Polyisocyanide Quaternary Ammonium Salts with Exceptionally Star-Shaped Structure for Enhanced Antibacterial Properties.

The development of non-polluting and non-hazardous polymeric antimicrobial agents has become a hot issue in current research and development. Among them, polymer quaternary ammonium salts are thought to be one of the most promising materials for antibacterial efficacy. Here, we present an efficient strategy for synthesizing polyisocyanide quaternary ammonium salts (PQASs) with a novel star-shaped structure. Benefitting from the novel structure, increased cation density and enhanced water solubility, the prepared star polyisocyanide quaternary ammonium salts (S-PQASs) exhibit excellent antibacterial properties against Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus). In particular, S-POcQAS-M50 (where M stands for isonitrile monomer and 50 stands for the initial feeding ratio) showed the best antimicrobial activity with minimum inhibitory concentration (MIC) of 17 and 20 µg/mL against E. coli and S. aureus, respectively. It was also found that the unique star-shaped structure can give QASs with improved antimicrobial performance compared with our previously prepared linear quaternary ammonium salts (L-PQASs). These results demonstrated that the antibacterial activity of QASs is closely related to its structure. This work provides an idea for the design of efficient polymeric antimicrobial agents.

1. Introduction

Due to the irrational use of antibiotic drugs, the increase in drug-resistant bacteria and viruses has led to an increasingly hostile environment for human survival [1,2,3,4]. The research and development of organic polymeric antibacterial compounds are recognized as one of the most effective strategies to alleviate the environmental issues caused by the increasing abuse of antibiotics [5,6,7,8], which is ascribed to its excellent processing properties, good stability and non-volatility [9,10,11]. Therefore, the design of an organic polymer with excellent antimicrobial properties using simple methods has received interest in the research community.

In the past decades, various organic polymeric antibacterial agents have been developed, such as polymer quaternary ammonium salts (PQASs) [12,13,14], polymer quaternary phosphonium salts (PQPSs) [15,16], chitosan [17,18,19,20], polymer guanidine [21], etc. Among them, PQASs has attracted great attention owing to its fast sterilization performance, excellent structure stability, and low cost [22,23,24]. To date, numerous studies have reported that various factors, including the structure and molecular weight of the polymer, structure of quaternary ammonium groups, water-solubility and biocompatibility of the polymer, can significantly affect the antimicrobial activity of PQASs. For example, Zhang et al. [25] have developed six β-pinene quaternary ammonium salts and found that compound 4a showed the best antibacterial action, suggesting that the antibacterial activity of quaternary ammonium salts is closely related to their structure. Dong et al. synthesized phosphonate/quaternary ammonium copolymers exhibiting remarkable antibacterial activity, and they found that the bactericidal ability of the copolymer increased with the increase in cationic content in the copolymer [26]. Brittany et al. found that the bactericidal activity of quaternary ammonium-modified poly(amidoamine) dendrimers was dependent on dendrimer generation [27]. However, despite the considerable endeavors that have been devoted, the current PQASs still have the following several limitations: (1) the poor stability of monomers; (2) the low quaternization of PQASs; and (3) the poor water-solubility of PQASs. Therefore, exploring novel PQASs that can overcome the above disadvantages is of great importance.

In our previous work, a series of L-PQASs have been reported for highly efficient antibacterial and antitumor applications [28]. In recent years, branched polymers with star-shaped, dendritic and hyperdendritic structures have attracted attention as potential antimicrobial agents. Many studies in the literature have reported that the antimicrobial activity of the dendrimers and the hyperbranched quaternary ammonium compounds was higher than the linear polymers [29,30]. For example, Lia A. T. W. Asri et al. found that hyperbranched Si-HB-PEI+ coatings exhibited high contact-killing efficacies against S. aureus. Furthermore, the hyperbranched Si-HB-PEI+ coatings (10 wt %) did not negatively affect the adhesion and proliferation of fibroblasts [31]. Yudan Wang et al. developed a novel crosslinked hyperbranched cationic polymer membranes (cHCPMs) and found that the membranes showed excellent broad-spectrum antibacterial and antibacterial adhesion properties against E. coli and S. aureus [32]. Chen et al. functionalized poly(propylene imine) and poly(amidoamine) dendrimers with dimethyl dodecyl ammonium groups and found that dendrimers macromolecules containing 16 QAS groups exhibited two orders of magnitude higher bactericidal efficiency against Gram-negative bacteria than those with mono-functional counterparts [33,34]. It is well known that polymers with dendritic structures possess many unique properties, such as high solubility, low viscosity, and large numbers of functional groups [35]. Hence, the development of hyperbranched quaternary ammonium salt antibacterial agents has become a hot focus of research.

In this work, we further optimized the experimental conditions and successfully synthesized S-PQASs with novel star-shaped hyperbranched structures by using polyisocyanide as the backbone. Due to the sequential insertion of isonitrile groups in the polyisocyanide backbone, a tightly structured carbon chain is formed in the polymer with a small spacing between the side chains, which enables effective synergistic effects between the functional groups on the side chains. Notably, the prepared S-POcQASs exhibited advanced antibacterial activity against both E. coli and S. aureus in comparison with our previously prepared L-POcQASs. The results of the experimental demonstrate that the improvement of antibacterial activity is mainly ascribed to the optimization of the structure of PQASs.

2. Materials and Methods

2.1. Materials

Tetrakis(triphenylphosphine)palladium(0) (Pd(PEt3)4) (CAS: 14221-01-3), copper iodide (CuI) (CAS: 7681-65-4), 1,3,5-tribromobenzene (C6H3Br3) (CAS: 626-39-1), triphenylphosphine (PPh3) (CAS: 603-35-0), (trimethylsilyl)acetylene (C5H10Si) (CAS: 1066-54-2), dichloromethane (DCM) (CAS: 75-09-2), anhydrous magnesium sulfate (MgSO4) (CAS: 139939-75-6), sodium bicarbonate (NaHCO3) (CAS: 144-55-8), petroleum ether (PE) (CAS: 8032-32-4), ethyl acetate (EA) (CAS: 141-78-6), tetrabutylammonium fluoride (TBAF), solution in THF (1M) (CAS: 429-41-4), N-hexane (C6H14) (CAS: 110-54-3), dichlorobis(triethylphosphine)-palladium(II) (Pd(PEt3)2Cl2) (CAS: 28425-04-9), cuprous chloride (CuCl) (CAS: 7758-89-6), triethylamine (NEt3) (CAS: 121-44-8), and N,N-dimethyloctylamine (C10H23N) (CAS: 7378-99-6) were purchased from Sigma-Aldrich (St. Louis, MO, USA). All other raw materials used for the preparation of monomers and S-POcQASs were the same as those used for our previous preparation of L-POcQASs. E. coli (ATCC25922) and S. aureus (ATCC25923) were obtained from the first affiliated hospital of Harbin Medical University (Harbin, China).

2.2. Synthesis of Palladium Catalyst with Star-Shaped Structure

The synthesis of a palladium catalyst with a star-shaped structure was carried out, and the reaction process is shown in Scheme 1.

Scheme 1

Chemical structure and synthetic procedures of palladium catalyst.

2.2.1. Synthesis of Intermediate a

The synthesis of intermediate a was conducted via a classical Sonogashria coupling reaction. First, 3.14 g (10.01 mmol) of 1,3,5-tribromobenzene, 0.69 g (0.60 mmol) of Pd(PEt3)4, 0.16 g (0.60 mmol) of PPh3, and 0.12 g (0.60 mmol) of CuI were added to a 100 mL two-necked flask containing 50 mL of triethylamine. Afterward, 5.1 mL (36.10 mmol) of (trimethylsilyl)acetylene was added to the reactor under N2 conditions, and the reaction was continued for 12 h under constant stirring. When the substrate was completely consumed using TLC analysis, the mixture was cooled to room temperature, and the filtered liquid was spin-dried under reduced pressure. The solid was washed with distilled water (2 × 30 mL) and saturated NaHCO3 solution (2 × 30 mL). Subsequently, it was dried over anhydrous MgSO4, filtered, and the filtered liquid was spin-dried under reduced pressure. Finally, the pale yellow intermediate a (2.09 g, yield 57.2%) was obtained by column chromatography (silica gel, PE:EA = 10:1 v/v) for further purification.

2.2.2. Synthesis of Intermediate b

First of all, the above-prepared intermediate a (1.83 g, 5.00 mmol) was dissolved in a two-necked flask containing dry THF (30 mL) under N2 conditions. Then, 5 mL of THF solution of tetrabutylammonium fluoride (TBAF) (1M) was added dropwise to the above solution under ice water bath conditions. After continuous stirring for 3 h at room temperature, the intermediate a was judged to be completely consumed by TLC analysis. Subsequently, the reaction was quenched by the immediate addition of deionized water. The organic phase was washed with deionized water, and the solvent was removed in a rotary evaporator to obtain the crude product. Finally, the crude product was recrystallized by N-hexane to obtain a needle-shaped crystal of intermediate b (0.47 g, yield 62.8%).

2.2.3. Synthesis of Palladium Catalyst

The palladium catalyst with a star-shaped structure was prepared according to the literature [36]. In detail, 0.08 g (0.55 mmol) of the above-prepared intermediate b, 0.75 g (1.80 mmol) of Pd(PEt3)2Cl2, and trace CuCl were added to a 50 mL two-neck flask. Then, 25.0 mL of CH2Cl2 and 5.0 mL of NEt3 were mixed into the solution under dark and N2 condition. After continuous stirring for 8 h, the intermediate b was judged to be completely consumed by TLC analysis. Subsequently, the solvent was removed, and column chromatography (silica gel, PE:EA = 7:1 v/v) was used to purify the solid. Finally, the crude product was recrystallized from DCM/N-hexane as a white crystal (0.24 g, yield 37.1%).

2.3. Synthesis of Isonitrile Monomer

The synthesis procedure and detailed characterization of the isonitrile monomer was reported in our previously published work, as shown in Scheme 2.

Scheme 2

Chemical structure and synthetic procedures of isonitrile monomer.

2.4. Synthesis of S-POcQASs

The S-POcQASs were synthesized by a process similar to L-POcQASs that was described in our previous work [28], except that the linear palladium catalyst was replaced by a star-shaped palladium catalyst. The synthesis route of targeted products is shown in Scheme 3.

Scheme 3

Chemical structure and synthetic procedures of S-POcQASs.

2.5. Antimicrobial Activity Test of S-POcQASs

The antimicrobial activity was evaluated by determining the minimum inhibitory concentrations (MICs) of the S-POcQASs against S. aureus and E. coli using the TTC colorimetric method. The inoculated and incubated E. coli and S. aureus cells were diluted to 106~107 CFU/mL with LB broth. Then, 100 µL of bacteria was incubated in each well of a 96-well plate and cultured in a 100 μL growth medium. The S-POcQASs were diluted with sterile water to a concentration of 20 mg/mL, and it was added to the 96-well plate and adjusted to the required concentration (2.5~5000 μg/mL). The negative control group consisted of the nutritional broth, and the positive control sample was made up without S-POcQASs. After that, the agar plates were incubated in an incubator at 37 °C for 24 h. Finally, the antibacterial activity evaluation was carried out three times under the same experimental conditions.

2.6. Characterizations

1H NMR spectra and 13C NMR spectra of the samples were recorded by a Bruker Avance III-500 (Bruker Co., Ltd., Fällanden, Switzerland) using DMSO-d6 and CDCl3 as the solvent. The chemical structures of the samples were verified by Fourier transform infrared (FTIR) spectra (Spectrum 100, PerkinElmer, Waltham, MA, USA).

3. Results

3.1. Characterization of Polymers and Compounds

The structures of the polymers and compounds were verified by 1H NMR spectroscopy, 13C NMR spectroscopy and gel permeation chromatography (GPC). The number-averaged molecular masses (M) and polydispersity indices (PDI = M/M) of these polymers are determined from GPC analysis and listed in Table 1. It was found that the molecular weight of these polyisocyanides increases gradually as the ratio of [Monomer]0/[Catalyst]0 increased from 50 to 150. In addition, all polymers exhibited a relatively high molecular weight and a relatively low polydispersity index (PDI ≈ 1.1). These results indicate that the polyisocyanides with different molecular weights can be polymerized in a controlled manner by initiating the polymerization of isocyanide monomers through palladium catalysts. It is worth noting that the molecular weight of S-P-Mn is nearly three times that of L-P-Mn when the [Monomer]0/[Catalyst]0 is the same, which can be ascribed to construction of a palladium catalyst with a star-shaped structure.

Table 1

Polymerization results of PQASs with palladium catalysts as initiator at 55 °C.

Entry	M/C a	Solvent	Polymer	Mn (g/mol) b	Mw/Mnc	Yield
1	50	THF	L-P-M50	10,125	1.12	88.2%
2	100	THF	L-P-M100	21,310	1.09	92.1%
3	150	THF	L-P-M150	30,051	1.13	79.4%
4	50	THF	S-P-M50	29,775	1.16	88.4%
5	100	THF	S-P-M100	59,611	1.11	94.6%

M/C = monomer to palladium catalyst. Determined by gel permeation chromatography (GPC) relative to polystyrene standards. Polydispersity index calculated by GPC.

Based on the above analysis, the catalyst plays a crucial role in the controllable preparation of polymers. Therefore, 1H NMR and 13C NMR were performed to confirm the structure of intermediates and the palladium catalyst. The 1H NMR spectrum of intermediate a is shown in Figure 1. The signal peak at δ = 7.59 ppm corresponded to unsubstituted hydrogen protons on the benzene ring, while the peak at δ = 0.19 ppm corresponded to hydrogen protons of –Si(CH3)3 groups, confirming the successful synthesis of intermediate a.

Figure 1

1H NMR spectrum of intermediate a.

Figure 2 showed the 1H NMR spectrum of intermediate b. The signal at δ = 7.66 ppm was attributed to unsubstituted hydrogen protons on the benzene ring, while the peak at δ = 3.13 ppm was assigned to hydrogen protons of –C≡C–H groups, which confirmed that intermediate b was synthesized successfully.

Figure 2

1H NMR spectrum of intermediate b.

Figure 3 shows the 1H NMR and 13C NMR spectra of the palladium catalyst. As shown in Figure 3a, the signal at δ = 6.98 ppm was attributed to hydrogen protons on the benzene ring, while characteristic peaks at 1.84–1.95 and 1.10–1.16 ppm were attributed to the protons of –CH2– and –CH3 of –PEt3, respectively, indicating the successful synthesis of the palladium catalyst. The structure and component of the palladium catalyst was further verified by the 13C NMR spectrum (Figure 3b). The peaks at δ = 130.79 ppm and δ = 128.61 ppm were the characteristic peaks of unsubstituted carbon atoms and carbon atoms directly connected to the alkyne group in the benzene ring block, respectively. The peak at δ = 104.22 ppm represented the carbon atom peak of the alkyne group near the benzene ring. The signal at δ = 96.64 ppm was attributed to carbon atoms coordinated with palladium on the alkyne group. The peaks at δ = 15.44 ppm and δ = 8.33 ppm are the characteristic peaks of carbon atoms of –CH2– and –CH3 of –PEt3, respectively. These results demonstrate that the palladium catalyst with a star-shaped structure was constructed by introducing –Pd(PEt3)2Cl groups into intermediate b.

Figure 3

(a) 1H NMR and (b) 13C NMR spectra of palladium catalyst.

Figure 4 shows the 1H NMR and FT-IR spectra of the isonitrile monomer. As shown in Figure 4a, δ = 8.06–8.08 ppm are the peaks of hydrogen protons on the benzene ring near the ester group, while δ = 7.44–7.46 ppm are the peaks of the other hydrogen protons on the benzene ring. Meanwhile, δ = 4.34–4.36 ppm at position c are the peaks of hydrogen protons in the methylene group connected to the ester group, δ = 3.59–3.62 ppm at position f are the peaks of hydrogen protons in the methylene group connected to the chlorine atom, and δ = 1.94–1.96 ppm at position d and position e are the peaks of the other hydrogen protons on 4-chloro-1-butyl group. In addition, FT-IR spectrometry was used to further characterize the structure of isonitrile monomer (Figure 4b). The band at 2125 cm−1 represents the C≡N stretching vibration. Furthermore, Figure 4b exhibited the characteristic bands corresponding to C=O stretching (1723 cm−1), C=C aromatic stretching (1607 cm−1), methylene C-H bending (1275 cm−1), and ester group C-O stretching (1119 cm−1). These results showed that the isonitrile monomer had been successfully obtained.

Figure 4

(a) 1H NMR and (b) FT-IR spectra of isonitrile monomer (The + in the graph represents the peak position).

In our previous work, L-POcQAS-Mn was successfully obtained by a simple quaternization reaction using N, N-dimethyloctylamine as the alkyl halides [28]. In this work, the S-POcQAS-Mn was synthesized by a process that was similar to L-POcQASs, except that the linear palladium catalyst was replaced by a star-shaped palladium catalyst. The molecular structure of the S-POcQAS-M50 sample was investigated by 1H NMR and FT-IR spectra. As shown in Figure 5a, the broad signal at δ = 3.16 ppm was attributed to hydrogen protons of –N+(CH3)2–, while characteristic peaks at 1.26 and 0.86 ppm were attributed to the protons of –CH2– and –CH3 of –(CH2)7CH3, respectively. Similar signal peaks could be obtained in the L-POcQAS-M50 sample, indicating that the S-POcQAS-M50 had been successfully obtained. In the FT-IR spectrum presented in Figure 5b, the absorption band located at 2883 cm−1 originated from the C-H stretching vibration in the introduced quaternary ammonium groups. Furthermore, the absorption peak of C-Cl stretching vibration at 645 cm−1 can hardly be observed, indicating that the quaternization of polyisocyanine was successfully achieved by using N, N-dimethyloctylamine, which matches well with the 1H NMR results.

Figure 5

(a) 1H NMR and (b) FT-IR spectra of S-POcQAS-M50.

3.2. Solubility and Antibacterial Properties

Since the antimicrobial properties of antimicrobial agents are mainly studied in aqueous solutions, the antimicrobial activity of QAS is suggested to be more related to its solubility. It is well known that polyisocyanide is insoluble in water. However, its solubility can be changed by modification of its side chain using tertiary amines. Our previous study found that L-POcQASs with n-octyl side chains had the best water solubility, suggesting that the optimal length of the alkyl side chain was 8. In this work, we also found that both S-POcQAS-M50 and S-POcQAS-M100 with n-octyl side chains exhibited high water solubility (Table 2). On the one hand, S-POcQAS has a branched structure, and on the other hand, the star-shaped structure of S-POcQAS increases the density of hydrophilic cations.

Table 2

Water solubility of L-POcQASs and S-POcQASs.

Entry	1	2	3	4
Compound	L-POcQAS-M50	L-POcQAS-M100	S-POcQAS-M50	S-POcQAS-M100
Solubility	14 mg/mL	11 mg/mL	13 mg/mL	9 mg/mL

The antibacterial performance of the L-POcQAS-M50, L-POcQAS-M100, S-POcQAS-M50, and S-POcQAS-M100 were evaluated by using the minimum inhibitory concentrations (MICs) method. As shown in Figure 6, L-POcQASs demonstrated an apparent antibacterial effect, but it was obviously lower than that of S-POcQASs. Both S-POcQAS-M50 and S-POcQAS-M100 exhibited excellent antibacterial activity with MICs of 17 µg/mL and 19 μg/mL against E. coli, and 20 µg/mL and 17 μg/mL against S. aureus, respectively. However, the MICs of L-POcQAS-M50 and L-POcQAS-M100 were 27 µg/mL and 33 µg/mL on E. coli, and 32 µg/mL and 43 µg/mL on S. aureus, respectively. Remarkably, the improvement in antibacterial activity may be ascribed to their structural differences. It is well known that the sterilization mechanism of polymerized QAS antimicrobial materials is mostly contact-killing antibacterial. In other words, when the water-soluble QASs come into contact with the bacterial solution, the alkyl side chains of QASs can puncture the cell wall of bacteria, and then, the QAS groups with positive charge can be adsorbed onto the bacterial cell membrane with negative charge to cause the destruction of the cell membrane and the leakage of the inner cell contents, which will lead to the death of the bacteria. Generally, polymers with a branched structure have better water solubility than linear polymers. Furthermore, the unique star-shaped structure will increase the density of hydrophilic cations of POcQASs compared to the linear structure. Thus, as a result of the aforementioned factors, S-POcQASs exhibited approximately 1.5 times higher antibacterial activity than those of L-POcQASs. This work provides an effective and promising method for designing and preparing QAS antimicrobial materials.

Figure 6

Minimum inhibitory concentrations (MICs) (μg/mL) of synthesized L-POcQASs and S-POcQASs against S. aureus and E. coli.

4. Conclusions

In summary, two water-soluble antibacterials, L-POcQASs and S-POcQASs, have been synthesized by initiating the polymerization of isonitrile monomers with linear and star-shaped palladium catalysts, respectively, which is followed by quaternization by N, N-dimethyloctylamine. All the 1H NMR, FT-IR, and GPC results indicated that S-POcQASs were synthesized successfully. In addition, the synthesized S-POcQASs have high water solubility, and they exhibited higher antibacterial activities than L-POcQASs. Compared with L-POcQASs, S-POcQASs with hyperbranched structure possess more peripheral alkyl chains, more terminal cations and higher positive charge density, which facilitates the attraction and lysis of bacteria. Notably, the MIC of S-POcQAS-M50 to Gram-negative E. coli and Gram-positive S. aureus was as low as 17 µg/mL and 20 µg/mL, respectively, which suggests that S-POcQAS-M50 is a potential candidate as a kind of broad-spectrum antimicrobial agent.