UW News

November 20, 1996

Twenty years of bird songs: It’s a record filled with scientific high notes

WASHINGTON, D.C. — While punk, disco and acid rock have given way to new country, rap and grunge over the last two decades, neuroscientists have been making beautiful science studying the melodies produced by some of nature’s sweetest voices — songbirds.

A panel of five researchers today marked the 20th anniversary of two landmark papers by Rockefeller University researcher Fernando Nottebohm and his colleagues that launched the modern era of neuroscience research into the study of bird songs, as a model of the human brain and language.

The panel at the annual meeting of the Society for Neuroscience was organized by Eliot Brenowitz, a University of Washington professor of psychology and zoology. It focused on the achievements of two decades of avian brain research, new findings and future directions for investigation.

“The avian song system is one of the great models for studying biological learning. With it we have one of the best natural models to learn from, instead of, for example, studying rats as they find their way through a laboratory maze,” said Brenowitz.

“Because juvenile birds must learn their songs by listening to adults of their species, they provide a superb animal model for studies of vocal development in humans,” he explained. “Also, young birds go through an initial ‘babbling’ stage when they first begin to sing, in a manner very similar to human infants when they start speaking. In addition, young birds must be able to hear themselves while they are learning to sing. Deafened juvenile birds never develop normal songs. This is very similar to what is seen in deaf infants.”

Other members of the panel were Daniel Margoliash, a University of Chicago neurophysiologist; Barney Schlinger, a UCLA biochemist and endocrinologist; Kathy Nordeen, a University of Rochester developmental neuroscientist and David Clayton, a University of Illinois neurobiologist.

Brenowitz said that among the key findings made by Nottebohm and other avian researchers, which set the stage for today’s reports are:

? There are pronounced sex differences in the size of brain nuclei that control song in birds where only males sing. The discovery that male songbirds have larger brain nuclei was the first example found of such striking gender differences in brain structure. Similar examples have since been observed in many vertebrate animals, including humans. Steroid hormones, such as testosterone and estrogen, are critical for the development of gender differences in these areas of the bird brain.

? Special regions of the bird brain are necessary to learn song. Lesions of these regions in juvenile birds prevent them from learning songs, but have no effect when made in adult bird species that only learn songs as juveniles. However, these same lesions disrupt seasonal song learning in species that can learn new songs each year.

? There are dramatic seasonal changes in the size of brain nuclei that control song. These size changes parallel seasonal changes in song behavior.

? In adult song birds, brain cells or neurons continue to be born and be incorporated into brain pathways important for song behavior. It has been estimated that in one song-control nucleus, up to 1 percent of the neurons are born each day. In this regard, birds are very different from humans, in whom neurons in most parts of the brain are not generated after birth.

? The number of versions of a song that an individual bird sings is closely related to the size of the song-control nuclei in the brain. Birds with bigger song nuclei sing more songs. However, the size of the song nuclei isn’t the only factor determining the number of songs a bird sings. Hormonal or other as yet unidentified factors seem to determine the size of the brain nuclei, which then determines how many songs a young bird can learn.

? Exposing a bird to a playback of the songs of his or her own species activates a class of genes within the brain. Playing the songs of other species results in little or no activation of these genes. Activation appears to be important in the formation of song-related memories.

The other panelists reported on their recent work and findings that included:

Snapshots of neuronal activity — Clayton’s laboratory has found five genes that are turned on and off in the brain circuit to control singing in adolescent zebra finches (Taenopygia guttata) as song learning progresses in young birds.

Three of the genes are most active early in the period critical for song memorization. Clayton said these genes promote plasticity or flexibility at the synaptic level. Synapses are the junctions between brain cells where information is stored. Plasticity is the ability of a synapse to undergo the changes necessary to store information.

The other two genes found by Clayton are turned on later in the song-control circuit at the time when a song becomes fixed or crystallized in a bird’s memory. These genes encode proteins to stabilize existing pathways for neural communications.

“We are learning how neurons orchestrate and modulate the storage of new information and the retention of old information,” he explained. “Our results show that a neuron’s cell nucleus and the genes that are stored there, have a much more active role in neural plasticity and learning than may have been imagined. We are uncovering evidence of a major role for several specific genes in guiding or supporting information storage in the brain. These genes are likely to have important functions in learning and memory in humans as well as songbirds.”

The stuff of memories — Nordeen’s research has shown that a chemical neurotransmitter called n-methyl-d-aspartate (NMDA) appears to play a critical role in forging the long-term changes in brain- cell communication that help form memory.

Brain cells communicate with each other by secreting neurotransmitters that plug into receptors on other cells. She said NMDA receptors normally are activated when a young male zebra finch is learning song from an older or tutor bird. But when birds in her laboratory that were being tutored were given an agent that blocks NMDA receptors in and around a brain area called the lNAM, the young birds’ ability to learn song was seriously impaired. They were only able to correctly repeat smaller fragments of song.

“The NMDA receptor is found in the nervous system of creatures throughout the animal kingdom, including humans,” said Nordeen.” This means studies with songbirds can open the door to fundamental understanding of how the brain processes experiences into learning and memories.”

She also reported on data showing that the number of new neurons added during song learning to another brain region, the HVC, related to how much of a tutor’s song can be accurately reproduced. By adulthood, the number of these brain cells varies greatly among birds, she said. Among a group of finches raised together with the same pair of tutors, birds with the larger number of HVC neurons learned more song material than birds with fewer neurons.

Recording the responses of single brain cells: Margoliash’s lab is developing insight into how the birdsong system is organized. He’s doing this by recording the activity of individual neurons in two brain areas, the HVC and RA, when the birds sing and when they hear a recorded song. This work is unique because it was done with awake, freely moving zebra finches, rather than with anesthetized birds.

Margoliash reported two major findings. First, when a bird sings, neurons in the HVC have a signature or activity pattern which is unique to individual syllables or groups of notes in a song. Neurons in the RA have activity patterns which are unique for even smaller bits of a song. Thus, he believes, there is a hierarchical organization of the song motor program.

His second discovery resulted from experiments when a bird hears recorded song. In awake birds, neurons in the HVC exhibited highly selective responses that preferred the bird’s own song over others.  This has been previously shown with anesthetized birds. In the RA, however, neurons in awake birds exhibited little or no auditory responses.  Prior data showing RA auditory responses in anesthetized birds have resulted in theories of song perception that may be false, according to Margoliash.


            Responding to steroids: Songbirds have emerged as an important animal model for studying how steroid hormones influence brain development and Schlinger reported that an enzyme called aromatase converts male hormones to female hormones in several  brain regions of two bird species.


            This enzyme is produced at such high levels in zebra finches and brown-headed cowbirds (Molothrus ater) that the estrogen content of blood increases by passing through the brains of  these birds, Schlinger said.  Aromatase was produced at the highest levels in two brain regions that are thought to process auditory information and relay information to the song system. 


            In addition, these brain regions demonstrated specific electrophysiological  and neurochemical reactions to distinct types of song, Schlinger said. This leads him to believe  that estrogen formed locally  may play a role in song recognition and/or song learning.


###


For more information contact:


Brenowitz  at (206) 543-8534 or {eliotb@u.washington.edu}.


Clayton at (217) 244-4525  or {dclayton@uiuc.edu}.


Margoliash at (312) 702-8090 or {dan@fred.uchicago.edu}.


Nordeen at (716) 275-8452 or {knordeen@bcs.rochester.edu}.


Schlinger at (310) 825-5716 or {schlinge@mail0.lifesci.ucla.edu}.