Biofilm eradication and antifungal mechanism of action against Candida albicans of cationic dicephalic surfactants with a labile linker

Our research aims to expand the knowledge on relationships between the structure of cationic dicephalic surfactants—N,N-bis[3,3_-(dimethylamine)propyl]alkylamide dihydrochlorides and N,N-bis[3,3_-(trimethylammonio)propyl]alkylamide dibromides (alkyl: n-C9H19, n-C11H23, n-C13H27, n-C15H31)—and their antifungal mechanism of action on Candida albicans. The mentioned groups of amphiphilic substances are characterized by the presence of a weak, hydrochloride cationic center readily undergoing deprotonation, as well as a stable, strong quaternary ammonium group and alkyl chains capable of strong interactions with fungal cells. Strong fungicidal properties and the role in creation and eradication of biofilm of those compounds were discussed in our earlier works, yet their mechanism of action remained unclear. It was shown that investigated surfactants induce strong oxidative stress and cause increase in cell membrane permeability without compromising its continuity, as indicated by increased potassium ion (K+) leakage. Thus experiments carried out on the investigated opportunistic pathogen indicate that the mechanism of action of the researched surfactants is different than in the case of the majority of known surfactants. Results presented in this paper significantly broaden the understanding on multifunctional cationic surfactants and their mechanism of action, as well as suggest their possible future applications as surface coating antiadhesives, fungicides and antibiofilm agents in medicine or industry.


Surfactants. The dicephalic
N,N-bis [3,3′-(dimethylamine)propyl]alkylamide dihydrochlorides (C n (DAPACl) 2 ) and N,N-bis [3,3′-(trimethylammonio) propyl]alkylamide dibromides (C n (TAPABr) 2 ) were synthesized according to procedures described before 2,24 . The greener synthetic routes of the mentioned surfactants' intermediates 24 enabled us to use fatty acids as raw materials instead of their chlorides. Moreover, solventless synthetic route, utilizing use of solvent still-head distillation apparatus, made it possible to avoid employment of chlorocarbon solvent and extraction steps, leading to formation of harmful waste. Briefly, tetradecanoic or hexadecanoic acid (54 mmol), 3,3′-iminobis(N,N-dimethylpropylamine (68 mmol), and NaF (5 mmol) were placed in a reaction vessel, equipped with a solvent still-head distillation apparatus filled with dry Al 2 O 3 , followed by heating at 180-185 °C with continuous dry N 2 flow for 24 h. After reaction completion, residual 3,3′-iminobis(N,N- Multifunctional mechanism of action of dicephalic cationic surfactants. Multifunctional mechanism of action of investigated surfactants, taking into account the relation of abiotic surfaces with formation and eradication of C. albicans biofilm especially, as well as the structure of yeast cells and the molecular mechanism of action of the surfactants. dimethylpropylamine was removed in vacuo and the crude products were dissolved in acetone, filtered and evaporated to dryness under reduced pressure. The obtained semiproducts (40 mmol) were allowed to react with 0.5 m HCl (200 mL) at 5 °C for 5 h and the resulting products C14(DAPACl)2 and C16(DAPACl)2 were isolated by freeze-drying. Yield: 70-75%. For C14(TAPABr)2 and C16(TAPABr)2 the semiproducts (40 mmol), obtained in reaction of appropriate fatty acid with 3,3′-iminobis(N,N-dimethylpropylamine, were allowed to react with excess of bromomethane solution in anhydrous ethyl ether (400 mL) at 0 °C for 24 h. The precipitated products were filtered, washed with two portions of cold ethyl ether (2 × 25 mL) and dried in vacuo. Yield: 75-80%. The most fungicidally active surfactants were chosen to be investigated: (C 14 (TAPABr) 2 ,C 16 (TAPABr) 2 , C 14 (DAPACl) 2 , C 16 (DAPACl) 2 ), based on previous research 18 . The chemical structures of dicephalic surfactants are shown in Fig. 2. N,N-bis [3,3′-(dimethylamine)propyl]alkylamide dihydrochlorides (C n (DAPACl) 2 are pH sensitive compounds, due to presence of amine hydrochloride cationic moiety, with tendency to form free amine in basic solutions, resulting in significant drop of aqueous solubility. On the other hand N,N-bis [3,3′-(trimethylammonio)propyl]alkylamide dibromides (C n (TAPABr) 2 comprise group of relatively pH-stable cationic surfactants with quaternary ammonium salt as hydrophilic group, although undergoing degradation in strongly basic environments. The chemical structures of dicephalic surfactants are shown in Fig. 2 and extended data in Electronic Supplementary Materials (Table 1).
Strains and growth conditions. Candida albicans (ATCC 10231) was used to study the mechanism of action of the double-headed cationic surfactants. C. albicans was purchased from the American Type Culture Collection (LGC France SARL, Strasbourg, France). Yeast Peptone Glucose (YPG; 1% Difco Yeast extract, 1% Difco peptone, 2% Difco glucose) was used to cultivate the strains. Obtained cultures were centrifuged, washed with PBS (pH 7.4) and suspended in fresh YPG so suitable optical density was achieved, according to experimenter's judgment.

Minimal inhibitory and fungicidal concentration. The values of the minimal inhibitory concentration
(MIC) and minimal fungicidal concentration (MFC) were assessed according to published protocols 18 . MFC was expressed as the concentration of the dicephalic surfactant that reduced the number of colony forming units on YPG medium (CFU) by 99.9% after 24 h of incubation at 37 °C 25 . MIC and MFC of tested compounds were also measured for this work, being evaluated by dilution in liquid RMPI 1640 medium in 96 well microplates, using methodology M27-A4 of the CLSI 26 . Microscopy. Candida albicans ATCC 10231 cultures were centrifuged and diluted in PBS to OD 0.6. Surfactants were added to the cultures to the final concentration of ½ MIC, unless specified otherwise. Untreated cells were used as a control. Carl Zeiss Axio Imager M1 microscope with an AxioCam MRc5 camera was used to visualize the results. Three random fields of view per experimental condition were observed. Acquired images were processed and analyzed in the Fiji/ImageJ software (NIH). The areas from binarized images were then transferred onto original background channel (DIC or Calcofluor white MR2) and for individual fluorescent probes channels MIP images and mean fluorescence intensities of all detected objects per field of view were calculated using the ImageJ's Analyze Particles function (Software Fiji/ImageJ software ver. 1.53c).

Cell viability in biofilm (CLSM).
Aliquots of 3 ml of C. albicans ATCC 10,231 in RMPI 1640 MOPS buffered medium 10 6 CFU/ml were added to the wells of sterile 6 well plates. Sterile microscopic slides (Ø 15 mm) were put in the wells and the resultant cultures were incubated at 37 °C for 24 h with shaking (240 rpm). The slides were washed twice with sterile physiological salt solution and transferred to fresh 6 well plates. C. albicans biofilm on glass was treated with chosen surfactants: C 14 (DAPACl) 2 ; C 16 (DAPACl) 2 at 50 and 1000 μM were stained with 3 µl propidium iodide (Ex λ = 543 nm) and 3 µl SYTO 9 (Ex λ = 488) for 3 ml using LIVE/DEAD BacLight Bacterial Viability Kit (Thermo Fisher Scientific). The imaging was performed on an upright Leica SP8 resonant scanning confocal system equipped with spectral PMT detectors (Leica Microsystem). The stacks of confocal 12-bit images with pixel size of 0.455 μm and a 0.684 μm Z step were acquired using a dry 20 × objective (NA 0.75). The pinhole was set to 1 AU and line average 8 was applied. Syto9 fluorescence was excited with a 488 nm laser line and 492-526 nm emission range was recorded; PI was excited with a 552 nm laser line and 565-611 nm emission range was collected. The acquisition was performed in a sequential mode. Five random fields of view per experimental condition were imaged. Acquired images were processed and analyzed in the software Fiji/ImageJ software ver. 1.53c (NIH). First, maximum intensity projections (MIP) were obtained from stacks of images. Next, live and dead biofilm areas were established after background noise removal by thresholding and median filtering (radius 1) of respective channels. The areas from binarized images were then transferred onto original live and dead channel MIP images and mean fluorescence intensities of all detected objects per field of view were calculated using the ImageJ's Analyze Particles function.
Statistical analysis. In this work variance analysis was performed using software Statistica 13 ver. 13.3.721.0 (ANOVA analysis). Results for which p < 0.05 were treated as significant.
Ethics approval and consent to participate. This article does not contain any studies.

Results
Antifungal activity. Candida albicans ATCC 10231 was selected as an opportunistic pathogen for the study of the antifungal mechanism of action of the dicephalic surfactants investigated in this study. The sensitivity to the dicephalic surfactants-minimal inhibitory concentrations and fungicidal concentrations (MIC and MFC*) for C. albicans are as follows: C 14 (TAPABr) 2 (800; > 1000* µM); C 16 (TAPABr) 2 (400; 800* µM); C 14 18 . Aquired results for MIC and MFC were also confirmed according to M27-A4 recommendation of the CLSI, which yelded conforming results.

Microscopy.
The results of our research on the mechanism of action of dicephalic cationic surfactants using differential interference contrast microscopy (DIC), fluorescence microscopy (FM) and transmission electron microscopy (TEM) techniques are shown below.
General oxidative stress. All of the researched multifunctional cationic surfactants caused oxidative stress in C. albicans cells. In the untreated control about 2% of cells were phosphorescent. Selected quaternary ammonium salt (QAS) derivatives caused an increase in oxidative stress. A compound with a 14-carbon long hydrophobic chain caused an increase of oxidative stress by 32% compared to the control, while the surfactant with a 16-carbon alkyl chain caused an increase of oxidative stress by 18%. Dimethylamine derivates caused an increase in phosphorescence of cells to 40% in the case of C 14 (DAPACl) 2 , while C 16 (DAPACl) 2 increased it to 36%. Fluorescence microscopy allowed us to confirm the induction of oxidative stress in C. albicans cells treated with investigated surfactants (Fig. 3A,B).
Production of superoxide anion. Use of fluorescence microscopy allowed us to observe an impact of dicephalic cationic surfactants on the production of the intracellular anion radical superoxide in C. albicans. It was found that QAS derivatives caused an increase in cells showing phosphorescence to 8% C 14 (TAPABr) 2 and to 9% (C 16 (TAPABr) 2 . Surfactants which were diethylamine derivates caused a more significant accumulation of anion radical superoxide. C 14 (DAPACl) 2 caused growth in the proportion phosphorescing cells to 63%, while C 16 (DAPACl) 2 increased it to 83%. These results prove that the impact of diethylamine derivates on synthesis of anion radical superoxide is significantly stronger than those of QAS derivates (Fig. 3C,D).
Mitochondrial oxidative stress. Our research has shown that the surfactants we investigated could cause mitochondrial oxidative stress of C. albicans cells. QAS derivatives caused an increase in oxidative stress: C 14 (TAPABr) 2 increased by 32% and C 16 (TAPABr) 2 by 21% the number of phosphorescing cells. Surfactants derived from dimethylamine caused significantly stronger mitochondrial oxidative stress. In the presence of C 14 (DAPACl) 2 the number of phosphorescing cells increased by 48%, in the case of C 16 (DAPACl) 2 by 39%. Untreated cells showed no increase in phosphorescence. Diethylamine derivates induced mitochondrial oxidative stress to a further degree than QAS derivates (Fig. 3E).

Interruption of the cells membrane.
None of the tested surfactants showed a significant ability to disrupt continuity of the C. albicans cell wall. QAS derivates caused quite an insignificant increase in the number of phosphorescing cells: C 14 (TAPABr) 2 by 2% and C 16 (TAPABr) 2 by 5%. However, C n (DAPACl) 2 surfactants caused an increase in the number of cells with disrupted cell wall continuity. C 14 (DAPACl) 2 caused an increase in the number of phosphorescing cells by 21%, while C 16 (DAPACl) 2 caused an increase to 29% of C. albicans cells with disrupted membranes. No phosphorescence of control cells was observed (Fig. 3F).

Morphology of C. albicans in TEM.
Transmission electron microscopy showed that the investigated surfactants could cause changes in C. albicans cells. In control conditions the yeast cells had shown normal morphology: vacuole, nucleus, numerous mitochondria and singular lipid droplets. Yeast cells incubated with researched QAS derivates with a 14-carbon alkyl chain had shown significant thickening of the cell wall, numerous lipid droplets in the cytoplasm and a highly granular nucleus (Fig. 3G). The 16-carbon alkyl chain surfactant had a less pronounced impact on cell morphology of yeast cells; however, numerous lipid aggregates in the cytoplasm and a slight thickening of the cell wall were noticeable. Cationic multifunctional surfactants derived from diethylamine had a stronger impact than QAS derivates. C 14 (DAPACl) 2 and C 16 (DAPACl) 2 caused pronounced changes in cell morphology in treated yeast cells. Cell walls were disproportionally thickened; there were numerous lipid droplets in the cells and acute morphological changes made identification of some of the organelles impossible. Dimethylamine-derived surfactants with a longer, 16-carbon alkyl chain showed a more pronounced effect on the cells than those with shorter, 14-carbon alkyl chains (Fig. 3G).

Leakage of potassium and calcium ions.
All of the analyzed surfactants significantly increased the permeability of the C. albicans cell membrane to potassium ions, but only C 16 (DAPA)Cl 2 caused a small yet significant increase of permeability to calcium ions (P < 0.05). Permeability of the cell membrane to potassium ions was increased by about 40% by the investigated surfactants when compared to the negative control of untreated cells. In our research we did not observe a significant difference in effect, either between the groups of compounds or between compounds with different length alkyl chains, on cell membrane permeability (P > 0.05) (Fig. 4).

Biofilm eradication.
The most effective C. albicans biofilm eradication on glass surface for the concentration range 100-1000 μM was observed for dimethylamine derivatives (P < 0.05). Comparison of antibiofilm properties of cationic dicephalic surfactants shows a significant impact of the group of dimethylamine derivatives (P < 0.05). The length of the alkyl chain of a given derivative seemed to be of significance for biofilm eradication properties, as the strongest properties were exhibited by the derivate with the longest chain, 16 carbons long (P < 0.05) (Fig. 5).  properties of the investigated surfactants at a low concentration (50 μM) of dimethylamine derivates compared to the control on the glass surface. In the control young biofilm with almost exlusively living cells is visible. After treatment with dimethylamine derivates very significant eradication occurs, with many dead, yet still adhering to the surface, C. albicans cells visible, especially clearly at a concentration of 800 μM. C 14 (DAPACl) 2 caused a strong decrease in the viability of C. albicans cells, ranging from 45% (50 μM) to 1% (1000 μM) viable cells. Also surfactant C 16 (DAPACl) 2 induced a similar yet less significant decrease of viability in biofilm, reducing, C. albicans cell viability from 43% (50 μM) to 36% (1000 μM). Analysis of the surface of the biofilm has shown that the mode of eradication of biofilm depends on the length of alkyl chain of the surfactant. Compounds with C14 long alkyl chain induced stronger fungicidal effect, while those with C16 long chain eradicated already established biofilm more efficiently (Figs. 5, 6).

Discussion
The broad antibacterial and antifungal properties of dicephalic cationic surfactants allow their potential application in many fields. Due to their amphiphilic properties conferred by a double hydrophilic head and singular n-variable hydrophobic alkyl tail, these compounds demonstrate the ability to adsorb on many surfaces and to cover them with a layer. This in turn may cause a reduction in the ability of microorganisms to adhere to such surfaces 18,19,29 . Compounds exhibiting the abovementioned properties are especially valuable in the light of reported resistance of Candida biofilms to many fungicides, including those that show an inhibitory effect on planktonic forms 30 . This could significantly impact the complex deposition of surfactants on a surface and adhesion of yeast cells to said surface as well, including subsequent filamentation and biofilm formation 31,32 , as it has been established that cell surface hydrophobicity plays an important part in biofilm formation, so it stands to reason that changing the hydrophobicity of a surface may disrupt the process of biofilm formation or even induce its eradication 17 .
Previous research carried on cationic multifunctional surfactants seems to suggest that they may penetrate into the cell and cause disruptions in the cell metabolism, including induction of production of superoxide anion and resulting oxidative stress. Observed stronger induction of oxidative stress by 14-carbon derivates than 16-carbon ones could be best explained by the fact that surfactants which are less spatially significant may penetrate into the cells with greater ease 18 . Increased production of reactive oxygen species in cells of microorganisms results in a cytotoxic effect, often leading to apoptosis. Similar results were obtained in previous research, where  www.nature.com/scientificreports/ antimicrobial peptides caused oxidative stress, which in turn lead to cell membrane disruption 33 . Oxidation of cell membrane lipids and sterols could also lead to increased permeability 34 . Mitochondrial oxidative stress observed in our experiments was probably the effect of disruption of mitochondrial membranes and malfunctions of yeast electron transport chain. Increased phosphorescence results from accumulation of reactive oxygen species in endoplasmic reticulum, produced and released by mitochondria damaged by surfactants 35 or possibly by decrease in activity of mitochondrial cytochrome c oxidase (COX) 36 . It seems that high intracellular granularity that we observed correlates with severe oxidative stress, suggesting that induction of that stress, resulting in DNA damage and finally apoptosis, may be the main mechanism of action of cationic surfactants in general, especially so that none of the surfactants that we investigated was able to cause a significant induction of cell membrane perforation. Other papers have documented those cationic surfactants such as benzalkonium chloride (BAC) could cause a genotoxic effect through induction of single and double strand DNA breakage 37 . This is further demonstrated by our observations of potassium and calcium ions from surfactant treated cells. All investigated surfactants caused increased permeability of the C. albicans cell membrane to potassium ions, albeit only C 16 (DAPA)Cl 2 had such effect regarding calcium ions. This may be rather due to induction of changes in permeability, possibly through formation of pores in the membrane, rather than its complete disintegration, as none of the surfactants caused a permeability increase comparable with that induced by positive control. Similar effect was observed in previous research, where monomeric quaternary ammonium salts were able to destabilize the cell membrane without causing its perforation 38,39 . Increase in permeability may also be an indirect effect of surfactant penetrating into a cell, causing oxidative stress and inducing oxidation of lipids and sterols by the released ROS 34,40 .
These processes may in turn negatively impact the cell. Negative metabolic changes caused by penetration of surfactants into the cell, as disruptions leading to morphological changes, are shown by TEM electron microscopy: cell wall thickening, cell membrane deformations, and lipid droplets in cytoplasm or oxidative stress. The mechanism of action of the researched compounds may also be dependent on the length of the alkyl chain and the precise type of hydrophilic head of a given compound, as was previously observed for surfactants 18,41 . Among the investigated surfactants a correlation between antiadhesive and antibiofilm properties on one side and the length of the alkyl chain was observed, with the surfactants with a longer chain showing greater activity. Significantly higher activity was also demonstrated by surfactants whose hydrophilic heads were derivatives of diethylamine. It is worth noting however that the activity of researched compounds depends on the material the surface they adhere to is made of 29,42 . The observed increase in activity of the compounds concurrent with increase in the length of their alkyl chains may be a result of increasing hydrophobicity of the compounds, which in turn may lead to a stronger interaction with phospholipids of the cell membranes and subsequently a disruption of their functions. Alkyl chains of the surfactants being integrated between the phospholipids of the plasma membrane could destabilize it, impairing its functions, and due to the increase in hydrophobicity lead to alkylation of highly hydrophobic surface proteins through increased affinity to them 38,43 . Under favorable conditions C. albicans cells adhering to a surface may modify their mosaic of surface structures during the filamentation process, among others. It impacts the hydrophobicity of cell surfaces significantly, as these structures are synthesized during filamentation. This in turn greatly impacts the adhesion process and virulence of Candida cells (Fig. 7) 44 . www.nature.com/scientificreports/ Shorter alkyl chain cationic surfactants are generally characterized by weaker surface interactions and consequently exhibit weaker antifungal activity, showing a certain gradation, which confirms the relation between alkyl chain length and activity observed above, which is in agreement with previous research 7 . It is worth noting however that smaller particles may penetrate into the biofilm more easily and into the cells as well, for example through transmembrane water channels 45 . Cationic surfactants possessing a shorter chain are less lipophilic, so they have less of a predisposition to interact with cell membranes. Retardation of morphogenesis in C. albicans could be connected to disruption of polarized growth and damage of mitochondria, which seems to be confirmed by previous research in which strong oxidative stress in cells treated by surfactants was detected 46 . On a polystyrene surface a correlation between retardation of filamentation and adhesion processes and reduction of the amount of formed biofilm was observed, which was later confirmed on a glass surface 18 .
Research carried out to date suggests a positive correlation between the ability to eradicate biofilm and the length of the alkyl chain. High concentrations of surfactants could have a fungicidal effect and lead to disintegration of biofilm structure, which in the case of concentrations exceeding CMC (critical micelization concentration) could be assisted by micelization 20 . Further research carried out in this study concerning the impact of multifunctional cationic surfactants on vitality of the cells has shown its significant decrease after treatment with surfactants, especially dimethylamine derivates. It is worth noting, that our confocal microscopy observations have established a correlation not only between the effectiveness of biofilm eradication and concentration of tested surfactants, but also between said effectiveness and the length of the alkyl chain of the dimethylamine derivates. Here the positive correlation between the length of alkyl chain and ability to eradicate biofilm mentioned in the literature was confirmed for our compounds 18 . A general observation was also made, that compounds with C14 long alkyl chain induced stronger fungicidal effect, while those with C16 long chain eradicated already established biofilm more efficiently, regardless of the nature of their respective cationic heads. Figure 7. Molecular mechanism of action of dicephalic cationic surfactants on C. albicans cells, which takes into consideration domain structure of yeast cells, morphological changes (filamentation) of yeast-like cells to pseudohyphal and hyphal cells as well as ionic and hydrophobic interactions. Yeast cell at the first stage of adhesion starts to synthesize increased amounts of EPS (including exopolysaccharides), which determine the increase in negative surface charge of the cell, together with other cell structures. It results in the ability of double-headed cationic surfactants with their double positive charges to bind to negatively charged cell structures, orienting their hydrophilic heads towards the cell surface. Then the surfactants could interact with other areas of the yeast cell, including more hydrophobic ones, due to increased expression of proteins, which allows the surfactants to penetrate into the cell, change orientation and interact with hydrophobic molecules and cell membranes with their n-variable alkyl chains. Penetration of surfactants into the cell may cause impairment of the mitochondrial process, marked "M" in the picture, and severe oxidative stress through high production of reactive oxygen species (ROS), which in turn could cause further intracellular damage, including damage to the endoplasmic reticulum (ER) and increased synthesis of lipid droplets (LD) as well as damage to cell membranes. The multimodal mechanism of action of dicephalic surfactants may lead to a wide spectrum of intracellular damage in yeast cells, which in turn could lead to disruption of the filamentation process and biofilm formation, or lead to its significant eradication.

Conclusions
The research presented here enabled us to understand the biological activity of de novo synthetized cationic multifunctional surfactants, derivates of tertiary ammonia salts and dimethylamine. The investigated compounds varied in the length of their alkyl chain, which made it possible to correlate their chemical structure and mechanism of action, taking into account various models of surfaces with attention paid especially to the glass surface. Dimethylamine derived surfactants investigated in this study may be used as effective surface covering agents, limiting the ability of C. albicans cells to adhere to such surfaces, significantly reducing their ability to form biofilm and in higher concentrations even causing its eradication. Our research implies that the molecular mechanism of action of the investigated surfactant towards the opportunistic pathogen C. albicans is based on synergistic action of oxidative stress induction and disturbance of cell membranes and lipid droplet accumulation. Due to their strong antifungal activity, the surfactants which were investigated in this study could therefore be applied as effective agents eradicated biofilm (surface active agents; disinfectants).

Data availability
The corresponding authors will make the data available upon request.