ShineGAL4 drivers for tissue and cell-type specific optogenetics in Drosophila
Victor Girard, Sebastian Sorge, Joachim Kurth, Cyrille Alexandre, Alex P. Gould
TL;DR
Researchers developed ShineGAL4, a new optogenetic tool for precise control of gene expression in specific tissues and cells in fruit flies.
Contribution
The study introduces ShineGAL4 drivers, enabling light-controlled gene expression in diverse Drosophila tissues and cell types.
Findings
ShineGAL4 drivers allow rapid and specific gene induction in tissues like fat body, muscles, and neurons.
The system uses CRISPR to replace GAL4 with a photoswitchable version for spatiotemporal control.
An optogenetic cassette was developed for activating GAL4 in previously silent clones.
Abstract
An optogenetic split-GAL4 system, ShineGAL4, allows genes to be manipulated with unprecedented spatiotemporal precision. Here, we convert a panel of 14 GAL4 drivers widely used in Drosophila research into their ShineGAL4 counterparts. Homology assisted CRISPR knock-in (HACK) is used to replace GAL4 with the GAL4 DNA binding domain fused to a Magnet photoswitch. We show that the resulting ShineGAL4 drivers enable gene expression to be rapidly induced by light specifically in fat body, muscles, enterocytes, oenocytes, Malpighian tubules, neurons, neuroblast lineages, glial subtypes or in all glia. We also develop an optogenetic cassette for photoactivation of GAL4 in ‘silent’ FLP-out clones. This panel of optogenetic tools will enable precise spatiotemporal control of gene expression in a wide range of different Drosophila tissues and cell-types. Summary: A panel of optogenetic GAL4…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Fig. 1
Fig. 2
Fig. 3| Tissue/cell type | GAL4 line | Converted | Chr. |
|---|---|---|---|
| Mature neurons |
|
| 3 |
| Pan neuronal |
|
| X |
| Neuroblast lineage |
|
| 3 |
| Pan glial |
|
| 3 |
| Cortex glia |
|
| 3 |
| Subperineurial glia |
|
| 2 |
| Astrocyte-like glia |
|
| 3 |
| Fat body |
|
| 3 |
| Enterocytes |
|
| 2 |
| Enterocytes |
|
| 2 |
| Muscles |
|
| 3 |
| Oenocytes |
|
| 2 |
| Malpighian tubule |
|
| 2 |
| Actin FLP-out clones |
|
| 3 |
- —Cancer Research UKhttp://dx.doi.org/10.13039/501100000289
- —Medical Research Councilhttp://dx.doi.org/10.13039/501100000265
- —Wellcome Trusthttp://dx.doi.org/10.13039/100010269
- —EMBOhttp://dx.doi.org/10.13039/100004410
- —Francis Crick Institutehttp://dx.doi.org/10.13039/100010438
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsLight effects on plants · Photoreceptor and optogenetics research · Photochromic and Fluorescence Chemistry
INTRODUCTION
Genetic research in Drosophila has benefited greatly from the GAL4/UAS binary system for modulating gene expression in a cell-type specific manner (Brand and Perrimon, 1993). This system relies on two genetic elements derived from budding yeast (Saccharomyces cerevisiae): a transcriptional activator (GAL4) with its cognate DNA binding site and the upstream activating sequence (UAS). When GAL4 is expressed under a tissue-specific enhancer, it binds to the UAS and initiates DNA polymerase II transcription of downstream sequences. The GAL4/UAS system has been refined several times over the years to enable more precise spatial and/or temporal control of gene expression. For example, GAL4 has been separated into two components (split-GAL4), whereby the GAL4 DNA binding domain [GAL4(DBD)] and a p65 transactivation domain (AD) are fused to a heterodimerising leucine zipper domain and each expressed from an independent promoter (Luan et al., 2006). Dimerisation of the leucine zipper domain brings AD and GAL4(DBD) together, thus reconstituting GAL4 activity only in those cells that express both promoters. This intersectional strategy has been used to great effect to refine GAL4 expression to very specific cell populations and many large sets of split-GAL4 drivers are now available, especially for specific neuronal subtypes (Ehrhardt et al., 2025 preprint; Meissner et al., 2023, 2025; Soffers et al., 2025). Another way to refine the spatial expression pattern of GAL4 is to use its natural repressor GAL80 (Ma and Ptashne, 1987; Salmeron et al., 1990). GAL80 can thus be used to switch off GAL4 activity in undesired sites within its expression domain. GAL80 repression of GAL4 activity is also an important component of mosaic analysis with a repressible cell marker (MARCM), a technique for marking and/or genetically manipulating clones of dividing cells (Lee and Luo, 1999, 2001; Lee et al., 2000).
A key modification of the GAL4/UAS system has been to provide it with temporal control. This was made possible with the temporal and regional gene expression targeting (TARGET) system, which uses a temperature sensitive allele of GAL80 (GAL80^ts^) that is active around the permissive temperature (18°C) but degraded at the restrictive temperature (29°C) (Matsumoto et al., 1978; McGuire et al., 2003). TARGET has been widely used with great success for temporally restricting genetic manipulations, but the temperature shifts can impact aspects of development, metabolism and physiology that complicate analysis. In addition, the degradation of GAL80^ts^ at the restrictive temperature is not immediate and GAL4-driven transcription is therefore only activated to detectable levels with a delay of 7-10 h (di Pietro et al., 2021). Additional modifications of the GAL4/UAS-system include variants where GAL4 activity is induced by exogenous hormones. For example, GeneSwitch is based on a fusion between the GAL4(DBD) and a hormone-sensitive activation domain that is only active in the presence of a synthetic steroid, mifepristone (RU-486) (Burcin et al., 1999; Nicholson et al., 2008; Osterwalder et al., 2001; Roman et al., 2001). An auxin-inducible gene expression system (AGES) has also been developed, which uses a GAL80 variant that is degraded in the presence of the plant hormone auxin (McClure et al., 2022). The GeneSwitch and AGES approaches allow dose-dependent GAL4 activation without side effects associated with temperature shifts. Nevertheless, they require Drosophila to be exposed to exogenous hormones usually administered in the food, which limits the precision of temporal regulation.
Recently, an optogenetic version of the Drosophila GAL4/UAS system was developed (di Pietro et al., 2021). This uses a modified split-GAL4, called ShineGAL4, where the GAL4(DBD) and AD are bound together by a pair of light-sensitive proteins known as Magnets (di Pietro et al., 2021; Kawano et al., 2015). Upon light exposure, Magnets of opposite ‘polarity’ heterodimerise and reconstitute an active GAL4 transcription factor (Fig. 1A). Major advantages of this system are that GAL4 can be temporally activated very rapidly and also, if required, spatially restricted to small regions, or even to single cells, using localised photoactivation (di Pietro et al., 2021). Here, we convert a panel of widely used GAL4 drivers, and also an FLP-out cassette, into ShineGAL4 lines that are publicly available as a new optogenetic resource for the Drosophila community.
Photoactivation of tissue and cell-type specific ShineGAL4 drivers. (A) Schematic of optogenetic activation mechanism for ShineGAL4. The dark and light states are shown, adapted from di Pietro et al. (2021). (B-E) Confocal images (from top to bottom) showing the fat body (B), gut (C), oenocytes (D), Malpighian tubule principal cells (E) and myocytes (F) expressing CD8::GFP (mGFP) under the control of the indicated tissue or cell-type specific ShineGAL4 drivers, exposed to the dark or white light. In B-E, mGFP immunostaining (left) and DAPI-stained nuclei (right) after 4- or 24-h light exposure are shown. In F, mGFP immunostaining (left) and DAPI-stained nuclei and phalloidin-stained F-actin (right) after 24-h light exposure are shown. Full ShineGAL4 driver genotypes are as follows: ubi-pMag::AD; Lpp-GAL4DBD::nMag/UAS-CD8::GFP for fat body, ubi-pMag::AD, Desat1-GAL4(DBD)::nMag; UAS-CD8::GFP for oenocytes, ubi-pMag::AD, Uro-GAL4(DBD)::nMag[3xP3-Cherry]; UAS-CD8::GFP for Malpighian tubule principal cells, ubi-pMag::AD/Mex1-GAL4(DBD)::nMag[3xP3-Cherry]; UAS-CD8::GFP for enterocytes and ubi-pMag::AD; Mef2-GAL4(DBD)::nMag/UAS-CD8::GFP for myocytes. For Desat1-GAL4(DBD)::nMag expression in the dark, the weak signal in oenocytes is due to background not to mGFP expression (see Fig. S1E). Scale bars: 50 µm.
RESULTS AND DISCUSSION
We selected a panel of tissue or cell-type specific GAL4 drivers that are widely used in Drosophila research (Table 1). For each driver, homology assisted CRISPR knock-in (HACK) was used to replace GAL4 sequences with a GAL4(DBD) fused to a negative polarity Magnet photoswitch (henceforth called GAL4(DBD)::nMag) (Chen et al., 2020; di Pietro et al., 2021; Lin and Potter, 2016; Poernbacher et al., 2019 preprint). In brief, double strand breaks were generated in the GAL4 sequence downstream of the DBD using Cas9 and a specific guide RNA. These breaks were then repaired using a template encoding nMag and containing a Cre-excisable 3xP3-Cherry selection cassette, all flanked by GAL4 homology arms (Fig. S1A). Candidates carrying GAL4(DBD)::nMag were selected using a fluorescent marker (3xP3-Cherry) that was subsequently excised in most cases (Fig. S1B). Newly generated GAL4(DBD)::nMag drivers (Table 1) were tested for photoactivation by crosses to flies carrying UAS-CD8::GFP and a ubiquitously expressed Magnet photoswitch of the opposite polarity fused to a p65 transactivation domain (henceforth called ubi-pMag::AD). Progeny carrying all components of the ShineGAL4 system were raised in the dark until the late-L3 larval stage before exposure to white light for a defined period of time (Fig. 1A). Larvae carrying GAL4(DBD)::nMag under the control of Lpp (also known as apolpp; fat body), Myo10A or Mex1 (enterocytes), or Desat1 (also known as PromE; oenocytes) displayed robust and cell-type specific CD8::GFP (mGFP) expression after 4 h of light exposure but not if they were raised in the dark (Fig. 1B-D). In the case of larvae carrying GAL4(DBD)::nMag under the control of Mef2 (myocytes) and Uro (Malpighian tubule principal cells), GFP was barely detectable after light exposure for 4 h but robustly expressed after 24 h of illumination (Fig. 1E,F). Timecourses of photoactivation for Lpp-ShineGAL4 and Desat1-ShineGAL4 show that native mGFP fluorescence was detectable from 3-6 h and was strong at 24 h of exposure to white light (Fig. S1C-F). In summary, there was some variation between different tissue-specific ShineGAL4 drivers in the time required to detect the onset of mGFP expression but, in most cases, it occurred after 3-6 h of light exposure. Importantly, we did not detect expression of mGFP in the dark, indicating that leakiness of the ShineGAL4 system is minimal.
The Drosophila CNS contains a wide variety of diverse cell types including neural stem cells (neuroblasts) and hundreds of different neuronal subtypes (Scheffer et al., 2020). In addition, there are at least five different subtypes of glia including the subperineurial glia of the blood brain barrier, the cortex glia that surround neuroblast lineages, and the astrocyte-like glia of the neuropil (Yildirim et al., 2019). Commonly used GAL4 drivers specific for these neural cell types were selected and converted into their GAL4(DBD)::nMag equivalents using HACK. We chose elav^c155^-GAL4 and nSyb-GAL4 to target mature neurons and nab-GAL4 to target postembryonic neuroblast lineages. repo-GAL4, Cyp4g15-GAL4, moody-GAL4 and alrm-GAL4 were selected to target all glia, cortex, subperineurial and astrocyte-like glia, respectively (Table 1). As the 3xP3-Cherry selection cassette is expressed in the larval CNS, including in cortex glia, it was excised from the final GAL4(DBD)::nMag element, thus freeing the red channel for other uses. However, in the case of nSyb-ShineGal4 and two non-CNS drivers (Uro-ShineGal4 and Mex1-ShineGal4), the 3xP3-Cherry cassette was retained as its deletion decreased GAL4 expression. After 24 h of white light exposure, all seven CNS drivers showed robust mGFP expression, detected via immunofluorescence, whereas no mGFP signal was detected in larvae shielded from light (Fig. S2A-G). To investigate the dynamics of photoactivation, we quantified native mGFP fluorescence at four time points (0, 3, 6 and 24 h of light exposure) for each of the CNS ShineGal4 drivers. For most drivers, mGFP fluorescence in cell-type specific patterns was detectable at 3 h and increased substantially by 24 h of light exposure (Fig. 2A-G). This analysis shows that ShineGal4 drivers enable rapid optogenetic induction of UAS transgene expression in most of the major cell types of the CNS.
*Photoactivation timecourses of CNS cell-type specific ShineGAL4 drivers. (A-G) Confocal projections of CNSs from late L3 larvae expressing CD8::GFP (mGFP) under the control of the indicated cell-type specific ShineGAL4 drivers, following exposure to white light for 0 h (dark), 3 h, 6 h or 24 h (light). Graphs (right panels) show corresponding quantifications of native mGFP fluorescence normalised to the 24 h photoactivation condition (mean±s.d.). Each dot represents one animal (N=3-10 animals per condition). Significant differences from dark controls (0 h) are indicated (*P<0.05, **P<0.01, **P<0.001; one way ANOVA and Dunett's post hoc test). Full driver genotypes are as follows: (A) for postembryonic neuroblast lineages (NB+secondary neurons): ;; ubi-pMag::AD, nab-GAL4(DBD)::nMag/UAS-CD8::GFP (B) for mature neurons (primary neurons): ; ubi-pMag::AD; nSyb-GAL4(DBD)::nMag[3xP3-Cherry]/UAS-CD8::GFP (C) for neuroblast lineages including mature neurons (NB+neurons): elavC155-GAL4(DBD)::nMag; ubi-pMag::AD; UAS-CD8::GFP (D) for pan-glia: ;; ubi-pMag::AD, repo-GAL4(DBD)::nMag/UAS-CD8::GFP (E) for cortex glia: ; ubi-pMag::AD; Cyp4g15-GAL4(DBD)::nMag/UAS-CD8::GFP (F) for subperineurial (SPG) glia: ; ubi-pMag::AD, moody-GAL4(DBD)::nMag; UAS-CD8::GFP (G) for astrocyte-like glia: ubi-pMag::AD; alrm-GAL4(DBD)::nMag/UAS-CD8::GFP.
We next compared expression of ShineGAL4 drivers after 24 h of photoactivation with their original GAL4 counterparts, using the same UAS-CD8::mGFP reporter. In each case, the spatial patterns appeared to be similar but, in about half of the drivers, mGFP expression was noticeably weaker with the ShineGAL4 version (Fig. S3). More specifically, ShineGAL4 drivers for Desat1, Lpp, nab, Cyp4g15 and moody gave comparable expression to their GAL4 counterparts, whereas for repo, alrm and elav comparable expression was only detected in a subset of the original GAL4 domain (Fig. S3). These differences in expression may be due either to GFP accumulation driven by the original GAL4 drivers over a time period longer than 24 h and/or to cell-type differences in expression levels of the ubi-pMag::AD split driver. Either way, ShineGAL4 expression was sufficient to be detectable by GFP fluorescence and more strongly by anti-GFP immunostaining (Fig. 2; Fig. S2).
The ability to generate marked clones of cells expressing a gene of interest is a mainstay of the Drosophila genetic toolkit. One widely used method for this is the FLP-out system, whereby a ubiquitous promoter is separated from GAL4 by an FRT-STOP-FRT cassette. Upon brief induction of FLP recombinase, often via a heat shock promotor, the STOP cassette is excised in only a small proportion of cells and GAL4 and its UAS target transgenes are then expressed in these cells and their descendants (Golic and Lindquist, 1989; Ito et al., 1997; Pignoni and Zipursky, 1997). Importantly, once the FLP-out excision event has occurred, GAL4 is then continuously expressed. To provide an opportunity for more precise temporal regulation of GAL4, we used HACK to convert the widely used Actin5C>CD2>GAL4 FLP-out cassette into its ShineGAL4 equivalent, Actin5C>CD2>GAL4(DBD)::nMag. This ShineGAL4 FLP-out cassette was tested in the fat body by combining it with ubi-pMag::AD, hs-FLP1.22 and a UAS-CD8::GFP reporter (Fig. 3A). In principle, ShineGAL4 FLP-out allows clone induction (in the dark) and GAL4 activation (in the light) to be split into two temporally separable events (Fig. 3B). Consistent with this, we observed clones of one or two fat body cells that had lost the CD2 marker, thus validating excision of the FRT-CD2-FRT stop cassette (Fig. 3C). If larvae were raised in the dark, these fat body clones did not express detectable mGFP but, following 24 h of light exposure, robust mGFP expression was observed (Fig. 3C). This demonstrates that ShineGAL4 FLP-out enables UAS-transgene expression to be temporally regulated within clones. As clone formation and GAL4 activation are independent events, it is thus possible to grow ‘silent clones’ and, only much later, to induce expression from the transgene of interest. To illustrate this, we used a manipulation that is known to decrease cell size in fat body clones – co-expression of two negative regulators of Tor signalling, Tsc1 and Tsc2 (Gig) (Scott et al., 2004). If a standard GAL4 FLP-out cassette is used to generate Tsc1/2-expressing clones in fat body cells before the onset of endoreplicative cycles, then both cell and nuclear size are dramatically reduced (Scott et al., 2004). In contrast, the ShineGAL4 FLP-out system can be used to generate silent fat body clones that only express Tsc1/2 after photoactivation. Hence, in ShineGAL4 FLP-out clones, nuclear volume is unaffected in the dark (0 h) but after light exposure for 24 h or 48 h it progressively decreases (Fig. 3D,E). Hence, the duration of light exposure can be modulated in order to vary the severity of clonal phenotypes without altering their genetic backgrounds. ShineGAL4 FLP-out therefore enables the events of clone induction and GAL4 activation to be separated in time, so that the acute effects of severe phenotypes can be studied in viable clones. In principle, this method should be equally useful in tissues that grow via cell proliferation or endoreplication.
*Photoactivation of ShineGAL4 in FLP-out clones. (A) Schematic of the four genetic elements used to photoactivate ShineGAL4 in FLP-out clones. (B) ShineGAL4 FLP-out clone strategy. FLP excision (+FLP) in the dark generates a silent clone and subsequent optogenetic induction with light switches this to an activated clone, in this case expressing mGFP. (C) Confocal images of fat body cells from larvae at the late-L3 stage showing DAPI-stained nuclei and immunostainings for CD2 and mGFP. FLP-out ShineGAL4 clones exposed to light for 0 h (dark) or 24 h (light) are shown. Clones where the FLP-out cassette has been excised (indicated with red asterisk and dashed line) do not express CD2 in either the dark or light conditions but express mGFP only following photoactivation. Full genotype: hs-FLP1.22; ubi-pMag::AD; Act5C-FRT-CD2-FRT-GAL4(DBD)::nMag[3xP3-Cherry]/UAS-CD8::GFP. Scale bar: 50 µm. (D) Confocal images of spontaneous ShineGAL4 FLP-out clones in the fat body at 0, 24 or 48 h of photoactivation. Clones, visualised by loss of CD2 expression, express mCherry (magenta) and Tsc1/2 at 24 h and 48 h but not at 0 h of light exposure. Full genotype: hs-FLP1.22; ubi-pMag::AD; Act5C-FRT-CD2-FRT-GAL4(DBD)::nMag[3xP3-Cherry], UAS-mCherry/UAS-Tsc1, UAS-Tsc2. (E) Quantification of relative nuclear volumes in control fat body cells (CD2+) and in ShineGAL4 FLP-out cells (CD2−) at 0, 24 and 48 h of photoactivation (0 h: N=19 clones; 24 h: N=63 clones; 48 h: N=31 clones). The boxplots show the median, with whiskers indicating the 25th and 75th percentiles. Photoactivation of Tsc1/2 co-expression decreases nuclear volume compared to surrounding non clonal cells. Each dot represents the relative volume of one nucleus and significant differences between CD2+ and CD2− cells are indicated (**P<0.001, using two-way ANOVA and Tukey HSD post hoc test; ns, not significant).
Together, the findings in this study extend an optogenetic capability to many of the GAL4 drivers commonly used in Drosophila physiology, metabolism and neurobiology research. They also highlight that ShineGAL4 is a robust method applicable to the photoactivation of GAL4 in different genetic contexts. The unprecedented temporal control of gene expression possible with ShineGAL4 has many applications. These include bypassing the early lethality of genetic manipulations and enabling pulse and pulse-chase experiments, such as those used to define the time at which genes are required for a biological process. In addition, large split-GAL4 collections that target very specific neuronal subtypes have recently been generated (Ehrhardt et al., 2025 preprint; Meissner et al., 2023, 2025; Soffers et al., 2025). We suggest that it may now be a worthwhile investment to use large-scale HACK to convert many of these split-GAL4 lines into their Gal4(DBD)::nMag and pMag::AD counterparts.
MATERIALS AND METHODS
Drosophila stocks
Drosophila melanogaster flies were maintained on a standard yeast-based diet and all experiments were performed at 25°C. Most stocks used in this study were obtained from Bloomington Drosophila Stock Center including nSyb-GAL4 (#51635), Mef2-GAL4 (#27390), Mex1-GAL4 (#91368), Uro-GAL4 (#91415), Desat1-GAL4 (#65404), Act-FRT-CD2-FRT-GAL4 (#4780), Cyp4g15-GAL4 (#39103), moody-GAL4 (#90883), repo-GAL4 (*#*7415), elav^C155^-GAL4 (#458) and alrm-GAL4 (#67032). nab-GAL4 was obtained from the Kyoto Drosophila Stock Center (#112622) (Maurange et al., 2008). The other GAL4 driver lines were Lpp-GAL4 (Brankatschk and Eaton, 2010) and MyoA1-GAL4/NP1-GAL4 (Jiang and Edgar, 2009; Zaidman-Remy et al., 2006). ubi-pMagHigh1::p65(AD)/CyO, Dfd::YFP; UAS-CD8::GFP and If/CyO; ubi-pMagHigh1::p65(AD)/TM6B (di Pietro et al., 2021) and nos-Cas9-mSA (Poernbacher et al., 2019 preprint) were gifts from Jean-Paul Vincent (Francis Crick Institute, UK). hs-Cre/CyO; TM2/TM6B was used to remove 3xP3-Cherry marker, and UAS-Tsc1, UAS-Tsc2 compound stock (UAS-Tsc1/2) (Tapon et al., 2001) was used. We used FlyBase (Jenkins et al., 2022; Öztürk-Çolak et al., 2024) release FB2023_05 and later versions to find information on Drosophila genetic elements and stocks. GAL4(DBD)::nMagHigh1 lines generated in this study are listed in Table 1 and Table S1.
HACK conversion of GAL4 into GAL4(DBD)::nMag
GAL4 drivers were converted into ShineGAL4 drivers using a previously described HACK strategy (di Pietro et al., 2021) (Fig. S1A,B). Briefly, ∼100 males of the target GAL4 driver were crossed with ∼200 virgin females carrying a germ-line source of Cas9 (nos-Cas9-mSA). After 3-4 days, ∼500 embryos were collected on agar-grape juice plates and co-injected with a mix of two vectors before pole cell formation. One vector carried a gRNA tandem repeat (pCFD4–GAL4T, 300 ng/µl) to induce a double-strand break in GAL4 and the other contained the Magnet repair template (pTV-GAL4(DBD)::2xnMagHigh1, 500 ng/µl). Surviving mosaic larvae (F1) were collected, raised to adulthood and amplified by intercrossing. In the next generation, five single F2 males carrying 3xP3-Cherry were selected by red eye fluorescence at the adult stage and crossed to ubi-pMagHigh1::p65(AD)/CyO; UAS-CD8::GFP females. Non-CyO F3 progeny larvae were then tested for photoactivation of mGFP expression. Successful candidates were then further processed by crossing to hs-Cre/CyO; TM2/TM6B flies to excise the 3xP3-Cherry cassette and combined with ubi-pMag::p65(AD) to generate stable ShineGAL4 stocks.
Photoactivation of ShineGAL4
Virgin ubi-pMagHigh1::p65(AD)/Cyo; UAS-CD8::GFP females were crossed to males from the newly generated GAL4(DBD)::nMagHigh1 stocks. Eggs were collected in a light-tight cardboard box and allowed to develop for 96 h to the late L3 larval stage. In a room with ambient red light, larvae were then placed in 35 mm diameter Petri dishes containing a ∼3-4 mm layer of food. Half of the Petri dishes were exposed to white light using spotlight illuminators on the brightest setting (Leica, LED3000 SLI) for 4-24 h (details provided in figure legends), while the other half were kept shielded from light for the same amount of time (dark condition). Larvae of the correct genotype were subsequently selected, dissected in phosphate buffered saline (PBS) and fixed in 4% paraformaldehyde/PBS for 30 min before proceeding to immunostaining.
Immunostaining
After fixation, tissues were washed and permeabilised with three successive 0.2% Triton X-100/PBS (PBT) rinses and blocked in 4% normal goat serum in PBT. Tissues were then incubated with 1:400 rabbit anti-GFP (Invitrogen, A11122) in PBT overnight at 4°C. After three PBT washes over 15 min, tissues were incubated in 1:400 goat anti-rabbit Alexa-Fluor488 (Invitrogen, A11034) in PBT. For myocyte staining, 1:400 phalloidin-iFluor647 (Abcam, #ab176759) was also added at this step. Three PBT washes over 15 min were then repeated, and tissues incubated 1:10,000 DAPI (Invitrogen, #D1306) in PBS, briefly rinsed in PBS and mounted in anti-fade medium (Vectashield, #H-1000). Imaging was carried out using a STELLARIS 8 confocal microscope (Leica), using either 20× (HC APO CS2 20×/0.75, Leica) or 40× (HC PL APO CS2 40×/1.30 oil, Leica) objectives. For image acquisition, laser parameters were set using the light-induced sample and kept the same for the dark sample. Image contrast was slightly enhanced using ImageJ (v1.54i) for presentation purposes, but adjustments were identical between the light and dark condition. Figures were prepared using Adobe Illustrator version 29.3.
Quantifications of native GFP fluorescence and nuclear volume
All image quantifications were performed in ImageJ (v1.54i). For native GFP fluorescence, confocal stacks were processed with maximum-intensity projections, and a manually drawn brain region of interest (ROI) was thresholded using a lower bound determined from specimens raised in the dark to exclude background. Total pixel intensity above this threshold was quantified and normalised to the total area of the brain ROI. Values are expressed relative to the 24-h photoactivated sample.
For nuclear quantifications, confocal stacks were converted to maximum intensity projections, binarised, hole-filled, and then separated by watershed. Particles larger than 20 µm² were analysed using the ImageJ Analyze Particles function, and nuclear volume was calculated from the extracted major and minor axes assuming an ellipsoid shape. Values are expressed relative to the mean CD2^+^ nuclear volume of each sample.
Photoactivation of ShineGAL4 in FLP-out clones
An Act5C>FRT-CD2-FRT-GAL4 driver, previously developed for generating FLP-out clones (Pignoni and Zipursky, 1997) was converted into a GAL4(DBD)::nMagHigh1 counterpart using the same HACK strategy described for the other GAL4 drivers. Fat body clones were induced spontaneously at room temperature (25°C) in larvae of the genotype hs-FLP^1.22^; ubi-pMagHigh1::p65(AD); Act5C-FRT-CD2-FRT-GAL4(DBD)::nMagHigh1/UAS-CD8::GFP (Britton et al., 2002). ShineGAL4 was then photoactivated for 24 or 48 h as described above. Fat bodies were subsequently dissected in PBS and immunostained according to the procedure above but also including 1:100 mouse PE or FITC-conjugated anti-rat CD2 (BioLegend, #201305 and #201303) to score for FLP-out excision of the FRT-CD2-FRT stop cassette.
Supplementary Material
10.1242/develop.204981_sup1Supplementary information
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Brand, A. H. and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401-415. 10.1242/dev.118.2.4018223268 · doi ↗ · pubmed ↗
- 2Brankatschk, M. and Eaton, S. (2010). Lipoprotein particles cross the blood-brain barrier in Drosophila. J. Neurosci. 30, 10441-10447. 10.1523/JNEUROSCI.5943-09.201020685986 PMC 6634680 · doi ↗ · pubmed ↗
- 3Britton, J. S., Lockwood, W. K., Li, L., Cohen, S. M. and Edgar, B. A. (2002). Drosophila's insulin/PI 3-kinase pathway coordinates cellular metabolism with nutritional conditions. Dev. Cell 2, 239-249. 10.1016/S 1534-5807(02)00117-X 11832249 · doi ↗ · pubmed ↗
- 4Burcin, M. M., Schiedner, G., Kochanek, S., Tsai, S. Y. and O'Malley, B. W. (1999). Adenovirus-mediated regulable target gene expression in vivo. Proc. Natl. Acad. Sci. USA 96, 355-360. 10.1073/pnas.96.2.3559892637 PMC 15140 · doi ↗ · pubmed ↗
- 5Chen, H. M., Yao, X., Ren, Q., Chang, C. C., Liu, L. Y., Miyares, R. L. and Lee, T. (2020). Enhanced Golic+: highly effective CRISPR gene targeting and transgene HAC King in Drosophila. Development 147, dev 181974. 10.1242/dev.18197432467238 · doi ↗ · pubmed ↗
- 6di Pietro, F., Herszterg, S., Huang, A., Bosveld, F., Alexandre, C., Sancere, L., Pelletier, S., Joudat, A., Kapoor, V., Vincent, J. P. et al. (2021). Rapid and robust optogenetic control of gene expression in Drosophila. Dev. Cell 56, 3393-3404.e 7. 10.1016/j.devcel.2021.11.01634879263 PMC 8693864 · doi ↗ · pubmed ↗
- 7Ehrhardt, E., Whitehead, S. C., Namiki, S., Minegishi, R., Siwanowicz, I., Feng, K., Otsuna, H., Fly Light Project, T., Meissner, G. W., Stern, D. et al. (2025). Single-cell type analysis of wing premotor circuits in the ventral nerve cord of Drosophila melanogaster. bio Rxiv. 10.1101/2023.05.31.542897 · doi ↗
- 8Golic, K. G. and Lindquist, S. (1989). The FLP recombinase of yeast catalyzes site-specific recombination in the Drosophila genome. Cell 59, 499-509. 10.1016/0092-8674(89)90033-02509077 · doi ↗ · pubmed ↗