Age-dependent failure of axon regeneration in organotypic culture of gerbil auditory midbrain

Axonal Sprouting and Neuronal Connectivity following Central Nervous System Insult: Implications for Occupational Therapy

Based on selected contemporary research, this paper presents a critical analysis of central nervous system (CNS) reorganisation following insult and the need for therapists better to understand the processes that constitute reorganisation and their possible contribution to the development of spasticity. In the treatment of the sequelae of CNS lesions, the synaptic reorganisation as a result of losses caused by injury - in the form of axonal sprouting - is illustrated, focusing on neuronal reconnectivity. Critical analysis of laboratory, electron microscopy and other animal and human studies is also conducted to integrate the controversies identified and to highlight the concepts that become relevant for occupational therapists, in order to optimise therapeutic intervention for maximising restitution in patients with CNS insult. The paper further discusses the capacity of the CNS to compensate and the need to utilise occupational therapy interventions, such as imagining, mental rehearsals, constraint-induced therapy, virtual reality, biofeedback and the traditional repetitive tasks, which leads to ensuring and facilitating the emergence of new synapses to perform motor tasks and manual skills and to prevent secondary changes. These external stimulations provided by the therapists are likely to stimulate both the damaged hemisphere cross-innervation and/or collateral sprouting. These scientifically based treatment strategies and neurological rehabilitation programmes would, in turn, contribute to improving the quality of life of people with CNS insult.

Dissociated neurons and glial cells derived from rat inferior colliculi after digestion with papain

The formation of gliosis around implant electrodes for deep brain stimulation impairs electrode–tissue interaction. Unspecific growth of glial tissue around the electrodes can be hindered by altering physicochemical material properties. However, in vitro screening of neural tissue–material interaction requires an adequate cell culture system. No adequate model for cells dissociated from the inferior colliculus (IC) has been described and was thus the aim of this study. Therefore, IC were isolated from neonatal rats (P3_5) and a dissociated cell culture was established. In screening experiments using four dissociation methods (Neural Tissue Dissociation Kit [NTDK] T, NTDK P; NTDK PN, and a validated protocol for the dissociation of spiral ganglion neurons [SGN]), the optimal media, and seeding densities were identified. Thereafter, a dissociation protocol containing only the proteolytic enzymes of interest (trypsin or papain) was tested. For analysis, cells were fixed and immunolabeled using glial- and neuron-specific antibodies. Adhesion and survival of dissociated neurons and glial cells isolated from the IC were demonstrated in all experimental settings. Hence, preservation of type-specific cytoarchitecture with sufficient neuronal networks only occurred in cultures dissociated with NTDK P, NTDK PN, and fresh prepared papain solution. However, cultures obtained after dissociation with papain, seeded at a density of 2×10⁴ cells/well and cultivated with Neuro Medium for 6 days reliably revealed the highest neuronal yield with excellent cytoarchitecture of neurons and glial cells. The herein described dissociated culture can be utilized as in vitro model to screen interactions between cells of the IC and surface modifications of the electrode.

Comparison of neurogenesis in the dentate gyrus between the adult and aged gerbil following transient global cerebral ischemia

In the present study, we compared differences in cell proliferation, neuroblast differentiation and neuronal maturation in the hippocampal dentate gyrus (DG) between the adult and aged gerbil induced by 5 min of transient global cerebral ischemia using Ki-67 and BrdU (markers for cell proliferation), doublecortin (DCX, a marker for neuroblast differentiation) and neuronal nuclei (NeuN, a marker for mature neuron). The number of Ki-67-immunoreactive (⁺) cells in the DG of both the groups peaked 7 days after ischemia/reperfusion (I/R). However, the number in the aged DG was 40.6 ± 1.8% of that in the adult DG. Thereafter, the number decreased with time. After ischemic damage, DCX immunoreactivity and its protein level in the adult and aged DG peaked at 10 and 15 days post-ischemia, respectively. However, DCX immunoreactivity and its protein levels in the aged DG were much lower than those in the adult. DCX immunoreactivity and its protein level in the aged DG were 11.1 ± 0.6% and 34.4 ± 2.1% of the adult DG, respectively. In addition, the number of Ki-67⁺ cells and DCX immunoreactivity in both groups were similar to those in the sham at 60 days postischemia. At 30 days post-ischemia, the number of BrdU⁺ cells and BrdU⁺/NeuN⁺ cells in the adult-group were much higher (281.2 ± 23.4% and 126.4 ± 7.4%, respectively) than the aged-group (35.6 ± 6.8% and 79.5 ± 6.1%, respectively). These results suggest that the ability of neurogenesis in the ischemic aged DG is much lower than that in the ischemic adult DG.

Comparison of newly generated doublecortin-immunoreactive neuronal progenitors in the main olfactory bulb among variously aged gerbils

In the present study, we investigated age-related differences in neuronal progenitors in the gerbil main olfactory bulb (MOB) using doublecortin (DCX), a marker for neuronal progenitors which differentiate into neurons in the brain. No difference in the number of neuronal nuclei (NeuN)-immunoreactive neurons was found in the MOB at variously aged gerbils. At postnatal month (PM) 1, DCX immunoreaction was detected in all layers of the MOB except for the olfactory nerve layer. At this time point, DCX-immunoreactive cells (neuronal progenitors) were very abundant; however, they did not have fully developed-processes. From PM 3, the number of DCX-immunoreactive neuronal progenitors was decreased with age. At PM 6, DCX-immunoreactive cells showed very well-developed processes. In western blot analysis, DCX protein level in the MOB was highest at PM 1. Thereafter, levels of DCX protein were decreased with age. In the subventricular zone of the lateral ventricle, the number of Ki-67-immunoractive cells (proliferating cells) was also significantly decreased with age. In addition, increases of α-synuclein-immunoreactive structures were observed in the MOB with age. These results suggest that decrease in DCX-immunoreactive neuronal progenitors and its protein levels in the MOB with age may be associated with reduction of cell proliferation in the SVZ and with an increase in α-synuclein in the MOB.

Organotypic slice co-cultures reveal that early postnatal hippocampal axons lose the ability to grow along the fimbria, while retaining the ability to invade and arborise in septal neuropil

The failure of cut axons to grow along fibre tracts in the adult CNS contrasts with their ability to do so in development. Organotypic slices culture of a number of areas enables the time of failure to be pinpointed to around the second week of postnatal life in the rat. 'Heterochronic' co-culture of slices above and below this age shows that the failure is due to the inability of the older axons to grow into either the same age or younger targets. Using hippocampo-septal slices the present experiments show that this failure is due to an inability to recognise the glial pathway of the fimbria, even when this is of a younger age. However, the older hippocampal neurons retain the ability to grow axons into septal target tissue when they are placed in direct contact with it. This exactly mirrors the inability of cut central axons to regenerate along their previous fibre pathways while they retain their ability to reinnervate neuropil.

Is otospiralin inner ear specific? Evidence for its expression in mouse brain

The small protein otospiralin has initially been identified as an inner ear specific molecule. However, compelling evidence from high throughput sequencing projects suggested that otospiralin is likely expressed in the central nervous system. Here, we tested this hypothesis using a combination of molecular biology, immunological, and histological techniques, and found that otospiralin is expressed in numerous regions of the central nervous system in mouse. In situ hybridization and immunohistochemistry revealed that otospiralin is widely expressed in neuronal cell bodies and glia. Ultrastructural observations in the cerebral cortex located the small protein in close proximity to membranous organelles in perikarya, the inner face of post-synaptic neuronal membranes, and in astrocytic processes. These results are in agreement with the predicted structure of the protein which revealed a single N-terminal transmembrane helix domain followed by a C-terminus cytosolic tail. Interestingly, 2 weeks after a mechanical trauma in the cerebral cortex, otospiralin expression increased in reactive astrocytes located within the vicinity of the site of injury, but not in neurons. Collectively, our observations suggest that otospiralin is possibly involved in signaling pathways, and could play a role in repair mechanisms subsequent to an injury in the central nervous system.

Age-related differentiation in newly generated DCX immunoreactive neurons in the subgranular zone of the gerbil dentate gyrus

In the present study, we investigated age-related changes of newborn neurons in the gerbil dentate gyrus using doublecortin (DCX), a marker of neuronal progenitors which differentiate into neurons in the brain. In the postnatal month 1 (PM 1) group, DCX immunoreactivity was detected in the subgranular zone of the dentate gyrus, but DCX immunoreactive neurons did not have fully developed processes. Thereafter, DCX immunoreactivity and its protein levels in the dentate gyrus were found to decrease with age. Between PM 3 and PM 18, DCX immunoreactive neuronal progenitors showed well-developed processes which projected to the granular layer of the dentate gyrus, but at PM 24, a few DCX immunoreactive neuronal progenitors were detected in the subgranular zone of the dentate gyrus. DCX protein level in the dentate gyrus at PM 1 was high, thereafter levels of DCX were decreased with time. The authors suggest that a decrease of DCX immunoreactivity and its protein level with age may be associated with aging processes in the hippocampal dentate gyrus.

Age-Dependent Changes in Iron Deposition in the Gerbil Hippocampus

In this study, we focused on age-dependent changes in intracellular iron deposition in the gerbil hippocampus. At 1 month of age (PM 1), iron reactivity was weak in the gerbil hippocampus. At this time, cells in the polymorphic layer of the dentate gyrus showed weak iron reactivity. At PM 3, iron reactivity in cells had not changed significantly. Thereafter, iron reactivity in the CA1-3 regions and in the dentate gyrus increased with time until PM 18. At PM 24, iron reactivity in all the subfields was similar to that at PM 18. In animals aged PM 18-24, iron positive cells had various shapes, and had processes which contained iron. These results suggest that the increase of iron deposition may be associated with normal aging and that the iron deposition in the aged hippocampus is different according to hippocampal subfields.

Correlations between neuronal loss, decrease of memory, and decrease expression of brain-derived neurotrophic factor in the gerbil hippocampus during normal aging

It is known that the hippocampus has vital functions in learning and memory, behavioral regulation, and activity-dependent synaptic plasticity, and that the hippocampus contains high levels of brain-derived neurotrophic factor (BDNF). In the present study, we followed age-dependent changes of BDNF immunoreactivity and protein level in the gerbil hippocampus to identify the correlation between BDNF and aging. BDNF immunoreactivity and its protein level significantly increased at postnatal month (PM) 12 in the hippocampus and thereafter reduced. At PM 24, BDNF immunoreactivity in the hippocampal CA1 region and dentate gyrus was similar to that in the PM 1 group, whereas BDNF immunoreactivity in the CA2/3 region at PM 24 was higher than that at PM 1. In the PM 24 group, an age-related neuronal loss and the decrease of reference and working memory were observed. In conclusion, our results suggest that observed reduction in BDNF and reference memory may be associated with age-dependent neuronal loss in the hippocampal CA1 region.

Chronological distribution of Rip immunoreactivity in the gerbil hippocampus during normal aging

Age-dependent studies on oligodendrocytes, which are the myelinating cells in the central nervous system, have been relatively less investigated. We examined age-dependent changes in Rip immunoreactivity and its protein level in the gerbil hippocampus during normal aging using immunohistochemistry and Western blot analysis with Rip antibody, an oligodendrocyte marker. Rip immunoreactivity and its protein level in the hippocampal CA1 region significantly increased at postnatal month 3 (PM 3). Thereafter, they decreased in the hippocampal CA1 region with age. At PM 24, Rip immunoreactive processes in the hippocampal CA1 region markedly decreased in the stratum radiatum. In the hippocampal CA2/3 region and dentate gyrus, the pattern of changes in Rip immunoreactivity and its protein level was similar to those in the hippocampal CA1 region; however, no significant changes were found in the CA2/3 region and dentate gyrus at various age stages. These results indicate that Rip immunoreactivity and protein level in the hippocampal CA1 region decreases significantly at PM 24 compared to the CA2/3 region and dentate gyrus.

Expression analysis of neuroleukin, calmodulin, cortactin, and Rho7/Rnd2 in the intact and injured mouse brain

Subtracted cDNA libraries from the mouse developing inferior colliculus were previously constructed between postnatal day (P) 6 and 10. In the P10-P6 subtracted library, neuroleukin, calmodulin I, cortactin, and Rho7 were identified. The goal of the present study was to analyze their distribution, at the mRNA and protein levels, in both the adult and the developing mouse brain. The four molecules showed a wide expression throughout the brain, with a neuronal-enriched localization in structures such as the cortex, the hippocampus, the cerebellum, and the inferior colliculus. The level of expression of their corresponding mRNAs increased during brain postnatal development. The expression of these molecules was also investigated 2 weeks after a mechanical lesion in the adult cerebral cortex. Neuroleukin and cortactin were found to be expressed by reactive astrocytes, while there were no changes in the expression of calmodulin and Rho7. The expression of neuroleukin, calmodulin, cortactin, and Rho7 is discussed in the context of their putative role in the maturation of the brain and in the axonal regeneration process.

Identification of differentially expressed genes in the developing mouse inferior colliculus

Although injured neurons of inferior colliculus (IC) display a robust axonal outgrowth through a lesion site at postnatal day six (P6) in vitro, and are capable to re-innervate their target cells, injured neurons from P10 IC are unable to regenerate their axons across the lesion site. This axonal regenerative failure has been attributed to an increase of expression of inhibitory molecules in endogenous tissue, during development. As a first step to identify such inhibitory molecules, the present study reports the isolation of molecules differentially expressed in the IC during development. A two-directional (forward and backward) suppression subtractive hybridization (SSH) was performed on IC tissue between P6 and P10 stages. One hundred cDNAs from P6 (P6-P10) and 200 cDNAs from P10 (P10-P6)-subtracted libraries were randomly sequenced. A dot-blot screening of sequenced cDNAs revealed the differential expression for the majority of these cDNAs at their respective developmental stage. Then, the analysis of sequenced clones showed that P6 library was highly enriched in molecules expressed early in the development, such as GAP43 or vimentin proteins. By contrast, the P10 library contained mostly molecules expressed at later stages of development in the central nervous system, such as myelin-related proteins. Our results show that SSH is a suitable method for identifying differentially expressed genes in the developing IC. In addition, these results provide a foundation for further studies dealing with molecules involved in the IC development before and at the onset of hearing, some of which being probably involved in the axonal outgrowth mechanism.

Reconnecting neuronal networks in the auditory brainstem following unilateral deafening

When we disturbed the auditory input of the adult rat by cochleotomy or noise trauma on one side, several substantial anatomical, cellular, and molecular changes took place in the auditory brainstem. We found that: (1) cochleotomy or severe noise trauma both lead to a considerable increase of immunoreactivity of the growth-associated protein GAP-43 in the ventral cochlear nucleus (VCN) of the affected side; (2) the expression of GAP-43 in VCN is restricted to presynaptic endings and short fiber segments; (3) axon collaterals of the cholinergic medial olivocochlear (MOC) neurons are the path along which GAP-43 reaches VCN; (4) partial cochlear lesions induce the emergence of GAP-43 positive presynaptic endings only in regions tonotopically corresponding to the extent of the lesion; (5) judging from the presence of immature fibers and growth cones in VCN on the deafened side, at least part of the GAP-43 positive presynaptic endings appear to be newly formed neuronal contacts following axonal sprouting while others may be modified pre-existing contacts; and (6) GAP-43 positive synapses are formed only on specific postsynaptic profiles, i.e., glutamatergic, glycinergic and calretinin containing cell bodies, but not GABAergic cell bodies. We conclude that unilateral deafening, be it partial or total, induces complex patterns of reconnecting neurons in the adult auditory brainstem, and we evaluate the possibility that the deafness-induced chain of events is optimized to remedy the loss of a bilaterally balanced activity in the auditory brainstem.

Inhibition of collagen synthesis overrides the age-related failure of regeneration of nigrostriatal dopaminergic axons

To investigate the mechanism of the age-related failure of regeneration of transected axons, nigrostriatal dopaminergic axons were unilaterally transected in the lateral hypothalamus in adult mice and in immature mice aged postnatal days 7, 14, and 21. Ten days after the transection, tyrosine hydroxylase-immunoreactive axons had regenerated from caudally to rostrally across the lesion site in mice transected at postnatal day 7, whereas they stopped and did not extend across the lesion site in mice transected at postnatal day 14 or older. Reactive astrocytes bearing chondroitin sulfate proteoglycans were observed around the lesion in mice transected at all ages. However, a fibrotic scar containing type IV collagen-immunoreactive deposits, which was consistently formed at the lesion site in mice transected at postnatal day 14 or older, was not formed in mice lesioned at postnatal day 7. When 2,2'-dipyridyl, an inhibitor of collagen synthesis, was injected into the lesion site at the time of transection in both postnatal day 14 and adult mice, the deposition of type IV collagen and the formation of a fibrotic scar were completely prevented, and large numbers of tyrosine hydroxylase-immunoreactive axons extended across the lesion and reinnervated the striatum. These results imply that the fibrotic scar formed in the lesion site is a crucial impediment to the regeneration of ascending dopaminergic axons in adult mice and suggest that type IV collagen is required for the development of the fibrotic response to adult brain injury.

Cell death and axon regeneration of Purkinje cells after axotomy: challenges of classical hypotheses of axon regeneration

Although adult mammalian neurons are able to regenerate their axons in the peripheral nervous system under certain conditions, they are not able to do it in the central nervous system. The environment surrounding the severed axons appears to be a key factor for axon regeneration. Many studies aiming to enhance axon regeneration in the CNS of adult mammals have successfully manipulated this environment by adding growth permissive molecules and/or neutralizing growth inhibitory molecules. In both cases, the number of axons able to regenerate was low and the different neuronal populations were not equal in their regenerative response, suggesting that manipulation of the environment is not always sufficient. This is particularly well illustrated in the cerebellar system, in which axotomized inferior olivary neurons regenerate when confronted with a permissive environment, whereas mature Purkinje cells do not. The intrinsic ability of a neuron to regenerate its axon is generally correlated with the intensity of its reaction to axotomy (expression of molecules, probability to die). Furthermore, molecules such as GAP-43 (growth-associated molecule) and c-Jun are involved in both axon regeneration and cell death suggesting that these two processes are linked. Surprisingly, Purkinje cells lose their capacity to regenerate their axon (even in the absence of myelin) during development before losing their capacity to react to an axotomy by cell death. These results emphasize the different reactions to axotomy between neuron types and underline that in Purkinje cells, the two cell decisions (axon regeneration and cell death) are differently regulated and therefore not part of the same signaling pathway.

Age-related changes of γ-aminobutyric acid transaminase immunoreactivity in the hippocampus and dentate gyrus of the Mongolian gerbil

We investigated the age-related changes of gamma-aminobutyric acid transaminase (GABA-T, a GABA degradation enzyme) in the hippocampus and dentate gyrus of the gerbil at postnatal month 1 (PM 1), PM 3, PM 6, PM 12, and PM 24. Age-related changes of GABA-T immunoreactivity were distinct in the hippocampal CA1 region and in the dentate gyrus. GABA-T immunoreactivity was weak at PM 1, but at PM 3, it had increased significantly, and then increased further. Between PM 6 and PM 12, strong GABA-T immunoreactivity was found in nonpyramidal cells (GABAergic) in the stratum pyramidale of the CA1 region, and at PM 6, strong GABA-T immunoreactivity was found in neurons of the dentate gyrus subgranular zone. At PM 24, CA1 pyramidal cells showed strong GABA-T immunoreactivity. Western blot analysis showed a pattern of GABA-T expression similar to that shown by immunohistochemistry at various ages. In conclusion, our results suggest that the age-related changes of GABA-T provide important information about the aged brain with GABA dysfunction.

In vitro analysis of mechanisms underlying age-dependent failure of axon regeneration

Severed axons of the inferior colliculus (IC) commissure can regenerate across a lesion in organotypic cultures from postnatal day (P) 6 gerbils, but this regenerative capacity is lost by P12 (Hafidi et al. [ 1995] J Neurosci 15:1298-1307, [1999] J Neurobiol 41:267-280). In the present study, we examined the mechanisms underlying this age-dependent failure of axons to regenerate. In P6-P12 heterochronic cultures, the P12 axons failed to cross the lesion site and project to the contralateral P6 IC lobe. In contrast, axons originating from the P6 lobe could regenerate through the lesion and invade the contralateral P12 IC lobe. To determine whether this age-dependent change in regenerative capacity can develop in organotypic cultures, IC slices with an intact commissure were obtained from P6 animals, grown in vitro for 6 days, and then lesioned at the commissure. In these slices, axon regeneration failure was similar to that observed in normal P12 tissue. Several in vitro treatments enhanced axon regeneration: removal of the entire midline region, inhibition of protein synthesis at the lesion site, and exposure to ABC chondroitinase. Furthermore, when the injured commissural axons were provided with a carpet of C6-R cells (a radial glia-like cell line), significantly more axons projected to the contralateral lobe of the IC. Taken together, these results suggest that the maturation of nonneuronal cells within the lesion site lead to failed axon regeneration in mature animals, and show that ameliorative strategies can be evaluated in vitro.

Macroglia distribution in the developing and adult inferior colliculus

Macroglia distribution in the developing and adult gerbil inferior colliculus (IC) was investigated using two oligodendrocytic [myelin-associated-glycoprotein (MAG) and oligodendrocyte-specific molecule (Rip)] and two astrocytic [glial fibrillary acidic protein (GFAP) and S100] markers and immunohistochemistry. There was a spatio-temporal pattern of myelin marker expression starting in the ventral area of the IC and continuing to the dorsal part of the nucleus. Myelination, as revealed by MAG and Rip markers, starts in the IC during the second postnatal week. The intensity of myelination increased between stages P15 and P21 and extended to the whole IC. The appearance of myelin proteins in the IC may suggest a possible axonal outgrowth inhibition by oligodendrocytes in this structure. A differential pattern of staining was obtained with S100 and GFAP antibodies. Astrocytes identified as S100 immunoreactive cells were observed in the IC by birth and the staining was localized to their cell body and processes. S100 positive cells were homogeneously distributed within the IC nucleus. S100 pattern of staining remained the same in stages P7, P15 and P21. In adult IC, S100 positive cell processes were in contact with neuronal cell bodies, other S100 positive cells and blood vessels. Quantitative analysis showed an increase in the density of positive cells during the first postnatal week and a decrease then after through to adulthood. Unlike S100, GFAP immunoreactivity showed a different pattern of staining. At birth GFAP positive astrocytes were observed along the collicular brain midline and around the IC nucleus delimiting its boundaries. The GFAP pattern of labelling remained the same during development and in the adult. This data suggests the presence of two astrocytes subtypes with different locations in the IC nucleus. The GFAP positive astrocytes were located along the edge of the nucleus, while the S100 positive ones displayed a homogeneous distribution across the nucleus.

Intrinsic properties inhibit axonal outgrowth from neonatal rat spinal cord explant

Lumbar spinal cord explants, harvested from neonatal rat pups aged between postnatal day 0 (P0) and P6, were cultured for a period of 48 hrs in the chemically defined medium R(12) on a poly-ethylene-imine (PEI) and on poly-D-lysin (PDL) coated surface. The outgrowth outside the explant was quantified. Lumbar explants from the same rat and embedded in a collagen matrix, and cortical explants from a P0 rat were used as controls. Statistical analysis demonstrated a clear relation between age-at-explantation and the number of neurites in the corona surrounding the explant. The number of outgrowing neurites decreased sharply with age-at-explantation. The average number of neurites per explant obeyed to the expression log (n) = -0.736x + 3.294 on PEI, and log (n) = -0.721x + 2.295 on PDL; x epsilon in [P0 - P6] (n, the number of neurites per explant; x, the age-at-explantation expressed in postnatal days). A similar observed age-related decrease of outgrowth has been described when culturing the lumbar explant inside a collagen matrix. The phenomenon appears to be an intrinsic property of the explant. We review growth inhibitory properties in different models and propose that the phenomenon occurs here at the interface explant-world.

Regeneration of entorhinal fibers in mouse slice cultures is age dependent and can be stimulated by NT-4, GDNF, and modulators of G-proteins and protein kinase C

Axonal regeneration after lesions is normally not possible in the mature central nervous system, but occurs in the embryonic and neonatal nervous system. Slice cultures offer a convenient experimental system to study the decline of axonal regeneration with increasing maturation of central nervous system tissue. We have used mouse entorhinohippocampal slice cultures to assess regeneration of entorhinal fibers after mechanical lesions in vitro. We found that entorhinal axons regenerate well in cultures derived from postnatal days 5-7 mouse pups when the lesion is made at the second and fourth days in vitro (DIV 2 and DIV 4). Only little regenerative outgrowth is seen after lesions made at DIV 6 and DIV 10. This indicates that a maturation of the cultures occurs within a short time period in vitro resulting in a loss of the regenerative potential. We have used this system to screen for neurotrophic factors and pharmacological compounds that may promote axonal regeneration. Treatments were added to the cultures 1 day before the lesion was made. We found that most added factors did not promote regeneration. Only treatment with the neurotrophic factors NT-4 and GDNF stimulated regeneration in cultures where normally little regeneration is found. A similar improvement of regeneration was found after treatment with pertussis toxin, an inhibitor of G(i)-proteins, and with GF109203X, an inhibitor of protein kinase C. These substances may promote regeneration by interfering with intracellular signaling pathways activated by outgrowth inhibitors. Our findings indicate that the application of neurotrophic factors and the modulation of intracellular signal transduction pathways could be useful strategies to enhance axonal regeneration in a complex microenvironment.