Dendritic spines are small postsynaptic protrusions on a dendrite that receive most of the excitatory synaptic input in the brain. The cytoskeleton of the dendritic spines is predominately composed of actin filaments which form the structural and func...
Dendritic spines are small postsynaptic protrusions on a dendrite that receive most of the excitatory synaptic input in the brain. The cytoskeleton of the dendritic spines is predominately composed of actin filaments which form the structural and functional network associating specialized substructures like postsynaptic density (PSD). The morphology of spines is highly variable and dynamically regulated with neuronal activity by actin-regulating proteins in PSD.
An increasing number of postsynaptic proteins such as SAPAPs and SHANKs are implicated in different forms of mood disorders such as bipolar disorders, autism spectrum disorders, obsessive-compulsive disorders, and schizophrenia although their underlying mechanisms have not been fully understood. Accumulating evidence from recent studies suggests that the structural remodeling of dendritic spines is critical for synaptic plasticity and the mechanisms regulating actin cytoskeleton may contribute to spine pathology in these neuropsychiatric disorders.
nArgBP2 was originally identified as a protein that directly interacts with SAPAP3, and. the previous study found that ArgBP2/nArgBP2 controls the balance between adhesion and motility by coordinating multiple signaling pathways converging on the actin cytoskeleton. A recent study found that genetic deletion of ArgBP2/nArgBP2 (SORBS2) in mice is known to cause behavioral phenotypes resembling human intellectual disability (ID). It has been, however, mostly unknown that how nArgBP2 deficiency leads to phenotypes observed in ID and more importantly, how nArgBP2 functions at postsynapses and its relevance to the underlying cellular and molecular mechanisms that might be related to ID.
To investigate the roles of nArgBP2 at synapses, based on the results from previous studies, I set up the following research hypotheses in my dissertation. 1) nArgBP2 is one of the key protein that regulates actin cytoskeleton at postsynapses, 2) nArgBP2 regulates the morphological changes of dendritic spines, 3) given that dendritic spines are major sites that receive most of the excitatory synaptic inputs, nArgBP2 controls the formation of excitatory synapses, 4) since excitatory-inhibitory synaptic balance(E/I balance) is the key mechanism that maintains homeostatic functional properties of nervous system, E/I imbalance caused by nArgBP2 deficiency might be the underlying factor associated with synaptic dysfunction observed in ID.
I found that the knockdown (KD) of nArgBP2 by specific shRNA resulted in a dramatic change in dendritic spine morphology. The nArgBP2 KD also impaired the formation of excitatory synapses which largely terminated at dendritic shafts instead of dendritic spine heads in spiny neurons. The aberrant formation of excitatory synapses resulted in a reduced mean frequency of miniature excitatory postsynaptic currents. I also found that the morphological changes were associated with increased WAVE1/PAK/cofilin phosphorylation, and this effect was rescued by either inhibiting PAK or activating cofilin combined to sequestration of WAVE. Using live-cell imaging technique, I confirmed that a marked increase of actin cytoskeleton dynamics resulted in a significant increase in the motility of dendritic spines in nArgBP2 KD neurons.
Surprisingly, nArgBP2 KD did not cause any morphological defect in the mature stage when the dendritic spines were stabilized. I inferred that nArgBP2 may be needed when significant structural remodeling is needed, such as developing stage. To test this idea, I decided to induce chemically-induced Long Term Potentiation (cLTP) in mature neurons. It is known to mimic many features of developing stages, including dramatic remodeling of pre-and postsynaptic structures in mature neurons. The cLTP significantly increased the size of the spine heads of control neurons while remained almost the same in nArgBP2 KD neurons. I also measured the 3D morphological features of dendritic spines under the same conditions and found that cLTP in nArgBP2 KD neurons could not induce normal head enlargement in dendritic spines. These results support my idea that nArgBP2 controls the actin cytoskeleton dynamics also in mature neurons.
Together, my research suggests that nArgBP2 functions to regulate the actin cytoskeleton dynamics in dendritic spines. It plays a particularly important role when active structural remodeling is needed, such as spine morphogenesis and subsequent spine-synapse formation in developing stages and during synaptic plasticity in mature stages. The results also raise the possibility that the aberrant regulation of synaptic actin dynamics caused by reduced nArgBP2 expression may contribute to the synaptic excitatory/inhibitory imbalance observed in ID.