Osmolyte protection of Escherichia coli and Salmonella enterica against inactivation by acetic acid

viii The antibacterial activity of acetic acid is usually explained by weak acid theory, involving diffusion of the undissociated moiety followed by dissociation and cytoplasm acidification, and acetate anion accumulation and toxicity. These proposed mechanisms are ultimately mediated at the level of the cell envelope and, in particular, the cytoplasmic membrane. The most obvious changes due to osmotic stress affect the structure and composition of the cell envelope, suggesting a possible mechanism of protection against acetic acid inactivation, i.e., that nonmonotonic inactivation in response to increasing osmolarity in the presence of acetic acid arises from damage to, or changes in, the cell envelope. Improved survival at ‘intermediate’ (hypertonic) osmolarities could arise from: 1) one type of damage increasing with osmolarity above or below the optima, or 2) one type of damage increasing with increasing osmolarity and a second type of damage increasing with decreasing (hypotonic) osmolarity. Studies of E. coli and S. enterica recovery in the presence of bile salts and crystal violet suggested that damage to the outer membrane in the presence of acetic, but not hydrochloric, acid is non-monotonic with increasing osmolarity. Flow cytometry and Three Dimensional Structured Illumination Microscopy (3D-SIM) were used to assess membrane changes in the total population of E. coli and S. enterica. Substantial outer membrane damage by acetic acid was confirmed in this manner using the fluorescent dyes, hexidium iodide and SYTO 9, with the latter changing non-monotonically with increasing osmolarity. Outer membrane damage to E. coli and S. enterica by acetic acid has not previously been reported. Outer membrane damage to E. coli was also observed to be substantially slowed by storage at 5°C, correlating with improved survival. Substantial loss of cytoplasmic membrane integrity, as assessed by flow cytometry with propidium iodide, was not observed. However 3D-SIM showed the development of distinct, brightly SYTO 9stained membrane domains in response to increasing exposure time, osmolarity and acidity, suggesting that changes in cytoplasmic membrane structure, if not integrity, did occur. Using 3DSIM and staining with nonyl acridine orange (NAO), it was shown that the membrane domains were enriched with cardiolipin. Cardiolipin has previously been shown to be involved in the response of E. coli to osmotic stress, but production in response to acid, and acetic acid, stress