We describe here the use of a pH-sensitive green fluorescent protein variant, pHluorin, to study the spatio-temporal dynamics of axon guidance receptors trafficking at the cell surface. The pHluorin-tagged receptor is expressed both in cell culture and in vivo, using electroporation of the chick embryo.
During development, axon guidance receptors play a crucial role in regulating axons sensitivity to both attractive and repulsive cues. Indeed, activation of the guidance receptors is the first step of the signaling mechanisms allowing axon tips, the growth cones, to respond to the ligands. As such, the modulation of their availability at the cell surface is one of the mechanisms that participate in setting the growth cone sensitivity. We describe here a method to precisely visualize the spatio-temporal cell surface dynamics of an axon guidance receptor both in vitro and in vivo in the developing chick spinal cord. We took advantage of the pH-dependent fluorescence property of a green fluorescent protein (GFP) variant to specifically detect the fraction of the axon guidance receptor that is addressed to the plasma membrane. We first describe the in vitro validation of such pH-dependent constructs and we further detail their use in vivo, in the chick spinal chord, to assess the spatio-temporal dynamics of the axon guidance receptor of interest.
During their navigation, axons integrate multiple environmental cues that guide them towards their target. These cues activate guidance receptors at the surface of axon terminals, the growth cones, which in turn initiate an appropriate signaling pathway. Thus, the temporal and spatial regulation of the cell surface distribution of the receptors is critical to set the sensitivity of the growth cone1. In this context, midline crossing by commissural axons is an excellent model to investigate the regulation of receptor cell surface levels. In the developing spinal cord, commissural axons are initially attracted towards the ventral floor plate where they cross the midline. After crossing, they lose their responsiveness to the floor plate attractants and gain response to floor plate repellents so that they can exit the floor plate and navigate towards their final destination in the contralateral side of the nervous system2,3. Regulation of receptor availability at the growth cone surface is one of the mechanisms underlying the switch of responsiveness to midline cues4,5. Thus, selective monitoring of the receptors present at the plasma membrane of growth cones is of prime importance. We describe here a method based on the pH-dependent fluorescence property of a green fluorescent protein (GFP) variant to specifically visualize the axon guidance receptors that are addressed to the plasma membrane in vitro and in vivo, in the developing chick spinal cord.
Rothman and colleagues engineered by point mutations pH-sensitive variants of GFP including the ecliptic pHluorin6. Ecliptic pHluorin has the property of being nonfluorescent when exposed to acidic pH (<6), while being fluorescent at neutral pH. This allows distinguishing nonfluorescent receptors localized in intracellular acidic compartments (i.e. endosomes, trafficking vesicles) from fluorescent receptors incorporated to the plasma membrane and thus exposed to the extracellular neutral pH7. We took advantage of this to monitor the plasma membrane localization of plexinA1, an axon guidance receptor mediating the growth cone response to the midline repellent semaphorin 3B5 (Figure 1A). We describe here the in vitro characterization of a pHluorin-plexinA1 construct, along with in ovo electroporation8-10 of this construct in the developing chick spinal cord followed by the microscopic analysis of cryosections which enable to follow the axon guidance receptor dynamics in vivo with both spatial and temporal resolutions.
1. Cloning Strategy to Tag PlexinA1 Receptor with pHluorin
2. Characterization of pHluorin-tagged Receptor In vitro in COS7 Cells
The ability of the fusion protein to reach the plasma membrane and its reversible loss of fluorescence as pH is lowered can be confirmed using the following procedure.
3. In ovo Electroporation of pHluorin-plexinA1 Construct
4. Embryos Embedding and Cryosectioning
5. Microscopic Analysis of Cryosections
Representative images of pHluorin-plexinA1 and eGFP expression in the chick embryo spinal cord are shown in Figure 4.
Figure 1. A. Scheme of the pHluorin-plexinA1 fluorescence properties in a cellular context. PHluorin is nonfluorescent in intracellular compartments where the pH is acidic (<6) such as in trafficking vesicules or in endosomes and is fluorescent when exposed to the extracellular medium where the pH is neutral. This allows visualizing only the cell surface pool of the pHluorin-plexinA1 receptor. B. Map of the pBK-CMV-pHluorin-plexinA1 cloning site. Mouse receptor plexinA1 expressing vector was used as a backbone. This vector was engineered to achieve efficient VSV-tagged receptor insertion at the plasma membrane. To do so, semaphorin 3a peptide signal and cleavage site were fused upstream of the VSV-tagged receptor ablated from its own signal peptide and cleavage site. By PCR, pHluorin sequence was inserted in frame between VSV-tag and PlxnA1 coding sequences using NruI/Kpn2I restriction sites. Single restriction sites are indicated for designing cloning strategy of other receptors. The scheme is adapted from bioinformatic software.
Figure 2. Characterization of the pHluorin-plexinA1 construct pH-dependent fluorescent properties in COS7 cells. A. COS7 cells transfected with the pHluorin-plexinA1 construct where observed with a confocal microscope allowing visualization of pHluorin-plexinA1 fluorescence at the plasma membrane. Scale bar: 20 µM. B. COS7 cells transfected with the pHluorin-plexinA1 construct where observed through live imaging in a pH 7.4 culture medium, after acidification of the culture medium up to pH 5.5 and after restoration of pH 7.4 in the culture medium. Cell outlines are indicated with a dashed line on GFP fluorescence acquisition. Scale bar: 10 µM. Click here to view larger image.
Figure 3. In ovo electroporation procedure. A. 1. Albumen removal from the largest side of the shell. 2. Windowing the egg shell to access the embryo. B. DNA/Fast Green mix injection in the neural tube. C. Placing electrodes on either sides of the neural tube, avoiding the heart and big vessels, and electroporation of the chick embryo.
Figure 4. Microscopic analysis of chick neural tube cryosections after pHluorin-plexinA1 electroporation. A comparison between eGFP (upper panels) and pHluorin-plexinA1 (lower panels) expression pattern in the electroporated chick spinal cord is shown. Enlarged panels show the membranous fluorescence of pHluorin-plexinA1 (a’) as compared to the diffuse subcellular localization of eGFP (a). Regarding plexinA1, enlarged panel shows that this receptor is specifically enriched at the plasma membrane upon floor plate crossing (b’) which is not the case for eGFP that is equally expressed before and after floor plate crossing (b). Scale bars: 100 µM. FP: Floor Plate; Pre: Precrossing; Post: Postcrossing. Click here to view larger image.
This protocol provides a step-by-step procedure to follow the dynamics of an axon guidance receptor both in cell culture and in the developmental context of the chick embryo spinal cord.
To design a de novo pHluorin tagged protein, two points need to be considered regarding the cloning strategy. First, the pHluorin tag should be exposed to the lumen of the acidic endosomes, and consequently, to the extracellular compartment in order to visualize the plasma membrane receptor pool. Thus, the correct positioning of the pHluorin sequence regarding the receptor sequence is directly dependent on the type of receptor studied (i.e. whether the N-terminal or the C-terminal part of the receptor is exposed to the extracellular compartment). Second, as explained in the protocol, the position of the pHluorin sequence relative to the signal peptide should be considered to avoid subsequent cleavage between the pHluorin and the receptor of interest.
A limitation of the technique is linked to the trafficking of the receptors after their activation at the plasma membrane. Commonly, receptors are internalized and progress through the endocytic pathway composed of functionally and physically distinct compartments15. Due to ATP-driven proton pumps, endosomes maintain an acidic pH around 6 in early endosomes, around 5 in late endosomes, lower than 5 in lysosomes, but only around 6.4 in recycling endosomes or 7.0 in caveosomes allowing fluorescence in these two endosomic compartments. This issue can be resolved in vitro using total internal reflection fluorescence (TIRF) microscopy16. A second limitation inherent to this technique is that, due to its relatively big size, pHluorin tag could disrupt the receptor activity, depending on the tag insertion site.
Despite its advantages, pHluorin has not been widely used to study the dynamics and regulation of membrane proteins during axon navigation. Seminal works have used pHluorin to study the spatio-temporal dynamics of exocytosis during growth cone turning in vitro17. In the present protocol, we illustrate how pHluorin could be used to investigate the distribution of receptors at the plasma membrane of axons in vivo. Since pHluorins are genetically encoded, they are particularly appropriate to in vivo analysis. Although we describe its use in chick after in ovo electroporation, this approach can be used in other species. Indeed electroporation works efficiently in various animal models18,19. In addition, pHluorins have been successfully used in C. elegans, Drosophila, or mouse transgenic animals20-23.
Since pH-dependent change occurs in the millisecond range, the pHluorins are particularly adapted for live imaging6,24. PHluorin fusion has, for example, been used to monitor the in vivo dynamics of synaptobrevin exocytosis in olfactory sensory neurons from transgenic mice23. In vivo live imaging of pHluorin fusions holds considerable promise as confocal microscopes adapted to live imaging (fast imaging and low phototoxicity) become more efficient and accessible25. Live imaging in vivo could also be combined with fluorescence recovery after photobleaching (FRAP) so that exocytosis or diffusion could be assessed more directly26.
The use of other pH-sensitive variants and their combinations can further extend the range of applications. The dynamic distribution of the receptor in different organelles could be monitored by using ratiometric pHluorin that changes fluorescence according to the pH of the compartment6,27. Similarly, the addition of a pH-insensitive fluorescent protein to a membrane protein fused to an ecliptic pHluorin could provide important insights into its trafficking28. In addition, the recent cloning of pHtomato29 will enable monitoring of two receptors simultaneously. This could provide important insights into the formation of receptor complexes. Since pHluorin can also be used to tag guidance cues30, dual imaging of the ligand, and its receptor is also feasible.
In this protocol, the expression rate of the fusion protein is a critical point and particularly depends on the strength of the promoter, the stability of the fusion protein and the quantity of plasmid used to transfect the cells. Indeed, when overexpressed, fusion proteins may accumulate in intracellular vesicles and a fluorescent background may appear. Thus, several optimizations may be necessary to achieve a precise visualization of the pHluorin fused protein at the cell surface. Moreover, the use of pHluorin-fusion protein in fixed tissue requires particular care during and after fixation. Indeed, the stabilization of the conformations of pHluorin fusion proteins may only be temporary. Thus, pH must be kept above 7 to prevent a loss of the pHluorin signal. Moreover observations have to be carried out shortly after fixation. Optimal timing may need to be defined by users according to their particular pHluorin fusion protein.
The authors have nothing to disclose.
We thank Homaira Nawabi, Frederic Moret and Isabelle Sanyas for their help. This work is supported by CNRS, Association Francaise contre les Myopathies (AFM), ANR YADDLE, Labex DevWeCan, Labex Cortex, ERC YODA to V.C.; C.D-B and A.J are supported by a La Ligue contre le cancer and Labex DevWeCan fellowships, respectively.
COS7 cells | ATCC | CRL-1651 | |
DMEM GlutaMAX | GIBCO | 61965-026 | |
Sodium pyruvate | GIBCO | 11360-039 | |
Amphotericin B | Sigma | A2942 | |
Fetal bovine serum | GIBCO | 10270-106 | |
Penicillin/Streptomycin | GIBCO | 15140-122 | |
Exgen500 reagent | Euromedex Fermentas | ET0250 | |
PBS -Ca2+ -Mg2+ | GIBCO | 14190-094 | |
Fast green dye | Sigma | F7252 | |
32% Paraformaldehyde aqueous solution | Electron Microscopy | 15714-S | Dilute extemporaneously in PBS to achieve a 4% solution |
Gelatin from cold water fish skin | Sigma | G7041 | |
Sucrose | Sigma | S0389 | |
Cryomount | Histolab | 00890 | |
Hoechst 34580 | Invitrogen | H21486 | |
Mowiol 4-88 | Fluka | 81381 | |
Consumables | |||
Bottom-glass 35 mm dish | MatTek | P35G-1.5-14-C | |
5 ml Syringe | Terumo | SS-05S | |
Needles 0.9 mm x 25 mm | Terumo | NN-2025R | |
Capillaries | CML | PP230PO | capillaries are stretched manually in the flame |
Superfrost Plus Slides | Thermo Scientific | 4951PLUS | |
Material | |||
Curved scissors | FST | 129-10 | |
Microscalpel | FST | 10316-14 | |
Forceps | FST | Dumont #5 REF#11254 | |
Equipment/software | |||
Time lapse microscope | Zeiss | Observer 1 | |
Temp module S | PECON for Zeiss | ||
CO2 module S | PECON for Zeiss | ||
Metamorph software | Metamorph | ||
Eggs incubator | Sanyo | MIR154 | |
Electroporator apparatus | Nepa Gene CO., LTD | CUY21 | |
Electrodes | Nepa Gene CO., LTD | CUY611P7-4 | 4 mm platinum electrodes |
Fluorescence stereomicroscope | LEICA | MZ10F | |
Cryostat | MICROM | HM550 | |
Confocal microscope | Olympus | FV1000, X81 | |
Fluoview software | Olympus | ||
CLC Main Workbench software | CLC Bio |