Study of a microbial commensalism

TR Number



Journal Title

Journal ISSN

Volume Title


Virginia Polytechnic Institute


Saccharomyces cerevisiae V. P. I. Strain No. 1 was found to estimate the growth of Proteus vulgaris V. P. I. Strain No. 3 in a chemically-defined medium when growth curves of the viable counts of the two organisms in pure and in mixed culture were studied. The yeast stimulated the growth of the bacterium without its own numbers being affected, indicating a true commensalism. The stimulation began by the 8th hour of incubation and continued throughout the growth period. The organisms were plated on differential plating media designed to specifically permit the growth of one organism while inhibiting the other. The medium used for P. vulgaris contained bile salts to inhibit the yeast, while the medium used for S. cerevisiae contained penicillin to which the bacterium was sensitive.

During the phase of active growth of the yeast, a temporary decrease or dip in viable count occurred to the extent of about 1/2 to 1 log unit, regardless of whether the yeast was in pure of mixed culture. Total counts of the yeast cells (by direct microscopic methods) failed to show any corresponding decrease. A rough correspondence of the appearance of amounts of ethanol detectable by gas chromatography in the medium was observed; however, the factors causing the viable count dip remain obscure. It is possible that clumping of yeast cells is closely connected with this dip.

Separation of the yeast and bacterium by a dialysis membrane failed to halt the stimulatory action of the yeast. The ability of P. vulgaris to exhibit a growth response toward nicotinic acid, nicotinamide, or nicotinamide adenine dinucleotide was demonstrated, suggesting that the stimulatory factor formed by the yeast was a compound of this group. The amount of stimulatory factor formed by the yeast in pure culture after 48 hr corresponded to 0.20 µg of nicotinamide adenine dinucleotide/ml. Exhaustive attempts to identify the stimulatory factor by bioautography were unsuccessful; the amount of factor was too slight to be detectable by even this highly sensitive method. Evidence is presented to indicate that the high amount of biotin employed in the defined medium may have resulted in production of only small quantities of nicotinic acid-like factor in comparison to the amounts formed when less biotin was present.