Learning to Fly: Decoding Neural Circuits in Drosophila melanogaster

Model organisms are a special type of technology. Scientists can make then do things that we value as humans but that they might not have done in nature. From C.elegans (nematode worm)(1) to Mus musculus (laboratory mouse)(2), scientists research an assortment of creatures. I have recently gone down the evolutionary ladder and transitioned from work with mice to work with fruit flies. In this way, I have come to appreciate the neural activity patterns of Drosophila melanogaster (the fruit fly) and current efforts to decode neural circuits in the fly.

What is a neural circuit?

The term circuit is derived from Latin circuitus, which means “to go around”(7). Thus, one begins and ends in the same place in a circuit. A neural circuit is an ensemble of interconnected neurons that can regulate its own activity through feedback loops (3). Neural circuits emerge during development in order to produce a behavior:

Development → Circuit → Behavior

For a researcher studying neural circuits, the first step is to identify all the cellular components (often neurons) of that circuit. One must then characterize each neuron involved in the circuit. On October 20, 2015, I attended a Genetics Seminar relevant to neural circuitry given by Professor Chris Q. Doe(4) Doe is a Howard Hughes Medical Investigator (4) at the University of Oregon (5). He argues that researchers must consider how an individual neuron acquires its unique identity. He named five criteria for characterizing a neuron as follows:

1. Progenitor Lineage: In development, a progenitor cell gives rise to a distinct cell lineage by multiple cell divisions. In this way, Doe believes researchers must identify the progenitor or parent cell that gives rise to the neuron of interest.
2. Transcription Factor Code: A neuron is specified by a unique combination of transcription factors that is required to activate or repress a certain gene in a given cell.
3. Hox Code: The Hox (homeobox) code describes the expression of the Hox genes, which are critical for development. The Hox genes express a set of transcription factors critical to segment identity. In Drosophila, these genes decide if a segment of embryo becomes the head, thorax, or abdomen.
4. Spatial Identification: This simply refers to the position of the neuron in the tissue or embryo.
5. Temporal Identification: Doe defines this in two ways: 1. the changing intrinsic or extrinsic cues that act of a neural progenitor to specify the neuron of interest and 2. the progenitor cell’s response to changing cues (6).

Why the fly?

Drosophila (fruit fly) needs moisture (the name means “dew lover”) and feeds on yeast. In fact, the term fruit fly is a bit of a misnomer and yeast fly would be a more accurate description. In this blog post, I will argue that it is important to decode neural circuits in the fly because of its numerical simplicity (16). The Drosophila adult brain has 1000-fold fewer neurons than the mouse brain. Neurons, in theory, are in stereotyped positions in each fly. Although hermaphrodite C. elegans (nematode worm) is even simpler than the fly with 302 neurons, Drosophila has a wider range of behavior. In this way, Drosophila is a compromise between neuronal variation and behavioral repertoire.
There are many powerful tools available for Drosophila in genetics, molecular biology, and microscopy. Cornelia Bargmann (17), a Howard Hughes Medical Investigator at Rockefeller University (18), delivered a Presidential Lecture titled “Themes and Variations in Circuits and Behavior” at the Society for Neuroscience (SfN) 2015 meeting. In an interview I conducted with Professor Bargmann in July 2014, she emphasized the importance of building tools to decode neural circuits. She says:

On one level, we know a lot about the molecules in individual neurons. On the other level, we have fMRI19 where we look at big blobs of the brain and a single voxel is a million neurons. It is as though you could look at the Earth either with a microscope or a telescope but nothing in between. We can’t see the middle scale where people are interacting with each other and cars are moving on the freeway. And that’s what the new tools make possible. And that middle scale is where we believe the nature of mental process is— that is the stage where we believe neural codes and neural processing is taking place and gives rise to perception, motion, and memory.

As researchers aim to understand the patterns of information flow associated with the elusive brain, it is necessary to have the tools to do so. It remains a paradox that the organ that enables us to sense the outside world is also the most isolated from it.

Defining a Behavioral Paradigm for Drosophila

Early uses of Drosophila were quite varied and focused on identifying phototropisms (relevant to light) and chemotropisms (relevant to chemicals) (16). Scientists found that fruit flies move toward light when bothered and follow vapor trails of alcohol to rotting fruit. Recent publications such as “Time flies like an arrow: Fruit flies like crack?”(9) suggest that these tiny experimental creatures are not terribly different from their human counterparts.
Current uses of Drosophila are extensive and varied. Researchers have found that Drosophila can recall the spatial position of an object after it has been removed from their environment.10 In a different study, researchers discovered that a male fly can discriminate between female virgin and non-virgin flies (5). In the laboratory of Edward Kravitz at Harvard Medical School (20), Drosophila is used to model aggression (male and female fruit flies do fight and males even become territorial). Researchers have found that flies will fight over food or mates, and they can even alter their fighting strategy based on memories of previous encounters.

At the SfN 2015 meeting, Anita Burgos, a PhD candidate at Columbia University, presented her work on interneurons in Drosophila larvae that mediate nociception (pain-sensing) and are modulated by touch sensing neurons, which produce the behavior of rolling or 360 degree turns. To model real-life noxious situations for a Drosophila larva, such as a wasp attack, Burgos exposed larvae to heat that humans and flies alike find irritating. She then modeled larval locomotion in normal vs. noxious conditions.

The ultimate goal of a behavioral paradigm is to generate a phenotype, which is a distinct set of observable characteristics, for genetic manipulations. It is important to note that neurons of distinct circuits may mediate the same behavior. In other words, many neural circuits may act in parallel or in opposition for a single behavior. These neural circuits may share the same neurons but have distinct types of connections. In this way, the same set of neurons can be activated in a different order or at a different time to generate distinct behaviors. This is known as the multiplex perspective, as described in a 2014 PLOS ONE article.

Defining the components of a Neural Circuit: Characterization of Neurons

To identify neurons involved in a circuit(3), the GAL4/UAS System(14) is often used in Drosophila. The GAL4/UAS system has two parts: the GAL4 gene (encoding the yeast transcription activator protein Galactose-4) and the UAS (Upstream Activation Sequence). The gene of interest is placed next to the UAS, and GAL4 is used to activate transcription of the desired gene. In this way, the GAL4/UAS System can be used to identify sparsely labeled neuronal populations.

Interdisciplinary Approach to Decode Neural Circuits

The SfN 2015 meeting is undoubtedly the world’s largest conference focused on the brain. Nearly 30,000 researchers from 80 countries traveled to Chicago this year to attend SfN (21). They presented emerging discoveries, explored new technologies, and attended lectures. In particular, many developmental neurobiologists use SfN to forge collaborations with computational neuroscientists. For Drosophila, few mathematical models of neural circuits and behaviors exist. The advantages of modeling include revealing the operation of neural circuits by linking anatomical data (defining neurons) and functional data (defining behavioral paradigm). Moreover, mathematical models can be used to form hypotheses about neural activity and behavior.

Society for Neuroscience 2015

When I asked Anita Burgos, a graduate student in the laboratory of Wesley Grueber at Columbia University Medical Center, what she enjoyed most at SfN 2015, she says “the poster sessions, because you are able to look directly at data and ask questions to the person presenting. The same is not true of lectures where massive amounts of data are presented.” Anita enjoyed posters ranging from use of microRNAs to model Huntington’s disease(22) to the making of recombinant probes for synapse labeling (23). She spoke of how researchers at all career stages, ranging from graduate students to Nobel Laureates, were present in one venue to share their insights.
For many, SfN 2015 is a break from the rigors of laboratory work. Anita told me of a social event titled Squishies and Crunchies intended for researchers that work with worms (squishies) and flies (crunchies). She also added that she plans to come back every few years for the chance to meet with old friends and take a temporary break from research. But for now, she must continue to decode neural circuits and see how those larvae roll.

References

1. Caenorhabditis elegans. In Wikipedia. Retrieved October 28, 2015, from https://en.wikipedia.org/wiki/Caenorhabditis_elegans
2. Laboratory Mouse. In Wikipedia. Retrieved October 28, 2015, from
https://en.wikipedia.org/wiki/Laboratory_mouse
3. Biological Neural Network. In Wikipedia. Retrieved October 28, 2015, from
https://en.wikipedia.org/wiki/Biological_neural_network
4. http://www.hhmi.org/scientists/chris-q-doe
5. http://www.doelab.org/
6. Kohwi, M., & Doe, C. Q. (2013). Temporal fate specification and neural progenitor competence during development. Nature Reviews Neuroscience, 14(12), 823-838.
7. circuit. (n.). The American Heritage® Dictionary of the English Language, Fourth Edition. Retrieved October 28, 2015, from Dictionary.com website: http://dictionary.reference.com/browse/divers
8. Edward Tufte, Correlation/Causation, print on canvas, 52.75” x 27.5’, edition of 3. Retrieved October 28, 2015, from: www.edwardtufte.com
9. Hirsh, J. (2001). Time flies like an arrow. Fruit flies like crack?. The pharmacogenomics journal, 1(2), 97-100.
10. Neuser, K., Triphan, T., Mronz, M., Poeck, B., & Strauss, R. (2008). Analysis of a spatial orientation memory in Drosophila. Nature, 453(7199), 1244-1247.
11. Ejima, A., Smith, B. P., Lucas, C., Levine, J. D., & Griffith, L. C. (2005). Sequential learning of pheromonal cues modulates memory consolidation in trainer-specific associative courtship conditioning. Current biology, 15(3), 194-206.
12. Yurkovic, A., Wang, O., Basu, A. C., & Kravitz, E. A. (2006). Learning and memory associated with aggression in Drosophila melanogaster. Proceedings of the National Academy of Sciences, 103(46), 17519-17524.
13. Timme N, Ito S, Myroshnychenko M, Yeh F-C, Hiolski E, Hottowy P, et al. (2014) Multiplex Networks of Cortical and Hippocampal Neurons Revealed at Different Timescales. PLoS ONE 9(12): e115764. doi:10.1371/journal.pone.0115764
14. Duffy, J. B. (2002). GAL4 system in Drosophila: a fly geneticist’s Swiss army knife. Genesis, 34(1‐2), 1-15.
15. http://www.edwardtufte.com/tufte/
16. Olsen, S. R., & Wilson, R. I. (2008). Cracking neural circuits in a tiny brain: new approaches for understanding the neural circuitry of Drosophila. Trends in neurosciences, 31(10), 512-520.
17. Marino, M. (2005). Biography of Cornelia I. Bargmann. Proceedings of the National Academy of Sciences of the United States of America, 102(9), 3181-3183.
18. http://www.hhmi.org/scientists/cornelia-i-bargmann
19. Functional magnetic resonance imaging. In Wikipedia. Retrieved October 28, 2015, from https://en.wikipedia.org/wiki/Functional_magnetic_resonance_imaging
20. https://www.hms.harvard.edu/dms/neuroscience/fac/Kravitz.php
21. https://www.sfn.org/annual-meeting/neuroscience-2015
22. Richner, M., Victor, M. B., Liu, Y., Abernathy, D., & Yoo, A. S. (2015). MicroRNA-based conversion of human fibroblasts into striatal medium spiny neurons. Nature protocols, 10(10), 1543-1555.
23. Gross, G. G., Junge, J. A., Mora, R. J., Kwon, H. B., Olson, C. A., Takahashi, T. T., … & Arnold, D. B. (2013). Recombinant probes for visualizing endogenous synaptic proteins in living neurons. Neuron, 78(6), 971-985.

Naureen Ghani currently works at Columbia University Medical Center. She received her BS in Biomedical Engineering at Columbia University. In her spare time, she enjoys reading and painting watercolors.