Neurogenetics of Memory in Drosophila

views updated

Neurogenetics of Memory in Drosophila

Neurogenetic analysis of memory seeks to identify genes involved in behavioral plasticity, to characterize the cells and neuroanatomies in which they are expressed, and to define the biochemistries and cell biologies in which they participate. From the study of Drosophila (fruit flies) beginning in the 1970s, two fundamental notions became apparent. First, a complex circuitry, consisting of sensory inputs, central processing, and motor outputs from thousands of neurons, likely is required. In addition, hundreds to thousands of genes likely participate in the neuronal-synaptic plasticity underlying behavioral plasticity. Hence, a vertical integration of function, from gene to behavior, will entail five primary levels of experimental analysis: molecular-genetic, gene network computation, neurophysiological, neural network computation, and behavioral. Second, molecular biological work across the animal kingdom has revealed a remarkable degree of evolutionary conservation, from basic phenomenology of cell biology during development to behavioral biology of sleep and senescence. Thus, a horizontal integration of gene action from lower organisms to humans appears the rule rather than the exception. Given the economy of scale (short generation time, inexpensive rearing costs), extant experimental tools, and a knowledge base of Drosophila genetics that goes back to the nineteenth century, this model system will continue to yield valuable information on the neurobiological basis of plasticity. Moreover, functional insights gained from flies will inform the mammalian condition directly, thereby allowing discovery of genetic and pharmacological therapies for various forms of cognitive dysfunction in humans.

Behavioral Plasticity

In the final quarter of the twentieth century, a plethora of behavioral learning tasks were developed for Drosophila (Connolly and Tully, 1998). Nonassociative tasks exist for the landing response, cleaning reflex, proboscis extension response, and odor avoidance response. Associative tasks include conditioned task aversion, courtship conditioning, operant conditioning to visual cues in a flight simulator, and olfactory discriminative conditioning. The last task, in particular, has proved valuable. Because initial conditioned avoidance levels are robust and memory retention can last more than a week, this learning task has been used extensively to identify and characterize single-gene mutations (see below). Careful analyses of the learning/memory defects in these mutants has suggested that olfactory memory formation occurs in five genetically distinct phases: acquisition or learning (LRN), short-term memory (STM), middle-term memory (MTM), anesthesia-resistant memory (ARM), and long-term memory (LTM) (Tully et al., 1996; Tully, Preat, Boynton, and Del Vecchio, 1994). Processing through these memory phases seems to occur sequentially from LRN to STM to MTM, at which point memory is processed independently (in parallel) into ARM and LTM. These temporal stages of memory processing likely reflect a combination of neural activity in different anatomic sites and biochemical activity within each.


Three mutageneses have been conducted to screen for behavioral mutants with learning/memory defects. All have used an odor-avoidance procedure (Boynton and Tully, 1992; Dudai et al., 1976). The first screen, in Seymour Benzer's laboratory at the California Institute of Technology, yielded dunce, the first experimentally induced mutant gene that produced deficient learning. The second screen, in W. G. Quinn's laboratory at Princeton University, produced the "vegetable" mutants, rutabaga, radish, turnip, and cabbage, which also showed no odor-avoidance learning (Aceves-Pina et al., 1983). A modification of this behavioral screen also yielded the amnesiac mutant, which showed normal learning but defective memory retention thereafter (Quinn, Sziber, and Booker, 1979). The third screen, in T. Tully's laboratory at Brandeis University, looked for performance defects three hours after Pavlovian training and yielded the mutants latheo, linotte, nalyot, and golovan (Tully et al., 1996).

Anatomical experiments (see below) have revealed that olfactory memory formation involves the mushroom body structures, neuropillar structures in the central brain thought to integrate sensory input. Accordingly, another screen for memory mutants was accomplished by looking first for "enhancer trap" strains that expressed a beta-galactosidase reporter gene preferentially in mushroom bodies and then by screening for behavioral defects in olfactory memory. Mutants of dunce and rutabaga were reisolated with this approach, which also yielded the mutants PKAC1, leonardo, and Volado (Roman and Davis, 2001).


Early biochemical experiments (see Aceves-Pina et al., 1983) and subsequent molecular cloning (see Dubnau and Tully, 1998) have established that dunce encodes a cAMP (Cyclic adenosine monophosphate)-specific phosphodiesterase (PDE), rutabaga encodes a calcium-sensitive adenylyl cyclase (AC), and amnesiac encodes a neuropeptide similar to vertebrate PACAP (pituitary adenylyl cyclase activating peptide). Three reverse-genetic experiments, focusing on additional components of this pathway, have strengthened this notion: 1. Mutations targeted to the regulatory subunit of PKA (PKA-RI) produced flies with olfactory learning defects. 2. Overexpression of a dominant-negative mutation of the stimulatory G protein subunit (Gs) in transgenic flies produced olfactory learning defects. 3. Overexpression of a repressor isoform of the CREB (cAMP response element binding protein) transcription factor in transgenic flies blocked olfactory long-term memory. Together, these behavior-genetic experiments strongly support the notion that cAMP signaling is central to olfactory memory formation in Drosophila. These data also are consistent with behavioral, electrophysiological, and cellular experiments in Aplysia (Bailey, Bartsch, and Kandel, 1996) and in mice (Wong et al., 1999), suggesting that cAMP signaling is part of an evolutionarily conserved molecular mechanism underlying synaptic and behavioral plasticity.

The MAP (mitogen-activated protein) kinase signaling pathway also may be involved. Molecular cloning of leonardo revealed it to be a mutation of the 14-3-3 gene, which, among other tasks, regulates the function of Ras/Raf, two of several proteins involved in GTP (guanine triphosphate) exchange. The Volado mutation has been shown to reside in a gene encoding α -integrin, a cell surface adhesion molecule. This general class of molecule often activates the MAP kinase pathway via interactions with receptor tyrosine kinases, though such a connection has not yet been made for Volado.

Mysteries remain and puzzles present themselves. The mutants radish, turnip, and cabbage have yet to be cloned. The linotte mutation resides either in the receptor tyrosine kinase, derailed, or in a novel neighboring transcript. The nalyot mutation lies in the Adf1 transcription factor; golovan is in the neurogenetic locus, extra machrochaetea. Exactly how these molecular-genetic components fit into the cell biology of olfactory memory is not yet clear. Curiously, the latheo mutation has been shown to disrupt ORC3, which encodes a protein subunit of the Origin Recognition Complex involved in DNA replication during cell proliferation. Surprisingly, the protein LAT also is expressed in synapses of terminally differentiated neurons, thereby suggesting a completely novel function for this protein outside of the nucleus of dividing cells. Although such genetic pleiotropy is not unusual, this observation nevertheless emphasizes how genetic screens can break the bondage of hypothesis-driven research. These "dangling mutants" presage an ultimate understanding of a more complicated and complete genetic basis of memory.


Gaining electrophysiological access to neurons in the adult central nervous system that subserve olfactory memory has been challenging. Thus, initial physiological characterizations of memory mutants relied on neural circuitries underlying other behaviors. Corfas and Dudai (1990), for instance, found that habituation of the (bristle) cleaning reflex in dunce and rutabaga mutants was abnormal, but this behavioral effect resulted from opposite physiological defects in the underlying sensory neurons. Sensory fatigue was accelerated in dunce mutants, while it was retarded in rutabaga mutants. Research by Engel and Wu (1996) began to characterize normal and mutant electrophysiological responses in the giant-fiber neurons, which contribute to habituation of the jump response.

Much insight on a role for memory genes in synaptic plasticity has come from studies of the larval neuromuscular junction (NMJ). In short, all memory mutants known to date produce distinct defects in synaptic structure or function, or both, at the NMJ (Saitoe and Tully, 2000). Increases in neural activity or in cAMP levels lead to increases in the number of synaptic boutons onto muscles (structure) and to increased excitability of synaptic transmission (function). Adf1 (nalyot) or fasII (another cell adhesion molecule) appear involved in changes of synaptic structure but not function, while dCREB2 regulates synaptic function but not structure (Davis, Schuster, and Goodman, 1996; DeZazzo et al., 2000). Thus, a genetic dissection of this form of developmental plasticity is under way.


Early electrophysiological and lesion experiments in bees identified a distinct anatomical region of the insect brain, the mushroom body, to be involved in associative processes (Menzel et al., 1991). In Drosophila, mutants with structural defects in specific adult brain anatomies were identified by M. Heisenberg's laboratory in Wurzburg, and those with abnormal mushroom bodies also displayed olfactory learning defects (Heisenberg, Borst, Wagner, and Byers, 1985). Four subsequent experiments clearly established mushroom bodies as a neural substrate of adult olfactory memory:

  1. Chemical ablation of mushroom body neurons completely abolishes olfactory learning (de Belle and Heisenberg, 1994);
  2. Overexpression of a dominant-negative Gs protein specifically in mushroom bodies completely abolishes olfactory learning (Connolly et al., 1996);
  3. Transgenic expression of RUTABAGA (the protein encoded by the rutabaga gene) in mushroom bodies specifically rescues the learning defect of rutabaga mutants (Zars, Wolf, Davis, and Heisenberg, 2000); and
  4. Structural mutants with lesions restricted to the alpha lobes (axonal projections from intrinsic mushroom body neurons) specifically abolish long-term memory (Pascual and Preat, 2001).

Together, these observations suggest that olfactory learning and memory depend, at least in part, on the activity of mushroom body neurons. Though these neurons remain largely inaccessible to classic electrophysiological investigations (but see Wright and Zhong, 1995), less invasive imaging techniques are revealing some of their cellular properties in response to experience (Rosay, Armstrong, Wang, and Kaiser, 2001; Wang et al., 2001).

The primary strength of neurogenetic analysis of memory in Drosophila lies in the discovery of genes. Behavioral screens for memory mutants enable this discovery process without any preconceived (and naive) assumptions about the underlying biochemical or anatomical substrates. To this end, yet another behavioral screen by Tully and coworkers at Cold Spring Harbor laboratory has discovered fifty-seven new mutants, defining forty-seven new genes. DNA microarrays also have been used to identify more than one thousand candidate memory genes (CMGs), which are transcriptionally regulated in normal flies during olfactory long-term memory formation. Significant genetic overlap exists between these two experimental approaches. As outlined above for the first few (dunce, rutabaga, latheo), these genes become an experimental "common currency" with which to investigate mechanisms of plasticity at several biological levels of organization (biochemical, physiological, anatomical, and so forth) and across many animal models. Combined, these data will reveal in more detail the molecular and cellular mechanisms by which memories form. In humans, various types of heritable mental retardation will become associated with homologues of these "memory genes" and the neurobiological pathways in which they participate (Oike et al., 1999; Petrij et al., 1995; Zhang et al., 2001). Ultimately, these mammalian homologues will become targets of drug discovery, thereby yielding viable therapies for those who suffer from cognitive dysfunction (Scott et al., 2002).



Aceves-Pina, E. O., Booker, R., Duerr, J. S., Livingstone, M. S., Quinn, W. G., Smith, R. F., Sziber, P. P., Tempel, B. L., and Tully, T. P. (1983). Learning and memory in Drosophila, studied with mutants. Cold Spring Harbor Symposia on Quantitative Biology 48, 831-839.

Bailey, C. H., Bartsch, D., and Kandel, E. R. (1996). Toward a molecular definition of long-term memory storage. Proceedings of the National Academy of Sciences of the United States of America 93, 13,445-13,452.

Boynton, S., and Tully, T. (1992). latheo, a new gene involved in associative learning and memory in Drosophila melanogaster identified from P element mutagenesis. Genetics 131, 655-672.

Connolly, J. B., Roberts, I. J., Armstrong, J. D., Kaiser, K. M. F., Tully, T., and O'Kane, C. J. (1996). Associative learning disrupted by impaired Gs signaling in Drosophila mushroom bodies. Science 274, 2,104-2,107.

Connolly, J. B., and Tully, T. (1998). Behaviour, learning, and memory. In D. B. Roberts, ed., Drosophila: A practical approach, 2nd edition, pp. 265-318. Oxford, UK: Oxford University Press.

Corfas, G., and Dudai, Y. (1990). Adaptation and fatigue of a mechanosensory neuron in wild-type Drosophila and in memory mutants. Journal of Neuroscience 10, 491-499.

Davis, G. W., Schuster, C. M., and Goodman, C. S. (1996). Genetic dissection of structural and functional components of synaptic plasticity, Part 3: CREB is necessary for presynaptic functional plasticity. Neuron 17 (4), 669-679.

de Belle, J. S., and Heisenberg, M. (1994). Associative odor learning in Drosophila abolished by chemical ablation of mushroom bodies. Science 263, 692-695.

DeZazzo, J., Sandstrom, D., deBelle, S., Velinzon, K., Smith, P., Grady, L., DelVecchio, M., Ramaswami, M., and Tully, T. (2000). nalyot, a mutation of the Drosophila myb-related Adf1 transcription factor, disrupts synapse formation and olfactory memory. Neuron 27, 145-158.

Dubnau, J., and Tully, T. (1998). Gene discovery in Drosophila : New insights for learning and memory. Annual Review of Neuroscience 21, 407-444.

Dudai, Y., Jan, Y.-N., Byers, D., Quinn, W., and Benzer, S. (1976). dunce : A mutant of Drosophila melanogaster deficient in learning. Proceedings of the National Academy of Sciences of the United States of America 73, 1,684-1,688.

Engel, J. E., and Wu, C. F. (1996). Altered habituation of an identified escape circuit in Drosophila memory mutants. Journal of Neuroscience 16, 3,486-3,499.

Heisenberg, M., Borst, A., Wagner, S., and Byers, D. (1985). Drosophila mushroom body mutants are deficient in olfactory learning. Journal of Neurogenetics 2, 1-30.

Menzel, R., Hammer, M., Braun, G., Mauelshagen, J., and Sugawa, M. (1991). Neurobiology of learning and memory in honeybees. In R. C. Fisher, ed., The behavior and physiology of bees, pp. 323-353. London: CAB International.

Oike, Y., Hata, A., Mamiya, T., Kaname, T., Noda, Y., Suzuki, M., Yasue, H., Nabeshima, T., Araki, K., and Yamamura, K. (1999). Truncated CBP protein leads to classical Rubinstein-Taybi syndrome phenotypes in mice: Implications for a dominant-negative mechanism. Human Molecular Genetics 8, 387-396.

Pascual, A., and Preat, T. (2001). Localization of long-term memory within the Drosophila mushroom body. Science 294, 1,115-1,117.

Petrij, F., Giles, R., Dauwerse, H., Saris, J., Hennekam, R., Masuno, M., Tommerup, N., van Ommen, G.-J. B., Goodman, R., Peters, D., and Breuning, M. (1995). Rubinstein-Taybi syndrome caused by mutations in the transcriptional co-activator CBP. Nature 376, 348-351.

Quinn, W., Sziber, P. P., and Booker, R. (1979). The Drosophila memory mutant amnesiac. Nature 277, 212-214.

Roman, G., and Davis, R. L. (2001). Molecular biology and anatomy of Drosophila olfactory associative learning. Bioessays 23, 571-581.

Rosay, P., Armstrong, J. D., Wang, Z., and Kaiser, K. (2001). Synchronized neural activity in the Drosophila memory centers and its modulation by amnesiac. Neuron 30, 759-770.

Saitoe, M., and Tully, T. (2000). Making connections between synaptic and behavioral plasticity in Drosophila. In J. McEachern and C. Shaw, eds., Toward a theory of neuroplasticity, pp. 193-220. New York: Psychology Press.

Scott, R., Bourtchuladze, R., Gossweiler, S., Dubnau, J., and Tully, T. (2002). CREB and the discovery of cognitive enhancers. Journal of Molecular Neuroscience.

Tully, T., Bolwig, G., Christensen, J., Connolly, J., DelVecchio, M., DeZazzo, J., Dubnau, J., Pinto, S., Regulski, M., Svedberg, B., and Velinzon, K. (1996). A return to genetic dissection of memory in Drosophila, Cold Spring Harbor Symposium on Quantitative Biology 61, 207-218.

Tully, T., Preat, T., Boynton, S. C., and Del Vecchio, M. (1994). Genetic dissection of consolidated memory in Drosophila. Cell 79, 35-47.

Wang, Y., Wright, N. J., Guo, H., Xie, Z., Svoboda, K., Malinow, R., Smith, D. P., and Zhong, Y. (2001). Genetic manipulation of the odor-evoked distributed neural activity in the Drosophila mushroom body. Neuron 29, 267-276.

Wong, S., Athos, J., Figueroa, X., Pineda, V., Schaefer, M., Chavkin, C., Muglia, L., and Storm, D. (1999). Calcium-stimulated adenylyl cyclase activity is critical for hippocampus-dependent long-term memory and late phase LTP. Neuron 23, 787-798.

Wright, N. J. D., and Zhong, Y. (1995). Characterization of K+ currents and the cAMP-dependent modulation in cultured Drosophila mushroom body neurons identified by lacZ expression. Journal of Neuroscience 15, 1,025-1,034.

Zars, T., Wolf, R., Davis, R., and Heisenberg, M. (2000). Tissue-specific expression of a type I adenylyl cyclase rescues the rutabaga mutant memory defect: in search of the engram. Learning and Memory 7, 18-31.

Zhang, Y. Q., Bailey, A. M., Matthies, H. J., Renden, R. B., Smith, M. A., Speese, S. D., Rubin, G. M., and Broadie, K. (2001). Drosophila fragile X-related gene regulates the MAP1B homolog Futsch to control synaptic structure and function. Cell 107, 591-603.


Revised byTimTully