Metabolism of glycerol by an acrolein-forming lactobacillus.

Mills et al. (1954) and Serjak et al. (1954) have reported the production of acrolein in distillery grain mashes and identified the causative organisms as lactobacilli. In addition to the lactobacilli described by these investigators, bacteria representing at least four other genera have been observed to produce acrolein: Bacillus amaracrylus by Voisenet (1918) (see B. polymyxa in Bergey's Manual of Determinative Bacteriology 6th edition, p. 720, 1948); Clostridium perfringens by Humphreys (1924), Warcollier and Le Moal (1932), Warcollier et al. (1934); Escherichia (Citrobacter) freundii by Mickelson and Werkman (1940); and an aerobacter by Reynolds et al. (1939). Acrolein is formed only when glycerol is provided as a substrate and, then, only in minor amounts. Little work has been directed toward the mechanism of acrolein formation although the dehydration of glycerol in aqueous media poses intriguing biochemical questions. Both Mills et al. (1954) and Serjak et al. (1954) speculated that glycerol was attacked by acroleinproducing lactobacilli only after fermentable carbohydrate was largely utilized, i.e., that sugar has a sparing action on glycerol. This view was supported by the observations that acrolein is not formed in the presence of high levels of glucose and that glycerol is not actively fermented as a sole substrate. Voisenet (1914) studied the formation of acrolein from glycerol by B. amaracrylus and concluded that glycerol was first dehydrated to 3-hydroxypropionaldehyde. He then assumed a second dehydration to acrolein. He also noted that fermentation liquors contained trimethylene glycol which he assumed was formed from f-hydroxypropionaldehyde strictly by chemical reaction. Several workers have noted that acrolein is occasionally formed when Escherichia freundii ferments glycerol to trimethylene glycol. Mickelson and Werkman (1940) found the amount of acrolein formed could be increased by including CaHSO3 in the medium. However, attempts to show that acrolein is a precursor of trimethylene glycol ended in failure. They concluded that it is not a precursor, but rather is formed from some minor side reaction. Otsuka (1958) postulated that Bacterium succinicum reduces glycerol to trimethylene glycol which is then oxidized to 3-hydroxypropionaldehyde, and, in turn dehydrated to acrolein. According to Bergey's Manual of Determinative Bacteriology this organism belongs to the genus Escherichia, and since it produces trimethylene glycol from glycerol it probably should be classified as E. freundii. Studies on the kinetics of acrolein hydration by Pressman and Lucas (1942) revealed that the rate of hydration is slow in aqueous solution at room temperatures but is greatly accelerated in the presence of 0.5 M perchloric acid. The equilibrium constant for hydration was found to be about 0.03 compared with 0.003 for dehydration. When acidified solutions of acrolein were heated to 100 C, equilibrium with the hydrated compound was reached in approximately 5 min. The hydroxyaldehyde was found to be quite stable in acidic solution at room temperature. Since distillation is commonly employed to separate acrolein prior to its quantitative determination, it seems plausible that acrolein found in fermentation liquors may actually have existed (before distillation) as,-hydroxypropionaldehyde, the acrolein precursor suggested by Voisenet (1914). This study is concerned with the identification of the products of glycerol dissimilation by an organism found by Serjak et al. (1954) to produce acrolein. A metabolic pathway of glycerol metabolism is proposed. The results obtained tend to disprove that acrolein is a metabolic product of glycerol fermentation.