Self-Aggregation, Antimicrobial Activity and Cytotoxicity of Ester-Bonded Gemini Quaternary Ammonium Salts: The Role of the Spacer

Abstract

Five ester-bonded gemini quaternary ammonium surfactants C₁₂-E_n-C₁₂ (n = 2, 4, 6), with a flexible spacer group, and C₁₂-B_m-C₁₂ (m = 1, 2), with rigid benzene spacers, were synthesized via a two-step reaction and analyzed. Furthermore, the effects of the spacer structure, spacer length and polymerization degree on the self-aggregation, antimicrobial activity and cytotoxicity of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ and their corresponding monomer N-dodecyl-N,N,N-trimethyl ammonium chloride DTAC were investigated. The results showed that C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ had markedly lower critical micellar concentration (CMC) values and lower surface tension than DTAC. Moreover, the CMC values of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ decreased with increasing spacer length. In the case of equivalent chain length, the rigidity and steric hindrance of phenylene and 1,4-benzenediyl resulted in larger CMC values for C₁₂-B_m-C₁₂ than for C₁₂-E_n-C₁₂. The antibacterial ability of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ was assessed using Escherichia coli (E. coli) and Staphylococcus albus (S. aureus) based on minimum inhibitory concentrations (MICs). Furthermore, C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ exhibited higher antimicrobial activity than DTAC and had stronger function toward S. aureus than E. coli. The antimicrobial activity was enhanced by increasing the spacer chain length and decreased with the increased rigidity of the spacers. The cytotoxic effects of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ in cultured Hela cells were evaluated by the standard CCK8 method based on half-maximal inhibitory concentration (IC₅₀). The cytotoxicity of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ was significantly lower than alkanediyl-α,ω-bis(dimethyldodecylammonium) bromide surfactants and DTAC. The spacer structure and the spacer length could induce significant cytotoxic effects on Hela cells. These findings indicate that the five ester-bonded GQASs have stronger antibacterial activity and lower toxicity profile, and thus can be used in the pharmaceutical industry.

1. Introduction

Gemini quaternary ammonium surfactants (GQASs) contain two hydrophobic alkyl tails and two hydrophilic quaternary ammonium cations connected by a spacer at or near the head groups [1]. Many researchers have designed and synthesized GQASs with various structures by altering the spacer structure, spacer length, hydrophobic chain length and symmetry of the amphiphile molecule [2,3,4,5]. GQASs can form micelles and efficiently reduce surface tension. In addition, GQASs have better wetting ability as well as antimicrobial and anticorrosive properties as compared to their corresponding conventional single-chain surfactant counterparts [6]. Therefore, GQASs can be used in the biomedicine industry as antibacterial agents [7], corrosion inhibitors [8], drug carriers [9] and genetic vectors [10]. However, most GQASs are stable compounds with poor microbial and chemical degradability [11], posing risks to microorganisms in water and soil. GQASs can accumulate in soil and sediment and leak into underground water and wastewater, threatening human health and the Earth’s ecosystems [12,13]. Thus, the utilization of non-degradable GQASs has raised serious aquatic toxicity and environmental safety concerns, which has limited their application. As a result, some GQASs have been designed via different strategies to decrease environment pollution [13,14,15]. For example, biodegradable groups or cleavable bonds can be introduced in the GQAS structure to improve the environmental acceptability [15,16,17]. Enzymatic reaction/hydrolysis occurs in the presence of ester functional groups. Therefore, induction of ester groups in GQAS molecules may effectively improve their biodegradability and significantly reduce aquatic toxicity [18]. GQAS with ester groups have the mildness of surfactants containing ester groups and the high efficiency of quaternary ammonium surfactants [19]. Nazish et al. synthesized ester-bonded GQAS m-E₂-m containing diester groups in the spacer tail and found that the cytotoxicity of the GQAS on 3T3-L1 fibroblast cells decreases with increasing alkyl carbon number [20]. GQAS with ester linkages also have special aggregation behavior because of the formation of intramolecular or intermolecular hydrogen bonds [15]. Tehrani-Bagha et al. also synthesized ester-bonded GQAS 12Q2OCO1Q12 and found that GQASs with ester groups in the spacer unit can easily biodegrade compared with GQASs with ester groups in the hydrophobic alkyl tails [21]. They also showed that 12Q2OCO1Q12 can be hydrolyzed to produce two products. Cytotoxic evaluation showed that the EC₅₀ value of 12Q2OCO1Q12 is 0.27 mg·L⁻¹, indicating that 12Q2OCO1Q12 is toxic to aquatic organisms [21]. Garcia et al. found that the aquatic toxicity of ester-based GQASs increases regularly with increasing hydrophobic alkyl tails [12]. Several studies have investigated the physicochemical properties, cytotoxicity and antimicrobial activity of GQASs with ester bonds based on changes in hydrophobic alkyl tails.

In view of the above points and the continuity of our previous works, GQASs with diester groups in the spacer and dodecyl alkyl chains, the same head groups and counterions were synthesized. They had different spacer structures and spacer lengths; the reaction scheme and abbreviations of the ester-bonded GQASs are shown in Scheme 1. The spacer of C₁₂-E_n-C₁₂ (n = 2, 4, 6) contained a flexible spacer group, while the spacer of C₁₂-B_m-C₁₂ (m = 1, 2) possessed a rigid benzene ring. The self-aggregation, antimicrobial activity and cytotoxicity of ester-bonded GQASs and their corresponding monomer N-dodecyl-N,N,N-trimethyl ammonium chloride DTAC were also surveyed. Furthermore, the effects of spacer length, spacer structure and polymerization degree on the properties of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ were evaluated. Therefore, this study establishes the structure–property relationship of ester-bonded GQAS, providing the basis for the development of biocompatible and nontoxic gemini surfactants with high antimicrobial activity.

2. Results and Discussion

2.1. Surface Tension

Figure 1 illustrates the surface tension (γ) curves for the synthesized ester-bonded GQASs and DTAC as a function of logarithm concentration at 25 °C. The surface tension values noticeably decreased and then stabilized after a certain concentration, corresponding to CMC. The CMC and surface parameters of the synthesized ester-bonded GQASs and DTAC were determined from the γ-logC curves, which are listed in

Table 1. The surface activities of the ester-bonded GQASs and DTAC at 25 °C.

Surfactant	CMC 1 (mmol·L−1)	CMC 2 (mmol·L−1)	$\mathsf{\beta }$	γCMC (mN·m−1)	Γmax × 106 (mol·m−2)	Amin (nm2)	pC20	$\mathbf{\Delta }{\mathit{G}}_{\mathit{m}}^{\mathit{\theta }}$ (kJ·mol−1)
C12-E2-C12	0.815	0.895	0.62	36.92	0.91	1.82	3.07	−25.97
C12-E4-C12	0.528	0.706	0.51	35.58	1.01	1.65	3.19	−28.20
C12-E6-C12	0.378	0.502	0.45	34.29	1.17	1.42	3.36	−32.22
C12-B1-C12	0.618	0.813	0.49	34.58	0.94	1.77	3.15	−27.30
C12-B2-C12	0.467	0.597	0.58	32.82	1.05	1.58	3.28	−30.61
DTAC	10.25	11.88	0.66	38.57	2.32	0.72	2.95	−22.78

¹ Surface tension method. ² Electrical conductivity method.

.

As shown in Table 1, the observed CMC values of C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ were 0.815, 0.528, 0.378, 0.618 and 0.467 mmol·L⁻¹, respectively, which were significantly lower than that of DTAC (10.25 mmol·L⁻¹). This difference could be due to the presence of two hydrophobic alkyl tails and two hydrophilic quaternary ammonium cations at the molecular structures of the ester-bonded GQASs, which enhanced the hydrophobic synergism among the hydrophobic alkyl tails [22]. The additional hydration in the spacer overcomes the Coulombic repulsive forces between the head group quaternary ammonium cations [23]. The ester groups in the spacer for the ester-bonded GQASs, which are more polar than the methylene spacers, promoted water-mediated hydrogen bonding between ester groups.

For the ester-bonded GQASs, the CMC values followed the order C₁₂-E₄-C₁₂ < C₁₂-B₁-C₁₂ < C₁₂-E₆-C₁₂ < C₁₂-B₂-C₁₂ < C₁₂-E₂-C₁₂. Furthermore, C₁₂-E₂-C₁₂ had the highest CMC value among C₁₂-E_n-C₁₂. Two head group quaternary ammonium cations in C₁₂-E₂-C₁₂ were connected with ethylene: two ester and two methylene groups. The spacer group was constrained, thus remaining in extended conformation and lying prone at the micelle–water (or air–water) interface [22,23,24,25], which limited the micellization process. The hydrophobicity and flexibility of C₁₂-E_n-C₁₂ increased with the increasing number of n of the polymethylene spacer from 2 to 6. The loop orientation of the longer spacer was extended toward the hydrophobic micellar core, and the spacers adopted the folder conformation toward the air side, preventing the hydration of the micelle. As a result, the CMC values decreased compared with n = 2 and n = 4 to 6. Zana also studied a series of GQASs with 12-s-12 with hydrophilic flexible spacers (s represents the number of –CH₂ groups in the spacer tail) and found that CMC values increased up to s = 6 and then decreased for s = 8, 10, 12 [26,27]. C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂ had larger CMC values than C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂, respectively, indicating that the rigidity of the spacer influenced the micellar aggregation [27]. According to the report by Zhu et al., the contribution of the benzene ring to the molecular structure of surfactants in the surface properties is similar to that of 3.5–4.0 methylene units [27]. We choose C₁₂-E₄-C₁₂ and C₁₂-B₁-C₁₂ as examples for discussion. The spacer length of C₁₂-B₁-C₁₂ is almost equal to that of C₁₂-E₄-C₁₂. Therefore, the spacers of C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂ had n-hexanediyl and n-butanediyl units, respectively. The flexibility of these units allowed the spacers to easily adopt looped conformation on the air–water interface. C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂ had phenylene and 1,4-benzenediyl units, and the benzene ring in the spacer was rigid and large in volume, forming the steric inhibition of C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂ caused by the rigid spacer [28]. The rigidity and steric hindrance of phenylene and 1,4-benzenediyl can prevent its incorporation in the hydrophobic core of the micelle. This would inhibit micelle formation and lead to larger CMC values. Regarding the aggregation behavior of C₁₂-B_m-C₁₂, C₁₂-B₂-C₁₂ had a stronger tendency to aggregate than C₁₂-B₁-C₁₂. Such behavior could be attributed to the rigid spacer of C₁₂-B_m-C₁₂. The rigid hydrophobic spacer of C₁₂-B₁-C₁₂ lay flat at the surface at the micelle–water (or air–water) interface; the spacer of C₁₂-B₂-C₁₂ could be formed by adding two methylene groups to the spacer of C₁₂-B₁-C₁₂, which increased hydrophobic interaction and favored micellization.

In this study, γ_CMC was defined as the effectiveness of the ester-bonded GQASs or DTAC in reducing the surface tension of water. The γ_CMC values of C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ were 36.92, 35.58, 34.29, 34.58 and 32.82 mN·m⁻¹, respectively, which were slightly lower than that of DTAC (38.57 mN·m⁻¹). This indicates C₁₂-B_m-C₁₂ and C₁₂-E_n-C₁₂ are adsorbed strongly at the air–water interface and are effective in reducing the surface tension of water.

The minimum surface area (A_min) occupied per the synthesized ester-bonded GQASs or DTAC molecule and the surface excess concentration (Γ_max) were estimated based on the following Gibbs equations, Equations (1) and (2) [6]:

$${\Gamma }_{\mathit{max}}=-\frac{1}{2.303nRT}{\left(\frac{d\gamma }{d\mathit{log}C}\right)}_T$$

(1)

$$A_{\mathit{min}}=\frac{1}{N_A{\Gamma }_{\mathit{max}}}$$

(2)

where R = 8.314 J·mol⁻¹·K⁻¹, T = 298.15 K, and N_A = 6.023 × 10²³ mol⁻¹; γ and C represent the surface tensions and the concentration of ester-bonded GQASs or DTAC, respectively; dγ/dlogC is the slope of the line before the CMC in the γ-lgC curves. The parameter n is a constant corresponding to the number of individual ions for the surfactants adsorbed at the interface. n is taken as 3 for the ester-bonded GQASs, implying the ion paring and dissociation of two univalent counterions from the dicationic quaternary ammonium headgroups, while n is taken as 2 for DTAC, implying the surfactant ion and the chloridion counterion were monovalent.

As shown in Table 1, the A_min values of C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂, C₁₂-B₂-C₁₂ and DTAC were 1.82, 1.65, 1.42, 1.77, 1.58 and 0.72 nm², respectively. The ester-bonded GQASs had larger A_min values than DTAC, indicating the impact of the polymerization degree on the A_min value. C₁₂-E₂-C₁₂ had the highest A_min value among C₁₂-E_n-C₁₂, demonstrating that the spacer length impacts the A_min value. The flexibility and hydrophobicity of C₁₂-E_n-C₁₂ increases because of the lengthening of the spacer, and the spacer adopted the folder conformation to the water side of the air–water interface, thus decreasing A_min value and increasing the aggregation density of surfactant per unit area. As a result, the Γ_max value increased with increasing the spacer length. Moreover, C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂, with the rigid benzene ring in the spacer, had a larger A_min value than C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂, with a similar length of flexible spacer, respectively, indicating that the spacer structure can impact the A_min value. Compared with the flexible spacer of C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂, the benzene ring of C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂ increased the spacer rigidity, thus hindering the conformational change of C₁₂-B_m-C₁₂ at the air–water interface. The same change tendency for the spacer rigidity with regard to the A_min value was reported by Zheng [23]. Zheng et al. found that the gemini surfactant (Mor)m-P-m containing the benzene ring as the spacer had a larger A_min value, and the spacer tended to lay on the air–water interface [23]. However, introducing two flexible methylene units to the spacer of C₁₂-B₁-C₁₂ reduced the A_min value from 1.77 nm² to 1.58 nm² since the spacer of C₁₂-B₂-C₁₂ became more flexible and more hydrophobic.

The adsorption efficiency pC₂₀ refers to the negative logarithm concentration of the ester-bonded GQASs and DTAC to decrease the surface tension of the aqueous system by 20 mN·m⁻¹ [23]. Higher pC₂₀ values indicate greater efficiency in reducing surface tension. Table 1 shows that the pC₂₀ values of C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂, C₁₂-B₂-C₁₂ and DTAC are 3.07, 3.19, 3.36, 3.15, 3.28 and 2.95, respectively. Obviously, the pC₂₀ values of ester-bonded GQASs are higher than that of DTAC, suggesting that the ester-bonded GQASs are strongly adsorbed at the water–air interface and have greater efficiency in reducing surface tension. The pC₂₀ values for C₁₂-E_n-C₁₂ displayed an increasing trend from 3.07 to 3.36 with spacer lengthening. The pC₂₀ value of C₁₂-B₂-C₁₂ was larger than that of C₁₂-B₁-C₁₂, which may be associated with the arrangement of the spacer at the water–air interface.

2.2. Conductivity Measurements

The aggregation behavior of the ester-bonded GQASs or DTAC in aqueous solutions was investigated by the conductivity method at 25 °C. The representative conducto-metric profiles for C₁₂-E₂-C₁₂ at 25 °C are shown in Figure 2. The plots of κ vs. C for the other four ester-bonded GQASs and DTAC are provided in Figures S17–S21 in the Supplementary Materials. The CMC values were determined by the inflection point of κ vs C and are presented in Table 1. The CMC values for C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂, C₁₂-B₂-C₁₂ and DTAC were 0.985, 0.643, 0.489, 0.745, 0.579 and 11.88 mmol·L⁻¹, respectively, which were slightly higher than those investigated by surface tension measurements. However, the CMC values determined using the conductivity method and the surface tension measurements followed the same tendency. The degree of counterion binding (β), the ability of chloride counterions to bind to the stern layer of the micelle, was determined from the slope values before and after the CMC obtained by the conductivity plot. Table 1 shows that the β values for C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂, C₁₂-B₂-C₁₂ and DTAC are 0.62, 0.51, 0.45, 0.49, 0.58 and 0.66, respectively, and the β values of ester- bonded GQASs are all slightly lower than that of DTAC, possibly because of the charge-density of the head group polarity [6]. The ester-bonded GQASs had larger A_min values compared to DTAC because of the presence of the spacer, thus affecting binding between chloride counterions and micelles. The β values of the C₁₂-E_n-C₁₂ series decreased with an increase in the spacer chain length. Zhang and Hou et al. reported similar findings for different structures of cationic gemini surfactants [22,24]. Furthermore, the β values for C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ were lower than those of C₁₂-E₄-C₁₂ and C₁₂-E₆-C₁₂, respectively. Martin et al. also reported similar findings for alkanediyl-α-ω-bis (dodecyl-dimethylammonium) bromide [25]. The A_min values of C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ on the air–water interface were larger than those of C₁₂-E₄-C₁₂ and C₁₂-E₆-C₁₂. Rigidity and steric hindrance of phenylene and 1,4-benzenediyl of C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ could prevent incorporation in the nonpolar region of the micellar core [28]. Moreover, the rigid spacer of C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ hindered the binding between chloride counterions and micelles [28].

The Gibbs free energy of micellization, $\Delta {\mathrm{G}}_{\mathrm{m}}^{\mathrm{\theta }}$, of the synthesized ester-bonded GQASs or DTAC can be computed using Equation (3) below [6].

$$\Delta G_m^{\theta }=RT(1/2+\beta )\ln X_{CMC}$$

(3)

where X_CMC = CMC/55.4; 55.4 is derived from 1 L of water corresponding to 55.4 mol of water at 25 °C [25]. Table 1 shows that the $\Delta G_m^{\theta }$ values for C₁₂-E₂-C₁₂, C₁₂-E₄-C₁₂, C₁₂-E₆-C₁₂, C₁₂-B₁-C₁₂, C₁₂-B₂-C₁₂ and DTAC are −25.97, −28.20, −32.22, −27.30, −30.61 and −22.78 kJ·mol⁻¹, respectively. Notably, all $\Delta {\mathrm{G}}_{\mathrm{m}}^{\mathrm{\theta }}$ values of the ester-bonded GQASs and DTAC are negative, which indicates that the micellization is an energy-favorable process. The $\Delta G_m^{\theta }$ values of the ester-bonded GQASs were more negative than that of DTAC, indicating that the ester-bonded GQASs were more favored to form micelles. Moreover, the $\Delta G_m^{\theta }$ values decreased with the elongation of the spacer length for C₁₂-E_n-C₁₂. The $\Delta G_m^{\theta }$ values of C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ with rigid spacers were higher than those of C₁₂-E₄-C₁₂ and C₁₂-E₆-C₁₂, respectively, demonstrating that the spacer rigidity had an effect on the micellization.

2.3. Antimicrobial Activity

The antibacterial effect of the ester-bonded GQASs and DTAC was assessed using S. aureus and E. coli. The minimum inhibitory concentrations (MICs) were examined by the agar dilution method, and these data are listed in

Table 2. The MIC values of the ester-bonded GQASs and DTAC against E. coli and S. aureus (μg·mL⁻¹).

Strains	C12-E2-C12	C12-E4-C12	C12-E6-C12	C12-B1-C12	C12-B2-C12	DTAC
S. aureus	12	10	5	11	8	16
E. coli	22	15	13	19	17	68

. It was found that the five ester-bonded GQASs and DTAC exhibited stronger antibacterial activity toward S. aureus than E. coli, probably due to the different cell membrane structures of S. aureus and E. coli. The cell walls of S. aureus consist of many layers of peptidoglycan and teichoic acids, promoting the permeation of surfactants through the phospholipid layer of the cell membrane. However, the cell walls of E. coli contain a thin layer of peptidoglycan, lipopolysaccharides and porins, inhibiting the permeation of surfactants and the entrance of the hydrophobic alkyl tails of surfactants [22]. The MIC values of the five synthesized ester-bonded GQASs ranged from 13 to 22 μg·mL⁻¹ for E. coli and from 12 to 5 μg·mL⁻¹ for S. aureus, while the MIC values of DTAC were 68 μg·mL⁻¹ for E. coli and 16 μg·mL⁻¹ for S. aureus. Notably, the MIC values of the five ester-bonded GQASs on both E. coli and S. aureus were smaller than that of DTAC, indicating that the ester-bonded GQASs exhibited a stronger antimicrobial effect than DTAC. The antibacterial mechanism of cationic surfactants is induced by the electrostatic interaction between the positively charged headgroup of surfactants and the negatively charged phospholipid surface of the bacterial cell wall. In addition, hydrocarbon tails of cationic surfactants can penetrate and disturb the selective permeability of the membrane, leading to the leakage of cell contents and, ultimately, cell death [22,23]. The ester-bonded GQASs contained organic quaternary ammonium cations and two hydrophobic dodecyl chains in the molecule, facilitating stronger electrostatic and hydrophobic interaction with cell membranes than DTAC.

Amongst the C₁₂-E_n-C₁₂ series examined, the MIC values gradually decreased for both S. aureus and E. coli with increasing of the spacer chain length. Similar behavior has been observed for MIC values of different molecular structures of GQASs [9,22,23]. The increase in spacer length resulted in increasing hydrophobicity and flexibility, thus promoting the adsorption tendency of C₁₂-E_n-C₁₂ molecules at cell membrane surfaces. Meanwhile, the β values of C₁₂-E_n-C₁₂ decreased with the spacer chain lengthening due to the decrease of the charge density. This findings indicate that C₁₂-E₆-C₁₂ may have a weaker electrostatic interaction with the cell membrane than C₁₂-E₄-C₁₂ and C₁₂-E₂-C₁₂. The joint effects of electrostatic and hydrophobic interactions with cells made the MIC values of C₁₂-E_n-C₁₂ follow the pattern C₁₂-E₂-C₁₂ > C₁₂-E₄-C₁₂ > C₁₂-E₆-C₁₂, meaning that C₁₂-E₆-C₁₂ has the highest antimicrobial activity on S. aureus and E. coli. Nonetheless, the MIC values of C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂ on S. aureus and E. coli were lower than those of C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂, indicating that C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂ have stronger electrostatic and hydrophobic interactions with the cell wall and better antibacterial properties. This could be due to the spacer of the ester-bonded GQASs. C₁₂-B_m-C₁₂ with phenylene and 1,4-benzenediyl as the spacer had a more steric effect than C₁₂-E₆-C₁₂ and C₁₂-E₄-C₁₂ with n-hexanediyl and n-butanediyl as the spacer, thus reducing the permeability of the cell membrane. In addition, conformational changes of C₁₂-B₂-C₁₂ and C₁₂-B₁-C₁₂ can weaken the hydrophobic binding between the tails and the lipid layers of cells [21].

2.4. Cytotoxicity of Gemini Surfactants

The cytotoxicity profiles of ester-bonded GQASs and DTAC in cultured Hela cells were assessed by the standard CCK8 method. The value of half-maximal inhibitory concentration (IC₅₀) was quantified based on the profiles of the cell viability versus surfactant concentration (C) (Figure 3), and the results are shown in

Table 3. Cytotoxic effect (IC₅₀) of the ester-bonded GQASs and DTAC on Hela cells.

Surfactants	C12-E2-C12	C12-E4-C12	C12-E6-C12	C12-B1-C12	C12-B2-C12	DTAC
IC50 (μmol·L−1)	36.34 ± 2.04	29.36 ± 3.11	21.01 ± 2.11	14.55 ± 1.23	12.75 ± 2.45	5.04 ± 1.23

. The IC₅₀ values were 14.55 and 12.75 μmol·L⁻¹ for C₁₂-B_m-C₁₂ (m = 1, 2), respectively, and ranged from 36.24 to 21.01 μmol·L⁻¹ for C₁₂-E_n-C₁₂ with n from 2 to 6. The IC₅₀ values of ester-bonded GQASs were higher than that of DTAC, indicating that the ester-bonded GQASs are less toxic than DTAC. This was in accordance with the findings of Garcia [12]. In comparison with the traditional GQASs, the IC₅₀ values of ester-bonded GQASs were greater than those of 12-s-12, in the range from 2.8 to 3.1 μmol·L [22]. This indicated that the ester-bonded GQASs have lower cytotoxicity compared to those reported for the traditional GQASs, and the ester group within the spacer functionality can reduce their cytotoxicity, which was slightly analogous to the study by Pinazo with GQASs containing ester bonds in their spacers [29]. This is because the introduction of ester to GQASs can enhance the hydrophilicity; meanwhiles the ester bond linking the head group to the two hydrophobic chains had lower chemical stability and could be easily hydrolyzed, giving rise to non-toxic species [30,31]. Moreover, the IC₅₀ values of C₁₂-E_n- C₁₂ and C₁₂-B_m-C₁₂ decreased with the growth of the spacer chain length; then, the cytotoxicity increased slightly with elongation of spacer chain, which was in agreement with the data available on cytotoxicity activity of other gemini surfactants [31]. The IC₅₀ values of C₁₂-B₁-C₁₂ and C₁₂-B₂-C₁₂ were lower than those of C₁₂-E₄-C₁₂ and C₁₂-E₆-C₁₂, respectively, which showed that the structure of the spacer had an effect on the cytotoxicity activity of ester-bonded GQASs, which could be attributed to the combined effects of electrostatic and hydrophobic interactions of ester-bonded GQASs with Hela cells [21].

3. Materials and Methods

3.1. Materials

1,4-benzene-dimethanol (99%), hydroquinone (99.5%), N,N-dimethyldodecyl amine (95%), ethylene glycol (99%), hexane-1,6-diol (99%), butane-1,4-diol (99%) and DTAC were sourced from Aladdin Industrial Corporation. Chloroacetyl chloride was obtained from Xiya Reagent Factory. Beef-extract paste, agar powder and protein peptone were purchased from Sinopharm Reagent Co., Ltd. Bovine serum, streptomycin and penicillin were sourced from Sigma-Aldrich. E. coli and S. aureus were provided by Shanghai Microspectrum Chemical Technology Service Co., Ltd. Human cervical squamous cell carcinoma Hela cells was provided by the Department of Life Science, Changzhi University. All other reagents were of analytical grade. Milli-Q water, obtained by sub-boiling distilled water, was used for all experimental procedures.

3.2. Synthesis and Characterization

An ALPHA infrared spectrophotometer (Bruker BioSpin Co., Ettlingen, Germany) was used to study the FTIR spectra of the synthetic GQAS. Meanwhile, ¹H NMR spectra and ¹³C NMR spectra were determined using a BRUKER Ascend^TM 400 instrument (Bruker BioSpin Co., Ettlingen, Germany). A Compact mass spectrometer (Bruker Co., Bremen, Germany) was used to detect the mass spectra of ester-based GQASs using ESI as an ion source. A K100 tensiometer (Krüss Co., Hamburg, Germany) was used to determine surface tensions of the synthetic GQASs and DTAC, while a DDSJ-308F conductimeter (Shanghai Rex Electric Chemical Co., Shanghai, China) was used to determine the conductivity. A microplate reader (Synergy H1, Agilent BioTek, Santa Clara, CA, USA) was used to evaluate the cytotoxicity of the surfactants.

Take C₁₂-E₂-C₁₂ as an example for demonstrating the procedure of synthesis. Freshly distilled triethylamine (80 mmol, 8.52 g) as an acid-binding agent and ethylene glycol (30 mmol, 1.88 g) were added to 100 mL absolute dichloromethane as a solvent at room temperature and in a nitrogen atmosphere. Chloroacetyl chloride (90 mmol, 10.16 g) dissolved in 40 mL dichloromethane was added to the mixed solution dropwise at 0–5 °C for 5 h. The viscous liquid was achieved by washing with 0.5 mol·L⁻¹ Na₂CO₃ aqueous solution, drying with anhydrous MgSO₄, suction filtering, discolorating with active carbon and evaporating the solvent.

The obtained product (20 mmol, 3.72 g) was dissolved in 40 mL anhydrous ethyl acetate, followed by addition of 50 mmol 10.67 g N,N-dimethyldodecylamine under even stirring. The solution was refluxed for 48 h, cooled down to ambient temperature and filtered. The wet filter cake was recrystallized thrice using acetone and ethanol (1:1) to obtaining the product as fairy white powder.

Ethane-1,2-diyl bis(N,N-dimethyl-N-dodecylammonium acetoxy) dichloride (C₁₂-E₂-C₁₂). 9.71 g, 85% yield. FT-IR (KBr, cm⁻¹): 2915, 2847, 1738, 1047, 720. ¹H NMR (DMSO, ppm): 0.84–0.88 (t, J = 8.0 Hz, 6H), 1.25–1.28 (m, 36H), 1.66–1.73 (m, 4H), 3.25 (s, 12H), 3.51–3.55 (t, J = 8.0 Hz, 4H), 4.45 (s, 4H), 4.70 (s, 4H). ¹³C NMR (DMSO, ppm): 14.42, 22.25, 22.56, 26.19, 28.94, 29.18, 29.27, 29.42, 29.49, 31.60, 51.16, 60.92, 63.42, 64.93, 165.01. HRMS(ESI+): m/z calcd for [M-Cl]²⁺/2: 285.2663, found: 285.2679.

Butane-1,4-diyl bis(N,N-dimethyl-N-dodecylammonium acetoxy) dichloride (C₁₂-E₄ -C₁₂). 9.96 g, 83% yield. FT-IR (KBr, cm⁻¹): 2920, 2850, 1767, 1050, 718. ¹H NMR (DMSO, ppm): 0.84–0.88 (t, J = 8.0 Hz, 6H), 1.25–1.28 (m, 36H), 1.66–1.72 (m, 8H), 3.23 (s, 12H), 3.49–3.53 (t, J = 8.0 Hz, 4H), 4.20–4.23 (s, 4H), 4.56 (t, 4H). ¹³C NMR (DMSO, ppm): 14.58, 22.18, 22.56, 25.21, 26.26, 28.18, 28.90, 29.18, 29.21, 29.39, 29.48, 31.76, 51.28, 61.14, 64.75, 66.15, 165.37. HRMS(ESI+): m/z calculated for [M-Cl]²⁺/2: 299.2819, found: 299.2844.

Hexane-1,6-diyl bis(N,N-dimethyl-N-dodecylammonium acetoxy) dichloride (C₁₂-E₆-C₁₂). 10.53 g, 84% yield. FT-IR (KBr, cm⁻¹): 2913, 2850, 1746, 1062, 719. ¹H NMR (DMSO, ppm): 0.80–0.87 (t, J = 8.0 Hz, 6H), 1.25–1.37 (m, 40H), 1.61–1.71 (m, 8H), 3.20–3.24 (s, 12H), 3.50–3.55 (t, J = 8.0 Hz, 4H), 4.16–4.20 (t, J = 8.0 Hz, 4H), 4.60 (s, 4H). ¹³C NMR (DMSO, ppm): 14.42, 22.23, 22.56, 24.82, 26.14, 28.91, 29.18, 29.23, 29.40, 29.48, 31.76, 51.35, 60.95, 64.74, 65.64, 165.37. HRMS(ESI+): m/z calculated for [M-Cl]²⁺/2: 313.2976, found: 313.2995.

Benzene-1,4-diyl bis(N,N-dimethyl-N-dodecylammonium acetoxy) dichloride (C₁₂-B₁-C₁₂). 10.65 g, 86% yield. FT-IR (KBr, cm⁻¹): 2915, 2847, 1738, 1466, 1414, 720. ¹H NMR (DMSO, ppm): 0.85–0.88 (t, J = 8.0 Hz, 6H), 1.26–1.31 (m, 36H), 1.74–1.78 (m, 4H), 3.23 (s, 12H), 3.53–3.57 (t, J = 8.0 Hz, 4H), 4.70–4.79 (s, 4H), 7.39 (m, 4H). ¹³C NMR (DMSO, ppm): 14.42, 22.27, 22.56, 26.12, 28.91, 29.17, 29.22, 29.47, 31.76, 51.48, 62.26, 65.29, 123.50, 147.62, 164.32. HRMS(ESI+): m/z calculated for [M-Cl]²⁺/2: 309.2663, found: 309.2679.

Phenylene-1,2-diyl bis(N,N-dimethyl-N-dodecylammonium acetoxy) dichloride (C₁₂-B₂-C₁₂). 10.61 g, 82% yield. FT-IR (KBr, cm⁻¹): 2916, 2850, 1740, 1467, 1415, 722. ¹H NMR (DMSO, ppm): 0.87–0.92 (t, J = 8.0 Hz, 6H), 1.27–1.34(m, 36H), 1.72 (m, 4H), 3.55 (s, 12H), 3.84 (t, J = 8.0 Hz, 4H), 5.16 (s, 4H), 5.33(m, 4H), 7.35 (m, 4H). ¹³C NMR (DMSO, ppm): 14.42, 22.37, 22.66, 26.42, 29.01, 29.22, 29.32, 29.42, 29.57, 31.88, 51.60, 61.34, 65.37, 123.38, 147.91, 164.24. HRMS(ESI+): m/z calculated for [M-Cl]²⁺/2: 323.2819, found: 323.2808.

3.3. Surface Tension Measurement

A K100 automatic tensiometer was used to determine the surface tension of the ester-bonded GQASs and DTAC at the water–air interface with a Wilhelmy plate at 25 °C. Three consecutive readings with a standard deviation less than ±0.2 mN·m⁻¹ were recorded at each concentration. The surface tensions values were plotted against the logarithm of the surfactant concentration. The CMC and γ_CMC values of the ester-bonded GQASs and DTAC were determined based on the inflection point on the plot of γ against log C [6].

3.4. Conductivity Measurements

Conductivity of the synthesized ester-bonded GQAS and DTAC solutions was determined using a conductimeter equipped with platinum electrodes in conjunction with a conductivity cell with a cell constant of 0.98 cm⁻¹. The surfactant solutions were placed in thermostatic water bath at 25 °C, maintained using a temperature controller. The uncertainty of the conductivity measurements was ±0.2 μs·cm⁻¹ [6]. The CMC values were determined based on the intersection between two line segments of the plots [6]. The conductivity was measured three times for each solution.

3.5. Antimicrobial Activity

Antimicrobial activity of the ester-bonded GQASs and DTAC against gram-positive (S. aureus) bacterium and gram-negative (E. coli) bacterium was evaluated by the broth dilution method [21]. The bacterial strains were each cultured in the nutrient agar plates with an inoculating loop at 37 °C overnight, then diluted to about 10⁷ CFU·mL⁻¹. A total of 1 mL of the surfactant solution was added to the nutrient broths, followed by inoculation of 10 mL of bacterial suspension at 37 °C for 1 h [1]. The test organisms were coated on the Petri dishes by the tenfold dilution method, then inoculated at 37 °C for 24 h. The bacterial liquid without any surfactant solution was used as the blank control group. The minimum inhibitory concentration (MIC) values were quantified according to the concentration of the surfactant needed to achieve a sterilization rate of 90% [7]. All tests were performed in triplicate.

3.6. Cytotoxicity Assay (CCK-8 Test)

The cytotoxicity of the synthesized ester-bonded GQASs and DTAC in Hela cells was tested using the CCK-8 assay, as previously reported [32,33]. Hela cells (100 μL, 4×10³ per cell in a 96-well plate) were treated with 100 μL culture medium per well and grown in an incubator for 24 h with a temperature of 37 °C and 5% CO₂ [33]. Hela cells were treated with different concentrations of the ester-bonded GQASs or DTAC (0.0, 0.05, 0.1, 1, 5, 10, 20, 40, 60, 80 and 100 μmol·L⁻¹) overnight. A total of 10 μL of CCK8 solution was added to each well in 96-well plate (away from light) [23]. After 4 h, CCK-8–containing medium was shaken for 15 min. The absorption was then read at 450 nm with a microplate reader. The cell viability of the ester-bonded GQASs and DTAC was calculated according to the following equation:

$$Cell viablity/\%=\frac{A-A_0}{A_c^{}-A_0}\times 100\%$$

(4)

A, A₀ and A_c represent the absorbance (450 nm) in the treated cells, the blank in which the cells almost completely died, and the untreated cells, respectively [22]. The half maximal inhibitory concentration (IC₅₀) was quantified based on the plot of the cell viability versus the logarithm of surfactant concentration (C). All the data were expressed as mean ± standard deviations for six different experiments.

4. Conclusions

In summary, a series of ester-bonded GQASs with different spacer groups were synthesized, and their self-aggregation, antimicrobial activity and cytotoxicity were systematically studied. The CMC values of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ were found to be between 0.895 and 0.597 μmol·L⁻¹, which was markedly lower compared with DTAC. The CMC values of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ decreased with the spacer lengthening in the series of homologs. In case of the equivalent chain length, C₁₂-B_m-C₁₂ with the rigid spacer exhibited higher CMC values than C₁₂-E_n-C₁₂ containing the flexible spacer. C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ exhibited stronger antibacterial activity than DTAC, and they showed greater inhibition of S. aureus than E. coli. The MIC values of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ were found to decrease with the n changing from 2 to 6 and the m changing from 1 to 2. They decreased with the decrease in rigidity of the spacers. The cytotoxicity evaluation of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ in cultured Hela cells showed that the cytotoxicity of C₁₂-E_n-C₁₂ and C₁₂-B_m-C₁₂ was obviously lower than alkanediyl-α,ω-bis (dimethyldodecyl ammonium) bromide surfactants and DTAC. The spacer structure and the spacer length could induce significant cytotoxic effects on Hela cells. Therefore, this study offers deep insight into the interfacial properties, the antibacterial activity and the cytotoxicity of the ester-bonded GQASs and provides ideas for the potential technical and biological applications.