Point mutation in trkB causes loss of NT4-dependent neurons without major effects on diverse BDNF responses.

Neurotrophins are a family of soluble ligands that promote the survival and differentiation of peripheral and central neurons and regulate synaptic function. The two neurotrophins, brain-derived neurotrophic factor (BDNF) and neurotrophin-4 (NT4), bind and activate a single high-affinity receptor, TrkB. Experiments in cell culture have revealed that an intact Shc adaptor binding site on TrkB and subsequent activation of the Ras/MAPK pathway are important for neuronal survival and neurite outgrowth. To elucidate the intracellular signaling pathways that mediate the diverse effects of BDNF and NT4 in vivo, we have mutated in the mouse germline the Shc binding site in the trkB gene. This trkB(shc) mutation revealed distinctive responses to BDNF and NT4. While nearly all NT4-dependent sensory neurons were lost in trkB(shc/shc) mutant mice, BDNF-dependent neurons were only modestly affected. Activation of MAP kinases and in vitro survival of cultured trkB(shc/shc) neurons were reduced in response to both neurotrophins, with NT4 being less potent than BDNF, suggesting differential activation of TrkB by the two ligands. Moreover, while the Ras/MAPK pathway is required for in vitro differentiation of neuronal cells, trkB(shc/shc) mutant mice do not show any defects in BDNF-dependent differentiation of CNS neurons or in the function of sensory neurons that mediate innocuous touch.

Specific subtypes of cutaneous mechanoreceptors require neurotrophin-3 following peripheral target innervation.

Neurotrophin-3 (NT-3) is required for the development of most sensory neurons of the dorsal root ganglia. Using electrophysiological techniques in mice with null mutations of the NT-3 gene, we show that two functionally specific subsets of cutaneous afferents differentially require this factor: D-hair receptors and slowly adapting mechanoreceptors; other cutaneous receptors were unaffected. Merkel cells, which are the end organs of slowly adapting mechanoreceptors, are virtually absent in 14-day-old homozygous mutants and are severely reduced in adult NT-3 heterozygous animals. This loss of Merkel cells, together with their innervation, happens in the first postnatal weeks of life, in contrast to muscle spindles and afferents, which are never formed in the absence of NT-3. Thus, NT-3 is essential for the maintenance of specific cutaneous afferents known to subserve fine tactile discrimination in humans.

The DRASIC cation channel contributes to the detection of cutaneous touch and acid stimuli in mice.

Cation channels in the DEG/ENaC family are proposed to detect cutaneous stimuli in mammals. We localized one such channel, DRASIC, in several different specialized sensory nerve endings of skin, suggesting it might participate in mechanosensation and/or acid-evoked nociception. Disrupting the mouse DRASIC gene altered sensory transduction in specific and distinct ways. Loss of DRASIC increased the sensitivity of mechanoreceptors detecting light touch, but it reduced the sensitivity of a mechanoreceptor responding to noxious pinch and decreased the response of acid- and noxious heat-sensitive nociceptors. The data suggest that DRASIC subunits participate in heteromultimeric channel complexes in sensory neurons. Moreover, in different cellular contexts, DRASIC may respond to mechanical stimuli or to low pH to mediate normal touch and pain sensation.

A role for BDNF in mechanosensation.

Brain-derived neurotrophic factor (BDNF) is a survival factor for certain sensory neurons during development. Using electrophysiology in BDNF-deficient mice, we show here that slowly adapting mechanoreceptors (SAM), but not other types of cutaneous afferents, require BDNF in postnatal life for normal mechanotransduction. Neurons lacking BDNF did not die, but instead showed a profound and specific reduction in their mechanical sensitivity, which was quantitatively the same in BDNF -/- and BDNF +/- animals. Postnatal treatment of BDNF +/- mice with recombinant BDNF completely rescued the mechanosensitivity deficit. Therefore BDNF is important for regulating SAM mechanosensitivity, independent of any survival-promoting function.

Nerve growth factor and nociception.

Nerve growth factor (NGF) is thought of as a target-derived factor responsible for the survival and maintaining the phenotype of specific sets of peripheral and central neurons during development and maturation. Recently, using physiological techniques, we have shown that specific functional types of nociceptive sensory neurons require NGF, first for survival during development in utero and then for their normal phenotypic development (but not survival) in the early postnatal period. In adulthood, the physiological role of NGF changes dramatically and here it may serve as a link between inflammation and hyperalgesia. Despite apparent changes in NGF's mode of action as the animal matures, it always interacts specifically with nociceptive sensory neurons.

Central hyperexcitability triggered by noxious inputs.

Repetitive activity in unmyelinated sensory afferent neurones, arising from electrical stimuli, tissue injury or nerve damage, can induce long-lasting sensitization in dorsal horn neurones. This process can be blocked by antagonists of the NMDA receptor. In the past year it has emerged that sensory neuropeptides and nitric oxide are also essential mediators of this phenomenon.

Lack of neurotrophin-4 causes selective structural and chemical deficits in sympathetic ganglia and their preganglionic innervation.

Neurotrophin-4 (NT-4) is perhaps the still most enigmatic member of the neurotrophin family. We show here that NT-4 is expressed in neurons of paravertebral and prevertebral sympathetic ganglia, i.e., the superior cervical (SCG), stellate (SG), and celiac (CG) ganglion. Mice deficient for NT-4 showed a significant reduction (20-30%) of preganglionic sympathetic neurons in the intermediolateral column (IML) of the thoracic spinal cord. In contrast, neuron numbers in the SCG, SG, and CG were unchanged. Numbers of axons in the thoracic sympathetic trunk (TST) connecting the SG with lower paravertebral ganglia were also reduced, whereas axon numbers in the cervical sympathetic trunk (CST) were unaltered. Axon losses in the TST were paralleled by losses of synaptic terminals on SG neurons visualized by electron microscopy. Furthermore, immunoreactivity for the synaptic vesicle antigen SV2 was clearly reduced in the SG and CG. Levels of catecholamines and tyrosine hydroxylase immunoreactivity were dramatically reduced in the SG and the CG but not in the SCG. Despite this severe phenotype in the sympathetic system, blood pressure levels were not reduced and displayed a pattern more typical of deficits in baroreceptor afferents. Numbers of IML neurons were unaltered at postnatal day 4, suggesting a postnatal requirement for their maintenance. In light of these and previous data, we hypothesize that NT-4 provided by postganglionic sympathetic neurons is required for establishing and/or maintaining synapses of IML neurons on postganglionic cells. Impairment of synaptic connectivity may consequently reduce impulse flow, causing a reduction in transmitter synthesis in postganglionic neurons.

The sensory mechanotransduction ion channel ASIC2 (acid sensitive ion channel 2) is regulated by neurotrophin availability.

Almost all sensory neurons of the dorsal root ganglia have a mechanosensitive receptive field in the periphery. We have shown that the sensitivity to mechanical stimuli of a subset of sensory neurons that are slowly adapting mechanoreceptors (SAM) is strongly dependent on the availability of brain-derived neurotrophic factor (BDNF). Here we have investigated whether the ASIC2 sodium channel, recently shown by us to be necessary for normal SAM sensitivity, might be regulated by BDNF and thus partially account for the down-regulation of SAM sensitivity seen in BDNF deficient mice. We show that the mRNA for ASIC2 channels is reduced in the DRG of BDNF deficient mice indicating that BDNF might maintain its expression in vivo. We also made short-term cultures of sensory neurons from adult BDNF deficient mice and used a specific antibody to detect the presence of ASIC2 channels in different classes of sensory neurons. We observed that the channel protein was dramatically down-regulated selectively in medium and large diameter neurons and this expression could be rescued in a dose and time dependent manner by addition of BDNF to the culture (10-100 ng/ml). Drugs that block new transcription or protein synthesis also prevented the rescue effects of BDNF. We observed that ASIC2 channels were down-regulated in sensory neurons taken from neurotrophin-4 and neurotrophin-3 deficient mice; these effects might be due to a selective loss of neurons that normally express large amounts of ASIC2 channels. In summary, our data identify the ASIC2 channel as a target of BDNF signaling in vivo and suggest that the functional down-regulation of sensory mechanotransduction in BDNF deficient mice is in part due to loss of ASIC2 expression.

Neurotrophin-3 involvement in the regulation of hair follicle morphogenesis.

Hair follicle epithelium and nervous system share a common ectodermal origin, and some neurotrophins can modulate keratinocyte proliferation and apoptosis. It is therefore reasonable to ask whether growth factors that control neural development are also involved in the regulation of hair follicle morphogenesis. Focusing on neurotrophin-3 (NT-3) and its high-affinity-receptor [tyrosine kinase C (TrkC)], we show that hair placode keratinocytes express TrkC mRNA and immunoreactivity early during murine hair follicle morphogenesis. In later stages of hair follicle development, TrkC mRNA, TrkC-, and NT-3-immunoreactivity are seen in keratinocytes of the proximal hair bulb as well as in dermal papilla fibroblasts. Compared with the corresponding wild-type animals, early stages of hair follicle morphogenesis are significantly accelerated in newborn NT-3 overexpressing mice, whereas these are retarded in newborn heterozygous NT-3 knockout (+/-) mice. These observations suggest that NT-3 is an important growth modulator during morphogenesis and remodeling of neuroectodermal-mesenchymal interaction systems like the hair follicle.

An ultrastructural size principle.

Recent ultrastructural descriptions of synaptic contacts suggest that potential synaptic efficacy may be directly correlated with bouton size. The characteristics of a synaptic bouton which presumably underlie its potential physiological strength (such as vesicle number, active zone number and area, and mitochondrial volume) are all linearly related to the volume of the bouton. Furthermore, at synapses which contact dendritic spines in both the hippocampus and cerebellum, the volume of the spine is linearly related to bouton volume. The existence of these scaling relationships has widespread implications for interpreting synaptic anatomy and variability, and for examining synaptic plasticity. We review evidence in support of the "ultrastructural size principle" outlined above and its potential generality.

Quantitative analysis of peptide levels and neurogenic extravasation following regeneration of afferents to appropriate and inappropriate targets.

We have studied quantitatively the levels of substance P and calcitonin gene-related peptide in nerves innervating skin and muscle of rats, and examined the effects of cross-anastomosing these nerves so that they regenerate to an inappropriate target. We have also compared the ability of nerves to induce neurogenic extravasation with their peptide content. Peptide was measured by radioimmunoassay in the proximal section of ligated peripheral nerves, and neurogenic oedema was measured by determination of Evans Blue extravasation induced by either systemic capsaicin treatment or topical mustard oil application. The levels of these peptides are higher in cutaneous nerves than muscle nerves. This cannot be explained by differences in the number of fibres in the nerves studied. The levels of peptides fall when cutaneous afferents reinnervate muscle, and rise when muscle afferents reinnervate skin. We suggest that these changes occur because of some tissue-specific trophic influence arising from the tissue innervated. The ability to produce extravasation in skin is highly correlated with the substance P and calcitonin gene-related peptide levels of its innervation, even when this occurs in inappropriate nerves which do not normally produce extravasation.

Altered expression of nerve growth factor in the skin of transgenic mice leads to changes in response to mechanical stimuli.

It has recently become clear that the neurotrophic factor, nerve growth factor, interacts specifically with nociceptive sensory neurons during development and maturity. Indeed, it may serve as a critical link between inflammation and the hyperalgesia that ensues in adult animals. Nerve growth factor is normally expressed in limiting amounts in target tissues of sensory and postganglionic sympathetic neurons. In the present study we have altered the basal level of nerve growth factor expression in the skin by producing transgenic mice that express a fusion gene construct containing either a sense or antisense nerve growth factor complementary DNA linked to the K14 keratin promoter. The K14-nerve growth factor transgene (sense or antisense) is abundantly expressed in skin from approximately embryonic day 15 and is then constitutively expressed throughout the life of the animal. In light of the fact that systemic administration of nerve growth factor to neonatal or adult rats leads to hyperalgesia, we have asked whether mice expressing the sense K14-nerve growth factor transgene exhibit similar sensory abnormalities and whether mice expressing the antisense nerve growth factor complementary DNA were hypoalgesic. Here we show that mice over-expressing nerve growth factor in skin display a profound hyperalgesia to noxious mechanical stimulation. Additionally, K14-nerve growth factor antisense mice displayed a profound hypoalgesia to the same stimuli.

Regulation of myelinated nociceptor function by nerve growth factor in neonatal and adult rats.

The role of nerve growth factor (NGF) as a survival factor for sensory neurons during embryonic life has been well documented. Here we examine the actions of NGF or antisera against NGF (anti-NGF) on physiologically identified sensory neurons with myelinated axons later in life, after the dependence on NGF for survival ends. We find that the effects of NGF and anti-NGF are specific for sensory neurons which are nociceptors. Treatments were found to affect the biophysical properties, the development, or the physiological function of myelinated nociceptors. They also affect the animal's behavioral response to noxious stimulation, depending upon when the treatments were given: neonatally, from 2-5 weeks of age, or chronically, beginning at birth. Thus, we find that the actions of NGF are specific for nociceptors but that the function of this neurotrophic factor changes according to the developmental age of the animal.

Neurotrophins and the specification of neuronal phenotype.

Nerve growth factor, brain derived neurotrophic factor and neurotrophin-3 all influence sensory neurons derived from the dorsal root ganglia. Traditionally these neurotrophins have been thought of as survival factors for sensory neurons during their development. Recent evidence from experiments where the in vivo levels of these proteins has been manipulated indicates that they may influence the development of specific sensory neuron phenotypes. In this review these experiments are discussed in relation to the mechanisms by which neurotrophins could influence the phenotypic fate of sensory neurons. The first mechanism requires that when a neuron becomes dependent for survival on a neurotrophin the availability of the factor simply influences the number of neurons surviving with a certain modality. This model requires that neurotrophin responsiveness is a determinant of the possible modalities that the neuron may acquire. The second mechanism requires that the availability of a given neurotrophin influences how many neurons can differentiate into different sensory neuron phenotype independent of survival. The available experimental data is discussed in relation to these two models.

Neurotrophins, nociceptors and pain.

Nerve growth factor (NGF) is known to play a key role in the development of hyperalgesia after inflammatory injury. The increased levels of NGF that accompany injury lead to hyperalgesia via peripheral and central spinal mechanisms. New evidence reviewed here indicates that NGF can directly sensitize nociceptive neurones to noxious heat stimuli through rapid modulation of heat/vanilloid receptors or via de-novo increased expression of heat receptors. In addition, new data suggest that the central sensitization that can result from increased NGF may be mediated via central release of another neurotrophin, brain-derived neurotrophic factor.

Muscle afferents innervating skin form somatotopically appropriate connections in the adult rat dorsal horn.

We have studied the somatotopic reorganization in dorsal horn neurons after a disruption in the normal spatial arrangement of primary sensory axons in adult rats. Muscle afferents were redirected to skin by cutting and cross-anastomosing the hindlimb gastrocnemius nerve (GN) and sural nerve (SN). It has previously been shown that after 10-12 weeks GN afferents innervate the hairy skin of the lateral ankle and calf (previously innervated by SN afferents) and become potentially capable of relaying information on the location and intensity of stimuli applied to the skin. We determined the receptive field and response properties of dorsal horn neurons in the lumbar spinal cord, in regions where the lower hindlimb is normally represented. In control animals (with intact or self-anastomosed sural nerves) very few neurons (< 8%) received any synaptic input from the GN as assessed by electrical stimulation of the nerve. In contrast, when this nerve innervated skin, many cells responded to GN stimulation, and these nearly all had receptive field components in the former SN territory. Moreover, in animals with cross-anastomosed nerves, cells without GN inputs all had receptive fields outside the former SN skin territory. We have shown that in all likelihood GN afferents substituted for SN afferents in subserving the low and high threshold receptive fields of dorsal horn neurons. Furthermore, for many neurons, receptive fields were formed from inappropriately regrown GN afferents and adjacent intact cutaneous afferents (in the tibial or common peroneal nerves). Therefore, when GN afferents innervate skin in adult animals, they alter their central connectivity in an appropriate manner for their new peripheral terminations, so that an orderly somatotopic representation of the hind limb skin is maintained. We suggest that this plasticity of dorsal horn somatotopy is driven in part by activity-dependent mechanisms.

Physiological properties of primary sensory neurons appropriately and inappropriately innervating skin in the adult rat.

1. We have studied the physiology of sensory neurons innervating skin of the rat hindlimb, in three groups of animals: 1) normal animals; 2) animals in which the sural nerve (Sn) had regenerated to its original cutaneous target; and 3) animals in which the gastrocnemius muscle nerve (Gn) had previously been cut and cross anastomosed with the distal stump of the cut Sn so that its axons regenerated to a foreign target, skin. 2. Single-unit recordings were made from 222 afferents in normal, intact animals. They had conduction velocities of 0.5-53.1 m/s. The conduction velocity distribution had distinct peaks at approximately 37.5, 2.5, and 1.25 m/s, presumably corresponding to A alpha beta-, A delta-, and C-fiber populations. Eighty-two percent of the characterized myelinated fibers had low-threshold mechanosensitive receptive fields, whereas 16% were high threshold, and only 2% appeared to have no receptive field. The very large majority of low-threshold mechanosensitive receptive fields (87%) were rapidly adapting hair follicle afferents. 3. In animals with regenerated Sn, 308 afferents were recorded with conduction velocities of 0.4-58.8 m/s. However, the mean conduction velocity was lower than in control animals (P less than 0.05), and only one peak, at 27.5 m/s, was apparent for myelinated fibers. Eighty-six percent of myelinated fibers were low-threshold mechanosensitive afferents, 8.5% were high-threshold mechanoreceptors (HTMRs), and 5.5% appeared to have no receptive fields. Fewer low-threshold mechanoreceptors (LTMRs; compared with controls) were activated by hair movement (63 vs. 87%). Most of the remainder appeared to be field receptors (which were therefore more commonly observed here than in normal animals). 4. In animals in which the Gn had regenerated to skin, 430 afferents were recorded. These had conduction velocities ranging from 0.6 to 71.4 m/s, and again only one peak was apparent in the myelinated conduction velocity histogram, at approximately 17.5 m/s. Of the myelinated fibers, 79% had low-threshold mechanosensitive receptive fields in skin and 10% high-threshold mechanosensitive receptive fields. The remaining 11% apparently had no receptive field (cf. 5.5% in regenerated Sn). In contrast to normal or regrown sural afferents, only 58% of low-threshold gastrocnemius afferents in skin were rapidly adapting. Of the 42% slowly adapting afferents, many surprisingly responded to hair movement. Thus some gastrocnemius afferents seemed to have retained the adaptation properties characteristic of muscle afferents. Also surprisingly, given that the Gn contains fewer fibers than the Sn, receptive-field areas were not significantly different from regrown or normal sural fibers.(ABSTRACT TRUNCATED AT 400 WORDS)

Physiological properties of primary sensory neurons appropriately and inappropriately innervating skeletal muscle in adult rats.

1. We have studied the physiology of primary sensory neurons innervating rat hindlimb muscle in the following: 1) normal control animals; 2) animals in which the gastrocnemius nerve (Gn) had regenerated to its original muscle target; and 3) animals in which the cutaneous sural nerve (Sn) had regenerated to a foreign target, muscle. 2. Single-unit recordings were made from 115 afferents in normal, intact animals. They had conduction velocities of 0.8-67.2 m/s, which were distributed with peaks at approximately 1.25, 17.5, and 47.5 m/s. Of the myelinated fibers, 88% had low-threshold mechanosensitive receptive fields and responded to ramp-and-hold stretches of the muscle. The large majority of these fibers (85%) gave slowly adapting responses to ramp-and-hold stretches or direct muscle probing. Stretch-sensitive afferents could be divided (on the basis of their responses to active muscle contraction) into in-parallel or in-series receptors (presumed muscle spindles and Golgi tendon organs, respectively). The in-parallel receptors outnumbered the in-series receptors by approximately 3:2. The 12% of fibers that were insensitive to stretches of the muscle in the physiological range could be divided into roughly equal groups of totally insensitive fibers and high-threshold fibers, which required excessive stretching of the muscle. 3. In the animals with regrown Gn, 94 single fibers with conduction velocities ranging from 11 to 60.6 m/s were studied. The myelinated conduction velocity distribution exhibited only one peak, at approximately 37.5 m/s. Only 67% of the afferents were stretch sensitive (vs. 88% in normal animals), and only about two-thirds of these (vs. 85% in normal animals) gave slowly adapting responses to ramp-and-hold stretches or muscle probing. The incidence of in-series receptors was also increased among regenerated gastrocnemius afferents. The 33% of fibers that were stretch insensitive were mostly unresponsive to even extreme forms of mechanical stimuli. This group presumably represents afferents that failed to make appropriate endings. 4. In the animals with Sn directed to muscle, 460 single afferents were recorded. Their conduction velocities ranged from 0.7 to 67.9 m/s, and the distribution exhibited only a single peak for myelinated fibers at approximately 22.5 m/s, significantly lower than for intact or regrown Gn. Only 41% of the myelinated fibers were stretch sensitive. Nearly all of these (98%) were rapidly adapting to ramp-and-hold stretches or muscle probing, in marked contrast to the other groups. Also, unlike other groups, nearly all stretch-sensitive afferents appeared to be in-series.(ABSTRACT TRUNCATED AT 400 WORDS)