Primary structure and mRNA localization of protein F1, a growth-related protein kinase C substrate associated with synaptic plasticity

Restoring Synaptic Function: How Intranasal Delivery of 3D-Cultured hUSSC Exosomes Improve Learning and Memory Deficits in Alzheimer's Disease

Memory problems are often the first signs of cognitive impairment related to Alzheimer's disease (AD), and stem cells and stem cell-derived exosomes (EXOs) have been studied for their therapeutic potential to improve the disease signs. While many studies have shown the anti-inflammatory and immunomodulatory effects of stem cells and exosomes on improving memory in different AD models, there is still insufficient data to determine how they modulate neural plasticity to enhance spatial memory and learning ability. Therefore, we conducted a study to investigate the effects of exosomes derived from 3D-cultured human Unrestricted Somatic Stem Cells (hUSSCs) on spatial memory and neuroplasticity markers in a sporadic rat model of AD. Using male Wistar rats induced by intracerebral ventricle injection of streptozotocin, we demonstrated that intranasal administration of hUSSC-derived exosomes could decrease Aβ accumulation and improve learning and memory in the Morris water maze test. We also observed an increase in the expression of pre-synaptic and post-synaptic molecules involved in neuronal plasticity, including NMDAR1, integrin β1, synaptophysin, pPKCα, and GAP-43, in the hippocampus. Our findings suggest that intranasal administration of exosomes can ameliorate spatial learning and memory deficits in rats, at least in part, by increasing the expression of neuroplasticity proteins. These results may encourage researchers to further investigate the molecular pathways involved in memory improvement after stem cell and exosome therapy, with the goal of increasing the efficacy and safety of exosome-based treatments for AD.

A Shift from a Pivotal to Supporting Role for the Growth-Associated Protein (GAP-43) in the Coordination of Axonal Structural and Functional Plasticity

In a number of animal species, the growth-associated protein (GAP), GAP-43 (aka: F1, neuromodulin, B-50, G50, pp46), has been implicated in the regulation of presynaptic vesicular function and axonal growth and plasticity via its own biochemical properties and interactions with a number of other presynaptic proteins. Changes in the expression of GAP-43 mRNA or distribution of the protein coincide with axonal outgrowth as a consequence of neuronal damage and presynaptic rearrangement that would occur following instances of elevated patterned neural activity including memory formation and development. While functional enhancement in GAP-43 mRNA and/or protein activity has historically been hypothesized as a central mediator of axonal neuroplastic and regenerative responses in the central nervous system, it does not appear to be the crucial substrate sufficient for driving these responses. This review explores the historical discovery of GAP-43 (and associated monikers), its transcriptional, post-transcriptional and post-translational regulation and current understanding of protein interactions and regulation with respect to its role in axonal function. While GAP-43 itself appears to have moved from a pivotal to a supporting factor, there is no doubt that investigations into its functions have provided a clearer understanding of the biochemical underpinnings of axonal plasticity.

Overexpression of GAP-43 reveals unexpected properties of hippocampal mossy fibers

The mossy fiber (MF) system targets the apical dendrites of CA3 pyramidal cells in the stratum lucidum (SL). In mice overexpressing the growth-associated protein GAP-43 there is an apparent ectopic growth of these MFs into the stratum oriens (SO) targeting the basal dendrites of these same pyramidal cells (Aigner et al. (1995) Cell 83:269-278). This is the first evidence to our knowledge that links increased GAP-43 expression with growth of central axons. Here we studied the Aigner et al. transgenic mice but were unable to confirm such growth into SO. However, using quantitative methods we did observe enhanced growth within the regions normally targeted by MFs, for example, the SL in the CA3a region. These contrasting results led us to study MFs with double-immunostaining using an immunohistochemical marker for MFs, the zinc transporter, ZnT3, to visualize the colocalization of transgenic GAP-43 within MFs. Unexpectedly, using both fluorescence and confocal microscopy, we were unable to detect colocalization of GAP-43-positive axons with ZnT3-positive MF axons within the MF pathways, either in the region of the MF axons or in the SL, where MF terminals are abundant. In contrast, the plasma membrane-associated presynaptic marker SNAP-25 did colocalize with transgenic GAP-43-positive terminals in the SL. Synaptophysin, the vesicle-associated presynaptic terminal marker, colocalized with ZnT3 but did not appear to colocalize with GAP-43. The present findings raise important questions about the properties of granule cells and the MF mechanisms that differentially regulate axonal remodeling in the adult hippocampus: (1) Because there appears to be at least two populations of granule cells defined by their differential protein expression, this points to the existence of an intrinsic heterogeneity of granule cell expression beyond that contributed by adult neurogenesis; (2) Giventhe present evidence that growth is induced in mice overexpressing GAP-43 in adjacent non-GAP-43 containing MFs, the potential exists for a heretofore unexplored interaxonal communication mechanism.

Ectopic growth of hippocampal mossy fibers in a mutated GAP-43 transgenic mouse with impaired spatial memory retention

In a previous study, it was shown that transgenic mice, designated G-NonP, forget the location of a water maze hidden platform when tested 7 days after the last training day (Holahan and Routtenberg (2008) Hippocampus 18:1099-1102). The memory loss in G-NonP mice might be related to altered hippocampal architecture suggested by the fact that in the rat, 7 days after water maze training, there is discernible mossy fiber (MF) growth (Holahan et al. (2006) Hippocampus 16:560-570; Rekart et al. (2007) Learn Mem 14:416-421). In the present report, we studied the distribution of the MF system within the hippocampus of naïve, untrained, G-NonP mouse. In WT mice, the MF projection was restricted to the stratum lucidum of CA3 with no detectable MF innervation in distal stratum oriens (dSO). In G-NonP mice, in contrast, there was an ectopic projection terminating in the CA3 dSO. Unexpectedly, there was nearly a complete loss of immunostaining for the axonal marker Tau1 in the G-NonP transgenic mice in the MF terminal fields indicating that transgenesis itself leads to off-target consequences (Routtenberg (1996) Trends Neurosci 19:471-472). Because transgenic mice overexpressing nonmutated, wild type GAP-43 do not show this ectopic growth (Rekart et al., in press) and the G-NonP mice overexpress a mutated form of GAP-43 precluding its phosphorylation by protein kinase C (PKC), the possibility exists that permanently dephosphorylated GAP-43 disrupts normal axonal fasciculation which gives rise to the ectopic growth into dSO.

Expression of protein kinase-C substrate mRNA in the motor cortex of adult and infant macaque monkeys

To understand the molecular and cellular bases of plasticity in the primate motor cortex, we investigated the expression of three protein kinase-C (PKC) substrates: GAP-43, myristoylated alanine-rich C-kinase substrate (MARCKS), and neurogranin, which are key molecules regulating synaptic plasticity. Prominent signals for the three mRNAs were primarily observed in pyramidal cells. Large pyramidal cells in layer V, from which the descending motor tract originates, contained weaker hybridization signals for GAP-43 and neurogranin mRNAs than did the smaller pyramidal cells. We also performed double-label in situ hybridization showing that GAP-43 and neurogranin mRNAs were expressed in a subset of MARCKS-positive neurons. Quantitative analysis showed that the expression was different between the layers: layer VI contained the strongest and layer II the weakest signals for all three mRNAs. The expression levels of GAP-43 and MARCKS mRNA in layer V were higher than in layer III, while the expression level of neurogranin mRNA in layer V was almost the same as in layer III. Developmental analysis from the newborn to adult indicated that the expression levels of the three mRNAs were higher in the infant motor cortex than in the adult. The expression of both GAP-43 and neurogranin mRNAs transiently increased over several months postnatally. The present study showed that the expression of the three PKC substrates was specific to cell types, cortical layers, and postnatal developmental stage. The specific expression may reflect functional specialization for plasticity in the motor cortex of both infants and adults.

Agents that act by different mechanisms modulate the activity of protein kinase CβII isozyme in the rat spinal cord during peripheral inflammation

Hyperalgesia following unilateral complete Freund's adjuvant-induced inflammation was characterized by paw withdrawal latency to thermal stimulus. Paw withdrawal latencies were significantly shorter on the complete Freund's adjuvant-treated paw than on the contralateral paw of the complete Freund's adjuvant- and the sham-treated rats. Total cytosolic protein kinase C activity in the lumbar enlargement was unchanged on the sides of the spinal cord ipsi- and contra-lateral to the inflamed paw. Membrane-associated activities of protein kinase Calpha, protein kinase CbetaI and protein kinase Cgamma did not change significantly on the sides of the cord ipsi- and contra-lateral to the inflammation. However, membrane-associated activity of protein kinase CbetaII was increased in the cord section ipsilateral to the inflammation, suggesting that increased translocation/activation of protein kinase CbetaII is related to thermal hyperalgesia. Dextrorphan (an N-methyl-D-aspartate receptor antagonist), L-703,606 (an NK-1 receptor antagonist) and an antisense oligodeoxynucleotide for a selective knockdown of protein kinase Cbeta, reduced complete Freund's adjuvant-induced hyperalgesia, and reversed significant changes in the membrane activity of protein kinase CbetaII on the spinal cord section ipsilateral to the inflamed paw. Dextrorphan and protein kinase Cbeta antisense oligodeoxynucleotide were effective in reversing complete Freund's adjuvant-induced increase in the activity of protein kinase CbetaII ipsilateral to the inflammation at all the doses tested, but L-703,606 was effective only at the highest dose. Furthermore, in the presence of inflammatory stimulus, dextrorphan and L-703,606 did not alter the activities of membrane-associated protein kinase Calpha, protein kinase CbetaI, and protein kinase Cgamma in the section of the spinal cord ipsi- and contra-lateral to the inflammation. Protein kinase Cbeta antisense oligodeoxynucleotide had no significant effect on the membrane-associated activities of protein kinase Calpha and protein kinase Cgamma, but decreased the activities of both protein kinase CbetaI and protein kinase CbetaII and the expression of protein kinase Cbeta isozyme in the spinal cord. The data provide evidence that a common molecular event that converges to initiate and maintain hyperalgesia may include the translocation and activation of protein kinase CbetaII in the spinal dorsal horn.

Expression of protein kinase C-substrate mRNAs in the basal ganglia of adult and infant macaque monkeys

We performed in situ hybridization histochemistry on the monkey basal ganglia to investigate the mRNA localization of three protein kinase C substrates (GAP-43, MARCKS, and neurogranin), of which expression plays a role in structural changes in neurites and synapses. Weak hybridization signals for GAP-43 mRNA and intense signals for both MARCKS and neurogranin mRNAs were observed in the adult neostriatum. All three of the mRNAs were expressed in both substance P-positive direct pathway neurons and enkephalin-positive indirect pathway neurons. In the nucleus accumbens, the hybridization signals for the three mRNAs were weaker than those in the neostriatum. Double-label in situ hybridization histochemistry in the neostriatum revealed that GAP-43 and neurogranin mRNAs were expressed in a subset of MARCKS-positive neurons. While intense hybridization signals for MARCKS mRNA were observed in all of the other basal ganglia regions such as the globus pallidus, substantia innominata, subthalamic nucleus, and substantia nigra, intense signals for GAP-43 mRNA were restricted to the substantia innominata and substantia nigra pars compacta. No signal for neurogranin mRNA was observed in the basal ganglia regions outside the neostriatum and the nucleus accumbens. These results indicate that the protein kinase C substrates are abundant in some specific connections in cortico-basal ganglia circuits. Developmental analysis showed that the expression level in the putamen and nucleus accumbens, but not in the caudate nucleus, was higher in the infant than in the adult, suggesting that synaptic maturation in the caudate nucleus occurs earlier than that in the putamen and nucleus accumbens.

Hippocampal-dependent memory is impaired in heterozygous GAP-43 knockout mice

Cajal proposed that the rearrangement and growth of neurites and synaptic terminals is a substrate for the formation and storage of long-term memories. Proteins that regulate this learning-dependent growth are therefore likely to be "core determinants" (Sanes and Lichtman, Nat Neurosci 1999; 2:597-604) of such information storage processes. Although the growth-associated, protein kinase C (PKC) substrate GAP-43 has been oft-implicated in synaptic plasticity and memory, it has never been demonstrated that a reduction in the level of this protein has a deleterious effect on memory, because most homozygotes die perinatally. In this report, we observe significant memory impairments in heterozygous GAP-43 knockout mice with GAP-43 levels reduced by one-half. Impaired memory for a context was demonstrated in contextual fear conditioning. Importantly, no significant impairments in cued conditioning or on tests of nociceptive or auditory perception were observed in the heterozygous knockout, indicating that the observed impairments were unlikely related to performance or acquisition factors and are the result of reduced GAP-43 levels in the hippocampus. The present results, taken together with the prior demonstration of enhanced memory in transgenic mice overexpressing GAP-43, provide strong evidence for a pivotal role of hippocampal GAP-43 in the bidirectional regulation of mnemonic processing.

Expression of neurogranin and neuromodulin is affected in the striatum of vitamin A-deprived rats

Our previous data showed that vitamin A deficiency (VAD) induces, in whole brain, a reduced amount of mRNA for brain retinoic acid (RA) and triiodothyronine (T3) nuclear receptors (i.e., RAR, RXR, and TR, respectively), which is accompanied by reduced amounts of mRNA and protein of neurogranin (RC3, a neuronal protein involved in synaptic plasticity) as well as selective behavioral impairment. Given the important role of retinoids for optimal brain functioning, the effects of vitamin A depletion and subsequent administration of RA or T3 on the mRNA levels of RA and T3 nuclear receptors and on two target genes' (RC3 and neuromodulin or GAP43) mRNA and protein levels were examined in the hippocampus, striatum, and cerebral cortex. A quantitative real-time polymerase chain reaction (PCR), in situ hybridization, and Western blot analysis demonstrated that the striatal region is the brain site where both RA and T3 signaling pathways are most affected by VAD. Indeed, rats fed a vitamin A-free diet for 10 weeks exhibited decreased expression of RAR, RXR, TR, RC3, and GAP43 in the striatum. The administration of T3 to these vitamin A-deprived rats reversed the reduction in mRNA levels of RA and T3 nuclear receptors and in mRNA and protein levels of target genes in this region. These data suggest that modifications that appear preferentially in the striatum, a region highly sensitive to vitamin A bioavailability, may contribute to neurobiological alterations and the spatial learning impairment that occurs in vitamin A-deprived animals.

Cell type- and region-specific expression of protein kinase C-substrate mRNAs in the cerebellum of the macaque monkey

We performed nonradioactive in situ hybridization histochemistry in the monkey cerebellum to investigate the localization of protein kinase C-substrate (growth-associated protein-43 [GAP-43], myristoylated alanine-rich C-kinase substrate [MARCKS], and neurogranin) mRNAs. Hybridization signals for GAP-43 mRNA were observed in the molecular and granule cell layers of both infant and adult cerebellar cortices. Signals for MARCKS mRNA were observed in the molecular, Purkinje cell, and granule cell layers of both infant and adult cortices. Moreover, both GAP-43 and MARCKS mRNAs were expressed in the external granule cell layer of the infant cortex. In the adult cerebellar vermis, signals for both GAP-43 and MARCKS mRNAs were more intense in lobules I, IX, and X than in the remaining lobules. In the adult hemisphere, both mRNAs were more intense in the flocculus and the dorsal paraflocculus than in other lobules. Such lobule-specific expressions were not prominent in the infant cerebellar cortex. Signals for neurogranin, a postsynaptic substrate for protein kinase C, were weak or not detectable in any regions of either the infant or adult cerebellar cortex. The prominent signals for MARCKS mRNA were observed in the deep cerebellar nuclei, but signals for both GAP-43 and neurogranin mRNAs were weak or not detectable. The prominent signals for both GAP-43 and MARCKS mRNAs were observed in the inferior olive, but signals for neurogranin were weak or not detectable. The cell type- and region-specific expression of GAP-43 and MARCKS mRNAs in the cerebellum may be related to functional specialization regarding plasticity in each type of cell and each region of the cerebellum.

Down-regulation of GAP-43 During Oligodendrocyte Development and Lack of Expression by Astrocytes In Vivo: Implications for Macroglial Differentiation

The discovery of molecular markers which are selectively expressed during the development of specific classes of rat central nervous system macroglia has greatly advanced our understanding of how these cells are related. In particular, it has been shown in tissue culture that oligodendrocytes and some astrocytes (type-2) may be derived from a common progenitor cell (O-2A progenitor). However, the existence of type-2 astrocytes in vivo has yet to be unequivocally established. Recently, it has been reported that the neural-specific growth-associated protein-43 (GAP-43, otherwise known as B-50, F1, pp46 and neuromodulin) may be expressed by cells of the O-2A lineage in vitro. We set out to examine the cellular specificity of GAP-43 in O-2A progenitors and their descendants in vitro and in vivo. Using a polyclonal antiserum against a GAP-43 fusion protein we have shown the presence of immunoreactive GAP-43 in the membranes of bipotential O-2A glial progenitor cells and type-2 astrocytes by Western blotting and immunocytochemistry of cells in culture. In contrast to previous studies, double labelling with mature oligodendrocyte markers showed that GAP-43 is down-regulated during oligodendrocyte differentiation in vitro. Immunohistochemical staining of sections of developing rat brain demonstrated the same developmental regulation of GAP-43, suggesting that oligodendrocytes only express GAP-43 at immature stages. In addition, normal and reactive astrocytes in tissue sections were not labelled with GAP-43.

B-50/GAP43 Expression Correlates with Process Outgrowth in the Embryonic Mouse Nervous System

The hypothesis that B-50/GAP43, a membrane-associated phosphoprotein, is involved in process outgrowth has been tested by studying the developmental pattern of expression of B-50/GAP43 mRNA and protein during mouse neuroembryogenesis. B-50/GAP43 mRNA is first detectable at embryonic day 8.5 (E8.5) in the presumptive acoustico-facialis ganglion. Subsequently, both B-50/GAP43 mRNA and protein were co-expressed in a series of neural structures: in the ventral neural tube (from E9.5) and dorsal root ganglia (from E10.5), in the marginal layer of the neuroepithelium surrounding the brain vesicles and in the cranial ganglia (from E9.5), in the autonomic nervous system (from E10.5), in the olfactory neuroepithelium and in the mesenteric nervous system (from E11.5), in a continuum of brain regions (from E12.5) and in the retina (from E13.5). Immunoreactive fibers were always seen arising from these regions when they expressed B-50/GAP43 mRNA. The spatial and temporal pattern of B-50/GAP43 expression demonstrates that this protein is absent from neuroblasts and consistently appears in neurons committed to fiber outgrowth. The expression of the protein in immature neurons is independent of their embryological origin. Our detailed study of B-50/GAP43 expression during mouse neuroembryogenesis supports the view that this protein is involved in a process common to all neurons elaborating fibers.

Cloning and Characterization of the Rat Gene Encoding GAP-43

GAP-43 is a gene expressed only in the nervous system. The protein product is believed to be important to neuronal growth and plasticity. Most, and likely all, neurons express high levels of GAP-43 during periods of neurite elongation. To initiate studies of GAP-43 gene regulation we have cloned the rat gene encoding GAP-43. The GAP-43 gene includes three exons. The first exon encodes only the amino terminal 10 amino acids, which corresponds to the membrane targeting domain of GAP-43. The second exon encodes a putative calmodulin binding domain and a protein kinase C phosphorylation site. The 5'-flanking sequence is unusual in that it lacks CAAT or TATA elements, and directs RNA transcription initiation from several sites. Some of the transcription start sites are used to a different degree in the central and peripheral nervous systems.

Palmitoylcarnitine modulates interaction protein kinase C delta-GAP-43

Palmitoylcarnitine, reported previously to promote neuronal differentiation, was observed to affect distribution of protein kinase C (PKC) isoforms in neuroblastoma NB-2a cells, leading to retardation in cytoplasm of high molecular weight species of PKCbeta and delta. Growth cone protein-GAP-43, a PKC substrate, was co-immunoprecipitated with all the conventional and novel PKCs: palmitoylcarnitine, however, decreased its amount exclusively in the complex with PKCdelta. Administration of palmitoylcarnitine, although did not change the subcellular distribution of GAP-43, decreased its phosphorylation, which could regulate other signal transduction pathways (calmodulin and G(0)-dependent).

Hyperalgesia induced by peripheral inflammation is mediated by protein kinase C betaII isozyme in the rat spinal cord

We have addressed the molecular mechanism(s) of hyperalgesia, which depends on increased excitability of dorsal horn neurons and on sensitization of primary afferent nociceptors, during peripheral inflammation. Following unilateral adjuvant-induced inflammation in the rat hind paw, time-course changes in behavioral hyperalgesia and functional activities of Ca2+/phospholipid-dependent protein kinase C isozymes were examined. Inflammation was characterized by increase in paw diameter, and behavioral hyperalgesia was quantified as paw withdrawal latency from a radiant heat source. Behavioral hyperalgesia on the injected paw was significantly increased. This was accompanied by a significant increase in total functional membrane-associated protein kinase C activity, whereas total cytosolic protein kinase C activity was unchanged on the sides of the lumbar spinal cord both contralateral and ipsilateral to the inflammation. Importantly, on the side of lumbar cord ipsilateral to the inflamed paw, the activity of membrane-associated protein kinase CbetaII was increased following the same time-course as the paw withdrawal latency decrease, suggesting an increased translocation of protein kinase Cbetall to the membrane related to behavioral hyperalgesia. A defined mixture of purified gangliosides, which inhibits intracellular protein kinase C translocation and activation, decreased inflammation-induced paw withdrawal latency, and specifically decreased the activity of membrane-associated protein kinase Cbetall on the side of the spinal cord ipsilateral to the inflammation. Quantitative immunohistochemical analyses demonstrated intensified protein kinase CbetaII-like immunoreactivity on the side of the spinal cord ipsilateral to the inflammation. Time-course for increases in the activity of membrane-associated protein kinase CbetaII, and in intensity of protein kinase CbetaII-immunoreactivity, paralleled inflammation-mediated changes in paw withdrawal latency and paw diameter. Our findings indicate an apparent involvement of protein kinase CbetaII isozyme specifically in the molecular mechanism(s) of thermal hyperalgesia.

Neuronal BC1 RNA: co-expression with growth-associated protein-43 messenger RNA

Brain-specific cytoplasmic RNA 1 (BC1-RNA), a non-coding RNA polymerase III transcript, is a neuronal RNA that is specifically targeted to dendritic domains. It is co-localized with components of the dendritic protein synthetic machinery, and it has been suggested to operate in the regulation of local translation-related processes in postsynaptic microdomains, thus subserving long-term synaptic plasticity in neurons. To probe the relevance of BC1 expression in neuronal plasticity, we have analyzed the expression pattern of BC1 RNA in the rat nervous system. We found that BC1 RNA is expressed by a specific subset of neurons (but not by non-neuronal cells) in the central and peripheral nervous system of the adult rat. The BC1 labeling pattern indicates that the subcellular location of the RNA is typically postsynaptic which, depending on cell type, manifests itself in a predominantly somatic, somatodendritic, or dendritic location. Our results further show that BC1-expressing neurons typically co-express the messenger RNA for growth-associated protein-43 (GAP-43). Such co-expression was observed in diverse brain areas, including the olfactory bulb, neocortex, and hippocampus, among others. While BC1 RNA was in many neuronal cell types detectable in distal dendritic domains, GAP-43 messenger RNA was typically more restricted to neuronal perikarya. In the mature nervous system, expression of GAP-43 has been described as an intrinsic determinant of predominantly presynaptic plasticity, while BC1 RNA has been implicated in postsynaptic plasticity. Co-expression of both RNAs, as reported here, thus identifies a distinct subset of neurons in the rat nervous system that exhibits both types of plasticity.

Absence of hippocampal mossy fiber sprouting in transgenic mice overexpressing brain-derived neurotrophic factor

Excess neuronal activity upregulates the expression of two neurotrophins, nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF) in adult hippocampus. Nerve growth factor has been shown to contribute the induction of aberrant hippocampal mossy fiber sprouting in the inner molecular layer of the dentate gyrus, however the role of prolonged brain-derived neurotrophic factor exposure is uncertain. We examined the distribution and plasticity of mossy fibers in transgenic mice with developmental overexpression of brain-derived neurotrophic factor. Despite 2--3-fold elevated BDNF levels in the hippocampus sufficient to increase the intensity of neuropeptide Y immunoreactivity in interneurons, no visible changes in mossy fiber Timm staining patterns were observed in the inner molecular layer of adult mutant hippocampus compared to wild-type mice. In addition, no changes of the mRNA expression of two growth-associated proteins, GAP-43 and SCG-10 were found. These data suggest that early and persistent elevations of brain-derived neurotrophic factor in granule cells are not sufficient to elicit this pattern of axonal plasticity in the hippocampus.

Site-specific calcium-dependent proteolysis of neuronal protein GAP-43

GAP-43 is a presynaptic protein participating in signal transduction processes in nerve terminals. GAP-43 exists in neurons along with two truncated forms devoid of 4 and 40 N-terminal residues. In this report, we show that these forms of GAP-43 are proteolytic fragments derived from calcium-dependent cleavage of GAP-43 molecule at 5th and 41st residues. GAP-43 site-specific proteolysis in synaptosome and cytosol fractions proved to be dependent on the addition of millimolar amounts of calcium. This fact together with inhibition of GAP-43 proteolysis by calpain inhibitors as well as local composition of the cleavage sites indicates to the participation of calpain in this process. The proteolysis disturbs some properties characteristic for whole GAP-43 molecules, in particular, calmodulin binding and Ser-41 phosphorylation, when the cleavage occurs at 41st residue. Some other GAP-43 properties (G(o) protein activation and membrane attachment) are retained by separate fragments. Therefore, calcium controlled site-specific proteolysis of GAP-43 can be of great physiological significance.

Role of GAP-43 in mediating the responsiveness of cerebellar and precerebellar neurons to axotomy

To determine whether the competence for axonal sprouting and/or regeneration in the cerebellar system correlates with GAP-43 expression, we have studied GAP-43 mRNA and protein expression in the postlesioned cerebellum and inferior olive. Purkinje cells transiently express GAP-43 during their developmental phase (from E15 to P5 in the rat) which consists of fast axonal growth and the formation of the corticonuclear projection. Adult Purkinje cells, which in control adult rats do not express GAP-43, are extremely resistant to the effects of axotomy but cannot regenerate axons. However, a late and protracted sprouting of axotomized Purkinje cells occurs spontaneously and correlates with a mild expression of GAP-43 mRNA. In contrast, inferior olivary neurons, despite their high constitutive expression of GAP-43, do not sprout but retract their axons and die after axotomy. Furthermore, mature Purkinje cells in cerebellar explants of transgenic mice that overexpress GAP-43 do not regenerate after axotomy, even in the presence of a permissive substrate (cerebellar embryonic tissue) and, contrary to the case in wild-type mice, they do not survive in the in vitro conditions and undergo massive cell death. These results show that the expression of GAP-43 is not only associated with axonal growth, but also with neuronal death.

Differential regulation of primary protein kinase C substrate (MARCKS, MLP, GAP-43, RC3) mRNAs in the hippocampus during kainic acid-induced seizures and synaptic reorganization

In the mature hippocampus, kainic acid seizures lead to excitotoxic cell death and synaptic reorganization in which granule cell axons (mossy fibers) form ectopic synapses on granule cell dendrites. In the present study, we examined the expression of four major, developmentally regulated protein kinase C (PKC) substrates (MARCKS, MLP, GAP-43, RC3), which have different subcellular and regional localizations in the hippocampus at several time points (6 hr, 12 hr, 18 hr, 24 hr, 48 hr, 5 days, or 15 days) following kainic acid seizures using in situ hybridization. Consistent with previous reports, following kainate seizures, GAP-43 mRNA expression exhibited a delayed and protracted elevation in the granule cell layer, which peaked at 24 hr, whereas expression in fields CA1 and CA3 remained relatively unchanged. Conversely, RC3 mRNA expression exhibited a delayed reduction in the granule cell layer that was maximal at 18 hr, as well as a reduction CA1 at 48 hr, whereas CA3 levels did not change. MARCKS mRNA expression in the granule cell layer and CA1 remained stable following kainate, although an elevation was observed in subfield CA3c at 12 hr. Similarly, MLP mRNA expression did not change in the granule cell layer or CA1 following kainate but exhibited a protracted elevation in subfields CA3b,c beginning at 6 hr post-kainate. Collectively these data demonstrate that different PKC substrate mRNAs exhibit unique expression profiles and regulation in the different cell fields of the mature hippocampus following kainic acid seizures and during subsequent synaptic reorganization. The expression profiles following kainate seizures bear resemblance to those observed during postnatal hippocampal development, which may indicate the recruitment of common regulatory mechanisms.

Transcriptional and post-transcriptional regulation of a brain growth protein: regional differentiation and regeneration induction of GAP-43

During axonal regeneration synthesis of different growth-associated proteins is increased. As yet there is no clear picture of the specific contribution made by the transcriptional and post-transcriptional machinery that provides the gene products necessary for process outgrowth. Here we focus our study on the transcriptional processes in neurons by using intron-directed in situ hybridization to the primary transcript of a brain growth protein GAP-43. In most brain regions, levels of primary transcript expression of GAP-43 were highly correlated with levels of its mRNA. However, there were notable dissociations: in hippocampal granule cells, high levels of primary transcript were evident yet no GAP-43 mRNA was detected. In locus coeruleus the reverse was true; there were high levels of GAP-43 mRNA but no detectable primary transcript. A primary transcript antitermination mechanism is proposed to explain the first dissociation, and a post-transcriptional mRNA stabilization mechanism to explain the second. Transcriptional activation during nerve regeneration was monitored by assessing primary transcript induction of GAP-43 in mouse facial motor neurons. This induction, as well as its mRNA, was restricted to the side of the facial nerve crush. Increases were first observed at 24 h with a rapid increase in both measures up to 3 days. To our knowledge, this is the first in vivo evidence demonstrating transcriptional activation of a brain growth protein in regenerating neurons. The present study points to the GAP-43 transcriptional mechanism as a key determinant of GAP-43 synthesis. Along with the recruitment of post-transcriptional mechanisms, such synthesis occurs in response to both intrinsic developmental programs and extrinsic environmental signals.

Genetic dissection of the signals that induce synaptic reorganization

Synaptic reorganization of mossy fibers following kainic acid (KA) administration has been reported to contribute to the formation of recurrent excitatory circuits, resulting in an epileptogenic state. It is unclear, however, whether KA-induced mossy fiber sprouting results from neuronal cell loss or the seizure activity that KA induces. We have recently demonstrated that certain strains of mice are resistant to excitotoxic cell death, yet exhibit seizure activity similar to what has been observed in rodents susceptible to KA. The present study takes advantage of these strain differences to explore the roles of seizure activity vs cell loss in triggering mossy fiber sprouting. In order to understand the relationships between gene induction, cell death, and the sprouting response, we assessed the regulation of two molecules associated with the sprouting response, c-fos and GAP-43, in mice resistant (C57BL/6) and susceptible (FVB/N) to KA-induced cell death. Following administration of KA, increases in c-fos immunoreactivity were observed in both strains, although prolonged induction of c-fos was present only in the hippocampal neurons of FVB/N mice. Mossy fiber sprouting following KA administration was also only observed in FVB/N mice, while induction of GAP-43, a marker associated with mossy fiber sprouting, was not observed in either strain. These results indicate that: (i) KA-induced seizure activity alone is insufficient to induce mossy fiber sprouting; (ii) mossy fiber sprouting may be due to the loss of hilar neurons following kainate administration; and (iii) induction of GAP-43 is not a necessary component of the sprouting response that occurs following KA in mice.

Microscale purification of proteins exhibiting anomalous electrophoretic migration: application to the analysis of GAP-43 phosphorylation

Quite often, in vivo analysis of posttranslational protein modifications is complicated by the lack of specific antibodies or unsatisfactory immunoprecipitation efficiency. Here, we present an alternative method to immunoprecipitation that takes advantage of the anomalous electrophoretic behavior exhibited by GAP-43. This method can be applied to other proteins which show similar characteristics, such as myristoylated alanine-rich C kinase, NAP-22, and Neurogranin, among others. All these proteins display relative mobility values that depend on the concentration of polyacrylamide used in the resolving gel. Cell extracts or tissue homogenates are first separated by SDS-PAGE in 15% polyacrylamide gels, and the bands containing GAP-43 are identified, excised from the gel, and rerun on a second SDS-PAGE in 7.5% polyacrylamide/6 M urea gels. To quickly identify the position of GAP-43 in the first gel, a small amount of fluorescein-labeled recombinant GAP-43 was added to the initial extracts. The method, applied to the analysis of GAP-43 phosphorylation in rat hippocampal slices, can be typically completed in less than 4 h. The excellent yields of purification obtained contributed to a greater accuracy and increased reliability of the radioactivity measurements. It also allowed further processing of the samples, including the analysis of the different phosphorylation sites by proteolytic digestion and peptide mapping.

Generation of aberrant sprouting in the adult rat brain by GAP-43 somatic gene transfer

The expression of GAP-43 was modulated genetically in the adult rat nigrostriatal or septohippocampal pathway using recombinant adeno-associated virus (rAAV) vectors incorporating the neuron specific enolase (NSE) promoter and either a rat GAP-43 cDNA or the corresponding antisense sequence. Bicistronic expression of green fluorescent protein (GFP) enabled us to evaluate transduced neurons selectively. Single injections of rAAV into the substantia nigra pars compacta (SNc) transduced both dopaminergic and non-dopaminergic neurons stably for the 3-month duration of the study. Transduction with the GAP-43 vector in this region: (1) increased GAP-43 mRNA levels 2-fold compared to controls; (2) led to GAP-43 immunoreactivity in neuronal perikarya, axons, and dendrites that was not observed otherwise; and (3) resulted in GAP-43/ GFP-positive axons that were traced to the striatum where they formed clusters of aberrant nets. The GAP-43 antisense vector, in contrast, decreased neuropil GAP-43 immunoreactivity compared to controls in the SNc. In septum, injections of the GAP-43 expressing vector also caused aberrant clusters of GAP-43 labelled fibers in terminal fields, i.e., fornix and hippocampus, that were not observed in control tissues. It therefore appears that rAAV vectors provide a novel approach for modulating intraneuronal GAP-43 expression in the adult brain.

Cloning of novel injury-regulated genes. Implications for an important role of the muscle-specific protein skNAC in muscle repair

To gain insight into the molecular mechanisms underlying the wound repair process, we searched for genes that are regulated by skin injury. Using the differential display reverse transcription-polymerase chain reaction technique, we identified a gene that was strongly induced as early as 12 h after wounding. Sequence analysis revealed the identity of the corresponding protein with skeletal muscle nascent polypeptide-associated complex (skNAC), a recently identified muscle-specific transcription factor. By in situ hybridization and immunohistochemistry, we demonstrated the specific expression of skNAC in skeletal muscle cells of the panniculus carnosus at the wound edge. Furthermore, in vitro studies with cultured myoblasts revealed expression of skNAC in differentiating and differentiated, but not in proliferating, nondifferentiated cells. Differentiation of cultured myoblasts was accompanied by simultaneous expression of skNAC and the muscle-specific transcription factor myogenin. Our results provide the first evidence for a role of skNAC in muscle repair processes. Furthermore, they demonstrate the usefulness of our approach in identifying new players in wound repair.

Different types of ROS-scavenging enzymes are expressed during cutaneous wound repair

Injury to the skin initiates a series of events including inflammation, new tissue formation, and matrix remodeling. During the early inflammatory phase, polymorphonuclear leukocytes and macrophages infiltrate the wounded tissue. Once activated, they produce large amounts of reactive oxygen species (ROS) as part of their defense mechanism. Although this process is beneficial, increased levels of ROS can inhibit cell migration and proliferation and can even cause severe tissue damage. Therefore, cells must develop strategies for the detoxification of these molecules. To gain insight into the mechanisms which underlie this process, we analyzed the temporal and spatial expression pattern of various ROS-scavenging enzymes during the healing process of full-thickness excisional wounds in mice. Here we demonstrate a strong mRNA expression of two types of superoxide dismutase (SOD), as well as of catalase, and the selenoenzymes glutathione peroxidase (SeGPx) and phospholipid hydroperoxide glutathione peroxidase in normal and wounded skin. Most importantly, mRNA levels of the SODs and of SeGPx increased strongly after skin injury. In situ hybridization and immunofluorescence studies revealed the presence of these transcripts at multiple places in the wound, whereby particularly high expression levels were detected in the hyperproliferative epithelium and the hair follicles at the wound edge. These data suggest an important role of ROS-scavenging enzymes in the detoxification of ROS during cutaneous wound repair.

Hypoxia/ischemia induces dephosphorylation of rat brain neuromodulin/GAP-43 in vivo

The in vivo state of phosphorylation and the modification of two Cys residues of neuromodulin/ GAP-43 (Nm) were analyzed by electrospray ionization-mass spectrometry (ES-MS). The protein was purified from rat brain with homogenization buffer containing 1% Nonidet P-40, protease inhibitors, protein phosphatase inhibitors, and sulfhydryl reagent, 4-vinylpyridine. Nm was purified by HPLC and ion-exchange chromatography, and the various fractions were identified by ES-MS as unphosphorylated and mono-, di-, tri-, and tetraphosphorylated species. All of these Nm species contained 2 mol of added 4-vinylpyridine per mol of Nm, suggesting that the two Cys residues are in the reduced form in the brain. In vivo, the majority of Nm is in the phosphorylated form (approximately 80%), of which the levels of the mono- and diphospho forms are higher than those of the tri- and tetraphospho species. Four in vivo phosphorylation sites, Ser41, Thr95, Ser142, and Thr172, were identified by amino acid sequencing and tandem ES-MS of the peptides derived from Lys-C endoproteinase digestion. Among these sites, only Ser41 is a known target of PKC, whereas the kinases responsible for the phosphorylation of the other three novel sites are unknown. Hypoxia/ischemia caused a preferential dephosphorylation of Ser41 and Thr172, whereas Thr95 is the least susceptible to dephosphorylation.

cis-acting regulatory elements in the GAP-43 mRNA 3'-untranslated region can function in trans to suppress endogenous GAP-43 gene expression

The expression of the GAP-43 gene is controlled partly by changes in the stability of its mRNA, a process that is mediated by the interaction of specific sequences in the 3'-untranslated region (3'UTR) with neuronal-specific RNA-binding proteins. Limiting amounts of these trans-acting factors are available in the cell, thus we proposed that overexpression of the GAP-43 3'UTR could affect the levels of the endogenous mRNA via competitive binding to specific RNA-binding proteins. In this study, we show that chronic expression of GAP-43 3'UTR sequences in PC12 cells causes the depletion of the endogenous mRNA and consequent reduction of GAP-43 protein levels. The levels of the mRNAs for c-fos, the amyloid precursor protein (APP) and the microtubule associated protein tau, all three containing similar 3'UTR sequences, were not affected by the treatment. These results thus suggest that the effect of excess GAP-43 3'UTR is specific for its corresponding mRNA. We also used an HSV (herpes simplex virus)-1 vector and a mammalian expression vector with an inducible promoter to acutely express a 10 to 50 fold excess of 3'UTR sequences. Under these conditions, we found that transient expression of the GAP-43 3'UTR was effective in inhibiting both GAP-43 gene expression and neurite outgrowth in nerve growth factor (NGF)-treated PC12 cells and in primary neuronal cultures. These results underscore the role of 3'UTR sequences in the control of GAP-43 gene expression and suggest that overexpression of specific 3'UTR sequences could be used as a potential tool for probing the function of other post-transcriptionally-regulated proteins during neuronal differentiation.

The protein kinase C (PKC) substrate GAP-43 is already expressed in neural precursor cells, colocalizes with PKCeta and binds calmodulin

Expression of the growth-associated protein of 43-kDa (GAP-43), which is described as a postmitotic, neuron-specific major protein kinase C (PKC) substrate, was investigated in the murine embryonic carcinoma cell line PCC7-Mz1 which develops into a brain-tissue-like pattern of neuronal, fibroblast-like and astroglial cells upon stimulation with all-trans retinoic acid (RA). GAP-43 expression was very low in stem cells, but increased on mRNA and protein level within the 12 h after differentiation was initiated. While the P1 promoter of the GAP-43 gene gave rise to a 1.6-kb mRNA and was already active at a very low level in PCC7-Mz1 stem cells, transcription of the P2 promoter, which resulted in a 1.4-kb mRNA, was completely blocked in stem cells but increased rapidly after RA treatment. Within the first 2 days of neural differentiation, GAP-43 was localized with the cytoplasmic membrane and the Golgi complex of proliferating neural precursor cells. Then, GAP-43 was translocated to the growth cones and neurites, and from day 6, when neurons began to acquire polarity, the protein was found in the axons. GAP-43 was never detected in the non-neuronal PCC7-Mz1 derivatives, i.e. in fibroblasts or glial cells. In the foetal rat brain (prenatal day F11), GAP-43 was expressed in the optic stalk, the lense plakode and in the postmitotic neurons of the marginal zone of the hindbrain. Moreover, in a layer between the ventricular and marginal zone of the hindbrain (F13) and forebrain (F15), GAP-43 was already expressed in mitotic neural precursor cells. In PCC7-Mz1 cultures, 2 days after addition of RA, GAP-43 became phosphorylated upon activation of PKC, and colocalized specifically with the novel PKC isoform eta. Phosphorylation of GAP-43 caused a disruption of its complex with calmodulin. These data demonstrate that GAP-43 is already a functional PKC substrate in prolific neuronal precursor cells, and may participate in neuronal cell lineage determination.

Gene expression of growth-associated proteins, GAP-43 and SCG10, in the hippocampal formation of the macaque monkey: nonradioactive in situ hybridization study

We performed nonradioactive in situ hybridization histochemistry in the monkey hippocampal formation that includes the hippocampus, the subicular complex, and the entorhinal cortex to detect the expression of mRNA for two growth-associated proteins: GAP-43 and SCG10. Overall, the distribution patterns overlapped but were partially distinct. In the hippocampus, the intense hybridization signals for both GAP-43 and SCG10 mRNAs were observed in the pyramidal cell layer of Ammon's horn, especially in CA3 subfields. The intense hybridization signals were also observed in the stratum oriens of Ammon's horn and the polymorphic layer of the dentate gyrus. In the granule cell layer of the dentate gyrus, many GAP-43 mRNA-positive cells were observed, whereas a few positive cells with weak signals were observed for SCG10 mRNA. Throughout the subicular complex, the hybridization signals for both mRNAs were weak. In the entorhinal cortex, both mRNAs were abundant in the caudal field. These subregion-specific expression of the growth-associated proteins may reflect the functional specialization regarding plasticity in each region of the monkey hippocampal formation.

Connective tissue growth factor: a novel regulator of mucosal repair and fibrosis in inflammatory bowel disease?

Inflammatory bowel disease (IBD) is a multifactorial disorder which is characterized by massive damage of the epithelium and the underlying mesenchyme of the intestine. Due to the potent effect of connective tissue growth factor (CTGF) on fibroblast proliferation and connective tissue deposition we speculated about a possible role of this mitogen in IBD. Here we demonstrate a strikingly increased expression of CTGF mRNA in surgical specimens of patients suffering from two forms of IBD, Crohn's disease and ulcerative colitis. In most specimens, the levels of CTGF mRNA correlated with the degree of inflammation as assessed by histological analysis of adjacent tissue samples and by expression analysis of the pro-inflammatory cytokine interleukin-1 beta. However, areas of little inflammation which were characterized by severe fibrosis also revealed high levels of CTGF mRNA. Expression of transforming growth factor beta-1 (TGF-beta 1), the only known inducer of CTGF so far, as well as of the CTGF target genes collagen I alpha 1, fibronectin and integrin alpha 5 revealed a strong correlation with the expression of CTGF. These data suggest a prominent role of CTGF in the repair of mucosal injury in IBD and in the aberrant deposition of extracellular matrix leading to fibrosis and stenosis, one major complication in IBD, especially in Crohn's disease.

Matrix metalloproteinases (MMPs) and their physiological inhibitors (TIMPs) are differentially expressed during excisional skin wound repair

During cutaneous wound healing a number of migratory and remodeling events occur that require the action of matrix metalloproteinases (MMPs) and their natural inhibitors (TIMPs). In this study, we analyzed the temporal and spatial expression patterns of these molecules during the healing of murine excisional skin wounds. Our data imply that defined phases of repair rely on distinct repertoires of MMP activity and TIMP counterregulation. Reepithelialization was found to be associated with active production of collagenase, 92-kDa gelatinase, and stromelysins-1 and -2 by distinct subpopulations of keratinocytes at the migrating border. Notably, no TIMP transcripts were expressed in the epidermis, but TIMP-1 expression in the wound colocalized with expression of collagenase, 92-kDa gelatinase, and stromelysin-1, albeit in distinct cells. Concomitant with the formation of an extensive hyperproliferative epithelium, TIMP-1 transcripts accumulated at the mesenchymal/epidermal border of the granulation tissue. During later phases of wound repair, we observed an increase in 72-kDa gelatinase and MT1-MMP expression, whereby the transcripts of these colocalizing MMPs were detected exclusively and at high levels in the granulation tissue. At completion of reepithelialization, the expression levels of the MMPs and TIMP-1 seen in epidermal and dermal compartments declined to near-basal levels, whereas the macrophage-specific metalloelastase (MME) reached maximum expression. In reepithelialized wound tissue, MME transcripts were detected in deep layers of reconstituted dermis and seemed to cluster around vascular structures. Systemic glucocorticoid treatment, which is known to result in impaired wound healing, led to a nearly complete shut-off of MME expression. These observations imply an additional role of macrophage-related proteolysis, independent of its classical roles during earlier, inflammatory phases of cutaneous wound repair.

Modulation of GAP-43 mRNA by GABA and glutamate in cultured cerebellar granule cells

Expression of GAP-43 in the cerebellum and selected regions of the brain has been shown to be developmentally regulated. Localization of GAP-43 mRNA within granule cells of the immature and mature rat cerebellum has been demonstrated by in situ hybridization. Higher levels are detected in the neonate compared to the adult. To determine if the cerebellar neurotransmitters, GABA (gamma-amino-butyric acid) and glutamate are involved in the modulation of GAP-43 expression, cultured cerebellar granule cells were exposed to these transmitters. Cultures were treated with glutamate, GABA, or the agonists/antagonists to their receptors in serum-free media for 5-7 days. Analysis of the levels of GAP-43 mRNA by in situ hybridization indicated that a 7-day exposure to GABA (25 and 50 microM) significantly lowered levels of granule cell GAP-43 mRNA. Specific agonists to the GABAA (muscimol) and GABAB (baclofen) receptors produced a decrease similar to that observed for GABA. Results from these studies also indicated that exposure to non-NMDA (CNQX) and NMDA (CPP, MK-801) glutamate receptor antagonists, and a metabotropic receptor glutamate agonist (ACPD), decreased the level of GAP-43 mRNA. The involvement of GABA and glutamate in the modulation of GAP-43 expression was corroborated by Northern hybridization. These studies revealed that a 5-day exposure to GABA decreased the cellular content of GAP-43 mRNA by 21% whereas exposure to glutamate resulted in a 37% increase. Findings from the studies reported here, using an in vitro cerebellar granule cell model, suggest that levels of GAP-43 mRNA, in vivo, are modulated by input from both excitatory glutamatergic mossy fibers and inhibitory GABAergic Golgi interneurons. Thus, modulation of GAP-43 mRNA by these neurotransmitters may influence granule cell maturation during development in the neonate and neuroplasticity in the adult, possibly at the parallel fiber-Purkinje cell synapse.

Activation of gp130 by IL-6/soluble IL-6 receptor induces neuronal differentiation

Interleukin-6 (IL-6) on target cells binds to the specific IL-6 receptor (IL-6R) and subsequently induces homodimerization of the signal-transducing protein gp130. Cells which express gp130 but no IL-6R and which therefore do not respond to IL-6 can be stimulated by the complex of IL-6 and soluble IL-6R (slL-6R). Here we show that on rat pheochromocytoma cells (PC12), the combination of IL-6 and slL-6R but not IL-6 alone induces expression of c-fos, GAP-43 and neuron-specific enolase followed by neuron-specific differentiation and formation of a neuronal network. The differentiation was dose-and time-dependent and followed the same kinetics as nerve-growth factor (NGF)-induced differentiation. The responses of PC12 cells to IL-6/sIL-6R and NGF were additive, suggesting independent signaling pathways. We demonstrate that activation of gp130 generates a neuronal differentiation signal that is equivalent to and independent of trk/NGF receptor tyrosine kinase. Interestingly, the failure of IL-6 to induce differentiation of PC12 cells is not due to lack of surface expression of IL-6R as IL-6 alone triggered expression of GAP-43 mRNA and protein. We hypothesize that PC12 cells express more gp130 than IL-6R and that the extent of activated gp130 molecules determines the quality of the response.

B-50, the growth associated protein-43: modulation of cell morphology and communication in the nervous system

The growth-associated protein B-50 (GAP-43) is a presynaptic protein. Its expression is largely restricted to the nervous system. B-50 is frequently used as a marker for sprouting, because it is located in growth cones, maximally expressed during nervous system development and re-induced in injured and regenerating neural tissues. The B-50 gene is highly conserved during evolution. The B-50 gene contains two promoters and three exons which specify functional domains of the protein. The first exon encoding the 1-10 sequence, harbors the palmitoylation site for attachment to the axolemma and the minimal domain for interaction with G0 protein. The second exon contains the "GAP module", including the calmodulin binding and the protein kinase C phosphorylation domain which is shared by the family of IQ proteins. Downstream sequences of the second and non-coding sequences in the third exon encode species variability. The third exon also contains a conserved domain for phosphorylation by casein kinase II. Functional interference experiments using antisense oligonucleotides or antibodies, have shown inhibition of neurite outgrowth and neurotransmitter release. Overexpression of B-50 in cells or transgenic mice results in excessive sprouting. The various interactions, specified by the structural domains, are thought to underlie the role of B-50 in synaptic plasticity, participating in membrane extension during neuritogenesis, in neurotransmitter release and long-term potentiation. Apparently, B-50 null-mutant mice do not display gross phenotypic changes of the nervous system, although the B-50 deletion affects neuronal pathfinding and reduces postnatal survival. The experimental evidence suggests that neuronal morphology and communication are critically modulated by, but not absolutely dependent on, (enhanced) B-50 presence.

Multiple subcellular mRNA distribution patterns in neurons: a nonisotopic in situ hybridization analysis

Previous studies have established that most of the mRNAs that neurons express are localized in the cell body and very proximal dendrites, whereas a small subset of mRNAs is present at relatively high levels in dendrites. It is not clear, however, whether particular mRNAs have the same subcellular distribution in different types of neurons or whether different types of neurons sort mRNAs in different ways. The present study was undertaken to address these questions. Nonisotopic in situ hybridization techniques were used to define the subcellular localization of representative mRNAs including beta-tubulin, low-molecular-weight neurofilament protein (NF-68), high-molecular-weight microtubule-associated protein (MAP2), growth-associated protein 43 (F1/GAP43), the alpha subunit of calcium/calmodulin-dependent protein kinase II (alpha CaMII kinase), and poly (A+) mRNA. The mRNAs for beta-tubulin, neurofilament 68, and F1/GAP43 were restricted to the region of the cell body and very proximal dendrites in most neurons. In some neuron types, however, labeling for NF-68 extended for considerable distances into dendrites. In some neurons that express MAP2, the mRNA was present at the highest levels in the proximal third to half of the dendritic arbor, whereas in other neurons the highest levels of labeling were in the cell body. In most neurons that express alpha CaMII kinase, the highest levels of the mRNA were in the cell body, but labeling was also present throughout dendrites. However, in a few types of neurons, alpha CaMII kinase mRNA was largely restricted to the cell body. The fact that there are no general rules for mRNA localization that apply to all neuron types implies the existence of neuron type-specific mechanisms that regulate mRNA distribution.

GAP-43: an intrinsic determinant of neuronal development and plasticity

Several lines of investigation have helped clarify the role of GAP-43 (FI, B-50 or neuromodulin) in regulating the growth state of axon terminals. In transgenic mice, overexpression of GAP-43 leads to the spontaneous formation of new synapses and enhanced sprouting after injury. Null mutation of the GAP-43 gene disrupts axonal pathfinding and is generally lethal shortly after birth. Manipulations of GAP-43 expression likewise have profound effects on neurite outgrowth for cells in culture. GAP-43 appears to be involved in transducing intra- and extracellular signals to regulate cytoskeletal organization in the nerve ending. Phosphorylation by protein kinase C is particularly significant in this regard, and is linked with both nerve-terminal sprouting and long-term potentiation. In the brains of humans and other primates, high levels of GAP-43 persist in neocortical association areas and in the limbic system throughout life, where the protein might play an important role in mediating experience-dependent plasticity.

B-50/growth-associated protein-43, a marker of neural development in Xenopus laevis

To study the regulation and function of the growth-associated protein B-50/growth-associated protein-43 (mol. wt 43,000) in Xenopus laevis, B-50/growth-associated protein-43 complementary DNAs were isolated and characterized. The deduced amino acid sequence revealed potential functional domains of Xenopus B-50/growth-associated protein-43 that may be involved in G-protein interaction, membrane-binding, calmodulin-binding and protein kinase C phosphorylation. The expression of B-50/growth-associated protein-43 at the RNA and protein level during development was investigated using the Xenopus complementary DNA and the monoclonal B-50/growth-associated protein-43 antibody NM2. The antibody NM2 recognized the gene product on western blot and in whole-mount immunocytochemistry of Xenopus embryos. Moreover, visualization of the developmentally regulated appearance of B-50/growth-associated protein-43 immunoreactivity showed that this mode of detection may be used to monitor axonogenesis under various experimental conditions. In the adult Xenopus, XB-50/growth-associated protein-43 messenger RNA was shown to be expressed at high levels in brain, spinal cord and eye using northern blotting. The earliest expression detected on northern blot was at developmental stage 13 with poly(A) RNA. By whole-mount immunofluorescence, applying the confocal laser scanning microscope, the protein was first detected in embryos from stage 20, where it was expressed in the developing trigeminal ganglion. Also later in development the expression of the B-50/growth-associated protein-43 gene was restricted to the nervous system in Xenopus, as was previously found for the mouse. In conclusion, we find that XB-50/growth-associated protein-43 is a good marker to study the development of the nervous system in Xenopus laevis.

Distinctions between hippocampus of mouse and rat: protein F1/GAP-43 gene expression, promoter activity, and spatial memory

We began these experiments as an attempt to replicate in the mouse the induction by kainate (KA) of F1/GAP-43 mRNA we observed in adult rat hippocampal granule cells [Mol. Brain Res., 33 (1995) 22-28]. However, even though KA induced behavioral seizures in the mouse similar to those in the rat, neither induction of F1/GAP-43 mRNA nor subsequent mossy fiber sprouting observed in the rat was detected in three different mouse strains. It was also surprising that the distribution of constitutive levels of F1/GAP-43 mRNA in mouse and rat hippocampus was qualitatively different. Indeed, F1/GAP-43 expression in CA3 pyramidal cells was significantly greater in rat than mouse, while F1/GAP-43 expression in CA1 cells of rat and mouse was equivalent using densitometric analysis. Thus, F1/GAP-43 expression in rat is quantitatively higher in CA3 and CA1 pyramidal cells. In mouse, expression was equivalent in these two subfields. In a transgenic mouse bearing a rat F1/GAP-43 promoter-reporter (lacZ) construct (line 252), in-vivo promoter activity of F1/GAP-43 was studied in hippocampal cells. Transgene expression in hippocampal pyramidal subfields, high in CA3, low in CA1 pyramidal cells, paralleled the distribution of rat F1/GAP-43 mRNA levels, not mouse. Differences in the constitutive F1/GAP-43 expression pattern in hippocampus between rat and mouse may therefore be determined by different recognition elements present on the F1/GAP-43 promoter. KA injected into the line 252 transgenic mouse did not activate the rat F1/GAP-43 promoter in mouse hippocampal granule cells. The absence of both F1/GAP-43 mRNA expression induction and promoter activation in mouse granule cells after KA is likely related to genera differences in transcriptional regulatory mechanisms, though post-transcriptional mechanisms cannot be excluded. Since the different hippocampal chemistry of F1/GAP-43 in rat and mouse likely extends to other molecular species, behaviors in rat and mouse that depend on hippocampal function might be different as well. We therefore evaluated spatial memory ability in a delayed matching-to-sample task. In contrast to rat, we were surprised to find no evidence of the ability to learn this task in three different mouse strains. Since interest in mouse genetics in relation to behavior and memory functions of hippocampus is growing, generalizations concerning hippocampal function from studies carried out on the mouse need to be made with caution considering the specific behavioral, pharmacological, and general molecular differences observed here. One can also be opportunistic and exploit the natural variations between these two genera to gain insight into the molecular mechanisms underlying information storage.

Distribution of GAP-43 mRNA in the immature and adult cerebellum: a role for GAP-43 in cerebellar development and neuroplasticity

Expression of GAP-43 mRNA in the rat cerebellum and inferior olivary nucleus was examined at birth, during postnatal development and in the adult by both Northern and in situ hybridization. Northern blot analysis revealed that cerebellar GAP-43 mRNA expression increases from birth to postnatal day (PD) 7 and then declines to a lower level in the adult. At birth, in situ hybridization experiments showed intense labeling of GAP-43 mRNA in the premigratory, but not the germinal, zone of the cerebellar external granule cell layer. Localization of GAP-43 within the premigratory zone, a layer containing post-mitotic granule cells, indicates that granule cells begin expressing GAP-43 mRNA after final mitosis and during axonal outgrowth of the parallel fibers. The deep cerebellar nuclei and the inferior olive were also intensely labeled at birth. GAP-43 mRNA was localized in granule cells during their migration through the molecular layer of the developing cerebellum and after their arrival in the internal granule cell layer. By PD 21, the pattern of GAP-43 expression was similar to that observed in the adult; GAP-43 mRNA was localized to the internal granule layer and the inferior olive with minimal to no hybridization in the deep cerebellar nuclei and none in the molecular layer. Purkinje cells were devoid of GAP-43 mRNA throughout the postnatal and adult periods. In light of our observations, we propose that GAP-43 is a critical factor in granule cell differentiation/migration, as well as in the parallel and climbing fiber axonal outgrowth and synaptogenesis during development. Localization of GAP-43 mRNA within granule and inferior olivary cells of adult animals indicates that GAP-43 protein observed in the molecular layer is transported from these cells to their terminals in the molecular layer suggesting that GAP-43 is also an intrinsic presynaptic determinant in cerebellar neuroplasticity.

Keratinocyte growth factor is highly overexpressed in inflammatory bowel disease

Recently we demonstrated an important function of keratinocyte growth factor (KGF) in wound re-epithelialization. As KGF is mitogenic for various epithelial cells, we speculated about a role of KGF in epithelial repair processes of other organs as seen in a variety of inflammatory diseases. Here we demonstrate a strikingly increased expression of KGF in surgical specimens from patients suffering from Crohn's disease and ulcerative colitis. The levels of KGF expression strongly correlated with the degree of inflammation as assessed by histological analysis of adjacent tissue and expression analysis of the pro-inflammatory cytokine interleukin-1 beta. The highest levels of KGF mRNA and protein were found in mesenchymal cells of the lamina propria, particularly in highly inflamed areas. As the KGF receptor is expressed in intestinal epithelial cells, KGF seems to act in a paracrine manner to stimulate proliferation of these cells. These data suggest a crucial role of KGF in epithelial repair after injury caused by inflammatory processes.

Growth-associated protein-43 (GAP-43) facilitates peptide hormone secretion in mouse anterior pituitary AtT-20 cells

The neuronal growth-associated protein (GAP)-43 (neuromodulin, B-50, F1), which is concentrated in the growth cones of elongating axons during neuronal development and in nerve terminals in restricted regions of the adult nervous system, has been implicated in the release of neurotransmitter. To study the role of GAP-43 in evoked secretion, we transfected mouse anterior pituitary AtT-20 cells with the rat GAP-43 cDNA and derived stably transfected cell lines. Depolarization-mediated beta-endorphin secretion was greatly enhanced in the GAP-43-expressing AtT-20 cells without a significant change in Ca2+ influx; in contrast, expression of GAP-43 did not alter corticotropin-releasing factor-evoked hormone secretion. The transfected cells also displayed a flattened morphology and extended processes when plated on laminin-coated substrates. These results suggest that AtT-20 cells are a useful model system for further investigations on the precise biological function(s) of GAP-43.

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.

Regulation of phosphatidylinositide transduction system in the rat spinal cord during aging

Age-related functional alterations in a variety of neurotransmitter systems result in modulation of interneuronal communications which has some relevance in neurological deficits observed in the aging process. The synergistic interactions between protein kinase and inositol 1,4,5-trisphosphate (insP3)/Ca2+ pathways underlie a variety of cellular responses to external stimuli. To determine whether age-dependent changes occur in the regulation of protein kinase C and inositol 1,4,5-trisphosphate/Ca2+ pathways, insP3 contents as a marker for the release of intracellular calcium, saturation binding analysis of Ins P3 receptor using [3H]inositol 1,4,5-trisphosphate, slot/northern blot analysis of Ins P3 receptor-encoding mRNA transcripts, and the activities of Ca2+/phospholipid-dependent protein kinase C isozymes were investigated in the rat spinal cord. Inositol 1,4,5-trisphosphate content and [3H]inositol 1,4,5-trisphosphate binding site density (Bmax) were quantified in the spinal cords of young (three months old), adult (12 months old) and senescent (25 months old) male Fischer 344 rats. Spinal cord content of inositol 1,4,5-trisphosphate was increased (P < 0.01) in the 25-month old compared to the three- and 12-month old animals. The density of Ins P3 receptor in particulate membranes derived from the 25-month old rats was reduced (P < or = 0.01), but the binding affinity (Kd) was increased (P < or = 0.04) by a factor of 2.2 and 3.2 at 25 months of age when compared with three- and 12-month old animals, respectively. Young and middle-aged animals showed no differences in both inositol 1,4,5-trisphosphate contents and [3H]inositol 1,4,5-trisphosphate binding site density. The quantity of Ins P3 receptor mRNA was significantly increased with age in the order 25 >> 12 > 3 months of age. Total functional cytosolic and membrane-associated PKC activities were decreased (P < or = 0.05) in the 25-month compared to the three- and 12-month old rats in which activity remained unchanged. Total membrane/cytosolic activity ratios were unchanged by the aging process. In all cases, the activities of membrane-associated conventional protein kinase C isozymes (alpha, beta and gamma), determined by immunoprecipitation followed by in situ quantification of protein kinase C activities in the immunoprecipitates, showed age-dependent decline. The activities of protein kinase C-alpha and beta were significantly decreased in age-related manner. However, the activity of the gamma-isozyme was not significantly changed at 12- and 25-months of age, although it was higher (P < or = 0.03) in young rats. Western blot analyses using affinity purified polyclonal antibodies specific for each isozyme indicated a single protein with an apparent molecular mass of approximately 80 x 10(3) molec. weight for all isozymes except for the beta isozyme that also had an appreciable immunoreactive band at approximately 36 x 10(3) molec. weight. Overall, the aging process did not affect the electropheretic mobility of each isozyme. With decreased protein kinase C activity, the present data suggest that the aging process would decrease protein kinase C-induced phosphorylation of membrane proteins including Ins P3 receptor. A significant change in Ins P3 receptor affinity combined with increased levels of Ins P3 receptor mRNA-encoding transcripts in senescent rats suggests not only a modification (possibly by phosphorylation) of Ins P3 receptor protein but also the existence of multiple (spliced) variants of Ins P3 receptor in spinal neurons with increasing age. The present data indicate that the spinal contents of inositol 1,4,5-trisphosphate increased with age, but with decreased efficacy and number of inositol 1,4,5-trisphosphate-activatable Ca2+ channels in the spinal cord of senescent rats. These age-related changes may contribute to the attenuated responsiveness of spinal cord neurons by phosphoinositide-coupled receptors during the aging process.

SCG10, a neuron-specific growth-associated protein in Alzheimer's disease

Neuronal growth-associated proteins (nGAPs) are markers of neuronal process outgrowth and are associated with both degenerative and sprouting responses in Alzheimer's disease (AD) brain. To study possible involvement of SCG10, an nGAP, in AD, we cloned human SCG10 cDNA and analyzed SCG-10 at mRNA and protein levels in control and AD brains. The deduced amino acid sequence of human SCG10 was 69% identical to stathmin, another nGAP. By in situ hybridization, both SCG10 and stathmin mRNAs were detected in selected neuronal populations in aged human brains. Quantitative analysis by RNase protection revealed that levels of neither SCG10 nor stathmin mRNAs were significantly altered in AD. Using an SCG10-specific antibody, Western blot analysis did not reveal any quantitative changes of SCG10 in AD. However, when the concentration of SCG10 protein was plotted against the number of tangles, a positive correlation was found. SCG10 levels did not correlate with plaque numbers. Furthermore, immunohistochemical study revealed that neuronal SCG10 protein accumulated in the cell bodies in AD-affected regions. Thus, SCG10 compartmentalization and metabolism may be altered in AD possibly due to mechanisms related to tangle formation in this disease.

Aberrant GAP-43 gene expression in Alzheimer's disease

GAP-43 is a growth-associated phosphoprotein expressed at high levels in neurons during development, axonal regeneration, and neuritic sprouting. GAP-43 gene expression in mature neurons is probably functionally important for the structural remodeling of synapses as required for learning and establishing new memory. The widespread aberrant neuritic growth accompanied by impaired synaptic plasticity in Alzheimer's disease (AD) suggests that abnormal GAP-43 gene expression may contribute to the cascade of neurodegeneration. In the present study, end-stage AD brains exhibited reduced neuronal expression but increased glial cell levels of GAP-43 mRNA and protein. Glial cell localization of GAP-43 gene expression was confirmed by in situ hybridization of cerebral tissue, Northern blot analysis of microdissected cerebral white matter, and independent analysis of astrocytoma cell lines and primary malignant astrocytomas. In addition, in AD, GAP-43 immunoreactivity was translocated from the cytosol to membranes of swollen neuritic (dendritic) and glial cell processes throughout cerebral cortex and white matter. Downregulated and aberrant neuronal GAP-43 gene expression appears to reflect an important molecular lesion that precedes and progresses with the widespread synaptic disconnection and dementia in AD. At the same time, the presence of similar neuronal abnormalities in Pick's disease, diffuse Lewy body disease, Parkinson's disease, and Down syndrome suggests common mechanisms in the respective cascades of neurodegeneration. Finally, the finding of aberrantly increased glial cell GAP-43 gene expression in AD exposes a previously unrecognized neurodegenerative change that may account for the axonal loss and white matter atrophy detected early in the course of disease.

Immediate upstream promoter regions required for neurospecific expression of SNAP-25

The promoter structure and regulation of Snap, a gene encoding the presynaptic t-SNARE SNAP-25 implicated in synaptic vesicle docking and fusion, was studied. Transcription start-site analysis revealed two major start sites located 42 nucleotides apart. Nucleotide sequence of a promoter region 2073 nucleotides upstream of the first transcription site contains three AP-1, one CRE sequence, and three Sp1-like sites close to the TATA box. Further upstream of these sites two TG repeats were found. The ability of regions within the 5' upstream sequence to promote basal neural-specific expression in tissue culture cells was evaluated using a series of constructs containing both Snap gene start sites with progressively restricted 5' sequence linked to the chloramphenicol acetyl transferase (CAT) reporter gene. CAT expression was maximal in neuron-like undifferentiated ND7 and PC12 cells transfected with constructs containing Snap sequences up to 127 bp from the start site. In contrast, nonneuronal fibroblast cell lines did not express significant amounts of CAT, suggesting that this short 127-bp sequence is sufficient to drive neural specific expression of SNAP-25. Band shift analysis of oligonucleotides spanning from -127 to -41 bp of the Snap promoter revealed three distinct DNA-protein complexes generated by brain nuclear extracts and one by liver nuclear extracts, indicating that transcription factors that bind to this 86-bp sequence located just upstream of the TATA box are involved in regulation of basal neurospecific expression of this gene.