|Home | About | Journals | Submit | Contact Us | Français|
Studies on the formation of connections in the developing nervous system are greatly aided by methods that permit the differential visualization and manipulation of pre- and postsynaptic partner neurons. This has been facilitated by the advent of the LexA-based, GAL4/UAS-independent, binary expression system. On the molecular side, the introduction of DNA sequences into expression vectors has been simplified by the Invitrogen Gateway cloning technology. We have developed cloning vectors that combine the Gateway cloning technology with the LexA-based genetic expression system. These vectors facilitate the creation of driver and reporter constructs for the generation of Drosophila transgenic lines for the new LexA-based binary transcriptional system. We further report a new LexA::GAD sensory neuron driver and a red fluorescent membrane targeted lexAop reporter designed to complement the existing GFP-based lexAop reporter. Using these transgenic lines we have been able to differentially label motor and sensory neuron projections in the ventral nerve cord of Drosophila larvae.
The fruit fly Drosophila melanogaster has been successfully used as a model organism for studying the development of neuronal networks.1-4 Major advances in this endeavor have been facilitated by the GAL4/UAS binary gene expression system developed by Brand and Perrimon5 more than a decade ago. This by now well established Gal4/UAS expression system enables researchers to genetically manipulate and visualize specific subsets of neurons and thus characterize the anatomy and function of these cells.6,7 The visualization and manipulation of pre- and postsynaptic partners independently of one another has been facilitated by the recent addition of a second binary Drosophila expression system, which is GAL4/UAS-independent and based on the bacterial LexA DNA binding protein.8 To expedite the generation of new driver and reporter lines for this LexA system we designed vectors that integrate the efficient Gateway cloning technique with LexA system vectors.9 For the efficient and reliable generation of promoter-LexA system driver lines we designed Gateway-LexA::GAD (Gal4 Activation Domain) vectors and lexAop-Gateway (lexA operator) vectors for the creation of new reporter lines. To demonstrate the application of these vectors we generated a peripheral nervous system-specific LexA::GAD driver line using a regulatory fragment from the Choline acetyl transferase (Cha)10 locus. To complement the existing lexAop-rCD2::GFP reporter, we also made a red fluorescent mCherry11 (monomeric fast maturing variant of mRFP) lexA-operator reporter where the fluorophore is targeted to the cell membrane via a myristoylation site, lexAop-myr-mCherry. These vectors and resultant transgenic stocks have enabled us to look for potential postsynaptic partners of sensory neurons in the Drosophila nervous system.
We designed a vector that permits the straightforward insertion of identified regulatory regions upstream of the transcriptional LexA::GAD activation unit and thus facilitates the creation of new driver lines for the LexA binary expression system. We designated this vector “pCasper-Gateway(W)-LexA::GAD”, which contains a pCasper vector backbone, the so-called “W” Gateway cloning cassette followed by DNA coding for the LexA::GAD (Gal4 activation domain) fusion protein. This pCasper-Gateway (W)-LexA::GAD destination vector was generated by replacing the Repo regulatory sequences in the pCasper-Repo-LexA::GAD vector8 with the Gateway reading frame B, which contains all essential parts for the Gateway cloning technique (Fig. 1). The Gateway version of the lexAop destination reporter vector, called “pLOT-W”, was generated by ligating the Gateway reading frame into the existing pLOT-vector8 (Fig. 1). This new Gateway vector expedites the generation of new lexAop reporter lines using the Gateway cloning method.
To verify the functionality of these new Gateway destination vectors, we used them to generate transgenic LexA driver and reporter lines. We recombined a 3.3 kb fragment of the Cha regulatory region into the pCasper-W-LexA::GAD destination vector. The expression of the resultant transgenic lines is specific to sensory neurons with the exception of a few cells in the brain lobes. Strong expression is detected in chordotonal neurons in embryonic and all larval stages (Fig. 2A and C). Weaker expression can be seen in the dbd (dorsal bipolar dendritic) neurons and the multidendritic neurons of the v’ cluster in the peripheral nervous system. Axonal projections of sensory cells into the ventral nerve cord are clearly visible with the lexAop-rCD2::GFP reporter (Fig. 2A).
To complement the existing lexAop-rCD2::GFP reporter we produced a membrane targeted red fluorescent (mCherry)11 variant. This construct was assembled using the MultiSite Gateway cloning technique, thus demonstrating the functionality of the new pLOT-W vector also for this more versatile cloning method. To target the mCherry fluorophore to the plasma membrane we tagged it N-terminally with a myristoylation site consensus sequence.12 Expression of lexAop-myr-mCherry under control of the Cha3.3-LexA::GAD driver shows that this reporter is excellent for charting neuronal projections (Fig. 2B–D). We note however, that in cell bodies the localization of the myr-mCherry is less pronounced than that of rCD2::GFP (data not shown). We could not detect any ectopic expression in salivary glands as reported for constructs based on Stinger or Pelican vectors.13
Finally, we combined the GAL4/UAS and the LexA/lexAop binary expression systems to ask whether sensory neuron terminals overlapped with motor neuron dendrites in the larval CNS, as has been demonstrated for stretch receptors in Manduca sexta.14 We generated first instar larvae in which sensory projections were labeled with lexAop-rCD2::GFP and motor neurons with OK371-Gal4,15 driving UAS-mCD8::RFP16 (Fig. 2E). In these larvae, both populations of neurons and their projections into the neurophile were clearly differentially labeled. Digital cross-sections from confocal image stacks of such nerve cords did not show any obvious apposition of motor neuron dendrites and sensory neuron terminals (n = 6) (Fig. 2E and F).
(i) The pCasper-Gateway(W)-LexA::GAD vector was created by excision of the Repo promotor fragment with NotI and AccI (Roche) from pCasper-Repo-LexA::GAD.8 Blunt ends were created with the Klenow fragment of DNA Polymerase III (Promega), then the Gateway Reading Frame B (Invitrogen) was ligated into the vector using T4 DNA Ligase (Promega). The plasmid was transformed into DB3.1 cells (Invitrogen) and selected for Kanamycin resistance. DNA was extracted and verified by sequencing. The sequence of the vector is available as a supplementary data file (Suppl. 1).
(ii) The pLOT-Gateway vector was created by opening the pLexOp-vector8 with EcoRI (Roche). Ends were blunted, the Gateway Reading Frame B (Invitrogen) was ligated in and the vector was introduced into DB3.1 cells as outlined above. The sequence of the vector is available as a supplementary data file (Suppl. 2).
(iii) Sensory LexA::GAD driver line: A 3.3 kb fragment from the Cha promotor was amplified by Polymerase Chain Reaction (PCR) using the following Gateway primers: attB1-GAA TTC TTA ATT GAA AAT AAA CAT TAA GG and attB2-GGA TCC GGT TGG TTT GGC CCC TTT TTC TTT GTC GCT. The PCR product was introduced into pDONRTM221 (Invitrogen) via Gateway cloning to create the pEntry-vector. In the subsequent Gateway cloning reaction this vector was combined with the newly developed pCasper-Gateway-LexA::GAD (above) to recombine the Cha promotor fragment with the LexA::GAD. DNA was purified using a Qiagen Midi Kit and transgenic lines were generated by BestGene Inc., (Chino Hills, CA, USA).
(iv) lexAop-myr-mCherry was created using the Multisite Gateway technique. One pEntry vector for the myristoylation site was created using the primers: attB1-ATG GGC AAC AAA TGC TCC AG and attB5r-TGG TCT GAT GAT GTC AAC CCC. Another, pEntry-mCherry entry vector was generated using the primers attB5-AAGAGCTCCGCCACCATGG and attB2-GGT TTA CGT CAC GTG GAC CGG TG. Both pEntry-vectors were combined into the new pLOT-Gateway vector using the MultiSite Gateway Kit from Invitrogen. Clones were selected on Ampicillin plates and DNA isolated and sequenced. Purified DNA was sent for embryo injection services to BestGene Inc., (Chino Hills, CA, USA).
Expression patterns were recorded in intact young 1st instar larvae or isolated ventral nerve cords of the same stage using a Zeiss Axiophot widefield fluorescence microscope equipped with an AxioCam MR operated by AxioVision software (Zeiss). Confocal image stacks were taken with an Olympus 60x/1.2NA water immersion objective using a CSU-22 spinning disk confocal system (Yokagawa), mounted on an Olympus BX51-WI microscope, operated with MetaMorph (Molecular Devices). Images were processed using ImageJ (NIH) and Photoshop (Adobe) software.
By changing the available LexA-vectors (pCasper-Repo-LexA::GAD and pCasper-LOT) into Gateway destination vectors we have made this recently developed binary transcriptional system accessible to a highly efficient cloning technology. These new vectors will allow Drosophila researchers to generate new driver and reporter lines for the LexA-system rapidly and efficiently.
The sensory neuron-specific LexA::GAD driver line will be a useful tool with which to identify candidate partner neurons, for instance GAL4 expressing neurons that innervate the same neuropile as the sensory terminals.
The new membrane targeted red fluorescent lexAop reporter lexAop-myr-mCherry that we have introduced complements the existing lexAop-rCD2::GFP reporter. It is ideal for use in combination with the plethora of GFP-tagged GAL4 constructs, e.g., those marking presynaptic or other subcellular structures.17-19
We have illustrated the potential of using the GAL4/UAS and the LexA/lexAop systems in combination for studying connectivity in the nervous system. Using a sensory neuron-specific LexA driver and a motor neuron-specific GAL4 driver we were able to survey the neuropile for potential contact sites between these two neuronal sets, though we did not find evidence for direct sensory-motor neuron connections at early larval stages.
We thank S. Lai and T. Lee for the generous gift of the original LexA-vectors, P. Salvaterra for the 7.4 Cha DNA, B. Ye and T. Lee for fly stocks and A. North for helpful comments. This work supported by the Wellcome Trust Programme grant (075934) to Micheal Bate and Matthias Landgraf.
Supplementary materials can be found at: www.landesbioscience.com/supplement/DiegelmannFLY2-4-Sup.pdf