Brain spectrin isoforms during development of chick dorsal root ganglion cells in vivo and in vitro

Brain spectrin is one of the major cytoskeletal proteins associated with the plasma membrane. In many tissues this protein occurs in a variety of isoforms, for which at least three have been described in the brain: i) brain spectrin 240/235 is localized in neurons most prominently in axons and is present early during brain development. ii) Brain spectrin 240/235E is immunologicaly related to erythrocyte spectrin and restricted to somato-dendritic regions in neurons and to glia. It appears late in brain development. iii) A third form, brain spectrin 240/ 235A, is found exclusively in astrocytes. In this study we have investigated the appearance and distribution of brain spectrins 240/235 and 240/235E during embryonic chick dorsal root ganglia development in vivo and in vitro. This system provides a unique model due to the lack of dendrites on developing sensory neurons. Both isoforms first appeared at embryonic day 6. Brain spectrin 240/235 increased transiently around embryonic day 10 and 14, and was first expressed in ventrolateral neurons. It was localized abundantly in perikarya and their axons. This somato-axonal distribution pattern found in situ was also observed in vitro. In contrast, brain spectrin 240/235E only slightly increased between E6 and E15 and remained unchanged thereafter. It was localized mainly in small neurons of the mediodorsal area, where it was found as punctate staining in the cytoplasm, forming first a nuclear cap and in subsequent stages becoming distributed evenly throughout cytoplasm. This brain spectrin isoform was absent from axons, both in situ and in vitro. In conclusion, this study suggests i) that brain spectrin 240/235 may contribute towards the outgrowth, elongation and possibly maintenance of axonal processes, ii) that brain spcctrin 240/235E could be involved in the stablization of the cytoarchilecture of cell bodies in a sclected population of ganglion cells, and iii) that isoform expression of brain spectrin 240/235E in DRG cells may depend on environmental factors.

Inducing effect of skeletal muscle extracts on the appearance of calbindin-immunoreactive dorsal root ganglion cells in culture.

Calbindin D-28k is a calcium-binding protein which is not expressed by dorsal root ganglion cells cultured from 6-day-old (E6) chick embryos. When soluble muscle extracts from embryos at E11, E18 or chickens 2 weeks after hatching were added immediately after seeding, dorsal root ganglia cells grown at E6 displayed neuronal subpopulations expressing calbindin immunoreactivity with time; the effect of muscle extract on the percentage of calbindin-immunoreactive dorsal root ganglia cells followed a dose-response curve. When muscle extract was added to cultures after a 3 day delay, the percentage of calbindin-expressing neurons was unchanged. The effect produced by muscle extract and, to a lesser degree, skin extract on the appearance of calbindin-positive neurons was not reproduced by brain or liver extracts while all four exerted a trophic action on cultured neurons. Hence it is assumed that muscle extract contains a factor which produces an inductive effect on the initiation of calbindin-expression by uncommitted subpopulations of sensory neurons rather than a trophic influence on the selective survival of covertly committed neuronal subpopulations. The fact that muscle extract promoted calbindin expression by dorsal root ganglia cells in neuron-enriched as well as in mixed dorsal root ganglion cell cultures indicates that the factor would act directly on sensory neurons rather than indirectly through mediation of non-neuronal cells. Since the active muscular factor was non-dialysable, heat-inactivated, trypsin-sensitive and retained by molecular filters with a cut-off of 30 K, this factor is probably a protein.

Neuronal phenotypes in mouse dorsal root ganglion cell cultures: enrichment of substance P and calbindin D-28k expressing neurons in a defined medium.

The primary sensory neurons in mouse dorsal root ganglia consist of diversified subpopulations which express distinct phenotypic characteristics such as substance P or calbindin D-28k. To determine whether neuronal phenotypes are altered or not in in vitro cultures carried out in a defined synthetic medium, dissociated dorsal root ganglion cells from newborn mice were grown in the alpha-modified minimum essential medium either supplemented with 10% fetal calf serum or serum-free. About 80% of the neurons survived after 5 days of culture in both media, but only 35% or 65% were rescued after 12 days in serum-free or fetal calf serum supplemented medium, respectively. The neuronal subpopulations expressing substance P or calbindin D-28k displayed similar morphological properties in both media and a higher resistance to culture conditions than the whole neuronal cell population, especially in serum-free medium. It is therefore concluded that a defined synthetic medium offers reproducible conditions to culture dorsal root ganglion cells for at least 5 days, stimulates the expression of substance P and enriches preferentially neuronal phenotypes expressing substance P or calbindin D-28k, for a longer period of culture.

Triiodothyronine and nerve growth factor are required to induce cytoplasmic dynein expression in rat dorsal root ganglion cultures.

Beside the several growth factors which play a crucial role in the development and regeneration of the nervous system, thyroid hormones also contribute to the normal development of the central and peripheral nervous system. In our previous work, we demonstrated that triiodothyronine (T3) in physiological concentration enhances neurite outgrowth of primary sensory neurons in cultures. Neurite outgrowth requires microtubules and microtubule associated proteins (MAPs). Therefore the effects of exogenous T3 or/and nerve growth factors (NGF) were tested on the expression of cytoskeletal proteins in primary sensory neurons. Dorsal root ganglia (DRG) from 19 day old rat embryos were cultured under four conditions: (1) control cultures in which explants were grown in the absence of T3 and NGF, (2) cultures grown in the presence of NGF alone, (3) in the presence of T3 alone or (4) in the presence of NGF and T3 together. Analysis of proteins by SDS-polyacrylamide gel electrophoresis revealed the presence of several proteins in the molecular weight region around 240 kDa. NGF and T3 together induced the expression of one protein, in particular, with a molecular weight above 240 kDa, which was identified by an antibody against MAP1c, a protein also known as cytoplasmic dynein. The immunocytochemical detection confirmed that this protein was expressed only in DRG explants grown in the presence of NGF and T3 together. Neither control explants nor explants treated with either NGF or T3 alone expressed dynein. In conclusion, a combination of nerve growth factor and thyroid hormone is necessary to regulate the expression of cytoplasmic dynein, a protein that is involved in retrograde axonal transport.

Maintenance of neuronal expression of calbindin by a muscular extract in cultures of chick dorsal root ganglion cells.

Calbindin D-28K is a calcium-binding protein which is expressed by subpopulations of dorsal root ganglion cells cultured from 10-day-old (E10) chick embryos. After 7 or 10 days of culture, more than 20% of the ganglion cells are immunostained by an anticalbindin-antiserum; however, after 14 days of culture, the proportion drops to 10%. This fall can be prevented by addition of muscle extract to cultures at 10 days. Thus the transitory expression of calbindin-immunoreactivity by responsive sensory neurons would be not only induced but also maintained by a differentiation factor of muscular origin.

Expression of calbindin immunoreactivity by subpopulations of primary sensory neurons in chick embryo dorsal root ganglion cells grown in coculture or conditioned medium.

Primary sensory neurons which innervate neuromuscular spindles in the chicken are calbindin-immunoreactive. The influence exerted by developing skeletal muscle on the expression of calbindin immunoreactivity by subpopulations of dorsal root ganglion (DRG) cells in the chick embryo was tested in vitro in coculture with myoblasts, in conditioned medium (CM) prepared from myoblasts and in control cultures of DRG cells alone. Control cultures of DRG cells grown at the 6th embryonic day (E6) did not show any calbindin-immunostained ganglion cell. In coculture of myoblasts previously grown for 14 days, about 3% of calbindin-immunoreactive ganglion cells were detected while about 1% were observed in some cultures grown in CM. Fibroblasts from various sources were devoid of effect. Skin or kidney cells were more active than myoblasts to initiate calbindin expression by subpopulations of DRG cells in coculture or, to a lesser degree, in CM. The results suggest that cellular factors would rather induce calbindin expression in certain sensory neurons than ensure a selective neuronal survival.

Carbonic anhydrase activity in primary sensory neurons. II. Influence of environmental factors on the phenotypic expression of the enzyme in dissociated cultures of chicken dorsal root ganglion cells.

Neuronal subpopulations of dorsal root ganglion (DRG) cells in the chicken exhibit carbonic anhydrase (CA) activity. To determine whether CA activity is expressed by DRG cells maintained in in vitro cultures, dissociated DRG cells from 10-day-old chick embryos were cultured on a collagen substrate. The influence exerted by environmental factors on the enzyme expression was tested under various conditions of culture. Neuron-enriched cell cultures and mixed DRG-cell cultures (including numerous non-neuronal cells) were performed either in a defined medium or in a horse serum-supplemented medium. In all the tested conditions, subpopulations of cultured sensory neurons expressed CA activity in their cell bodies, while their neurites were rarely stained; in each case, the percentage of CA-positive neurons declined with the age of the cultures. The number and the persistence of neurons possessing CA activity as well as the intensity of the reaction were enhanced by addition of horse serum. In contrast, the expression of the neuronal CA activity was not affected by the presence of non-neuronal cells or by the rise of CO2 concentration. Thus, the appearance and disappearance of neuronal subpopulations expressing CA activity may be decisively influenced by factors contained in the horse serum. The loss of CA-positive neurons with time could result from a cell selection or from genetic repression. Analysis of the time curves does not support a preferential cell death of CA-positive neurons but suggests that the eventual conversion of CA-positive neurons into CA-negative neurons results from a loss of the enzyme activity. These results indicate that the phenotypic expression of cultured sensory neurons is dependent on defined environmental factors.

Influence of environmental factors on the expression of phenotypic characters by chicken dorsal root ganglion cells in culture.

Primary sensory neurons were grown under four conditions of culture. The influence of nonneuronal cells, horse serum or both was studied on the phenotypic expression of certain neuronal subpopulations. The number of neurons expressing acetylcholinesterase, alpha-bungarotoxin-binding sites or a high uptake capacity for glutamine was enhanced by nonneuronal cells. The horse serum increases the neuronal subpopulation exhibiting a carbonic anhydrase activity. Certain phenotypic changes fit conditions consistent with an epigenetic induction rather than a cell selection.

Gene expression mapping in neuropathic pain : molecular plasticity in nociceptors and superficial dorsal horn

RESUME : La douleur neuropathique est le résultat d'une lésion ou d'un dysfonctionnement du système nerveux. Les symptômes qui suivent la douleur neuropathique sont sévères et leur traitement inefficace. Une meilleure approche thérapeutique peut être proposée en se basant sur les mécanismes pathologiques de la douleur neuropathique. Lors d'une lésion périphérique une douleur neuropathique peut se développer et affecter le territoire des nerfs lésés mais aussi les territoires adjacents des nerfs non-lésés. Une hyperexcitabilité des neurones apparaît au niveau des ganglions spinaux (DRG) et de la corne dorsale (DH) de la moelle épinière. Le but de ce travail consiste à mettre en évidence les modifications moléculaires associées aux nocicepteurs lésés et non-lésés au niveau des DRG et des laminae I et II de la corne dorsale, là où l'information nociceptive est intégrée. Pour étudier les changements moléculaires liés à la douleur neuropathique nous utilisons le modèle animal d'épargne du nerf sural (spared nerve injury model, SNI) une semaine après la lésion. Pour la sélection du tissu d'intérêt nous avons employé la technique de la microdissection au laser, afin de sélectionner une sous-population spécifique de cellules (notamment les nocicepteurs lésés ou non-lésés) mais également de prélever le tissu correspondant dans les laminae superficielles. Ce travail est couplé à l'analyse à large spectre du transcriptome par puce ADN (microarray). Par ailleurs, nous avons étudié les courants électriques et les propriétés biophysiques des canaux sodiques (Na,,ls) dans les neurones lésés et non-lésés des DRG. Aussi bien dans le système nerveux périphérique, entre les neurones lésés et non-lésés, qu'au niveau central avec les aires recevant les projections des nocicepteurs lésés ou non-lésés, l'analyse du transcriptome montre des différences de profil d'expression. En effet, nous avons constaté des changements transcriptionnels importants dans les nocicepteurs lésés (1561 gènes, > 1.5x et pairwise comparaison > 77%) ainsi que dans les laminae correspondantes (618 gènes), alors que ces modifications transcriptionelles sont mineures au niveau des nocicepteurs non-lésés (60 gènes), mais important dans leurs laminae de projection (459 gènes). Au niveau des nocicepteurs, en utilisant la classification par groupes fonctionnels (Gene Ontology), nous avons observé que plusieurs processus biologiques sont modifiés. Ainsi des fonctions telles que la traduction des signaux cellulaires, l'organisation du cytosquelette ainsi que les mécanismes de réponse au stress sont affectés. Par contre dans les neurones non-lésés seuls les processus biologiques liés au métabolisme et au développement sont modifiés. Au niveau de la corne dorsale de la moelle, nous avons observé des modifications importantes des processus immuno-inflammatoires dans l'aire affectée par les nerfs lésés et des changements associés à l'organisation et la transmission synaptique au niveau de l'aire des nerfs non-lésés. L'analyse approfondie des canaux sodiques a démontré plusieurs changements d'expression, principalement dans les neurones lésés. Les analyses fonctionnelles n'indiquent aucune différence entre les densités de courant tétrodotoxine-sensible (TTX-S) dans les neurones lésés et non-lésés même si les niveaux d'expression des ARNm des sous-unités TTX-S sont modifiés dans les neurones lésés. L'inactivation basale dépendante du voltage des canaux tétrodotoxine-insensible (TTX-R) est déplacée vers des potentiels positifs dans les cellules lésées et non-lésées. En revanche la vitesse de récupération des courants TTX-S et TTX-R après inactivation est accélérée dans les neurones lésés. Ces changements pourraient être à l'origine de l'altération de l'activité électrique des neurones sensoriels dans le contexte des douleurs neuropathiques. En résumé, ces résultats suggèrent l'existence de mécanismes différenciés affectant les neurones lésés et les neurones adjacents non-lésés lors de la mise en place la douleur neuropathique. De plus, les changements centraux au niveau de la moelle épinière qui surviennent après lésion sont probablement intégrés différemment selon la perception de signaux des neurones périphériques lésés ou non-lésés. En conclusion, ces modulations complexes et distinctes sont probablement des acteurs essentiels impliqués dans la genèse et la persistance des douleurs neuropathiques. ABSTRACT : Neuropathic pain (NP) results from damage or dysfunction of the peripheral or central nervous system. Symptoms associated with NP are severe and difficult to treat. Targeting NP mechanisms and their translation into symptoms may offer a better therapeutic approach.Hyperexcitability of the peripheral and central nervous system occurs in the dorsal root ganglia (DRG) and the dorsal horn (DH) of the spinal cord. We aimed to identify transcriptional variations in injured and in adjacent non-injured nociceptors as well as in corresponding laminae I and II of DH receiving their inputs.We investigated changes one week after the injury induced by the spared nerve injury model of NP. We employed the laser capture microdissection (LCM) for the procurement of specific cell-types (enrichment in nociceptors of injured/non-injured neurons) and laminae in combination with transcriptional analysis by microarray. In addition, we studied functionál properties and currents of sodium channels (Nav1s) in injured and neighboring non-injured DRG neurons.Microarray analysis at the periphery between injured and non-injured DRG neurons and centrally between the area of central projections from injured and non-injured neurons show significant and differential expression patterns. We reported changes in injured nociceptors (1561 genes, > 1.5 fold, >77% pairwise comparison) and in corresponding DH laminae (618 genes), while less modifications occurred in non-injured nociceptors (60 genes) and in corresponding DH laminae (459 genes). At the periphery, we observed by Gene Ontology the involvement of multiple biological processes in injured neurons such as signal transduction, cytoskeleton organization or stress responses. On contrast, functional overrepresentations in non-injured neurons were noted only in metabolic or developmentally related mechanisms. At the level of superficial laminae of the dorsal horn, we reported changes of immune and inflammatory processes in injured-related DH and changes associated with synaptic organization and transmission in DH corresponding to non-injured neurons. Further transcriptional analysis of Nav1s indicated several changes in injured neurons. Functional analyses of Nav1s have established no difference in tetrodotoxin-sensitive (TTX-S) current densities in both injured and non-injured neurons, despite changes in TTX-S Nav1s subunit mRNA levels. The tetrodotoxin-resistant (TTX-R) voltage dependence of steady state inactivation was shifted to more positive potentials in both injured and non-injured neurons, and the rate of recovery from inactivation of TTX-S and TTX-R currents was accelerated in injured neurons. These changes may lead to alterations in neuronal electrogenesis. Taken together, these findings suggest different mechanisms occurring in the injured neurons and the adjacent non-injured ones. Moreover, central changes after injury are probably driven in a different manner if they receive inputs from injured or non-injured neurons. Together, these distinct and complex modulations may contribute to NP.

Differential proteoglycan expression in two spinal cord regions after dorsal root injury.

Dorsal root injury leads to reactive gliosis in the spinal cord dorsal root entry zone and dorsal column, two regions that undergo Wallerian degeneration, but have distinct growth-inhibitory properties. This disparity could in part be due to differences in the number of degenerating sensory fibers, differences in glial cell activation, and/or to differential expression of growth-inhibitory molecules such as chondroitin sulfate proteoglycans. Laser capture microdissection of these two spinal cord white matter regions, followed by quantitative analysis of mRNA expression by real-time PCR, revealed that glial marker transcripts were differentially expressed post-injury and that the chondroitin sulfate proteoglycans Brevican and Versican V1 and V2 were preferentially up-regulated in the dorsal root entry zone, but not the dorsal column. These results indicate that reactive gliosis differs between these two regions and that Brevican and Versican are potential key molecules participating in the highly inhibitory properties of the dorsal root entry zone.

Exploring Parallels Between Molecular Changes Induced in PNS by Aging and Demyelinating Neuropathies

The peripheral nervous system (PNS) is involved in many age-dependent neurological deficits, including numbness, pain, restless legs, trouble with walking and balance that are commonly found in the elderly. These symptoms generally result from demyelination and/or loss of axonal integrity. However, the precise identity of age-regulated molecular changes in either neuronal or glial compartments of the nerve is unclear. Interestingly, these deficiencies are also present in inherited neuropathies, where the expressivity of the rapid and early onset phenotypes is undeniably more severe than in normal aging. Nevertheless, especially the molecular changes underlying loss of axonal integrity in neuropathy condition are also poorly understood. To unravel molecular mechanisms affected by PNS aging, we used wildtype mice at 17 time-points from day of birth until senescence (28 months-old). For the neuropathy study, we focused on 56 day-old Schwann cell-specific neuropathy-inducing mutants, MPZCre/1/ LpinfE2-3/fE2-3 and MPZCre/1/ScapfE1/fE1 mice, that have, at this age, already developed neuropathic symptoms. Transcriptomes of dissected Schwann cell-containing endoneurium or sensory neuron-containing dorsal root ganglia have been analyzed throughout time or genotypes, using Illumina Bead Chips. Following data validation, we identified groups of differentially expressed genes in the development, aging and in the neuropathic mutants, in both glial and neuronal compartments. We detected substantial differences in the dynamics of changes in gene expression during development and aging between these two compartments. Furthermore, considering the above-mentioned phenotypic similarities, we integrated aging and mutant data. Interestingly, we observed that there are some parallels at the molecular level between processes involved in aging, which leads to less severe and more progressive PNS alterations, and in the rapid onset peripheral neuropathies. Apart from helping the understanding of molecular alterations underlying age-related PNS phenotypes, this data should also contribute to the identification of pathways that could be used as targets for therapeutical approaches to prevent complications associated with both aging and inherited forms of neuropathies.

Recombinant adeno-associated virus serotype 6 (rAAV2/6)-mediated gene transfer to nociceptive neurons through different routes of delivery.

BACKGROUND: Gene transfer to nociceptive neurons of the dorsal root ganglia (DRG) is a promising approach to dissect mechanisms of pain in rodents and is a potential therapeutic strategy for the treatment of persistent pain disorders such as neuropathic pain. A number of studies have demonstrated transduction of DRG neurons using herpes simplex virus, adenovirus and more recently, adeno-associated virus (AAV). Recombinant AAV are currently the gene transfer vehicles of choice for the nervous system and have several advantages over other vectors, including stable and safe gene expression. We have explored the capacity of recombinant AAV serotype 6 (rAAV2/6) to deliver genes to DRG neurons and characterized the transduction of nociceptors through five different routes of administration in mice. RESULTS: Direct injection of rAAV2/6 expressing green fluorescent protein (eGFP) into the sciatic nerve resulted in transduction of up to 30% eGFP-positive cells of L4 DRG neurons in a dose dependent manner. More than 90% of transduced cells were small and medium sized neurons (< 700 microm 2), predominantly colocalized with markers of nociceptive neurons, and had eGFP-positive central terminal fibers in the superficial lamina of the spinal cord dorsal horn. The efficiency and profile of transduction was independent of mouse genetic background. Intrathecal administration of rAAV2/6 gave the highest level of transduction (approximately 60%) and had a similar size profile and colocalization with nociceptive neurons. Intrathecal administration also transduced DRG neurons at cervical and thoracic levels and resulted in comparable levels of transduction in a mouse model for neuropathic pain. Subcutaneous and intramuscular delivery resulted in low levels of transduction in the L4 DRG. Likewise, delivery via tail vein injection resulted in relatively few eGFP-positive cells within the DRG, however, this transduction was observed at all vertebral levels and corresponded to large non-nociceptive cell types. CONCLUSION: We have found that rAAV2/6 is an efficient vector to deliver transgenes to nociceptive neurons in mice. Furthermore, the characterization of the transduction profile may facilitate gene transfer studies to dissect mechanisms behind neuropathic pain.

Thyroid hormone enhances transected axonal regeneration and muscle reinnervation following rat sciatic nerve injury.

Improvement of nerve regeneration and functional recovery following nerve injury is a challenging problem in clinical research. We have already shown that following rat sciatic nerve transection, the local administration of triiodothyronine (T3) significantly increased the number and the myelination of regenerated axons. Functional recovery is a sum of the number of regenerated axons and reinnervation of denervated peripheral targets. In the present study, we investigated whether the increased number of regenerated axons by T3-treatment is linked to improved reinnervation of hind limb muscles. After transection of rat sciatic nerves, silicone or biodegradable nerve guides were implanted and filled with either T3 or phosphate buffer solution (PBS). Neuromuscular junctions (NMJs) were analyzed on gastrocnemius and plantar muscle sections stained with rhodamine alpha-bungarotoxin and neurofilament antibody. Four weeks after surgery, most end-plates (EPs) of operated limbs were still denervated and no effect of T3 on muscle reinnervation was detected at this stage of nerve repair. In contrast, after 14 weeks of nerve regeneration, T3 clearly enhanced the reinnervation of gastrocnemius and plantar EPs, demonstrated by significantly higher recovery of size and shape complexity of reinnervated EPs and also by increased acetylcholine receptor (AChRs) density on post synaptic membranes compared to PBS-treated EPs. The stimulating effect of T3 on EP reinnervation is confirmed by a higher index of compound muscle action potentials recorded in gastrocnemius muscles. In conclusion, our results provide for the first time strong evidence that T3 enhances the restoration of NMJ structure and improves synaptic transmission.

The expression of nuclear 3,5,3' triiodothyronine receptors is induced in Schwann cells by nerve transection.

The effects of thyroid hormones on the nervous system are mediated by the presence of nuclear T3 receptors (NT3R). In this study, the expression of NT3R was investigated in spinal cord, dorsal root ganglia (DRG), or sciatic nerve of adult rats after immunostaining with a 2B3-NT3R monoclonal antibody which recognizes both alpha and beta types of NT3R. The specificity of this monoclonal antibody was confirmed by Western blots. The 2B3-NT3R monoclonal antibody recognized one band corresponding to a molecular weight of 57 kDa in extract of spinal cord or DRG. No staining was observed on immunoblot of intact sciatic nerve. In the spinal cord, the nuclei of the neurons and glial cells including both astrocytes and oligodendrocytes exhibited 2B3-NT3R immunoreactivity. While all the nuclei of the DRG sensory neurons expressed the NT3R, all the nuclei of the satellite and Schwann cells were devoid of any immunoreaction. In the sciatic nerve, the nuclei of the Schwann cells also lacked 2B3-NT3R-immunoreactivity. After sciatic nerve transection in vivo, Schwann cell nuclei, which never expressed NT3R in intact nerves of adult rats, displayed a clear 2B3-NT3R immunoreaction in proximal and distal stumps adjacent to the section. Double immunostaining with antibodies raised to 3-sulfogalactosylceramide or S100 confirmed that most of the NT3R containing nuclei belong to Schwann cells. In dissociated cell cultures grown in vitro from sciatic nerves, Schwann cells exhibited 2B3-NT3R immunoreactivity. These data suggest that the inhibition of NT3R expression in Schwann cells ensheathing axons in intact nerve is reversed when the axons are degenerating or lacking.(ABSTRACT TRUNCATED AT 250 WORDS)

Triiodothyronine exerts a trophic action on rat sensory neuron survival and neurite outgrowth through different pathways.

Apart from several growth factors which play a crucial role in the survival and development of the central and peripheral nervous systems, thyroid hormones can affect different processes involved in the differentiation and maturation of neurons. The present study was initiated to determine whether triiodothyronine (T3) affects the survival and neurite outgrowth of primary sensory neurons in vitro. Dorsal root ganglia (DRG) from 19-day-old embryos or newborn rats were plated in explant or dissociated cell cultures. The effect of T3 on neuron survival was tested, either in mixed DRG cell cultures, where neurons grow with non-neuronal cells, or in neuron-enriched cultures where non-neuronal cells were eliminated at the outset. T3, in physiological concentrations, promoted the growth of neurons in mixed DRG cell cultures as well as in neuron-enriched cultures without added nerve growth factor (NGF). Since neuron survival in neuron-enriched cultures cannot be promoted by endogenous neurotrophic factors synthesized by non-neuronal cells, the increased number of surviving neurons was due to a direct trophic action of T3. Another trophic effect was revealed in this study: T3 sustained the neurite outgrowth of sensory neurons in DRG explants. The stimulatory effect of T3 on nerve fibre outgrowth was considerably reduced when non-neuronal cell proliferation was inhibited by the antimitotic agent cytosine arabinoside, and was completely suppressed when the great majority of non-neuronal cells were eliminated in neuron-enriched cultures. These results indicate that the stimulatory effect of T3 on neurite outgrowth is mediated through non-neuronal cells. It is conceivable that T3 up-regulates Schwann cell expression of a neurotrophic factor, which in turn stimulates axon growth of sensory neurons. Together, these results demonstrate that T3 promotes both survival and neurite outgrowth of primary sensory neurons in DRG cell cultures. The trophic actions of T3 on neuron survival and neurite outgrowth operate under two different pathways.