Holstein steers were fed carbohydrate-rich, short-fiber basal diets with and without added sodium sulfate. Steers fed the high-sulfate diet developed the CNS disorder polioencephalomalacia (PEM). The onset of signs of PEM was associated with increased sulfide concentration in the rumen fluid. Over the course of the disease, anaerobic rumen bacteria were enumerated in roll tubes by use of the Hungate method to determine the effect of dietary sulfate on sulfate-reducing bacterial numbers. Media used included a general type for total counts and sulfate-containing media with and without cysteine to assess sulfate-reducing bacteria. Changes in total and sulfate-reducing bacterial numbers attributable to dietary sulfate content were not observed. The capacity to generate hydrogen sulfide from sulfate in fresh rumen fluid in vitro was substantially increased only after steers had been fed the high-sulfate diet for 10 to 12 days, which coincided with the onset of signs of PEM. The low capacity for hydrogen sulfide production of rumen fluid taken at earlier times in the feeding period suggests that rumen microorganisms must adapt to higher dietary sulfate content before they are capable of generating potentially toxic concentrations of sulfide.

To study their role in sulfate reduction, anaerobic bacteria were cultured from rumen fluid samples of cattle fed high-carbohydrate, short-fiber diets with and without added sulfate. The steers fed the diet with added sulfate developed polioencephalomalacia. Microbiological methods included colony-type profiles, molybdate sensitivity, presence of desulfoviridin, sulfate reduction rates of pure and mixed cultures, and incubation time effects on sulfate reduction. Colony-type profiles indicated decreased diversity, but no relative change in numbers of sulfate-reducing bacteria in rumen fluid from cattle fed diets with and without added sulfate. Thirteen bacterial isolates were selected for further study on the basis of colony type, sulfate-reducing activity, and growth in lactate, sulfate, and yeast extract media. Seven of the isolates had Desulfovibrio-like characteristics (ie, they were gram-negative, motile rods that reduced sulfate, were inhibited by molybdate, and contained the pigment desulfoviridin). The remaining 6 isolates were gram-negative, nonmotile rods. Four of these released sulfide from cysteine, and 2 generated only limited amounts of sulfide from sulfate or cysteine. The 7 sulfate-reducing isolates generated sulfide in rumen fluid broth medium at greater rates than those observed in fresh rumen fluid. Sulfate reduction could be sustained in cultures for prolonged incubation times if the gas phase containing hydrogen sulfide was replaced at frequent intervals. Variations in the amount of sulfate reduced by the pure cultures were most pronounced at short incubation times. Sulfate reduction was not inhibited in mixed cultures of sulfate-reducing and nonsulfate-reducing bacteria.