- Email: [email protected]
Regulation of neurogenesis by neurotrophins in developing spinal sensory ganglia
Brain Research Bulletin, Vol. 57, No. 6, pp. 809 – 816, 2002 Copyright © 2002 Elsevier Science Inc. Printed in the USA. All rights reserved 0361-9230/02/$–see front matter
PII S0361-9230(01)00767-5
Regulation of neurogenesis by neurotrophins in developing spinal sensory ganglia Isabel Farin˜as,* Marife´ Cano-Jaimez, Elena Bellmunt and Mario Soriano Departamento de Biologı´a Celular, Universidad de Valencia, Burjassot, Spain ABSTRACT: Neurons and glia in spinal sensory ganglia derive from multipotent neural crest-derived stem cells. In contrast to neural progenitor cells in the central nervous system, neural crest progenitors coexist with differentiated sensory neurons all throughout the neurogenic period. Thus, developing sensory ganglia are advantageous for determining the possible influence of cell– cell interactions in the regulation of precursor proliferation and neurogenesis. Neurotrophins are important regulators of neuronal survival in the developing vertebrate nervous system and, in addition, they appear to influence precursor behavior in vitro. Studies in mice carrying mutations in neurotrophin genes provide a good system in which to analyze essential actions of these factors on the different developing neural populations. © 2002 Elsevier Science Inc.
lation of neuron number by cell loss, coming largely from the identification of molecules such as neurotrophic factors, secreted factors that promote neuronal survival. Seminal experiments of target ablation and addition in developing vertebrate embryos provided, several decades ago, many examples where the process of naturally occurring cell death undergone by many neurons, at times mostly coincident with the arrival of their axons to their final destinations, appeared to be regulated by the target territories [44]. The neurotrophic hypothesis postulates that neurotrophic factors are produced by target cells in limited amounts, thus engendering a competition among all the neurons projecting to that target for these molecules such that only a fraction of the initial neuronal population can survive (see [32]). This process is thought to ensure an adequate balance between the size of the presynaptic population and that of its postsynaptic target. The best known neurotrophic factors are those of the neurotrophin family (Fig. 1B) which, in mammals, comprises several highly related polypeptides, both functionally and structurally: nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and NT-4/5. These factors appear to function as non-covalently associated homodimers that activate protein tyrosine kinases of the trk family of tyrosine kinase receptors, which includes three members: trkA, trkB, and trkC. TrkA is the receptor for NGF, both BDNF and NT-4/5 activate trkB and NT-3 is the ligand for trkC and is also able to activate trkA and trkC although less efficiently [4,21,45] (Fig. 1B). These interactions and the signal transduction pathways activated by them appear to mediate the major survival-promoting actions of the neurotrophins. Ligand binding to these receptors has been shown to activate several intracellular signaling pathways, including ras, PI-3 kinase/Akt, MEK/MAP kinases, and PLC1, some of which promote cell survival and some of which promote differentiation [21,25]. In addition, all neurotrophins bind with comparable affinity to p75NTR, a transmembrane glycoprotein distantly related to the tumor necrosis factor receptor family, whose function is still controversial and that will not be considered in the present review ([see 17]). Following purification and cloning of the different members of the neurotrophin family, in vitro studies using explants or dissociated neuron-enriched cultures had indicated that each neurotrophin supports the survival of different, albeit overlapping, neuronal populations [8]. Expression studies using in situ hybridization to determine patterns of trk expression (e.g., [41,54]) had also indicated some of the types of neurons that were likely to be responsive to a particular neurotrophin in vivo. More recently, mouse
KEY WORDS: Neurotrophic factors, Mutant mice, Dorsal root ganglia, Precursors, Cell death.
INTRODUCTION Neural function depends on the organization of large numbers of neurons into operational synaptic circuits. The intricacy of the process poses a problem as to how correct numbers of neurons are generated in each population from a pool of precursor cells. Moreover, during development of the vertebrate nervous system an excess of neurons is initially produced than is subsequently pruned down to final numbers by apoptosis [44]. Therefore, the study of how neuron number is regulated must take into account the processes of neuron production and neuron elimination. The final size of any neuronal population is largely determined by the number of neurons generated from their embryonic precursors and those neurons subsequently lost owing to cell death. Heterogeneity among precursor populations in the nervous system is difficult to assess because of the lack of morphological distinguishing features and of appropriate biochemical markers, but different stages of neural precursors are likely to coexist during development, from long-term stem cells, multipotent cells with indefinite self-renewal capacity, to non-self-renewing multipotent progenitors that give rise to progenitors with restrictions in their differentiation potential and from which functionally mature fully differentiated cells will derive (Fig. 1A). The cell dynamics and proportions along this irreversible maturational line will largely determine how many cells of different cell varieties will be generated. Little is known about the mechanisms that regulate the initial pool size, proliferation rate, and number of divisions of neural progenitor cells as well as the process of neurogenesis for a particular neuronal population. We have more knowledge, however, about the regu-
* Address for correspondence: Isabel Farin˜as, Departamento de Biologı´a Celular, Universidad de Valencia, Doctor Moliner 50, 46100 Burjassot, Spain. Fax: 34-96-3864372; E-mail: [email protected]
809
˜ AS ET AL. FARIN
810
strains carrying mutations in genes encoding all neurotrophins and their receptors have provided us with excellent in vivo models to analyze the essential roles of these molecules during the development and differentiation of the mammalian nervous system [47]. In many instances, the findings in the mutant strains have confirmed the predicted specificities. In cases where it had not been possible in vitro to determine the requirements of neurons that cannot be identified on an outgrowth assay, the analysis in vivo has allowed precise identification of the neurons affected in the absence of a particular neurotrophin or its receptor, based on their attributes, such as innervation patterns, terminal endings, cytochemical markers, etc. The study of the peripheral nervous system (PNS) and, particularly, that of the sensory system in these mutant animals has provided us with information as to which neurotrophins are essential for the development of different neuronal populations in vivo [16] (Table 1). Spinal primary sensory neurons comprise morphologically and functionally heterogeneous groups of neurons in the PNS, specialized in the transfer of different modalities of sensory information. (see reviews in [49]). They are grouped in a collection of paired ganglia along the spinal cord, the so-called spinal or dorsal root ganglia (DRG). Each DRG innervates a full array of targets in the periphery, including skin, muscle, and viscera and individual DRG neurons supply specific types of sensory receptors in these tissues and organs, conveying information about position in space (proprioception), pain (nociception), distension or touch (mechanoception), etc to the central nervous system (CNS). SENSORY PHENOTYPES OF NEUROTROPHINDEFICIENT MICE
FIG. 1. (A) Neural populations develop from progenitor cells with different proliferation and differentiation potentials. The neurotrophic hypothesis postulates that during embryogenesis neurons are initially generated in excess and that, coinciding with axon target encounter, they compete for limited amounts of neurotrophic factors, such that only neurons exposed to the factors will eventually survive. (B) Schematic drawing depicting ligand-receptor specificities within the neurotrophin family. Nerve growth factor (NGF) is the ligand for trkA, brain derived neurotrophic factor (BDNF), and neurotrophin (NT)-4/5 can activate trkB and trkC is the receptor for NT-3, although the latter can also activate trkA and trkB at least in some cellular contexts.
In the sensory portion of the PNS, neurotrophins play an essential role in the maintenance of a normal complement of neurons because it appears that virtually all sensory neurons require the presence of at least one neurotrophin during development [47] (Table 1). Moreover, different types of sensory neurons appear to express specific trk receptors and neurotrophins are found to be expressed in peripheral tissues innervated by these neurons. Therefore, sensory populations represent a good system in which to analyze the specificity of action of the different neurotrophins at the cellular level. DRG neurons express all trk receptors throughout development and during postnatal life (see, e.g., [15,38,41,59]. In addition, cultures of embryonic DRG neurons show some survival in response to each of the four neurotrophins [8] and most mouse neurotrophin mutant strains show losses of DRG neurons to a variable extent (Table 1). Because the different mutations result in relatively selective losses, initial analyses in the mutant strains had emphasized the apparent specificity
TABLE 1 PERCENTAGES OF NEURONS AND SENSORY MODALITIES LOST IN DORSAL ROOT GANGLIA OF MUTANTS OF THE NEUROTROPHIN FAMILY Mutation
trkA NGF TrkB BDNF NT-4/5 TrkC NT-3
Neuronal Loss (%)
80 80 30 30 0 20–30 60
Sensory Modality
Nociceptors, thermoreceptors, low-threshold mechanoreceptors Nociceptors, thermoreceptors, low-threshold mechanoreceptors Mechanoreceptors (Meissner) Mechanoreceptors (Meissner) – Proprioceptors Proprioceptors, D-hair, and mechanoreceptors
NGF, nerve growth factor; BDNF, brain derived neurotrophic factor.
NEUROGENESIS IN SENSORY GANGLIA of some neuronal types for a particular neurotrophin signaling pathway, although phenotypic comparisons among mice with different mutations (Table 1) have revealed a considerable overlap in terms of requirements. Some populations appear to require different neurotrophins for survival and/or maintenance during development because a large fraction of spinal sensory neurons exhibit multiple dependencies. Approximately 70%, 30%, and 60% of lumbar sensory neurons are lost in the absence of NGF, BDNF, and NT-3, respectively (Table 1). Although approximately 50% of DRG neurons express trkA in normal mice postnatally [38], more than 80% of all DRG neurons express this receptor during embryonic development [15,41,59]. Accordingly, around 70 – 80% of the DRG neurons are missing at birth in animals lacking this receptor or its ligand [7,40,51,53] (Table 1). The neurons lost include all nociceptive neurons that mediate pain perception and neurons implicated in innocuous thermal and low-threshold mechano-receptive stimuli reception [51]. In agreement with the loss of all nociceptors, DRGs are depleted of small diameter neurons that express calcitonin generelated peptide and substance P, and projections to the most superficial layers of the spinal cord are lost. Physiologically, animals deprived of NGF/trkA signalling during embryonic life have reduced sensitivity to painful stimuli at birth [7,53]. The loss of the non-nociceptive population of trkA dependent neurons is reflected in the elimination of the population of small neurons that specifically bind the IB4 lectin in postnatal DRGs and of axonal projections to the inner portion of layer II in the spinal cord [51]. Nociceptive neurons have been shown to continually express trkA throughout embryonic development and during postnatal life [38]. Other small neurons express trkA during prenatal development and are, therefore, lost following trkA signaling elimination but these neurons stop expressing the receptor after birth. The initial reports on the characterization of mice deficient for trkB or BDNF indicated that approximately 30 –35% of the DRG neurons were lost in these animals 2 weeks after birth [11,23,28] and that NT-4/5 deficient mice did not show any neuron deficits in the DRGs at birth [6,35] (Table 1). More recent work in the same strains indicates that neuronal deficits in trkB and BDNF deficient animals occur postnatally because there appears to be no deficit at birth ([40,52], and unpublished observations), although prenatal deficits in a different BDNF mutant strain of mice have been reported [33,34]. Two week-old mice with targeted deletions of either trkB or BDNF genes appear to lack Meissner corpuscles in the digital glabrous skin along with a small fraction of myelinated axons in the digital nerves supplying the footpads [unpublished results]. The development of these sensory corpuscles requires sensory afferent innervation and, therefore, their absence is a clear indication of a loss of a type of rapidly adapting cutaneous afferents in the DRG of the mutant animals. A similar result has been reported for the palate of the same strain of trkB null mice [22]. As the small number of Meissner-innervating neurons cannot account for the entire neuronal deficit seen in the deficient DRG, other sensory modalities are likely to disappear in the absence of BDNFdependent trkB signaling. NT-3 activation of trkC has been shown to be essential for the survival of the relatively small (⬍20%) population of large-diameter parvalbumin-positive DRG proprioceptive neurons, which supply stretch and tension receptors in muscle and joints and convey information about position of limbs in space. Therefore, NT-3 and trkC-deficient mice show abnormal postures and movements, an absence of the Ia afferent projection of spinal proprioceptive neurons to motor neurons, and a failure of muscle spindles and Golgi tendon organs to form [12,13,29,55,56,58]. Because afferent innervation of the muscle is required for the induction of muscle spindles, the fact that the muscle is never innervated by
811 presumptive proprioceptive afferents, accounts for their absence in NT-3 deficient animals [30]. Comparison between the phenotypes of the NT-3 and trkC deficient animals (Table 1, Fig. 1) has revealed that many more DRG sensory neurons are lost in NT-3 deficient (around 60%) [12,13,33,34,55–57] than in trkC-deficient mice (approximately 20 –30%) [29,52,56]. Interestingly, the lack of NT-3, but not trkC, results in deficits in specific cutaneous afferents, as seen in skin-nerve preparations: D-hair afferents and SA slowly-adapting (SA) mechanoreceptors that innervate specialized Merkel cells in the touch domes of the hairy skin [1]. NEUROTROPHINS AND EARLY SENSORY DEVELOPMENT The neurotrophic hypothesis postulates that neurons become dependent on a particular neurotrophin when their axons come into contact with their final targets and, therefore, the deficits observed in the different neurotrophin mutants could be explained by increased cell death in discrete neuronal populations as a result of the lack of trophic support from their innervating territories. Nevertheless, neurotrophins are expressed during development earlier than would be expected for factors implicated solely in the regulation of target-dependent neuronal survival (see, e.g., [5,14,43,48, 59,60]) and analyses of the trophic requirements of cultured immature sensory neurons isolated from early embryos have indicated that these neurons may be dependent on neurotrophins at times during gangliogenesis, before their axons have established contact with their final targets [5,10]. In addition, some observations have suggested that neurotrophic factors can regulate the proliferation, survival, and differentiation in vitro of certain neural and glial precursors, including sensory, sympathetic, motor, renal, and enteric neural progenitor populations [2,19,24,26,46,50,58]. In particular, NT-3 promotes both the survival and proliferation of isolated chick neural crest cells [19,24,46] and of murine sensory precursors from early DRGs [39]. Therefore, neurotrophins could potentially influence the neuronal complement of a ganglion by regulating neuronal survival, or by affecting precursor cell proliferation, differentiation, and/or survival, or by multiple actions. In order to differentiate among these possibilities, it is necessary to obtain precise data in vivo on the timing of neurotrophin/trk signaling requirements and on the cell types affected by the neurotrophin deficiencies. Because of the precise anatomical location and limits of developing ganglia, sensory neurons are among the most accessible neuronal populations for experimental studies during development, both in vitro and in vivo. The number of neurons in each sensory ganglion appears to be rather constant and can be easily determined during development as well as in different experimental and genetic conditions. Moreover, developing sensory ganglia are advantageous for the study of neuron number establishment because peripheral neurons and their precursors, unlike their CNS counterparts, coexist in the ganglia all throughout development (Fig. 2). Thus, sensory ganglia represent good models in which to analyze how cell numbers are controlled by secreted as well as by cell– cell interactions during development. All DRG neurons derive from neural crest cells that delaminate from the neural tube and migrate away at embryonic stage (E)9 and coalesce to form the first discernible ganglia at E10 in the mouse (Fig. 2). After coalescence, neural crest-derived sensorycommitted precursor cells have the potential to proliferate to increase their number and the potential to differentiate into neurons. Precursor proliferation and neurogenesis take place over the same developmental period, as cells that incorporate the thymidine analogue bromodeoxyuridine (BrdU) and cells that express markers of postmitotic neurons appear to coexist in the ganglia at stages
˜ AS ET AL. FARIN
812
FIG. 2. (A) Spinal ganglia neurons and glial cells are generated from neural crest-derived precursors that migrate at E9 in the mouse. Between E10 and E13 the final number of neurons present in newborn ganglia are generated. Different cell processes, such as precursor proliferation, neurogenesis, and cell death occur simultaneously over that period. (B) Whole-mount staining of an E10 embryo with anti-neurofilament antibodies showing developing peripheral sensory ganglia and their growing projections. (C) Dorsal root ganglia from an E10 embryos stained with cresyl violet. Neurons and precursor cells coexist at this stage but they cannot be recognized based solely on morphology. (D) Precursor, proliferating, cells can be labeled by BrdU immunocytochemistry (black, nuclear) and neurons can be labeled by antibodies to neurofilament (red, cytoplasmic).
immediately after ganglion coalescence [9,14,33]. Birthdating studies in the mouse have indicated that virtually all neurons are generated between E10 and E13 [31]. DRG neuron numbers increase progressively until E13, when the final number of neurons present at birth is reached and, therefore, developmental processes taking place prior to this stage are the primary determinants of the final size of the spinal ganglia (see [14]). Between E12 and E13 the increase in the number of neurons is much more rapid than at earlier stages, a possible indication of a process of terminal neurogenesis. Because neurons cannot be reliably recognized before E15 on morphological criteria alone, the progressive accumulation of postmitotic neurons in early DRGs can be analyzed with antibodies that recognize neuronal-specific antigens, such as neurofilaments [14]. Proliferating cells in the ganglia account for around 50 – 60% of the total number of cells present in the ganglia during the most active period of neurogenesis and they can be marked by cumulative labeling with BrdU. Using repetitive pulses of BrdU
we calculated the cell cycle parameters at E11, such as S-phase duration or length of cell cycle [42] as well as the growth fraction, the proportion of the total population that is proliferating. The growth fraction in the developing ganglia, calculated according to this method, corresponds to the proportion of neurofilament-negative cells in the ganglia [14], suggesting that all precursors found in early ganglia are cycling and that we can estimate the number of precursors as that of neurofilament-negative cells. It is unclear how heterogeneous the precursor population may be but recent genetic studies suggest that more than one type of sensory neuron precursor and at least one type of glial progenitor are likely to coexist. Analyses of mice carrying targeted mutations in genes coding for Drosophila atonal-related bHLH transcription factors neurogenin-1 and -2 (ngn1 and ngn2) have provided genetic evidence for at least two different sensory lineages [36]. It is, unclear, however, how the proliferation and/or differentiation of these different precursor types, or of others yet to be identified, is regulated. More-
NEUROGENESIS IN SENSORY GANGLIA over, even though tritiated thymidine-based birthdating studies have suggested that gliogenesis peaks at E13 [31], glial progenitor cells and differentiated Schwann and satellite cells are already present at E11 in DRGs and spinal nerves [3]. Analyses of neurotrophin mutants suggest that the patterns of expression of trk receptors in early developing, but not in postnatal DRG, are the most consistent with the deficits produced by deletion of corresponding neurotrophins. In early DRGs, cells immunolabeled with antibodies that specifically recognize trkA, trkB, and trkC have a clear neuronal morphology, express neurofilament and do not incorporate the proliferation marker BrdU [15]. Therefore, in spite of the high levels of mRNA for trkB and trkC previously observed in migrating neural crest cells, trk protein expression does not appear to precede neurogenesis. An interesting possibility is that there are high levels of trkB and trkC receptor mRNA expression [27,55] whose translation might be repressed in precursor cells and initiated concomitantly with neuronal differentiation. Initial steps in sensory neurogenesis appear to include induction of trk receptor protein expression, as trk receptor-expressing cells were observed at essentially the same time as neurofilament-expressing cells in these ganglia. As a parallel, in chick embryos there is also strong evidence for the presence of trkC mRNA in migrating neural crest cells with neurogenic potential [19] but trkC protein has been observed only in very few neural crest cells [31a] indicating that the same translational repression may be involved. Interestingly, in zebrafish trkC mRNA expression does not precede neurogenesis [37]. The expression patterns of the different trk proteins are highly dynamic at times of sensory neurogenesis. At E11, trkC appears to be expressed in a majority of neurons, but comparatively few neurons continue to express this receptor at E13. Conversely, while trkA-expressing cells are present at moderate numbers at E11 the vast majority of neurons at E13 express this receptor [15]. Classical birthdating studies had shown that large neurons (presumably including the trkC-positive proprioceptive populations) are born earlier than small neurons (most of which express trkA) in the DRG, suggesting different waves of neurogenesis for DRG neurons [31]. The observation that the number of neurons expressing different trk receptors peak at different embryonic stages could be also interpreted as an indication that the generation of the trk expressing populations follows an orderly neurogenic schedule as it has been recently shown for the trigeminal ganglion (TRG), using a combination of birthdating BrdU labelings and neurofilament immunocytochemistry [20]. This type of study has indicated that while most trkB and trkC-expressing neurons are generated between E10 and E11, the vast majority of trkA expressing neurons is generated between E11 and E13. Moreover, the number of neurons coexpressing multiple trk receptors at these embryonic stages is not significant. Similarly, analysis of trk expression in DRGs of ngn mutants has revealed that ngn2 is transiently required only for the differentiation of trkB and trkC-positive neurons, whereas ngn1 is required later for most or all trkA-positive neurons and a fraction of the trkB and trkC-positive ones [36]. All these data together suggest that changes in receptor expression and neurotrophin dependence could be explained by different waves of neurogenesis and that many neurons, at least in DRG and TRG, express only one receptor from the time they become postmitotic [15,20], although a certain fraction of neurons can switch neurotrophin dependence over time [10]. DRG sensory neurons appear to initiate axon elongation immediately after they are generated (Fig. 2). In the mouse, sensory projections exit the ganglia at E10 but many DRG do not reach their final targets until late in embryogenesis. In the trunk, epidermis is innervated at E13 but target tissues in limbs are innervated at around E13 in the proximal region and as late as E18 in the most
813
FIG. 3. (A) The graph shows the dynamics of neuronal growth in wild-type embryos. The number of neurons increases progressively from the time of coalescence to E13 and then is kept constant. Thus, developmental processes taking place prior to E13 in normal mice are the primary determinants of the final size of the spinal ganglia. The more rapid increase in the number of neurons between E12 and E13 is likely to be due to terminal neurogenesis and therefore to the massive differentiation of neural precursors at the end of the neurogenic period. Apoptosis occurs during the neurogenic period as part of the normal development of these ganglia. (B) In neurotrophin-3 deficient embryos apoptosis is significantly enhanced at E11 and E12 as many more Nissl-stained pyknotic profiles or TUNEL labeled apoptotic nuclei can be observed. In nerve growth factor null embryos, cell death is also significantly increased at E11 as compared to wild types and it remains elevated up to E13.
distal growing region (see [14]). Interestingly, cell death in DRGs of normal embryos peaks at E11 and E12, long before sensory axons have reached their final destinations (Fig. 3) [9,14,59]. Low levels of apoptosis are seen after the initial pruning period but primary modulation of the final number of neurons by cell loss appears to occur before neurons are actually exposed to targetderived neurotrophic factors. Patterns of neurotrophin expression during embryonic development and the deficits seen in certain mutants are in agreement with this early dependence of spinal sensory neurons from tissues other than final targets [14,59]. There are no deficits in the number of precursors or neurons at the time of ganglion formation at E10 in any of the neurotrophin mutants. This indicates that neural crest cells, while migrating, are not affected by the lack of neurotrophins. The finding that neural crest cells do not express detectable levels of trk receptor proteins
˜ AS ET AL. FARIN
814 is consistent with the apparent normal migration of trunk neural crest cells and initial formation of spinal ganglia observed in the mutants ([9,14,33,59]; and unpublished results). In addition, we did not find any deficits in the absence of BDNF, neither in the number of cells nor in the number of neurons during embryogenesis (unpublished results) (but see [33,34]). On the contrary, both the NGF and the NT-3 deficiencies result in loss of neurons between E11 and E13 ([9,14,59]; and unpublished results). At E13, when the full complement of neurons is basically achieved in normal animals, mutants for NGF and NT-3 show losses in the number of neurons that are basically those found in the corresponding mutants at birth, that is 70 – 80% reduction in the number of neurons in the absence of NGF and 60% loss in the absence of NT-3. These neurotrophins, therefore, are likely to play a crucial role in determining the final number of neurons during the most active period of neurogenesis, between E11 and E13, when the full complement of DRG neurons is being generated in normal animals. Moreover, the deficits in the numbers of neurons seen at these stages follow a rather complex pattern if we compare the number of neurons in wild-type and mutant littermates, indicating that these neurotrophins may be involved in more than one aspect of the development of these ganglia [14]: in mutant embryos there is a significant loss in the number of DRG neurons at E11, but at E12, surprisingly, mutant embryos have as many neurons as wildtype embryos. Both NT-3 and NGF mutants show significantly reduced number of neurons in DRGs at E11. Consistently, cell death is significantly elevated in NT-3 and NGF mutants at E11 and in NGF mutants it remains elevated for a longer time ([9,14,59]; and unpublished results). Mutant ganglia also have a higher number of cells expressing activated caspase 3 and caspase 9 (unpublished results). This indicates that neurotrophin deprivation activates at least the cell death pathway controlled by the Apaf1/citochrome c/caspase 9 complex [61]. Interestingly, caspase 3 and 9 are also activated in cells dying during normal development indicating that NT-3 and NGF deficiencies in the DRGs result in apoptosis mediated by the same effectors as in normal development. In our analyses, cell death appears to selectively affect neurons [14,15] and mutations in the NT-3 and the NGF genes selectively deplete different classes of DRG neurons. TrkA-positive neurons selectively undergo apoptosis in the absence of NGF [unpublished results]. As for NT-3, its deficiency results in death of trkB and of trkC expressing neurons [15]. The neurons lost in NT-3 null embryos are likely to include the large proprioceptive neurons that are missing in newborn NT-3 and trkC mutant mice. Large cells in DRGs, including the proprioceptive neurons, are the earliest born neurons and are already present between E10 and E11 [31]. The finding that trkC mutants also have elevated cell death in their DRGs at E11 [59] suggests that the loss of proprioceptive neurons occurs at the same developmental stage in both mutants. At E12, NT-3 mutant embryos have as many neurons as wildtype embryos and fewer precursor cells ([14; and unpublished results) suggesting that neurogenesis occurs at an abnormally higher rate in mutant embryos during the E11–12 interval and that precursors abandon the cell cycle and differentiate into neurons prematurely. This interpretation is supported by our observation that the reduction in the numbers of proliferating cells is not accompanied by changes in their death or proliferation rates [14]. However, some reports have suggested that NT-3 is essential in vivo for the survival of precursor cells in developing DRGs, because cells that incorporated BrdU in NT-3 mutant mice of a different strain were observed to be also positive for the TUNEL labeling of apoptotic cells [9]. However, when lower, non-toxic concentrations of the analog were used, proliferating cells did not co-label with the apoptotic marker [14]; therefore, survival of
precursors does not appear to be compromised in these animals. The lack of NT-3, therefore, could deplete the precursor pool before the normal round of cell divisions occurred. Reduced numbers of precursor cells would prevent in the mutants the increase in the number of neurons that takes place in normal embryos between E12 and E13, at the time of terminal neurogenesis. Thus, much of the final deficit in neuron number found in these mutant mice at birth occurs because some of the neurons were never generated during the major period of neurogenesis. These neurons are likely to include trkA-positive cells born late from ngn1-expressing precursors. These results could suggest that, in normal development, NT-3 keeps the sensory precursors cycling, acting either as mitogen or as repressor of their differentiation program. Evidence for actions of NT-3 on precursor cells have been obtained in vitro in different systems [2,19,24,26,46,50,58]. The data on trk receptor expression, however, argue that the in vivo reported effects of the NT-3 deficiency on sensory neuron precursors are likely to be indirect, as these cells do not appear to express detectable levels of trk proteins [15]. Expression of NT-3 in the limbs at E11 is very high around the growing tips of the projecting sensory and motor axons as well as in close proximity to the DRGs [14,59]. On the contrary, expression of NGF is seen in the skin of developing limbs at a distance from the developing DRG, and, therefore, it is likely to be accessible to projecting neurons but not to precursor cells [59]. Altogether, these results suggest that large reductions in sensory neurons of either type create an environment in which precursor cell differentiation is enhanced, probably by removal of inhibitors of neurogenesis (see, e.g., [18]). Consequently, we have proposed that the premature differentiation of the sensory progenitor cells is a consequence of the enhanced neuronal death [15]. Thus, it could be suggested that the in vivo and in vitro studies are not directly comparable, because the loss of cell-cell interactions or another aspect of cell culture results in expression of the trkC protein and NT-3 responsiveness. Different factors, both intrinsic and extrinsic, are likely to regulate progenitor proliferation and/or differentiation in the developing nervous system. During the development of the mammalian nervous system, many processes such as cell death, proliferation, neurogenesis, and differentiation occur simultaneously for a given neuronal population, complicating the analysis. Moreover, in the complex context of organismic development, the complete elimination of the product of one gene can result in indirect or compensatory changes that occur as a consequence of the primary effects of the mutation. Detailed in vivo analysis will be germane to our understanding of how molecular intrinsic programs imbricate with cell-cell interactions in time and space to produce an appropriate final number of neurons. ACKNOWLEDGEMENTS
Work in the authors’ laboratory has been supported by grants from the Ministerio de Ciencia y Tecnologı´a (SAF1999-0119-C02-01, 1FD972153). M.C.-J. is the recipient of a predoctoral fellowship from Generalitat Valenciana and E.B. is a recipient of a predoctoral fellowship from the Ministerio de Ciencia y Tecnologı´a (FPI).
REFERENCES 1. Airaksinen, M. S.; Koltzenburg, M.; Lewin, G. R.; Masu, Y.; Helbig, C.; Wolf, E.; Brem, G.; Toyka, K. V.; Thoenen, H.; Meyer, M. Specific subtypes of cutaneous mechanoreceptors require neurotrophin-3 following peripheral target innervation. Neuron 16:287–295; 1996. 2. Averbuch-Heller, L.; Pruginin, M.; Kahane, N.; Tsoulfas, P.; Parada, L. F.; Rosenthal, A.; Kalcheim, C. Neurotrophin 3 stimulates the
NEUROGENESIS IN SENSORY GANGLIA
3.
4. 5. 6.
7.
8. 9. 10. 11. 12. 13. 14. 15.
16. 17. 18. 19.
20.
21. 22. 23. 24.
differentiation of motoneurons from avian neural tube progenitor cells. Proc. Natl. Acad. Sci. USA 91:3247–3251; 1994. Britsch, S.; Goerich, D. E.; Riethmacher, D.; Peirano, R. I.; Rossner, M.; Nave, K.-A.; Birchmeier, C.; Wegner, M. The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev. 15:66 –78; 2001. Bothwell, M. Functional interactions of neurotrophins and neurotrophin receptors. Annu. Rev. Neurosci. 18:223–253; 1995. Buchman, V. L.; Davies, A. M. Different neurotrophins are expressed and act in a developmental sequence to promote the survival of embryonic sensory neurons. Development 118:989 –1001; 1993. Conover, J. C.; Erickson, J. T.; Katz, D. M.; Bianchi, L. M.; Poueymirou, W. T.; McClain, J.; Pan, L.; Helgren, M.; Ip, N. Y.; Boland, P.; Friedman, B.; Wiegand, S.; Vejsada, R.; Kato, A. C.; DeChiara, T. M.; Yancopoulos, G. D. Neuronal deficits, not involving motor neurons, in mice lacking BDNF and/or NT4. Nature 375:235–238; 1995. Crowley, C.; Spencer, S. D.; Nishimura, M. C.; Chen, K. S.; PittsMeek, S.; Armanini, M. P.; Ling, G. H.; McMahon, S. B.; Shelton, D. L.; Levinson, A. D.; Phillips, H. S. Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons. Cell 76:1001–1011; 1994. Davies, A. M. The role of neurotrophins in the developing nervous system. J. Neurobiol. 25:1334 –1348; 1994. ElShamy, W. M.; Ernfors, P. A local action of neurotrophin-3 prevents the death of proliferating sensory neuron precursor cells. Neuron 16:963–972; 1996. Enokido, Y.; Wyatt, S.; Davies, A. M. Developmental changes in the response of trigeminal neurons to neurotrophins: Influence of birthdate and the ganglion environment. Development 126:4365– 4373; 1999. Ernfors, P.; Lee, K. F.; Jaenisch, R. Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature 368:147–150; 1994. Ernfors, P.; Lee, K. F.; Kucera, J.; Jaenisch, R. Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents. Cell 77:503–512; 1994. Farin˜ as, I.; Jones, K. R.; Backus, C.; Wang, X.-Y.; Reichardt, L. F. Severe sensory and sympathetic deficits in mice lacking neurotrophin-3. Nature 369:658 – 661; 1994. Farin˜ as, I.; Yoshida, C. K.; Backus, C.; Reichardt, L. F. Lack of neurotrophin-3 results in death of spinal sensory neurons and premature differentiation of their precursors. Neuron 17:1065–1078; 1996. Farin˜ as, I.; Wilkinson, G. A.; Backus, C.; Reichardt, L. F.; Patapoutian, A. Characterization of neurotrophin and trk receptor functions in developing sensory ganglia: Direct NT-3 activation of trkB neurons in vivo. Neuron 21:325–334; 1998. Farin˜ as, I. Neurotrophin actions during the development of the peripheral nervous system. Microsc. Res. Tech. 45:233–242; 1999. Frade, J. M.; Barde, Y. A. Nerve growth factor: Two receptors, multiple functions. Bioessays 2:137–145; 1998. Haddon, C.; Smithers, L.; Schneider-Maunoury, S.; Coche, T.; Henrique, D.; Lewis, J. Multiple delta genes and lateral inhibition in zebrafish primary neurogenesis. Development 125:359 –370; 1998. Henion, P. D.; Garner, A. S.; Large, T. H.; Weston, J. A. trkCmediated NT-3 signalling is required for the early development of a subpopulation of neurogenic neural crest cells. Dev. Biol. 172:602– 613; 1995. Huang, E. J.; Wilkinson, G. A.; Farinas, I.; Backus, C.; Zang, K.; Wong, S. L.; Reichardt, L. F. Expression of Trk receptors in the developing mouse trigeminal ganglion: In vivo evidence for NT-3 activation of TrkA and TrkB in addition to TrkC. Development 126: 2191–2203; 1999. Huang, E. J.; Reichardt, L. F. Neurotrophins: Roles in neuronal development and function. Annu. Rev. Neurosci. 24:677–736; 2001. Ichikawa, H.; Matsuo, S.; Silos-Santiago, I.; Sugimoto, T. Developmental dependency of Meissner corpuscles on trkB but not trkA or trkC. Neuroreport 11:259 –262; 2000. Jones, K. R.; Farin˜ as, I.; Backus, C.; Reichardt, L. F. Targeted disruption of the BDNF gene perturbs brain and sensory neuron development but not motor neuron development. Cell 76:989 –999; 1994. Kalcheim, C.; Carmeli, C.; Rosenthal, A. Neurotrophin 3 is a mitogen
815 for cultured neural crest cells. Proc. Natl. Acad. Sci. USA 89:1661– 1665; 1992. 25. Kaplan, D. R.; Miller, F. D. Neurotrophin signal transduction in the nervous system. Curr. Opin. Neurobiol. 3:381–391; 2000. 26. Karavanov, A.; Sainio, K.; Palgi, J.; Saarma, M.; Saxen, L.; Sariola, H. Neurotrophin 3 rescues neuronal precursors from apoptosis and promotes neuronal differentiation in the embryonic metanephric kidney. Proc. Natl. Acad. Sci. USA 92:11279 –11283; 1995. 27. Klein, R.; Martı´n-Zanca, D.; Barbacid, M.; Parada, L. F. Expression of the tyrosine kinase receptor gene trkB is confined to the murine embryonic and adult nervous system. Development 109:845– 850; 1990. 28. Klein, R.; Smeyne, R. J.; Wurst, W.; Long, L. K.; Auerbach, B. A.; Joyner, A. L.; Barbacid, M. Targeted disruption of the trkB neurotrophin receptor gene results in nervous system lesions and neonatal death. Cell 75:113–122; 1993. 29. Klein, R.; Silos-Santiago, I.; Smeyne, R. J.; Lira, S. A.; Bambrilla, R.; Bryant, S.; Zhang, L.; Snider, W. D.; Barbacid, M. Disruption of the neurotrophin-3 receptor gene trkC eliminates Ia muscle afferents and results in abnormal movements. Nature 368:249 –251; 1994. 30. Kucera, J.; Fan, G.; Jaenisch, R.; Linnarsson, S.; Ernfors, P. Dependence of developing group Ia afferents on neurotrophin-3. J. Comp. Neurol. 363:307–320; 1995. 31. Lawson, S.; Biscoe, T. J. Development of mouse dorsal root ganglia: An autoradiographic and quantitative study. J. Neurocytol. 8:265–274; 1979. 31a.Lefcort, F.; Clary, D. O.; Rusoff, A. C.; Reichardt, L. F. Inhibition of the NT-3 receptor TrkC, early in chick embryogenesis, results in severe reductions in multiple neuronal subpopulations in the dorsal root ganglia. J. Neurosci. 16:3704 –3713; 1996. 32. Levi-Montalcini, R. The nerve growth factor 35 years later. Science 237:1154 –1162; 1987. 33. Liebl, D. J.; Tessarollo, L.; Palko, M. E.; Parada, L. F. Absence of sensory neurons before target innervation in brain derived neurotrophic factor-, neurotrophin3-, and trkC-deficient embryonic mice. J. Neurosci. 17:9113–9121; 1997. 34. Liebl, D. J.; Klesse, L. J.; Tessarollo, L.; Wohlman, T.; Parada, L. F. Loss of brain-derived neurotrophic factor-dependent neural crest-derived sensory neurons in neurotrophin-4 mutant mice. Proc. Natl. Acad. Sci. USA 97:2297–2302; 2000. 35. Liu, X.; Ernfors, P.; Wu, H.; Jaenisch, R. Sensory but not motor neuron deficits in mice lacking NT4 and BDNF. Nature 375:238 –241; 1995. 36. Ma, Q.; Fode, C.; Guillemot, F.; Anderson, D. J. Neurogenin1 and neurogenin2 control two distinct waves of neurogenesis in developing dorsal root ganglia. Genes Dev. 13:1717–1728; 1999. 37. Martin, S. C.; Sandell, J. H.; Heinrich, G. Zebrafish trkC1 and trkC2 receptors define two different cell populations in the nervous system during the period of axonogenesis. Dev. Biol. 195:114 –130; 1998. 38. McMahon, S. B.; Armanini, M. P.; Ling, L. H.; Phillips, H. S. Expression and coexpression of trk receptors in subpopulations of adult primary sensory neurons projecting to identified peripheral targets. Neuron 12:1161–1171; 1994. 39. Memberg, S. P.; Hall, A. K. Proliferation, differentiation, and survival of rat sensory neuron precursors in vitro require specific trophic factors. Mol. Cell Neurosci. 6:323–335; 1995. 40. Minichiello, L.; Piehl, F.; Vazquez, E.; Schimmang, T.; Ho¨ kfelt, T.; Represa, J.; Klein, R. Differential effects of combined trk receptor mutations on dorsal root ganglion and inner ear sensory neurons. Development 121:4067– 4075; 1995. 41. Mu, X.; Silos-Santiago, I.; Carroll, S. L.; Snider, W. D. Neurotrophin receptor genes are expressed in distinct patterns in developing dorsal root ganglia. J. Neurosci. 13:4029 – 4041; 1993. 42. Nowakowski, R. S.; Lewin, S. B.; Miller, M. W. Bromodeoxyuridine immuno-histochemical determination of the lengths of the cell cycle and the synthetic phase for an anatomically defined population. J. Neurocytol. 12:311–318; 1989. 43. O’Conner, R.; Tessier-Lavigne, M. Identification of maxillary factor, a maxillary process-derived chemoattractant for developing trigeminal sensory axons. Neuron 24:165–178; 1999. 44. Oppenheim, R. W. Cell death during development of the nervous system. Annu. Rev. Neurosci. 14:453–501; 1991.
˜ AS ET AL. FARIN
816 45. Patapoutian, A.; Reichardt, L. F. Trk receptors: Mediators of neurotrophin action. Curr. Opin. Neurobiol. 11:272–280; 2001. 46. Pinco, O.; Carmeli, C.; Rosenthal, A.; Kalcheim, C. Neurotrophin-3 affects proliferation and differentiation of distinct neural crest cells and is present in the early neural tube of avian embryos. J. Neurobiol. 24:1626 –1641; 1993. 47. Reichardt, L. F.; Farin˜ as, I. Neurotrophic factors and their receptors: Roles in neuronal development and function. In: Cowan, M. W.; Jessell, T. M.; Zipurski, L., eds. Molecular approaches to neural development. New York: Oxford University Press; 1997:220 –263. 48. Schecterson, L. C.; Bothwell, M. Novel roles for neurotrophins are suggested by BDNF and NT-3 mRNA expression in developing neurons. Neuron 9:449 – 466; 1992. 49. Scott, S. A. Sensory neurons. diversity, development, and plasticity. New York: Oxford University Press; 1992. 50. Sieber-Blum, M. Role of the neurotrophic factors BDNF and NGF in the commitment of pluripotent neural crest cells. Neuron 6:949 –955; 1991. 51. Silos-Santiago, A.; Molliver, D. C.; Ozaki, S.; Smeyne, R. J.; Fagan, A. M.; Barbacid, M.; Snider, W. D. Non-trkA-expressing small DRG neurons are lost in trkA deficient mice. J. Neurosci. 15:5929 –5942; 1995. 52. Silos-Santiago, I.; Fagan, A. M.; Garber, M.; Fritzsch, B.; Barbacid, M. Severe sensory deficits but normal CNS development in newborn mice lacking trkB and trkC tyrosine protein kinase receptors. Eur. J. Neurosci. 9:2045–2056; 1997. 53. Smeyne, R. J.; Klein, R.; Schnapp, A.; Long, L. K.; Bryant, S.; Lewin, S. A.; Barbacid, M. Severe sensory and sympathetic neuropathies in mice carrying a disrupted Trk/NGF. Nature 368:246 –248; 1994. 54. Tessarollo, L.; Tsoulfas, P.; Martin-Zanca, D.; Gilbert, D. J.; Jenkins,
55.
56.
57.
58.
59.
60.
61.
N. A.; Copeland, N. G.; Parada, L. F. trkC, a receptor for neurotrophin-3, is widely expressed in the developing nervous system and in non-neuronal tissues. Development 118:463– 475; 1993. Tessarollo, L.; Vogel, K. S.; Palko, M. E.; Reid, S. W.; Parada, L. F. Targeted mutation in the neurotrophin-3 gene results in loss of muscle sensory neurons. Proc. Natl. Acad. Sci. USA 91:11844 –11848; 1994. Tessarollo, L.; Tsoulfas, P.; Donovan, M. J.; Palko, M. E.; Blair-Flynn, J.; Hempstead, B. L.; Parada, L. F. Targeted deletion of all isoforms of the trkC gene suggests the use of alternate receptors by its ligand neurotrophin-3 in neuronal development and implicates trkC in normal cardiogenesis. Proc. Natl. Acad. Sci. USA 94:14776 –14781; 1997. Tojo, H.; Kaisho, Y.; Nakata, M.; Matsuoka, K.; Kitagawa, M.; Abe, T.; Takami, K.; Yamamoto, M.; Shino, A.; Igarashi, K., et al. Targeted disruption of the neurotrophin-3 gene with lacZ induces loss of trkCpositive neurons in sensory ganglia but not in spinal cords. Brain Res. 669:163–175; 1995. Verdi, J. M.; Anderson, D. J. Neurotrophins regulate sequential changes in neurotrophin receptor expression by sympathetic neuroblasts. Neuron 13:1359 –1372; 1994. White, F. A.; Silos-Santiago, I.; Molliver, D. C.; Nishimura, M.; Phillips, H.; Barbacid, M.; Snider, W. D. Synchronous onset of NGF and trkA survival dependence in developing dorsal root ganglia. J. Neurosci. 16:4662– 4672; 1996. Wilkinson, G. A.; Farin˜ as, I.; Backus, C.; Yoshida, C. K.; Reichardt, L. F. Neurotrophin-3 is a survival factor in vivo for early mouse trigeminal neurons. J. Neurosci. 16:7661–7669; 1996. Yuan, J.; Yankner, B. A. Neuronal apoptosis sculpts the developing brain and has a potentially important role in neurodegenerative diseases. Nature 407:802– 809; 2000.
PII S0361-9230(01)00767-5
Regulation of neurogenesis by neurotrophins in developing spinal sensory ganglia Isabel Farin˜as,* Marife´ Cano-Jaimez, Elena Bellmunt and Mario Soriano Departamento de Biologı´a Celular, Universidad de Valencia, Burjassot, Spain ABSTRACT: Neurons and glia in spinal sensory ganglia derive from multipotent neural crest-derived stem cells. In contrast to neural progenitor cells in the central nervous system, neural crest progenitors coexist with differentiated sensory neurons all throughout the neurogenic period. Thus, developing sensory ganglia are advantageous for determining the possible influence of cell– cell interactions in the regulation of precursor proliferation and neurogenesis. Neurotrophins are important regulators of neuronal survival in the developing vertebrate nervous system and, in addition, they appear to influence precursor behavior in vitro. Studies in mice carrying mutations in neurotrophin genes provide a good system in which to analyze essential actions of these factors on the different developing neural populations. © 2002 Elsevier Science Inc.
lation of neuron number by cell loss, coming largely from the identification of molecules such as neurotrophic factors, secreted factors that promote neuronal survival. Seminal experiments of target ablation and addition in developing vertebrate embryos provided, several decades ago, many examples where the process of naturally occurring cell death undergone by many neurons, at times mostly coincident with the arrival of their axons to their final destinations, appeared to be regulated by the target territories [44]. The neurotrophic hypothesis postulates that neurotrophic factors are produced by target cells in limited amounts, thus engendering a competition among all the neurons projecting to that target for these molecules such that only a fraction of the initial neuronal population can survive (see [32]). This process is thought to ensure an adequate balance between the size of the presynaptic population and that of its postsynaptic target. The best known neurotrophic factors are those of the neurotrophin family (Fig. 1B) which, in mammals, comprises several highly related polypeptides, both functionally and structurally: nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and NT-4/5. These factors appear to function as non-covalently associated homodimers that activate protein tyrosine kinases of the trk family of tyrosine kinase receptors, which includes three members: trkA, trkB, and trkC. TrkA is the receptor for NGF, both BDNF and NT-4/5 activate trkB and NT-3 is the ligand for trkC and is also able to activate trkA and trkC although less efficiently [4,21,45] (Fig. 1B). These interactions and the signal transduction pathways activated by them appear to mediate the major survival-promoting actions of the neurotrophins. Ligand binding to these receptors has been shown to activate several intracellular signaling pathways, including ras, PI-3 kinase/Akt, MEK/MAP kinases, and PLC1, some of which promote cell survival and some of which promote differentiation [21,25]. In addition, all neurotrophins bind with comparable affinity to p75NTR, a transmembrane glycoprotein distantly related to the tumor necrosis factor receptor family, whose function is still controversial and that will not be considered in the present review ([see 17]). Following purification and cloning of the different members of the neurotrophin family, in vitro studies using explants or dissociated neuron-enriched cultures had indicated that each neurotrophin supports the survival of different, albeit overlapping, neuronal populations [8]. Expression studies using in situ hybridization to determine patterns of trk expression (e.g., [41,54]) had also indicated some of the types of neurons that were likely to be responsive to a particular neurotrophin in vivo. More recently, mouse
KEY WORDS: Neurotrophic factors, Mutant mice, Dorsal root ganglia, Precursors, Cell death.
INTRODUCTION Neural function depends on the organization of large numbers of neurons into operational synaptic circuits. The intricacy of the process poses a problem as to how correct numbers of neurons are generated in each population from a pool of precursor cells. Moreover, during development of the vertebrate nervous system an excess of neurons is initially produced than is subsequently pruned down to final numbers by apoptosis [44]. Therefore, the study of how neuron number is regulated must take into account the processes of neuron production and neuron elimination. The final size of any neuronal population is largely determined by the number of neurons generated from their embryonic precursors and those neurons subsequently lost owing to cell death. Heterogeneity among precursor populations in the nervous system is difficult to assess because of the lack of morphological distinguishing features and of appropriate biochemical markers, but different stages of neural precursors are likely to coexist during development, from long-term stem cells, multipotent cells with indefinite self-renewal capacity, to non-self-renewing multipotent progenitors that give rise to progenitors with restrictions in their differentiation potential and from which functionally mature fully differentiated cells will derive (Fig. 1A). The cell dynamics and proportions along this irreversible maturational line will largely determine how many cells of different cell varieties will be generated. Little is known about the mechanisms that regulate the initial pool size, proliferation rate, and number of divisions of neural progenitor cells as well as the process of neurogenesis for a particular neuronal population. We have more knowledge, however, about the regu-
* Address for correspondence: Isabel Farin˜as, Departamento de Biologı´a Celular, Universidad de Valencia, Doctor Moliner 50, 46100 Burjassot, Spain. Fax: 34-96-3864372; E-mail: [email protected]
809
˜ AS ET AL. FARIN
810
strains carrying mutations in genes encoding all neurotrophins and their receptors have provided us with excellent in vivo models to analyze the essential roles of these molecules during the development and differentiation of the mammalian nervous system [47]. In many instances, the findings in the mutant strains have confirmed the predicted specificities. In cases where it had not been possible in vitro to determine the requirements of neurons that cannot be identified on an outgrowth assay, the analysis in vivo has allowed precise identification of the neurons affected in the absence of a particular neurotrophin or its receptor, based on their attributes, such as innervation patterns, terminal endings, cytochemical markers, etc. The study of the peripheral nervous system (PNS) and, particularly, that of the sensory system in these mutant animals has provided us with information as to which neurotrophins are essential for the development of different neuronal populations in vivo [16] (Table 1). Spinal primary sensory neurons comprise morphologically and functionally heterogeneous groups of neurons in the PNS, specialized in the transfer of different modalities of sensory information. (see reviews in [49]). They are grouped in a collection of paired ganglia along the spinal cord, the so-called spinal or dorsal root ganglia (DRG). Each DRG innervates a full array of targets in the periphery, including skin, muscle, and viscera and individual DRG neurons supply specific types of sensory receptors in these tissues and organs, conveying information about position in space (proprioception), pain (nociception), distension or touch (mechanoception), etc to the central nervous system (CNS). SENSORY PHENOTYPES OF NEUROTROPHINDEFICIENT MICE
FIG. 1. (A) Neural populations develop from progenitor cells with different proliferation and differentiation potentials. The neurotrophic hypothesis postulates that during embryogenesis neurons are initially generated in excess and that, coinciding with axon target encounter, they compete for limited amounts of neurotrophic factors, such that only neurons exposed to the factors will eventually survive. (B) Schematic drawing depicting ligand-receptor specificities within the neurotrophin family. Nerve growth factor (NGF) is the ligand for trkA, brain derived neurotrophic factor (BDNF), and neurotrophin (NT)-4/5 can activate trkB and trkC is the receptor for NT-3, although the latter can also activate trkA and trkB at least in some cellular contexts.
In the sensory portion of the PNS, neurotrophins play an essential role in the maintenance of a normal complement of neurons because it appears that virtually all sensory neurons require the presence of at least one neurotrophin during development [47] (Table 1). Moreover, different types of sensory neurons appear to express specific trk receptors and neurotrophins are found to be expressed in peripheral tissues innervated by these neurons. Therefore, sensory populations represent a good system in which to analyze the specificity of action of the different neurotrophins at the cellular level. DRG neurons express all trk receptors throughout development and during postnatal life (see, e.g., [15,38,41,59]. In addition, cultures of embryonic DRG neurons show some survival in response to each of the four neurotrophins [8] and most mouse neurotrophin mutant strains show losses of DRG neurons to a variable extent (Table 1). Because the different mutations result in relatively selective losses, initial analyses in the mutant strains had emphasized the apparent specificity
TABLE 1 PERCENTAGES OF NEURONS AND SENSORY MODALITIES LOST IN DORSAL ROOT GANGLIA OF MUTANTS OF THE NEUROTROPHIN FAMILY Mutation
trkA NGF TrkB BDNF NT-4/5 TrkC NT-3
Neuronal Loss (%)
80 80 30 30 0 20–30 60
Sensory Modality
Nociceptors, thermoreceptors, low-threshold mechanoreceptors Nociceptors, thermoreceptors, low-threshold mechanoreceptors Mechanoreceptors (Meissner) Mechanoreceptors (Meissner) – Proprioceptors Proprioceptors, D-hair, and mechanoreceptors
NGF, nerve growth factor; BDNF, brain derived neurotrophic factor.
NEUROGENESIS IN SENSORY GANGLIA of some neuronal types for a particular neurotrophin signaling pathway, although phenotypic comparisons among mice with different mutations (Table 1) have revealed a considerable overlap in terms of requirements. Some populations appear to require different neurotrophins for survival and/or maintenance during development because a large fraction of spinal sensory neurons exhibit multiple dependencies. Approximately 70%, 30%, and 60% of lumbar sensory neurons are lost in the absence of NGF, BDNF, and NT-3, respectively (Table 1). Although approximately 50% of DRG neurons express trkA in normal mice postnatally [38], more than 80% of all DRG neurons express this receptor during embryonic development [15,41,59]. Accordingly, around 70 – 80% of the DRG neurons are missing at birth in animals lacking this receptor or its ligand [7,40,51,53] (Table 1). The neurons lost include all nociceptive neurons that mediate pain perception and neurons implicated in innocuous thermal and low-threshold mechano-receptive stimuli reception [51]. In agreement with the loss of all nociceptors, DRGs are depleted of small diameter neurons that express calcitonin generelated peptide and substance P, and projections to the most superficial layers of the spinal cord are lost. Physiologically, animals deprived of NGF/trkA signalling during embryonic life have reduced sensitivity to painful stimuli at birth [7,53]. The loss of the non-nociceptive population of trkA dependent neurons is reflected in the elimination of the population of small neurons that specifically bind the IB4 lectin in postnatal DRGs and of axonal projections to the inner portion of layer II in the spinal cord [51]. Nociceptive neurons have been shown to continually express trkA throughout embryonic development and during postnatal life [38]. Other small neurons express trkA during prenatal development and are, therefore, lost following trkA signaling elimination but these neurons stop expressing the receptor after birth. The initial reports on the characterization of mice deficient for trkB or BDNF indicated that approximately 30 –35% of the DRG neurons were lost in these animals 2 weeks after birth [11,23,28] and that NT-4/5 deficient mice did not show any neuron deficits in the DRGs at birth [6,35] (Table 1). More recent work in the same strains indicates that neuronal deficits in trkB and BDNF deficient animals occur postnatally because there appears to be no deficit at birth ([40,52], and unpublished observations), although prenatal deficits in a different BDNF mutant strain of mice have been reported [33,34]. Two week-old mice with targeted deletions of either trkB or BDNF genes appear to lack Meissner corpuscles in the digital glabrous skin along with a small fraction of myelinated axons in the digital nerves supplying the footpads [unpublished results]. The development of these sensory corpuscles requires sensory afferent innervation and, therefore, their absence is a clear indication of a loss of a type of rapidly adapting cutaneous afferents in the DRG of the mutant animals. A similar result has been reported for the palate of the same strain of trkB null mice [22]. As the small number of Meissner-innervating neurons cannot account for the entire neuronal deficit seen in the deficient DRG, other sensory modalities are likely to disappear in the absence of BDNFdependent trkB signaling. NT-3 activation of trkC has been shown to be essential for the survival of the relatively small (⬍20%) population of large-diameter parvalbumin-positive DRG proprioceptive neurons, which supply stretch and tension receptors in muscle and joints and convey information about position of limbs in space. Therefore, NT-3 and trkC-deficient mice show abnormal postures and movements, an absence of the Ia afferent projection of spinal proprioceptive neurons to motor neurons, and a failure of muscle spindles and Golgi tendon organs to form [12,13,29,55,56,58]. Because afferent innervation of the muscle is required for the induction of muscle spindles, the fact that the muscle is never innervated by
811 presumptive proprioceptive afferents, accounts for their absence in NT-3 deficient animals [30]. Comparison between the phenotypes of the NT-3 and trkC deficient animals (Table 1, Fig. 1) has revealed that many more DRG sensory neurons are lost in NT-3 deficient (around 60%) [12,13,33,34,55–57] than in trkC-deficient mice (approximately 20 –30%) [29,52,56]. Interestingly, the lack of NT-3, but not trkC, results in deficits in specific cutaneous afferents, as seen in skin-nerve preparations: D-hair afferents and SA slowly-adapting (SA) mechanoreceptors that innervate specialized Merkel cells in the touch domes of the hairy skin [1]. NEUROTROPHINS AND EARLY SENSORY DEVELOPMENT The neurotrophic hypothesis postulates that neurons become dependent on a particular neurotrophin when their axons come into contact with their final targets and, therefore, the deficits observed in the different neurotrophin mutants could be explained by increased cell death in discrete neuronal populations as a result of the lack of trophic support from their innervating territories. Nevertheless, neurotrophins are expressed during development earlier than would be expected for factors implicated solely in the regulation of target-dependent neuronal survival (see, e.g., [5,14,43,48, 59,60]) and analyses of the trophic requirements of cultured immature sensory neurons isolated from early embryos have indicated that these neurons may be dependent on neurotrophins at times during gangliogenesis, before their axons have established contact with their final targets [5,10]. In addition, some observations have suggested that neurotrophic factors can regulate the proliferation, survival, and differentiation in vitro of certain neural and glial precursors, including sensory, sympathetic, motor, renal, and enteric neural progenitor populations [2,19,24,26,46,50,58]. In particular, NT-3 promotes both the survival and proliferation of isolated chick neural crest cells [19,24,46] and of murine sensory precursors from early DRGs [39]. Therefore, neurotrophins could potentially influence the neuronal complement of a ganglion by regulating neuronal survival, or by affecting precursor cell proliferation, differentiation, and/or survival, or by multiple actions. In order to differentiate among these possibilities, it is necessary to obtain precise data in vivo on the timing of neurotrophin/trk signaling requirements and on the cell types affected by the neurotrophin deficiencies. Because of the precise anatomical location and limits of developing ganglia, sensory neurons are among the most accessible neuronal populations for experimental studies during development, both in vitro and in vivo. The number of neurons in each sensory ganglion appears to be rather constant and can be easily determined during development as well as in different experimental and genetic conditions. Moreover, developing sensory ganglia are advantageous for the study of neuron number establishment because peripheral neurons and their precursors, unlike their CNS counterparts, coexist in the ganglia all throughout development (Fig. 2). Thus, sensory ganglia represent good models in which to analyze how cell numbers are controlled by secreted as well as by cell– cell interactions during development. All DRG neurons derive from neural crest cells that delaminate from the neural tube and migrate away at embryonic stage (E)9 and coalesce to form the first discernible ganglia at E10 in the mouse (Fig. 2). After coalescence, neural crest-derived sensorycommitted precursor cells have the potential to proliferate to increase their number and the potential to differentiate into neurons. Precursor proliferation and neurogenesis take place over the same developmental period, as cells that incorporate the thymidine analogue bromodeoxyuridine (BrdU) and cells that express markers of postmitotic neurons appear to coexist in the ganglia at stages
˜ AS ET AL. FARIN
812
FIG. 2. (A) Spinal ganglia neurons and glial cells are generated from neural crest-derived precursors that migrate at E9 in the mouse. Between E10 and E13 the final number of neurons present in newborn ganglia are generated. Different cell processes, such as precursor proliferation, neurogenesis, and cell death occur simultaneously over that period. (B) Whole-mount staining of an E10 embryo with anti-neurofilament antibodies showing developing peripheral sensory ganglia and their growing projections. (C) Dorsal root ganglia from an E10 embryos stained with cresyl violet. Neurons and precursor cells coexist at this stage but they cannot be recognized based solely on morphology. (D) Precursor, proliferating, cells can be labeled by BrdU immunocytochemistry (black, nuclear) and neurons can be labeled by antibodies to neurofilament (red, cytoplasmic).
immediately after ganglion coalescence [9,14,33]. Birthdating studies in the mouse have indicated that virtually all neurons are generated between E10 and E13 [31]. DRG neuron numbers increase progressively until E13, when the final number of neurons present at birth is reached and, therefore, developmental processes taking place prior to this stage are the primary determinants of the final size of the spinal ganglia (see [14]). Between E12 and E13 the increase in the number of neurons is much more rapid than at earlier stages, a possible indication of a process of terminal neurogenesis. Because neurons cannot be reliably recognized before E15 on morphological criteria alone, the progressive accumulation of postmitotic neurons in early DRGs can be analyzed with antibodies that recognize neuronal-specific antigens, such as neurofilaments [14]. Proliferating cells in the ganglia account for around 50 – 60% of the total number of cells present in the ganglia during the most active period of neurogenesis and they can be marked by cumulative labeling with BrdU. Using repetitive pulses of BrdU
we calculated the cell cycle parameters at E11, such as S-phase duration or length of cell cycle [42] as well as the growth fraction, the proportion of the total population that is proliferating. The growth fraction in the developing ganglia, calculated according to this method, corresponds to the proportion of neurofilament-negative cells in the ganglia [14], suggesting that all precursors found in early ganglia are cycling and that we can estimate the number of precursors as that of neurofilament-negative cells. It is unclear how heterogeneous the precursor population may be but recent genetic studies suggest that more than one type of sensory neuron precursor and at least one type of glial progenitor are likely to coexist. Analyses of mice carrying targeted mutations in genes coding for Drosophila atonal-related bHLH transcription factors neurogenin-1 and -2 (ngn1 and ngn2) have provided genetic evidence for at least two different sensory lineages [36]. It is, unclear, however, how the proliferation and/or differentiation of these different precursor types, or of others yet to be identified, is regulated. More-
NEUROGENESIS IN SENSORY GANGLIA over, even though tritiated thymidine-based birthdating studies have suggested that gliogenesis peaks at E13 [31], glial progenitor cells and differentiated Schwann and satellite cells are already present at E11 in DRGs and spinal nerves [3]. Analyses of neurotrophin mutants suggest that the patterns of expression of trk receptors in early developing, but not in postnatal DRG, are the most consistent with the deficits produced by deletion of corresponding neurotrophins. In early DRGs, cells immunolabeled with antibodies that specifically recognize trkA, trkB, and trkC have a clear neuronal morphology, express neurofilament and do not incorporate the proliferation marker BrdU [15]. Therefore, in spite of the high levels of mRNA for trkB and trkC previously observed in migrating neural crest cells, trk protein expression does not appear to precede neurogenesis. An interesting possibility is that there are high levels of trkB and trkC receptor mRNA expression [27,55] whose translation might be repressed in precursor cells and initiated concomitantly with neuronal differentiation. Initial steps in sensory neurogenesis appear to include induction of trk receptor protein expression, as trk receptor-expressing cells were observed at essentially the same time as neurofilament-expressing cells in these ganglia. As a parallel, in chick embryos there is also strong evidence for the presence of trkC mRNA in migrating neural crest cells with neurogenic potential [19] but trkC protein has been observed only in very few neural crest cells [31a] indicating that the same translational repression may be involved. Interestingly, in zebrafish trkC mRNA expression does not precede neurogenesis [37]. The expression patterns of the different trk proteins are highly dynamic at times of sensory neurogenesis. At E11, trkC appears to be expressed in a majority of neurons, but comparatively few neurons continue to express this receptor at E13. Conversely, while trkA-expressing cells are present at moderate numbers at E11 the vast majority of neurons at E13 express this receptor [15]. Classical birthdating studies had shown that large neurons (presumably including the trkC-positive proprioceptive populations) are born earlier than small neurons (most of which express trkA) in the DRG, suggesting different waves of neurogenesis for DRG neurons [31]. The observation that the number of neurons expressing different trk receptors peak at different embryonic stages could be also interpreted as an indication that the generation of the trk expressing populations follows an orderly neurogenic schedule as it has been recently shown for the trigeminal ganglion (TRG), using a combination of birthdating BrdU labelings and neurofilament immunocytochemistry [20]. This type of study has indicated that while most trkB and trkC-expressing neurons are generated between E10 and E11, the vast majority of trkA expressing neurons is generated between E11 and E13. Moreover, the number of neurons coexpressing multiple trk receptors at these embryonic stages is not significant. Similarly, analysis of trk expression in DRGs of ngn mutants has revealed that ngn2 is transiently required only for the differentiation of trkB and trkC-positive neurons, whereas ngn1 is required later for most or all trkA-positive neurons and a fraction of the trkB and trkC-positive ones [36]. All these data together suggest that changes in receptor expression and neurotrophin dependence could be explained by different waves of neurogenesis and that many neurons, at least in DRG and TRG, express only one receptor from the time they become postmitotic [15,20], although a certain fraction of neurons can switch neurotrophin dependence over time [10]. DRG sensory neurons appear to initiate axon elongation immediately after they are generated (Fig. 2). In the mouse, sensory projections exit the ganglia at E10 but many DRG do not reach their final targets until late in embryogenesis. In the trunk, epidermis is innervated at E13 but target tissues in limbs are innervated at around E13 in the proximal region and as late as E18 in the most
813
FIG. 3. (A) The graph shows the dynamics of neuronal growth in wild-type embryos. The number of neurons increases progressively from the time of coalescence to E13 and then is kept constant. Thus, developmental processes taking place prior to E13 in normal mice are the primary determinants of the final size of the spinal ganglia. The more rapid increase in the number of neurons between E12 and E13 is likely to be due to terminal neurogenesis and therefore to the massive differentiation of neural precursors at the end of the neurogenic period. Apoptosis occurs during the neurogenic period as part of the normal development of these ganglia. (B) In neurotrophin-3 deficient embryos apoptosis is significantly enhanced at E11 and E12 as many more Nissl-stained pyknotic profiles or TUNEL labeled apoptotic nuclei can be observed. In nerve growth factor null embryos, cell death is also significantly increased at E11 as compared to wild types and it remains elevated up to E13.
distal growing region (see [14]). Interestingly, cell death in DRGs of normal embryos peaks at E11 and E12, long before sensory axons have reached their final destinations (Fig. 3) [9,14,59]. Low levels of apoptosis are seen after the initial pruning period but primary modulation of the final number of neurons by cell loss appears to occur before neurons are actually exposed to targetderived neurotrophic factors. Patterns of neurotrophin expression during embryonic development and the deficits seen in certain mutants are in agreement with this early dependence of spinal sensory neurons from tissues other than final targets [14,59]. There are no deficits in the number of precursors or neurons at the time of ganglion formation at E10 in any of the neurotrophin mutants. This indicates that neural crest cells, while migrating, are not affected by the lack of neurotrophins. The finding that neural crest cells do not express detectable levels of trk receptor proteins
˜ AS ET AL. FARIN
814 is consistent with the apparent normal migration of trunk neural crest cells and initial formation of spinal ganglia observed in the mutants ([9,14,33,59]; and unpublished results). In addition, we did not find any deficits in the absence of BDNF, neither in the number of cells nor in the number of neurons during embryogenesis (unpublished results) (but see [33,34]). On the contrary, both the NGF and the NT-3 deficiencies result in loss of neurons between E11 and E13 ([9,14,59]; and unpublished results). At E13, when the full complement of neurons is basically achieved in normal animals, mutants for NGF and NT-3 show losses in the number of neurons that are basically those found in the corresponding mutants at birth, that is 70 – 80% reduction in the number of neurons in the absence of NGF and 60% loss in the absence of NT-3. These neurotrophins, therefore, are likely to play a crucial role in determining the final number of neurons during the most active period of neurogenesis, between E11 and E13, when the full complement of DRG neurons is being generated in normal animals. Moreover, the deficits in the numbers of neurons seen at these stages follow a rather complex pattern if we compare the number of neurons in wild-type and mutant littermates, indicating that these neurotrophins may be involved in more than one aspect of the development of these ganglia [14]: in mutant embryos there is a significant loss in the number of DRG neurons at E11, but at E12, surprisingly, mutant embryos have as many neurons as wildtype embryos. Both NT-3 and NGF mutants show significantly reduced number of neurons in DRGs at E11. Consistently, cell death is significantly elevated in NT-3 and NGF mutants at E11 and in NGF mutants it remains elevated for a longer time ([9,14,59]; and unpublished results). Mutant ganglia also have a higher number of cells expressing activated caspase 3 and caspase 9 (unpublished results). This indicates that neurotrophin deprivation activates at least the cell death pathway controlled by the Apaf1/citochrome c/caspase 9 complex [61]. Interestingly, caspase 3 and 9 are also activated in cells dying during normal development indicating that NT-3 and NGF deficiencies in the DRGs result in apoptosis mediated by the same effectors as in normal development. In our analyses, cell death appears to selectively affect neurons [14,15] and mutations in the NT-3 and the NGF genes selectively deplete different classes of DRG neurons. TrkA-positive neurons selectively undergo apoptosis in the absence of NGF [unpublished results]. As for NT-3, its deficiency results in death of trkB and of trkC expressing neurons [15]. The neurons lost in NT-3 null embryos are likely to include the large proprioceptive neurons that are missing in newborn NT-3 and trkC mutant mice. Large cells in DRGs, including the proprioceptive neurons, are the earliest born neurons and are already present between E10 and E11 [31]. The finding that trkC mutants also have elevated cell death in their DRGs at E11 [59] suggests that the loss of proprioceptive neurons occurs at the same developmental stage in both mutants. At E12, NT-3 mutant embryos have as many neurons as wildtype embryos and fewer precursor cells ([14; and unpublished results) suggesting that neurogenesis occurs at an abnormally higher rate in mutant embryos during the E11–12 interval and that precursors abandon the cell cycle and differentiate into neurons prematurely. This interpretation is supported by our observation that the reduction in the numbers of proliferating cells is not accompanied by changes in their death or proliferation rates [14]. However, some reports have suggested that NT-3 is essential in vivo for the survival of precursor cells in developing DRGs, because cells that incorporated BrdU in NT-3 mutant mice of a different strain were observed to be also positive for the TUNEL labeling of apoptotic cells [9]. However, when lower, non-toxic concentrations of the analog were used, proliferating cells did not co-label with the apoptotic marker [14]; therefore, survival of
precursors does not appear to be compromised in these animals. The lack of NT-3, therefore, could deplete the precursor pool before the normal round of cell divisions occurred. Reduced numbers of precursor cells would prevent in the mutants the increase in the number of neurons that takes place in normal embryos between E12 and E13, at the time of terminal neurogenesis. Thus, much of the final deficit in neuron number found in these mutant mice at birth occurs because some of the neurons were never generated during the major period of neurogenesis. These neurons are likely to include trkA-positive cells born late from ngn1-expressing precursors. These results could suggest that, in normal development, NT-3 keeps the sensory precursors cycling, acting either as mitogen or as repressor of their differentiation program. Evidence for actions of NT-3 on precursor cells have been obtained in vitro in different systems [2,19,24,26,46,50,58]. The data on trk receptor expression, however, argue that the in vivo reported effects of the NT-3 deficiency on sensory neuron precursors are likely to be indirect, as these cells do not appear to express detectable levels of trk proteins [15]. Expression of NT-3 in the limbs at E11 is very high around the growing tips of the projecting sensory and motor axons as well as in close proximity to the DRGs [14,59]. On the contrary, expression of NGF is seen in the skin of developing limbs at a distance from the developing DRG, and, therefore, it is likely to be accessible to projecting neurons but not to precursor cells [59]. Altogether, these results suggest that large reductions in sensory neurons of either type create an environment in which precursor cell differentiation is enhanced, probably by removal of inhibitors of neurogenesis (see, e.g., [18]). Consequently, we have proposed that the premature differentiation of the sensory progenitor cells is a consequence of the enhanced neuronal death [15]. Thus, it could be suggested that the in vivo and in vitro studies are not directly comparable, because the loss of cell-cell interactions or another aspect of cell culture results in expression of the trkC protein and NT-3 responsiveness. Different factors, both intrinsic and extrinsic, are likely to regulate progenitor proliferation and/or differentiation in the developing nervous system. During the development of the mammalian nervous system, many processes such as cell death, proliferation, neurogenesis, and differentiation occur simultaneously for a given neuronal population, complicating the analysis. Moreover, in the complex context of organismic development, the complete elimination of the product of one gene can result in indirect or compensatory changes that occur as a consequence of the primary effects of the mutation. Detailed in vivo analysis will be germane to our understanding of how molecular intrinsic programs imbricate with cell-cell interactions in time and space to produce an appropriate final number of neurons. ACKNOWLEDGEMENTS
Work in the authors’ laboratory has been supported by grants from the Ministerio de Ciencia y Tecnologı´a (SAF1999-0119-C02-01, 1FD972153). M.C.-J. is the recipient of a predoctoral fellowship from Generalitat Valenciana and E.B. is a recipient of a predoctoral fellowship from the Ministerio de Ciencia y Tecnologı´a (FPI).
REFERENCES 1. Airaksinen, M. S.; Koltzenburg, M.; Lewin, G. R.; Masu, Y.; Helbig, C.; Wolf, E.; Brem, G.; Toyka, K. V.; Thoenen, H.; Meyer, M. Specific subtypes of cutaneous mechanoreceptors require neurotrophin-3 following peripheral target innervation. Neuron 16:287–295; 1996. 2. Averbuch-Heller, L.; Pruginin, M.; Kahane, N.; Tsoulfas, P.; Parada, L. F.; Rosenthal, A.; Kalcheim, C. Neurotrophin 3 stimulates the
NEUROGENESIS IN SENSORY GANGLIA
3.
4. 5. 6.
7.
8. 9. 10. 11. 12. 13. 14. 15.
16. 17. 18. 19.
20.
21. 22. 23. 24.
differentiation of motoneurons from avian neural tube progenitor cells. Proc. Natl. Acad. Sci. USA 91:3247–3251; 1994. Britsch, S.; Goerich, D. E.; Riethmacher, D.; Peirano, R. I.; Rossner, M.; Nave, K.-A.; Birchmeier, C.; Wegner, M. The transcription factor Sox10 is a key regulator of peripheral glial development. Genes Dev. 15:66 –78; 2001. Bothwell, M. Functional interactions of neurotrophins and neurotrophin receptors. Annu. Rev. Neurosci. 18:223–253; 1995. Buchman, V. L.; Davies, A. M. Different neurotrophins are expressed and act in a developmental sequence to promote the survival of embryonic sensory neurons. Development 118:989 –1001; 1993. Conover, J. C.; Erickson, J. T.; Katz, D. M.; Bianchi, L. M.; Poueymirou, W. T.; McClain, J.; Pan, L.; Helgren, M.; Ip, N. Y.; Boland, P.; Friedman, B.; Wiegand, S.; Vejsada, R.; Kato, A. C.; DeChiara, T. M.; Yancopoulos, G. D. Neuronal deficits, not involving motor neurons, in mice lacking BDNF and/or NT4. Nature 375:235–238; 1995. Crowley, C.; Spencer, S. D.; Nishimura, M. C.; Chen, K. S.; PittsMeek, S.; Armanini, M. P.; Ling, G. H.; McMahon, S. B.; Shelton, D. L.; Levinson, A. D.; Phillips, H. S. Mice lacking nerve growth factor display perinatal loss of sensory and sympathetic neurons yet develop basal forebrain cholinergic neurons. Cell 76:1001–1011; 1994. Davies, A. M. The role of neurotrophins in the developing nervous system. J. Neurobiol. 25:1334 –1348; 1994. ElShamy, W. M.; Ernfors, P. A local action of neurotrophin-3 prevents the death of proliferating sensory neuron precursor cells. Neuron 16:963–972; 1996. Enokido, Y.; Wyatt, S.; Davies, A. M. Developmental changes in the response of trigeminal neurons to neurotrophins: Influence of birthdate and the ganglion environment. Development 126:4365– 4373; 1999. Ernfors, P.; Lee, K. F.; Jaenisch, R. Mice lacking brain-derived neurotrophic factor develop with sensory deficits. Nature 368:147–150; 1994. Ernfors, P.; Lee, K. F.; Kucera, J.; Jaenisch, R. Lack of neurotrophin-3 leads to deficiencies in the peripheral nervous system and loss of limb proprioceptive afferents. Cell 77:503–512; 1994. Farin˜ as, I.; Jones, K. R.; Backus, C.; Wang, X.-Y.; Reichardt, L. F. Severe sensory and sympathetic deficits in mice lacking neurotrophin-3. Nature 369:658 – 661; 1994. Farin˜ as, I.; Yoshida, C. K.; Backus, C.; Reichardt, L. F. Lack of neurotrophin-3 results in death of spinal sensory neurons and premature differentiation of their precursors. Neuron 17:1065–1078; 1996. Farin˜ as, I.; Wilkinson, G. A.; Backus, C.; Reichardt, L. F.; Patapoutian, A. Characterization of neurotrophin and trk receptor functions in developing sensory ganglia: Direct NT-3 activation of trkB neurons in vivo. Neuron 21:325–334; 1998. Farin˜ as, I. Neurotrophin actions during the development of the peripheral nervous system. Microsc. Res. Tech. 45:233–242; 1999. Frade, J. M.; Barde, Y. A. Nerve growth factor: Two receptors, multiple functions. Bioessays 2:137–145; 1998. Haddon, C.; Smithers, L.; Schneider-Maunoury, S.; Coche, T.; Henrique, D.; Lewis, J. Multiple delta genes and lateral inhibition in zebrafish primary neurogenesis. Development 125:359 –370; 1998. Henion, P. D.; Garner, A. S.; Large, T. H.; Weston, J. A. trkCmediated NT-3 signalling is required for the early development of a subpopulation of neurogenic neural crest cells. Dev. Biol. 172:602– 613; 1995. Huang, E. J.; Wilkinson, G. A.; Farinas, I.; Backus, C.; Zang, K.; Wong, S. L.; Reichardt, L. F. Expression of Trk receptors in the developing mouse trigeminal ganglion: In vivo evidence for NT-3 activation of TrkA and TrkB in addition to TrkC. Development 126: 2191–2203; 1999. Huang, E. J.; Reichardt, L. F. Neurotrophins: Roles in neuronal development and function. Annu. Rev. Neurosci. 24:677–736; 2001. Ichikawa, H.; Matsuo, S.; Silos-Santiago, I.; Sugimoto, T. Developmental dependency of Meissner corpuscles on trkB but not trkA or trkC. Neuroreport 11:259 –262; 2000. Jones, K. R.; Farin˜ as, I.; Backus, C.; Reichardt, L. F. Targeted disruption of the BDNF gene perturbs brain and sensory neuron development but not motor neuron development. Cell 76:989 –999; 1994. Kalcheim, C.; Carmeli, C.; Rosenthal, A. Neurotrophin 3 is a mitogen
815 for cultured neural crest cells. Proc. Natl. Acad. Sci. USA 89:1661– 1665; 1992. 25. Kaplan, D. R.; Miller, F. D. Neurotrophin signal transduction in the nervous system. Curr. Opin. Neurobiol. 3:381–391; 2000. 26. Karavanov, A.; Sainio, K.; Palgi, J.; Saarma, M.; Saxen, L.; Sariola, H. Neurotrophin 3 rescues neuronal precursors from apoptosis and promotes neuronal differentiation in the embryonic metanephric kidney. Proc. Natl. Acad. Sci. USA 92:11279 –11283; 1995. 27. Klein, R.; Martı´n-Zanca, D.; Barbacid, M.; Parada, L. F. Expression of the tyrosine kinase receptor gene trkB is confined to the murine embryonic and adult nervous system. Development 109:845– 850; 1990. 28. Klein, R.; Smeyne, R. J.; Wurst, W.; Long, L. K.; Auerbach, B. A.; Joyner, A. L.; Barbacid, M. Targeted disruption of the trkB neurotrophin receptor gene results in nervous system lesions and neonatal death. Cell 75:113–122; 1993. 29. Klein, R.; Silos-Santiago, I.; Smeyne, R. J.; Lira, S. A.; Bambrilla, R.; Bryant, S.; Zhang, L.; Snider, W. D.; Barbacid, M. Disruption of the neurotrophin-3 receptor gene trkC eliminates Ia muscle afferents and results in abnormal movements. Nature 368:249 –251; 1994. 30. Kucera, J.; Fan, G.; Jaenisch, R.; Linnarsson, S.; Ernfors, P. Dependence of developing group Ia afferents on neurotrophin-3. J. Comp. Neurol. 363:307–320; 1995. 31. Lawson, S.; Biscoe, T. J. Development of mouse dorsal root ganglia: An autoradiographic and quantitative study. J. Neurocytol. 8:265–274; 1979. 31a.Lefcort, F.; Clary, D. O.; Rusoff, A. C.; Reichardt, L. F. Inhibition of the NT-3 receptor TrkC, early in chick embryogenesis, results in severe reductions in multiple neuronal subpopulations in the dorsal root ganglia. J. Neurosci. 16:3704 –3713; 1996. 32. Levi-Montalcini, R. The nerve growth factor 35 years later. Science 237:1154 –1162; 1987. 33. Liebl, D. J.; Tessarollo, L.; Palko, M. E.; Parada, L. F. Absence of sensory neurons before target innervation in brain derived neurotrophic factor-, neurotrophin3-, and trkC-deficient embryonic mice. J. Neurosci. 17:9113–9121; 1997. 34. Liebl, D. J.; Klesse, L. J.; Tessarollo, L.; Wohlman, T.; Parada, L. F. Loss of brain-derived neurotrophic factor-dependent neural crest-derived sensory neurons in neurotrophin-4 mutant mice. Proc. Natl. Acad. Sci. USA 97:2297–2302; 2000. 35. Liu, X.; Ernfors, P.; Wu, H.; Jaenisch, R. Sensory but not motor neuron deficits in mice lacking NT4 and BDNF. Nature 375:238 –241; 1995. 36. Ma, Q.; Fode, C.; Guillemot, F.; Anderson, D. J. Neurogenin1 and neurogenin2 control two distinct waves of neurogenesis in developing dorsal root ganglia. Genes Dev. 13:1717–1728; 1999. 37. Martin, S. C.; Sandell, J. H.; Heinrich, G. Zebrafish trkC1 and trkC2 receptors define two different cell populations in the nervous system during the period of axonogenesis. Dev. Biol. 195:114 –130; 1998. 38. McMahon, S. B.; Armanini, M. P.; Ling, L. H.; Phillips, H. S. Expression and coexpression of trk receptors in subpopulations of adult primary sensory neurons projecting to identified peripheral targets. Neuron 12:1161–1171; 1994. 39. Memberg, S. P.; Hall, A. K. Proliferation, differentiation, and survival of rat sensory neuron precursors in vitro require specific trophic factors. Mol. Cell Neurosci. 6:323–335; 1995. 40. Minichiello, L.; Piehl, F.; Vazquez, E.; Schimmang, T.; Ho¨ kfelt, T.; Represa, J.; Klein, R. Differential effects of combined trk receptor mutations on dorsal root ganglion and inner ear sensory neurons. Development 121:4067– 4075; 1995. 41. Mu, X.; Silos-Santiago, I.; Carroll, S. L.; Snider, W. D. Neurotrophin receptor genes are expressed in distinct patterns in developing dorsal root ganglia. J. Neurosci. 13:4029 – 4041; 1993. 42. Nowakowski, R. S.; Lewin, S. B.; Miller, M. W. Bromodeoxyuridine immuno-histochemical determination of the lengths of the cell cycle and the synthetic phase for an anatomically defined population. J. Neurocytol. 12:311–318; 1989. 43. O’Conner, R.; Tessier-Lavigne, M. Identification of maxillary factor, a maxillary process-derived chemoattractant for developing trigeminal sensory axons. Neuron 24:165–178; 1999. 44. Oppenheim, R. W. Cell death during development of the nervous system. Annu. Rev. Neurosci. 14:453–501; 1991.
˜ AS ET AL. FARIN
816 45. Patapoutian, A.; Reichardt, L. F. Trk receptors: Mediators of neurotrophin action. Curr. Opin. Neurobiol. 11:272–280; 2001. 46. Pinco, O.; Carmeli, C.; Rosenthal, A.; Kalcheim, C. Neurotrophin-3 affects proliferation and differentiation of distinct neural crest cells and is present in the early neural tube of avian embryos. J. Neurobiol. 24:1626 –1641; 1993. 47. Reichardt, L. F.; Farin˜ as, I. Neurotrophic factors and their receptors: Roles in neuronal development and function. In: Cowan, M. W.; Jessell, T. M.; Zipurski, L., eds. Molecular approaches to neural development. New York: Oxford University Press; 1997:220 –263. 48. Schecterson, L. C.; Bothwell, M. Novel roles for neurotrophins are suggested by BDNF and NT-3 mRNA expression in developing neurons. Neuron 9:449 – 466; 1992. 49. Scott, S. A. Sensory neurons. diversity, development, and plasticity. New York: Oxford University Press; 1992. 50. Sieber-Blum, M. Role of the neurotrophic factors BDNF and NGF in the commitment of pluripotent neural crest cells. Neuron 6:949 –955; 1991. 51. Silos-Santiago, A.; Molliver, D. C.; Ozaki, S.; Smeyne, R. J.; Fagan, A. M.; Barbacid, M.; Snider, W. D. Non-trkA-expressing small DRG neurons are lost in trkA deficient mice. J. Neurosci. 15:5929 –5942; 1995. 52. Silos-Santiago, I.; Fagan, A. M.; Garber, M.; Fritzsch, B.; Barbacid, M. Severe sensory deficits but normal CNS development in newborn mice lacking trkB and trkC tyrosine protein kinase receptors. Eur. J. Neurosci. 9:2045–2056; 1997. 53. Smeyne, R. J.; Klein, R.; Schnapp, A.; Long, L. K.; Bryant, S.; Lewin, S. A.; Barbacid, M. Severe sensory and sympathetic neuropathies in mice carrying a disrupted Trk/NGF. Nature 368:246 –248; 1994. 54. Tessarollo, L.; Tsoulfas, P.; Martin-Zanca, D.; Gilbert, D. J.; Jenkins,
55.
56.
57.
58.
59.
60.
61.
N. A.; Copeland, N. G.; Parada, L. F. trkC, a receptor for neurotrophin-3, is widely expressed in the developing nervous system and in non-neuronal tissues. Development 118:463– 475; 1993. Tessarollo, L.; Vogel, K. S.; Palko, M. E.; Reid, S. W.; Parada, L. F. Targeted mutation in the neurotrophin-3 gene results in loss of muscle sensory neurons. Proc. Natl. Acad. Sci. USA 91:11844 –11848; 1994. Tessarollo, L.; Tsoulfas, P.; Donovan, M. J.; Palko, M. E.; Blair-Flynn, J.; Hempstead, B. L.; Parada, L. F. Targeted deletion of all isoforms of the trkC gene suggests the use of alternate receptors by its ligand neurotrophin-3 in neuronal development and implicates trkC in normal cardiogenesis. Proc. Natl. Acad. Sci. USA 94:14776 –14781; 1997. Tojo, H.; Kaisho, Y.; Nakata, M.; Matsuoka, K.; Kitagawa, M.; Abe, T.; Takami, K.; Yamamoto, M.; Shino, A.; Igarashi, K., et al. Targeted disruption of the neurotrophin-3 gene with lacZ induces loss of trkCpositive neurons in sensory ganglia but not in spinal cords. Brain Res. 669:163–175; 1995. Verdi, J. M.; Anderson, D. J. Neurotrophins regulate sequential changes in neurotrophin receptor expression by sympathetic neuroblasts. Neuron 13:1359 –1372; 1994. White, F. A.; Silos-Santiago, I.; Molliver, D. C.; Nishimura, M.; Phillips, H.; Barbacid, M.; Snider, W. D. Synchronous onset of NGF and trkA survival dependence in developing dorsal root ganglia. J. Neurosci. 16:4662– 4672; 1996. Wilkinson, G. A.; Farin˜ as, I.; Backus, C.; Yoshida, C. K.; Reichardt, L. F. Neurotrophin-3 is a survival factor in vivo for early mouse trigeminal neurons. J. Neurosci. 16:7661–7669; 1996. Yuan, J.; Yankner, B. A. Neuronal apoptosis sculpts the developing brain and has a potentially important role in neurodegenerative diseases. Nature 407:802– 809; 2000.