Dramatic increases in number of cerebellar granule-cell-Purkinje-cell synapses across several mammals

The classical comparative literature on mammalian brain evolution has mainly focused on brain mass measurements because larger brains are more likely to have more neurons to process information. The phylogenetic expansion in the mass of the cerebellum is as significant as that of the cerebral cortex. The synapse, however, has recently been recognized as the basic unit of neuronal information processing, including neuroplasticity. Here we hypothesize significant absolute and relative increases in the functionally important granule-cell-Purkinje-cell (gcPc) synapses as a salient feature of the evolving cerebellum. To probe evolutionary constraints, we define the gcPc circuitry with ten degrees of freedom, including number of granule cells, Purkinje cells, lengths of the granule cell axonal segments, linear densities of synapses along them, and physical dimensions of Purkinje as well as granule cell dendritic structures. We show that although only two of the ten parameters are not constrained and therefore can exhibit independent, comparative changes, there is a dramatic increase in the number of gcPc synapses from the rodent to the human cerebellum. By assigning a value of unity for the mouse, the ratio of the number of gcPc synapses from mouse, rat, cat, non-human primate, and human is 1:5.5:236:620:20,000, which greatly exceeds the ratio of increase in cerebellar mass (1:6:48:180:3000). Dramatic changes in the number of gcPc synapses can therefore occur despite evolutionary constraints and only modest changes in parameters of the neuronal circuitry. Increases in the number of gcPc synapses have important functional consequences as these synapses enhance the capacity of the cerebellum to code and process information, which directly impact memory and learning in both motor and non-motor tasks.