Peptide regulation of cofilin activity in the CNS: A novel therapeutic approach for treatment of multiple neurological disorders

Abstract

Cofilin is a ubiquitous protein which cooperates with many other actin-binding proteins in regulating actin dynamics. Cofilin has essential functions in nervous system development including neuritogenesis, neurite elongation, growth cone pathfinding, dendritic spine formation, and the regulation of neurotransmission and spine function, components of synaptic plasticity essential for learning and memory. Cofilin's phosphoregulation is a downstream target of many transmembrane signaling processes, and its misregulation in neurons has been linked in rodent models to many different neurodegenerative and neurological disorders including Alzheimer disease (AD), aggression due to neonatal isolation, autism, manic/bipolar disorder, and sleep deprivation. Cognitive and behavioral deficits of these rodent models have been largely abrogated by modulation of cofilin activity using viral-mediated, genetic, and/or small molecule or peptide therapeutic approaches. Neuropathic pain in rats from sciatic nerve compression has also been reduced by modulating the cofilin pathway within neurons of the dorsal root ganglia. Neuroinflammation, which occurs following cerebral ischemia/reperfusion, but which also accompanies many other neurodegenerative syndromes, is markedly reduced by peptides targeting specific chemokine receptors, which also modulate cofilin activity. Thus, peptide therapeutics offer potential for cost-effective treatment of a wide variety of neurological disorders. Here we discuss some recent results from rodent models using therapeutic peptides with a surprising ability to cross the rodent blood brain barrier and alter cofilin activity in brain. We also offer suggestions as to how neuronal-specific cofilin regulation might be achieved.

Keywords: Cofilin phosphoregulation; Cognitive disorders; Dendritic spines; Neuropathic pain; Psychiatric disorders; Rodent models; Sleep deprivation.

Figure 1. Modulation of actin and dendritic spine morphology by cofilin and ancillary proteins

Actin dynamics and network formation depend upon nucleation factors, cross-linkers and G-actin subunit availability. Nucleation factors in spines include formins, involved in linear filament growth such as filopodial extensions for initial spine formation, and the Arp2/3 complex, which nucleates branched filament networks (Spence and Soderling, 2015). The role of cross-linkers, such as filamin, drebrin, and α-actinin, and the cross-linking and contractile motor myosin II, in creating the actin meshworks that contribute to spine morphology is still being elucidiated (Ivanov et al., 2009; Korobova & Svitkina, 2010; Chazeau & Giannone, 2016; Segura et al., 2016; Kneussel & Wagner, 2013). In addition to changes in dendritic spine shape and volume during LTP and LTD (structural plasticity), changes in ion channel surface expression and activity (functional plasticity) mediate synaptic transmission efficiency. Both processes rely on regulation of the actin cytoskeleton and myosin motors (Kneussel & Wagner, 2013; Spence & Soderling, 2015; Lei et al., 2016; Chazeau & Giannone, 2016). Cofilin works in concert with many other proteins to regulate actin filament dynamics, most notably as a factor to promote actin severing and depolymerization, allowing spine shrinkage for LTD, although cofilin activity may be transiently required for actin polymerization-dependent processes such as LTP (Gu et al., 2010). At saturating concentrations (*) cofilin promotes stabilization of F-actin but inhibits myosin II from binding actin and thus modulates myosin II-mediated contractile activity, which is necessary for synaptic functions including ion channel trafficking. In neurons undergoing oxidative stress, cofilin-saturated filaments can bundle into rods, sequestering cofilin and compromising synaptic function (Bamburg & Bernstein, 2016). Cofilin can also remove Arp2/3 complex-branched filament networks (Chan et al., 2009), although a cofilin-related protein (glial maturation factor, GMF) has evolved with greater efficiency in this function (Ydenberg et al., 2013; Poukkula et al., 2014). Cofilin competes for F-actin binding with some proteins, such as drebrin (Grintsevich & Reisler, 2014) and long isoforms of tropomyosins (Tpms), but short Tpms allow and may enhance cofilin effects (Bryce et al., 2003; Janco et al., 2016). Other proteins work in conjunction with cofilin to modulate actin dynamics: coronin1b (crn) enhances the recruitment of cofilin to F-actin (Mikati et al., 2015); actin interacting protein 1 (Aip) enhances the severing ability of cofilin and its complete disassembly of F-actin (Ono et al., 2004; Nadkarni and Brieher, 2014; Gressin et al., 2015; Jansen et al., 2015). Srv2 enhances the dissociation of cofilin from cofilin-actin monomers, or it can enhance cofilin-mediated filament depolymerization from both ends of uncapped filaments (Balcer et al., 2003; Chaudhry et al., 2014; Johnston et al., 2015). The source and maintenance of the G-actin pool in spines is less well understood. Monomer pools in brain are composed primarily of profilin-actin and Tβ4-actin complexes (Devineni et al., 1999), with profilin-actin being preferentially used for formin-mediated filament elongation (Lee et al., 2013). Profilin and Tβ4 inhibit Arp2/3 complex nucleation of filaments and thus serve as a gatekeeper for the switch between Arp2/3-branched networks and linear arrays (Suarez et al., 2015; Rotty et al., 2015; Vitriol et al., 2015; Pernier et al., 2016).

Figure 2. Overview of some dendritic spine signaling pathways of importance to this review

Cofilin is inhibited in binding to actin by phosphorylation and is activated by dephosphorylation. Active cofilin can sever F-actin, or bind cooperatively to saturate and stabilize pieces of F-actin, as elaborated in Figure 1. Cofilin binding to F-actin also competes with myosin II, inhibiting its contractile activity. The sites of inhibition by the cofilin-derived S3 and phosphoS3 (pS3) peptides are shown. LIM kinases (LIMK) and slingshot phosphatases (SSH) that regulate cofilin are themselves subject to regulation in response to many different receptors in dendritic spines, only some of which are shown here. The C-terminal domain (CTD) of NLG1 is generated by the activity of the same protease (γ-secretase) that cleaves APP to generate Aβ in AD. Abbreviations used in this figure and not defined elsewhere are: Pak-interacting exchange factor β (PIX), an activator of Rac1 and Rap1; Pak inhibitory domain (PID); G-protein coupled receptor (GPCR), which, when activated, stimulates nucleotide exchange on the alpha subunit of a heterotrimeric G protein, stimulating adenylcyclase (AC) to make cyclic AMP (cAMP), which activates protein kinase A (PKA) anchored via A kinase anchoring protein (AKAP), but whose response is limited by lowering cAMP via activation of a phosphodiesterase (PDE); ionotropic glutamate receptors and signaling proteins are organized by multidomain scaffolding proteins including postsynaptic density protein 95 (PSD-95), and synapse-associated protein PSD-95-associated protein/guanylate kinase-associated protein (SAPAP/GKAP); protein kinase D1 (PKD) phosphorylates and inactivates SSH; RanBP9 interacts with integrin dimers that contain the β1 subunit (α/β1).