Neurodegenerative stimuli induce persistent ADF/cofilin-actin rods that disrupt distal neurite function.

Inclusions containing actin-depolymerizing factor (ADF) and cofilin, abundant proteins in adult human brain, are prominent in hippocampal and cortical neurites of the post-mortem brains of Alzheimer's patients, especially in neurites contacting amyloid deposits. The origin and role of these inclusions in neurodegeneration are, however, unknown. Here we show that mediators of neurodegeneration induce the rapid formation of transient or persistent rod-like inclusions containing ADF/cofilin and actin in axons and dendrites of cultured hippocampal neurons. Rods form spontaneously within neurons overexpressing active ADF/cofilin, suggesting that the activation (by dephosphorylation) of ADF/cofilin that occurs in response to neurodegenerative stimuli is sufficient to induce rod formation. Persistent rods that span the diameter of the neurite disrupt microtubules and cause degeneration of the distal neurite without killing the neuron. These findings suggest a common pathway that can lead to loss of synapses.

Increase in neurite outgrowth mediated by overexpression of actin depolymerizing factor.

Growth cone motility is regulated by changes in actin dynamics. Actin depolymerizing factor (ADF) is an important regulator of actin dynamics, and extracellular signal-induced changes in ADF activity may influence growth cone motility and neurite extension. To determine this directly, we overexpressed ADF in primary neurons and analyzed neurite lengths. Recombinant adenoviruses were constructed that express wild-type Xenopus ADF/cofilin [XAC(wt)], as well as two mutant forms of XAC, the active but nonphosphorylatable XAC(A3) and the less active, pseudophosphorylated XAC(E3). XAC expression was detectable on Western blots 24 hr after infection and peaked at 3 d in cultured rat cortical neurons. Peak expression was approximately 75% that of endogenous ADF. XAC(wt) expression caused a slight increase in growth cone area and filopodia but decreased filopodia numbers on neurite shafts. At maximal XAC levels, neurite lengths increased >50% compared with controls infected with a green fluorescent protein-expressing adenovirus. Increased neurite extension was directly related to the expression of active XAC. Expression of the XAC(E3) mutant did not increase neurite extension, whereas expression of the XAC(A3) mutant increased neurite extension but to a lesser extent than XAC(wt), which was partially phosphorylated. XAC expression had minimal, if any, impact on F-actin levels and did not result in compensatory changes in the expression of endogenous ADF or actin. However, F-actin turnover appeared to increase based on F-actin loss after treatment with drugs that block actin polymerization. These results provide direct evidence that increased ADF activity promotes process extension and neurite outgrowth.

Signal-regulated ADF/cofilin activity and growth cone motility.

It is becoming increasingly evident that proteins of the actin depolymerizing factor (ADF)/cofilin family are essential regulators of actin turnover required for many actin-based cellular processes, including motility. ADF can increase actin turnover by either increasing the rate of actin filament treadmilling or by severing actin filaments. In neurons ADF is highly expressed in neuronal growth cones and its activity is regulated by many signals that affect growth cone motility. In addition, increased activity of ADF causes an increase in neurite extension. ADF activity is inhibited upon phosphorylation by LIM kinases (LIMK), kinases activated by members of the Rho family of small GTPases. ADF become dephosphorylated downstream of signal pathways that activate PI-3 kinase or increase levels of intracellular calcium. The growth-regulating effects of ADF together with its ability to be regulated by a wide variety of guidance cues, suggest that ADF may regulate growth cone advance and navigation.

Selective expression of protein F1/(GAP-43) mRNA in pyramidal but not granule cells of the hippocampus.

Protein F1/GAP-43 is a protein kinase C substrate associated with axonal growth and synaptic plasticity. We used in situ hybridization in rat brain to determine the cellular distribution of its gene expression. Throughout the septotemporal axis of the adult hippocampus, pyramidal cells express F1/GAP-43 mRNA, but granule cells do not. To determine if F1/GAP-43 expression in granule cells ever occurs, we studied its expression in development during mossy fiber outgrowth, when expression should be maximal. Quantitation of relative hybridization levels in the hippocampus revealed a modest increase in granule cell F1/GAP-43 mRNA coincident with mossy fiber outgrowth. But even the peak hybridization in granule cells on day 16 was 75% less than in pyramidal cells. The distribution of grains was over the entire granule cell layer at day 9, but was restricted by day 20 to the inner aspect of the layer, the site of the youngest cells which are still sending out axonal processes. Cell-selective expression of F1/GAP-43 within a particular brain structure was not restricted to the hippocampus. In cerebellum, F1/GAP-43 hybridization was detected in granule cells but not Purkinje cells; in olfactory bulb, mitral cells but not internal granule cells; in habenula, cells in the lateral but not medial nucleus; in substantia nigra, pars compacta cells but not cells in pars reticulata. Neurons containing biogenic amines exhibited intense F1/GAP-43 hybridization: substantia nigra pars compacta (dopamine), the locus coeruleus (norepinephrine), and dorsal raphe (serotonin). In contrast, cholinergic neurons exhibited little (basal forebrain) or no (medial habenula) hybridization. F1/GAP-43 expression is not restricted to a specific cell type and is not correlated with axon length. High F1/GAP-43 expression is apparent in many neurons having either neuromodulatory or memory storage functions. We propose that F1/GAP-43 is important for accelerating process outgrowth and synaptic remodeling, rather than directing growth itself.

Induction of F1/GAP-43 gene expression in hippocampal granule cells after seizures [corrected].

In the adult rat hippocampus mRNA of F1/GAP-43, an axonal growth-associated protein, is highly expressed in pyramidal cells, but is absent in granule cells. To determine whether granule cells can be induced to express mRNA of F1/GAP-43, transcript levels were studied after limbic seizures, which can induce sprouting of granule cell mossy fibers. Seizure-inducing electrolytic lesions were made in the dentate gyrus hilus with stainless-steel electrodes and mRNA levels were measured in contralateral hippocampus by quantitative in situ hybridization. Induction of F1/GAP-43 mRNA expression was observed in granule cells at 24 h, but not at 6 or 12 h, after the hilar lesion. When equivalent sized hilar lesions were made with platinum electrodes, which do not induce seizures, no hybridization was apparent over the granule cells. Hybridization over granule cells had declined by 48 h post-lesion, but even at 10 days it was still slightly higher than in control rats. F1/GAP-43 mRNA expression was also increased 2-fold in CA1 pyramidal cells with peak expression at 48 h post-lesion. These are the first data to our knowledge that demonstrate that F1/GAP-43 gene expression can be altered in neurons located within the adult brain. Induction of F1/GAP-43 mRNA expression in the granule cells may be important for the sprouting of mossy fibers and could be triggered by the elevated levels of brain-derived neurotrophic factor in CA3 cells which precede the increased F1/GAP-43 gene expression in granule cells.

Protein F1/GAP-43 and PKC gene expression patterns in hippocampus are altered 1-2 h after LTP.

Three days after long-term potentiation (LTP) there is a decrease in the gene expression of protein F1 (GAP-43) and gamma-PKC in CA3 pyramidal cells that is correlated with the magnitude of LTP. We predicted these decreases would be preceded by an increment in gene expression. At 1 h, but not at 2 h after LTP, F1/GAP-43 and gamma-PKC mRNA hybridization were increased, but increases were also observed after control stimulation. At both 1 and 2 h after LTP, changes in F1/GAP-43 hybridization were positively correlated with gamma-PKC hybridization and negatively correlated with LTP magnitude. These data indicate that correlated alterations in F1/GAP-43 gene expression and synaptic efficacy can occur as early as 1 h after LTP and persist for days.

MARCKS and protein F1/GAP-43 mRNA in chick brain: effects of imprinting.

The phosphorylation of MARCKS, but not protein F1/GAP-43, is increased in the intermediate and medial portion of the hyperstriatum ventrale (IMHV) after chick imprinting. Here we investigated if MARCKS, but not F1/GAP-43, gene expression would also be altered after imprinting. We first investigated the constitutive mRNA distribution of MARCKS and F1/GAP-43 in chick brain. MARCKS mRNA was expressed in most cells and exhibited a relatively homogeneous distribution. In contrast, F1/GAP-43 mRNA levels were elevated in discrete brain regions, as we had observed in mammals. The highest F1/GAP-43 mRNA levels in the chick brain were in sensory and associational structures such as the hyperstriatal complex and neostriatum, and lower levels were in structures involved in motor control, such as paleostriatum. These results in chick are consistent with the previously drawn generalization that F1/GAP-43 mRNA is expressed in those brain regions which exhibit synaptic plasticity. After imprinting, MARCKS mRNA levels in IMHV were higher in good learners than poor learners. In contrast, analysis of F1/GAP-43 mRNA levels revealed no differences related to training in any brain region sampled. These selective results for MARCKS but not F1/GAP-43 parallel the prior findings on their phosphorylation, and are consistent with our hypothesis that the very same proteins that are post-translationally modified in association with learning and memory also undergo alterations in their gene expression.

Gene expression of the transcription factor NF-kappa B in hippocampus: regulation by synaptic activity.

NF-kappa B is a potent transcriptional activator that resides in latent form in the cytoplasm complexed to its inhibitor I kappa B. Phosphorylation of I kappa B by protein kinase C (PKC) releases NF-kappa B, enabling its translocation to the nucleus. Since PKC can activate NF-kappa B and PKC is activated by long-term potentiation (LTP), we investigated NF-kappa B expression after hippocampal LTP induced in vivo. We first described the expression of the NF-kappa B subunits, p50 and p65, and I kappa B alpha mRNAs, in each cell field of the hippocampus. In other brain locations I kappa B alpha mRNA exhibited a more selective expression than p50 and p65. We then demonstrated specific NF-kappa B-like DNA-binding activity in hippocampal whole-cell extracts and in synaptosomes using electrophoretic mobility shift assays by the following criteria: (1) latent binding was revealed after deoxycholate treatment; (2) binding was competed off by unlabeled kappa B oligonucleotides; and (3) antibodies to either p50 or p65 blocked binding. Since p50 gene expression is auto-regulated by NF-kappa B, we used its expression as a reporter for NF-kappa B activity using quantitative in situ hybridization. Both p50 and p65 increased their expression in response to either LTP-inducing or low-frequency control stimulation, although the increase in p65 mRNA levels was greater after LTP than control stimulation. In contrast to p50 and p65, I kappa B alpha hybridization levels were not increased, but were inversely correlated with the magnitude of LTP. Since NF-kappa B subunit gene expression in the hippocampus is increased by augmented synaptic activity, NF-kappa B activation may contribute to alterations in target gene expression that accompany activity-dependent synaptic plasticity, but only in a combinatorial fashion with other transcription factors.

Calcium-dependent alterations in dendritic architecture of hippocampal pyramidal neurons.

Dendritic arbor formation and the underlying mechanisms are crucial for the functional connectivity and plasticity of neurons. We used a focal electric field to locally raise calcium levels in individual dendritic shafts of isolated hippocampal pyramidal neurons, in order to develop an accessible system for studying dendritic branch formation, and to test the role of calcium as an intrinsic signal that may participate in arborization. Filopodia were induced in a manner temporally and spatially related to induced calcium rises. Certain filopodia also thickened and were transformed into dendritic branches. These results suggest that calcium-mediated signaling can induce branching in dendrites, and describe an accessible system for studying the intracellular machinery that drives dendritic arborization.

Is the lack of protein F1/GAP-43 mRNA in granule cells target-dependent?

Protein F1/GAP-43 is differentially expressed in brain with high levels present in regions associated with memory functions. However, in hippocampus the granule cells lack F1/GAP-43 expression. To determine if this lack of expression is due to inhibitory signals from the target cells, we selectively destroyed CA3 pyramidal cells unilaterally using microinjections of excitotoxins. Kainate lesions induced F1/GAP-43 mRNA expression bilaterally in granule cells at 24 h post-injection. Since the induction contralateral to the lesion was not due to loss of target cells, that induction may be ascribed to consequences of seizure activity. However, F1/GAP-43 mRNA hybridization decreased by 3 d post-lesion and was at background levels by 6 d, indicating that the lack of F1/GAP-43 expression in granule cells is restored despite a lack of target neurons. Unilateral lesions of CA3 cells using ibotenate, which are not as complete as kainate but do not cause seizures, did not induce F1/GAP-43 mRNA in granule cells on either the contralateral or, in 4 of 5 cases, the ipsilateral side. Taken together, these data suggest that the CA3 target is not essential for the absence of F1/GAP-43 expression in granule cells. To compare the extent of damage caused by the lesions, we investigated the location of astrocytes undergoing reactive gliosis, employing as a reporter glial fibrillary acidic protein (GFAP) gene expression. After both kainate and ibotenate injections GFAP hybridization increased in the lesioned area as well as in the contralateral hippocampus. These results indicate that injections of kainate, and possibly ibotenate to a lesser extent, may affect behavior not only by damaging cells at the injection site, but also by altering gene expression in cells at distant sites.

Protein kinase C and F1/GAP-43 gene expression in hippocampus inversely related to synaptic enhancement lasting 3 days.

The mRNA levels of protein F1 (also known as GAP-43), and protein kinase C (PKC) subtypes were measured 3 days after the induction of long-term enhancement (also known as long-term potentiation) in the hippocampus of chronically prepared conscious rats by quantitative in situ hybridization. Altered mRNA levels correlated significantly with alternations in synaptic efficacy; such correlations have not been reported previously. Rats with greater synaptic enhancement had lower gene expression in the CA3 subfield of F1/GAP-43 and both beta-PKC and gamma-PKC, but not alpha-PKC. For microtubule-associated protein 2 (MAP-2), neurogranin, and the glutamate receptor subtype B-flip, no correlation was observed in any cell field between synaptic enhancement and hybridization to the mRNA. To our surprise, alterations in mRNA levels of F1/GAP-43 and gamma-PKC were highly correlated (r = +0.928, P < 0.001), suggesting coordinate regulation. Since F1/GAP-43 is associated with neurite growth, its lowered expression at 3 days would reduce potential growth, leading to synaptic stabilization. We propose that long-term synaptic change is mediated by gene expression of the very same proteins initially modified posttranslationally.

Actin depolymerizing factor and cofilin phosphorylation dynamics: response to signals that regulate neurite extension.

The actin assembly-regulating activity of actin depolymerizing factor (ADF)/ cofilin is inhibited by phosphorylation. Studies were undertaken to characterize the signaling pathways and phosphatases involved in activating phosphorylated ADF (pADF), emphasizing signals related to neuronal process extension. Western blots using antibodies to ADF and cofilin, as well as an ADF/cofilin phosphoepitope-specific antibody characterized in this paper, were used to measure changes in the phosphorylation state and phosphate turnover of ADF/cofilin in response to inhibitors and agents known to influence growth cone motility. Increases in both [Ca2+]i and cAMP levels induced rapid pADF dephosphorylation in HT4 and cortical neurons. Calcium-dependent dephosphorylation depended on the activation of protein phosphatase 2B (PP2B), while cAMP-dependent dephosphorylation was likely through activation of PP1. Growth factors such as NGF and insulin also induced rapid pADF/pcofilin dephosphorylation, with NGF-stimulated dephosphorylation in PC12 cells correlated with the translocation of ADF/cofilin to ruffling membranes. Of special interest was the finding that the rate of phosphate turnover on both pADF and pcofilin could be enhanced by growth factors without changing net pADF levels, demonstrating that growth factors can activate bifurcating pathways that promote both phosphorylation and dephosphorylation of ADF/cofilin. All experimental results indicated that dynamics of phosphorylation on ADF and cofilin are coordinately regulated. Signals that decreased pADF levels are associated with increased process extension, while agents that increased pADF levels, such as lysophosphatidic acid, inhibit process extension. These data indicate that dephosphorylation/activation of pADF is a significant response to the activation of signal pathways that regulate actin dynamics and alter cell morphology and neuronal outgrowth.

Regulating actin dynamics in neuronal growth cones by ADF/cofilin and rho family GTPases.

Growth cone motility and navigation in response to extracellular signals are regulated by actin dynamics. To better understand actin involvement in these processes we determined how and in what form actin reaches growth cones, and once there, how actin assembly is regulated. A continuous supply of actin is maintained at the axon tip by slow transport, the mobile component consisting of an unassembled form of actin. Actin is co-transported with actin-binding proteins, including ADF and cofilin, structurally related proteins essential for rapid turnover of actin filaments in vivo. ADF and cofilin activity is regulated through phosphorylation by LIM kinases, downstream effectors of the Rho family of GTPases, Cdc42, Rac and Rho. Attractive and repulsive extracellular guidance cues might locally alter actin dynamics by binding specific GTPase-linked receptors, activating LIM kinases, and subsequently modulating the activity of ADF/cofilin. ADF is enriched in growth cones and is required for neurite outgrowth. In addition, signals that influence growth cone behavior alter ADF/cofilin phosphorylation, and overexpression of ADF enhances neurite outgrowth. Growth promoting effects of laminin are mimicked by expression of constitutively active Cdc42 and blocked by expression of the dominant negative Cdc42. Repulsive effects of myelin and sema3D on growth cones are blocked by expression of constitutively active Rac1 and dominant negative Rac1, respectively. Thus a series of complex pathways must exist for regulating effectors of actin dynamics. The bifurcating nature of the ADF/cofilin phosphorylation pathway may provide the integration necessary for this complex regulation.