Photomodulation of bacterial growth and biofilm formation using carbohydrate-based surfactants

The photocontrollable antibacterial and biofilm modulatory activity of a panel of light responsive carbohydrate-based surfactants is reported.

Deprotected carbohydrate-based surfactants were purified by reversed-phase (C18) preparative HPLC using an Agilent 1260 preparative HPLC system equipped with an automated fraction collector. Separations were performed on an Agilent Zorbax SB300 5 μm (20 mm x 150 mm) C18 column using a linear gradient of 0.1% formic acid in water (Solvent A) and 0.1% formic acid in acetonitrile (Solvent B) as mobile phase, operating at a flow rate of 10 mL/min. Carbohydrate-based surfactants were purified using a linear gradient of 20% solvent B to 100% solvent B over 40 minutes (monitoring at 280 nm).Purified fractions were subsequently combined and lyophilized.

S4
Nutrient Broth (NB), Brain Heart Infusion (BHI) broth, Tryptic Soy Broth (TSB), tryptone, yeast extract, Mueller-Hinton Broth (MHB) and agar were purchased from Amyl Media PTY LTD (NB); Becton, Dickinson and Company (BHI); Oxoid PTY LTD (TSB, tryptone, yeast extract and MHB) and Sigma-Aldrich (agar). All media components were used as received. Escherichia coli (DH5α) and methicillin-resistant Staphylococcus aureus (MRSA ATCC 1698) were procured from Southern Biologicals and maintained on Nutrient agar plates, as per standard microbiological protocol. Two multi-drug resistant Pseudomonas aeruginosa strains (MDR283/1-6 and MDR283/1-23), previously isolated from the clinical samples were maintained on NA plates and used for biofilm assays and swarming motility assays. Crystal violet used in biofilm assays was purchasedfrom E. Merck (Darmstadt, Germany). Glucose used in swarming motility assays was from Sigma.
Photoswitching experiments were performed using a UV light source which was comprised of four 9 Watt halogen tubes, delivering a total power of 36 W at λmax of 361 nm at the source (Nail curing lamp, eBay). The working distance between the source and the centre of the samples for the photo-modulated biological studies was fixed at 4.5 cm. UV-vis spectra were recorded in 96-well format using a Perkin Elmer EnVision TM 2104 Multilabel Plate Reader (instrument linear operating range 0-4 OD; see antibacterial assay section for more details).

UV-Vis stability studies on cis isomers
The native trans isomers of carbohydrate surfactants were converted into cis states by illumination of their aqueous solutions under ambient conditions using a UV lamp with λmax at 361 nm in a time-dependant manner (1-90 min). One minute of photoexcitation was found sufficient to convert trans isomers into cis state. All the studies involving bacteria utilised cis isomers obtained after 5 min of illumination with λ361nm. Since bacterial experiments were performed at 37 °C, the thermal stability of both the native trans isomer as well as the photoinduced cis isomer was studied at 37 °C using UVvisible spectroscopy. Azobenzene trans-cis photoisomerisation was found to be complete within 1 minute of UV irradiation. The native trans isomers remained stable under ambient lighting conditions at least up to 24 h, as evident from insignificant changes in their UV-vis absorbance spectra over this period. The thermal relaxation S5 studies were performed at 37 °C for 24 h in deionised water as well as in the growth media employed for the antibacterial studies against E. coli (NB) and S. aureus (BHI).
The photoinduced cis isomers were found to thermally relax back to their native configuration within 24 h at 37 °C in deionized water, NB and BHI media and the data is summarized in Table S1. The half-lives of the cis isomers in water and two bacterial media were estimated from thermal relaxation studies by plotting the ratios of peak area under 350 nm peak and that under 440 nm peak over a period of 24 h. These peaks were chosen as cis-trans relaxation results in increase in peak intensity at ~350 nm, with a corresponding decrease in peak intensity at 440 nm.      Table S1. S10 Figure S5. Ratios of absorbance of surfactants (0.1 mM) at 350 and 440 nm during timedependent thermal relaxation of photoexcited cis isomers to their native trans isomers at 37 °C in brain heart infusion broth (BHI). The half-lives of cis isomers calculated from these graphs in BHI at 37 °C are indicated in Table S1.

Surface tension measurements
Surface tension measurements were made on a custom-designed pendant drop instrument. 1 A time series was taken (that is, surface tension as a function of time) and values were noted for 100 s to ensure full equilibration of interfacial adsorption. Once a stable surface tension had been attained, this was recorded. Drop volumes were measured throughout and changes of <5% throughout the course of a measurement were a requirement for the data shown. Critical micelle concentration (CMC) values were obtained from the intersection of lines extrapolated from surface tension values in the near pre-and post-CMC regions. Surface tension data for surfactants not shown in the main paper is provided in Figure S6. S12 Figure S6. Photocontrollable surface tension data for photosurfactants. The surface tension data for AzoGlc has been published previously. 1

Small angle neutron scattering (SANS)
SANS measurements were made on the Quokka instrument at the Bragg Institute, ANSTO, Lucas Heights NSW, Australia. For all samples, raw scattering counts were collected on a 128 x 128 element area detector, where the sample-detector distances used were 2 m and 14 m, with no detector off-set. An incident neutron wavelength of 5 Å was used with a typical spread of 10%, thus giving an effective q-range of 0.004 -0.4 Å -1 . Samples were prepared in circular 12.5 mm Hellma quartz cells with a path-length of 2 mm, and a thermostatically-controlled automatic sample changer ensured that a temperature of 25 +/-0.05 o C was maintained. Data were converted from raw counts at the detector into 1D scattering spectra by first subtracting the scattering from an empty cell and then radially averaging the resulting spectrum, normalising for the measured sample transmission. A D2O background was then subtracted from the final 1D sample data to ensure that the scattering signals seen are from the surfactant only. All samples were run at a fixed concentration of 4 mM, well above the CMC for each.
Fitting was performed using standard equations for ellipsoidal and cylindrical form factors as described previously. 1,2 Experimental and fitted SANS spectra are shown in Table S2.

Antibacterial assays
The antibacterial activity of these compounds was tested against Gram-negative E. coli, Grampositive S. aureus and Gram-negative P. aeruginosa using an OD600 method. 3

Minimum inhibitory concentration (MIC) test
The antibacterial activity of AzoTAB and non-surface active AzoTEG were tested against

Biofilm assays
The photoswitchable surfactants and control compounds were screened for their influence on components. The plates were then kept at 55 °C for 1 h to rigidly fix the biofilms to the plates, followed by staining the biofilms with 125 µL crystal violet (0.1% w/v in water) at room temperature for 10 min. The plates were then exposed to running tap water until free crystal violet stops releasing from the biofilms. The biofilms were then dried overnight in air, followed S20 by addition of 125 µL of acetic acid (30% v/v in water) to each well to solubilize the crystal violet taken up by the bacterial biofilms. The solubilized crystal violet was monitored at OD550 using a Perkin Elmer EnVision TM 2104 Multilabel Plate Reader. All biofilm assays were performed in triplicates; tests were repeated independently three times; each well was read five times. The average of 45 readings (3 x 3 x 5) in each case was calculated and plotted along with standard deviation. The data was background corrected by subtracting the OD550 value obtained from the blank control from all other samples ( Figure S10 and S11). Figure S10. Concentration-dependent influence of trans and cis photoisomers of surfactants on biofilm forming ability of S. aureus at 37 °C after 24 h of exposure. Figure S11. Concentration-dependent influence of trans and cis photoisomers of surfactants on biofilm forming ability of P. aeruginosa at 37 °C after 24 h of exposure.

Bacteria preparation for differential dynamic microscopy (DDM)
Overnight cultures of P. aeruginosa were grown in Luria-Bertani (LB) broth (10% tryptone, 5% yeast extract and 5% NaCl) using a shaking incubator at 37 °C and shaking speed of 120 rpm. A fresh culture was inoculated as 1:100 dilution of overnight grown cells in tryptone broth (TB) and grown for further 4 h to obtain actively growing late exponential phase bacteria.
Bacterial suspensions were gently mixed with the surfactant solutions before use.

Swarming motility assays
Compounds were tested for their influence on swarming motility of Gram-negative P.
aeruginosa. Swarm plate (0.8% nutrient broth, 0.5% glucose, 0.5% agar) was air dried for 5-10 min before use. The compounds were dissolved in MilliQ deionised water and then added to each agar medium at a final concentration of 1 to 100 µg/mL. An equal amount of deionised water without any compound was added to the agar medium to serve as a control. A fixed concentration of overnight grown bacterial culture (2 µL) was inoculated at the centre of the agar surface, and the plates were incubated at 30 °C for 22 h. Figure S13. Influence of the increasing concentration of trans photo-isomer of photosurfactants and controls on swarming motility of P. aeruginosa MDR283/1-6, a swarming-positive strain.

Synthesis and analytical data
Carbohydrate surfactants 1-6 were synthesized as previously described (Scheme S1). 1 The reaction mixture was extracted into ethyl acetate (10 mL), and washed with water (10 mL) and brine (10 mL). The organic layer was dried (Na2SO4), filtered and evaporated to dryness. completion. The reaction was neutralized with Amberlite IR-120 acidic ion exchange resin, filtered and concentrated under reduced pressure. Surfactants 1-6 were subsequently purified by reversed-phase preparative HPLC and isolated as deep yellow solids following lyophilization (21-67% yields).