The cyanobacterium Anabaena grows on a source of fixed nitrogen as filaments of seemingly indistinguishable cells, but when nitrogen is removed from the medium, differentiated cells, heterocysts, appear within one or two generation times at nearly regular intervals. This simple pattern provides the opportunity for a deep understanding of pattern formation in a multicellular organism. Evidence has accumulated that favors a model in which pattern evolves in two stages: (a) nonrandom initiation in groups of cells already predisposed to differentiate, leading to a crude initial pattern, and (b) resolution of differentiating groups by the action of a diffusible inhibitor, leading to a pattern of isolated heterocysts. The position of a cell in its cell cycle may be a factor that determines the ability of the cell to differentiate.
One way to test this idea is to perturb DNA replication and note how heterocyst differentiation and spacing of heterocysts are affected. To this end, three genes (dmnA, dmnB, and dmnC) from Anabaena, encoding previously uncharacterized DNA modification enzymes, have been cloned. Attempts to mutate dmnA indicate that the gene is essential for viability, but even partial inactivation alters the heterocyst spacing pattern. Mutation of dmnB yields a strain that is blocked in heterocyst differentiation at a very early stage.
In order to examine the connection between DNA modification and patterned heterocyst differentiation, two of the genes, dmnA and dmnB will be conditionally expressed in Anabaena and the resulting mutants characterized by genetic and biochemical means. Suppressor mutations of a dmnB- mutant that enable the mutant to grow on N2 will be sought, hoping to gain insight into the mechanism by which failure to modify DNA leads to blockage of differentiation. Experiments also are aimed at testing directly the connection between cell cycle and competence of a cell to differentiate by monitoring cell division and differentiation in individual cells within filaments.