NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Alzate O, editor. Neuroproteomics. Boca Raton (FL): CRC Press/Taylor & Francis; 2010.

Cover of Neuroproteomics


Show details

Chapter 13Behaviorally Regulated mRNA and Protein Expression in the Songbird Brain

and .


Most biological processes are the result of the regulated expression of genes into their protein products in a defined temporal and spatial manner. The study of these dynamic protein changes in the brain is called neuroproteomics. A class of proteins expressed in the brain that have generated significant interests are those regulated by neural activity, called activity-dependent genes. The activity of neurons in the brain is important for normal brain function and expression of behavior. One system where the study of these features has been brought together is that of the songbird vocal learning system. Songbirds, and a limited number of animals (i.e., parrots, hummingbirds, bats, cetaceans, seals, elephants, and humans) are capable of vocal learning, the ability to produced imitative and improvisational sounds. Most animals do not have vocal learning but produce species-specific innate sounds used for alarm (e.g., predator) or other communication functions (e.g., alert for food or attracting a mate). The production of learned vocalizations requires the animal to process what it hears in the auditory pathway of the brain and then to produce the sounds heard, as song in songbirds or speech in humans, through specialized motor learning pathways. In songbirds and other vocal learning birds, this song learning motor pathway is organized into anatomically discrete nuclei that are not found in vocal non-learning birds. In this regard, songbirds represent a relatively unique animal model suited to the study of molecular mechanisms of a learned behavior with parallels to human speech. In recent years, a number of genes and their protein products have been identified that are regulated by neural activity during hearing and singing in the auditory and vocal brain pathways of songbirds. This chapter focuses on proteins regulated in the vocal pathway of songbirds performing learned vocalization. A discussion of proteins regulated in the auditory pathway by hearing is presented by Pinaud et al. in Chapter 14.


All avian species, whether vocal learners or vocal non-learners, share similar brain auditory pathways (Figure 13.1a) (1). Sounds activate ear cells, which synapse onto sensory neurons that project to cochlear and lemniscal nuclei in the brainstem. These neurons in turn project to the midbrain auditory mesencephalic lateral dorsal nucleus (MLd) and to the thalamic nucleus uvaeformis (Uva). The MLd projects to the thalamic auditory nucleus ovoidalis (Ov). Ov projects to primary auditory cell populations in field L2 of the forebrain pallium. L2 neurons then form a complex network to higher order auditory neurons in the caudal pallium, to fields L3 and L1 that project to the caudal medial nidopallium (NCM) and the caudal mesopallium (CM), respectively. In addition, L2 also projects to a shelf region below the HVC (used as a letter-based name) song nucleus and to an auditory region of the caudal striatum (CSt). A descending auditory pathway sequentially connects the HVC shelf to the intermediate arcopallium (Ai), which is adjacent to the song nucleus called the robust nucleus of the arcopallium (RA), known as the RA cup. Then the RA cup connects to the shell regions around Ov and MLd. This descending pathway is thought to modulate ascending auditory information (2–6).

FIGURE 13.1. Diagram of auditory (a) and vocal (b) pathways of the songbird brain.


Diagram of auditory (a) and vocal (b) pathways of the songbird brain. Only the most prominent or most studied projections are indicated. For the auditory pathway, NCM actually lies in a parasagittal plane medial to that depicted and reciprocal connections (more...)

Unliker the auditory brain pathway that is found in all avian species whether a vocal learner or non-learner, a forebrain vocal pathway is found only in species that learn their vocalizations. This pathway consists of seven brain regions, distributed into two sub-pathways, the posterior vocal pathway and the anterior vocal pathway (Figure 13.1b) (1). In the posterior pathway, the interfacial nucleus of the nidopallium (NIf) projects to HVC that in turn projects to RA; RA projects out of the forebrain to the dorsal medial nucleus of the midbrain (DM) and to the vocal motor neurons in the tracheosyringeal subdivision of the hypoglossal nucleus, nXIIts. The nXIIts projects to the muscles of syrinx, the avian vocal organ. The posterior vocal pathway is required for the production of learned vocalizations, as lesions to posterior vocal pathway nuclei HVC and RA either prevent singing or for HVC result in variable random-like singing, called subsong (7–9). The anterior vocal pathway is composed of the magnocellular nucleus of the anterior nidopallium (MAN; separated into lateral and medial parts, LMAN and MMAN) that projects to the striatal song nucleus Area X (also separated into lateral and medial parts, LArea X and MArea X). Area X projects to nuclei within the dorsal thalamus (DLM, the medial nucleus of the dorsalateral thalamus and DIP, dorsal intermediate posterior nucleus), which projects back to MAN forming closed loops (10–12). A mesopallium oval (MO) nucleus is part of this pathway in parrots, but its connectivity in songbirds is not known (13–15). The anterior vocal pathway is necessary for song learning, as lesions to anterior vocal pathway nuclei prevent juvenile song learning and adult song plasticity (16, 17). Connections from the posterior to the anterior vocal pathway occur via HVC to Area X, and from the anterior to the posterior occur via LMAN to RA and MMAN to HVC (12).

The source of auditory pathway input into the vocal pathway is an unresolved question. It has been proposed that input from the auditory pathway to the posterior vocal pathway can be by way of the HVC shelf to HVC, the RA cup to RA, or L2 into NIf (3). Other alternatives are that CM sends projections to NIf, and then NIf sends the auditory information to HVC and that CM sends a direct auditory projection to HVC (18). The thalamic nucleus Uva also sends projections to NIf, which sends a projection to HVC as well a direct projection from Uva to HVC (Figure 13.1a). It is conceivable that there are multiple contributing inputs. With this anatomical background, we now describe behaviorally regulated mRNA and protein expression in the song system.


13.3.1. Initial Discoveries with the ZENK GENE

Increased neuronal firing in the auditory pathway of birds hearing song and in the vocal pathway of birds singing is associated with high increases in the synthesis of mRNA and protein products of the immediate early gene (IEG) zenk (19, 20). ZENK protein (also known as Zif268, Egr-1, NGFIA, and Krox 24) is a transcription factor that regulates the expression of other genes. In the brain, IEG, its mRNA expression, is activity dependent; zenk requires increased neuronal firing for its increased expression (21). The study of Zenk has led to a sub-field of research on the study of mRNA and protein regulation in the brain by natural behaviors, using songbirds. The regulated expression that occurs in the auditory pathway is discussed in Chapter 14. In this chapter, we discuss the behaviorally regulated expression found in the vocal pathways during singing.

ZENK expression increases in vocal nuclei (also known as song nuclei) when birds sing (Figure 13.2). The increase occurs in the absence of hearing, i.e., in deafened birds that sing, and in the absence of somatosensory feedback from the syrinx muscle, i.e., in muted birds from nXIIts nerve cuts that attempt to sing but cannot move their syrinx muscles (silent song) (19). The amount of increased zenk mRNA and protein expression per 30 minutes correlates with the amount of song the bird produces per 30-minute period. Increased expression of zenk mRNA peaks at 30 minutes after singing starts, and then partly habituates to a steady state lower level in the song nuclei, as a bird continues to sing (19). A similar profile appears to occur with the protein with a delay of about 30 minutes, i.e., 60 minutes for peak expression (22).

FIGURE 13.2. ZENK mRNA expression in canary brain.


ZENK mRNA expression in canary brain. Shown are dark field views of cresyl violet stained sagittal brain sections reacted in situ hybridization experiments with a 35S-zenk riboprobe (white grains). One can see a subset of structures (arrows), as outlined (more...)

Within the song nucleus HVC, depending on the amount of singing, up to 40%-80% of the neurons that project to RA, called RA-projecting neurons, and a comparable percentage of neurons that project to Area X, called X-projecting neurons, show singing-induced ZENK protein expression (14). These findings suggest that a large portion of HVC neurons participate in singing behavior.

Some of the initial discoveries of behaviorally regulated gene expression in songbirds were also made with another IEG transcription factor, c-fos (23). Expression of the c-Fos protein was found to be induced in the RA-projecting neurons of HVC and in the RA nucleus itself, but very little in the X-projecting neurons of HVC or in Area X itself. However, this apparent difference in ZENK expression could be due to social context (see Social Context Regulation of Gene Expression). As seen for ZENK, c-Fos expression in song nuclei depends on the act of singing and the amount of protein product correlates with the amount of song produced (23).

13.3.2. Social Context Regulation of Gene Expression

Expression of ZENK in specific song nuclei was also found to depend on the social context in which a songbird sings. When a zebra finch male sings directed song facing another bird, ZENK expression is lower in the singing male’s song nuclei RA, LMAN, and LArea X than when he sings undirected song not facing another bird or by himself (Figure 13.3) (14). This differential IEG expression is associated with similar electrophysiological changes in LArea X and LMAN, where the neural firing is less and more distinct during directed singing relative to undirected singing (24). The source of this modulatory effect on both gene expression and electrical activity in LArea X and LMAN may come from outside the song system, in particular dopaminergic modulation from the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc) of the midbrain into Area X and RA (Figure 13.1b) (25–27) and from neuroadrenergic modulation of the locus ceoruleus (LoC) also possibly into Area X (26–28). The VTA-SNc neurons have higher levels of firing with associated higher levels of dopamine release in Area X during directed singing relative to undirected singing (25, 26). Dopamine is a modulatory neurotransmitter that reinforces motivation behaviors such as sexual motivation (31). Within the VTA-SNc, the GABAergic inhibitory interneurons showed increased ZENK expression during directed singing (27), suggesting that the depressed amount of ZENK expression in Area X during directed singing could be modulated by the inhibitory activities of the GABAergic interneurons. However, although lesions of VTA-SNc prevent high levels of singing-induced ZENK expression in Area X, they do not prevent the social context differences (27).

FIGURE 13.3. Social-context-dependent gene expression in the song system of adult male zebra finches.


Social-context-dependent gene expression in the song system of adult male zebra finches. (a) Top panels: Sagittal brain sections (dark field images) showing high singing-driven zenk mRNA expression (white) in all song nuclei during undirected singing (more...)

The song syllables produced during undirected singing are slightly more variable than during directed singing and this variability is controlled by activity from LMAN into RA (17, 32). This variability is associated with vocal exploration, and as such, the increased ZENK protein expression in LMAN and RA could be associated with consolidating newly explored song syllables. Female zebra finches also detect this variability, preferring to perch next to speakers with playbacks of the more stereotyped directed song versus undirected song; this preference is associated with selective increased ZENK protein expression in the CM auditory area of the female brain (33). Not only are auditory neurons sensitive to these two different song types, but they also have differences of ZENK expression depending on whether the listening birds know other birds are present (in the dark); hearing song-induced expression is higher in birds with experience of being housed together than when housed alone (34). Thus, social context regulation of ZENK mRNA and protein expression can occur in both vocal and auditory pathways.

Later studies found that social context regulation occurs for other genes, but with different patterns than those found for ZENK. One such gene is FoxP2, a transcription factor associated with spoken language function in humans and song learning in songbirds (35–37). Starting at high basal levels, FoxP2 mRNA decreases in expression in Area X when zebra finch males sing undirected song, and no change occurs when they sing directed song (Figure 13.3) (38). Further, even in the absence of singing, when juvenile zebra finches are in their developmental, plastic to stereotype phase of song learning or canaries are in their seasonal stereotyped song phase, basal levels of FoxP2 expression are higher in Area X than during other stages (35). These correlations suggest that down-regulation of FoxP2 in Area X may be linked with exploratory singing behavior and song learning. This idea was tested in manipulation experiments using RNAi to knock down FoxP2 protein in Area X of zebra finch juveniles learning to sing. Compared to RNAi controls, juvenile zebra finches with knocked down FoxP2 protein could still sing, but they were unable to properly imitate the song syllables and song sequences of their tutors (Figure 13.3) (39). If FoxP2 has a similar function during adult undirected singing, then it might also be involved in modulating song exploration. FoxP2 is thought to suppress the expression of most of its target genes (40). Thus, when FoxP2 is down-regulated, its target genes are presumably up-regulated. The regulation during undirected singing suggests that those target genes could be involved in synaptic plasticity.

Another gene, synaptotagmin IV, which produces a synapse-associated protein, was found to be regulated in a social-context-dependent manner. During undirected singing, synaptotagmin IV expression is up-regulated in LMAN, HVC, and RA, and down-regulated in Area X (41, 42). During directed singing, there are no changes in synaptotagmin IV expression in any of the song nuclei. Synaptotagmin IV is a presynaptic protein that modulates neurotransmitter vesicle release at nerve terminals, but there is no consensus on whether it suppresses or enhances the vesicle release (41). Synaptotagmin IV null mutant mice show enhanced short-term synaptic plasticity as well as deficits in associative passive avoidance memory (43, 44). Thus, one possible consequence of increased synaptotagmin IV in palliai song nuclei (LMAN, HVC, and RA) during undirected singing would be to subsequently increase the ability to release synaptic vesicles for future singing events and maintain song memories. Another would be to simply replace synaptotagmin IV protein that was used up in the act of singing.

Overall, the social context experiments demonstrate that the song system is highly engaged during undirected singing when the birds are producing more variable song, that each activity-dependent gene tested to date has its own regulatory pattern in different social context, and that there is not a simple link between neural activity and gene regulation in a behaving animal. The findings also indicate that there are other behaviorally regulated genes to be discovered, as discussed in the next section.

13.3.3. An Apparent Cascade of Genes

Since the initial studies with zenk and c-fos, a total of 36 genes regulated by singing have been discovered (Figures 13.4 and 13.5a). The discoveries were based upon either prior knowledge of activity-regulated genes in other organisms (41, 42,45,46), serendipitous findings (38, 47), or by non-biased screenings with cDNA microarrays hybridized to mRNA probes from song nuclei of birds that sang (42, 48). The discovered genes of these and past studies generate known proteins that span a diverse set of functions. These include signal transduction proteins (egr-1, c-fos, FoxP2, c-jun, similar to junB, Atf4, Hspbl, UbE2vl, HnrpH3, Shfdgl, and Madh2), chromosome scaffold proteins (H3f3B and H2AfX), actin-interacting cytoskeletal proteins (Arc, sim Fmnl, Tagln2, ARHGEF9, and ß-actin), a Ca2+-regulating protein (Cacyb), cytoplasmic proteins with enzymatic activity (Prkarla, Atp6vlb2, Ndufa5, and GADPH), protein kinase (Gadd45b), folding (Hsp70–8), binding, and transporting functions (Hsp40, Hsp90a, Hsp25, UCHL1), and membrane (Stard7, Syt4, and Ebag9) and synaptically released proteins (JSC, BDNF, and Penk; Figure 13.5a-c). Like FoxP2, several of these (ARHGEF9 and similar to NPD014) are down-regulated in Area X as a result of singing. This diversity of recruited genes suggests that large signal transduction networks are activated during production of a learned behavior. However, the networks may be song nuclei specific.

FIGURE 13.4. In situ hybridizations of 33 singing-driven genes in song nuclei.


In situ hybridizations of 33 singing-driven genes in song nuclei. Shown are inverse images of autoradiographs; white is mRNA expression. Images are ordered from top to bottom according to four overall expression patterns and from left to right in temporal (more...)

FIGURE 13.5. Summary of singing-driven genes in songbird song nuclei.


Summary of singing-driven genes in songbird song nuclei. (a) Table of 36 singing-regulated genes verified to date. Shown are the inferred cellular locations, molecular functions, and biological processes based on ontology definitions of homologous genes (more...)

The study of Wada et al. (2006) (42) was able to group the singing-regulated genes into four anatomical categories: (a) those regulated in all four major song nuclei; (b) a combination of one or two palliai song nuclei and in the striatal song nucleus Area X; (c) those regulated in one or more palliai song nuclei only (HVC, RA, and LMAN); and (d) those regulated in the striatal nucleus only (Figure 13.4). Of these, Area X and HVC had the highest percentages (94% and 76%, respectively) of genes regulated by singing, and LMAN and RA the lowest (30% and 33%, respectively). This suggests that Area X and HVC are more molecularly dynamic than the other song nuclei, and thus potentially open to greater neural plasticity. It also indicates that in each song nucleus, overlapping or different gene networks are activated during production of song.

These gene networks appear to involve molecules of various functions and cellular locations. The largest proportion (36%) are involved in the nucleus with transcription factors being the largest functional group (18%) (Figures 13.5b and c). The second largest are involved in cytoplasmic (25%) and cytoskelatal (15%) functions, with smaller percentages being membrane and synaptically released proteins. These proportional relationships suggest a trajectory of activation from the nucleus to released proteins at the membrane. If this trajectory were a cascade, though, then a question is why the proportional representation of the cellular categories does not increase from the nucleus to the membrane. One reason could be that the screening for singing-regulated genes focused on early time points (within one hour) after singing (42, 48), where a focus on all time points after singing could reveal a higher proportion of genes involved in extra-nuclear functions. A time course analysis, however, of the currently identified genes does suggest a cascade from the nucleus out (Figures 13.413.6).

FIGURE 13.6. Time course of singing-regulated expression of 33 genes in four song nuclei.


Time course of singing-regulated expression of 33 genes in four song nuclei. The genes are categorized into one of six types of temporal expression patterns by Wada et al. (2006) based on expression peak time (0.5, 1, or 3 hours) and expression profile. (more...)

As shown by Wada et al. (2006) (42), the first genes that show the first peak of expression from the start of singing, at 30 minutes, are mainly transcription factors (ZENK, c-fos, c-jun, sim junB, Atf4) (Figure 13.4). The only structural molecules with peak expression to date at 30 minutes are Arc and a gene similar to formin-like protein, involved in actin filament elongation (Figure 13.5a). Thereafter, expression of these genes either remains high or decreases as singing continues. Many of the cytoplasmic, cytoskeleton, and other non-nuclear genes show peak expression at one hour, again with some remaining high and some decreasing as singing continues. Besides basic cytoplasmic proteins, these include RNA binding (HnrpH3) and protein binding/heat shock proteins (Hsp25, Hsp40, Hsp70–8, Hsp90a) (Figures 134–13.6). By three hours after singing, mainly structural molecules (ß-actin, Stard7, H3F3B) show peak expression.

Some of the non-transcription factor genes have been studied in further detail during singing. A discussion of several of them may help with understanding the functional consequences of behaviorally regulated gene expression. One gene that has received a lot of attention is BDNF, i.e., brain derived neurotrophic factor. BDNF is a synaptically released protein at axon endings, and has a variety of functions including enhancement of neuronal survival. Singing drives increased BDNF expression in the palliai song nuclei (HVC, RA, and LMAN) (42, 45). Within HVC, BDNF shows a preference of increased expression in the RA-projecting neurons, with very little to no induction in the X-projecting neurons. Like ZENK, the amount of BDNF in the RA-projecting neurons correlates with the amount of singing. Intriguingly, the RA-projecting neurons in HVC are renewed throughout the animals’ life, whereas the X-projecting neurons are not renewed and remain stable. In this regard, it was found that more singing leads to enhanced survival of new RA-projecting neurons that enter HVC (45, 49). These findings suggest that singing induces BDNF in the RA-projecting neurons, which in turn enhances the survival of those same or adjacent RA-projecting neurons. The survival role of BDNF may be explained by BDNF acting in local networks with several other IEGs. In the mouse regulatory region of BDNF, the CCAAT enhancer binding protein (C/EBP), AP-1 (composed of a dimer of c-Jun and c-Fos), and ZENK binding sites were identified (50). A study by Calella et al. (51), using microarray analysis, showed that the binding of BDNF to its receptor TrkB initiates the transcription of a number of transcription factor such as c-fos, zenk 1, and zenk 2. Thus, it is possible that BDNF, ZENK, and c-fos may interact in a molecular pathway that completes a circular loop whereby the neurotrophin stimulates transcription of immediate early genes whose protein products regulate the expression of the same neurotrophic factors and other proteins required for neuron maintenance and/or strengthening of neuronal connections during and after the act of singing.

A related but not identical finding was seen with the ubiquitin carboxy-terminal hydrolase 1 (UCHL1) gene (47). UCHL1 is part of a ubiquitin-proteasome system necessary for protein degradation, and has been implicated in the death of neurons in neurodegenerative diseases, such as Parkinson’s, Alzheimer’s, and Huntington’s. With RNA from laser captured microdissected neurons of HVC hybridized to cDNA microarrays, Lombardino et al. (47, 48) found that UCHL1 was expressed at high levels in the X-projecting neurons and at low levels in the RA-projecting neurons. When the birds sang, UCHL1 levels increased in the RA-projecting neurons, but there was no change in the already high levels of the X-projecting neurons. The singing-induced expression in the RA-projecting neurons was still lower than the expression in the X-projecting neurons. A mouse mutant for UCHL1 shows axonal loss and cell death (52). Thus, it has been suggested that naturally low UCHL1 levels in replaceable neurons allow for these neurons to undergo natural cell death, and that the singing up-regulation in these neurons extends their life (47).

The above two examples are of genes that function in neuron survival and death. As such, the turnover of the RA-projecting neurons has been proposed to be involved in memory functions (53). Another type of change one would expect to be involved in memory formation is whether expression of any of the singing-regulated genes differs when birds are producing well learned song versus when they are practicing singing in the early juvenile stages of life. Such a difference has been found for proenkephalin (Penk). Penk shows higher singing-induced expression in HVC when juveniles sing plastic song than in adults singing stable stereotyped song (42). Proenkephalin is a precursor protein that contains several copies of enkephalins. Enkephalins are pentapeptides, released at synapses, and are best known for their ability to compete with and mimic the effect of opiate drugs. They are also involved in memory and emotional functions (54–56). How high singing-regulated enkephalin expression in song nuclei during juvenile development functions in the development of song is a question that remains to be answered. Most other singing-regulated genes that have been tested differ little or at all during singing in juveniles versus adults (14, 19,42,57). ZENK is higher in song nuclei when juveniles sing, but basal levels are already higher.

Future experiments will be necessary to determine the functional, cellular, and behavioral consequences of individual genes regulated during production of a learned behavior, or during the actual learning of the behavior. The experimental designs will have to consider how faithful the behaviorally regulated protein changes reflect the mRNA changes, the subject of the next section (14, 57).


Most genes show concordant singing-regulated mRNA and protein expression, in which protein and mRNA have been examined. Singing-induced expression of mRNA and protein has been shown for the transcription factor ZENK in all forebrain song nuclei, with the protein localized in the nucleus (14, 19,58) as expected for a transcription factor. The peak of protein expression lags that of mRNA expression by 30 minutes (3). Not only the mRNA, but ZENK protein expression in Area X, LMAN, and RA depends on the social context in which singing occurs, with increased expression in birds that produce undirected song and less of an increase when they sing directed song to females (14). However, in the song nucleus RA, translation of zenk mRNA and into ZENK protein depends on whether the birds sing in the presence of other birds and on hearing (59). In the absence of other birds, zenk mRNA synthesis is induced in RA, but only a proportion of the mRNA is translated into protein. When other birds are present, then singing-driven zenk mRNA is faithfully translated into ZENK protein (Figure 13.7a and b). This effect of translation into protein requires that the singing male hear the others birds while singing, because when he is deaf the lower translation levels are found (57). This finding indicates that in RA, another layer of regulation, potentially on the translational level from mRNA to protein, is at work. This regulation could involve microRNAs or other repression mechanisms of mRNA translation by the binding of a repressor to a specific sequence in the 5′ end of the mRNA. An alternative mechanism would be rapid ZENK post-translational protein degradation, such as through the ubiquitin pathway, when birds sing alone. The effect is not observed in HVC where relative amounts of zenk mRNA matched the protein levels (Figure 13.7b).

FIGURE 13.7. In RA singing drives zenk mRNA, while ZENK protein varies as a function of social context.


In RA singing drives zenk mRNA, while ZENK protein varies as a function of social context. (a) (Left) ZENK protein-labeled cells are shown at low magnification in RA for birds that sang alone (solo) or in the presence of another singing male (duo), for (more...)

Comparable spatial and temporal regulation of singing-induced mRNA and protein was confirmed for several other genes by in situ hybridization and immu-nocytochemistry: c-fos, BDNF, c-jun, and Penk (42). For c-fos and BDNF, protein expression changes follow the mRNA expression changes, although not all song nuclei have been tested (23, 42,45). It is also difficult to assess BDNF protein expression, as it is synaptically released and difficult to detect. The transcription factor c-jun had the expected nuclear localization within neuronal cells and enkephalin, a protein product of proenkephalin, was localized to neuronal processes (42). However, for ß-actin, no changes in protein expression were detected in song nuclei one to two hours after singing. It is possible that the increase in the actin mRNA is compensated by a decrease in the protein expression, or that an increase in protein expression occurs at a later time point. In general, five of six genes tested show singing upregulation at both the mRNA and protein levels (39). However, the mRNA and protein changes do not faithfully follow each other in all instances, indicating that it is necessary to study regulation at both levels of expression.


There has been some overlap of genes regulated by hearing in the auditory pathway and singing in the vocal pathway. These include ZENK, ARC, c-Fos, and c-Jun. The activity-dependent regulation of these genes was discovered first either by hearing or singing, and then checked for the other condition. At first it was thought that c-Fos protein was not regulated in NCM by hearing song (23), but this negative result has since not been confirmed, as hearing song can induce c-Fos in NCM (46). The discrepancy between the studies could be due to an apparent higher threshold of activity needed to induce c-Fos, where the threshold may have been reached in most but not all studies (15). When using unbiased gene screening approaches, there has been little overlap between the genes discovered that are regulated by hearing in the auditory pathway and singing in the vocal pathway. The reasons for this non-overlap may be due to differences in approaches used: microarray analysis for the vocal pathway (42) and proteomics for the auditory pathway (60) (also see Pinaud et al., Chapter 14). It may also be due to real differences in expression of proteins as seen in the expression of 19 glutamate receptor subtypes in the zebra finch brain (61). Glutamate receptors, upon binding the neurotransmitter glutamate, regulate the expression of activity-dependent genes such as ZENK, as determined in the mammalian brain (62). Of the 21 glutamate receptor subtypes studied in songbirds, 19 showed differential expression in song nuclei relative to the respective surrounding brain subdivisions, whereas the auditory pathways had expression with few differences relative to the surrounding areas (60). In this regard, it is possible that the differential expression of neurotransmitter receptors may result in different, but overlapping sets of genes regulated by singing in song nuclei and hearing in the auditory nuclei. A resolution to this difference may be resolved in future studies that test both the vocal and auditory pathway using the same methods, and that test singing-regulated genes discovered in the vocal pathway for their potential regulation in the auditory pathway and vice versa. Such an initial comparison was recently made while this chaper was in press, where more overlap was in the hearing and singing related genes (63).


It has been proposed that activity-dependent regulatory cascades are the basic contributory mechanisms for long-term memory formation in the nervous system of a wide range of organisms from sea mollusks to mammals (64–66). The exact causal relationship between neuronal activation and gene expression response is unknown. One hypothesis (46, 67) has been that presynaptic neurotransmitter release leads to postsynaptic neurotransmitter receptor activation, which in turn induces calcium-dependent entry through voltage-dependent channels (Figure 13.8) (68). This will then lead to the activation of calcium-dependent signal transduction pathways (e.g., kinases) and activation of temporally inactive transcription factor proteins such as CREB or C/EBP (20). CREB, C/EBP, and other factors, some already bound to the promoter regions, then initiate the transcription of IEG mRNAs such as ZENK, c-Fos, c-Jun, and ARC. The extent to which electrophysiological and gene activation are coupled in neurons depends on the location and type of neuron affected (69, 70). This coupling of action potentials to gene activation has been referred to as the genomic potential (71). In this regard, it has been argued that the rate of neuronal firing of the presynaptic neuron controls the amount of behaviorally regulated gene expression in the postsynaptic neuron (67). In support of a presynaptic-postsynaptic relationship, lesion of a presynaptic song nucleus either prevents or disrupts the modulation of singing-regulated ZENK expression in its downstream postsynaptic targets (12).

FIGURE 13.8. Hypothesis of synaptic mechanism for immediate early gene induction by neural activity associated with behavior.


Hypothesis of synaptic mechanism for immediate early gene induction by neural activity associated with behavior. Shown is a schematic representation of the cascade (arrows) triggered upon synaptically activated presynaptic neurotransmitter release onto (more...)

Another open question is what is the role of the singing-regulated mRNA and protein expression in cellular physiology and behavior? Two possibilities include neural plasticity (i.e., learning) and neural homeostasis (i.e., behavior maintenance). The genome as it concerns the neuron must express genes that are necessary for the homeostasis of the cell. Homeostasis presumably would include genes that have singing-regulated expression levels in the song brain nuclei at relatively the same levels in adult and juvenile animals. Examples are NADH dehydrogenase, glycer-aldehyde-3-phosphate dehydrogenase, ß-actin, and actin-associated molecules (42). These gene products may need to be replenished on a regular basis as they get used up or damaged during the active process of the genomic potential. Consistent with this idea is the fact that a number of the singing-regulated genes are heat shock proteins (Figure 13.5a). As such, one can imagine that neural activity causes some cellular fatigue. The heat shock proteins, known to be involved in neuroprotection (72), could protect the cells from further fatigue and death. In contrast, genes involved in plasticity may be expected to be regulated differentially in juveniles and adults. Examples are Penk, zenk, and foxp2. All are expressed at higher levels in specific juvenile song nuclei relative to adults (14, 35,42). Thus, in juveniles actively learning how to sing, the neuroprotective mechanisms may be at work, overlaid with synaptic plasticity mechanisms to form new motor memories of the songs the birds imitate. Future experiments are necessary to test these hypotheses, using gene manipulation tools and assessing the cellular and behavioral consequences.


In future experiments, it will be necessary to link the behaviorally regulated genes into gene regulatory networks. This can be accomplished with a combination of gene manipulation and computational studies. With computational analyses, these molecules can be placed into inferred gene and/or protein networks to gain insight into how these interactions affect hearing and singing behavior (73, 74). For example, one can generate protein-DNA, protein-protein, and other interaction networks of the singing-regulated genes using an interaction algorithm that utilizes known molecular interactions in the literature (Ingenuity System, Path Designer Network software, 2000–2008). We applied the Ingenuity algorithm to the 36 singing-regulated genes and found that the highest scoring network was a neural molecular network that included 18 of the genes, with the immediate early gene transcription factors (JUN, JUNB, FOS, and EGR-1 [aka ZENK]) as central hubs with direct or indirect interactions to the other genes (Figure 13.9). In this network, for example, the binding of JUN to proenkephalin (Penk) DNA (PD [1], direct protein-DNA interaction Figure 13.9) is consistent with the temporal (c-jun on first) and anatomical regulation found in singing animals. Although this is a small-scale analysis, it serves as a testable hypotheses of how the cascade of singing-regulated genes may interact with each other and other genes. Once inferred, these networks can to be used to perform gene manipulation experiments in live animals via either RNAi or over-expression to either inhibit or enhance a gene product in the song nuclei of birds and observe a result on cellular physiology and singing behavior. These challenges are within the grasp of the foreseeable future and promise to unlock the mystery of behaviorally regulated gene expression.

FIGURE 13.9. Hypothesized interaction pathway of the singing-regulated genes (Figure 13.


Hypothesized interaction pathway of the singing-regulated genes (Figure 13.5), according to pathway analyses interactions of the Ingenuity Systems software (Sept. 2008 version). Filled objects (nodes) are the gene/protein-products-regulated singing. Open (more...)


Jarvis E.D., Gunturkun O., Bruce L., et al. Avian brains and a new understanding of vertebrate brain evolution. Nat Rev Neurosci. 2005;6:151–9. [PMC free article: PMC2507884] [PubMed: 15685220]
Vates G.E., Broome B.M., Mello C. V., Nottebohm F. Auditory pathways of caudal telencephalon and their relation to the song system of adult male zebra finches. J Comp Neurol. 1996;366:613–42. [PubMed: 8833113]
Mello C.V., Vates G.E., Okuhata S., Nottebohm F. Descending auditory pathways in the adult male zebra finch (Taeniopygia guttata) J Comp Neurol. 1998;395:137–60. [PubMed: 9603369]
Gentner T. Q., Margoliash D. Neuronal populations and single cells representing learned auditory objects. Nature. 2003;424:669–74. [PMC free article: PMC2631575] [PubMed: 12904792]
Theunissen F.E., Amin N., Shaevitz S.S., et al. Song selectivity in the song system and in the auditory forebrain. Ann N Y Acad Sci. 2004;1016:222–45. [PubMed: 15313778]
Coleman M.J., Roy A., Wild J. M., Mooney R. Thalamic gating of auditory responses in telencephalic song control nuclei. J Neurosci. 2007;27:10024–36. [PubMed: 17855617]
Nottebohm F., Stokes T. M., Leonard C.M. Central control of song in the canary, Serinus canarius. J Comp Neurol. 1976;165:457–86. [PubMed: 1262540]
Simpson H. B., Vicario D.S. Brain pathways for learned and unlearned vocalizations differ in zebra finches. J Neurosci. 1990;10:1541–56. [PubMed: 2332796]
Aronov D., Andalman A. S., Fee M.S. A specialized forebrain circuit for vocal babbling in the juvenile songbird. Science. 2008;320:630–4. [PubMed: 18451295]
Vates G. E., Nottebohm F. Feedback circuitry within a song–learning pathway. Proc Natl Acad Sci USA. 1995;92:5139–43. [PMC free article: PMC41864] [PubMed: 7761463]
Luo M., Ding L., Perkel D.J. An avian basal ganglia pathway essential for vocal learning forms a closed topographic loop. J Neurosci. 2001;21:6836–45. [PubMed: 11517271]
Kubikova L., Turner E. A., Jarvis E.D. The pallial basal ganglia pathway modulates the behaviorally driven gene expression of the motor pathway. Eur J Neurosci. 2007;25:2145–60. [PMC free article: PMC2561259] [PubMed: 17419760]
Durand S.E., Heaton J.T., Amateau S. K., Brauth S.E. Vocal control pathways through the anterior forebrain of a parrot (Melopsittacus undulatus) J Comp Neurol. 1997;377:179–206. [PubMed: 8986880]
Jarvis E.D., Scharff C., Grossman M.R., Ramos J. A., Nottebohm F. For whom the bird sings: context–dependent gene expression. Neuron. 1998;21:775–88. [PubMed: 9808464]
Feenders G., Liedvogel M., Rivas M., et al. Molecular mapping of movement– associated areas in the avian brain: a motor theory for vocal learning origin. PLoS ONE. 2008;3:e1768. [PMC free article: PMC2258151] [PubMed: 18335043]
Scharff C., Nottebohm F. A comparative study of the behavioral deficits following lesions of various parts of the zebra finch song system: implications for vocal learning. J Neurosci. 1991;11:2896–913. [PubMed: 1880555]
Kao M.H., Doupe A. J., Brainard M.S. Contributions of an avian basal ganglia–forebrain circuit to real-time modulation of song. Nature. 2005;433:638–43. [PubMed: 15703748]
Bauer E.E., Coleman M.J., Roberts T.F., et al. A synaptic basis for auditory– vocal integration in the songbird. J Neurosci, 2008;28:1509–22. [PubMed: 18256272]
Jarvis E. D., Nottebohm F. Motor-driven gene expression. Proc Natl Acad Sci USA. 1997;94:4097–102. [PMC free article: PMC20574] [PubMed: 9108111]
Mello C.V., Vicario D. S., Clayton D.F. Song presentation induces gene expression in the songbird forebrain. Proc Natl Acad Sci USA. 1992;89:6818–22. [PMC free article: PMC49595] [PubMed: 1495970]
Sheng M., McFadden G., Greenberg M.E. Membrane depolarization and calcium induce c–fos transcription via phosphorylation of transcription factor CREB. Neuron. 1990;4:571–82. [PubMed: 2157471]
Mello C. V., Ribeiro S. ZENK protein regulation by song in the brain of songbirds. J Comp Neurol. 1998;393:426–38. [PubMed: 9550149]
Kimpo R. R., Doupe A.J. FOS is induced by singing in distinct neuronal populations in a motor network. Neuron. 1997;18:315–25. [PubMed: 9052801]
Hessler N. A., Doupe A.J. Social context modulates singing-related neural activity in the songbird forebrain. Nat Neurosci. 1999;2:209–11. [PubMed: 10195211]
Sasaki A., Sotnikova T.D., Gainetdinov R. R., Jarvis E.D. Social context– dependent singing–regulated dopamine. J Neurosci. 2006;26:9010–4. [PMC free article: PMC2474783] [PubMed: 16943558]
Yanagihara S., Hessler N.A. Modulation of singing-related activity in the songbird ventral tegmental area by social context. Eur J Neurosci. 2006;24:3619–27. [PubMed: 17229110]
Hara E., Kubikova L., Hessler N. A., Jarvis E.D. Role of the midbrain dopaminergic system in modulation of vocal brain activation by social context. Eur J Neurosci. 2007;25:3406–16. [PMC free article: PMC2561249] [PubMed: 17553009]
Castelino C. B., Ball G.F. A role for norepinephrine in the regulation of context–dependent ZENK expression in male zebra finches (Taeniopygia guttata) Eur J Neurosci. 2005;21:1962–72. [PubMed: 15869489]
Castelino C.B., Diekamp B., Ball G.F. Noradrenergic projections to the song control nucleus area X of the medial striatum in male zebra finches (Taeniopygia guttata) J Comp Neurol. 2007;502:544–62. [PubMed: 17394158]
Cornil C.A., Castelino C. B., Ball G.F. Dopamine binds to alpha(2)-adrenergic receptors in the song control system of zebra finches (Taeniopygia guttata) J Chem Neuroanat. 2008;35:202–15. [PMC free article: PMC2329819] [PubMed: 18155403]
Young L. J., Wang Z. The neurobiology of pair bonding. Nat Neurosci. 2004;7:1048–54.. [PubMed: 15452576]
Olveczky B.P., Andalman A. S., Fee M.S. Vocal experimentation in the juvenile songbird requires a basal ganglia circuit. PLoS Biol. 2005;3:e153. [PMC free article: PMC1069649] [PubMed: 15826219]
Woolley S. C., Doupe A.J. Social context-induced song variation affects female behavior and gene expression. PLoS Biol. 2008;6:e62. [PMC free article: PMC2267820] [PubMed: 18351801]
Vignal C., Andru J., Mathevon N. Social context modulates behavioural and brain immediate early gene responses to sound in male songbird. Eur J Neurosci. 2005;22:949–55. [PubMed: 16115218]
Haesler S., Wada K., Nshdejan A., et al. FoxP2 expression in avian vocal learners and non-learners. J Neurosci. 2004;24:3164–75. [PubMed: 15056696]
Marcus G. F., Fisher S.E. FOXP2 in focus: what can genes tell us about speech and language? Trends Cogn Sci. 2003;7:257–262. [PubMed: 12804692]
Teramitsu I., Kudo L.C., London S.E., Geschwind D. H., White S.A. Parallel FoxP1 and FoxP2 expression in songbird and human brain predicts functional interaction. J Neurosci. 2004;24:3152–63. [PubMed: 15056695]
Teramitsu I., White S.A. FoxP2 regulation during undirected singing in adult songbirds. J Neurosci. 2006;26:7390–4. [PMC free article: PMC2683919] [PubMed: 16837586]
Haesler S., Rochefort C., Georgi B., et al. Incomplete and inaccurate vocal imitation after knockdown of FoxP2 in songbird basal ganglia nucleus Area X. PLoS Biol. 2007;5:e321. [PMC free article: PMC2100148] [PubMed: 18052609]
Spiteri E., Konopka G., Coppola G., et al. Identification of the transcriptional targets of FOXP2, a gene linked to speech and language, in developing human brain. Am J Hum Genet. 2007;81:1144–57. [PMC free article: PMC2276350] [PubMed: 17999357]
Poopatanapong A., Teramitsu I., Byun J.S., et al. Singing, but not seizure, induces synaptotagmin IV in zebra finch song circuit nuclei. J Neurobiol. 2006;66:1613–29. [PMC free article: PMC2694668] [PubMed: 17058190]
Wada K., Howard J.T., McConnell P., et al. A molecular neuroethological approach for identifying and characterizing a cascade of behaviorally regulated genes. Proc Natl Acad Sci USA. 2006;103:15212–7. [PMC free article: PMC1622802] [PubMed: 17018643]
Ferguson G.D., Anagnostaras S.G., Silva A. J., Herschman H.R. Deficits in memory and motor performance in synaptotagmin IV mutant mice. Proc Natl Acad Sci USA. 2000;97:5598–603. [PMC free article: PMC25874] [PubMed: 10792055]
Ferguson G.D., Wang H., Herschman H. R., Storm D.R. Altered hippocampal short-term plasticity and associative memory in synaptotagmin IV (–/–) mice. Hippocampus. 2004;14:964–74. [PubMed: 15390175]
Li X.C., Jarvis E.D., Alvarez–Borda B., Lim D. A., Nottebohm F. A relationship between behavior, neurotrophin expression, and new neuron survival. Proc Natl Acad Sci USA. 2000;97:8584–9. [PMC free article: PMC26991] [PubMed: 10890902]
Velho T.A., Pinaud R., Rodrigues P. V., Mello C.V. Co-induction of activitydependent genes in songbirds. Eur J Neurosci. 2005;22:1667–78. [PubMed: 16197507]
Lombardino A.J., Li X.C., Hertel M., Nottebohm F. Replaceable neurons and neurodegenerative disease share depressed UCHL1 levels. Proc Natl Acad Sci USA. 2005;102:8036–41. [PMC free article: PMC1142397] [PubMed: 15911766]
Lombardino A.J., Hertel M., Li X.C., et al. Expression profiling of intermingled long-range projection neurons harvested by laser capture microdissection. J Neurosci Methods. 2006;157:195–207. [PubMed: 16750569]
Alvarez-Borda B., Nottebohm F. Gonads and singing play separate, additive roles in new neuron recruitment in adult canary brain. J Neurosci. 2002;22:8684–90. [PubMed: 12351743]
Hayes V.Y., Towner M. D., Isackson P.J. Organization, sequence and functional analysis of a mouse BDNF promoter. Brain Res Mol Brain Res. 1997;45:189–98. [PubMed: 9149093]
alella A.M., Nerlov C., Lopez R.G., et al. Neurotrophin/Trk receptor signaling mediates C/EBPalpha, -beta and NeuroD recruitment to immediate-early gene promoters in neuronal cells and requires C/EBPs to induce immediate-early gene transcription. Neural Develop. 2007;2:4.. [PMC free article: PMC1796876] [PubMed: 17254333]
Mukoyama M., Yamazaki K., Kikuchi T., Tomita T. Neuropathology of gracile axonal dystrophy (GAD) mouse. An animal model of central distal axonopathy in primary sensory neurons. Acta Neuropathol. 1989;79:294–9. [PubMed: 2558488]
Nottebohm F. The road we travelled: discovery, choreography, and significance of brain replaceable neurons. Ann N Y Acad Sci. 2004;1016:628–58. [PubMed: 15313798]
Dudas B., Merchenthaler I. Topography and associations of leu-enkephalin and luteinizing hormone-releasing hormone neuronal systems in the human diencephalon. J Clin Endocrinol Metab. 2003;88:1842–8. [PubMed: 12679482]
Owczarek D., Garlicka M., Pierzchala-Koziec K., Skulina D., Szulewski P. [Met-enkephalin plasma concentration and content in liver tissue in patients with primary biliary cirrhosis] Przegl Lek. 2003;60:461–6. [PubMed: 14750419]
Zhang M., Balmadrid C., Kelley A.E. Nucleus accumbens opioid, GABaergic, and dopaminergic modulation of palatable food motivation: contrasting effects revealed by a progressive ratio study in the rat. Behav Neurosci. 2003;117:202–11. [PubMed: 12708516]
Jin H., Clayton D.F. Localized changes in immediate-early gene regulation during sensory and motor learning in zebra finches. Neuron. 1997;19:1049–59. [PubMed: 9390518]
Ribeiro S., Cecchi G.A., Magnasco M. O., Mello C.V. Toward a song code: evidence for a syllabic representation in the canary brain. Neuron. 1998;21:359–71. [PubMed: 9728917]
Whitney O., Johnson F. Motor-induced transcription but sensory-regulated translation of ZENK in socially interactive songbirds. J Neurobiol. 2005;65:251–259. [PMC free article: PMC4264577] [PubMed: 16155900]
Pinaud R., Osorio C., Alzate O., Jarvis E.D. Profiling of experience-regulated proteins in the songbird auditory forebrain using quantitative proteomics. Eur J Neurosci. 2008;27:1409–22. [PMC free article: PMC2517219] [PubMed: 18364021]
Wada K., Sakaguchi H., Jarvis E. D., Hagiwara M. Differential expression of glutamate receptors in avian neural pathways for learned vocalization. J Comp Neurol. 2004;476:44–64. [PMC free article: PMC2517240] [PubMed: 15236466]
Lerea L. Glutamate receptors and gene induction: signalling from receptor to nucleus. Cell Signal. 1997;9:219–226. [PubMed: 9218121]
Dong S., Replogle K.L., Hasadsri L., et al. Discrete molecular states in the brain accompany changing responses to a vocal signal. Proc Natl Acad Sci USA. 2009;106:11364–9. [PMC free article: PMC2708759] [PubMed: 19541599]
Goelet P., Castellucci V.F., Schacher S., Kandel E.R. The long and the short of long-term memory––a molecular framework. Nature. 1986;322:419–22. [PubMed: 2874497]
Robertson H.A. Immediate-early genes, neuronal plasticity, and memory. Biochem Cell Biol. 1992;70:729–37. [PubMed: 1482550]
Knapska E., Kaczmarek L. A gene for neuronal plasticity in the mammalian brain: Zif268/Egr-1/NGFI-A/Krox-24/TIS8/ZENK? Prog Neurobiol. 2004;74:183–211. [PubMed: 15556287]
Jarvis E.D. Brains and birdsong. In: Marler P., Slabbekoorn H., editors. Nature’s Music: The Science of Bird Song. San Diego: Elsevier–Academic Press; 2004. pp. 226–271.
Mello C. V., Jarvis E.D. Behavior-dependent expression of inducible genes in vocal learning birds. In: Zeigler H. P., Marler P., editors. Neuroscience of birdsong. Cambridge University Press; 2008. pp. 381–397.
Mello C. V., Clayton D.F. Differential induction of the ZENK gene in the avian forebrain and song control circuit after metrazole-induced depolarization. J Neurobiol. 1995;26:145–61. [PubMed: 7536234]
Jarvis E. D., Mello C.V. Molecular mapping of brain areas involved in parrot vocal communication. J Comp Neurol. 2000;419:1–31. [PMC free article: PMC2538445] [PubMed: 10717637]
Clayton D.F. The genomic action potential. Neurobiology of Learning & Memory. 2000;74:185–216. [PubMed: 11031127]
Yenari M.A. Heat shock proteins and neuroprotection. Adv Exp Med Biol. 2002;513:281–99. [PubMed: 12575825]
Smith V.A., Jarvis E. D., Hartemink A.J. Evaluating functional network inference using simulations of complex biological systems. Bioinformatics. 2002;18 Suppl 1:S216–24. [PubMed: 12169550]
Yu J., Smith V.A., Wang P.P., Hartemink A. J., Jarvis E.D. Advances to Bayesian network inference for generating causal networks from observational biological data. Bioinformatics. 2004;20:3594–603. [PubMed: 15284094]
Copyright © 2010 by Taylor and Francis Group, LLC.
Bookshelf ID: NBK56007PMID: 21882440


  • PubReader
  • Print View
  • Cite this Page

Other titles in this collection

Related information

  • PMC
    PubMed Central citations
  • PubMed
    Links to PubMed

Similar articles in PubMed

See reviews...See all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...