ, as a direct target. Since it is known that the lead agent TXA497 8 targets both the bacterial membrane in addition to FtsZ, with some cells exhibiting the effects of membrane disruption [31], the focus of future studies on the most potent subset of optimised compounds will address both FtsZ and the bacterial cell membrane, and also potential targets elsewhere.