The ecology, physiology, and biological control of mat -forming cyanobacteria

Kathryn Joan Wilkinson, Purdue University


Mat-forming cyanobacteria are becoming a common weed problem in the Midwest, but little is known about their distribution and physiology. A survey was conducted of Indiana ponds, and mat-forming cyanobacteria were found at nuisance levels throughout the state, with one species, Oscillatoria limosa , being the most prevalent. Seven species (Oscillatoria limosa, O. jasorvensis, O. amoena, O. deflexoides, O. pseudogeminata, Phormidium ambiguum , and Lyngbya martensiana) of cyanobacteria were isolated and subjected to six different light irradiances (0, 10, 50, 100, 500, and 700 μmol photons·m-2·S-1) and three different temperature regimes (15, 25, and 35°C). These species grow best at 25°C to 35°C and that they become light saturated at low light irradiances. These seven species (and an additional isolate of O. amoena) were then tested for sensitivity to a potential new biological control agent, bacterium SG-3. Bacterium SG-3 lysed cells of all species tested. SG-3 killed mats (algicidal activity) of O. amoena (#5) and slowed the growth and spread (algistatic activity) of mats of the seven other isolates. Differences in sensitivity may be due to differences in mat construction, which may decrease penetration of the bacteria into the mat. Control was observed at 15, 25 and 35°C and cutting mats of O. limosa into smaller pieces did not increase SG-3 activity. Data indicated that applications of SG-3 at 107 PFU·mL-1 (plaque forming units·mL-1) every 22 days should slow or prevent mat growth. The planktonic cyanobacteria Pseudanabaena limnetica and Cylindrospermopsis raciborskii, were also sensitive to treatment with SG-3. Two isolates of C. raciborskii were susceptible to low concentrations of SG-3 (EC50 = 1.3 X 10 3 PFU·mL-1), and the cyanobacterial populations tested were within population ranges normally encountered under bloom conditions. The data also indicated that slow or partial kill of a heavy bloom of C. raciborskii could be obtained by adjusting SG-3 concentrations. DNA sequence analysis (16S rRNA) showed that SG-3 is most closely related to Lysobacter brunescens (99% homology). Experiments were conducted to determine the possible mode of action of SG-3. The possibility that lytic activity is due to a bacteriophage or a secreted compound was not supported. The data suggested that like most lytic lysobacters, SG-3 must come into brief contact with the cyanobacterial cell to cause lysis and that lysis may be due to enzymes present on the surface of the bacterium. Based on observations made from these studies, SG-3 appears to be a viable candidate as an algaecide for many planktonic cyanobacteria and as an algistat for many mat-forming species.




Lembi, Purdue University.

Subject Area


Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server