Neurons come in a bewildering variety of shapes and sizes, with the structure of each type underlying its specific functions. A major preoccupation of developmental biologists is to understand how each neuronal type acquires these defining features. In their study, Peng, James and colleagues took the tack of comparing two neuronal types that are similar in most respects but differ in a key feature. They reasoned that molecular difference between the two would also be few in number, making it easier to link molecules to structure and function than is the case when the cells being compared differ in myriad ways. In fact, several such pairs exist. Among them are pairs of retinal neurons called paramorphic, in which one member responds to increases in illumination (called ON responses) and the other to decreases (OFF responses). Peng et al. studied one such pair, the starburst amacrine cells. Because inputs that endow retinal neurons with ON and OFF responses are restricted to the inner and outer portions of a region called the inner plexiform layer, the key structural difference between ON and OFF starbursts is their position with respect to this layer: they sit on the inner and outer margins of the layer, respectively, giving them preferred access to the proper inputs. Other than that, it has been close to impossible to tell the ON and OFF starbursts apart.
To find molecular differences between ON and OFF starbursts, Peng et al. used a method called scRNA-seq to identify the genes expressed by large numbers of individual starbursts. They were able to divide the cells into two groups based on differences in gene expression, and to establish that they corresponded to the ON and OFF types. They then looked for transcription factors present in only one type, reasoning that genes in this class control developmental programs. They found one, called Fezf1, that is exoressed in starbursts as soon as they appear, but in only half of them: those destined to become ON but not OFF starbursts. They then went on to show that Fezf1 controls the fate switch that the difference between ON and OFF types: removing it from ON starbursts turns them into OFF starbursts and putting it into OFF starbursts makes them more like ON starbursts. They were then able to return to their scRNA-seq data and find another gene, called Rnd3 that is part of the program controlled by Fezf1 and helps the ON and OFF starburst migrate to their distinct positions on either side of the plexiform layer. Thus, Peng and colleagues were able to find a “needle in the haystack” allowing them to begin mapping out a genetic program that controls diversification of a paramorphic neuronal pair. They are hopeful that this approach can be applied to other cell pairs, providing a way to efficiently hone in on key genes that determine neuronal structure and function.