Distinct modulatory actions of TGF-β and LIF on neurotrophin-mediated survival of developing sensory neurons

The neurotrophins nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and neurotrophin-3 (NT-3) are important for the regulation of survival and differentiation of distinct, largely non-overlapping populations of embryonic sensory neurons. We show here that the multifunctional cytokine transforming growth factor-β (TGF-β) fails to maintain sensory neurons cultured from embryonic day (E) 8 chick dorsal root ganglia (DRG), although DRG neurons are immunoreactive for the TGF-β receptor type II, which is essential for TGF-β signaling. However, in combination with various concentrations of NT-3 and NT-4, but not NGF, TGF-β3 causes a further significant increase in neuron survival. In DRG cell cultures treated with NGF, NT-3, and NT-4, a neutralizing antibody to TGF-β decreases neuron survival suggesting that endogenous TGF-β in these cultures affects the efficacies of neurotrophins. Consistent with this notion and a modulatory role of TGF-β in neurotrophin functions is the observation that TGF-β2 and-β3 immunoreactivities and TGF-β3 mRNA are located in embryonic chick DRG in close association with neurons from E5 onwards. We also show that leukemia inhibitory factor (LIF) significantly decreases NGF-mediated DRG neuron survival. Together, these data indicate that actions and efficacies of neurotrophins are under distinct control by TGF-β and LIF in vitro, and possibly also in vivo. Special issue dedicated to Dr. Hans Thoenen.     

Neurotrophin-4: The odd one out in the neurotrophin family

Neurotrophin-4 (NT-4) is a member of a family of neurotrophic factors, the neurotrophins, that control survival and differentiation of vertebrate neurons (2–4). Besides being the most recently discovered neurotrophin in mammals, and the least well understood, several aspects distinguish NT-4 from other members of the neurotrophin family. It is the most divergent member and, in contrast to the other neurotrophins, its expression is ubiquitous and appears to be less influenced by environmental signals. NT-4 seems to have the unique requirement of binding to the lowaffinity neurotrophin receptor (p75^LNGFR) for efficient signalling and retrograde transport in neurons. Moreover, while all other neurotrophin knock-outs have proven lethal during early postnatal development, mice deficient in NT-4 have so far only shown minor cellular deficits and develop normally to adulthood. Is NT-4 a recent addition to the neurotrophic factor repertoire in search of a crucial function, or is it an evolutionary relic, a kind of wisdom tooth of the neurotrophin family?     

Making and breaking the innervation of the ear: neurotrophic support during ear development and its clinical implications

Analyses of single and double mutants of members of the neurotrophin family and their receptors are reviewed. These data demonstrate that the two neurotrophins, brain-derived neurotrophic factor (BDNF) and neurotrophin 3 (NT-3), and their high-affinity receptors trkB and trkC, are the sole support for the developing afferent innervation of the ear. Neurotrophins are first expressed in the otocyst around the time afferent sensory neurons become postmitotic. They are crucial for the survival of certain topologically distinct populations of sensory neurons. BDNF supports all sensory neurons to the semicircular canals, most sensory neurons to the saccule and utricle, and many sensory neurons to the apex and middle turn of the cochlea. In contrast, NT-3 supports few sensory neurons to the utricle and saccule, all sensory neurons to the basal turn of the cochlea and most sensory neurons to the middle and apical turn. Some topologically restricted effects reflect the pattern of neurotrophin distribution as revealed by in situ hybridization (e.g., loss of all innervation to the semicircular canal sensory epithelia in BDNF or trkB mutants). However, other topologically restricted effects cannot be explained on the basis of current knowledge of neurotrophin or neurotrophin receptor distribution. Data on mutants also support the notion that BDNF may play a role in neonatal plastic reorganization of the pattern of innervation in the ear and possibly the brainstem. In contrast, data obtained thus far on the ability of neurotrophins to rescue adult sensory neuron after insults to cochlear hair cells are less compelling. The ear is a model system to test the interactions of the two neurotrophins, BDNF and NT-3, with their two high-affinity receptors, trkB and trkC.     

NT-3 Expression in Spared DRG and the Associated Spinal Laminae as well as Its Anterograde Transport in Sensory Neurons Following Removal of Adjacent DRG in Cats

Neurotrophin-3 plays an important role in survival and differentiation of sensory and sympathetic neurons, sprouting of neurites, synaptic reorganization, and axonal growth. The present study evaluated changes in expression of NT-3 in the spinal cord and L6 dorsal root ganglion (DRG), after ganglionectomy of adjacent dorsal roots in cats. NT-3 immunoreactivity increased at 3 days post-operation (dpo), but decreased at 10 dpo in spinal lamina II after ganglionectomy of L1–L5 and L7–S2 (leaving L6 DRG intact). Conversely, NT-3 immunoreactivity decreased on 3 dpo, but increased on 10 dpo in the nucleus dorsalis. Very little NT-3 mRNA signal was detected in the spinal cord, despite the changes in NT-3 expression. The above changes may be related to changes in NT-3 expression in the DRG. Ganglionectomy of L1–L5 and L7–S2 resulted in increase in NT-3 immunoreactivity and mRNA in small and medium-sized neurons, but decreased expression in large neurons of L6 DRG at 3 dpo. It is possible that increased NT-3 in spinal lamina II is derived from anterograde transport from small- and medium-sized neurons of L6 DRG, whereas decreased NT-3 immunoreactivity in the nucleus dorsalis is due to decreased transport of NT-3 from large neurons in the DRG at this time. This notion is supported by observations on NT-3 distribution in the dorsal root of L6 after ligation of the nerve root. The above results indicate that DRG may be a source of neurotrophic factors such as NT-3 to the spinal cord, and may contribute to plasticity in the spinal cord after injury.     

A chemoattractant role for NT-3 in proprioceptive axon guidance

Neurotrophin-3 (NT-3) is required for proprioceptive neuron survival. Deletion of the proapoptotic gene Bax in NT-3 knockout mice rescues these neurons and allows for examination of their axon growth in the absence of NT-3 signaling. TrkC-positive peripheral and central axons from dorsal root ganglia follow proper trajectories and arrive in close proximity to their targets but fail to innervate them. Peripherally, muscle spindles are absent and TrkC-positive axons do not enter their target muscles. Centrally, proprioceptive axons branch in ectopic regions of the spinal cord, even crossing the midline. In vitro assays reveal chemoattractant effects of NT-3 on dorsal root ganglion axons. Our results show that survival factor NT-3 acts as a short-distance axon guidance molecule for muscle sensory afferents as they approach their proper targets.     

Neurotrophin-5: a novel neurotrophic factor that activates trk and trkB.

In vertebrates, the formation and maintenance of neuronal connections are subject to regulation by multiple target-derived, diffusible (neurotrophic) factors. Here we describe the identification and characterization of a novel neurotrophic factor designated neurotrophin-5 (NT-5). NT-5 is structurally related to nerve growth factor and is expressed in embryonic as well as adult tissues. Recombinant NT-5 promotes the survival of peripheral sensory and sympathetic neurons and induces differentiation of the pheochromocytoma cell line PC12. NT-5 activates two trk-related tyrosine kinase receptors and shares these receptors with other neurotrophins. Activation of multiple receptors may permit a single neurotrophin to control target innervation by distinct neuronal populations. Receptor sharing could enable neurotrophic factors emanating from distinct targets to cooperate in regulating neurons with multiple connections.     

Selective dependence of mammalian dorsal root ganglion neurons on nerve growth factor during embryonic development.

We have investigated the NGF dependence of dorsal root ganglion (DRG) neurons in mammals using a paradigm of multiple in utero injections of a high titer anti-NGF antiserum. We have determined the specificity of our antiserum in relation to other members of the NGF neurotrophin family and found no cross-reactivity with brain-derived neurotrophic factor (BDNF) or neurotrophin-3 (NT-3). To identify various classes of DRG neurons, we have stained their characteristic central projections with Dil. We show here that the NGF dependence of DRG neurons is strikingly selective. Although a majority of DRG neurons are lost after NGF deprivation during embryonic life, these are almost exclusively small diameter neurons that project to laminae I and II of the dorsal horn and presumably subserve nociception and thermoreception. Larger neurons that project to more ventral spinal laminae and subserve other sensory modalities do not require NGF for survival. These NGF-independent DRG neurons likely require one of the more recently identified neurotrophins, BDNF or NT-3.     

Changes in skin levels of two neutotrophins （glial cell line derived neurotrohic factor and neurotrophin-3） cause alterations in cutaneous neuron responses to mechanical stimuli

Neurotrophins are important for the development and maintenance of both high and low threshold mechanoreceptors (HTMRs and LTMRs). In this series of studies, the effects of constitutive overexpression of two different neurotrophins, neurotrophin-3 (NT-3) and glial cell line derived neurotrohic factor (GDNF), were examined. Previous studies indicated that both of them may be implicated in the normal development of mouse dorsal root ganglion (DRG) neurons. Neurons from mice transgenically altered to overexpress NT-3 or GDNF (NT-3-OE or GDNF-OE mice) in the skin were examined using several physiological, immunohistochemi-cal and molecular techniques. Ex vivo skin/nerve/DRG/spinal cord and skin/nerve preparations were used to determine the response characteristics of the cutaneous neurons; immunohistochemistry was used to examine the biochemical phenotype of DRG cells and the skin; RT-PCR was used to examine the levels of candidate ion channels in skin and DRG that may correlate with changes in physiologi-cal responses. In GDNF-OE mice, I-isolectin B4 (IB4)-immunopositive C-HTMRs (nociceptors), a large percentage of which are sensitive to GDNF, had significantly lower mechanical thresholds than wildtype (WT) neurons. Heat thresholds for the same cells were not different. Mechanical sensitivity changes in GDNF-OE mice were correlated with significant increases in acid sensing ion channels 2a (ASIC2a) and 2b (ASIC2b) and transient receptor potential channel AI (TRPAI), all of which are putative mechanosensitive ion channels. Overexpression of NT-3 affected the responses of A-LTMRs and A-HTMRs, hut had no effect on C-HTMRs. Slowly adapting type 1 (SA1) LTMRs and A-HTMRs had increased mechanical sensitivity compared to WT. Mechanical sensitivity was correlated with significant increases in acid-sensing ion channels ASIC1 and ASIC3. This data indicates that both neurotrophins play roles in determining mechanical thresholds of cutaneous HTMRs and LTMRs and that sensitivity changes involve the ASIC family of putative mechanoreceptive ion channels.     

Binding of neurotrophin-3 to its neuronal receptors and interactions with nerve growth factor and brain-derived neurotrophic factor.

Neurotrophin-3 (NT-3) has low-affinity (Kd = 8 x 10(-10) M), as well as high-affinity receptors (Kd = 1.8 x 10(-11) M) on embryonic chick sensory neurons, the latter in surprisingly high numbers. Like the structurally related proteins nerve growth factor (NGF) and brain-derived neurotrophic factor (BDNF), NT-3 also binds to the low-affinity NGF receptor, a molecule that we suggest to designate low-affinity neurotrophin receptor (LANR). NT-3 dissociates from the LANR much more rapidly than BDNF, and more slowly than NGF. The binding of labelled NT-3 to the LANR can be reduced by half using a concentration of BDNF corresponding to the Kd of BDNF to the LANR. In contrast, the binding of NT-3 to its high-affinity neuronal receptors can only be prevented by BDNF or NGF when used at concentrations several thousand-fold higher than those corresponding to their Kd to their high-affinity neuronal receptors. Thus, specific high-affinity NT-3 receptors exist on sensory neurons that can readily discriminate between three structurally related ligands. These findings, including the remarkable property of the LANR to bind three related ligands with similar affinity, but different rate constants, are discussed.     

Fibroblast-like cells from rat plantar skin and neurotrophin-transfected 3T3 fibroblasts influence neurite growth from rat sensory neurons in vitro

Our previous finding that skin-derived and muscle-derived molecules can be used to sort regenerating rat sciatic nerve axons evoked questions concerning neuron-target interactions at the level of single cells, which prompted the present study. The results show that dorsal root ganglion (DRG) neurons co-cultured with fibroblast-like skin-derived cells emit many neurites. These have a proximal linear segment and a distal network of beaded branches in direct relation to skin-derived cells. Electron microscopic examination of such co-cultures showed bundles of neurites at some distance from the target cells and single profiles closely apposed to subjacent cells. RNase protection assay revealed that cultivated skin-derived cells express nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) and neurotrophin-4 (NT-4). In co-cultures of DRG neurons and 3T3 fibroblasts overexpressing either of the neurotrophins produced by skin-derived cells the picture varied. NT-3 transfected 3T3 fibroblasts gave a growth pattern similar to that seen with skin-derived cells. Neurons co-cultured with mock-transfected 3T3 fibroblasts were small and showed weak neurite growth. In co-cultures with a membrane insert between skin-derived cells or 3T3 fibroblasts and DRG neurons few neurons survived and neurite growth was very sparse. We conclude that skin-derived cells stimulate neurite growth from sensory neurons in vitro, that these cells produce NGF, BDNF, NT-3 and NT-4 and that 3T3 fibroblasts producing NT-3 mimic the effect of skin-derived cells on sensory neurons in co-culture. Finally the results suggest that cell surface molecules are important for neuritogenesis.     

Positive and negative interactions of GDNF, NTN and ART in developing sensory neuron subpopulations, and their collaboration with neurotrophins

Glial cell line-derived neurotrophic factor (GDNF), neurturin (NTN) and neublastin/artemin (ART) are distant members of the transforming growth factor beta family, and have been shown to elicit neurotrophic effects upon several classes of peripheral and central neurons. Limited information from in vitro and expression studies has also substantiated a role for GDNF family ligands in mammalian somatosensory neuron development. Here, we show that although dorsal root ganglion (DRG) sensory neurons express GDNF family receptors embryonically, they do not survive in response to their ligands. The regulation of survival emerges postnatally for all GDNF family ligands. GDNF and NTN support distinct subpopulations that can be separated with respect to their expression of GDNF family receptors, whereas ART supports neurons in populations that are also responsive to GDNF or NTN. Sensory neurons that coexpress GDNF family receptors are medium sized, whereas small-caliber nociceptive cells preferentially express a single receptor. In contrast to brain-derived neurotrophic factor (BDNF)-dependent neurons, embryonic nerve growth factor (NGF)-dependent nociceptive neurons switch dependency to GDNF, NTN and ART postnatally. Neurons that survive in the presence of neurotrophin 3 (NT3) or neurotrophin 4 (NT4), including proprioceptive afferents, Merkel end organs and D-hair afferents, are also supported by GDNF family ligands neonatally, although at postnatal stages they lose their dependency on GDNF and NTN. At late postnatal stages, ART prevents survival elicited by GDNF and NTN. These data provide new insights on the roles of GDNF family ligands in sensory neuron development.     

Role of motoneuron-derived neurotrophin 3 in survival and axonal projection of sensory neurons during neural circuit formation

Sensory neurons possess the central and peripheral branches and they form unique spinal neural circuits with motoneurons during development. Peripheral branches of sensory axons fasciculate with the motor axons that extend toward the peripheral muscles from the central nervous system (CNS), whereas the central branches of proprioceptive sensory neurons directly innervate motoneurons. Although anatomically well documented, the molecular mechanism underlying sensory-motor interaction during neural circuit formation is not fully understood. To investigate the role of motoneuron on sensory neuron development, we analyzed sensory neuron phenotypes in the dorsal root ganglia (DRG) of Olig2 knockout (KO) mouse embryos, which lack motoneurons. We found an increased number of apoptotic cells in the DRG of Olig2 KO embryos at embryonic day (E) 10.5. Furthermore, abnormal axonal projections of sensory neurons were observed in both the peripheral branches at E10.5 and central branches at E15.5. To understand the motoneuron-derived factor that regulates sensory neuron development, we focused on neurotrophin 3 (Ntf3; NT-3), because Ntf3 and its receptors (Trk) are strongly expressed in motoneurons and sensory neurons, respectively. The significance of motoneuron-derived Ntf3 was analyzed using Ntf3 conditional knockout (cKO) embryos, in which we observed increased apoptosis and abnormal projection of the central branch innervating motoneuron, the phenotypes being apparently comparable with that of Olig2 KO embryos. Taken together, we show that the motoneuron is a functional source of Ntf3 and motoneuron-derived Ntf3 is an essential pre-target neurotrophin for survival and axonal projection of sensory neurons.     

Effects of neurotrophin and neurotrophin receptor disruption on the afferent inner ear innervation

Two neurotrophins and their two receptors appear to regulate the survival of vestibular and cochlear neurons in the developing ear. Mice lacking either brain derived neurotrophic factor (BDNF) or its associated receptor, Trk B, show a severe reduction in the number of vestibular neurons and a loss of all innervation to the semicircular canals. Mice lacking NT-3 or its receptor, Trk C, show a severe reduction of spiral neurons in the basal turn of the cochlea. Mice lacking both BDNF and NT-3 or Trk B and Trk C, reportedly lose all innervation to the inner ear. These two neurotrophins and their associated receptors are necessary for the normal afferent innervation of the inner ear.     

Expression of neurotrophins and their receptors in peripheral lung cells of mice

Neurotrophins play an essential role in nerve systems. Recent reports indicated that neurotrophins [nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and neurotrophin-4/5 (NT-4/5)] have numerous effects on non-neural cells, especially on immune cells. However, whether lung cells express neurotrophins and/or their receptors (TrkA for NGF, TrkB for BDNF and NT-4/5, and TrkC for NT-3) has never been systematically investigated. We investigated constitutive expression of neurotrophin family and their Trk receptor family in alveolar macrophages and other peripheral lung cells of mice. New findings were: (1) RT-PCR for neurotrophins and their receptors detected NT-3 and NT-4/5 in alveolar macrophages, BDNF, NT-4/5, trkA, the truncated form of trkB, and trkC in lung homogenate, but no trks in alveolar macrophages, (2) immunohistochemistry for neurotrophin receptors detected TrkA in capillary cells, the truncated form of TrkB, and TrkC in interstitial macrophages, (3) immunoelectron microscopy for TrkC revealed expression of TrkC on the surface of interstitial macrophages, and (4) in situ hybridization for neurotrophins detected BDNF in interstitial macrophages and alveolar type I cells, NT-3 in alveolar macrophages, and NT-4/5 in alveolar and interstitial macrophages. These findings indicate that a previously unknown signal trafficking occurs through neurotrophins in peripheral lung.     

Constitutive phosphorylation of TrkC receptors in cultured cerebellar granule neurons might be responsible for the inability of NT-3 to increase neuronal survival and to activate p21 ras

The neurotrophins brain derived neurotrophic factor (BDNF) and neurotrophin-3 (NT-3) are both expressed in developing cerebellum in addition to their tyrosine kinase receptors, TrkB and TrkC. In contrast to BDNF, NT-3 has only a negligible or a transient survival activity on cultured cerebellar granule neurons. The granule neurons however, express both TrkC and Trk B receptors which suggests a basic difference in signaling between BDNF and NT-3 in these neurons. Here we have studied whether this difference can be attributed to the presence of alternative TrkC receptor variants on the granule neurons and which signaling pathway is specifically activated by BDNF but not by NT-3 in these neurons. Using RT-PCR it was shown that the cerebellar granule neurons express the full length TrkC receptor, in addition to variant receptors containing small inserts in the receptor tyrosine kinase domain. There was no dramatic change in the relative amounts of different TrkC receptors during development. However, we found the TrkC receptor constitutively phosphorylated even in the absence of added ligand suggesting an interaction of TrkC with endogenously produced NT-3. In addition, NT-3 was able to phosphorylate the BDNF receptor, TrkB but only at higher concentration (50 ng/ml). There were also distinct differences in the activation of intracellular molecules by BDNF and NT-3. Thus, p21 Ras and PLCγ were activated by BDNF but not by NT-3 whereas both BDNF and NT-3 increased calcium and c-fos mRNA in the granule neurons. These results show that differential activation of specific intracellular pathways such as that of p21 Ras determines the specific effects of BDNF and NT-3 on granule neuron survival. In addition, since calcium is increased by NT-3 in the cerebellar granule neurons, this neurotrophin might have some unknown important effects on these neurons. Special issue dedicated to Dr. Hans Thoenen.     

Differential Age-Dependent Trophic Responses of Nodose,Sensory, and Sympathetic Neurons to Neurotrophins and GDNF: Potencies for Neurite Extension in Explant Culture

The age-dependent trophic responses of sympathetic, sensory, and nodose neurons to the neuro-trophins NGF, BDNF, and NT-3 and to glial cell line-derived neurotrophic factor (GDNF) were examined by an explant culture system. Superior cervical ganglia (SCG), dorsal root ganglia (DRG), and nodose ganglia (NG) were removed from rat embryos (E18), neonatals ( 1 day old), young adults (3–6 months old), and aged adults (>24 months old). The ganglia were cultured with and without each neurotrophic factor; the neurite extension and neurite density were then assessed. The SCG from rats of all ages were significantly influenced by NGF, NT-3, and GDNF; the effects of NT-3 and GDNF were reduced after maturation. The DRG from embryos and neonates were influenced by all neurotrophic factors; however, the effects of BDNF and NT-3 disappeared after maturation. The GDNF showed little effect on adult DRG and no effect on aged DRG. The effect of NGF was preserved over all ages of DRG. The NG from embryonic rats were significantly responsive to BDNF and GDNF; their effects decreased in the neonatal NG, but a minimum effect remained in the aged NG. These results indicate that age-dependent profiles of trophic effects differ extensively among the lineages of the peripheral nervous system and also among the individual neurotrophic factors.     

Axonal outgrowth from adult mouse nodose ganglia in vitro is stimulated by neurotrophin-4 in a Trk receptor and mitogen-activated protein kinase-dependent way

The actions of neurotrophic factors on sensory neurons of the adult nodose ganglion were studied in vitro. The ganglia were explanted in an extracellular matrix-based gel that permitted observation of the growing axons. Neurotrophin-4 (NT-4) was a very efficient stimulator of outgrowth of axons from the nodose ganglion and had almost doubled the outgrowth length when this was analyzed after 2 days in culture. Brain-derived neurotrophic factor also stimulated outgrowth, but to a lesser degree, whereas NT-3 gave only weak stimulatory tendencies. Nerve growth factor and glial cell line-derived neurotrophic factor both lacked stimulatory effects. NT-4 is known to act via TrkB receptors, and the presence of these on growing nodose neurons was demonstrated immunohistochemically. In line with a Trk-mediated growth effect, the NT-4 stimulation was abolished by K252a, a selective inhibitor of neurotrophin receptor-associated tyrosine kinase activity. K252a had no effect on the unstimulated preparation. NT-4 treatment led to activation of the mitogen-activated protein kinase and inhibition of the latter pathway by PD98059 significantly reduced the NT-4 stimulated outgrowth, whereas the drug had no effect on the unstimulated growth. In conclusion, the data suggest that NT-4 can serve as a powerful growth factor for neurons of adult nodose ganglia and that the growth stimulation involves TrkB- and mitogen-activated protein kinase.     

Differential Trk expression in explant and dissociated trigeminal ganglion cell cultures

During embryonic development, expression of neurotrophin receptor tyrosine kinases (Trks) by sensory ganglia is continuously and dynamically regulated. Neurotrophin signaling promotes selective survival and axonal differentiation of sensory neurons. In embryonic day (E) 15 rat trigeminal ganglion (TG), NGF receptor TrkA is expressed by small diameter neurons, NT-3 receptor TrkC and BDNF receptor TrkB are expressed by large diameter neurons. Organotypic explant and dissociated cell cultures of the TG (and dorsal root ganglia) are commonly used to assay neurotrophin effects on developing sensory neurons. In this study, we compared Trk expression in E15 rat TG explant and dissociated cell cultures with or without neurotrophin treatment. Only a subset of TG cells express each of the three Trk receptors in wholemount explant cultures as in vivo conditions. In contrast, all TG neurons co-express all three Trk receptors upon dissociation, regardless of neurotrophin treatment. Neurons cultured in low concentrations of one neurotrophin first, and switched to higher concentrations of another after 1 day, survive and display morphological characteristics of neurons cultured in a mixture of both neurotrophins for 3 days. Our results indicate that wholemount explant cultures of sensory ganglia represent in vivo conditions in terms of Trk expression patterns; whereas dissociation dramatically alters Trk expression by primary sensory neurons.