mTORC2 controls neuron size and Purkinje cell morphology independent of mTORC1

Prenatal brain development is mainly accomplished by extensive proliferation of neuronal precursor cells whereas postnatal brain growth in mammals is mainly mediated by the growth of those post-mitotic nerve cells. The neuron size and the branching pattern of the dendritic tree are highly controlled during development to enable the proper connectivity of neuronal circuits and the accurate electrical transmission in the adult which is a prerequisite for the brain to function normally. Aberrations in size, morphology or connectivity have been shown to be the cause for various brain disorders. Neuron size and dendrite development are controlled by intrinsic mechanisms, trophic factors and neuronal activity, processes that need the concerted action of a plethora of signaling molecules. A central integrator of various signaling cascades is the mammalian target of rapamycin (mTOR) and as such it contributes to brain development and function and is thus also implicated in the pathophysiology of psychiatric disorders.
mTOR is a serine threonine protein kinase that is highly conserved from yeast to humans and has been found to be part of at least two multi-protein complexes mTORC1 and mTORC2. The formation of mTORC1 is dependent on the protein raptor whereas mTORC2 assembly relies on the protein rictor. In recent years a complex picture about the function of mTORC1 has emerged by use of rapamycin, an immunosuppressive drug that acutely inhibits mTORC1 formation and activity and has attributed mTORC1 a major role in the regulation of cell size and proliferation. However, because the activity of mTORC2 is only depleted upon long term application of rapamycin, research advancement on its function was thus far impeded. Due to the early embryonic lethality of raptor or rictor knockout in mammals conditional knockout models were constructed. Whereas tissue specific knockout of raptor led to characteristic alterations, knockout of rictor in several organs such as skeletal muscle and adipose tissue provided none or only a weak phenotype. Several cell culture studies assigned mTORC2 a role in cytoskeletal modifications but in vivo confirmation is still lacking. The current knowledge about mTORC2 is restricted to the downstream targets Akt/PKB (proteinkinase B) and PKC (protein kinase C) which belong to the AGC kinase family. Those kinases are reported to influence cell morphology, growth and survival and are also essential regulators of brain development and function. PKCs are involved in synaptic plasticity and neurotransmitter release and, hence, also in the pathophysiological mechanisms of psychiatric disorders especially in schizophrenia and bipolar disorder. Concordantly, several psychiatric agents have been shown to alter PKC signaling. This emphasizes the urge to analyze the role of mTORC2 in the central nervous system.
In this dissertation the role of mTORC2 was analyzed in the central nervous system and in specific sub-populations of neurons by deletion of rictor. I discovered, that in contrast to all other organs analyzed so far, rictor knockout in the brain reveals a pronounced phenotype. The brain-size of those mice shows an enormous reduction to almost half of that of control mice which is caused mainly by the reduction of neuron size. The reduced cell size is observed in neurons derived from different brain areas in vitro and in vivo but is most prominent in Purkinje cells of the cerebellum, the cell type with highest rictor expression. In addition, dendrite morphology is majorly disrupted and the formation of dendritic spines is affected which correlates with a decreased neuronal activity. The Purkinje cell phenotype can also be reproduced in a Purkinje cell specific knockout of rictor and thus demonstrates that the effect of rictor deletion in neurons is cell autonomous. Moreover, Purkinje cell axonal path-finding is affected which correlates with the decrease in phosphorylation of the neuron specific PKC target protein GAP-43, a known regulator for axon growth and path-finding. Molecular analysis reveals that rictor is essential for the activity of all conventional PKC isoforms and the novel PKCε in vivo and in vitro in neurons which influences the function of downstream targets important for cytoskeleton modifications such as GAP-43, MARCKs and neurofascin. In addition, rictor controls the phosphorylation of Akt but does not alter mTORC1 signaling towards its downstream effectors. In summary it becomes clear that rictor is important in the development and maturation of neurons and controls their size and neuron structure which influences the entire brain function and affects the behavior of the mice. Thus, these data encompass a new role of rictor in CNS disorders.