Examining the Evidence for Regulated and Programmed Cell Death in Cyanobacteria. How Significant Are Different Forms of Cell Death in Cyanobacteria Population Dynamics?

Cyanobacteria are ancient and versatile members of almost all aquatic food webs. In freshwater ecosystems some cyanobacteria form "bloom" populations containing potent toxins and such blooms are therefore a key focus of study. Bloom populations can be ephemeral, with rapid population declines possible, though the factors causing such declines are generally poorly understood. Cell death could be a significant factor linked to population decline. Broadly, three forms of cell death are currently recognized - accidental, regulated and programmed - and efforts are underway to identify these and standardize the use of cell death terminology, guided by work on better-studied cells. For cyanobacteria, the study of such differing forms of cell death has received little attention, and classifying cell death across the group, and within complex natural populations, is therefore hard and experimentally difficult. The population dynamics of photosynthetic microbes have, in the past, been principally explained through reference to abiotic ("bottom-up") factors. However, it has become clearer that in general, only a partial linkage exists between abiotic conditions and cyanobacteria population fluctuations in many situations. Instead, a range of biotic interactions both within and between cyanobacteria, and their competitors, pathogens and consumers, can be seen as the major drivers of the observed population fluctuations. Whilst some evolutionary processes may theoretically account for the existence of an intrinsic form of cell death in cyanobacteria, a range of biotic interactions are also likely to frequently cause the ecological incidence of cell death. New theoretical models and single-cell techniques are being developed to illuminate this area. The importance of such work is underlined by both (a) predictions of increasing cyanobacteria dominance due to anthropogenic factors and (b) the realization that influential ecosystem modeling work includes mortality terms with scant foundation, even though such terms can have a very large impact on model predictions. These ideas are explored and a prioritization of research needs is proposed.