Abstract
Neurons develop two types of processes, axons and dendrites, whose growth must be independently controlled. Recent research has identified the small guanosine triphosphatase Rit as a differential regulator of neurite growth. Activation of Rit enhances axonal growth, whereas inhibition of Rit promotes dendritic growth. These results imply that the reciprocal regulation of a single molecule in the same cell can achieve simultaneous regulation of axonal and dendritic growth.
Neurons extend two types of neurites—axons and dendrites—whose distinctive properties underlie the vectorial flow of information between nerve cells (1). Early in development, cultured hippocampal neurons (a model for analyzing the development of neuronal polarity) establish several short, apparently identical neurites. Then an abrupt change occurs: Growth is largely confined to a single process, which becomes the axon (2). This symmetry-breaking event represents the first step in the establishment of neuronal polarity. Over the past few years, a number of studies have begun to elucidate the signaling events that underlie axonal specification. One key pathway involves phosphoinositide 3-kinase, whose actions are mediated by a scaffolding complex consisting of mPar3, mPar6, and atypical protein kinase C (3, 4). Calcium- and calmodulin-dependent protein kinases, mitogen-activated protein kinase (MAPK), glycogen synthase kinase 3β (GSK3β), and c-Jun N-terminal kinase (JNK) also are involved (5–8). A role for small guanosine triphosphatases (GTPases) has also emerged (9), particularly members of the Ras (Ras and Rap1) and the Rho (Rac1 and cdc42) superfamilies (10, 11). Sometime after the axon is specified, dendrites begin to elongate. Little is known about the changes in signaling that accompany this second phase of polarity development, when axons and dendrites grow simultaneously.
Lein and colleagues (12) now identify the small GTPase Rit (Ras-like protein in all tissues) as another important player in the signaling cascades that regulate the development of neuronal polarity. They demonstrate that Rit is a yin-yang regulator of axonal and dendritic growth, enhancing axonal and inhibiting dendritic growth. Of particular interest, their results shed new light on the changes in signaling that occur during the period of combined axonal and dendritic growth.
Rit is a member of the newest subfamily of Ras-related proteins, which also includes Rin (Ras-like protein in neurons) and Drosophila Ric (Ras-related protein that interacts with calmodulin) (13, 14). Both Rit and Rin show roughly 50% amino acid sequence identity with Ras but lack the C-terminal sequences required for prenylation, a posttranslational modification that is generally required for plasma membrane localization and function of Ras family proteins (14). Although both Rit and Rin are expressed in mammalian neurons, only Rit is expressed early in development. Rit, but not Rin, induces neurite formation in rat pheochromocytoma 6 cells (PC6, a PC12 subline) (15, 16). Like other small GTPases, Rit is activated by guanine nucleotide exchange of GDP for GTP and is inactivated by hydrolysis of GTP to GDP (14). These actions are likely governed by specific guanine nucleotide exchange factors (GEFs) and GTPase activating proteins (GAPs), respectively, although the particular GEFs and GAPs that regulate Rit have not been identified.
Lein et al. (12) use two culture models to elucidate the role of Rit in polarity development. Hippocampal neurons establish axons and dendrites through a cell-autonomous program of development (2). In contrast, sympathetic neurons cultured in the presence of nerve growth factor (NGF), which is required for cell survival, form axons but do not develop dendrites (17). Bone morphogenetic protein 7 (BMP7), a member of the transforming growth factor–β (TGFβ) superfamily of growth factors, induces dendrite extension in sympathetic neurons in culture (17). BMP7 also enhances, but is not required for, dendritic growth in hippocampal neurons (18).
Rit function in cells can be investigated by overriding endogenous, regulated Rit activity through overexpression of mutant Rit proteins locked in either a constitutively active, GTP-bound form (caRit) or a dominant-negative, GDP-bound form (dnRit). Lein et al. (12) found that expression of caRit in young hippocampal neurons more than doubled axonal length; similarly, total axonal length of sympathetic neurons expressing caRit was several times greater than that in control neurons, in part because more neurons extended multiple axons. The same manipulation (globally increasing Rit activity) had the opposite effect on dendritic growth. Both hippocampal and sympathetic neurons extended fewer and shorter dendrites when transfected with caRit. Because activation of Rit stimulates axonal growth but stunts dendritic growth, decreasing Rit activity globally by overexpression of dnRit might be expected to increase dendritic growth but inhibit axonal growth. Indeed, axonal outgrowth was inhibited in hippocampal neurons expressing dnRit, whereas dendrites grew longer than in control neurons. Likewise, dnRit reduced the average axon length in sympathetic neurons to less than half of normal. Dendritic growth induced by BMP treatment of sympathetic neurons was enhanced in cells expressing dnRIT; this effect was most evident at suboptimal BMP7 concentrations. Taken together, these data fit a model in which Rit activity promotes axonal growth but inhibits the growth of dendrites in the same neuron.
In both hippocampal and sympathetic neurons, the promotion of axonal growth and inhibition of dendritic growth induced by caRIT were blocked by MEK (mitogen-activated or extracellular signal–regulated protein kinase kinase) inhibitors, establishing a role for extracellular signal–regulated kinases (ERKs) in mediating these actions of Rit. MEK inhibition had no noticeable effect on axonal or dendritic growth in control hippocampal or sympathetic neurons. This may indicate that the increased ectopic Rit activation achieved by expressing caRit may activate ERK pathways not normally required for regulating neurite growth. Alternatively, localized Rit-ERK signals may escape full inhibition by the pharmacological agent. To better elucidate the role of the ERK pathway in Rit function, more specific approaches to inhibiting this pathway will have to be employed.
The enhancement of dendritic growth produced by inhibiting Rit parallels the effects of treating neurons with BMP7. To determine whether BMP7 directly inhibits Rit activity, the authors used biochemical methods to assay GTP loading of Rit in PC6 cells (cultures of sympathetic neurons do not provide enough material for such assays). Previous studies had shown that NGF increases the fraction of active, GTP-bound Rit in these cells (16). In the present study, BMP7 also increased GTP-bound Rit, but when cells were treated with both BMP7 and NGF, the fraction of GTP-bound Rit declined. The biochemical implications of these findings are provocative, suggesting that crosstalk between BMP7- and NGF-mediated signaling pathways may rapidly increase the activity of a Rit-GAP. This crosstalk could also occur through decreased activity of a Rit-GEF, although, given the rapid inactivation of Rit by BMP7, this scenario would also require concomitant rapid inactivation of Rit-GTP by Rit-GAPs. Treating hippocampal neurons with BMP, which enhances dendritic growth (18), also decreased GTP loading of Rit (12).
For axons and dendrites to grow simultaneously, Rit must be differentially regulated in the two compartments, activated in the axon and inhibited in the dendrites (Fig. 1). The effects of NGF and BMP7 on Rit activation in PC6 cells suggest a simple model through which this might occur. In sympathetic neurons, NGF is likely involved in axonal signaling, given the well-documented effects of NGF on the growth and maintenance of sympathetic axons (19). Both NGF- and BMP7-mediated pathways may be active in dendrites, thereby suppressing Rit activation, whereas only the NGF pathway may be active within axons, leading to Rit activation. In this model, how are BMP signals excluded from the axon? One simple explanation could be that BMP receptors are selectively expressed on dendrites, whereas TrkA receptors are expressed on both axons and dendrites (Fig. 1). Spatial regulation of Rit could also be achieved by other mechanisms (20). The signals that might play a role analogous to that of NGF to modulate BMP-mediated Rit activation during dendritic development in hippocampal neurons are less clear, although brain-derived neurotrophic factor might be a candidate (21).
Activation of Rit enhances axonal growth and inhibits dendritic growth. At stages of development during which both types of neurites elongate simultaneously, Rit must be activated in axons (Rit-GTP, orange) and inactivated in dendrites (Rit-GDP, yellow). Individually, BMP7 and NGF activate Rit, but together they reduce Rit activation, suggesting a model in which neurotrophins activate Rit in axons, whereas the combined effects of neurotrophins and BMPs inhibit Rit in dendrites. BMP-R, BMP receptor; Trk, neurotrophin receptor.
A number of questions remain. How activation of Rit promotes the growth of one type of neurite but inhibits that of another is not understood. Nor is it known how Rit fits with existing models of axonal and dendritic specification. Is Rit activation coupled to other small GTPase–dependent signaling pathways such as Rap1, cdc42, and Rac1? Is it linked to the Par complex, the GSK3β and collapsin response mediator protein (CRMP-2) pathway, or other as-yet-undefined pathways involved in establishing neuronal polarity? In fact, a physical link between Rit and members of the Par complex has already been mapped (22, 23). Does axonal enrichment of the Par complex contribute to the activation of Rit in axons? The role of ERKs in endogenous Rit signaling also remains unresolved. Indeed, some actions of Rit on neuronal growth are ERK independent (24). Are other downstream targets of Rit, such as p38 MAPK (25), involved in axonal specification? The identification of Rit as a regulator of axonal and dendritic growth thus provides neuroscientists a number of exciting new avenues and thoroughfares to pursue.
