Search for a command to run...
The conversion of lignocellulosic biomass to biofuels and bioproducts is limited by small inhibitory molecules generated during the pretreatment process. The diverse compounds within the cell wall hydrolysate include aldehydes, carboxylic acids, phenolics and alcohols inhibit microbial growth by disrupting microbial membranes. Although Zymomonas mobilis combines high ethanol productivity with hopanoid rich membranes that confer solvent tolerance, hydrolysate compounds still inhibit microbial growth. One proposed mitigation strategy is to pump these molecules out through active transport processes, reducing the direct toxicity. However, if passive permeation rates are high, active transport would create a futile cycle where the exported molecules diffuse back in, creating a net drag on fitness. To investigate this hypothesis, we measure the passive permeation for thirty-three lignocellulose derived compounds using atomistic molecular dynamics simulations with enhanced sampling. Permeability coefficients determined spanned more than eight orders of magnitude driven largely by the balance between hydrophobicity and polarity. Hydrophobic inhibitors such as aromatic acids, aldehydes and short chain alcohols readily partitioned into the membrane and showed high permeability consistent with rapid passive influx and strong toxicity. Most polar molecules have lower but still appreciable permeation rates, with only the most polar sugars having low permeabilities. By integrating these predictions together with hydrolysate composition information, we find that most of these molecules are so permeable that futile cycles where passive permeation overwhelms efflux is highly likely. These findings provide a predictive framework for strain engineering, highlighting that lowering hydrolysate inhibitor concentrations through dilution may be the most effective strategy to productivity in industrial biorefineries.