Date of Award


Degree Type


Degree Name

Master of Science (MS)

Author's Department


First Advisor

Terry M. Weidner


Interstrain competition of Bradyrhizobium japonicum for nodulation of soybean (Glycine max) presents a serious problem to the field commercial inoculation practice. The more efficient, introduced strains are often outcompeted by the less efficient, indigenous strains. This study was conducted to determine if the competition could be effectively controlled by the massive introduction of a more efficient strain and/or the large scale reduction of a indigenous strain by the application of a bacteriophage specific to the indigenous one.

B. japonicum strain 61A123, common in Midwestern U.S. soils, and strain USDA110, representative of the strains used in commercial inocula, were chosen for the nodulation competition studies. A bacteriophage (P135pl) specific to strain 123 was used as a biological control agent. The experiments were in a split-plot, factorial design for factors of phage inoculating ratio (PIR) to strain 123 and the percentage of strain 110 in inocula. Soybeans were grown in a sterilized sand-vermiculite mixture under controlled conditions. The numbers of the bacteria and the phage used in inocula were determined by routine plate counts and a standard curve. Strain identification and the determination of the number of bacteria from nodule samples were performed by enzyme-linked immunosorbent assays (ELISA) and by developing standard curves in ELISA, which showed an exponential relationship between the number of the bacteria and the absorbance at 405 nm, measured in ELISA. Nodulation outcome was expressed by apparent nodule occupancies, defined as percentages of the nodules per plant occupied by a certain strain, and by actual nodule occupancies, defined as percentages of the number of bacteria of a certain strain in nodules per plant.

The nodulation competition experiments showed significant increases of the apparent or actual nodule occupancies of strain 110, and significant decreases of the apparent or actual nodule occupancies of strain 123, by the strain 110 input and the phage input at different levels in inocula (P<0.01). Generally, the strain 110 occupancies were increased and the strain 123 occupancies were decreased as the strain 110 percentage in inocula or the number of phage in inocula was increased. Even though the mean occupancies of a few of the treatments were not found to be statistically different, the actual nodule occupancies appear to be a more sensitive indicator than the apparent nodule occupancies. No significant combined effects of the phage input and the strain 110 input in inocula in changing the nodule occupancies of the either strain were found.

A significant linear relationship was found between the actual nodule occupancy and the strain 110 percentage in inocula (P<0.01). The elevations of the regression lines were regarded as the competitive indexes. They are representative of the nodulating competitiveness of a particular strain. It is shown that strain 110 competitive ability was increased and strain 123 competitive ability was decreased as the phage input in inocula was increased. The competitive indexes of each strain were all statistically different from one another at all PIRs. The relationship between the competitive index and the PIR was found to fit a semilogarithmic model (Y=A+B*LOG(X)). The A coefficient was proposed to be the phage efficiency index.

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Included in

Microbiology Commons