Noninvasive Imaging of 5-HT3 Receptor Trafficking in Live Cells

Sequential stages in the life cycle of the ionotropic 5-HT3 receptor (5-HT3R) were resolved temporally and spatially in live cells by multicolor fluorescence confocal microscopy. The insertion of the enhanced cyan fluorescent protein into the large intracellular loop delivered a fluorescent 5-HT3R fully functional in terms of ligand binding specificity and channel activity, which allowed for the first time a complete real-time visualization and documentation of intracellular biogenesis, membrane targeting, and ligand-mediated internalization of a receptor belonging to the ligand-gated ion channel superfamily. Fluorescence signals of newly expressed receptors were detectable in the endoplasmic reticulum about 3 h after transfection onset. At this stage receptor subunits assembled to form active ligand binding sites as demonstrated in situ by binding of a fluorescent 5-HT3R-specific antagonist. After novel protein synthesis was chemically blocked, the 5-HT3 R populations in the endoplasmic reticulum and Golgi cisternae moved virtually quantitatively to the cell surface, indicating efficient receptor folding and assembly. Intracellular 5-HT3 receptors were trafficking in vesicle-like structures along microtubules to the cell surface at a velocity generally below 1 μm/s and were inserted into the plasma membrane in a characteristic cluster distribution overlapping with actin-rich domains. Internalization of cell surface 5-HT3 receptors was observed within minutes after exposure to an extracellular agonist. Our orchestrated use of spectrally distinguishable fluorescent labels for the receptor, its cognate ligand, and specific organelle markers can be regarded as a general approach allowing subcellular insights into dynamic processes of membrane receptor trafficking.

Sequential stages in the life cycle of the ionotropic 5-HT 3 receptor (5-HT 3 R) were resolved temporally and spatially in live cells by multicolor fluorescence confocal microscopy. The insertion of the enhanced cyan fluorescent protein into the large intracellular loop delivered a fluorescent 5-HT 3 R fully functional in terms of ligand binding specificity and channel activity, which allowed for the first time a complete real-time visualization and documentation of intracellular biogenesis, membrane targeting, and ligand-mediated internalization of a receptor belonging to the ligand-gated ion channel superfamily. Fluorescence signals of newly expressed receptors were detectable in the endoplasmic reticulum about 3 h after transfection onset. At this stage receptor subunits assembled to form active ligand binding sites as demonstrated in situ by binding of a fluorescent 5-HT 3 R-specific antagonist. After novel protein synthesis was chemically blocked, the 5-HT 3 R populations in the endoplasmic reticulum and Golgi cisternae moved virtually quantitatively to the cell surface, indicating efficient receptor folding and assembly. Intracellular 5-HT 3 receptors were trafficking in vesiclelike structures along microtubules to the cell surface at a velocity generally below 1 m/s and were inserted into the plasma membrane in a characteristic cluster distribution overlapping with actin-rich domains. Internalization of cell surface 5-HT 3 receptors was observed within minutes after exposure to an extracellular agonist. Our orchestrated use of spectrally distinguishable fluorescent labels for the receptor, its cognate ligand, and specific organelle markers can be regarded as a general approach allowing subcellular insights into dynamic processes of membrane receptor trafficking.
A fundamental concern in neurobiology is the study of processes involved in expression, assembly and subcellular trafficking of neuroreceptors. Of particular interest are members of the large family of ligand-gated ion channels (LGIC) 1 that include nicotinic acetylcholine (nAChR), serotonin (5-HT 3 R), ␥-aminobutyric acid (GABA A ), glycine, and ionotropic glutamate receptors (1,2). All of them are oligomeric membrane proteins composed of subunits, which surround an ion channel that opens upon neurotransmitter binding to the receptor. The structural relationship between the different LGICs suggests that their assembly and trafficking involves similar molecular events (3). In general, it is thought that subunit assembly and ligand binding site formation occurs shortly after biosynthesis in the endoplasmic reticulum (ER) followed by trafficking to the plasma membrane (4 -8). Methodologies allowing the dynamic observation of these processes as well as a direct probing of receptor functions in intracellular compartments have still to be elaborated. In the present study, we focus on the ionotropic 5-HT 3 R as a representative member of the LGICs and explore new strategies to monitor receptor biogenesis in real time starting with the delivery of their coding DNA into living cells. We will resolve events, occurring after the receptors "birth" from those leading to cell membrane insertion of mature receptors and finally to ligand-induced re-absorption of receptors into the cell reflecting the end point of the receptors lifespan. Improved understanding of these processes can provide valuable information for the therapeutic targeting of LGICs at specific stages in their life cycle.
DNA Constructs-All constructs are based on a vector containing the short splicing variant of the murine 5-hydroxytryptamine type 3A subunit cDNA (23) preceded by the human cytomegalovirus gene promoter, as described before (9). The vector p5HT3R-ECFP, containing the ECFP-labeled receptor, was obtained as follows: the original vector was first mutated using the oligonucleotides 5Ј-CTG ATG ACT GCT CAA TCG ATG CCA TGG GAA ACC-3Ј and 5Ј-GGT TTC CCA TGG CAT CGA TTG AGC AGT CAT CAG-3Ј, adding a ClaI restriction site in the large cytoplasmic loop sequence between the third and fourth predicted membrane-spanning domains. ECFP was introduced between Ser 359 and Ala 360 of the receptor (corresponding to Swiss-Prot entry p23979, mature sequence numbering) to give 5-HT 3 R-ECFP. The ECFP insert was obtained by PCR amplification on the template pECFP-N1 (Clontech) using the synthetic oligonucleotides 5Ј-CCA TCG ATA TGG TGA GCA AGG GCG AGG-3Ј and 5Ј-CCA TCG ATC TTG TAC AGC TCG TCC ATG CCG-3Ј and ligated into the receptor ClaI restriction site. The restriction sites HindIII and NotI were added to the 5Ј and 3Ј ends of the 5-HT 3 R-EGFP sequence by PCR amplification using the oligonucleotides 5Ј-CGA TAA GCT TCA CCA TGC GGC TCT GCA TCC CGC AGG TG-3Ј and 5Ј-GCT GTG CCC ACG CGG CCG CTC AAG AAT AAT GCC AAA TGG ACC AGA G-3Ј, and the purified 2197-bp fragment was subcloned into the HindIII/NotI cut vector pEAK8 (Edge BioSystems, Gaithersburg, MD), yielding the vector p5HT3R-ECFP. The vector p5HT3R was obtained by subcloning the non-mutated receptor gene into the plasmid pEAK8 between HindIII and NotI, using the same oligonucleotides as for constructing p5HT3R-ECFP. The plasmid constructs were confirmed by restriction mapping and DNA sequencing.
Cell Culture, Transfections, and Permeabilization-Adherent human embryonic kidney (HEK293) and N1E-115 cells were grown in Dulbecco's modified Eagle medium/F-12 (Invitrogen) supplemented with 2.2 and 10% fetal calf serum (Invitrogen), respectively, using plastic flasks from TPP AG (Trasadingen, Switzerland). The cultures were split regularly and kept at 37°C in a humidified atmosphere with 5% CO 2 . For electrophysiology and confocal microscopy measurements, respectively, cells were seeded in 35-mm cell culture dishes and 6-well plates containing 22-mm diameter glass coverslips at a density of 150,000 cells/ml and transfected using LipofectAMINE 2000 reagent (Invitrogen).
Transfection efficiencies were determined by confocal fluorescence microscopy on cell samples which were cotransfected with an enhanced GFP (pEGFP-N1; Clontech) reporter DNA.
Fixation of cells was achieved by 10-min incubation at room temperature in a solution containing 3.7% formaldehyde in PBS. Subsequent permeabilization was performed by 5-min incubation in the presence of 0.1% Triton X-100 in PBS. Triton X-100 was used for cell permeabilization because this detergent was shown not to interfere with ligand binding affinity (12,24). Cells were washed three times with PBS between incubations.
Radioactive Binding Assays-Receptor concentrations as well as ligand affinities were measured by radioligand binding assay. 100 l containing ϳ1 ϫ 10 6 cells resuspended in 10 mM Hepes, pH 7.4, were incubated for 30 min at room temperature in solutions of 10 mM Hepes, pH 7.4, with varying concentrations of the specific antagonist [ 3 H]GR65630 in a final volume of 1 ml. A rapid filtration through Whatman GF/B filters (presoaked for 15 min in 0.5% (w/v) polyethylenimine) followed by two washes with 3 ml of ice-cold 10 mM Hepes at pH 7.4 terminated the incubation. Filters were then transferred into scintillation vials and 4 ml of Ultima Gold (Packard, Meridan, CT) was added. The radioactivity was measured in a Tri-Carb 2200CA liquid scintillation counter (Packard). Nonspecific binding was determined in the presence of 1 M quipazine.
Binding assay on permeabilized cells were processed as above, except that the cells were pre-treated with 0.1% saponin in 10 mM Hepes pH 7.4 for 5 min at room temperature before radioligand binding. All experiments were done in triplicate.
The dissociation constant K d of [ 3 H]GR65630 and the Hill coefficient n were calculated by fitting the experimental data to the following binding equation.
Electrophysiology-We used the standard patch-clamp technique in whole-cell voltage clamp to Ϫ60 mV employing an EPC-9 patch clamp amplifier (HEKA Elektronik GmbH, Lambrecht, Germany). For data acquisition and storage the software PULSE 8.3 (HEKA Elektronik GmbH) was used. Borosilicate glass pipettes were heat polished and had resistances of 2-5 M⍀. Pipettes were filled with 140 mM KCl, 5 mM MgCl 2 , 10 mM EGTA, 10 mM Hepes-KOH, pH 7.3. The external solution was 147 mM NaCl, 12 mM glucose, 2 mM KCl, 2 mM CaCl 2 , 1 mM MgCl 2 , 10 mM Hepes-NaOH, pH 7.4. Ligands dissolved in the external solution were applied with a RSC-200 or the MSC-200 perfusion system (Bio-Logic, Claix, France). During experiments the cells were continuously perfused. All experiments were performed at room temperature.
Data were evaluated by fitting to the Hill following equations, where I is the peak current at a certain ligand concentration, I max is the maximal peak current achievable, I 0 is the peak current in absence of any antagonist, EC 50 and IC 50 are the half-maximal effective and inhibitory concentrations, respectively, and n is the Hill coefficient. For antagonist-agonist competition experiments, cells were perfused with granisetron-containing solutions for 3 min before the addition of serotonin (5-HT) at a concentration of 30 M.
Laser Scanning Confocal Microscopy-Laser scanning confocal micrographs were recorded using 458/488-, 543-, and 633-nm laser lines on a Zeiss LSM 510 microscope (Carl Zeiss AG) equipped with a 63ϫ water objective (1.2 numerical aperture). Detection and distinction of fluorescence signals was achieved by appropriate filter sets using a multitracking mode. Scanning speed and laser intensity were adjusted to avoid photobleaching of the fluorescent probes and damage or morphological changes of the cells. The microscope was equipped with a microcultivation system (Incubator S, CTI controller 3700 digital, Zeiss) to control temperature, humidity, and CO 2 for maintaining physiological conditions during long-term experiments. Image analysis and fluorescence signal quantification was performed using Zeiss LSM software.

RESULTS
We investigated the time course of 5-HT 3 R plasma membrane expression by analyzing non-permeabilized HEK293 cell samples at different times after transfection via radioligand binding assay. 5-HT 3 R cell surface expression increased over time and reached a maximum number of 2.6 ϫ 10 6 receptors per cell 34 h after transfection (Fig. 1). When these cells were permeabilized with saponin it was possible to detect in addition to cell surface receptors also 5-HT 3 R-specific ligand binding activity in the cytoplasm, which accounted for about 40% of the total ligand binding sites. In N1E-115 cells, which endogenously express 5-HT 3 receptors, about 60% of the ligand binding activity was localized inside the cell (Table I). These results indicate that a substantial fraction of intracellular receptors is capable of binding 5-HT 3 R-specific ligands before plasma membrane integration. We then analyzed intracellular 5-HT 3 R formation and trafficking in more detail. To observe in real-time 5-HT 3 R biosynthesis after cell delivery of its coding DNA, we inserted the coding sequence of the ECFP into the large cytoplasmic loop between the third and fourth predicted membrane-spanning domains of the receptor (Fig. 2A). The preserved functionality of the fusion construct was confirmed by radioligand binding assay and whole cell patch clamp measurements (Table II).
Using laser scanning confocal microscopy we found that the ECFP-labeled receptors start to form in the endoplasmic reticulum typically 3 h after transfection onset. The receptor-derived fluorescence signal could be colocalized with a spectrally distinguishable enhanced yellow fluorescent ER marker (Fig.  2B), which was transfected together with the ECFP-tagged receptor DNA. In parallel experiments, we were able to overlap fluorescence images arising from the receptor with EYFP targeted to the Golgi apparatus at ϳ4 h after transfection (Fig.  2C). Receptor trafficking from the appearance in the Golgi to plasma membrane insertion occurred in about 30 min. The integration of mature receptors into the plasma membrane could be confirmed by binding of a fluorescence-labeled 5-HT 3 R-specific antagonist (Fig. 2D).
Although the fluorescence emitted from the 5-HT 3 R-ECFP is primarily an indication of a folded ECFP marker protein, it allows also monitoring of 5-HT 3 receptor biogenesis, since the ECFP label is directly integrated into the receptor sequence. Information about the assembly of 5-HT 3 R subunits can be assessed via detection of the formation of the ligand binding sites. It has been shown for the structurally related acetylcholine receptor (25) as well as by amino acid sequence homology studies on the 5-HT 3 R (26, 27) that the ligand binding sites are located between neighboring subunits; in consequence, ligand binding can only be observed once the subunits are fully assembled to a functional pentameric receptor.
After monitoring 5-HT 3 R biosynthesis in cytoplasmic compartments we investigated the presence of intracellular 5-HT 3specific ligand binding sites, which indicate the functional assembly of receptor subunits. Ligand binding to intracellular 5-HT 3 receptors was first investigated by permeabilizing fixed HEK293 cells at different times after transfection and subsequent labeling with fluorescent ligands. The binding specificity of those ligands to 5-HT 3 receptors in intracellular compartments was confirmed by ligand displacement experiments (see supplemental data).
ECFP-labeled 5-HT 3 Rs were localized in the endoplasmic reticulum or Golgi apparatus by comparing fluorescence images of the receptor with those of organelles labeled with EYFP as described above. The diffusion of a Cy5-labeled 5-HT 3 Rspecific antagonist (GR-Cy5) into permeabilized cells allowed us to detect ligand binding activity in ER and Golgi apparatus (Fig. 3). This reports on receptor assembly at an early stage of expression in the ER. The GR-Cy5 fluorescence signal located in the Golgi was stronger than in the ER indicating substantially higher amounts of assembled 5-HT 3 receptors in the Golgi apparatus.
The amount of 5-HT 3 R ligand binding sites on the cell surface receptors as compared to those formed inside the cell was then evaluated by double ligand binding experiments (Fig. 3, G-I). Fluorescent rhodamine-labeled antagonist GR-Rho was applied to non-permeabilized cells expressing 5-HT 3 R-ECFP to label the receptors located on the plasma membrane. After fixation and permeabilization of the cells, the antagonist GR-Cy5 was applied to label the receptors inside the cell. Due to the low off-rate of GR-Rho from 5-HT 3 R, it was not replaced by GR-Cy5, as described before (9), and thus allowed us to visualize and to distinguish between extra-and intracellular ligand binding sites. Analyzing images arising from the fluorescence of Cy5-and rhodamine-labeled ligands in consecutive confocal cross-sections covering the whole cell revealed that 43 Ϯ 14% of the receptor binding sites were located on the plasma membrane and the rest inside the cell (value averaged on 20 cells). This value is in the same range as obtained by radioligand binding assays on cell populations (average on 1 ϫ 10 6 cells). Receptor trafficking from the Golgi apparatus to the plasma membrane was then analyzed in detail by imaging 5-HT 3 R-ECFP-expressing HEK293 cells (frame rate of 2 images/s; Fig.  4, A and B). The ECFP label allowed us to detect vesicle-like structures carrying 5-HT 3 receptors in their membrane, which moved inside the cytoplasm. The velocity of these carrier vesicles was generally below 1 m/s, measured as the distance between the pixel coordinates in consecutive images. The receptor-carrying vesicles seemed to be located in tubulin-rich regions visualized by cotransfecting EYFP-labeled tubulin (pEYFP-Tub). To verify whether tubulin filaments are required for 5-HT 3 receptor transport to the plasma membrane, we studied the influence of colchicine on receptor trafficking (Fig.  4, C and D); colchicine is known to inhibit microtubule formation (28 -30). 5 h after adding 100 g/ml cycloheximide to cells, which interferes with the translation machinery (31), the receptors were almost completely located on the cell membrane. However, when the cells were treated 2 h before with 50 g/ml colchicine, the receptors did not completely localize at the membrane, demonstrating that the tubulin filaments are required for proper receptor trafficking and final membrane insertion. We then investigated the effects of agonist application on the subcellular distribution of 5-HT 3 receptors. 34 h after transient expression of 5-HT 3 R-ECFP, HEK293 cells were treated with 100 g/ml cycloheximide for 5 h to stop new protein synthesis and in turn allowing delivery of intracellular receptors to the cell membrane and thus depleting intracellular receptor fluorescence. Subsequent incubation of non-permeabilized cells with 12 nM GR-Cy5 yielded a complete spatial overlap of the ECFP receptor and the GR-Cy5 ligand fluorescence signals on the cell surface (Fig. 5, A-E). The fluorescent antagonist could be replaced by adding an excess (500 nM) of the non-fluorescent 5-HT 3 R-specific agonist mCPBG. After incubation at room temperature for 30 min, the cells were intensively washed with PBS buffer. We then applied the fluorescent antagonist GR-Cy5 for a second time using the same conditions as before (Fig.  5, F-J). The absence of fluorescence labeling on the cell surface was a first indication that agonist mCPBG might have induced internalization of the 5-HT 3 R. The fact that the ECFP fluorescence of the 5-HT 3 R was still located near the intracellular site of plasma membrane might show receptors internalized in early endosomes. This hypothesis was studied in more detail in the following.
Quantification of receptor-derived ECFP fluorescence intensities on one hand, and fluorescence signals of the receptorbound antagonist GR-Cy5 on the other hand, allowed us to estimate the number of accessible receptor binding sites on the cell surface. The time course of receptor internalization was studied by exposing the cells to 500 nM mCPBG for increasing periods of time at 37°C. The agonist was then quickly removed by three washes with PBS buffer. Subsequent labeling with GR-Cy5 allowed visualization of remaining 5-HT 3 receptors on the cell surface. We observed that more than half of the receptors were internalized after 5 min of constant exposure to the agonist (Fig. 6A). The internalized receptors progressively moved away from the cell surface after endocytosis, and after 2 h of constant exposure to the agonist, labeling of the cell membrane with R18 clearly revealed no overlap of the ECFP receptor fluorescence with the membrane dye indicating complete internalization of the receptor (Fig. 6, B-D).
During agonist-induced receptor internalization we observed a progressive disappearance of receptor clusters on the membrane to a more even distribution. By coexpression of 5-HT 3 R-ECFP together with the actin marker EYFP-actin, we could observe that the receptor clusters were colocalized with the actin marker, suggesting the involvement of actin filaments in

H]GR65630 to 5-HT 3 R on whole cells (1) and on saponin-permeabilized cells (2)
The values represent average number of 5-HT 3 3. Distribution of 5-HT 3 R and receptor ligand binding sites at the surface and inside of HEK293 cells imaged by confocal fluorescence microscopy at different stages of receptor expression. A-F, the 5-HT 3 R-ECFP (cyan, A and D) was coexpressed with EYFP (yellow) targeted to either the endoplasmic reticulum (B) or the Golgi apparatus (E). Comparing images of A and D with those of B and E shows the presence of 5-HT 3 R both in the ER and in the Golgi 3 and 4 h after transfection onset, respectively. After cell fixation and membrane permeabilization, the fluorescent ligand GR-Cy5 was added, resulting in specific binding to 5-HT 3 R (red images in C and F). Comparison of the red images in C and F with the yellow images in B and E shows the presence of functional receptor ligand binding sites (i.e. properly folded receptors) in the particular organelles. G-I, GR-Rho (green) and GR-Cy5 (red) were applied, respectively, before and after permeabilization to investigate the percentage of membrane-localized as compared with whole cell receptor binding sites. Comparing the image in I with the cyan image in G (reflecting the 5-HT 3 R-ECFP localization) demonstrates additionally that the major portion of intracellular 5-HT 3 Rs are correctly folded because they expose functional ligand binding sites. Size bars represent 5 m.
the membrane organization of the 5-HT 3 R (Fig. 6, E and F). Analysis of images of cells expressing 5-HT 3 R-ECFP over time indicated that the fraction of clustered receptors, which before agonist addition was originally almost 50% of total membrane receptors, decreased by a factor of two after ϳ4 min incubation with 500 nM mCPBG (Fig. 6G), a time period comparable with that of internalization as measured conjointly with the fluorescent antagonist.

DISCUSSION
Receptor labeling by genetic fusion with the enhanced version of the cyan fluorescent protein (ECFP) allowed monitoring of 5-HT 3 R expression, trafficking, plasma membrane targeting, and ligand-induced endocytosis in real time in living cells. The fluorescent receptor retained ligand binding and channel activity and permitted in vivo observation of its "birth" after delivery of its coding DNA into HEK293 cells. These cells were used as host system for this trafficking study because they efficiently express 5-HT 3 receptors in a functional form (9).
By colocalization with EYFP-labeled subcellular markers, it was possible to monitor the receptor successive appearance in different organelles over time. Furthermore, the formation of ligand binding sites was observable in situ at an early step in 5-HT 3 R synthesis in the endoplasmic reticulum as visualized by labeling with fluorescent antagonists on permeabilized cells. Recent studies on the ligand binding domain of nicotinic receptors indicated that the assembly of subunits was required for proper functioning of its binding site (25). Homology studies between acetylcholine receptor and 5-HT 3 R amino acid sequences point to the fact that the ligand binding site might be located at the interface between two subunits (26,27,32) indicating that assembly of the 5-HT 3 subunits might be required for proper functioning of the binding site. Thus this present study indicates that the assembly of the 5-HT 3 R subunits to a functional receptor is likely to occur in the endoplasmic reticulum, since the labeling with the 5-HT 3 R-specific antagonist GR-Cy5 correlates in time and space with the fluorescence signal of the 5-HT 3 R-ECFP in this subcellular compartment. Thus, we provide the first evidence obtained in situ on individual live cells that the first formation of functional 5-HT 3 R specific binding sites occurs in the ER. Until now only indirect studies have been reported for 5-HT 3 R subunit assembly using cell fractionation to assay ER-retained 5-HT 3 receptors (12). In analogy to our findings the assembly of nAchR subunits and thus the formation of ␣-bungarotoxin binding sites was also found to occur in the endoplasmic reticulum (33,34). The recent discovery of an ER retention signal in 5-HT 3B subunits (35) and the fact that these subunits can only be "rescued" from the ER and assemble to functional receptors by coexpression of 5-HT 3 type A subunits in COS 7 cells (8), are complementary indirect indications in support of our observations. We were able to follow receptor trafficking from intracellular compartments up to the plasma membrane as a microtubule-based transport; receptor-containing vesicles leaving the Golgi apparatus could not reach the membrane when the microtubule formation was blocked. Another support for this hypothesis was provided by measuring the transport of receptor vesicles along microtuble inside living cells. We obtained values of v ϭ 220 -1850 nm/s by consecutive fluorescence imaging, which is similar as for the Golgi-to-plasma membrane transport via post-Golgi carriers on microtubules measured elsewhere (36 -40). The first appearance of receptors on the plasma membrane occurred about 4 -5 h after transfection onset and exhibited a cluster distribution, in agreement with recent experiments from our laboratory (9). Fluorescence images of an actin marker showed that these receptor clusters colocalized with actin-rich membrane domains, suggesting an actin-dependent membrane localization, supporting previous investigations (41). Moreover, we could observe high receptor densities on cell-cell interfaces, corroborating with the large quantity of actin present in these domains (42). It has been shown for other ligand-gated channels such as nicotinic acetylcholine (43), GABA A (44), and P2X 1 (45) receptors that long time course desensitization provoke internalization by endocytosis. This property, which could furnish a pathway for receptor downregulation in response to tonic levels of agonist, was so far not observed for the 5-HT 3 R. We could monitor ligand-evoked endocytosis by concerted imaging of 5-HT 3 R-ECFP and the antagonist GR-Cy5 after application of the potent specific agonist mCPBG for different time lapses. We observed that after 5 min of incubation the number of receptors present on the plasma membrane diminished almost by a factor of three and in parallel that its cluster organization was consistently overtaken by a more homogeneous distribution, presumably in endosomes located below the cell membrane. The loss of cell surface binding of GR-Cy5 was the first indication for 5-HT 3 R internalization. The localization of internalized receptors below the plasma membrane could be confirmed by employing a lipophilic membrane dye, which did not overlap with the bioluminescence signal of the receptor. In conclusion, the functionally silent insertion of ECFP into the 5-HT 3 R proved to be a valuable tool FIG. 6. Temporal resolution of agonist-induced internalization of 5-HT 3 R binding sites and disappearance of 5-HT 3 R clusters colocalized with actin filaments as revealed by confocal fluorescence microscopy. A, live HEK293 cells expressing the 5-HT 3 R-ECFP chimera were exposed to 500 nM mCPBG for the indicated lengths of time. After washing with PBS, cells were incubated with 12 nM GR-Cy5; comparing ECFP and Cy5 fluorescence intensities allowed for the quantification of non-internalized receptors. Each horizontal dash represents one experimental value measured on a cell and the line connects the corresponding mean values for one particular incubation time. B-D, confocal fluorescence microscopy images of one representative experiment on HEK293 cells expressing 5-HT 3 R-ECFP 34 h after transfection. Cells were first incubated for 2 h in a solution of 500 nM agonist mCPBG, and then the cell membrane was stained by R18, a membrane-integrating fluorescence dye, which yielded red images of membrane structures together with cyan-colored intracellular 5-HT 3 R-ECFP (B). A magnified image reveals that the cyan receptor fluorescence does not overlap with the red fluorescence of the membrane-integrated dye (C). The tridimensional projection of confocal images of the ECFP label shows a partial clustered organization of the internalized receptors (D). The 5-HT 3 R-ECFP has a cluster distribution on the plasma membrane of a HEK293 cell (E), which colocalizes with actin-rich domains visualized by coexpression of EYFP-actin (F). Arrows in E and F indicate regions of colocalization. After inducing receptor activation by applying 500 nM of the agonist mCPBG, the percentage of membrane receptors in cluster form is rapidly decreasing, leading to a more homogeneous distribution of the receptors below the cell membrane (G). Size bars represent 5 m. enabling spatial and temporal resolution of the receptor trafficking in single living cells. We have demonstrated that multicolor imaging could be used to study the receptor translocation to the plasma membrane after biosynthesis, appearing to follow the microtubule cytoskeleton in post-Golgi carriers whereas its localization and clustering on the plasma membrane would involve actin filaments. Furthermore, this is the first report to demonstrate in situ the ER-localized formation of the 5-HT 3 R ligand binding sites. We could additionally track dynamic changes in the receptor localization by direct observation of ligand-induced internalization of the receptor in response to a 5-HT 3 R-specific agonist. We believe the same methodology using multicolor dynamic tracking in living cells could be applied to a variety of other receptors to elucidate number of unraveled cellular mechanisms.