Labeling Neuronal Membrane Receptors with Quantum Dots
Adapted from Imaging in Neuroscience (ed. Helmchen and Konnerth). CHSL Press, Cold Spring Harbor, NY, USA, 2011.INTRODUCTION
This protocol describes a highly sensitive approach for tracking the motion of membrane molecules over extended time periods with single-molecule resolution. This technique uses nanometer-sized quantum dots (QDs) linked to the extracellular part of the proteins to be followed. Single-fluorophore epifluorescence imaging then reveals the membrane diffusion of the particle of interest. Two methods are presented for labeling neurons with the primary antibody of choice along with a secondary anti-Fab antibody that is either biotinylated or directly coupled to the desired QD. The behavior of QD-labeled molecules can then be followed within the cell using epifluorescence imaging.
MATERIALS
Reagents
Antibody, primary
Biotinylated secondary Fab antibody (Jackson ImmunoResearch Laboratories) (for labeling Method A) or QD F(ab′)2 secondary IgG antibody (Invitrogen) (for labeling Method B)
Casein (1×) (for labeling Method B)
FM4-64 (N-(3-triethylammoniumpropyl)-4-(6-(4-diethylamino)phenyl)hexatrienyl)pyridinium dibromide), 1 µM in 40 mM KCl
Use this medium at 37°C for incubations, washes, and imaging.
Neurons for labeling
Equipment
Imaging setup
This procedure used an inverted microscope (Olympus, IX70 or 71) equipped with a 60× objective (numerical aperture [NA] = 1.45; Olympus). QD-605 and FM 4-64 were detected using a mercury arc lamp (excitation filter 525DF45) and appropriate emission filters (595DF60 and 695AF55 from Omega Filters). For Cy3 excitation, to achieve single-dye detection, the sample was illuminated with a frequency-doubled YAG crystal laser at 532 nm (∼0.5 kW/cm2). For detection of single QDs, the sample was illuminated with a mercury lamp. Real-time QD and Cy3 recordings were obtained at 13 Hz and 10 Hz, respectively, using a charge-coupled device (CCD) camera (Micromax 512EBFT, Cascade + 128, Roper Scientific or ORCA II ER, Hamamatsu Photonics) with up to approximately 1000 consecutive frames under Metaview (Universal Imaging). Single-molecule trajectories were analyzed with custom-written routines (Bonneau et al. 2005) in MatLab (MathWorks). Similar algorithms are freely available (Jaqaman et al. 2008; Sergé et al. 2008).
Incubator preset to 37°C
METHOD
-
1. Label neurons using one of the following two methods:
Method A
-
i. Incubate the neurons for 5 min with a high dilution (∼1–10 µg/mL) of primary antibody to label a small number of molecules.
-
ii. Wash the neurons three times without waiting, and then incubate them for 5 min in ∼10 µg/mL biotinylated secondary Fab antibody.
-
iii. Incubate the neurons for 1 min in ∼0.5–2 nM QD streptavidin conjugate solution.
Method B
-
iv. Mix 5 nM primary antibody with 30 nM QD F(ab′)2 secondary IgG antibody for 30 min at room temperature and then for an additional 15 min with 1× casein to block nonspecific binding.
-
v. Incubate the neurons for 10 min at 37°C with 0.06 nM the precoupled QD-antibody.
-
vi. Rinse off excess antibody.
If sufficient primary antibody is available, the primary antibody can be bound directly to QD (IgG/QD ratio = 1:1) with the QD antibody conjugation kit (Invitrogen) in a fast (few hours) and specific (coupling of thiols to maleimide groups) manner.
-
-
2. Label presynaptic boutons for 30 sec with 1 µM FM4-64 in 40 mM KCl.
See Troubleshooting.
-
3. Wash the cells, mount the coverslip in a recording chamber, and image the cells for up to 30 min in the imaging medium.
In our experiments, we perform QD real-time imaging at 13 Hz for approximately 500 consecutive frames. Time-lapse recording is preferred in experiments intended to determine dwell times in compartments (e.g., synaptic) where the molecules might reside for long durations. We often capture a single image of the cell morphology with transmitted light and of the synaptic marker (FM4-64, Venus::gephyrin, or DsRed::Homer1c) with fluorescent light before the QD image sequence.
TROUBLESHOOTING
Problem: The diffusion coefficients of the molecule of interest may be changed by neuronal activity stimulation.
[Step 2]
Solution: Neuronal activity regulates the neurotransmitter receptor's lateral diffusion properties (Tardin et al. 2003; Groc et al. 2004; Ehlers et al. 2007; Lévi et al. 2008; Bannai et al. 2009). Therefore, the KCl-induced FM4-64 synaptic vesicle loading may change the diffusion coefficients of the molecule of interest. As an alternative, one can use markers that do not require activity stimulation to be integrated into synaptic boutons (e.g., Mitotrackers; Invitrogen). Another option is to lipotransfect neurons with the main components of the inhibitory and excitatory postsynaptic differentiations, such as Gephyrin and Homer1c recombinant proteins tagged with GFP variants.
DISCUSSION
Interpretation of Imaging Data
After imaging, the trajectory is reconstructed from the image sequence of a single QD identified by the intermittency in its
fluorescence emission. QDs are classified as synaptic upon colocalization with a synaptic marker, such as glycine receptor
(GlyR; see below) (e.g., see Fig. 1). For each QD, we calculate the mean square displacement (MSD), diffusion coefficient (D), confinement area, transition between compartments, and dwell time within a compartment. The MSD is determined fromwhere τ is the frame acquisition time, N is the total number of frames, and n and i are positive integers with n determining the time increment. For simple two-dimensional (2D) Brownian mobility, the MSD as a function of time is linear
with a slope of 4D. If the MSD as a function of time tends to a constant value L, the diffusion is confined in a domain of size L. The diffusion coefficient (D) is determined by a fit on the first four points of the MSD as a function of time, with MSD(nτ) = 4Dnτ + b, where b is a constant reflecting the spot localization accuracy. The area in which diffusion is confined can be estimated by fitting
the MSD as a function of time to
where L2 is the confined area in which diffusion is restricted and Dmac is the diffusion coefficient on a long time scale.
QD trajectory analysis. (A) Example of surface exploration by an itinerant GlyR-QD visualized on a reconstructed trajectory. QD trajectories in synaptic areas #1 (Syn #1, green) and #2 (Syn #2, black); FM4-64-stained synapses (red). Note the large surface area explored by the GlyR-QD outside (blue) the synaptic areas. Scale bar, 1 µm. (B) Average diffusion coefficients of the QD shown in A during its extrasynaptic (blue) and synaptic journey (green and blue). Note the changes in the QD diffusion coefficient when exiting synapse #1 or entering synapse #2, since there is one sudden increase and one drop. (C) Time-averaged MSD function of individual QD shown in A during its exploration of extrasynaptic (blue) and synaptic (green) loci during a recording sequence. The same QD displayed an extrasynaptic linear MSD curve and a synaptic negatively bent MSD curve, characteristic of random walk and confined movement, respectively.
Using QDs to Measure GlyR Diffusion
GlyR lateral diffusion was studied in cultured neurons using a single-particle tracking (SPT) approach (Dahan et al. 2003). Data obtained with Cy3 fluorophore coupled directly to primary antibody (Fig. 2A) were compared with data obtained using QD-605-streptavidin conjugates (Fig. 2B,C). Cy3- and QD-tagged receptors were detected in extrasynaptic and synaptic regions. Individual Cy3 molecules were identified by their single-step bleaching. Cy3-GlyR could be tracked only for short durations (∼2.5 sec in Fig. 2A, white arrow). In contrast, the photostability of QDs allowed QD-GlyR trajectories to be visualized (Fig. 2B,C) for unprecedented durations (20 min) (Fig. 2C; Movie 1). Single QDs were identified by the random intermittency of their fluorescence emission (Nirmal et al. 1996). For example, one QD (white arrow in Fig. 2B) temporarily disappeared after 31.5 sec of recording. Long imaging duration enabled the observation of exchanges between extrasynaptic and synaptic domains, in which a GlyR alternated between free and confined diffusion states, respectively (white arrow in Fig. 2B). Silver-intensified and gold-toned QDs were detected using transmission electron microscopy (EM) with the same QD immunolabeling protocol (Fig. 2D). EM analysis provided evidence that QD-GlyR could access the core of the synapse (Fig. 2D). QD trajectories were reconstructed from recordings with custom “Sinema” software written in MatLab (Bonneau et al. 2005). Exploratory maps of trajectories indicated that individual QDs can exchange between extrasynaptic and synaptic compartments (e.g., blue and green trajectories, Fig. 1A). QDs diffused over broad areas of the extrasynaptic plasma membrane, whereas the QD exploratory map is reduced at synapses.
Diffusion and stabilization of single QD-GlyRs (green). Time-lapse recording (1200 images at 1 Hz; acquisition time, 75 msec). Synapses are labeled with FM4-64 (red).
Comparison of lateral GlyR motion analyzed with Cy3 and QD probes. GlyRs were detected in cultured neurons with Cy3 (A) or QD-streptavidin (B,C). (Green) GlyR spots; (red) FM4-64-labeled synaptic boutons. (A,B) Images were extracted from a sequence of 35 and 1024 images with an acquisition time of 100 msec for Cy3 and 75 msec for QD. The time after the start of recording (in seconds) is indicated on each image. The last image is a maximum projection of the entire stack of images corresponding to the GlyR trajectory. (C) Projection of time-lapse recording (1 Hz, 20 min) of QD-GlyR trajectory (green) overlaid with FM 4-64 staining (red) and bright-field image. (A–C) Cy3 and QD diffused rapidly in the extrasynaptic region (white arrows), whereas synaptic GlyRs were stable (orange arrows). Note the short membrane surface explored by Cy3-GlyR compared with QD-GlyR. Long QD imaging duration enabled the observation of a synaptic entry (white arrow in B). (D) EM detection of QD-GlyR within the synaptic cleft. (d) Dendrite; (b) synaptic bouton. The edges of the cleft are outlined. Scale bars, 1 µm (A,B); 5 µm (C); 500 nm (D).
As exemplified (Fig. 1B) for the QD trajectory shown in Figure 1A, the average diffusion coefficient (D) of the QD was lower within the synaptic area (Fig. 2, orange arrows). The plots of mean square displacement function (MSD) versus time were linear and negatively bent in the extrasynaptic and synaptic membrane, respectively (Fig. 1C), indicating a stronger confinement at synapses. This reflects local molecular interactions.
- © 2011 Cold Spring Harbor Laboratory Press