Binary Expression Systems

Almost every light microscopy image on Virtual Fly Brain exists because someone put a reporter into a defined set of cells using a binary expression system. Understanding the two-part structure explains what an expression pattern on VFB is an image of, and why the same driver appears in dozens of different experiments.

The two parts

A binary system separates where from what:

• A driver transgene puts a transcription factor under the control of a piece of genomic DNA — usually an enhancer fragment — so the factor is made in whatever cells that fragment is active in.
• An effector (or reporter) transgene puts the gene you actually want expressed downstream of a binding site for that factor. On its own it does nothing.

Cross the two lines, and the effector is expressed only where the driver is active. The same driver can be crossed to a fluorescent reporter to image a pattern, to a silencer to switch those cells off, or to an optogenetic channel to switch them on — without rebuilding the driver.

Brand and Perrimon adapted the yeast GAL4 transcription factor and its UAS binding site for exactly this (Brand and Perrimon, 1993). It remains the dominant system in fly neurobiology, and every GAL4 line on VFB is an instance of it.

The systems you will meet on VFB

Having more than one orthogonal system matters because it lets you label two populations in different colours, or drive an effector in one population while reporting activity in another (Lai and Lee, 2006; Potter et al., 2010).

Control in time as well as space

GAL80 represses GAL4. A temperature-sensitive GAL80 therefore gates GAL4 activity by temperature, converting a spatial tool into a spatiotemporal one — the TARGET system, “temporal and regional gene expression targeting”, which McGuire et al. used to rescue a memory defect transiently in the adult mushroom bodies and so rule out a developmental cause (McGuire et al., 2003). GAL80 is also what makes MARCM work — see stochastic labelling.

Where the driver lines came from

Early drivers were enhancer traps: a transposon inserted at random, and the reporter next to it reported on whatever regulatory element it landed near (O’Kane and Gehring, 1987). Useful, but neither systematic nor reproducible in insertion site.

This is also where the w¹¹¹⁸ background in so many stock genotypes comes from. Transformation constructs carry a selectable marker, and the classical one is mini-white: the P{GawB} enhancer-trap construct behind many early GAL4 lines is marked with w^+mW.hs (FBtp0000352). A white⁺ marker restores eye pigment, so it can only be scored in a white-mutant animal — which is why driver line genotypes on VFB so often read w[1118];P{w[+mW.hs]=GawB}…, and why the first Drosophila mutant ever described is still the standard way of telling that a transgene went in.

Two changes made large collections possible. First, the φC31 integrase — which catalyses unidirectional site-specific recombination between attB and attP sites — was shown to work in Drosophila, and attP sites were integrated into the genome to act as defined landing sites (Groth et al., 2004). Constructs placed there differ only in their enhancer fragment, not in where they sit, so patterns can be compared without position effects confounding them. Second, defined genomic fragments were cloned systematically rather than trapped (Pfeiffer et al., 2008).

The two collections that dominate VFB’s light microscopy followed:

• The Janelia GAL4 collection — 7,000 lines, with CNS expression imaged and published for 6,650 (Jenett et al., 2012).
• The Vienna Tiles (VT) collection — 7,705 enhancer candidates characterised in vivo (Kvon et al., 2014).

Both are searchable on VFB, registered to the standard templates and classified with anatomy ontology terms, so you can ask which lines label a region or cell type rather than reading pattern images one at a time.

What “expression pattern” means on VFB, formally

The term has a precise definition behind it. VFB defines a relation expresses holding between an anatomical entity and a gene or transgene, and then defines an expression pattern as the anatomical entity consisting of the mereological sum of all cells that express that gene or transgene (Osumi-Sutherland et al., 2014).

That is why VFB distinguishes an expression pattern from an expression pattern fragment. An image of a single neuron or a clone picked out of a driver’s pattern depicts only part of that sum, so it is modelled as a part of the expression pattern rather than as the pattern itself. The distinction is not cosmetic: it is what stops a sparse MCFO image of one cell being treated as the full extent of what a driver labels.

What an expression pattern image is, and is not

• It is the whole pattern of that driver, not of one cell type. A GAL4 line usually labels several unrelated populations. Specificity is what split drivers address.
• It is one animal. Expression varies between individuals, and between rearing conditions. Patterns on VFB are registered so that many such images can be compared.
• Absence of signal is weak evidence. A pattern may fail to show a cell because the enhancer is not active there, or because the reporter was too dim, or because that part of the CNS was not imaged — see FlyLight imaging tiles.
• Reporter choice changes what you see. Membrane-targeted, nuclear and cytoplasmic reporters give visibly different images of the same driver.

Effectors beyond reporters

The same drivers carry non-imaging effectors, which is why a driver line found on VFB is usually the starting point for a functional experiment:

• Silencing — temperature-sensitive shibire to block synaptic vesicle recycling (Kitamoto, 2001), or Kir2.1 to hyperpolarise.
• Activation — channelrhodopsins such as CsChrimson (Klapoetke et al., 2014).
• Activity imaging — GCaMP calcium indicators (Chen et al., 2013).
• Knockdown — UAS-RNAi from the VDRC and TRiP collections (Dietzl et al., 2007; Perkins et al., 2015).
• Connectivity reporters — GRASP (Feinberg et al., 2008) and trans-Tango (Talay et al., 2017).

VFB indexes the expression patterns; the reagents themselves are catalogued in FlyBase and distributed by the stock centres, and VFB links out to both from each transgene record. See transgene expression curation for how that record is built.

Sources

• Brand AH, Perrimon N (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401–415. doi:10.1242/dev.118.2.401
• O’Kane CJ, Gehring WJ (1987) Detection in situ of genomic regulatory elements in Drosophila. PNAS 84:9123–9127. doi:10.1073/pnas.84.24.9123
• Kitamoto T (2001) Conditional modification of behavior in Drosophila by targeted expression of a temperature-sensitive shibire allele in defined neurons. J Neurobiol 47:81–92. doi:10.1002/neu.1018
• McGuire SE et al. (2003) Spatiotemporal rescue of memory dysfunction in Drosophila. Science 302:1765–1768. doi:10.1126/science.1089035
• Groth AC et al. (2004) Construction of transgenic Drosophila by using the site-specific integrase from phage φC31. Genetics 166:1775–1782. doi:10.1534/genetics.166.4.1775
• Lai S-L, Lee T (2006) Genetic mosaic with dual binary transcriptional systems in Drosophila. Nat Neurosci 9:703–709. doi:10.1038/nn1681
• Dietzl G et al. (2007) A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature 448:151–156. doi:10.1038/nature05954
• Feinberg EH et al. (2008) GFP reconstitution across synaptic partners (GRASP) defines cell contacts and synapses in living nervous systems. Neuron 57:353–363. doi:10.1016/j.neuron.2007.11.030
• Pfeiffer BD et al. (2008) Tools for neuroanatomy and neurogenetics in Drosophila. PNAS 105:9715–9720. doi:10.1073/pnas.0803697105
• Potter CJ et al. (2010) The Q system: a repressible binary system for transgene expression, lineage tracing, and mosaic analysis. Cell 141:536–548. doi:10.1016/j.cell.2010.02.025
• Jenett A et al. (2012) A GAL4-driver line resource for Drosophila neurobiology. Cell Rep 2:991–1001. doi:10.1016/j.celrep.2012.09.011
• Chen T-W et al. (2013) Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499:295–300. doi:10.1038/nature12354
• Klapoetke NC et al. (2014) Independent optical excitation of distinct neural populations. Nat Methods 11:338–346. doi:10.1038/nmeth.2836
• Kvon EZ et al. (2014) Genome-scale functional characterization of Drosophila developmental enhancers in vivo. Nature 512:91–95. doi:10.1038/nature13395
• Perkins LA et al. (2015) The Transgenic RNAi Project at Harvard Medical School: resources and validation. Genetics 201:843–852. doi:10.1534/genetics.115.180208
• Talay M et al. (2017) Transsynaptic mapping of second-order taste neurons in flies by trans-Tango. Neuron 96:783–795. doi:10.1016/j.neuron.2017.10.011
• Osumi-Sutherland D, Costa M, Court R, O’Kane CJ (2014) Virtual Fly Brain — using OWL to support the mapping and genetic dissection of the Drosophila brain. Proceedings of OWLED 2014, CEUR Workshop Proceedings 1265:85–96. PDF