XB-ART-57320Sci Rep September 2, 2020; 10 (1): 14437.
Diffusible GRAPHIC to visualize morphology of cells after specific cell-cell contact.
The ability to identify specific cell-cell contact in the highly heterogeneous mammalian body is crucial to revealing precise control of the body plan and correct function. To visualize local connections, we previously developed a genetically encoded fluorescent indicator, GRAPHIC, which labels cell-cell contacts by restricting the reconstituted green fluorescent protein (GFP) signal to the contact site. Here, we modify GRAPHIC to give the reconstituted GFP motility within the membrane, to detect cells that make contact with other specific cells. Removal of leucine zipper domains, located between the split GFP fragment and glycophosphatidylinositol anchor domain, allowed GFP reconstituted at the contact site to diffuse throughout the entire plasma membrane, revealing cell morphology. Further, depending on the structural spacers employed, the reconstituted GFP could be selectively targeted to N terminal (NT)- or C terminal (CT)-probe-expressing cells. Using these novel constructs, we demonstrated that we can specifically label NT-probe-expressing cells that made contact with CT-probe-expressing cells in an epithelial cell culture and in Xenopus 8-cell-stage blastomeres. Moreover, we showed that diffusible GRAPHIC (dGRAPHIC) can be used in neuronal circuits to trace neurons that make contact to reveal a connection map. Finally, application in the developing brain demonstrated that the dGRAPHIC signal remained on neurons that had transient contacts during circuit development to reveal the contact history. Altogether, dGRAPHIC is a unique probe that can visualize cells that made specific cell-cell contact.
PubMed ID: 32879377
PMC ID: PMC7468259
Article link: Sci Rep
Genes referenced: gpi h2bc21 nucb1
Article Images: [+] show captions
|Figure 1. Distribution of reconstituted GFP is dependent on GRAPHIC probe molecular structure. A red nucleus (labeled by H2B-mCherry) indicates an NT cell, and a blue nucleus (labeled by H2B-Azurite) indicates a CT cell. (A) GRAPHIC: NT probe containing an acidic leucine zipper domain (LZA) and CT probe containing a basic leucine zipper domain (LZB). Reconstituted GFP was observed only at the boundary between NT and CT cells. White box indicates position of higher magnification view. (B) dGRAPHIC: Leucine zipper domain was removed from both NT and CT probes. dGRAPHIC showed almost all reconstituted GFP signals on NT cells, compared with CT cells. (C) A probe pair consisting of NT and CT probes with 3 spacer domains (15 aa × 3) inserted between the GFP-CT fragment and GPI anchor domain. Reconstituted GFP was equally observed on both NT and CT cells. (D) Combination of NT and CT probes with 4 spacer domains (15 aa × 4). Reconstituted GFP was detected predominantly on CT cells. Scale bar, 40 µm and for higher magnification 20 μm.|
|Figure 2. Spatiotemporal characteristics of dGRAPHIC. (A) Confocal images reveal subcellular distribution of reconstituted GFP in co-cultured NT-probe-expressing and CT-probe-expressing LLCPK1 cells. Schema of z-positions of confocal images. (B) Stacked confocal images (z = 1.2 μm) at apical surface (top), middle position (center), and basal membrane (bottom). (C) Full stacked image (z = 7.2 μm) of reconstituted GFP, nuclei of NT-cells (red), and nuclei of CT-cells (blue). Scale bar, 20 µm. (D) Time lapse images for the generation of reconstituted GFP signal between NT-LLCPK1 cells (red nuclei) and CT-LLCPK1 cells (blue nuclei). Upper panels are bright field (differential interference contrast) images of the bottom fluorescence images. The reconstituted GFP signal was generated at cell–cell contact sites and gradually spread out over the whole plasma membrane of contacted NT cells (red nuclei). Scale bar, 100 μm.|
|Figure 3. dGRAPHIC application in Xenopus blastomere. (A) Schema of mRNA injection into 8-cell-stage Xenopus blastomeres. The mRNA of the NT probe is injected in the right animal-smaller blastomere of the Xenopus 8-cell stage, and the mRNAs of the CT probe and nuc-mCherry are injected together into the left animal-smaller blastomere. (B) Embryos are grown to the early neurula stage. Cells expressing the CT probe (red nuclei) are present only within the left half of the neural plate, while GFP signals can be detected only in the right half of the neural plate. Scale bar, 100 μm.|
|Figure 4. dGRAPHIC labels neurons in vitro and in vivo. (A) Experimental schema for in vitro dGRAPHIC. Each probe plasmid was electroporated into hippocampal neurons in individual cuvettes, and electroporated neurons were mixed and co-cultured. NT neurons show red fluorescence only in the nucleus and CT neurons show red fluorescence throughout the cell. (B) Reconstituted GFP in hippocampal cultures. In co-cultured hippocampal neurons, the reconstituted GFP signal was observed only in NT neurons (yellow arrows), which was equally distributed on soma, dendrites, and axons. No GFP signal was present in CT neurons (red arrows). Scale bar, 100 μm. (C) Experimental schema for in vivo dGRAPHIC. AAVs encoding NT probes (co-expression with red fluorescent nuclear label) and CT probes (co-expression with cytosolic mCherry) were stereotaxically injected into the S1 cortex and VB, respectively. After several weeks, the injected brains were sectioned and observed. (D–F) In coronal sections, the GFP and mCherry signals were observed in the cortex (D), white matter (E), and VB thalamus area (F). Scale bar, 500 μm.|
|Figure 5. dGRAPHIC can reconstitute GFP from synaptic contact. (A) In utero electroporation was performed at E13.5 to express NT probes (red nuclear label) in cortical layer IV. (B) After the electroporated mice had grown to adult age (about 2 months old), AAV encoding the CT probe (cytosolic mCherry) was stereotaxically injected into the VB. (C) Lower magnification images of a sample brain several weeks after injection. Reconstituted GFP and nucleic mCherry signals were observed in layer IV. Scale bar, 50 μm. (D) Cytosolic mCherry signal was observed in VB neurons, indicating strong expression of the CT probe. Scale bar, 50 μm. (E) Higher-magnification confocal images of layer IV neurons. The reconstituted GFP signal delineates a cortical spiny stellate cell body and dendrites (white arrowheads), which receive input from VB thalamocortical axons. There are also GFP-positive fibers in layer IV, which are processes of other cortical layer IV neurons that received input from VB axons. Scale bar, 100 μm.|
|Figure 6. Visualization of past thalamocortical connections in the developing mouse brain. (A) Experimental schema. In utero electroporation in the ventricular zone of diencephalon (developing thalamus, blue arrows) was performed at E11.5 to express the CT probe (co-expressed with blue fluorescent nuclear label). Two days later, in utero electroporation in the ventricular zone of telencephalon (pink arrows) was performed at E13.5 to express the NT probe (co-expressed with red nucleus label) in cortical layer IV. (B) Schema of thalamocortical axon projections to cortical layer IV in adult brain. (C) GFP and mCherry images of the cortical S1 area. Scattered GFP signals are not restricted to layer IV but observed in several cortical layers (white arrowheads). Scale bar, 200 μm. (D) CT-probe expression in VB was observed by a blue nuclear signal. (E) Higher-magnification images of white dotted box in C. Dotted white circles in mCherry image indicate nuclei of GFP-positive neurons. Scale bar, 100 μm.|
References [+] :
Callaway, Transneuronal circuit tracing with neurotropic viruses. 2009, Pubmed