 |
The Journal of Neuroscience, December 17, 2003, 23(37):11554-11567
 Previous Article | Next Article 
Cellular/Molecular
Assembly of  4 2 Nicotinic Acetylcholine Receptors Assessed with Functional Fluorescently Labeled Subunits: Effects of Localization, Trafficking, and Nicotine-Induced Upregulation in Clonal Mammalian Cells and in Cultured Midbrain Neurons
Raad Nashmi,1
Mary E. Dickinson,1,2
Sheri McKinney,1
Mark Jareb,1,3
Cesar Labarca,1
Scott E. Fraser,1,2 and
Henry A. Lester1
1Division of Biology, 2Biological Imaging Center, California Institute of Technology, Pasadena, California 91125, and 3Department of Biology, Sacred Heart University, Fairfield, Connecticut 06825
 |
Abstract
|
|---|
Fura-2 recording of Ca2+ influx was used to show that incubation in 1 µM nicotine (2-6 d) upregulates several pharmacological components of acetylcholine (ACh) responses in ventral midbrain cultures, including a MLA-resistant, DH E-sensitive component that presumably corresponds to  42 receptors. To study changes in 42 receptor levels and assembly during this upregulation, we incorporated yellow and cyan fluorescent proteins (YFPs and CFPs) into the 4 or 2 M3-M4 intracellular loops, and these subunits were coexpressed in human embryonic kidney (HEK) 293T cells and cultured ventral midbrain neurons. The fluorescent receptors resembled wild-type receptors in maximal responses to ACh, dose-response relations, ACh-induced Ca2+ influx, and somatic and dendritic distribution. Transfected midbrain neurons that were exposed to nicotine (1 d) displayed greater levels of fluorescent 4 and 2 nicotinic ACh receptor (nAChR) subunits. As expected from the hetero-multimeric nature of 42 receptors, coexpression of the 4-YFP and 2-CFP subunits resulted in robust fluorescence resonance energy transfer (FRET), with a FRET efficiency of 22%. In midbrain neurons, dendritic 42 nAChRs displayed greater FRET than receptors inside the soma, and in HEK293T cells, a similar increase was noted for receptors that were translocated to the surface during PKC stimulation. When cultured transfected midbrain neurons were incubated in 1 µM nicotine, there was increased FRET in the cell body, denoting increased assembly of 42 receptors. Thus, changes in 42 receptor assembly play a role in the regulation of 42 levels and responses in both clonal cell lines and midbrain neurons, and the regulation may result from Ca2+-stimulated pathways.
Key words: nicotine addiction; nicotinic receptors; receptor assembly; FRET; 42; fluorescent protein
|
Introduction
|
|---|
Increased nicotinic receptor levels are measured in tobacco smokers and in animals exposed chronically to nicotine (Marks et al., 1992 ; Breese et al., 1997; Perry et al., 1999). Several incisive reviews suggest that this upregulation plays a role in nicotine addiction (Dani and Heinemann, 1996; Buisson and Bertrand, 2002; Mansvelder and McGehee, 2002). Critical evaluation of this hypothesis is not yet possible, because we do not know whether this increase in nicotinic receptor levels is accompanied by increased receptor function, by decreased receptor function, or by no change.
Important clues are presented by the findings that, in clonal cell lines and Xenopus oocytes expressing nicotinic receptor subunits, exposure to nicotine does produce upregulation of functional receptors (Peng et al., 1994; Whiteaker et al., 1998; Buisson and Bertrand, 2001). One study indicates that both levels and function of 7 receptors increase in cultured chick cortical neurons during chronic exposure to nicotine (Kawai and Berg, 2001). Because the regulation of ion channels is likely to vary among cell types, however, it is important to test for such functional upregulation with other relevant nicotinic receptor subtypes and with other neurons of interest. We have chosen to study the mesolimbic system, because it is a major common target for nicotine and others drugs of abuse (Stolerman and Shoaib, 1991; Dani and Heinemann, 1996).
Which nicotinic receptors should one monitor? This study concentrates on 42 receptors, because receptors containing 4 and 2 nicotinic subunits comprise the most abundant nicotinic subunits in the brain and are most sensitive to nicotine (Wonnacott, 1990). Nicotine binding is much reduced in mice lacking the 4 nicotinic subunit (Marubio et al., 1999). 2 subunit-containing nicotinic receptors are responsible for mediating nicotine self-administration in mice, nicotine-elicited currents from midbrain neurons, and increased dopamine release in the mesolimbic dopamine system (Picciotto et al., 1998). Dopaminergic neurons express several other nicotinic acetylcholine receptor (nAChR) subunits, including, 2, 3, 5, 6, 7, and 3 (Di Chiara, 2000; Klink et al., 2001). Altered nicotinic receptor function is also implicated in several disorders such as epilepsy (Steinlein et al., 1995; Figl et al., 1998; Bertrand et al., 2002), Parkinson's disease (Banerjee et al., 2000; Martin-Ruiz et al., 2000), schizophrenia (Durany et al., 2000), and Alzheimer's disease (Court et al., 2000).
This paper gathers new data about the regulation of 42 nicotinic receptors by combining four types of in vivo measurements: (1) electrophysiology, (2) visualization of fluorescently labeled receptors; (3) fluorescence resonance energy transfer (FRET) between receptor subunits; and (4) fura-2 measurements of Ca2+ influx. We chose to design, test, and measure fluorescent 4 and 2 subunits both because there is no 42 counterpart of the convenient 7 ligand, -bungarotoxin, and because fluorescent subunits can be detected in real time while still functional.
We expected that FRET measurements in particular would be more sensitive for hetero-oligomeric channels (Riven et al., 2003) such as the 42 receptor than for homomeric ion channels, because it is possible to place donor or acceptor fluorophores in all subunits of each channel (although we acknowledge some ambiguity because the receptors might have either (4)2(2)3 or (4)3(22) stoichiometry (Nelson et al., 2003)). The data presented here fulfill this expectation. The efficiency of FRET varies inversely with the sixth power of distance separating the two fluorophores, and we have used this sensitive measure of distance to assess the fraction of assembled versus unassembled receptors. These fluorescently tagged 42 nicotinic receptor subunits now reveal that nicotine-induced nAChR assembly occurs primarily in the soma of midbrain neurons.
|
Materials and Methods
|
|---|
cDNA constructs of fluorescently tagged nicotinic subunits. Mouse 4 and 2 nAChR cDNAs were kindly provided by Jerry Stitzel (University of Michigan, Ann Arbor, MI). Both 4 and 2 were inserted in the EcoRI site of the pCI-neo vector (Promega, Madison, WI) in the T7 orientation. Yellow fluorescent protein (YFP) was inserted in frame into the N terminus of 4. YFP (Clontech, BD Biosciences, Bedford, MA) and an upstream hemagglutinin (HA) epitope tag were inserted three amino acids (aa) downstream from the putative signal peptide sequence (after residue 34) (Fig. 1A). The XmaI in the multiple cloning region of pCI-neo was removed by cutting with NotI and XbaI. DNA polymerase I Large Fragment (Klenow) was used to fill in the 5' overhangs, and a blunt-end ligation was performed. For one upstream linker of YFP, the forward primer 5'-TTT TCC CGG GAA TAT CCT TAT GAC GTC CCA GAC TAC GCC ATG GTG AGC AAG GGC GAG GAG CTG-3' and the reverse primer 5'-AAA AGG CGT AGT CTG GGA CGT CAT AAG GAT ACC CGG GTC TTG TAC AGC TCG TCC ATG CCG AG-3' were used to amplify by PCR pEYFP-N1 (Clontech, BD Biosciences). To create YFP with two flanking linkers with an upstream HA tag, the forward primer 5'-TTT TCC CGG GAA TAT CCT TAT GAC GTC CCA GAC TAC GCC CCT GAC GCC ATG GTG AGC AAG GGC GAG GAG CTG-3' and the reverse primer 5'-AAA AGG CGT AGT CTG GGA CGT CAT AAG GAT ACC CGG GTG TCA GGG TCA TAT GGG GCA TAG TAG ACC TTG TAC AGC TGG TCC ATG CCG AG-3' were used to amplify pEYFP-N1 by PCR. In both constructs, the PCR insert products and the vectors were cut with XmaI. The vector was dephosphorylated with shrimp alkaline phosphatase (SAP) and ligated to the insert.
Cyan fluorescent protein (CFP) (Clontech, BD Biosciences) was inserted in frame into the C terminus of 2. AgeI restriction site was introduced one residue upstream of the terminal codon of 2 by site-directed mutagenesis (QuickChange; Stratagene, La Jolla, CA) using the forward primer 5'-C TCA GCT CCC AGC TCC AAG ACC GGT TGA GGT CTC TCA TTT GCA G-3' and the reverse primer 5'-C TGC AAA TGA GAG ACC TCA ACC GGT CTT GGA GCT GGG AGC TGA G-3'. To create the CFP insert with an upstream c-myc tag, the forward primer 5'-TTTT ACC GGT GAG CAG AAG CTG ATC TCA GAG GAG GAT CTG ATG GTG AGC AAG GGC GAG GAG CTG-3' and the reverse primer 5'-AAAA CAG ATC CTC CTC TGA GAT CAG CTT CTG CTC ACC GGT CTT GTA CAG CTC GTC CAT GCC GAG-3' were used in PCR to amplify pECFP-N1 (Clontech, BD Biosciences). Both insert and vector were cut with AgeI. The vector containing 2 was dephosphorylated with SAP and ligated to the insert.
YFP was introduced in frame into the BstEII site in the M3-M4 intracellular loop of 4. YFP with an upstream HA epitope tag was inserted after residue 426 of the 4 protein. The 4 vector was cut with BstEII and dephosphorylated. PEYFP-N1 was amplified by PCR using the forward primer 5'-TTT TGG TCA CCC TAT CCT TAT GAC GTC CCA GAC TAC GCC ATG GTG AGC AAG GGC GAG GAG CTG-3' and the reverse primer 5'-AAAA GGC GTA GTC TGG GAC GTC ATA AGG ATA GG TGA CCT CTT GTA CAG CTC GTC CAT GCC GAG-3'. The YFP PCR product was cut with BstEII and ligated to the 4 vector.
CFP was inserted into the M3-M4 loop of 2. CFP with an upstream c-myc tag was introduced in frame into the PpuMI restriction site, which is at amino acid position 381 of 2. 2 was cut with PpuMI and dephosphorylated with SAP. The pECFP-N1 insert was amplified by PCR using the forward primer 5'-TTTT A GGT CCT GAG CAG AAG CTG ATC TCA GAG GAG GAT CTG ATG GTG AGC AAG GGC GAG GAG CTG-3' and the reverse primer 5'-AAAA CAG ATC CTC CTC TGA GAT CAG CTT CTG CTC AGG ACC TTT CTT GTA CAG CTC GTC CAT GCC GAG-3'. The PCR-amplified CFP product was cut with PpuMI and ligated to the 2-containing vector. All cDNA constructs were verified by sequencing.
The long intracellular M3-M4 loop of both subunits contains several motifs that may play a role in receptor function: putative phosphorylation sites (Fenster et al., 1999; Guo and Wecker, 2002) and potential trafficking signals (Williams et al., 1998). In designing the subunits that contained fluorescent tags in the M3-M4 loop, we avoided regions that included phosphorylation motifs (PKC, cAMP-dependent protein kinase, tyrosine kinase, casein kinase, calmodulin-dependent kinase, cyclin-dependent kinase 5), although one of four putative calmodulin kinase motifs was disrupted in 4-YFP-M (Fig. 1 B). We also avoided putative endocytosis motifs (D'Hondt et al., 2000), endoplasmic reticulum (ER) retention motifs (Zerangue et al., 1999), and ER export signals (Ma et al., 2001, 2002) (Fig. 1C). We found no apparent post-Golgi surface-promoting signal. Also we avoided the negatively charged amphipathic helix that may line the permeation pathway (Miyazawa et al., 1999).
HEK293T cell transfection. HEK293T cells were maintained in DMEM high-glucose medium supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin. Cells were plated onto 35 mm Petri dishes with glass coverslip bottoms that were coated with 1% gelatin in PBS, incubated at 37°C in an incubator until they were 60-70% confluent, and then transiently transfected using either TransFast (Promega) or Lipofectamine 2000 (Invitrogen, Carlsbad, CA). In both cases, for each dish, 1 µg of cDNA was mixed with the transfection reagent and medium according to the manufacturer's protocol. Experiments were performed 2-3 d later.
Electrophysiology. The cells were visualized under Hoffman modulation contrast illumination with an inverted microscope [Olympus IMT-2, DPlan 10x 0.25 numerical aperture (NA) and MPlan 60x 0.70 NA)] under fluorescence illumination (Hg lamp) to identify cells transfected with GFP or the fluorescently tagged nicotinic receptors. Patch electrodes (3-6 M ) were filled with pipette solutions containing (in mM): 88 KH2PO4, 4.5 MgCl2, 0.9 EGTA, 9 HEPES, 0.4 CaCl2, 14 creatine phosphate (Tris salt), 4 Mg-ATP, 0.3 GTP (Tris salt), pH 7.4 with KOH. The extracellular solution (ECS) was (in mM): 150 NaCl, 4 KCl, 2 CaCl2, 2 MgCl2, 10 HEPES, and 10 D-glucose, pH 7.4. Atropine (1 µM) was included in the ECS to block muscarinic ACh receptors. In patch recordings of midbrain neurons, sodium channels, NMDA, AMPA, and GABAA receptors were blocked by 0.5 µM TTX, 50 µM AP5, 10 µM CNQX, and 20 µM bicuculline. Standard whole-cell recordings were made using an Axopatch 1-D amplifier, low-pass filtered at 2-5 kHz, and digitized on-line at 20 kHz (pClamp 8; Axon Instruments, Foster City, CA). Series resistance was compensated 80%, and the membrane potential was held at -60 mV. Recorded potentials were corrected for junction potential.
ACh was delivered using a two-barrel glass theta tube (outer diameter  200 µm; pulled from 1.5 mm diameter theta borosilicate tubing) connected to a piezo-electric translator (Burleigh LSS-3100, Fishers, NY). Each barrel of the theta tube was fed from a 12-way manifold. This allowed up to 12 different solutions to be fed in either the control or agonist barrel. Agonists were applied for 50 msec, which was triggered by pCLAMP 8 software. The voltage input to the high-voltage amplifier (Burleigh PZ-150M) used to drive the piezo translator was filtered at 150 Hz by an eight-pole Bessel filter (Frequency Devices, Haverhill, MA) to reduce oscillations from rapid pipette movement. Solution exchange rates measured from open tip junction potential changes during application with 10% ECS were typically 300 µsec (10-90% peak time).
Calcium imaging. HEK293T cells were loaded in the dark for 1 hr with fura-2 AM (1.25 µM) in pluronic acid (0.01%). During imaging, cells were perfused continuously with extracellular solution. Images were obtained with an inverted fluorescence microscope (IX71 Olympus; Lambda LS 175 W Xe arc lamp; Sutter Instrument Co., Novato, CA). Ratio images were obtained by acquiring pairs of images at alternate excitation wavelengths (340/380 nm, Lambda 10-2 filter wheel, Sutter Instrument Co) with an 80 nm bandpass emission centered at 510 nm. Images were collected with a 40x oil immersion objective (UApo/340 1.35 NA) and captured (one pair per 2 sec) on a Photometrics Cascade CCD camera (16 bit resolution; Photometrics, Tucson, AZ) under the control of Slidebook 4.0 imaging software (Intelligent Imaging Innovations, Denver, CO).
ECSs with [methyllycaconitine (MLA) or dihydro--erythroidine (DHE) + MLA] and without nicotinic receptor blockers were delivered via bath perfusion using a peristaltic pump. For imaging in midbrain neurons, sodium channels, NMDA, AMPA, and GABAA receptors were blocked by including 0.5 µM TTX, 50 µM AP5, 10 µM CNQX, and 20 µM bicuculline in the bath perfusate. The agonist ACh (300 µM) was delivered for either 2 or 5 sec using a U-tube. A silica tube (150 µm inner diameter and 360 µm outer diameter; TSP150375; Polymicro Technologies, Phoenix, AZ) emanated from the pore of the U-tube and directed agonist outflow. Negative pressure was constantly applied to the U-tube to prevent agonist leakage and desensitization of receptors. For this reason, agonists were applied for at least 2 sec to overcome the delay (500 msec) in delivery of agonist produced by the negative pressure. The agonist delivery via the U-tube was tested each day before and intermittently during the experiment by visual inspection of the output of phenol red added to the ECS.
Selection of optimal fluorescently labeled constructs in HEK293 cells. We compared the functional expression of the fluorescently tagged receptors with that of wild type (WT) by recording voltage-clamp currents induced by 300 µM ACh in HEK293T cells. This ACh concentration elicited currents from several of the fluorescently tagged receptor constructs (Fig. 2 A), and all of the responses were antagonized reversibly by DHE (10 µM). At the peak of the response waveform, the WT 42 response was 396 ± 86 pA (Fig. 2 B). The 4-YFP-M 2 response (206 ± 62 pA) was 52% of the WT value. The 4 2-CFP-M (613 ± 169 pA) and the (4-YFP-M)(2-CFP-M) responses (641 ± 275 pA) were actually larger than the WT response (155 and 162%, respectively), but not significantly so. Constructs with fluorescent proteins outside the M3-M4 loop function less well (Fig. 2 B). The N-terminal 4-YFP (two flanking linkers) 2 receptor displayed no observable ACh response. The peak N-terminal 4-YFP 2 response was much smaller than for WT, but nonetheless measurable (41 ± 12 pA). Inserting CFP into the C terminus of 2 reduced the ACh response by 80% (81 ± 15 pA), relative to the WT. We compared the waveforms of the (4-YFP-M)(2-CFP-M) and WT ACh-induced currents. We found no differences. The (4-YFP-M)(2-CFP-M) and WT responses had decay time constants of 183 ± 45 msec (n = 5) and 159 ± 32 msec (n = 6), respectively. Furthermore, the total charge carried by WT 42 and (4-YFP-M)(2-CFP-M) currents did not differ significantly (140 ± 55 and 103 ± 24 pC, respectively; p = 0.6; t test).
View larger version (40K):
[in this window]
[in a new window]
|
Figure 2. Whole-cell recordings and fura-2 recordings of Ca2+ accumulation for the various fluorescently tagged nicotinic receptor constructs expressed in HEK293T cells. ACh (300 µM) was applied every 60 sec (duration 50 msec). A, Exemplar ACh-induced currents of each of the fluorescently tagged chimeras. Currents were inhibited with continuous coapplication of the competitive antagonist DHE (10 µM) and recovered during washing. B, Peak current response with 300 µM ACh of the fluorescently tagged and WT 42 receptors. Numbers represent mean ± SEM. Asterisks represent a significant difference (p < 0.05) as compared with WT. Ratiometric measurements (F340/F380) of fura-2 AM loaded in HEK293T cells show that intracellular calcium rises during a 5 sec application of 300 µM ACh for WT 42, 4-YFP-M 2, and 4 2-CFP-M (C). Application of 300 µM ACh to cells expressing 4-YFP-N1 2 and 4 2-CFP-C showed no change in levels of intracellular calcium. Each trace represents a single cell. D, Box plot of peak calcium fluxes for each of the fluorescently labeled constructs. Bottom, middle, and top line of the box represents 25%, median, and 75%, respectively. Bottom error bar represents 5%, and the top error bar represents 95%. Asterisks represent a significant difference (p < 0.05) as compared with WT.
|
|
We also compared Ca2+ flux through the fluorescently tagged nAChR constructs by ratiometric imaging of fura-2 AM in HEK293T cells. We found essentially normal responses for the constructs with fluorescent proteins in the M3-M4 loop. Application of 300 µM ACh for 5 sec resulted in detectable rises in intracellular calcium for these M3-M4 loop constructs (Fig. 2C). The activation of WT 42 nAChRs produced a 9.4% (7.0, 20.5%) [median (25th, 75th percentile)] increase in intracellular Ca2+ (Fig. 2 D). The 4-YFP-M 2 receptor displayed a comparable increase in intracellular Ca2+ (7.4, 4.9, 10.9%). Cells expressing the 4 2-CFP-M receptor showed increases in intracellular calcium of 15.1% (13.5, 19.3%).
On the other hand, cells with the N-terminal 4-YFP and C-terminal 2-CFP constructs displayed no significant ACh-induced Ca2+ increases. These results are consistent with the electrophysiological data and demonstrate that the insertion of fluorescent proteins in the N-terminal and C-terminal positions of 4 and 2 nicotinic subunits impairs the ACh response of the receptors.
To compare ACh sensitivity of the M3-M4 constructs (4-YFP-M, 2-CFP-M) with that of the WT, we measured their ACh concentration-response relations (Fig. 3). In all cases, the ACh responses peaked at 1 mM and decreased at 3 mM (data not shown), presumably because of open channel block by ACh at high concentrations. The concentration-response relations were fitted with the sum of two Hill equations (Buisson and Bertrand, 2001). Similar to previous results (Buisson and Bertrand, 2001), for WT 42 the high-affinity EC50 was 1.9 ± 1.2 µM (mean ± SEM), and the low-affinity EC50 was 228 ± 85 µM (Fig. 3A). The higher- and lower-affinity components of the dose-response relations probably represent populations of nicotinic receptors with stoichiometry (4)2(2)3 and (4)3(2)2, respectively (Nelson et al., 2003). The fluorescently tagged nicotinic receptors showed minimal changes in the high- and low-affinity EC50 for ACh (Fig. 3B-D).
Mesencephalic dopaminergic neuronal culture and transfection. Brain tissue containing primary mesencephalic neurons of the ventral tegmental area from embryonic day 14 BDF1 mouse embryos (Shimoda et al., 1992) was incubated in plating medium (Neurobasal, 2% B27, 0.5 mM glutamax, 5% horse serum) with 1 mg/ml papain at 37°C for 20 min. The digested tissue was mechanically dispersed, laid on 5% bovine serum albumin (Fraction V; Sigma, St. Louis, MO) in PBS, spun for 6 min at 300 x g, and resuspended in plating medium. Cells (150,000) were plated onto 35 mm Petri dishes with glass coverslip bottoms (MatTek Corp., Ashland, MA) that were coated with polyethyleneimine (CC-4195; BioWhittaker, Walkersville, MD) in borate buffer (CC-4196; BioWhittaker) and laminin (Sigma).
To transfect neurons, 1 µg of each cDNA was mixed with 120 µg of Nupherin-neuron (Biomol Research Laboratories, Plymouth Meeting, PA) in 400 µl of Neurobasal medium without phenol red; 3 µl of Lipofectamine 2000 was mixed in 400 µl of Neurobasal media. The two solutions were incubated for 15 min at room temperature, combined, and incubated for another 45 min at room temperature. The medium of the mesencephalic cultures was replaced with plating medium without serum, and the transfection medium was added. The culture dishes were spun at 250 x g for 5 min on a swinging bucket centrifuge and incubated for 30 min at 37°C. Then the transfection mixture was removed and replaced with the original medium. Experiments were performed on the cells 1-8 d after transfection.
Immunocytochemistry. Mesencephalic cultures were washed in PBS and fixed in chilled 4% paraformaldehyde in PBS. Cultures were rinsed twice for 10 min with PBS and then permeabilized with 0.25% Triton X-100 in PBS for 5 min. Cultures were rinsed twice for 5 min with PBS and blocked with 10% donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA) in PBS for 30 min at 37°C. The primary antibody was diluted in 3% donkey serum in PBS and incubated for 1 hr at 37°C. Cultures were then washed three times for 5 min with PBS. The secondary antibody was diluted in 3% donkey serum in PBS and incubated for 1 hr at 37°C. Finally, cultures were washed three times for 5 min with PBS.
The following primary and corresponding secondary antibodies and their dilutions were used in this study: rabbit anti-microtubule-associated protein 2 (MAP2) polyclonal antibody at 1:500 (AB5622; Chemicon, Temecula, CA), CY5-conjugated donkey anti-rabbit (Jackson ImmunoResearch Laboratories), mouse anti-dephosphorylated tau monoclonal antibody at 1:500 (MN1010; Pierce Endogen), CY5-conjugated donkey anti-mouse (Jackson ImmunoResearch Laboratories) at 1:500, guinea pig anti-4 nAChR polyclonal antibody at 1:50 (AB5590; Chemicon), Alexa Fluor 546-conjugated goat anti-guinea pig at 1:100 (A11074
[GenBank]
; Molecular Probes, Eugene, OR), rabbit anti-tyrosine hydroxylase (Pel-Freez, Rogers, AR) at 1:50, CY5-conjugated donkey anti-rabbit (Jackson ImmunoResearch Laboratories) at 1:100.
Direct confocal fluorescent imaging. Mesencephalic neuronal cultures were processed for immunocytochemistry and then imaged with an upright LSM 510 Meta laser-scanning confocal microscope (Zeiss) using an Achroplan IR 63x 0.9 NA water immersion objective. Pinhole diameter was 1 Airy Unit. Images were collected at a 12-bit intensity resolution over 1024 x 1024 pixels at a pixel dwell time of 6.4 µsec. Z-stacks of images were acquired with a Z step size that ranged between 0.7 and 1.4 µm. The 458 nm laser line was used to excite CFP, the 514 laser line was used to excite YFP, and the 633 nm laser line was used to excite CY5. CFP images were acquired using a 458 nm dichroic mirror and a bandpass filter of 475-495 nm. YFP fluorescence was collected using a 514 nm dichroic mirror and a long-pass filter of 530 nm. CY5 fluorescence was collected using a 633 dichroic mirror and a long-pass filter of 650 nm.
Confocal FRET images. Cultured mesencephalic neurons or HEK293T cells 2 d after transfection were visualized at room temperature in ECS. During the imaging sessions, 2 mM ascorbic acid was included in the ECS to prevent damage to dendrites during photobleaching.
A series of lambda stack X-Y images were collected from an upright LSM 510 Meta laser scanning confocal microscope using an Achroplan IR 63x 0.9 NA water immersion objective (Zeiss). For details, see Lansford et al. (2001) and Dickinson et al. (2001). Images were collected over a spectrum of wavelengths between 462.9 and 602 nm with bandwidths of 10.7 nm during excitation of CFP with the 458 nm laser line of an argon laser. Therefore, each pixel of the X-Y image has a complete spectral emission intensity between 462.9 and 602 nm, which is defined as the lambda stack. Pinhole was 1.32 Airy Units and a Z resolution of 2.0 µm. Images were collected at a 12-bit intensity resolution over 512 x 512 pixels at a pixel dwell time of 12.8 µsec. FRET was recorded by examining the dequenching of CFP during incremental photobleaching of YFP by the 514 nm argon laser line. Lambda stacks were acquired at various time points before and after photobleaching of YFP. These time points included 0, 0.13, 0.26, 0.52, 1.05, 1.57, 2.09, 6.81, and 12.05 min, which corresponded to 0, 1, 2, 4, 8, 12, 16, 52, and 92 complete photobleaching scans.
Although there is significant overlap in the emission spectra of YFP and CFP, we were able to separate the fluorescence contribution of each fluorescent protein at each pixel using a linear unmixing algorithm based on the spectral signatures of YFP and CFP created from reference lambda stack images of cells expressing either soluble YFP or soluble CFP, respectively (Dickinson et al., 2001; Lansford et al., 2001). These unmixed YFP and CFP images were analyzed for FRET efficiency.
Background fluorescence for both CFP and YFP was determined as the mean fluorescence intensity from an area containing cells that did not express the constructs, and it was subtracted from the overall intensity. Mean fluorescence intensities of CFP and YFP were determined by tracing the outer perimeter of the cell, the neuronal soma, or dendrite with a region of interest. In the case of FRET calculated for cell membrane surface receptors with the application of 1 µm phorbol 12,13 myristate acetate (PMA) (Sigma), however, mean fluorescence intensity was calculated for only the outer shell of fluorescence of the cell and did not include the cell interior. Mean fluorescence intensities were determined using ImageJ software v1.29u (National Institutes of Health, http://rsb.info.nih.gov/ij/). Fluorescence intensities of CFP and YFP at various time points after photobleaching were normalized to time 0 (100%). FRET efficiency (E) was calculated:
 | (1) |
IDA represents the normalized fluorescence intensity of CFP (100%) in the presence of both donor and acceptor. ID represents the normalized fluorescence intensity of CFP in the presence of donor only. The ID value was extrapolated from a scatter plot of the percentage increase of CFP versus the percentage decrease of YFP for each cell.
Statistics. Data are reported as mean ± SE or median with the 25th and 75th percentiles for parametric and nonparametric data, respectively. Significant differences (at p < 0.05) were determined between two groups using a t test for continuous data meeting parametric assumptions of normality and equal variance. Otherwise the Mann-Whitney rank sum test was used for nonparametric data. To compare significant differences between more than two groups of data meeting assumptions of normality and equal variance, a one-way ANOVA was performed followed by a Tukey test for all pair-wise comparisons. If the groups of data did not meet the parametric assumptions of normality and equal variance, then a Kruskal-Wallis one-way ANOVA on ranks was performed using the Dunn's method as post hoc pair-wise analysis.
|
Results
|
|---|
Incubation in nicotine upregulates function
Untransfected ventral midbrain cultures were incubated in 1 µM nicotine for periods of 1-6 d. Blood levels of nicotine in smokers usually range between 125 and 275 nM, occasionally reaching 500 nM (Rowell, 2002); however, brain levels of nicotine are approximately three times higher than in blood (Rowell, 2002), comparable with the 1 µM value in our experiments (Rowell and Li, 1997). We then measured fura-2 signals in these cultures to monitor changes in functional nAChRs. These signals were measured in the presence of atropine, tetrodotoxin, CNQX, AP5, and bicuculline, so that the Ca2+ signals mostly represent direct influx rather than the amplifying effects of other channels. We focused on data measured at 2.75 and 4.75 sec after the start of a 2 sec ACh pulse, to concentrate on fluxes linked most directly to the receptor activation. Furthermore, we noted that MLA blocked the fura-2 signals by 29% (grand average of >300 cells under all conditions), indicating that MLA-insensitive components of Ca2+ flux dominate the signals. Most researchers believe that MLA-sensitive components are primarily 7-mediated (Kawai and Berg, 2001; Alkondon and Albuquerque, 2002) and that the DHE sensitive signals arise in part via 4- and 2-containing receptors (Buisson et al., 1996; Marks et al., 1999). Signals were indeed blocked further by 32% by (MLA + DHE).
We found statistically significant upregulation of the MLA-insensitive response to 300 µM ACh after 2 d and after 4-6 d of nicotine incubation (Fig. 4), but not after just 1 d of nicotine exposure (data not shown). We also found statistically significant upregulation of the MLA-sensitive response to 300 µM ACh at 2 d (Fig. 4). There was a residual ACh-elicited response after block with MLA and DHE that showed no significant difference in functional expression with nicotine incubation. These data extend results obtained previously for 7 receptors in cortical cultures (Kawai and Berg, 2001) and show for the first time that midbrain neurons display upregulated nAChR-dependent Ca2+ influx during chronic exposure to nicotine.
Although the average block by (MLA + DHE) was 61%, the (MLA + DHE)-insensitive responses were found in some 60% of cells. In general, we found that the (MLA + DHE)-resistant responses persisted longer than MLA + DHE-sensitive components, which are represented primarily by the 4.75 sec time scale summarized in Figure 4, B and D. These observations agree with reports that the (MLA + DHE)-insensitive nicotinic responses, which are found in the VTA and substantia nigra, have slower kinetics and are activated by longer applications of ACh (2000 msec) (Wooltorton et al., 2003). These results also are consistent with observations that mouse brain displays two classes of nicotinic responses that are not blocked by -bungarotoxin: DHE-sensitive and DHE-resistant responses (Marks et al., 1999). The DHE-sensitive component is likely 42. The DHE-resistant receptor is widely distributed, contains the 2 subunit, but is not 42 (Marks et al., 1999).
The upregulation of apparent 42-receptor responses during chronic nicotine encouraged us to study 42 subunit assembly in individual cells. We chose FRET analysis because FRET permits the study of subunit assembly in subcellular compartments of living cells.
Correct localization of the M3-M4 fluorescent 4 and 2 constructs expressed in midbrain neurons
We inserted YFP and CFP at various locations into the 4 and 2 subunits, respectively (see Materials and Methods) (Fig. 1). The constructs with the fluorescent protein inserted into the intracellular M3-M4 loop of the 4 and 2 nicotinic receptors showed relatively normal function and expression levels, as judged by ACh-induced currents (Fig. 2), by ACh-induced Ca2+ accumulation (see Materials and Methods) (Fig. 2), and by dose-response relations (Fig. 3), and were used for these experiments. We used the Nupherin-Neuron system, which produces good expression in hippocampal cultures (Slimko et al., 2002), for efficient transient expression of the fluorescent constructs in ventral midbrain cultures. We found highly significant increases in nicotinic responses of the transfected neurons (Fig. 5).
These increased responses at the whole-cell level do not address the following question: are the fluorescent subunits expressed in the same locations as WT 42 receptors? In designing the fluorescent subunits, we avoided placing the fluorescent moieties in regions that might compromise putative phosphorylation sites, potential trafficking signals, and the negatively charged amphipathic helix that may line the permeation pathway (Miyazawa et al., 1999). To determine whether the fluorescently labeled nAChR constructs in fact were targeted correctly to subcellular neuronal regions, we compared the subcellular localization of fluorescently tagged 42 receptors with that of endogenous 4-containing receptors. In ventral midbrain cultures transfected with 4-YFP-M 2, 4 2-CFP-M, or a combination of the two fluorescent subunits, the receptors were localized mainly in the soma and dendrites of neurons and mostly excluded from axons. The distribution of 4-YFP-M or 2-CFP-M overlapped with the somatodendritic marker MAP2 (Fig. 6A,B,D), whereas neither 4-YFP-M nor 2-CFP-M colocalized with the axonal marker dephospho-tau (data shown only for 4-YFP-M) (Fig. 6C). Fluorescently tagged 4 or 2 were distributed evenly within the somata and dendrites, and most receptors were excluded from the nucleus (Fig. 6A-D). No strong surface fluorescence was evident, although the electrophysiological experiments confirmed the presence of functional surface receptors (Fig. 5A).
Immunocytochemistry was performed on untransfected midbrain neurons using an anti-4 antibody for endogenous 4-containing nAChRs (Fig. 6 E). Staining patterns resembled those of the transfected 4-YFP-M. Endogenous 4 nAChRs were also localized mainly in the soma and dendrites. Double labeling with tyrosine hydroxylase indicated that some 4-containing neurons were dopaminergic. Therefore, the fluorescently tagged nicotinic receptor constructs express in the correct subcellular regions.
Incubation in nicotine upregulates levels of 4 nAChR subunits
In transfected cultured midbrain neurons, we assessed the level of the fluorescently tagged 4 subunits in individually identified neurons over a 24 hr period. We found that neurons incubated in 1 µM nicotine showed a significant increase in 4 subunit fluorescence in the dendrites (143 ± 17%), whereas the soma showed only a slight increase (115 ± 31%) (Fig. 7). In contrast, controls without nicotine showed a decrease in 4 expression in both soma and dendrites (22 and 43%, respectively) over a 24 hr period, probably because of the transient nature of expression of the nicotinic receptor with transfection. Therefore the dendrites of cells incubated in nicotine show a highly significant 2.5-fold greater 4 receptor level than control cells not incubated in nicotine. We found similar increases in the CFP fluorescence levels from the 2 receptor subunits that were cotransfected with the 4-YFP receptors; however, because the YFP moieties in the 4 subunits quench the 2-CFP fluorescence (see below) and therefore distort the quantitative data, the 2-CFP data are not presented in detail. The data suggest that the functional upregulation of DHE-sensitive and MLA-resistant responses is contributed by increased 42 nAChRs in nicotine-treated neurons. Because in this set of experiments transfected neurons overexpress nicotinic receptors, we expected that the upregulation would be accelerated and would thus appear earlier than the functional upregulation for the endogenous receptors (Fig. 4).
FRET signals are large, reveal surface translocation, and generalize to heteromeric glutamate-gated chloride (GluCl) channels but retain specificity for correctly assembled receptors
We measured FRET between the CFP and YFP moieties of the (4-YFP-M)(2-CFP-M) receptors in both HEK293T cells (Fig. 8) and midbrain neurons (see Fig. 10). We first describe the results for HEK293T cells. Photobleaching the acceptor YFP dequenched the CFP fluorescence in nearly all cells, over a time course of 2 min (Fig. 8A,C-F). We fitted exponential time courses to the initial phase of either the increase in CFP fluorescence ( = 0.37 ± 0.09 min) or the decrease in YFP fluorescence ( = 0.36 ± 0.03 min). On average, the time constants for YFP bleach and for CFP dequench differed by <1%, as expected if the YFP bleach and the CFP dequench arise from the same process. Occasionally there were longer components to the bleaching and dequench processes (Fig. 8F), possibly arising from slight photobleaching of CFP, although we have not systematically analyzed this. To quantify FRET between the nicotinic subunits, we constructed a scatterplot of the photobleach-induced changes in CFP and YFP fluorescence of each cell during the entire 12 min of bleaching (Fig. 8G) and then fitted a regression line to the scatterplot. The slope of the regression line ranged between 0.08 and 1.12 in these experiments on 42 nAChRs.
We also exposed transfected HEK293T cells cotransfected with PKC to PMA. Transfected (4-YFP-M)(2-CFP-M) nAChRs in HEK293T cells displayed modest surface expression; however, incubation in 1 µM PMA (a PKC activator) for  30 min robustly translocated 42 nAChRs to the vicinity of the cell surface (Fig. 8A). Electrophysiological experiments confirmed that PMA increased surface nAChR function, because PMA also increased the ACh response by threefold (Fig. 8B) at [ACh] levels that mostly saturate the dose-response relation.
The PMA-treated cells displayed greater FRET efficiency, defined by the slope of the scatterplot, from 0.30 ± 0.06 in the untreated cells to 0.60 ± 0.10 in the PMA-treated cells. According to Equation 1, these values imply a FRET efficiency of 22 ± 4% for the untreated cells and 36 ± 4% for the PMA-treated cells, respectively (significantly different at p = 0.017; t test) (Fig. 8H). Therefore, FRET efficiency was greater for the surface than the intracellular 42 receptors (Table 1). These data show a larger proportion of assembled receptors on or near the cell surface than in the interior, stimulated by PKC.
GluCl is a member of the nicotinic receptor superfamily, consisting of and subunits. We studied GluCl-CFP and GluCl-YFP constructs (Slimko and Lester, 2003), with the fluorescent tags in the M3-M4 intracellular loops. Coexpression of the GluCl-CFP and GluCl-YFP subunits showed dequenching of CFP in response to YFP photobleaching (Fig. 9A,C,E). FRET efficiency between the and GluCl subunits (23 ± 3%) was similar to that between the 4-YFP-M and 2-CFP-M.
The large FRET efficiencies (25%) that we reported are consistent with that found for other heteromeric ion channels (Riven et al., 2003). We anticipated these relatively large signals because the two channel types studied here are heteromultimers consisting of the two subunit types, so that all subunits of each individual channel molecule are either donors or acceptors. Nonetheless, some knowledgeable colleagues were surprised at these FRET efficiencies, especially in view of the data reported in Figure 8 and below (Fig. 10) showing that even larger efficiencies occur in specific compartments and after pharmacological manipulations. Therefore we report additional observations that support these measurements. (1) Measurements performed with the 4 and 2 constructs in a separate lab, using the three-cube system, verified the FRET ratios that we observed (James Fisher and Baljit S. Khakh, personal communication). (2) We have also constructed the reciprocal constructs, 4-CFP and 2-YFP, and we find similarly large measures of FRET (J. Revie, unpublished data). Comparably large FRET efficiencies are also measured in homomeric ion channels when a large excess of acceptor is used, but presumably at the expense of overall signal amplitude (Glauner et al., 1999).
We wished to use the fluorescent subunits to monitor properly assembled receptors. To test the idea that FRET assesses assembled receptors, with no contribution from isolated 4 and 2 subunits, we exploited the fact that GluCl subunits do not form heteromers with nicotinic receptor subunits. We asked whether a FRET signal occurred when a nicotinic receptor subunit (4-YFP-M 2) was coexpressed with a GluCl receptor subunit (GluCl-CFP). Coexpression of the GluCl-CFP and 4-YFP-M 2 subunits produced little or no reduction in the CFP signal when YFP was photobleached (Fig. 9A,B,D). These experiments show that the FRET observed between 4-YFP-M and 2-CFP-M provides a sensitive assay for assembled nicotinic receptors.
Dendritic receptors show a higher percentage of assembly than somatic receptors
The results presented in Figures 8 and 9 show that FRET between transfected functional, fluorescently labeled 4 and 2 nAChR subunits provides a sensitive, specific assay for correctly assembled 42 receptors. We therefore studied FRET between the 4-YFP-M and 2-CFP-M nAChR subunits in transfected midbrain cultures (Fig. 10). As in HEK cells, we found robust FRET. We compared FRET in the dendrites with that in the soma. Photobleaching of YFP indicated a stronger dequenching of CFP fluorescence in the dendrites than the soma. The average slope of the scatterplots of CFP increases versus YFP decreases in the dendrites was significantly greater (p < 0.01) than that in the soma, corresponding to respective FRET efficiencies of 48 ± 3 and 32 ± 2%, respectively.
Incubation in nicotine upregulates assembly of somatic 42 receptors
These results show that FRET between transfected functional, fluorescently labeled 4 and 2 nAChR subunits can address the following question: does nicotine incubation increase the assembly of 42 receptors in midbrain neurons? We assessed assembly of 42 receptors under conditions of nicotine incubation in cultures transfected with the fluorescent 42 receptors. In each of two experiments, we observed significant increases in receptor assembly even after 1 d of nicotine incubation. When measured only in the soma, FRET was 27 ± 3% in control cells and 37 ± 2% in incubated cells. Transfected cells express much higher receptor levels than untransfected cells (Fig. 5); it was therefore not surprising to find significant changes in receptor assembly even after 1 d of nicotine incubation, although untransfected cells required at least 2 d of incubation with nicotine to show increased responses to agonist (Fig. 4).
We next asked whether the upregulation of functional receptors by nicotine incubation is also contributed by increased assembly of receptors in the dendrites of the cultured midbrain neurons. We found no significant nicotine-induced differences in dendritic FRET signals, in contrast to the FRET increase that we found in the soma (Fig. 11).
|
Discussion
|
|---|
|
Top
Abstract
Introduction
