Abstract
The identity of specific cell types in the nervous system is defined by the expression of neuron type–specific gene batteries. How the expression of such batteries is initiated during nervous system development has been under intensive study over the past few decades. However, comparatively little is known about how gene batteries that define the terminally differentiated state of a neuron type are maintained throughout the life of an animal. We provide here an overview of studies in invertebrate and vertebrate model systems that have carved out the general and not commonly appreciated principle that neuronal identity is maintained in postmitotic neurons by the sustained, and often autoregulated expression of the same transcription factors that have initiated terminal differentiation in a developing organism. Disruption of postmitotic maintenance mechanisms may result in neuropsychiatric and neurodegenerative conditions.
The classic definition of the identity of specific neuron types in the nervous system relies on anatomical and electrophysiological features. These features have provided a rich description of the enormous diversity of cell types in animal nervous systems1. With the advent of molecular biology, the definition of cell type identity has started to include molecular features of individual neurons2. While such molecular markers have remained relatively sparse for a great number of neuron types in vertebrate nervous systems, molecular profiling studies over the past decade have nevertheless significantly broadened the spectrum of genes known to be coexpressed in specific neuron types, thereby defining neuron–type specific gene batteries3,4.
While specific features of individual neurons are remarkably plastic5, many neuron identity features remain stable throughout the life of a terminally differentiated, postmitotic neuron. For example, with some notable exceptions to be discussed below6, neurotransmitter identities, as assessed by expression of enzymes and transporters that synthesize, synaptically package and reuptake a specific neurotransmitter, remain stable in many adult neuron types. One question that immediately arises from such enduring stability is how the tremendous diversity of neuron–types and their elaborate connectivity patterns are maintained for up to many decades of life in mitotically quiescent cells. A number of studies over the past few years have shown that there is a critical need for active maintenance of postmitotic neurons. What is the nature of the gene regulatory programs that are required across the lifespan of postmitotic neurons to preserve their subtype identity, morphology, and connectivity and thereby sustain normal neural circuit function? In contrast to the intense focus over the past two decades on understanding the gene regulatory programs that specify different neuron types7 the regulatory programs that function later in life to maintain postmitotic neuron–type identities has received far less attention. The recent application of these tools has revealed that at least one answer to how maintenance of neuronal identity is achieved lies in considering how terminally differentiated features of a postmitotic neuron are induced in the context of development8. Indeed, a common lesson to emerge is that gene regulatory programs launched early in fetal life to specify neuronal–types identities continue to function later in life to maintain postmitotic identities. Here we review key findings from these studies that illustrate the impact of disrupted maintenance mechanisms on neuronal identity, function and circuitry with potential relevance to neurodegenerative and neuropsychiatric disorders.
Maintenance of neuronal identity in invertebrates
The identity of neurons in the nervous system of the nematode C.elegans, composed of 118 anatomically distinct neuron types, have been well–defined not only on the level of anatomy but also lineage and function9. A key advantage of studying the neuronal identity problem in C. elegans is the availability of many hundreds of reporter genes that monitor the expression of terminal identity markers of individual neuron types, such as neurotransmitter receptors, neurotransmitter synthesizing enzymes and transporters, sensory receptors, ion channels and neuronal signaling factors. Classic screens for mutants that affect specific neuronal lineages identified transcription factors (TFs) such as the Atonal homolog lin–32 or the POU homeobox gene unc–86 (which gave the POU domain its “U”) which control neuroblast identity and fate9. Genetic screens for behavioral mutants (e.g. Unc – uncoordinated, Ttx – thermotaxis–defective, Odr – odortaxis defective, Che – chemotaxis defective), as well as screens for mutants in which the expression of gfp–tagged terminal identity features of specific neuron types is affected, have identified a large number of TFs that act in postmitotic neurons, either alone or in specific TF combinations, to initiate neuron–type specific differentiation programs9. The deep knowledge of functional, morphological and molecular features of individual worm neuron types has enabled a fine–grained probing of these TF mutant phenotypes. A shared theme of many of these TFs, dubbed “terminal selectors”10, is that their genetic removal results in the loss of many functionally unrelated terminal identity features, such as for example, neurotransmitter identity, expression of neurotransmitter receptors, ion channels etc.; however, generic features shared by all neurons, e.g. the expression of synaptic vesicle proteins, are generally unaffected. Currently, distinct TFs or TF combinations are known to initiate the differentiation program of almost one half of all the 118 neuron classes of the worm9. In many of these cases, it was demonstrated that terminal selectors directly act on cis–regulatory elements in proximity of terminal identity genes to initiate their expression11–16.
The maintained expression of C. elegans terminal selector transcription factors throughout the postmitotic life of the respective neurons suggests that these factors not only initiate but also actively maintain the expression of terminal identity features of individual neuron types. To address this possibility directly, a number of studies have tested whether postembryonic and hence postdevelopmental removal of a number of terminal selectors from fully differentiated neurons affects the maintenance of the differentiated state of specific neuron types, as assessed by examining the expression of continuously expressed terminal identity markers. Temporally controlled removal of terminal selectors was achieved in a number of different manners and in all cases examined, it was shown that the factors that initiate terminal differentiation programs are also continuously required to maintain the differentiated state (Table 1). For example, the Zn finger TF che–1 is continuously required to maintain the differentiated state of the glutamatergic ASE sensory neurons17, the Prd–type homeodomain TF unc–42 is continuously required to maintain ASH sensory neuron identity18, the EBF–type TF unc–3 is continuously required in cholinergic motor neurons12, and the combinatorial TF signature composed of the ETS–domain TF ast–1 and Dlx–type homeodomain TF ceh–43 is continuously required to maintain the differentiated state of all dopaminergic neurons in C. elegans (Table 1)16, 19.
Table 1.
Transcription factors known to maintain neuronal identity
| Species | Transcription factor name (type) | Neuron type that requires sustained TF expression to maintain its identity | Approach | Reference |
|---|---|---|---|---|
| C.elegans | che–1 (Zn finger) | ASEL/R gustatory neuron class | postdevelopmental RNAi | 17 |
| die–1 (Zn finger) | ASEL subtype identity | postdevelopmental RNAi | 21 | |
| MYST histone acetyltransferase | ASEL subtype identity | temperature–sensitive allele | 21 | |
| unc–3 (EBF) | Ventral nerve cord motor neurons | pulsing gene activity in a mutant background 1 | 12 | |
| ast–1 (ETS) | Dopaminergic neurons | pulsing gene activity in a mutant background 1 | 19 | |
| ceh–43 (Dlx–type homeobox) | Dopaminergic neurons | hypomorphic allele leads to initial generation, but then loss of identity | 16 | |
| mec–3 (LIM homeobox) | Touch sensory neurons | pulsing gene activity in a mutant background | 18 | |
| unc–42 (Prd–type homeobox) | ASH nociceptive neuron | postdevelopmental RNAi | 18 | |
| D. melanogaster | Eya, Ap, Dimm, Dac | Tv1/Tv4 neurons | Temporally controlled–RNAi 2 | 24 |
| Acj6, E93, Fer1, Onecut, Aim, Xbp1, Zf30c | Olfactory neurons | Temporally controlled–RNAi 2 | 22 | |
| ewg | Photoreceptors | Adult ewg rescue | 23 | |
| S. mediterranea | Pitx (paired–type homeobox) | 5HT neurons | RNAi | 45 |
| M. musculus | Hand2 (bHLH) | Sympathetic neurons | DBH–Cre | 28 |
| Gata–3 (Zn–GATA) | Sympathetic neurons 5HT neurons |
DBH–CreER ePet–Cre |
56 40 |
|
| Pet–1/FEV (ETS) | 5HT neurons |
ePet–Cre ePet–CreER |
61 | |
| Lmx1b (LIM homeobox) | 5HT neurons |
ePet–Cre Pet1–CreER |
29, 30 | |
| Gata–3 (Zn–GATA) | 5HT neurons | ePet–Cre | 40 | |
| Engrailed 1/2 | 5HT neurons Dopaminergic neurons |
ePet–Cre null alleles |
70 66, 67 |
|
| Nurr1 (NHR) | Dopaminergic neurons |
DAT–Cre AAV–Cre DAT–CreER |
26, 27 | |
| Foxa1/foxa2 (forkhead/winged helix | Dopaminergic neurons | DAT-Cre | 41 | |
| Pax–6 (paired–type homeobox) | Dopaminergic neurons (Olfactory bulb) | DAT–Cre | 55 | |
| COUP–TF1 (NHR) | Olfactory bulb DA neurons | Lenti–Cre | 72 | |
| Islet1 (LIM homeobox) | Dorsal root ganglia sensory neurons | Isl1–CreER | 43 | |
| Nkx2.1 (homeobox) | Forebrain cholinergic neurons | Chat–Cre | 44 | |
| Lhx7 (LIM homeobox) | Striatal interneurons | Chat–Cre | 42 | |
| Phox2 (homeobox) | MNs. Sympathetic and Locus coeruleus Norepin. neurons |
Isl1–Cre βactinCreER |
25 | |
| RORα (NHR) | Purkinje cells |
Pcp2–Cre Lenti–Cre |
58 | |
| Otx2 (homeobox) | Photoreceptors |
CreER Lenti–Cre |
61, 62 | |
| Nrl (bZIP) | Rod photoreceptors | CAG–CreER | 64 | |
| Lhx1 (LIM homeobox) | Inferior olivary neuron | Rosa25–CreER | 18 |
initial pulse to successfully rescue mutant phenotype in early stage, then let gene activity peter out, then scored for loss of identity.
TARGET system.
NHR, nuclear hormone receptor
A continuous requirement has not only been observed for C. elegans terminal selector–type TFs that very broadly define the identity of a neuron type, but also for TFs that only affect select subsets of identity features. Specifically, the die–1 Zn finger TF initiates the expression of a gene expression program that distinguishes the left and right ASE sensory neuron20. These two neurons are very similar to one another, but differ in the complement of chemoreceptors that they express. die–1 is expressed in the left neuron to initiate left–specific chemoreceptor and repress right–specific chemoreceptors. Postembryonic removal of die–1 demonstrated that die–1 is continuously required to define chemoreceptor gene expression21.
The die–1 example also provides an entry point to tie the role of chromatin modifications into maintenance of gene expression. Genetic analysis suggests that die–1 operates with the MYST–type histone acetyltransferase complex to initiate and maintain left vs. right ASE subtype features21. The use of a temperature–sensitive allele of a MYST complex component has underscored the continuous need of MYST in maintaining the identity of the left ASE neuron21. This specific case demonstrates chromatin modification states are not self–sufficient in maintaining a specific gene expression state, but also depend on the continuous presence of a transcription factor, die–1, that likely recruits histone modifying complexes.
Studies in Drosophila have provided additional support for the notion that TFs that initiate differentiation programs also continuously support the differentiated state (Table 1). A combination of seven transcription factors is required to initiate and maintain olfactory receptor expression patterns in fly olfactory neurons22. In photoreceptor neurons, active maintenance mechanisms ensure that the expression of distinct types of rhodopsin (rh) genes remains restricted to specific photoreceptor neurons. For example, the Erect wing (Ewg) TF is continuously required to prevent expression of Rh6 in R8–type photoreceptors23. Moreover, the peptidergic Tv1 and Tv4 neurons require a combination of TFs (including the bHLH TF Dimm and the LIM homeodomain TF Apterous) to acquire a specific identity and, as revealed through the use inducible dsRNA transgenes, also require these TFs for maintenance of their differentiated state24.
Maintenance of neuronal identity in vertebrates
In contrast to more short–lived invertebrate neurons, vertebrate neurons can be preserved through many decades of life. Like in invertebrates, many vertebrate transcription factors involved in the coordinate induction of batteries of genes encoding key features of neuron–type identity, such as the neurotransmitter phenotype, continue to be expressed across the life span in the mature neurons they initially helped to generate. Over the past several years, conditional Cre/Lox gene–targeting approaches have been implemented to investigate these terminal–selector–like TFs in the stage–specific maintenance of neuronal identity, morphology, and connectivity (Table 1).
Our current understanding of vertebrate neuronal maintenance mechanisms has been gained, in part, through studies of mouse serotonergic, dopaminergic, and noradrenergic neurotransmitter identity25–30. The majority of the TFs examined were previously shown to perform crucial functions in the acquisition of specific monoaminergic features through activation of neurotransmitter–type gene batteries that encode the capacity for synthesis, reuptake, and plasma or vesicular membrane transport of a particular monoamine31–39. The expression of all of these factors extends into adulthood and therefore conditional targeting strategies were used to determine their ongoing functions in the maintenance of identity features.
A consistent finding from the various studies of monoaminergic identity is that genetic removal of individual TFs within a few days after the acquisition of transmitter identity causes a profound loss of expression of various markers of transmitter phenotype within a particular postmitotic neuron type25–28, 30, 40,41. In fetal postmitotic 5–HT neurons, for example, expression of all markers of the 5–HT neuron–type gene battery and 5–HT it self was severely reduced within a day of targeting either Pet–1 or Lmx1b30, 40. Similarly, conditional targeting of Gata–3 in fetal 5–HT neurons resulted in early, albeit less severe, maintenance defects in Tph2, AADC, Sert, and VMAT2 expression40.
Severe losses of noradrenergic identity followed targeting of Hand2 and Phox2 in differentiated fetal sympathetic neurons25, 28. Targeting of the nuclear hormone receptor gene, Nurr1, in fetal or neonatal mesencephalic dopamine (DA) neurons led to a severe and progressive loss of DA identity and tissue DA transmitter levels particularly in the SN (Substantia nigra) but also in the VTA26, 27. SN DA neuron cell bodies disappeared after early fetal targeting. Likewise, ongoing expression of the LIM–homeodomain factors, Lhx7 and Islet1, are required at fetal stages to maintain cholinergic identity of postmitotic striatal interneurons and expression of several sensory neuron–type identity features in maturing dorsal root ganglia neurons, respectively42, 43.
Importantly, transcriptional maintenance programs continue to be crucial during early postnatal and even adult stages for preservation of monoaminergic–type and cholinergic–type neuron identity and transmitter levels44,26, 29, 40. For example, tamoxifen–inducible dopaminergic targeting of Nurr1 in early adulthood led to progressive and severe decreases in the expression of the DA neuron gene battery encoding tyrosine hydroxylase, dopamine transporter and vesicular monoamine transporter 2 (VMAT2) and reduced striatal DA levels but normal numbers of mutant cell bodies remained26. Whole genome RNA sequencing of adult DA neurons obtained by laser capture revealed that Nurr1 determines DA identity primarily through activation of gene expression. In addition to maintaining expression of DA transmitter identity genes, Nurr1 is also required for maintenance of expression of genes involved in oxidative phosphorylation and mitochondrial function, which as discussed below, is potentially of significant clinical relevance. In 5–HT neurons, sustained expression of tryptophan hydroxylase, serotonin transporter, VMAT2, and vesicular glutamate transporter III was significantly weakened after adult targeting of Pet–1 or Lmx1b29, 40. A paired homeodomain orthologue, Pitx, appears to function similarly to maintain 5–HT neuron identity in planaria45.
Although sustained expression of key identity genes, such as those encoding rate–limiting enzymes for monoaminergic neurotransmitter synthesis, are highly sensitive to loss of adult TFs other transmitter identity genes become independent of the TFs that were necessary for their initial expression in fetal postmitotic precursors. For example, Pet–1 is required for the initial fetal expression of AADC and the 5–HT autoreceptor, Htr1a; however, targeted loss of Pet–1 in adult 5–HT neurons had no effect on the maintenance of their expression. Similarly AADC expression becomes insensitive to knock out of Lmx1b in adult 5HT neurons29. Interestingly, while VMAT2 becomes resistant to loss of Pet–1 in adult 5–HT neurons it remains dependent on Lmx1b for its maintenance in adult 5–HT neurons29, 40. These findings indicate evolving terminal selector requirements as the animal matures for maintenance of individual identity features. Individual identity features that lose dependence on a particular terminal selector may remain dependent on another for stable life long expression. Alternatively, terminal selectors can compensate for one another and even cooperate synergistically in the joint regulation of individual identity features16, 46. These types of neuronal terminal selector interactions have not yet been explored in vertebrates and therefore it is not clear whether maintained expression of an identity feature that is resistant to knock out of one or another terminal selector is lost or significantly reduced when both factors are targeted.
Autoregulation
The continuous expression of terminal selectors raises the question of how their expression is sustained given that the upstream regulatory factors that induced their expression in postmitotic precursors are generally downregulated at this stage. In several cases, such maintained expression has been explicitly shown to be mediated by direct autoregulation, the simplest form of a classic feedback loop11, 14, 47,48. For example, the LIM homeobox TF mec–3 (the “M” in LIM domain) autoregulates in C. elegans touch sensory neurons, which require mec–3 for their terminal differentiation49, the Zn finger TFs che–1 and odr–7 autoregulate in the C.elegans ASE and AWA sensory neurons, respectively11, 50, in which they initiate the respective differentiation programs and the LIM homeobox TF ttx–3 and the Prd–type homeobox TF ceh–10 jointly autoregulate their expression in the cholinergic AIY interneurons, aside from initiating the terminal differentiation program of AIY14. In all cases examined, autoregulation is direct. For example, TTX–3 and CEH–10 jointly control downstream, terminal effector genes via a cognate binding site, the “AIY motif” and this AIY is found in the ttx–3 and ceh–10 loci themselves, where it is functionally required for their maintained expression14, 51. Not only do terminal selectors autoregulate, but factors that only regulate specific subaspects of a neuron type, such as the above-mentioned die-1 transcription factor that control chemoreceptor expression, also autoregulate52.
Experiments aimed at testing the autoregulatory concept for mouse terminal selector–type factors are limited but do provide clear support for it at least in adult 5–HT neurons. The ePet::CreERT2 transgenic construct developed for targeting Pet–1 in adult 5–HT neurons was used as a reporter to determine whether Pet–1 expression in adulthood was needed to maintain transgenic CreER expression. In control animals injected with vehicle strong expression of CreER mRNA was detected in the cell bodies of DRN 5–HT neurons. However, in tamoxifen treated mice, which resulted in a severe knock down of Pet–1 expression, the expression of CreER was not sustained. Chromatin immunoprecipitation assays with chromatin isolated from embryonic hindbrain demonstrated that Pet–1 occupies sites in the Pet–1 upstream regulatory region where conserved Pet–1 ETS binding sites are located. Together these studies suggest that sustained expression of Pet–1 requires positive autoregulation through direct binding of Pet–1 to its own cis regulatory region40.
Autoregulation of terminal selector transcription factors can be viewed as a “lock in” mechanism. Regulatory factors that act in immature neurons or neuroblasts to induce terminal selectors are usually only transiently expressed. Through the activation of autoregulating terminal selectors, a regulatory state that is driven by these earlier acting determinants is stably maintained (Fig. 1). This is well illustrated in the case of the AIY interneurons, in which transiently expressed Zic and bHLH transcription factors as well as a temporally controlled Wnt signal activate expression of TTX–3 and CEH–10, which then autoregulate their own expression51. A further illustration of this concept is provided by activation of autoregulating Pet–1 by transient Gata–2 expression in serotonergic precursors53. Gata–2 appears to directly control Pet–1 through binding to conserved Gata sites in the Pet–1 enhancer region54. However, Gata–2 expression begins to decline shortly after initiation of Pet–1 expression and is not detectable a few days later in 5–HT neurons40.
Figure 1. Initiation and maintenance of neuronal identity.

Regulatory factors expressed transiently in the lineage that gives rise to a specific neuron type induce the expression of terminal selector–type transcription factors. These factors then initiate terminal differentiation by directly activating the expression of terminal effector genes, such as genes coding for neurotransmitter synthesizing enzymes, transporter and receptors, ion channel, synaptic adhesion molecules and so on. The activated state of these effector genes is maintained via the terminal selectors that initiated their expression. Such maintenance is made possible through the often observed autoregulation of terminal selectors. Other factors that played no role in initiating effector genes may assist terminal selectors in maintaining the differentiated state. In at least one case examined, these maintenance factors are controlled by terminal selectors as well (see text).
Neuronal survival, morphology, and connectivity
A remarkable outcome of studies aimed at determining “late” functions of continuously expressed TFs in the vertebrate brain is that their roles are not limited to maintenance of neurotransmitter identity features. Indeed, over the past few years the critical need for neuronal maintenance throughout life has been powerfully reinforced with findings demonstrating that targeted disruption of postnatal TF expression trajectories can lead to dendritic and axonal atrophy, neuron death, transdifferentiation and dramatic behavioral changes that model human neurodegenerative and neuropsychiatric diseases.
An early indication of more diverse functions of continuously expressed TFs was the finding in conditionally targeted Lmx1b mice that following the loss of serotonergic neuron identity markers at fetal stages nearly all 5–HT neuron cell bodies were missing from the neonatal brain suggesting Lmx1b is also required for 5–HT neuron survival30. Pet–1 does not appear to be necessary for survival of 5–HT neurons as Pet–1 deficient cell bodies remain in normal numbers within the DRN after conditional targeting of Pet–1. Thus, although these TFs share many common targets in 5–HT postmitotic precursors, Lmx1b likely regulates a distinct gene battery important for early postnatal neuron survival. Lmx1b becomes dispensable for maintenance of 5–HT neuron survival at some undefined postnatal stage as targeting of Lmx1b in adult 5–HT neurons does not lead to disappearance of Pet–1+ 5–HT cell bodies29.
Interestingly, ongoing expression of the paired–type homeobox TF, Pax6, is intrinsically required, in a highly specific manner, for survival of dopaminergic neurons in the olfactory bulb through maintenance of crystalline αA expression, which prevents caspase–3–mediated programmed cell death55. Similarly, the zinc finger factor, Gata–3, is continuously required in adult sympathetic neurons for survival (but not identity) of virtually all sympathetic neurons through transcriptional maintenance of proper levels of expression of several anti–apoptotic and pro–apoptotic genes56. Lastly, a targeted investigation of the role of homeobox genes in the maintaining adult glutamatergic neurons in the mouse CNS has identified the LIM homeobox gene Lhx1 and the POU homeobox gene Brn3 as being continuously expressed in glutamatergic neurons in inferior olivary neurons in the brain stem. Removal of Lhx1 in adult mice resulted in the loss of this neuron type18. Whether Lhx1 is exclusively required for neuronal survival, or whether loss of these neurons is a secondary consequence of loss of the neurotransmitter phenotype of these neurons is currently not known.
The retinoic acid–related orphan nuclear receptor α (Rorα) is expressed at high levels in fetal Purkinje cells (PCs) and is essential for their survival and maturation57. As Rorα expression continues across the life span conditional targeting approaches were used to determine whether it is also required for maintenance of PCs58. Conditional disruption of ongoing Rorα expression between postnatal days 10–21, when PC maturation is nearly complete, led to dendritic atrophy and distal spine loss58. Furthermore, substantially decreased numbers of PCs were detected several weeks later. In addition, loss of Rorα led to reversion of the mature mono–innervation pattern of climbing fiber input onto PCs to a multi–innervation pattern indicative of an earlier stage of development. Significantly, dendritic atrophy and morphological alterations of climbing fiber input to PCs also occurred after targeting of Rorα in adult PCs. A progressive dendritic and axonal dystrophy also resulted from loss of Nurr1 in adult DA neurons26. These findings provide intriguing examples of the requirement for transcriptional maintenance of adult neural connectivity by the very TFs that initially setup the neural circuitry58.
Transcriptional maintenance in the adult retina has recently been demonstrated to be crucial for preservation of visual activity and the prevention of late onset photoreceptor degeneration. Otx2, a paired–type homeodomain protein, is an essential regulator of multiple early steps in eye development including determination of the retinal pigment epithelium, photoreceptor cell fate and bipolar cell differentiation59, 60. The persistent expression of Otx2 in the retinal pigment epithelium (RPE), photoreceptors and bipolar cells led to studies of its potential postnatal functions in preserving retinal integrity and function. Conditional targeting of Otx2 at P30 resulted in a progressive and ultimately an apoptotic–driven total loss of photoreceptors61. A more rapid degeneration of the RPE occurred resulting in disruption of RPE cell adhesive contacts with photoreceptors. RPE–specific targeting of Otx2 disrupted maintenance of RPE specific gene expression and led to a near total degeneration of photoreceptors similar in magnitude to that resulting from deficiency of Otx2 from all Otx2+ retinal cell types. These findings demonstrated that ongoing Otx2 function, intrinsic to RPE cells, regulates photoreceptor survival through cell non–autonomous interactions with the RPE62.
Another noteworthy study illustrated the requirement for Nrl, a basic leucine zipper TF, in the postnatal maintenance of retinal photoreceptors. Nrl initially functions in retinal precursors as a switch for specification of rod photoreceptors over cones63. Montana et al. have shown that conditional targeting of Nrl in adult retina led to loss of rod–enriched gene expression and derepression of some cones genes, hence a partial rods–to–cones transdifferentiation. Further evidence in support of transdifferentiation in adult targeted cells was the appearance of photoreceptors with the morphological features of cone–like cells and retinal physiological changes indicative of enhanced cone activity. The potential for rod to cones switching with targeted Nrl loss of function was highest at birth and progressively declined as the animal matured. The reprogramming of rods to cones in Nrl postnatal conditional mutant mice indicates that ongoing Nrl expression is necessary for maintenance of rod photoreceptor identity and suppression of cone traits64.
Dedicated maintenance factors
The maintenance TFs discussed so far are also necessary for initial differentiation of neuron–type identities. Are there additional kinds of maintenance factors whose expression does not commence until after acquisition of neuron–type identity and therefore are not involved in initial specification? In other words is there a separate class of TF that is dedicated to maintenance of neuronal identity or survival? Dedicated maintenance factors have been described in C.elegans. The ASE chemosensory neurons require the che–1 Zn finger transcription factor for the initiation and maintenance of many distinct terminal identity features of the ASE neurons, including the expression of putative chemoreceptors from the receptor–type guanylyl cyclase (rGC) family11. The Otx–type transcription factor ceh–36 has a minor role in initiating the expression of the rGC proteins, but becomes indispensable at later stages to maintain rGC expression52. Importantly, the ceh–36 maintenance factor is a direct target of che–111 and the both operate via distinct binding sites in a feed forward configuration to control rGC expression11. This demonstrates that initiators of differentiation programs also induce the expression of maintenance factors that will assist the initiating factors in maintaining gene expression programs (Fig. 1).
The engrailed (En) homeodomain paralogues, En1 and En265, are also crucial for maintenance of mesDA neurons but not acquisition of DA identity. Simon et al., showed with En1/tau–LacZ null and En2−/− double knock out mice that nearly normal numbers of TH+ mesDA neurons were generated and acquired their dopaminergic identity66. However, soon thereafter DA cell bodies in the SNc and VTA completely disappeared from the fetal midbrain. These results are consistent with the finding that onset of En expression in postmitotic mesDA neurons occurs 2–3 days after acquisition of TH identity67. Further studies suggested that En maintains postmitotic mesDA neurons cell autonomously by prevention of apoptosis67. Therefore, at least for the majority of mesDA neurons, En1/2 are not required for acquisition of DA neuron identity but instead are dedicated maintenance factors essential for their postmitotic survival68. In addition, their maintenance function may extend to neuroprotection as evidenced in MPTP, 6–OHDA, and α–synuclein models of Parkinson’s disease69.
En1/2 may play a similar dedicated maintenance role in serotonergic neurons. Conditional targeting of En1/2 specifically in postmitotic fetal 5–HT neurons did not interfere with specification or early maintenance of 5–HT neuron identity. However, beginning at E16.5 a progressive perinatal decline in the numbers of 5–HT+ neurons in the dorsal raphe nucleus (DRN) occurred as a result of failed maintenance of Tph2 expression. The loss of serotonergic identity was followed by apoptotic loss of 5–HT cell bodies70, which resulted in a significant deficit in adult forebrain 5HT. Since En1/2 is also expressed earlier in 5–HT neuron lineage71, it cannot be excluded that apart from its maintenance role, it is also required at the progenitors stage. An intriguing preliminary finding is that En1 expression appears to be reduced in Pet–1 deficient 5–HT neurons and conserved Pet–1 ETS binding sites are present in the En1 upstream region. This suggests that similar to direct control of ceh–36 by che–1, continuous autoregulated Pet–1 expression is required not only for direct maintenance of 5–HT identity features but also to directly recruit other 5–HT maintenance factors (Fig. 1).
A different DA neuron class, juxtaglomerular dopaminergic olfactory bulb (OB) interneurons, may also depend on a dedicated transcriptional maintenance factor, COUP–TF1 encoded by Nr2f1, for preservation of neuron–type identity. Although OB interneurons of various identities are continually generated throughout life from progenitors, COUP–TF1 expression is largely restricted to mature DA and GABAergic interneurons within the glomerular (GL) and granule cell (GcL) layers and is rarely detectable in subventricular zone–derived neuroblasts that will enter the bulb72. Early conditional targeting in the EMX1–derived OB interneuron lineage indicated that COUP–TF1 was not necessary for the generation or survival of DA interneurons. However, Nr2f1–deficiency led to a loss of TH expression in adult GL interneurons but not in DA interneurons generated in the early postnatal period. Moreover, targeting of Nr2f1 with lenti–viral Cre injection into the adult rostral migratory stream did not affect TH expression in the GL after thirty days but led to a significant decrease in GL TH expression at 60 days post–injection. Together these findings suggest COUP–TF1 is not involved in acquisition of DA interneuron identity but instead is required for maintenance of adult DA identity72.
Plasticity of maintenance programs
Interestingly, olfactory sensory deprivation through naris closure caused a dramatic decrease in the number of COUP–TF1+ neurons in the GL. Decreased expression of COUP–TF1 following sensory deprivation was accompanied by a severe reduction in the numbers of TH+ neurons in the GL, which is in keeping with the well–documented control of TH expression in OB DA neurons by afferent synaptic activity73–75. These findings support a model in which activity–dependent control of COUP–TF1 is essential for maintenance of OB DA neuron identity by COUP–TF172 and reveal the potential for environmentally driven plasticity of gene regulatory maintenance programs in postmitotic neurons. Are other transcriptional maintenance programs influenced by sensory–stimulation or other types of environmental signals such as secreted signaling molecules? Although gene regulatory programs that specify neuron–type identity are induced by the interaction of secreted morphogens, once launched, these programs appear to function cell autonomously in the specification of neuron–type precursors. However, recent studies in Xenopus leavis embryos and larvae have shown that early in development but after progenitor specification changes in spontaneous calcium spike activity can change the number of differentiating neurons that express serotonergic identity. Increased spike activity suppressed the number of Lmx1b+, Tph+ and 5–HT+ neurons, while decreases in spike activity led to increased numbers of these neurons, thus revealing early homeostatic control of serotonergic signaling in a reserve pool of neurons in response to altered activity6, 76. Further studies showed that Lmx1b is required for changes in 5–HT identity in response to altered calcium spike activity. Thus, Xenopus Lmx1b is sensitive to changes in spontaneous activity in a subset of differentiating neurons that are serotonergic or are poised to become serotonergic. These results reveal neuroplasticity of gene regulatory programs that govern the differentiation or early maintenance of 5–HT neuron identity and raise the intriguing question of whether later in life maintenance of 5–HT neuron identity can be influenced by synaptic activity–dependent changes in the expression of terminal selectors.
In addition to activity–dependent control of transmitter acquisition in early development and maintenance of OB DA interneuron identity, there are striking examples of postmitotic neurons that acquire new transmitter identities as a result of cell extrinsic signaling or sensory input. In the adult hypothalamus, interneurons in the periventricular and paraventricular nuclei switch off their dopaminergic identity and acquire a somatostatin (SST) peptide identity in response to short and long–day photoperiods77. The changes in peptide expression were correlated with parallel changes in mRNAs encoding TH and SST indicating that transcriptional activation and repression mediated the switch in identity. The transcriptional mechanisms underlying the switch have not been explored but perhaps involve activity–dependent modulation of continuous Nurr1 expression78.
A well–studied example of transmitter switching is the adrenergic to cholinergic transmitter switch that occurs postnatally in specific small populations of postmitotic sympathetic neurons that innervate rodent eccrine sweat glands in footpads79. The switch is physiologically important, as it is required for the acquisition and maintenance of sweat gland secretion80. Target–derived neuropoetic secreted cytokines acting retrogradely on sympathetic neurons through gp130 receptor signaling are required for transmitter switching81. Although the transcriptional mechanisms involved in switching are poorly understood a recent study in primary sympathetic neuron cultures has shown that knockdown of Satb2, a chromatin organizer that binds to AT sequence–rich nuclear matrix attachment regions, prevents cytokine–stimulated upregulation of the cholinergic markers, Chat and VAcht, and suppression of noradrenergic markers, TH and DBH82. An intriguing possibility is that cytokine signaling via the gp130 receptor modifies chromatin architecture in a way that suppresses Phox2b and/or Hand2–mediated maintenance of noradrenergic identity and induces cholinergic identity through increased expression of Chat and VAcht transcriptional activators.
The role of chromatin modifications
The studies discussed so far demonstrate that both in vertebrate and invertebrates, sequence–specific DNA binding proteins that initiate terminal differentiation programs in postmitotic neurons are also continuously required to maintain the differentiated state. Cis–regulatory control regions and transcribed genes are often marked by specific patterns of chromatin modifications. For example, histone H3 is methylated at lysine 4 across actively transcribed genes83. Such types of chromatin modifications are important in maintaining differentiated features of neurons, as for example illustrated through the use of a temperature–sensitive mutant allele in a histone modification enzyme21. The continuous requirement of this enzyme is mirrored by the continuous requirement of a Zn finger transcription factor that likely recruits this enzyme21. However, the continuous requirement of TFs illustrates that histone modifications are alone not sufficient to maintain regulatory states. In other words, a TF that initiates specific patterns of gene expression and thereby is responsible for the introduction of histone modification patterns that mark transcriptionally active regions does not become superfluous after introducing these modification patterns. The insufficiency of chromatin modifications in maintaining gene expression states is unfortunately not generally appreciated in the literature84. The importance of specific kinds of chromatin modifications in maintenance of neuronal identity is an intriguing question that will require conditional removal of histone writers and erasers in specific postmitotic neuron–types.
Potential late onset paths to nervous system dysfunction
The foregoing examples illustrate the necessity of transcriptional maintenance programs for preservation of long–lived postmitotic mammalian neurons. What is the potential relevance of disrupted postnatal neuronal maintenance to nervous system pathogenesis? Not surprisingly, given the severity of the molecular maintenance defects reported so far, disruption of ongoing TF expression causes clinically relevant late onset physiological phenotypes. For example, continuous Otx2 expression in the RPE is required for prevention of photoreceptor degeneration. Heterozygous mutations in human Otx2 have been linked to many different kinds of ocular malformations. Interestingly, several of the genes regulated by Otx2 in the RPE have been linked to retinal diseases including genes associated with increased susceptibility to age–related macular degeneration62. Thus, disruption of ongoing Otx2 expression is a potential path to susceptibility for late onset retinal degenerative diseases.
Postnatal Rorα deficiency in Purkinje cells caused a progressive impairment of various sensorimotor functions including deficits in motor coordination, muscle strength, and balance mimicking the severe ataxic phenotype of spontaneous (staggerer) or germ line targeted Rorα mutant mice58. Rorα is required for expression of several Purkinje cell genes involved in establishing normal motor learning and coordination85. Further, postnatal loss of Rorα expression promotes polyglutamine–expanded mutant ATAXIN1–mediated neurodegeneration of Purkinje cells86. Thus, continued Rorα expression might be required for maintained expression of key downstream genes important for preservation of normal cerebellar function and prevention of susceptibility to neurodegeneration.
The critical need for continuous transcriptional maintenance of neuron–type identity was demonstrated by the dramatically increased anxiety–like behaviors that resulted from adult targeting of Pet–140. These anxiety behaviors are reminiscent of the abnormal emotional behaviors that result from germ line targeting of Pet–1 and other serotonergic identity genes33, 87, 88. A large literature has demonstrated that altered expression of either Tph2 or Sert in animals can lead to a range of behavioral abnormalities. Moreover, suppression of Htr1a autoreceptor expression in 5HT neurons during the early postnatal period increased anxiety–like behaviors later in life89. Human genetic studies suggest allelic variation that influences Tph2 or Sert expression or function is associated with increased susceptibility for various neuropsychiatric disorders90–92. Tph2 and Sert are direct targets of Pet–140, 93 and their reduced expression along with the accompanying significantly reduced levels of forebrain 5–HT that resulted from adult targeting of Pet–1 might therefore be solely responsible for the abnormal anxiety phenotype. However, it is likely that Pet–1 is required for maintained expression of other genes in adult 5–HT neurons and therefore a complex combination of mis–expressed genes in adult 5HT neurons may have contributed to defective 5–HT signaling that precipitated altered anxiety behaviors. Although human susceptibility variants impacting expression and function of the Pet–1 human ortholog, FEV, have not been identified it remains an intriguing possibility that variation in FEV or its autoregulatory sites could disrupt maintenance of the serotonergic gene expression and consequently contribute risk for mental health disorders.
The cellular and molecular pathology present in midbrain DA neurons following postnatal targeting of Nurr1 suggests that deficient NURR1 expression in postmitotic DA neurons might contribute to Parkinson’s disease (PD) pathogenesis94. Significant locomotor impairment accompanied the progressive cellular and morphological pathology of the DA system after disruption of adult Nurr1 expression,26. Further, loss of adult Nurr1 led to dystrophic dopaminergic dendrites, which is a pathological feature reported in early stages of PD. Nurr1 has also been implicated is regulation of genes involved in neuroprotection against oxidative and excitotoxic stress95. NURR1 expression is decreased in remaining SN DA neurons present in confirmed cases of PD, which parallels the loss of TH in the PD brain96. Thus, although human NURR1 has not been identified as a major susceptibility locus for PD it remains possible that rare genetic mutations97 or environmental factors might lead to decreased NURR1 expression and function in adulthood and thereby accelerate PD disease progression94. For example, α–synuclein gene (SNCA) missense and gene duplication mutations are known to cause familial cases of PD and the level of α–synuclein overexpression correlates with disease severity98. Significantly, overexpression of SNCA in the adult rat SN decreased the expression of Nurr1 and the downstream DA neuron gene battery comprising Th, Aadc, Vmat2, Dat, demonstrating that α–synuclein interferes with Nurr1–dependent maintenance of DA neuron identity99. In addition, Nurr1 is required for maintenance of the GDNF neuroprotection signaling pathway in DA neurons, which is also disrupted with α–synuclein overexpression. Importantly, viral–mediated delivery of Nurr1 in α–synuclein–overexpressing rats largely reversed α–synuclein–induced toxic effects on locomotor function, GDNF signaling and DA neuron identity99. These findings demonstrate that disruption of transcriptional maintenance programs is likely to be a key element underlying the progression of PD pathogenesis and that restoration of Nurr1 may offer a therapeutic avenue for treatment of the disorder94.
Conclusions and Future Directions
The findings we have reviewed demonstrate that gene regulatory programs launched early in fetal life are required across the life span to maintain many vital features necessary for continued normal neuronal circuit function and behavior (Fig. 1). In addition, examples of TFs that are required exclusively for maintenance of neuron–type identity but do not appear to be involved in the initial specification process have also been described although maintenance of these factors depend on the initiation TFs. These studies constitute an important beginning in understanding the regulatory mechanisms that function to maintain postmitotic neuronal identity. Yet many fundamental questions remained unanswered. First, our knowledge of gene regulatory maintenance programs in specific neuron–types is sketchy. In addition, the breadth of neuronal features that are actively maintained is poorly understood although the findings we have reviewed have so far indicated that maintenance is not restricted to just neurotransmitter–type identity features. Whole genome expression profiling as was performed for Nurr1 in DA neurons will help to reveal the types of gene batteries and range of neuronal functions that are under active maintenance at different stages of life. These approaches may also reveal additional terminal selectors and other dedicated maintenance factors involved in postmitotic neuronal maintenance. Second, the requirements for terminal selector dependent control of individual identity features throughout life are far from clear, particularly in vertebrate neuron–types. Terminal identity genes that become independent of one terminal selector in postnatal life might remain dependent on another untested TF for stable life long expression or perhaps the dependency switches to another TF. Individual terminal selectors may also act cooperatively or compensate for one another in the joint control of identity genes. More complex targeting strategies will need to be implemented to investigate terminal selector compensation, cooperation, and potential switching. Third, sensory–stimuli and electrical activity can influence adult neuronal identity through changes in expression of neurotransmitters. It is presently unclear, however, how common this type of plasticity is in different neuron–types. An intriguing question is whether there are activity–dependent critical periods for transcriptional maintenance of certain features of terminal identity?
Although further investigation of gene regulatory maintenance programs will be more straightforward in C. elegans and will likely provide additional key conceptual advances into conserved maintenance mechanisms it will also be essential, depending on availability of suitable genetic tools, to continue the study of these programs in specific neuronal types within the vertebrate nervous system. Finally, an important untested hypothesis for future investigation is whether fetal or postnatal environmental influences (nutritional status, emotional stress, toxins, pathogens etc.) or the acquisition of de novo somatic mutations, perhaps in cis–regulatory regions of genes encoding neuronal identity features or terminal selector autoregulatory elements result in maintenance defects and consequent susceptibility for nervous system disease.
Box 1. KEY POINTS.
Studies in C. elegans, Drosophila, and mice have demonstrated that neuronal identity must be actively maintained throughout life.
Invertebrate and vertebrate terminal selector transcription factors that initially control acquisition of neuron–type identity continue to function throughout life to maintain identity.
Additional transcription factors whose expression begins in postmitotic neurons after acquisition of neuron–type identity have been implicated in maintenance of neuron–type identity.
Adult transcriptional maintenance programs play key roles in neuronal survival and preservation of neural circuit function.
Expression of terminal selectors in postmitotic neurons is maintained through direct autoregulation.
Chromatin modifiers are important for maintenance of differentiated states but depend on continuous recruitment by sequence–specific transcription factors.
Disruption of terminal selector expression in adult mice leads to human disease–related pathophysiology.
Acknowledgments
The authors are funded by the NIH (R01NS039996 and R01NS050266–03 to O.H. and RO1MH062723 and P50MH096972 to ESD.) and the HHMI (OH). We thank Doug Allan, Paschalis Kratsios, Nuria Flames, Piali Sengupta, Steven Wyler, Clay Spencer for comments prior to submission. We also thank the three referees for their constructive comments. We are indebted to Mark Ptashne for his insightful comments and suggestions throughout the writing of this review.
Contributor Information
Evan S. Deneris, Email: esd@case.edu.
Oliver Hobert, Email: or38@columbia.edu.
References
- 1.Masland RH. Neuronal cell types. Current biology : CB. 2004;14:R497–500. doi: 10.1016/j.cub.2004.06.035. [DOI] [PubMed] [Google Scholar]
- 2.Nelson SB, Sugino K, Hempel CM. The problem of neuronal cell types: a physiological genomics approach. Trends in neurosciences. 2006;29:339–345. doi: 10.1016/j.tins.2006.05.004. [DOI] [PubMed] [Google Scholar]
- 3.Hobert O, Carrera I, Stefanakis N. The molecular and gene regulatory signature of a neuron. Trends in neurosciences. 2010;33:435–445. doi: 10.1016/j.tins.2010.05.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fishell G, Heintz N. The neuron identity problem: form meets function. Neuron. 2013;80:602–612. doi: 10.1016/j.neuron.2013.10.035. [DOI] [PubMed] [Google Scholar]
- 5.Flavell SW, Greenberg ME. Signaling mechanisms linking neuronal activity to gene expression and plasticity of the nervous system. Annu Rev Neurosci. 2008;31:563–590. doi: 10.1146/annurev.neuro.31.060407.125631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Spitzer NC. Activity-dependent neurotransmitter respecification. Nat Rev Neurosci. 2012;13:94–106. doi: 10.1038/nrn3154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Achim K, Salminen M, Partanen J. Mechanisms regulating GABAergic neuron development. Cell Mol Life Sci. 2013 doi: 10.1007/s00018-013-1501-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Holmberg J, Perlmann T. Maintaining differentiated cellular identity. Nat Rev Genet. 2012;13:429–439. doi: 10.1038/nrg3209. [DOI] [PubMed] [Google Scholar]
- 9.Hobert O. Neurogenesis in the nematode Caenorhabditis elegans. WormBook : the online review of C. elegans biology. 2010:1–24. doi: 10.1895/wormbook.1.12.2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hobert O. Regulation of terminal differentiation programs in the nervous system. Annual review of cell and developmental biology. 2011;27:681–696. doi: 10.1146/annurev-cellbio-092910-154226. [DOI] [PubMed] [Google Scholar]
- 11.Etchberger JF, et al. The molecular signature and cis-regulatory architecture of a C. elegans gustatory neuron. Genes Dev. 2007;21:1653–1674. doi: 10.1101/gad.1560107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kratsios P, Stolfi A, Levine M, Hobert O. Coordinated regulation of cholinergic motor neuron traits through a conserved terminal selector gene. Nat Neurosci. 2012;15:205–214. doi: 10.1038/nn.2989. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Duggan A, Ma C, Chalfie M. Regulation of touch receptor differentiation by the Caenorhabditis elegans mec-3 and unc-86 genes. Development. 1998;125:4107–4119. doi: 10.1242/dev.125.20.4107. [DOI] [PubMed] [Google Scholar]
- 14.Wenick AS, Hobert O. Genomic cis-regulatory architecture and trans-acting regulators of a single interneuron-specific gene battery in C. elegans. Dev Cell. 2004;6:757–770. doi: 10.1016/j.devcel.2004.05.004. [DOI] [PubMed] [Google Scholar]
- 15.Eastman C, Horvitz HR, Jin Y. Coordinated transcriptional regulation of the unc-25 glutamic acid decarboxylase and the unc-47 GABA vesicular transporter by the Caenorhabditis elegans UNC-30 homeodomain protein. J Neurosci. 1999;19:6225–6234. doi: 10.1523/JNEUROSCI.19-15-06225.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Doitsidou M, et al. A combinatorial regulatory signature controls terminal differentiation of the dopaminergic nervous system in C. elegans. Genes Dev. 2013;27:1391–1405. doi: 10.1101/gad.217224.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Etchberger JF, Flowers EB, Poole RJ, Bashllari E, Hobert O. Cis-regulatory mechanisms of left/right asymmetric neuron-subtype specification in C. elegans. Development. 2009;136:147–160. doi: 10.1242/dev.030064. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Serrano-Saiz E, et al. Modular control of glutamatergic neuronal identity in C. elegans by distinct homeodomain proteins. Cell. 2013;155:659–673. doi: 10.1016/j.cell.2013.09.052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Flames N, Hobert O. Gene regulatory logic of dopamine neuron differentiation. Nature. 2009;458:885–889. doi: 10.1038/nature07929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chang S, Johnston RJ, Jr, Frokjaer-Jensen C, Lockery S, Hobert O. MicroRNAs act sequentially and asymmetrically to control chemosensory laterality in the nematode. Nature. 2004;430:785–789. doi: 10.1038/nature02752. [DOI] [PubMed] [Google Scholar]
- 21.O’Meara MM, Zhang F, Hobert O. Maintenance of neuronal laterality in Caenorhabditis elegans through MYST histone acetyltransferase complex components LSY-12, LSY-13 and LIN-49. Genetics. 2010;186:1497–1502. doi: 10.1534/genetics.110.123661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Jafari S, et al. Combinatorial activation and repression by seven transcription factors specify Drosophila odorant receptor expression. PLoS Biol. 2012;10:e1001280. doi: 10.1371/journal.pbio.1001280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hsiao HY, Jukam D, Johnston R, Desplan C. The neuronal transcription factor erect wing regulates specification and maintenance of Drosophila R8 photoreceptor subtypes. Dev Biol. 2013;381:482–490. doi: 10.1016/j.ydbio.2013.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Eade KT, Fancher HA, Ridyard MS, Allan DW. Developmental transcriptional networks are required to maintain neuronal subtype identity in the mature nervous system. PLoS Genet. 2012;8:e1002501. doi: 10.1371/journal.pgen.1002501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Coppola E, d’Autreaux F, Rijli FM, Brunet JF. Ongoing roles of Phox2 homeodomain transcription factors during neuronal differentiation. Development. 2010;137:4211–4220. doi: 10.1242/dev.056747. [DOI] [PubMed] [Google Scholar]
- 26.Kadkhodaei B, et al. Transcription factor Nurr1 maintains fiber integrity and nuclear-encoded mitochondrial gene expression in dopamine neurons. Proc Natl Acad Sci U S A. 2013;110:2360–2365. doi: 10.1073/pnas.1221077110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kadkhodaei B, et al. Nurr1 is required for maintenance of maturing and adult midbrain dopamine neurons. J Neurosci. 2009;29:15923–15932. doi: 10.1523/JNEUROSCI.3910-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Schmidt M, et al. The bHLH transcription factor Hand2 is essential for the maintenance of noradrenergic properties in differentiated sympathetic neurons. Dev Biol. 2009;329:191–200. doi: 10.1016/j.ydbio.2009.02.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Song NN, et al. Adult raphe-specific deletion of lmx1b leads to central serotonin deficiency. PLoS One. 2011;6:e15998. doi: 10.1371/journal.pone.0015998. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhao ZQ, et al. Lmx1b is required for maintenance of central serotonergic neurons and mice lacking central serotonergic system exhibit normal locomotor activity. J Neurosci. 2006;26:12781–12788. doi: 10.1523/JNEUROSCI.4143-06.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Zetterstrom RH, et al. Dopamine neuron agenesis in Nurr1-deficient mice. Science. 1997;276:248–250. doi: 10.1126/science.276.5310.248. [DOI] [PubMed] [Google Scholar]
- 32.Saucedo-Cardenas O, et al. Nurr1 is essential for the induction of the dopaminergic phenotype and the survival of ventral mesencephalic late dopaminergic precursor neurons. Proc Natl Acad Sci USA. 1998;95:4013–4018. doi: 10.1073/pnas.95.7.4013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Hendricks TJ, et al. Pet-1 ETS gene plays a critical role in 5-HT neuron development and is required for normal anxiety-like and aggressive behavior. Neuron. 2003;37:233–247. doi: 10.1016/s0896-6273(02)01167-4. [DOI] [PubMed] [Google Scholar]
- 34.Ding YQ, et al. Lmx1b is essential for the development of serotonergic neurons. Nat Neurosci. 2003;6:933–938. doi: 10.1038/nn1104. [DOI] [PubMed] [Google Scholar]
- 35.Cheng L, et al. Lmx1b, Pet-1, and Nkx2.2 coordinately specify serotonergic neurotransmitter phenotype. J Neurosci. 2003;23:9961–9967. doi: 10.1523/JNEUROSCI.23-31-09961.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Pattyn A, Goridis C, Brunet JF. Specification of the central noradrenergic phenotype by the homeodomain gene Phox2b. Mol Cell Neurosci. 2000;15:235–243. doi: 10.1006/mcne.1999.0826. [DOI] [PubMed] [Google Scholar]
- 37.Pattyn A, Morin X, Cremer H, Goridis C, Brunet JF. The homeobox gene Phox2b is essential for the development of autonomic neural crest derivatives. Nature. 1999;399:366–370. doi: 10.1038/20700. [DOI] [PubMed] [Google Scholar]
- 38.Tsarovina K, et al. Essential role of Gata transcription factors in sympathetic neuron development. Development. 2004;131:4775–4786. doi: 10.1242/dev.01370. [DOI] [PubMed] [Google Scholar]
- 39.Hendershot TJ, et al. Conditional deletion of Hand2 reveals critical functions in neurogenesis and cell type-specific gene expression for development of neural crest-derived noradrenergic sympathetic ganglion neurons. Dev Biol. 2008;319:179–191. doi: 10.1016/j.ydbio.2008.03.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Liu C, et al. Pet-1 is required across different stages of life to regulate serotonergic function. Nat Neurosci. 2010;13:1190–1198. doi: 10.1038/nn.2623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Stott SR, et al. Foxa1 and foxa2 are required for the maintenance of dopaminergic properties in ventral midbrain neurons at late embryonic stages. J Neurosci. 2013;33:8022–8034. doi: 10.1523/JNEUROSCI.4774-12.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Lopes R, Verhey van Wijk N, Neves G, Pachnis V. Transcription factor LIM homeobox 7 (Lhx7) maintains subtype identity of cholinergic interneurons in the mammalian striatum. Proc Natl Acad Sci U S A. 2012;109:3119–3124. doi: 10.1073/pnas.1109251109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sun Y, et al. A central role for Islet1 in sensory neuron development linking sensory and spinal gene regulatory programs. Nat Neurosci. 2008;11:1283–1293. doi: 10.1038/nn.2209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Magno L, et al. The integrity of cholinergic basal forebrain neurons depends on expression of Nkx2-1. Eur J Neurosci. 2011;34:1767–1782. doi: 10.1111/j.1460-9568.2011.07890.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Marz M, Seebeck F, Bartscherer K. A Pitx transcription factor controls the establishment and maintenance of the serotonergic lineage in planarians. Development. 2013;140:4499–4509. doi: 10.1242/dev.100081. [DOI] [PubMed] [Google Scholar]
- 46.Zhang F, et al. The LIM and POU homeobox genes ttx-3 and unc-86 act as terminal selectors in distinct cholinergic and serotonergic neuron types. Development. 2014;141:422–435. doi: 10.1242/dev.099721. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Xue D, Finney M, Ruvkun G, Chalfie M. Regulation of the mec-3 gene by the C.elegans homeoproteins UNC-86 and MEC-3. EMBO J. 1992;11:4969–4979. doi: 10.1002/j.1460-2075.1992.tb05604.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Ptashne M. The chemistry of regulation of genes and other things. J Biol Chem. 2014;289:5417–5435. doi: 10.1074/jbc.X114.547323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Way JC, Chalfie M. The mec-3 gene of Caenorhabditis elegans requires its own product for maintained expression and is expressed in three neuronal cell types. Genes Dev. 1989;3:1823–1833. doi: 10.1101/gad.3.12a.1823. [DOI] [PubMed] [Google Scholar]
- 50.Sarafi-Reinach TR, Melkman T, Hobert O, Sengupta P. The lin-11 LIM homeobox gene specifies olfactory and chemosensory neuron fates in C. elegans. Development. 2001;128:3269–3281. doi: 10.1242/dev.128.17.3269. [DOI] [PubMed] [Google Scholar]
- 51.Bertrand V, Hobert O. Linking asymmetric cell division to the terminal differentiation program of postmitotic neurons in C. elegans. Dev Cell. 2009;16:563–575. doi: 10.1016/j.devcel.2009.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Cochella L, et al. Two distinct types of neuronal asymmetries are controlled by the Caenorhabditis elegans zinc finger transcription factor die-1. Genes Dev. 2014;28:34–43. doi: 10.1101/gad.233643.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Craven SE, et al. Gata2 specifies serotonergic neurons downstream of sonic hedgehog. Development. 2004;131:1165–1173. doi: 10.1242/dev.01024. [DOI] [PubMed] [Google Scholar]
- 54.Krueger KC, Deneris ES. Serotonergic transcription of human FEV reveals direct GATA factor interactions and fate of Pet-1-deficient serotonin neuron precursors. J Neurosci. 2008;28:12748–12758. doi: 10.1523/JNEUROSCI.4349-08.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Ninkovic J, et al. The transcription factor Pax6 regulates survival of dopaminergic olfactory bulb neurons via crystallin alphaA. Neuron. 2010;68:682–694. doi: 10.1016/j.neuron.2010.09.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Tsarovina K, et al. The Gata3 transcription factor is required for the survival of embryonic and adult sympathetic neurons. J Neurosci. 2010;30:10833–10843. doi: 10.1523/JNEUROSCI.0175-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Boukhtouche F, et al. RORalpha, a pivotal nuclear receptor for Purkinje neuron survival and differentiation: from development to ageing. Cerebellum. 2006;5:97–104. doi: 10.1080/14734220600750184. [DOI] [PubMed] [Google Scholar]
- 58.Chen XR, et al. Mature Purkinje cells require the retinoic acid-related orphan receptor-alpha (RORalpha) to maintain climbing fiber mono-innervation and other adult characteristics. J Neurosci. 2013;33:9546–9562. doi: 10.1523/JNEUROSCI.2977-12.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Koike C, et al. Functional roles of Otx2 transcription factor in postnatal mouse retinal development. Mol Cell Biol. 2007;27:8318–8329. doi: 10.1128/MCB.01209-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Martinez-Morales JR, Signore M, Acampora D, Simeone A, Bovolenta P. Otx genes are required for tissue specification in the developing eye. Development. 2001;128:2019–2030. doi: 10.1242/dev.128.11.2019. [DOI] [PubMed] [Google Scholar]
- 61.Beby F, et al. Otx2 gene deletion in adult mouse retina induces rapid RPE dystrophy and slow photoreceptor degeneration. PLoS One. 2010;5:e11673. doi: 10.1371/journal.pone.0011673. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Housset M, et al. Loss of Otx2 in the adult retina disrupts retinal pigment epithelium function, causing photoreceptor degeneration. J Neurosci. 2013;33:9890–9904. doi: 10.1523/JNEUROSCI.1099-13.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 63.Mears AJ, et al. Nrl is required for rod photoreceptor development. Nat Genet. 2001;29:447–452. doi: 10.1038/ng774. [DOI] [PubMed] [Google Scholar]
- 64.Montana CL, et al. Reprogramming of adult rod photoreceptors prevents retinal degeneration. Proc Natl Acad Sci U S A. 2013;110:1732–1737. doi: 10.1073/pnas.1214387110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 65.Liu A, Joyner AL. Early anterior/posterior patterning of the midbrain and cerebellum. Annu Rev Neurosci. 2001;24:869–896. doi: 10.1146/annurev.neuro.24.1.869. [DOI] [PubMed] [Google Scholar]
- 66.Simon HH, Saueressig H, Wurst W, Goulding MD, O’Leary DD. Fate of midbrain dopaminergic neurons controlled by the engrailed genes. J Neurosci. 2001;21:3126–3134. doi: 10.1523/JNEUROSCI.21-09-03126.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Alberi L, Sgado P, Simon HH. Engrailed genes are cell-autonomously required to prevent apoptosis in mesencephalic dopaminergic neurons. Development. 2004;131:3229–3236. doi: 10.1242/dev.01128. [DOI] [PubMed] [Google Scholar]
- 68.Simon HH, Thuret S, Alberi L. Midbrain dopaminergic neurons: control of their cell fate by the engrailed transcription factors. Cell Tissue Res. 2004;318:53–61. doi: 10.1007/s00441-004-0973-8. [DOI] [PubMed] [Google Scholar]
- 69.Alvarez-Fischer D, et al. Engrailed protects mouse midbrain dopaminergic neurons against mitochondrial complex I insults. Nature neuroscience. 2011;14:1260–1266. doi: 10.1038/nn.2916. [DOI] [PubMed] [Google Scholar]
- 70.Fox SR, Deneris ES. Engrailed is required in maturing serotonin neurons to regulate the cytoarchitecture and survival of the dorsal raphe nucleus. J Neurosci. 2012;32:7832–7842. doi: 10.1523/JNEUROSCI.5829-11.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 71.Li JY, Lao Z, Joyner AL. Changing requirements for Gbx2 in development of the cerebellum and maintenance of the mid/hindbrain organizer. Neuron. 2002;36:31–43. doi: 10.1016/s0896-6273(02)00935-2. [DOI] [PubMed] [Google Scholar]
- 72.Bovetti S, et al. COUP-TFI controls activity-dependent tyrosine hydroxylase expression in adult dopaminergic olfactory bulb interneurons. Development. 2013;140:4850–4859. doi: 10.1242/dev.089961. [DOI] [PubMed] [Google Scholar]
- 73.Kosaka T, Kosaka K, Hama K, Wu JY, Nagatsu I. Differential effect of functional olfactory deprivation on the GABAergic and catecholaminergic traits in the rat main olfactory bulb. Brain Res. 1987;413:197–203. doi: 10.1016/0006-8993(87)90171-5. [DOI] [PubMed] [Google Scholar]
- 74.Baker H, Morel K, Stone DM, Maruniak JA. Adult naris closure profoundly reduces tyrosine hydroxylase expression in mouse olfactory bulb. Brain Res. 1993;614:109–116. doi: 10.1016/0006-8993(93)91023-l. [DOI] [PubMed] [Google Scholar]
- 75.Cave JW, Baker H. Dopamine systems in the forebrain. Adv Exp Med Biol. 2009;651:15–35. doi: 10.1007/978-1-4419-0322-8_2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 76.Demarque M, Spitzer NC. Activity-dependent expression of Lmx1b regulates specification of serotonergic neurons modulating swimming behavior. Neuron. 2010;67:321–334. doi: 10.1016/j.neuron.2010.06.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 77.Dulcis D, Jamshidi P, Leutgeb S, Spitzer NC. Neurotransmitter switching in the adult brain regulates behavior. Science. 2013;340:449–453. doi: 10.1126/science.1234152. [DOI] [PubMed] [Google Scholar]
- 78.Liu N, Baker H. Activity-dependent Nurr1 and NGFI-B gene expression in adult mouse olfactory bulb. Neuroreport. 1999;10:747–751. doi: 10.1097/00001756-199903170-00016. [DOI] [PubMed] [Google Scholar]
- 79.Francis NJ, Landis SC. Cellular and molecular determinants of sympathetic neuron development. Annu Rev Neurosci. 1999;22:541–566. doi: 10.1146/annurev.neuro.22.1.541. [DOI] [PubMed] [Google Scholar]
- 80.Grant MP, Francis NJ, Landis SC. The role of acetylcholine in regulating secretory responsiveness in rat sweat glands. Mol Cell Neurosci. 1995;6:32–42. doi: 10.1006/mcne.1995.1004. [DOI] [PubMed] [Google Scholar]
- 81.Stanke M, et al. Target-dependent specification of the neurotransmitter phenotype: cholinergic differentiation of sympathetic neurons is mediated in vivo by gp 130 signaling. Development. 2006;133:141–150. doi: 10.1242/dev.02189. [DOI] [PubMed] [Google Scholar]
- 82.Apostolova G, Loy B, Dorn R, Dechant G. The sympathetic neurotransmitter switch depends on the nuclear matrix protein Satb2. J Neurosci. 2010;30:16356–16364. doi: 10.1523/JNEUROSCI.3502-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83.Shilatifard A. Molecular implementation and physiological roles for histone H3 lysine 4 (H3K4) methylation. Curr Opin Cell Biol. 2008;20:341–348. doi: 10.1016/j.ceb.2008.03.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 84.Ptashne M. Epigenetics: core misconcept. Proc Natl Acad Sci U S A. 2013;110:7101–7103. doi: 10.1073/pnas.1305399110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 85.Gold DA, et al. RORalpha coordinates reciprocal signaling in cerebellar development through sonic hedgehog and calcium-dependent pathways. Neuron. 2003;40:1119–1131. doi: 10.1016/s0896-6273(03)00769-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 86.Serra HG, et al. RORalpha-mediated Purkinje cell development determines disease severity in adult SCA1 mice. Cell. 2006;127:697–708. doi: 10.1016/j.cell.2006.09.036. [DOI] [PubMed] [Google Scholar]
- 87.Holmes A, Yang RJ, Lesch KP, Crawley JN, Murphy DL. Mice lacking the serotonin transporter exhibit 5-HT(1A) receptor-mediated abnormalities in tests for anxiety-like behavior. Neuropsychopharmacology. 2003;28:2077–2088. doi: 10.1038/sj.npp.1300266. [DOI] [PubMed] [Google Scholar]
- 88.Richardson-Jones JW, et al. Serotonin-1A Autoreceptors Are Necessary and Sufficient for the Normal Formation of Circuits Underlying Innate Anxiety. J Neurosci. 2011;31:6008–6018. doi: 10.1523/JNEUROSCI.5836-10.2011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 89.Donaldson ZR, et al. Developmental effects of serotonin 1A autoreceptors on anxiety and social behavior. Neuropsychopharmacology. 2014;39:291–302. doi: 10.1038/npp.2013.185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90.Veenstra-VanderWeele J, Blakely RD. Networking in autism: leveraging genetic, biomarker and model system findings in the search for new treatments. Neuropsychopharmacology. 2012;37:196–212. doi: 10.1038/npp.2011.185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91.Waider J, Araragi N, Gutknecht L, Lesch KP. Tryptophan hydroxylase-2 (TPH2) in disorders of cognitive control and emotion regulation: a perspective. Psychoneuroendocrinology. 2011;36:393–405. doi: 10.1016/j.psyneuen.2010.12.012. [DOI] [PubMed] [Google Scholar]
- 92.Murphy DL, Lesch KP. Targeting the murine serotonin transporter: insights into human neurobiology. Nat Rev Neurosci. 2008;9:85–96. doi: 10.1038/nrn2284. [DOI] [PubMed] [Google Scholar]
- 93.Deneris ES, Wyler SC. Serotonergic transcriptional networks and potential importance to mental health. Nature neuroscience. 2012 doi: 10.1038/nn.3039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 94.Decressac M, Volakakis N, Bjorklund A, Perlmann T. NURR1 in Parkinson disease–from pathogenesis to therapeutic potential. Nature reviews Neurology. 2013;9:629–636. doi: 10.1038/nrneurol.2013.209. [DOI] [PubMed] [Google Scholar]
- 95.Volakakis N, et al. NR4A orphan nuclear receptors as mediators of CREB-dependent neuroprotection. Proc Natl Acad Sci U S A. 2010;107:12317–12322. doi: 10.1073/pnas.1007088107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 96.Chu Y, et al. Nurr1 in Parkinson’s disease and related disorders. J Comp Neurol. 2006;494:495–514. doi: 10.1002/cne.20828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 97.Le WD, et al. Mutations in NR4A2 associated with familial Parkinson disease. Nat Genet. 2003;33:85–89. doi: 10.1038/ng1066. [DOI] [PubMed] [Google Scholar]
- 98.Farrer M, et al. Comparison of kindreds with parkinsonism and alpha-synuclein genomic multiplications. Ann Neurol. 2004;55:174–179. doi: 10.1002/ana.10846. [DOI] [PubMed] [Google Scholar]
- 99.Decressac M, et al. alpha-Synuclein-induced down-regulation of Nurr1 disrupts GDNF signaling in nigral dopamine neurons. Sci Transl Med. 2012;4:163ra156. doi: 10.1126/scitranslmed.3004676. [DOI] [PubMed] [Google Scholar]
