Green Catanionic Gemini Surfactant – Lichenysin Mixture : Improved Surface , Antimicrobial and Physiological Properties

Catanionic surfactant mixtures form a wide variety of organized assemblies and aggregates with improved physicochemical and biological properties. The green catanionic mixture NN-bis(Ncaproylarginine) α, ω-propyldiamide (C3(CA)2):lichenysin (molar ratio 8:2) showed antimicrobial synergies against Yersinia enterocolitica, Bacillus subtilis, Escherichia coli O157:H7 and Candida albicans. Flow cytometry and viability studies indicated that this catanionic mixture increases the probability of Y. enterocolitica (38.2%) and B. subtilis (17.1%) cells entering a viable but nonculturable state. Zeta potential showed that one of the cationic charges of C3(CA)2 is neutralised by lichenysin. An isotherm study demonstrated the formation of a stable aggregate between the two surfactants. The catanionic aggregate was able to interact with bacterial phospholipids. The lowest hemolysis (22.1 μM) was obtained with the catanionic mixture, although an irritant potential (0.70) was characterised. According to the therapeutic index, the C3(CA)2:lichenysin mixture was the formulation least toxic to eukaryotic cells. Partial neutralisation of C3(CA)2 by lichenysin modified the mode of action that enhances the transition of bacterial cells into a viable but nonculturable state (VBNC) and improved the cell selectivity. Page 1 of 47 ACS Paragon Plus Environment ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

are thermodynamically stable for long periods of time. 10 Recently it has been demonstrated that catanionic mixtures may also have improved biological properties, attributed to synergistic effects, compared with those of the individual components. 11 In a previous study, the catanionic mixture of lichenysin and an arginine-based surfactant showed synergistic antimicrobial activity. 12 These results prompted us to explore new catanionic mixtures (lichenysin plus arginine-based surfactants) in order to define the molecular requirements in their chemical structures for synergistic antimicrobial activity, and to characterise their mode of action and physicochemical and physiological properties.   8 The inoculum was a suspension equivalent to independent measurements. Each measurement was in turn the average of ten sub-measurements of 20 s each.

Monolayer isotherms.
Monolayer isotherms (surface pressure versus mean molecular area, π-A) at 37ºC were measured using a Langmuir balance (KSV Instruments Minitrough, Finland) and a paper Wilhelmy plate (Whatman ashless) as detailed in Lozano et al. 15 The surface pressure is defined as π=γ0-γ, where γ0 is the water surface tension (72.3 mN/m). Tris buffer 20 mM (Merck, Germany) at pH 6.8 was used as the subphase. Monolayers of lichenysin, C3(CA)2, CAM and an E. coli total lipid extract (TLE) (Avanti Polar Lipids, USA) were studied. Weight composition and charges at pH 6.8 of the TLE were phosphatidylethanolamine 57.5% (zwitterionic), phosphatidylglycerol 15.1% (one negative charge), cardiolipin 9.8% (two negative charges) and an unknown fraction 17.6%. The mean molecular weight was estimated as 772.75 g/mol from the phospholipid fraction. Aliquots of 25 µL of single components, catanionic mixtures of surfactants:lichenysin (molar ratio 8:2) and mixtures with TLE (volume ratio 8:2), all prepared in chloroform:ethanol 9:1 (V:V) at 1 mg/mL, were spread on the surface with a microsyringe (Hamilton 50±1 µL). The subphase was agitated with a stirrer and evaporation of the solvent was allowed for 15 min. The rate of symmetric compression was 20 mm/min. The surface pressure was monitored by the plate's weight. Each isotherm was measured at least twice. The π-A curve was plotted using the statistical software OriginPro 8. Mixed isotherms were analyzed as binary mixtures of monolayers. The excess of free energy of the mixture (ΔGm ex ) was calculated considering ΔGm ex =0 π (A12-x1*A1-x2*A2)dπ where A12 is the mean molecular area of the binary mixture, x1 and x2 are the molar fractions of the components in the mixture and A1 and A2 are the respective molecular area of each component in a pure monolayer. 16 Integrations were performed from inverted π-A curves using OriginPro 8.

RESULTS AND DISCUSSION
Production, extraction and characterization of lichenysin. Lichenysin was produced in the culture supernatant, which after 72 hours of incubation had a surface tension of 30±2 mN/m.
Lichenysin was recovered using the lyophilization-based concentration method, with a final productivity of crude extract up to 351.5 mg/L. The organic extract and purified lichenysin were compared by TLC to check purity. From here onwards the organic extract will be referred to as lichenysin.  shows a π at collapse lower than TLE alone and lower than the isotherm of CAM:lichenysin. This indicates a lower number of molecules in the monolayer or that the molecules remaining in the monolayer have less surface activity.
As explained before, one molecule of the cationic surfactants C3(CA)2 or CAM interacts with one molecule of the anionic lichenysin. The resulting catanionic aggregates would be cationic and non-ionic, respectively, and in the case of C3(CA)2, interactions with the partially negativelycharged phospholipids of TLE would be preferred. The formation of a catanionic aggregate with a different charge may explain the differences in the monolayer behaviour.
A complementary analysis of the data obtained was performed in order to understand the interactions and miscibility of the components by calculating the excess of free energy of the mixtures, ΔGm ex . If all the components mix ideally, at a given surface pressure and temperature, the proportional sum of areas of single component isotherms would be equal to the experimental area of the mixed monolayer isotherm. Any deviation would be due to interactions between components and partial or total miscibility. Hence, in a binary system, an ideal behaviour without miscibility of any component would result in ΔGm ex =0, while any deviation from it will be due to molecular interactions, even with the subphase, or the miscibility of each component or mixed aggregates. 16 A negative value of the excess of free energy evidences strong interactions of the two components, which leads to a partial formation of miscible complexes or aggregations that are diluted in the subphase, whose quantity depends on the molar ratio of the components. On the other hand, a positive value of the excess free energy evidences that the interactions between the two components are weaker than the individual self-interactions. In this case, at least one component forms auto-aggregates and becomes diluted in the subphase. 20 ΔGm ex values for C3(CA)2, CAM and their mixtures with lichenysin and TLE are shown in Figure 3. Notably, the catanionic mixtures studied in this work can be considered green systems. For practical reasons, lichenysin was synthesized using commercial glucose as the carbon source, but this can be substituted by molasses. The gemini surfactant C3(CA)2 was synthesized using renewable raw materials, arginine and fatty acid, by a chemoenzymatic approach in which papain is deposited into cells. 21,22 The formulations were prepared without high mechanical energy, and moreover, biodegradation studies showed that CAM and C3(CA)2 are readily biodegradable surfactants. 23
The MIC of the catanionic mixtures was characterized in the search for antimicrobial synergies.
Based on previous work, a molar ratio of 8:2 was chosen. 12       Although a high proportion of Y. enterocolitica cells were not stained, suggesting that their cell envelope was unaltered, they were not culturable in solid media. An explanation might be that other alterations, not detected by FC, rendered the cells unable to recover in solid media. Such cells are considered to be in a transitory viable but nonculturable (VBNC) state and represented 57.2% of the cells treated with the mixture and 19% of those treated with C3(CA)2. 27 The VBNC state is associated with sub-lethal metabolic and genetic alterations in response to a stressful environment. 28 The higher percentage of VBNC cells after treatment with the mixture indicates that this treatment affected the cell envelope less than the gemini surfactant, but reduced the viability alike.
Results of E. coli O157:H7 ( Figure 4B  None of the cells presented nuclei, perhaps because they had been disrupted, or vacuoles, which were always present in control cells (not shown). And finally, numerous cells had begun the formation of one or more gems, all of which seem to have been interrupted at the same point, while in the control the gemmating processes had been stopped at different stages by the fixation procedure. Overall, the severe alterations of the cytoplasm seem to have caused the cell death detected in both the viability reduction and FC assays ( Figure 4D).
The microbial cell envelope acts as an effective permeability barrier against antibiotics or biocides. Some compounds with little or no antimicrobial activity are being used to block or bypass active or intrinsic bacterial resistance mechanisms or enhance antibiotic action to rescue the activity of existing drugs. These compounds are called antibiotic adjuvants. 32 A cationic molecule able to interact with a negatively charged bacterial cell envelope causing destabilization and permeability could allow antimicrobial compounds to enter the cell. 33 According to the results obtained, the catanionic mixture could fulfill this role and may be considered a possible antibiotic adjuvant to reduce the onset of resistance.

RBC hemolysis and therapeutic index.
In order to test the irritancy potential of lichenysin, C3(CA)2 and their 8:2 mixture (mol:mol), the hemolysis level of these surfactants was studied. The resulting hemolysis curves are shown in Figure 6 and values extracted from them in Table 2 and their mixture at a molar ratio 8:2 (mol:mol) after 20 minutes.
All the studied surfactant formulations were classified as irritants ( Having determined the superior antimicrobial activity of the catanionic mixture, the relationship between the hemolysis curves and their respective MICs against different microorganisms was key for establishing whether this heightened activity affects eukaryotic cells. Lichenysin showed no antimicrobial activity within the interval shown, so no MIC was drawn ( Figure 6A). The MICs of C3(CA)2 ( Figure 6B) and the mixture ( Figure 6C  The therapeutic index (TI) correlates MICs with the H50 to express the relative cell selectivity of the formulations against microorganisms (Table 1). TI values of lichenysin were the lowest, due to its lack of antimicrobial activity. Overall, TI values of the catanionic mixture were higher than those of C3(CA)2 for two possible reasons: the MICs of the mixture were lower, implying a higher antimicrobial activity, or the H50 was higher, resulting in less hemolytic activity. In this case, it has already been proven that the mixture is less hemolytic, which would improve the TI when MICs are equal. However, in addition, a synergic antimicrobial activity against determined strains was detected. Focusing on E. coli O157:H7, Y. enterocolitica, B. subtilis and C. albicans, it can be seen that the TIs of the mixture formulation improved at least three-fold compared to C3(CA)2. It was therefore more selective against microbial cells than against eukaryotic cells, while the other formulations were less selective.

CONCLUSIONS
The green catanionic mixture of C3(CA)2, a cationic arginine-based gemini surfactant, and lichenysin, an anionic cyclic lipopeptide biosurfactant, induces the formation of a catanionic aggregate with a significant synergic antimicrobial activity, in which lichenysin acts as an antimicrobial potentiator.
The partial neutralisation of the two cationic polar heads of the gemini surfactant by the anionic charge of the biosurfactant changes its mode of action. When cells are treated with the gemini surfactant alone, the cell envelope is first depolarized and finally disrupted. When cells are treated with the catanionic mixture, bacterial cell envelopes are not disrupted but cells are altered at a cytoplasmic level, which makes them more likely to enter a viable but nonculturable state.
Additionally, the catanionic mixture showed a strong fungicidal activity. Our hypothesis is that the free cationic charge allows the catanionic aggregate to approach the anionic cell envelope by electrostatic interactions and to interact with it by aggregation with the bacterial phospholipids, which enhances the antimicrobial effect. On the other hand, when the cationic surfactant is completely neutralised, as occurs in catanionic mixtures of lichenysin and monomeric argininebased surfactants, no synergic antimicrobial activity is detected. Finally, the therapeutic index of the catanionic mixture, and thus its selectivity, is better than that of the gemini surfactant because it has a reduced hemolytic activity at the minimal inhibitory concentration, although it is still considered irritant. This study offers new insights into the potential advantages of environmentally friendly catanionic mixtures of green surfactants with improved surface-antimicrobial properties for biotechnological applications.

Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.