Incomplete Selective Sweeps of Microcystis Population Detected by the Leader-End CRISPR Fragment Analysis in a Natural Pond

Shigeko Kimura, Mika Uehara, Daichi Morimoto, Momoko Yamanaka, Yoshihiko Sako, Takashi Yoshida

The freshwater cyanobacteriumMicrocystis aeruginosafrequently forms toxic massive blooms and exists in an arms race with its infectious phages in aquatic natural environments, and as a result, has evolved extremely diverse and elaborate antiviral defense systems, including the clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated genes (Cas) system. Here, to assessMicrocystispopulation dynamics associated with exogenous mobile genetic elements such as phages and plasmids, we examined the temporal variation in CRISPR genotypes (CTs) by analyzing spacer sequences detected in a natural pond between June and October 2013 when a cyanobacterial bloom occurred. A total of 463,954 high-quality leader-end CRISPR sequences were obtained and the sequences containing spacers were classified into 31 previously reported CTs and 68 new CTs based on the shared order of the leader-end spacers. CT19 was the most dominant genotype (32%) among the 16 most common CTs, followed by CT52 (14%) and CT58 (9%). Spacer repertoires of CT19 showed mainly two different types; CT19origin, which was identical to the CT19 spacer repertoire of previously isolated strains, and CT19new+, which contained a new spacer at the leader-end of the CRISPR region of CT19origin, which were present in almost equal abundance, accounting for up to 99.94% of CT19 sequences. Surprisingly, we observed the spacer repertoires of the second to tenth spacers of CT19originat the most leader-end of proto-genotype sequences of CT19origin. These were observed during the sampling in this study and our previous study at the same ecosystem in 2010 and 2011, suggesting these CTs persisted from 2011 to 2013 in spite of phage pressure. The leader-end variants were observed in other CT genotypes. These findings indicated an incomplete selective sweep ofMicrocystispopulations. We explained the phenomenon as follow; the abundance ofMicrocystisvaried seasonally and drastically, resulting thatMicrocystispopulations experience a bottleneck once a year, and thereby founder effects following a bottleneck mean that older CTs have an equal chance of increasing in prevalence as the CTs generated following acquisition of newer spacers.