Cationic Peptidomimetic Amphiphiles Having a N-Aryl- or N-Naphthyl-1,2,3-Triazole Core Structure Targeting Clostridioides (Clostridium) difficile: Synthesis, Antibacterial Evaluation, and an In Vivo C. difficile Infection Model

Clostridioides (also known as Clostridium) difficile is a Gram-positive anaerobic, spore producing bacterial pathogen that causes severe gastrointestinal infection in humans. The current chemotherapeutic options are inadequate, expensive, and limited, and thus inexpensive drug treatments for C. difficile infection (CDI) with improved efficacy and specificity are urgently needed. To improve the solubility of our cationic amphiphilic 1,1′-binaphthylpeptidomimetics developed earlier that showed promise in an in vivo murine CDI model we have synthesized related compounds with an N-arytriazole or N-naphthyltriazole moiety instead of the 1,1′-biphenyl or 1,1′-binaphthyl moiety. This modification was made to increase the polarity and thus water solubility of the overall peptidomimetics, while maintaining the aromatic character. The dicationic N-naphthyltriazole derivative 40 was identified as a C. difficile-selective antibacterial with MIC values of 8 µg/mL against C. difficile strains ATCC 700057 and 132 (both ribotype 027). This compound displayed increased water solubility and reduced hemolytic activity (32 µg/mL) in an in vitro hemolysis assay and reduced cytotoxicity (CC50 32 µg/mL against HEK293 cells) relative to lead compound 2. Compound 40 exhibited mild efficacy (with 80% survival observed after 24 h compared to the DMSO control of 40%) in an in vivo murine model of C. difficile infection by reducing the severity and slowing the onset of disease.


Introduction
Clostridioides (also known as Clostridium) difficile is a Gram-positive, anaerobic sporeforming bacterium that causes mild to serious infections in the gastrointestinal tract (GIT) due to the production of potent exotoxins (TcdA, TcdB, and CDT) that cause severe gastrointestinal damage [1][2][3]. The resilient endospores contaminate healthcare environments and facilitate disease initiation, dissemination, and re-infection. In the GIT, spores require glycine and cholate derivatives for germination. In a healthy GIT, the microbiota metabolizes cholate derivatives preventing germination of C. difficile spores. CDI occurs when the normal GIT microbiota is disrupted or killed by conventional broad-spectrum antimicrobials [1]. Under these conditions the metabolism of cholate is significantly compromised, facilitating the germination of spores into C. difficile vegetative cells [4,5].
CDI has a mortality rate of up to 8% [2] with the reoccurrence of infections occurring in up to 20% of cases treated with vancomycin or metronidazole [6]. A 2019 Antibiotic Resistance Threat Report from the US Centers for Disease Control and Prevention indicated that in the USA in 2017 an estimated 223,900 cases of CDI in hospitalized patients resulted in 12,800 deaths and $1 billion in attributed healthcare costs [7]. Thus, there is a significant and important incentive to develop novel therapeutics that show selectivity for C. difficile over other gut bacteria to effectively combat CDI. While fecal microbiota transplantation can be effective for recurrent CDI, there can be adverse effects and the long-term impacts are unknown [1, 2,8].
In our earlier work on the development of the cationic amphiphilic 1,1 -binaphthylpeptidomimetics, we established the pharmacophoric importance of a hydrophobic head group (e.g., a binaphthyl moiety) connected to a dicationic peptide in the development of broadspectrum antibacterial agents. This led to the identification of compound 1 with potent antibacterial activity against drug resistant Gram-positive bacteria with potential for topical applications (Figure 1) [25]. More recent work in our laboratory has identified compounds 2-4 from a class of small molecule cationic amphiphilic 1,1 -biarylpeptidomimetics that exert antibacterial activity through cytoplasmic membrane disruption [13,14]. These compounds have IC 50 values of 4-8 µg/mL against C. difficile (Figure 1). The efficacy of these compounds at treating CDI in an in vivo murine CDI model was assessed against vancomycin as a positive control with 10% DMSO as the negative control. Compound 2 appeared to protect the mice from disease at the 24 h point with a 50% survival rate (2/4 mice) vs. 0% survival in the 10% DMSO group; this was not statistically significant due to the small sample size. These results clearly showed that compound 2 exhibited a notable positive effect in the treatment of CDI. Unfortunately compound 3 showed poor solubility with precipitation during preparation in a 10% DMSO solution, and high in vitro hemolytic activity against HEK293 cells. While compound 4 showed promising in vitro properties, it performed poorly in the C. difficile murine model with a survival rate of 60% after 24 h, but a 0% rate after 48 h [13], despite its low hemolytic activity. Despite some positive results, more water-soluble derivatives with lower hemolytic activity for further in vivo murine CDI model studies needed to be developed. To achieve this aim, we replaced the hydrophobic binaphthyl group found in 2 and 3 with an N-arytriazole or N-naphthyltriazole moiety as shown in Figure 2. These modifications should retain the aromatic character of these molecules while inducing a better polarity profile and thereby increasing the water solubility of the overall peptidomimetics. It was not clear at the start what effect these modifications would have on the antibacterial activities of these newly proposed compounds or their specificity for C. difficile over other pathogenic bacteria. Herein, we disclose the results of this investigation.

Results and Discussion
Preparation of the target N-arytriazole or N-naphthyltriazole peptidomimetics required the synthesis of the carboxylic acid derivatives 5, 6, 17, 18, 27, and 28 based on scaffolds 1-4 ( Figure 2); the syntheses of acid 17 is described in the experimental section with the other acid syntheses described in the Supporting Information.
The synthesis of the new peptidomimetic derivatives is described in Schemes 1-3. In a typical example, derivative 40 (Scheme 3) was generated starting from acid 17 coupling with the protected azidodipetide 29 under standard peptide coupling conditions (EDCI/HOBt) [26,27] to give amide 32 in 67% yield. This was followed by a standard copper-catalyzed azide-alkyne cycloaddition reactions [28] with ethenylcyclohexane to give the corresponding 1,4-disubstituted 1,2,3-triazole product which was deprotected using TFA/CH2Cl2/H2O followed by treatment with ethereal HCl to yield the dicationic amphiphile 40 in 46% yield over two steps. The synthesis of the additional mono-and dicationic peptidomimetic amphiphiles 10-16, 21-26, and 36-50 followed an analogous strategy and is summarized in Schemes 1-3 with experimental and characterization details provided in the Supporting Information. MIC values against C. difficile in µg/mL. SR = solubility ratio relative to that of compound 1-see Ref [13].

Results and Discussion
Preparation of the target N-arytriazole or N-naphthyltriazole peptidomimetics required the synthesis of the carboxylic acid derivatives 5, 6, 17, 18, 27, and 28 based on scaffolds 1-4 ( Figure 2); the syntheses of acid 17 is described in the experimental section with the other acid syntheses described in the Supporting Information.
The synthesis of the new peptidomimetic derivatives is described in Schemes 1-3. In a typical example, derivative 40 (Scheme 3) was generated starting from acid 17 coupling with the protected azidodipetide 29 under standard peptide coupling conditions (EDCI/HOBt) [26,27] to give amide 32 in 67% yield. This was followed by a standard copper-catalyzed azide-alkyne cycloaddition reactions [28] with ethenylcyclohexane to give the corresponding 1,4-disubstituted 1,2,3-triazole product which was deprotected using TFA/CH2Cl2/H2O followed by treatment with ethereal HCl to yield the dicationic amphiphile 40 in 46% yield over two steps. The synthesis of the additional mono-and dicationic peptidomimetic amphiphiles 10-16, 21-26, and 36-50 followed an analogous strategy and is summarized in Schemes 1-3 with experimental and characterization details provided in the Supporting Information.

Results and Discussion
Preparation of the target N-arytriazole or N-naphthyltriazole peptidomimetics required the synthesis of the carboxylic acid derivatives 5, 6, 17, 18, 27, and 28 based on scaffolds 1-4 ( Figure 2); the syntheses of acid 17 is described in the experimental section with the other acid syntheses described in the Supporting Information.
The synthesis of the new peptidomimetic derivatives is described in Schemes 1-3. In a typical example, derivative 40 (Scheme 3) was generated starting from acid 17 coupling with the protected azidodipetide 29 under standard peptide coupling conditions (EDCI/HOBt) [26,27] to give amide 32 in 67% yield. This was followed by a standard copper-catalyzed azide-alkyne cycloaddition reactions [28] with ethenylcyclohexane to give the corresponding 1,4-disubstituted 1,2,3-triazole product which was deprotected using TFA/CH 2 Cl 2 /H 2 O followed by treatment with ethereal HCl to yield the dicationic amphiphile 40 in 46% yield over two steps. The synthesis of the additional mono-and  The N-arytriazole and N-naphthyltriazole peptidomimetics were subjected to antimicrobial screening. In the first instance, minimum inhibitory concentrations (MICs) were determined against a panel of Gram-positive (including two strains of C. difficile) and Gram-negative pathogenic bacteria with vancomycin and the commercially available peptide colistin as positive controls, respectively; the MICs are displayed in Table 1. The compounds were then tested against a second panel of Gram-positive and Gram-negative pathogenic bacteria and two fungi strains at the Community for Open Antimicrobial Drug Discovery (CO-ADD)-these results are reported in the Supporting Information (Table S1) [29]. A cytotoxicity concentration (CC50) assay was also performed by CO-ADD; the synthesized compounds were tested at concentrations ≤32 µg/mL on human embryonic kidney cells (HEK293 cells; ATCC CRL-1573) while hemolysis assays for lysis of human erythrocytes were also performed. Vancomycin, colistin, fluconazole, and tamoxifen were used as positive controls (see Table 1 for details). The CC50 and HC50 values are also shown in Table 1.
Preliminary screening revealed that compared to the previously synthesized compounds 1-4, the new N-naphthyltriazole dicationic derivatives 40 and 42 showed the best activities against the two C. difficile RT 027 strains, ATCC 700,057 and 132 with a similar activity of 8 µg/mL compared to compounds 1, 3, and 4. However, they were generally The N-arytriazole and N-naphthyltriazole peptidomimetics were subjected to antimicrobial screening. In the first instance, minimum inhibitory concentrations (MICs) were determined against a panel of Gram-positive (including two strains of C. difficile) and Gramnegative pathogenic bacteria with vancomycin and the commercially available peptide colistin as positive controls, respectively; the MICs are displayed in Table 1. The compounds were then tested against a second panel of Gram-positive and Gram-negative pathogenic bacteria and two fungi strains at the Community for Open Antimicrobial Drug Discovery (CO-ADD)-these results are reported in the Supporting Information (Table S1) [29]. A cytotoxicity concentration (CC 50 ) assay was also performed by CO-ADD; the synthesized compounds were tested at concentrations ≤32 µg/mL on human embryonic kidney cells (HEK293 cells; ATCC CRL-1573) while hemolysis assays for lysis of human erythrocytes were also performed. Vancomycin, colistin, fluconazole, and tamoxifen were used as positive controls (see Table 1 for details). The CC 50 and HC 50 values are also shown in Table 1. Preliminary screening revealed that compared to the previously synthesized compounds 1-4, the new N-naphthyltriazole dicationic derivatives 40 and 42 showed the best activities against the two C. difficile RT 027 strains, ATCC 700,057 and 132 with a similar activity of 8 µg/mL compared to compounds 1, 3, and 4. However, they were generally less active against the other Gram-positive and Gram-negative bacteria ( Table 1). The relative solubility ratios (relative to compound 1) [13] for 40 and 42 were 5 and 4 with CLogP values of 4.46 and 4.39, respectively, when compared to 1 with a ClogP of 7.47. Therefore, despite the better solubility profiles of these compounds, they failed to show better activity against C. difficile. However, the increased solubility (enhanced polarity) of derivatives 40-42 could be a factor in the reduced activities against the other bacteria, when compared to compounds 1-4 (see Table 2). None of the other derivatives synthesized in this study showed appreciable activity against C. difficile with MIC values ranging from 32 to 128 µg/mL (Table 1). Importantly, the remaining anti-bacterial results were generally poor, however for these specific derivatives, these reduced activities could indicate reduced capacity to interfere with normal GIT microbiota ( Table 2). Compounds 40 and 42 showed a slight reduction in cytotoxicity against HEK293 cells compared to compounds 2 and 4. The hemolytic activity of these compounds was 32 µg/mL against human erythrocytes, 2-fold more than their IC 50 values against C. difficile.

In Vivo Assay: Murine Model of CDI
Compound 40 was selected for further evaluation as an effective treatment for C. difficile using a murine model of CDI study because of its sustained antimicrobial potency against C. difficile and its better water solubility profile. The results from these studies are summarized in Figure 3.  Analysis of the anti-bacterial activities against other bacterial species indicated that the monocationic naphthyltriazole derivatives 21-26 showed appreciable activity against Staphylococcus aureus (including an MRSA strain) with MIC values between 4 and 8 µg/mL (Table 1). Additionally, compound 21 had notable MIC values of 4 µg/mL against Enterococcus faecalis and Streptococcus pneumoniae. An overview of activity shown in Table 1 showed "pockets" of activities focused on the naphthyl-based derivatives (21-26 and 40-45, columns 1-4), with the monocationic examples (21-26) producing better outcomes against the Gram positive strains. The second screening results (Table S1, Supporting Information) were consistent with these results with analogous trends in activity against an additional S. aureus strain.

In Vivo Assay: Murine Model of CDI
Compound 40 was selected for further evaluation as an effective treatment for C. difficile using a murine model of CDI study because of its sustained antimicrobial potency against C. difficile and its better water solubility profile. The results from these studies are summarized in Figure 3.  The mice treated with compound 40 (red) showed delayed disease onset compared to mice treated with DMSO (blue; Figure 3), although they still succumbed to infection by day 2. Notably, at day 1 post-infection, mice treated with compound 40 showed 40% greater survival compared with mice treated with DMSO (Figure 3a), although there was no effect on mouse weight (Figure 3b), or spore numbers shed in the feces of these animals (Figure 3c), suggesting that compound 40 was not impacting C. difficile colonization. Furthermore, on day 1 post-infection, treatment with compound 40 resulted in a lower overall cage appearance score when compared to DMSO (Figure 3d), which suggested that this compound was delaying diarrheal onset although there was no significant difference in individual fecal score (Figure 3e) or physiological appearance score (Figure 3f) detected between the two groups of mice ( Figure 3e). Thus, collectively these data suggest that compound 40 may reduce the severity of disease caused by C. difficile.

Materials and Methods
Synthetic methods and general characterization and analysis were as described previously. [13] Notes and other considerations. Known reagents that were not available commercially were prepared as reported using known methods and is detailed in the Supporting Information. [14,[32][33][34][35]

General Procedure I: Alkylation of phenols (with ethyl bromoacetate)
A solution of the phenol (1 eq) in dry DMF (5 mL/mmol substrate) was stirred during the addition of K2CO3 (3 eq). Ethyl bromoacetate (1.3 eq) was added at room temperature and stirring was continued at rt for 12 h, before being diluted with EtOAc (2 × 50 mL). The resulting mixture was washed with water (2 × 50 mL), brine (2 × 50 mL), dried (MgSO4), filtered, and concentrated under vacuum. The residue was subjected to silica gel flash column chromatography to afford the desired ester product.
General Procedure II: Ester hydrolysis A solution of the ester (1 eq) in ethanol (10 mL/mmol substrate) was stirred followed by the addition of 7% KOH solution (5 mL/mmol) at rt. The mixture was stirred at rt for 2 The mice treated with compound 40 (red) showed delayed disease onset compared to mice treated with DMSO (blue; Figure 3), although they still succumbed to infection by day 2. Notably, at day 1 post-infection, mice treated with compound 40 showed 40% greater survival compared with mice treated with DMSO (Figure 3a), although there was no effect on mouse weight (Figure 3b), or spore numbers shed in the feces of these animals (Figure 3c), suggesting that compound 40 was not impacting C. difficile colonization. Furthermore, on day 1 post-infection, treatment with compound 40 resulted in a lower overall cage appearance score when compared to DMSO (Figure 3d), which suggested that this compound was delaying diarrheal onset although there was no significant difference in individual fecal score (Figure 3e) or physiological appearance score (Figure 3f) detected between the two groups of mice ( Figure 3e). Thus, collectively these data suggest that compound 40 may reduce the severity of disease caused by C. difficile.

Materials and Methods
Synthetic methods and general characterization and analysis were as described previously [13].
Notes and other considerations. Known reagents that were not available commercially were prepared as reported using known methods and is detailed in the Supporting Information, [14,[32][33][34][35].

General Procedure I: Alkylation of Phenols (with Ethyl Bromoacetate)
A solution of the phenol (1 eq) in dry DMF (5 mL/mmol substrate) was stirred during the addition of K 2 CO 3 (3 eq). Ethyl bromoacetate (1.3 eq) was added at room temperature and stirring was continued at rt for 12 h, before being diluted with EtOAc (2 × 50 mL). The resulting mixture was washed with water (2 × 50 mL), brine (2 × 50 mL), dried (MgSO 4 ), filtered, and concentrated under vacuum. The residue was subjected to silica gel flash column chromatography to afford the desired ester product.

General Procedure II: Ester Hydrolysis
A solution of the ester (1 eq) in ethanol (10 mL/mmol substrate) was stirred followed by the addition of 7% KOH solution (5 mL/mmol) at rt. The mixture was stirred at rt for To a solution of the N-protected amine (1.0 eq) in CH 2 Cl 2 (30 mL/mmol substrate) (if the substrate contained an N-Pbf moiety, H 2 O (20.0 eq) was added to the solution) was added TFA (30.0 mL/mmol substrate) and then stirred at rt overnight (>16 h). The solvent was removed and the resulting residue dissolved in CH 2 Cl 2 (30 mL/mmol substrate). Excess anhydrous HCl (2.0 M in Et 2 O, 15 mL/mmol substrate, 30.0 eq) was added and the solvent was then removed. The residue was then dissolved in a minimal volume of CH 2 Cl 2 (or MeOH) and excess Et 2 O (25 mL for ≤0.1 mmol substrate) was added, resulting in a precipitate of the hydrochloride salt of the amine. The reaction mixture was filtered; the resulting filtrate collected, concentrated, triturated with Et 2 O (3 × 20 mL); and the solids then dissolved in MeOH. The solution was concentrated and dried in vacuo to yield the mono or di-hydrochloride salt as a thin, translucent film that usually required scratching with a spatula, producing a fine hygroscopic powder or amorphous gum. General Procedure III: Amide coupling A mixture of the amine (1.0 eq), carboxylic acid (1.0 eq), EDC.HCl (1.2 eq) eq), and TEA (1 eq) in dichloromethane/acetonitrile solution (10 mL/mmol a stirred at rt for the specified time. The mixture was concentrated (if >5.0 m methane/acetonitrile), and then the resulting residue dissolved in EtOAc (25 m tions that contained ≤1.0 mmol amine or 25 mL/mmol amine for larger scal and washed with aqueous HCl (1.0 M-2 × 25 mL), saturated aqueous NaHCO3 and brine (1 × 25 mL). The organic solution was dried (MgSO4), filtered, conce subjected to further purification via flash chromatography (if required) to furn geted amide product.

General Procedure VII: Amine deprotection (N-Boc and/or N-Pbf remo
To a solution of the N-protected amine (1.0 eq) in CH2Cl2 (30 mL/mmol s the substrate contained an N-Pbf moiety, H2O (20.0 eq) was added to the so added TFA (30.0 mL/mmol substrate) and then stirred at rt overnight (>16 h). was removed and the resulting residue dissolved in CH2Cl2 (30 mL/mmol sub cess anhydrous HCl (2.0 M in Et2O, 15 mL/mmol substrate, 30.0 eq) was add solvent was then removed. The residue was then dissolved in a minimal volum (or MeOH) and excess Et2O (25 mL for ≤0.1 mmol substrate) was added, re precipitate of the hydrochloride salt of the amine. The reaction mixture was resulting filtrate collected, concentrated, triturated with Et2O (3 × 20 mL); and then dissolved in MeOH. The solution was concentrated and dried in vacuo mono or di-hydrochloride salt as a thin, translucent film that usually required with a spatula, producing a fine hygroscopic powder or amorphous gum.

In Vivo Murine Model of CDI Treatment
Disease Treatment Model. These experiments were performed as previously described [39][40][41][42]. Mice were humanely killed at the onset of severe disease or at the end of the experiment (day 4), as previously described [43].
Statistical Analysis. Statistical analysis was performed using Prism 7 (GraphPad Software). The Kaplan-Meier survival curves were assessed using a log-rank (Mantel-Cox) test. Weight loss, spore shedding, fecal consistency, and physiological appearance data were analyzed by one-way ANOVA with a post hoc Tukey's multiple comparison test. Differences in data values were considered significant at a p value of <0.05.

Conclusions
This study reported the next generation of hydrophobic anchored cationic peptidomimetics as antibacterial agents, with a focus on targeting CDI. A major aim was to improve the solubility profile of these compounds to allow for sufficient solubility for efficient administration of the drug while maintaining gut availability and antibacterial activity. The naphthyltriazole derivates containing either a monocationic or dicationic amino acid side chain were generally the most effective, with compounds 40 and 42, possessing terminal cyclohexyl and benzyl moieties, respectively, exhibiting MIC values of 8 µg/mL. Naphthyltriazole 40 was selected for an in vivo murine model trials of CDI but exhibited only mild evidence of in vivo efficacy indicating that further investigation into the structural and biological parameters affecting the in vivo efficacy of these antibacterial peptidomimetics is required, as the observed in vitro efficacy did not translate directly into in vivo efficacy. We have already reported that a correlation exists between increased hemolytic activity and an increase in hydrophobic/cationic ratio [15]; unfortunately, compound 40 exhibited a slight increase in hemolytic activity relative to the majority of tested compounds in this class with an HC 50 value of 32 µg/mL. While the selectivity ratio could be more substantial, this is acceptable for the future development of these gastrointestinal focused compounds. We have previously reported a comparative solubility assay for this class of antimicrobial agents with increasing numerical values corresponding to better aqueous solubility relative to compound 1 (which possesses a value of 1) [13]. Compound 40 showed a better solubility ratio with an assay value of 5, relative to our lead compound 2 with a value 3-this is also reflected in the CLogP values of 4.46 and 5.76 for 41 vs. 2, respectively. These outcomes were confirmed with no issues during the mouse model trials with sufficient solubility in the dosage regimen. Variations on the triazole and O-naphthyl substituents could be made in future studies with the view of enhancing antibacterial activity against C. difficile.