Concentration of isoprene in artificial and thylakoid membranes

Abstract

Isoprene emission protects plants from a variety of abiotic stresses. It has been hypothesized to do so by partitioning into cellular membranes, particularly the thylakoid membrane. At sufficiently high concentrations, this partitioning may alter the physical properties of membranes. As much as several per cent of carbon taken up in photosynthesis is re-emitted as isoprene but the concentration of isoprene in the thylakoid membrane of rapidly emitting plants has seldom been considered. In this study, the intramembrane concentration of isoprene in phosphatidylcholine liposomes equilibrated to a physiologically relevant gas phase concentration of 20 μL L⁻¹ isoprene was less than predicted by ab initio calculations based on the octanol-water partitioning coefficient of isoprene while the concentration in thylakoid membranes was more. However, the concentration in both systems was roughly two orders of magnitude lower than previously assumed. High concentrations of isoprene (2000 μL L⁻¹ gas phase) failed to alter the viscosity of phosphatidylcholine liposomes as measured with perylene, a molecular probe of membrane structure. These results strongly suggest that the physiological concentration of isoprene within the leaves of highly emitting plants is too low to affect the dynamics of thylakoid membrane acyl lipids. It is speculated that isoprene may bind to and modulate the dynamics of thylakoid embedded proteins.

Appendixes

Appendix 1

Anisotropy was calculated using the induced optical anisotropy function:

$$ R(t)=\frac{I_{\Big\Vert }(t)-{I}_{\perp }(t)}{I_{\Big\Vert }(t)+2{I}_{\perp }(t)}, $$

(1)

where I_∥(t) and I⊥(t) are the fluorescence emission intensities parallel and perpendicular to the vertically polarized excitation pulse, respectively. The resulting anisotropy decay curves were fitted using Origin 8 software. Fit parameters for individual acquisitions were pooled for statistical analysis. For the transition dipole moment that was pumped and probed (S₁-S₀), the model for perylene reorientation as a prolate rotor is:

$$ R(t)=0.4{e}^{-6{D}_Zt} $$

(2)

The model for perylene reorientation as an oblate rotor is:

$$ R(t)=0.1{e}^{-\left(2{D}_X+6{D}_Z\right)t}+0.3{e}^{-6{D}_Xt} $$

(3)

In the model equations, D_x and D_z are the Cartesian components of the rotational diffusion constant (Jiang and Blanchard 1994). Data fitting to the prolate rotor model was by a single exponential decay function.

$$ R(t)=A{e}^{-\raisebox{1ex}{$t$}\!\left/ \!\raisebox{-1ex}{$\tau $}\right.} $$

(4)

Data fitting to the oblate rotor model was by a constrained double exponential decay function,

$$ R(t)=A{e}^{-\raisebox{1ex}{$t$}\!\left/ \!\raisebox{-1ex}{${\tau}_2$}\right.}+3A{e}^{-\raisebox{1ex}{$t$}\!\left/ \!\raisebox{-1ex}{${\tau}_1$}\right.}, $$

(5)

where constraints were “A” between 10⁻⁴ and 0.2, “τ₁” between 20 and 5000, “τ ₂” between 1 and 1000, and “τ ₁” between 1.33 and 20 multiples of “τ ₂”. Adjusted R-squared was used to compare the fit of the single and double exponential decay models to the data. For the single exponential decays that characterized the data, the Debye-Stokes-Einstein equation was used to calculate viscosity (Koan and Blanchard 2006).

$$ {\tau}_{OR}=\frac{\eta Vf}{k_BTS} $$

(6)

τ _OR is the anisotropy decay time constant, η is viscosity, V is the hydrodynamic volume of perylene, f is a solvent-solute frictional interaction coefficient, k_B is the Boltzmann constant, T is the absolute temperature, and S is a shape factor to account for the non-spherical shape of perylene. The hydrodynamic volume of perylene is 225 ų, and the shape factor is 0.7. The translational diffusion coefficient, D_T, was estimated using the Stokes-Einstein equation,

$$ {D}_T=\frac{k_BT}{6\pi R\eta }, $$

(7)

where R is the radius of perylene. R was calculated from the hydrodynamic volume of perylene, 225 ų, by assuming a spherical shape, yielding R = 3.77 Å.

Appendix 2

The concentration of isoprene in DMPC bilayers equilibrated to a gaseous standard state of 20 μL L⁻¹ was estimated from the Henry’s constant (H_I) and K_OW (K_OW,I) of isoprene. The system was assumed to be at 25 °C and 1 atm pressure. The Henry’s constant of a substance is the ratio of its concentration in water to its gas phase partial pressure, in a system at equilibrium. The K_OW of a substance is the ratio of its molar concentration in octanol to its molar concentration in water, in a system at equilibrium. At 25 °C, H_I = 7780 Pa m³ mol⁻¹ and K_OW,I = 263 (Copolovici and Niinemets 2005). The area per lipid (a_DMPC) and bilayer thickness (D_P-P,DMPC) of a DMPC bilayer were taken to be 0.63 nm² and 3.4 nm, respectively (Siwko et al. 2007).

The partial pressure of isoprene (P_I) was taken as the product of the total atmospheric pressure (P_T) and the mole fraction of isoprene in air (X_I):

$$ {P}_T{X}_I={P}_I $$

(8)

The molar concentration of isoprene in water (M_I,W) was calculated from P_I and H_I:

$$ \frac{P_I}{H_I}={M}_{I,W} $$

(9)

Multiplying the molar concentration of isoprene in water by the K_OW of isoprene yielded the molar concentration of isoprene in octanol (or in a bilayer) (M_I,O):

$$ {M}_{I,W}{K}_{OW,I}={M}_{I,O} $$

(10)

The molecular volume of DMPC in a DMPC bilayer (V_DMPC) was calculated from a_DMPC and D_P-P,DMPC:

$$ \frac{a_{DMPC}{D}_{P-P, DMPC}}{2}={V}_{DMPC} $$

(11)

The inverse of V_DMPC yielded the molar concentration of DMPC in a DMPC bilayer (M_DMPC):

$$ \frac{1}{V_{DMPC}}={M}_{DMPC} $$

(12)

Dividing the molarity of isoprene from Eq. 10 by the molarity of DMPC from Eq. 12 yielded the mole fraction of isoprene in a DMPC bilayer (X_I,DMPC):

$$ \frac{M_{I,O}}{M_{DMPC}}={X}_{I, DMPC} $$

(13)

Calculations were performed similarly for thylakoids, except that the area per lipid and membrane thickness were taken to be 0.66 nm² and 2.9 nm, respectively (van Eerden et al. 2015).