Myrcene

For each aerosol system (oleic acid and myrcene–SOA), a suitable molecular rotor was selected based on hydrophobic/hydrophilic match of the aerosol to molecular rotor. For the hydrophobic oleic acid aerosol the most suitable probe was BODIPY-C₁₀, which has previously been successfully used for lipid based systems, e.g. membranes of live cells,40 ,49 (link) model lipid bilayers50 (link) and encapsulated lipid microbubbles.41 (link) The fluorescence intensity and lifetime of BODIPY-C₁₀ was demonstrated to be sensitive across a wide range of viscosities 1 to 10⁴ mPa s.50 (link) We have previously demonstrated that fluorescence lifetime is a superior marker for microviscosity, compared to fluorescence intensity, since it is not sensitive to gradients in the probe concentration.22 (link),49 (link)
The fluorescence lifetime decays of BODIPY-C₁₀ and Cy3 were measured in mixtures of methanol/glycerol and sucrose/water, at a working dye concentration of ca. ∼10 μM. The dye mixtures were measured in quartz cuvettes (BODIPY-C₁₀) and 8-well μ-Slide chamber (Ibidi) (Cy3) via time correlated single photon counting (TCSPC). BODIPY-C₁₀ decays were measured on a Jobin Yvon IBH data station (5000F, HORBIA Scientific Ltd.) using a 467 nm 1 MHz pulsed NanoLED (N-467, HORBIA Scientific Ltd.) for excitation. Emission was captured at 515 ± 5 nm with a long pass filter at 470 nm until a peak count >10 000 was reached; 1024 ADC and collection rate <2% was maintained. Cy3 decays were measured using the FLIM system (see below). The calibration plot for BODIPY-C₁₀ is linear between ca. 5 and 1500 mPa s (ref. 50 (link)) and follows the Förster Hoffmann equation for fluorescence lifetime22 (link),51 (eqn (1)). where, τ_f is the fluorescence lifetime, k_r is the radiative rate constant, η is the viscosity and z and α are constants. For BODIPY-C₁₀ in methanol/glycerol mixtures in the viscosity range 5–1500 mPa s (Fig. S2a†) this equation becomes: where, τ_f is the fluorescence lifetime of the molecular rotor in ns and η is viscosity in mPa s. The BODIPY-C₁₀ decays, at all viscosities, are monoexponential as shown in Fig. S2c.† For the myrcene SOA aerosols the Cy3 dye was chosen as the molecular rotor, due to its good aqueous solubility. Cy3 was previously successfully used to determine microviscosity in the cell cytoplasm52 and in model sucrose aerosols.23 (link)
Previously we have shown that Cy3 lifetimes do not follow the Förster Hoffmann eqn (1) at viscosities greater than 30 mPa s.23 (link) For this work, the Cy3 viscosity–fluorescence lifetime response was calibrated for the viscosity range 1 to 10⁶ mPa s (Fig. S2b†) using calibrant solutions of sucrose in water at concentrations up to 80% w/w sucrose. The time resolved fluorescence signal from Cy3 is found to be biexponential over all viscosities probed (Fig. S2d†). The mean fluorescence lifetime can be linked to viscosity through the use of a Hill function53 (link) shown in eqn (3). Where the mean fluorescence lifetime, τ_f, is the intensity-weighted mean, defined as (A₁τ₁² + A₂τ₂²)/(A₁τ₁ + A₂τ₂), where τ₁ and τ₂ are individual fitted exponential lifetimes, and A₁ and A₂ are their relative amplitudes as percentages. The relationship between the mean fluorescence lifetime data for Cy3 and viscosity is accurately described by the Hill function, eqn (3), Fig. S2b.†
The viscosity of methanol/glycerol mixtures for the entire calibration range was measured with a viscometer (Stabinger Viscometer SVM 3000, Anton Paar). The Cy3 calibration solutions, that used non-saturated sucrose concentrations (<67% w/w), were prepared by mixing increasing concentrations of sucrose in water and adding <0.5% Cy3 stock, and the viscosities were measured using a rheometer (HR03, TA Instruments). However, to achieve higher viscosities, supersaturated sucrose/water solutions were prepared, with sucrose concentrations >67% w/w; for these the direct rheological measurements were impossible due to sucrose precipitation. Instead, the theoretical model using Gènotelle's equation54 was used to predict the solution viscosity. The supersaturated solutions were prepared by controlled water evaporation, in which solutions of 40% (w/w) sucrose/water (ca. 5 ml) containing <0.2% Cy3 stock were heated at 100 °C in a round bottom flask under vacuum (150 mbar) for between 5–10 min. The water activity in each sample was determined using a Karl Fischer titrator (Mettler Toledo).55 The known water percentage in the sample allowed Gènotelle's equation to be applied, as shown in eqn (4).log₁₀ η/η* = a₁ + a₂x + Φ(b₁ + b₂xⁿ); where, η is the dynamic viscosity, x is the mole fraction of sucrose, η* is standard viscosity (1 mPa s), t is temperature (°C), Φ is calculated from the known temperature as shown above and a₁, a₂, b₁, b₂, n are constants, a₁ = –0.1245, a₂ = 22.452, b₁ = 1.095, b₂ = 46.39 and n = 1.303.54

Table 
3 described plasmids, C. defragrans strain 65Phen (wild type as well as derivatives) and E. coli strains used in this study. In course of the text, abbreviations are: i) C. defragrans 65Phen-RIF is equivalent to C. defragrans RIF; ii) C. defragrans 65Phen-RIF Δldi is equivalent to C. defragrans Δldi; iii) C. defragrans 65Phen-RIF Δldicomp is equivalent to C. defragrans Δldicomp; iv) C. defragrans 65Phen-RIF ΔgeoA is equivalent to C. defragrans ΔgeoA; v) C. defragrans 65Phen-RIF ΔgeoAcompgeoA is equivalent to C. defragrans ΔgeoAcomp.

All the dyes (Table S1) were purchased from Sigma-Aldrich (Merck KGaA, St. Louis, MI, USA). Anhydrous ethanol (≥99.5%) and ethyl acetate were purchased from Sinopharm. Trans-2-hexenal and benzaldehyde were purchased from Sigma-Aldrich. Hexyl acetate, (+)-limonene, β-myrcene, 3-carene, and isoamyl acetate were purchased from Aladdin Biochemistry Technology Co., Ltd (Shanghai, China).