Coupling optogenetics and light-sheet microscopy, a method to study signal transduction in vivo

Optogenetics allows precise, fast and reversible intervention in biological processes. Light-sheet microscopy allows observation of the full course of embryonic development from egg to larva. Bringing the two approaches together allows unparalleled precision into the temporal regulation of signaling pathways and cellular processes in vivo. To develop this method, we investigated the regulation of canonical Wnt signaling during anterior-posterior patterning of the Drosophila embryonic epidermis. Cryptochrome 2 (CRY2) from Arabidopsis Thaliana was fused to mCherry fluorescent protein and Drosophila β–catenin to form an easy to visualize optogenetic switch. Blue light illumination caused oligomerization of the fusion protein and inhibited downstream Wnt signaling in vitro and in vivo. Temporal inactivation of β–catenin confirmed that Wnt signaling is required not only for Drosophila pattern formation, but also for maintenance later in development. We anticipate that this method will be easily extendable to other developmental signaling pathways and many other experimental systems.


Dissections of signaling pathways and their downstream processes have centered on in vitro
cell culture models and in vivo model organism studies using small molecule, knockdown and overexpression approaches. Although these approaches have been very useful, they do not lend themselves easily to the study of the spatial and temporal regulation of cell signaling pathways and cellular processes in the dynamic environment of a living organism. Normal genetic crosses using mutations and RNAi techniques are limited in their applications as they result in irreversible knockout or knockdown of a gene, which can be lethal or incomplete 1 .
approaches have been developed 9 . In addition, new, stable fluorescent proteins allow imaging of various cellular processes at the same time 10 .
The Wnt pathway is a highly conserved signal transduction pathway functioning in development and disease in many organisms 11 . In Drosophila, Wnt1 (Wingless or Wg) is a segment polarity gene that defines anterior-posterior patterning of the epidermis 12 .
Canonical Wnt signaling leads to inhibition of β -catenin (Armadillo or Arm in Drosophila) degradation, which subsequently activates downstream Wnt target genes in the nucleus. As β-catenin protein levels increase, it enters the nucleus where it regulates transcription of target genes by associating with the transcription factor TCF 13,14 . This basic signaling mechanism is widely conserved, and critical in embryogenesis, but whether β -catenin signaling is needed in a constant widespread manner, versus in a temporally and/or spatially restricted manner was not known, because the tools needed to address this question were lacking. Here we describe our method for applying CRY2 protein fusions to investigate the spatial and temporal regulation of β -catenin protein in the developing embryo.
was not reversible. To test this, we performed experiments where embryos were placed in the light sheet microscope and illuminated with blue light every 2.5, 5 or 10 minutes.
arm XM19 mutants expressing the Arm-CRY2-mCh construct illuminated every 2.5 or 5 minutes did not survive to the end of embryogenesis, but those illuminated only once every 10 minutes hatched. We propose that the persistence of puncta represents the cells' inability to degrade aggregated Arm protein quickly, but this does not affect Wnt signaling as enough Arm protein is made during the dark phase to compensate for the aggregates as long as 10 minutes are allowed to elapse. The system, therefore, is reversible but may require new protein synthesis rather than dis-aggregation.

Temporal regulation of Wnt signaling during development in arm XM19 embryos
Having established the efficacy of the Arm-CRY2-mCh transgene in embryos, we next performed a temporal experiment to answer the biological question of when Wnt signaling is required. Segment polarity genes are expressed downstream of pair rule genes early in development 22 . These genes turn on Wg expression 23 through a regulatory loop with Engrailed 24,25 . Experiments with a temperature sensitive allele of wg established four phases of Wg signaling in establishment and maintenance of embryonic patterning 26 . These can be subdivided into early en activation and later shaven baby repression in combination with EGF signals 24,27,28 . We used eggs exposed to light at different developmental stages 29 , and observed the cuticle phenotypes. Staging was approximate as the experiment was carried out at 21°C. Exposure to light from stages 1 to 11 (From cleavage stages to Germ band elongation) showed a variable lawn of denticles similar to wg mutants (Fig. 3A). Strong Wnt pathway loss of function embryos show a lawn of denticles and are much smaller than wildtype. Restoring some Wnt pathway function rescues the small size phenotype 12,30 . The lawn of denticles began getting broader (size of the embryo increased) when light was applied at stage 12 (onset germ band retraction) with some distance developing between the denticle bands ( Fig. 3A, stage 13 germ band retraction). This experiment showed that Arm-CRY2-mCh could be used to recapitulate previous findings with wg ts that showed effects into dorsal closure stages 26 .

Spatial perturbation of Wnt signaling during development in arm XM19 embryos
We next wanted to test the spatial resolution of this approach using light sheet microscopy as a blue light source as well as a recorder for live imaging. Light sheet microscopy is not an ideal tool for this experiment as the light is not delivered as a point (For a better approach to spatial resolution see 31 ). Half of an embryo overexpressing the Arm-CRY2-mCh transgene was illuminated with blue light and the whole embryo imaged in the red channel. The region of the embryo with blue light illumination showed similar defects as in the earlier results with the unilluminated half remaining more wild-type ( can be used to make and image mosaic embryos.

Light-responsive properties of Arm-CRY2-mCh can modulate Wnt signaling in HEK293
We have shown that Arm-CRY2-mCh functions in insect cells and in Drosophila embryos.
We next tested if this construct could be applied to other experimental systems by looking at whether it could be used to modulate Wnt signaling in mammalian cells. We used HEK293 (Human embryonic kidney cells) stably transfected with the Wnt3a gene and TOPFlash vector (HEK293-Wnt3a-TOPFlash). Cells transfected with the construct showed fluorescent puncta formation upon blue light illumination (Fig. 4A). The puncta dispersed rapidly upon withdrawal of blue light (Fig. 4A).
The effect of inducible Arm-CRY2-mCh clustering on Wnt signaling in mammalian cells was quantified using a TOPFlash reporter assay. Overexpression of Arm-CRY2-mCh plasmid in HEK293-Wnt3a-TOPFlash cells in the absence of light significantly increased TOPflash activity relative to control transfected cultures (Fig. 4B). This activation was inhibited when transfected cells were exposed to light. Our data suggest that oligomerization of the Arm-CRY2-mCh construct upon illumination with light could be applied to other experimental systems.

Discussion
Classical genetics tools allow tissue-specific manipulation of gene expression, but not precise temporal or spatial manipulation. Optogenetic tools can address this problem by allowing spatial and temporal regulation in a rapid, precise, and reversible manner 2,32,33 . Various optogenetic systems have been described to date such as the Cryptochrome 2 (CRY2), Phytochrome B (PHYB), LOV and Dronpa systems. However, only the CRY2 and LOV systems allow optogenetic control at a lower wavelength with live imaging at a higher wavelength. Both systems are activated within a few seconds, however, the CRY2 system has a slower turn off speed of approximately 5 minutes as compared to tens of seconds or minutes in the case of LOV 16 . This makes the CRY2 system ideal for live imaging with lower phototoxicity. Our study is the first to combine optogenetics with light-sheet microscopy for noninvasive, temporal and spatial control of cell signaling in vivo. Using these tools, we investigated the spatial and temporal regulation of β−catenin and Wnt signaling in vivo and in vitro proving the efficacy of this method. Wnt's role in segment polarity is well known 34,35 , and we observed the same phenotypes by inactivating Wnt signaling during early stages of development (stage 1 to 11). Light-sheet microscopy is not designed for single cell conversion of optogenetic switches 31 , but can be applied to sections of the embryo to study problems such as left-right asymmetry and anterior-posterior patterning by illuminating different regions of the embryo. Other microscopy approaches can be used to prevent light scattering 9 , where the optical molecule could be activated in unintended parts of the embryo, but the advantage of light sheet microscopy is that it allows whole embryo imaging over the entire course of the 24 hour Drosophila embryogenesis.
Wnt signaling in segment polarity is well studied, but many additional functions have been observed 36 . Recently, 37 reported that intestinal formation during embryo development was disrupted in embryos treated with ionomycin, an inhibitor of Wnt/β-catenin signaling. By coupling optogenetics with live imaging, we observed a similar defect in intestinal development in Arm-CRY2-mCh expressing arm XM19 mutant embryos. Our experiment with spatial inactivation of Arm-CRY2-mCh in only a specified region of the embryo, also confirmed these findings. We see the application of optogenetics with light-sheet microscopy will allow the exploration of many more Wnt pathway effects. The main limitation appears to be the buildup of CRY2 puncta that do not disaggregate. In our experiments this leads to a buildup of mCh fluorescence in the later stages especially in hemocyte cells making imaging in the red channel difficult at late stages. We anticipate that this can be overcome by using separate fluorescent proteins for imaging and for optogenetics.
Our optogenetics approach is a non-invasive and versatile tool that allows for manipulation as well as live tracking of protein activity to interrogate complex phenotypes in vivo 38 . We foresee applying this tool to cell culture, organoid studies, vertebrate development in zebrafish, and many other systems 39 . It provides an alternative to small molecule studies which can only be performed on druggable targets. We have made a human β−catenin-CRY2 version that functions similarly, and are applying the optogenetic methods to study adhesion by using embryos with a stronger loss of function arm O43A01 allele that disrupts junctions but is rescued by Arm-CRY2-mCh 40-42 . Further, we find that NFκB/Dorsal and Erk pathways are also effectively studied in this way as our Dorsal-CRY2 and Ras-CRY2 constructs function well in tissue culture and hopefully soon in flies.

Crosses and expression of UAS constructs
The dominant female sterile technique was used to generate maternally mutant eggs 21 .
Please see Flybase for details on mutants used (flybase.bio.indiana.edu). For mis-expression experiments, the ArmGAL4 2 nd chromosome was used. FRT 101 was used to generate the X chromosome mutant.
The following cross was conducted to generate germline clone embryos: y -, arm XM19 FRT101/ovoD1 FRT101; Arm-Gal4 x w -, y + ; UAS-Arm-CRY2-mCh Embryos were imaged in the light sheet microscope showing mCh expression, and were further cuticle prepped after imaging to confirm the genotype and phenotype as described for y -, arm XM19 mutation, an assay based on the pigmentation of the denticles 40,41,48 .
The following cross was conducted to generate zygotic mutant rescue flies: In this cross, all males receive the X-chromosome from their mothers so only balancer (FM6, Bar) males are expected unless the transgene can rescue the lethality of arm XM19 . Identical crosses were set up with three kept entirely in the dark and three under normal laboratory light conditions for two weeks. The number and phenotype of males from all the crosses was scored.

Light-sheet microscopy
Embryos were dechorionated in bleach, washed in water and dried on a paper towel. They were then embedded into a capillary containing 1% agarose, low gelling temperature Type VII-A (Sigma) dissolved in water, using a fine wire such that the embryo was upright with the anterior-posterior axis aligned with the axis or perpendicular to the axis of the glass capillary. For imaging, the agarose was pushed out of the capillary and the sample was suspended freely in the sample chamber containing water as our lightsheet set-up utilized a water immersion objective.
Live imaging of Drosophila embryos was carried out on the light-sheet Z.1 fluorescence microscope (Carl Zeiss, Germany) with the Lightsheet Z.1 10x/0.2 Illumination Optics and W Plan-Apochromat 20×/1.0 UV-VIS detection objective (Carl Zeiss, Germany). The lightsheet microscope was equipped with the Lightsheet Z.1 detection module "Standard", 30mW 488nm and 20mW 561nm solid state lasers with BP505-545 and LP585 emission filters respectively. Whole embryos (unless otherwise stated) were imaged using dual-side illumination by a light-sheet modulated into a pivot scan mode. The 488nm laser was used at 6% power with 7.5ms exposure time and the 561nm laser was used at 13% laser power with 12.5ms exposure time. Since the oligomerization is reversible with removal of blue light, the whole embryo was excited with both laser lines every 2.5 min in Z-stack mode at 1μm step size for 24 hrs. For control embryos not exposed to blue light, the same settings were used but with illumination using only the 561nm laser line. All single plane illumination (SPIM) data was saved in the LSM format and processed using the ZEN 2014 SP1 software (Carl Zeiss, Germany). Background fluorescence was automatically calculated by the ZEN 2014 software using the background correction function (ZEN Software Guide, Carl Zeiss, Germany). Z-stacks that were deep in the embryo and out of focus were removed and maximum intensity projections of z-stacks in focus were generated for each embryo 49 .
Images for figures were processed and assembled using Photoshop and Illustrator programs from Adobe.

Antibodies and Immunofluorescence
Embryos were dechorionated in bleach and fixed with heptane/4% formaldehyde in phosphate buffer (0.1M NaPO4 pH 7.4) for 20 minutes 40 . The aqueous phase was removed and an equal amount of methanol was added to devitellinize the embryos. Antibody stainings were done in PBT (PBS, 0.1% Triton X-100, 1% bovine serum albumin, 0.1% Azide). The

anti-Engrailed antibody mAbs (4D9) was obtained from the Developmental Studies
Hybridoma Bank, created by the NICHD of the NIH and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242 and used at a dilution of 1:10. Alexa Fluor 488-anti-mouse secondary antibody was used (ThermoFisher Scientific, USA) at a 1:1000 dilution. Embryos were mounted in Aquapolymount ® (Polysciences, Inc.). Images were acquired on the Zeiss AxioImager Z1 with Apotome (Carl Zeiss, Germany) and images were processed using the AxioVision Rel. 4.8 (Carl Zeiss, Germany).

Live-cell imaging.
Time-lapse microscopy of activated CRY2 fusions in S2R+ and endogenously expressing WNT3A in a HEK 293 cell line that contains an integrated TOPFlash reporter (A kind gift from Dr Alan Prem Kumar, Cancer Science Institute of Singapore) was performed using a Zeiss AxioImager Z1 microscope (Carl Zeiss, Germany). Clustering visualization was carried out at room temperature. Blue light exposure and mCherry imaging were performed simultaneously by imaging in both 488nm and 561nm laser channels. Images were acquired and processed using the AxioVision Rel. 4.8 (Carl Zeiss, Germany). Puncta counts were determined using the Cell Counter plugin in ImageJ.

TOPflash assay
TOPflash luciferase assays (TCF/LEF reporter assays) were performed to assess the effect of light on canonical Wnt-signaling in cells transfected with the Arm-CRY2-mCh plasmid.
S2R+ cells were co-transfected with TOPflash 13 , Renilla luciferase-Pol III (Renilla luciferase-Pol III was a gift from Norbert Perrimon (Addgene plasmid # 37380)) 50 and the pAW-Arm-CRY2-mCh plasmid using lipofectamine 3000 (ThermoFisher Scientific) according to the manufacturer's instructions. In the case of the HEK 293 cell line endogenously expressing WNT3A and the TOPFlash reporter, Renilla luciferase expressing vector (pRL-CMV, Promega, USA) and the pDEST40-Arm-CRY2-mCh plasmid were transfected using lipofectamine 3000 (ThermoFisher Scientific) according to the manufacturer's instructions. Cell lysates were prepared 24 hours after transfection and luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega) according to the manufacturer's instructions. The relative TOPflash luciferase activity was measured using the ratio of firefly/renilla luciferase activity and the data was presented as mean ± SD. Statistical analyses were performed using Student's t test unless the data showed evidence of unequal standard deviation (F-test, p < 0.05). In these cases, the Mann-Whitney nonparametric test was used.

RNA Extraction, cDNA Synthesis and qPCR
Mixed stage and genotype embryos, 0 to 16 hours after deposition, were exposed to light or no light for 6 hours, dechorionated in bleach, washed in water and collected. Total RNA was extracted for each treatment using the RNeasy Mini Kit (Qiagen) as per the manufacturer's protocol. Total RNA concentration and purity was measured using the Cytation 3 Cell Imaging Multi-Mode Reader (BioTek). One µg of total RNA was reverse transcribed in a 20µl reaction volume using the QuantiTect reverse transcription kit (Qiagen) according to the manufacturer's protocol. Gene specific primer sequences were obtained from Fly Primer Bank 51 . Quantification of mRNA was performed using SYBR® Green Assay (Thermo Fisher Scientific) on the PikoReal™ Real-Time PCR System (Thermo Fisher Scientific) and a PCR product dissociation curve was generated to ensure specificity of amplification. For expression analysis, the mRNA data were normalized to the endogenous control, RPL32, followed by calibration to cultures not exposed to light using relative quantification (2 −ΔΔCT ).
Results were generated from 3 technical replicates for each mRNA. The average relative expression ± standard deviation (SD) was determined. Statistical analyses were performed using Student's t test unless the data showed evidence of unequal standard deviation (F-test, p < 0.05). In these cases, the Mann-Whitney nonparametric test was used.
Contributions P.K. and N.S.T. designed and performed the experiments. N.S.T., P.K. and T.E.M. wrote the paper. All authors reviewed manuscript.

Competing interests
The authors declare no competing financial interests.   Arm-CRY2-mCh were collected and exposed to light at various developmental stages.
Cuticle preparations were carried out after light exposure. Exposure to light earlier (stage 1 to 11) during development resulted in a wg mutant phenotype. The denticle patterning began spreading out thereafter until stage 13 after which a mostly wild-type phenotype was  TOPFlash activity after exposure to light. Graph is representative of three independent experiments and the average of three replicates (mean ± SD). The variance was found to be equal as determined by the F-test (p>0.05). Statistical significance relative to control samples was determined using the Student's t-test. **p<0.01 relative to control, ## p<0.01.

Supplementary Video 1. Arm-CRY2-mCh puncta formation in Drosophila embryos
overexpressing Arm-CRY2-mCh in a wild-type background upon exposure to 488nm laser light. The whole embryo was illuminated with 488nm laser light at 2.5 minute intervals for half an hour to activate oligomerization. Puncta formation observed during illumination was reversible and oligomerization could be induced repeatedly.

Supplementary Video 2A.
Arm-CRY2-mCh overexpressing arm XM19 mutant embryos not exposed to light showed diffuse localization of Arm-CRY2-mCh and normal embryonic development.
Supplementary Video 2B. Arm-CRY2-mCh overexpressing arm XM19 mutant embryos exposed to 488nm light showed distinct puncta formation of Arm-CRY2-mCh and various developmental defects as compared to embryos not exposed to light. The posterior region of the embryo (with blue light illumination) showed distinct clustering and incomplete germ band retraction.  T  i  s  c  h  e  r  ,  D  .  &  W  e  i  n  e  r  ,  O  .  D  .  I  l  l  u  m  i  n  a  t  i  n  g  c  e  l  l  s  i  g  n  a  l  l  i  n  g  w  i  t  h  o  p  t  o  g  e  n  e  t  i  c  t  o  o  l  s  .   N  a  t  R  e  v  M  o  l  C  e  l  l  B  i  o  l   1  5   ,  5  5  1  -5  5  8  ,  d  o  i  :  1  0  .  1  0  3  8  /  n  r  m  3  8  3  7  (  2  0  1  4  )  .   1  7  W  i  e  s  c  h  a  u  s  ,  E  .  ,  N  u  s  s  l  e  i  n  -V  o  l  h  a  r  d  ,  C  .  &  J  u  r  g  e  n  s  ,  G  .  M  u  t  a  t  i  o  n  s  a  f  f  e  c  t  i  n  g  t  h  e  p  a  t  t  e  r  n  o  f  t  h  e   l  a  r  v  a  l  c  u  t  i  c  l  e  i  n  D  r  o  s  o  p  h  i  l  a  m  e  l  a  n  o  g  a  s  t  e