|Home | About | Journals | Submit | Contact Us | Français|
We report here the construction of a novel knock-in mouse expressing chimeric α3 nAChR subunits with pharmacological sensitivity to α-bungarotoxin (αBTX). Sensitivity was generated by substituting five amino acids in the loop C (β9–β10) region of the mouse α3 subunit with the corresponding residues from the α1 subunit of the muscle type receptor from Torpedo californica. To demonstrate the utility of the underlying concept, expressed α3 subunits were characterized in the superior cervical ganglia (SCG) of homozygous knock-in mice, where the synaptic architecture of postsynaptic α3-containing nAChR clusters could now, for the first time, be directly visualized and interrogated by live-staining with rhodamine-conjugated αBTX. Consistent with the postsynaptic localization of ganglionic nAChRs, the αBTX-labeled puncta colocalized with a marker for synaptic varicosities. Following in vivo deafferentation, these puncta persisted but with significant changes in intensity and distribution that varied with the length of the recovery period. Compound action potentials and excitatory postsynaptic potentials recorded from SCG of mice homozygous for α3 were abolished by 100 nM αBTX, even in an α7 null background, demonstrating that synaptic throughput in the SCG is completely dependent on the α3-subunit. In addition, we observed that the genetic background of various inbred and outbred mouse lines greatly affects the functional expression of α3-nAChRs, suggesting a powerful new approach for exploring the molecular mechanisms underlying receptor assembly and trafficking. As αBTX-sensitive sequences can be readily introduced into other nicotinic receptor subunits normally insensitive to αBTX, the findings described here should be applicable to many other receptors.
Multiple nicotinic acetylcholine receptor (nAChR) subtypes populate the peripheral and central nervous systems of vertebrates. At present, 5 neuromuscular nicotinic subunits (α1, β1, γ, δ, and ε) and 12 neuronal nicotinic subunits (α2–α10 and β2–β4) have been identified and cloned (Karlin, 2002; Colquhoun et al., 2003, Millar and Gotti, 2009). Unlike muscle-type nAChRs, which are constrained to a single pentameric arrangement in adult tissue, neuronal nAChRs can assemble with varied subunit compositions and stoichiometries, resulting in an undetermined number of receptor isoforms (McGehee and Role, 1995; Role and Berg, 1996; Hogg et al., 2003). Alterations in nicotinic receptor function are thought to contribute to the pathological mechanisms underlying a wide range of neurological and psychiatric disorders, including Alzheimer’s disease, Parkinson’s disease, schizophrenia, epilepsy, and drug addiction. Despite their recognized importance, the relationships between the subunit composition of most neuronal nAChR isoforms and their physiological roles in vivo remain unclear. New tools and approaches would help address these critical questions (Gotti et al., 2006; Gotti et al., 2007; Hogg and Bertrand, 2007).
Neuronal nAChRs containing the α3 subunit mediate fast excitatory neurotransmission in both sympathetic and parasympathetic autonomic ganglia in mammals (Mandelzys et al., 1995; Perry et al., 2002; Rassadi et al., 2005). Of all the neuronal nAChR genes, α3 is the only single subunit that produces a lethal phenotype when deleted in inbred mice (Xu et al., 1999). There is also mounting evidence that α3-containing receptors are expressed in select regions of the central nervous system, including the medial habenula, pineal gland, interpeduncular nucleus, dorsal horn of the spinal cord, hypothalamus, cerebellum, locus coeruleus, inferior colliculus, and hippocampus (Sudweeks and Yakel, 2000; Yeh et al., 2001; Perry et al., 2002; Whiteaker et al., 2002; Liu et al., 2003; Hernandez et al., 2004; Turner and Kellar, 2005; Grady et al., 2009).
Previously, using heterologous expression studies in Xenopus oocytes, we demonstrated that the α3 neuronal nAChR subunit can be mutated into a αBTX-sensitive form by substituting from one to five amino acids in the loop C region of the α3 subunit with the corresponding residues from the α1 subunit of Torpedo californica (Levandoski et al., 1999). In the oocyte expression system, these chimeric α3-containing nAChRs exhibited EC50 values for ACh ranging from 50–500 µM and IC50 values for block by αBTX ranging from 20–200 nM, depending on (1) the specific amino acid substituted, (2) the total number of residues mutated, and (3) whether β2 or β4 was co-expressed. These studies led us to hypothesize that minimally mutated αBTX-sensitive α3 nAChR subunits could be utilized as part of a knock-in strategy for the study of α3-containing neuronal nAChRs in vivo. Toward this goal, a gene-targeting vector was designed to introduce five mutations in the mouse α3 nAChR subunit: Y184W, E187W, I188V, K189Y, and N191T (Torpedo numbering) (Levandoski et al., 1999). Here we report that these mutations have been introduced into the genome of mouse ES cells using gene replacement technology and that functional, αBTX-sensitive α3-containing nAChRs are expressed in SCG neurons, opening the door to new possibilities of probing the function and anatomical distribution of α3-containing nAChRs in the central nervous system. Furthermore, this novel knock-in mouse has increased our understanding of the well-characterized SCG system by enabling an imaging analysis of the dynamics of neuronal nAChR cluster size and distribution following preganglionic denervation at a level of analysis previously unattainable.
A gene-targeting vector containing genomic, mouse α3 nAChR DNA and a neomycin-resistance cassette in pBluescript SK (−) (Stratagene, La Jolla, CA) was the generous gift of Dr. Changhai Cui, Salk Institute. The genomic DNA sequence in this construct included exonV of the CHRNA3 gene along with ~2,000 to 3,000 of flanking, homologous nucleotides. PCR-based mutagenesis (QuikChange XL, Stratagene) was used to generate two alterations in the targeting vector. First, seven nucleotides were altered to produce the α3 substitutions in the α3 gene product (YKHEIKYN to WKHWVYYT), along with a BstZ17I restriction enzyme cleavage site. Mutagenic PCR primers were designed according to the guidelines in the QuikChange technical manual: forward (5'-GCCATCATTAAAGCCCCGGGCTGGAAACATTGGGTAT ACTACACCTGCTGT GAGGAGATCTACC-3') and reverse (5'-GGTAGATCTCCTCACAGCAGGTGTAGT ATACCCAATGTTTCCAGCC CGGGGCTTTAATGATGGC-3'). Positive clones were identified by a restriction analysis using BstZ17I.
Second, a 57 nucleotide region containing a Cre-recombinase recognition sequence (loxP) positioned just 5' to exonV in the original targeting vector was removed. The sequences of the mutagenic primers used for the loxP removal were as follows: forward (5'-GGTGATAAGTGTGGCAAATTATGTGCC AGCAGAGGCGGG GGTGGTGG TGGTGAATAACCAATGTGGG-3’) and reverse (5’-CCCACATTGGTTATTC ACCACCACCACCCCCGCCTCTGCTGGCACAT AATTTGCCACACTTATCACC-3'). Positive clones were identified by standard PCR methods. Plasmid DNA was isolated using Mini or Maxi Plasmid Kits (Qiagen Venlo, NL).
AB2.2 ES cells derived from a 129S7 male embryo were grown on mitotically inactive SNL76/7 cells and used for targeting. Ten million ES cells were electroporated with 25 µg of α3 targeting vector linearized with SalI, and G418 selection was initiated after 24 h. One-hundred and ninety-two neomycin resistant clones were selected for further analysis. Correctly targeted ES cell clones were identified as described (Ramirez-Solis et al., 1995).
All breeding and procedures were carried out according to approved institutional procedures at Brown University Animal Facility and in agreement with the NIH Guide for the Use and Care of Laboratory Animals. The targeted ES cells were grown to 90% confluence and trypsinized before injection. E3.5 blastocysts were derived from C57BL/6-Tyrc-Brd female mice and injected with 12–20 ES cells. The injected blastocysts were implanted into the uteri of day 2.5 pseudo-pregnant females for generation of chimeras. About 8–10 injected embryos were implanted per uterine horn.
The resulting male chimeras were mated with C57BL/6-Tyrc-Brd females to obtain F1 progeny. Thus, the strain carrying the germ line transmitted α3 allele (designated Chrna3tm1.0Hwrt) (Fig. 1A) was obtained and initially maintained on a mixed C57BL/6-129S7 background. To obtain mutant mice carrying α3 with Neo cassette deleted (allele designated Chrna3tm1.1Hwrt), heterozygous Chrna3tm1.0Hwrt/+ mutants were mated with 129S1-Hprt1tm1(cre)Mnn/J mice (The Jackson Laboratory, Bar Harbor, ME) and the offspring were screened for the presence of the α3 mutations and for the deletion of Neo (Chrna3tm1.1Hwrt/+ mice). Congenic mice were generated by 10 consecutive backcrosses of Chrna3tm1.1Hwrt/+ to wild-type C57BL/6 mice. To obtain Chrna3tm1.1Hwrt/tm1.1Hwrt Chrna7tm1Bay/tm1Bay double homozygous animals (α3-homozygous α7(−/−)), Chrna3tm1.1Hwrt/+ animals were mated with Chrna7tm1Bay/tm1Bay mice (B6.129S7-Chrna7tm1Bay/J; the Jackson Laboratory, Bar Harbor, ME) and double heterozygous mice were selected. In the next step Chrna3tm1.1Hwrt/+ Chrna7tm1Bay/+ animals were intercrossed and screened for Chrna3tm1.1Hwrt/+ Chrna7tm1Bay/tm1Bay. In the final step, Chrna3tm1.1Hwrt/+ Chrna7tm1Bay/tm1Bay and Chrna3tm1.1Hwrt/+ Chrna7tm1Bay/+ animals were intercrossed and offspring selected for Chrna3tm1.1Hwrt/tm1.1HwrtChrna7tm1Bay/tm1Bay.
ES cell clones and transgenic animals were screened for the knock-in allele using Southern blot analysis. The Southern strategy utilized a KpnI restriction cleavage site in the targeting construct. Digestion with KpnI was expected to produce 17.4 kb fragment from the wild-type allele, and 8.9 kb and 10.5 kb fragment in the case of the mutated allele. Two 0.6 kb nucleotide probes were designed to anneal outside of the targeted region of Chrna3. The Southern probes were generated by PCR, purified using QIAquick PCR Purification Kit (Qiagen). The probes were labeled with 32P-dCTP by the random primer labeling method according to the manufacturer’s protocol in the Megaprime DNA Labeling Systems (Amersham, GE Healthcare UK). KpnI digested genomic DNA (10–15 µg /sample) from ES cell clones or transgenic animals was separated by agarose gel electrophoresis (1% gel; run at 50V for 16 h) and transferred to a nylon membrane, Hybond-XL (Amersham, GE Healthcare UK), by the capillary blotting method. Southern hybridizations were performed using standard protocols.
Mice were housed in Brown University Animal Care Facility and McGill University Animal Resource Center mouse rooms. Routine cage maintenance, including feeding and watering, was performed by the animal facility staff. Mice were maintained on a 12 hour light/dark cycle. Mating cages were typically set up with one male and one female. Pups were weaned at 21 days. Tail biopsies were used as tissue sources for DNA extraction and genotyping. All survival surgeries were performed using sterile procedures within the animal care facility mouse rooms. In cases in which euthanasia was necessary, mice were asphyxiated with CO2 according to Brown University’s policy on the “Use of CO2 as a Euthanasia Agent for Small Laboratory Animals” and guidelines established by the Canadian Council on Animal Care. All animal use procedures described in this study were reviewed and approved by the designated institutional animal use committee.
Pups were screened using two PCR-based strategies. The first method screened directly for the wild-type or recombinant allele using the following forward primers: GCTGGAAACATTGGGTATACTACACC specific for the knock-in allele and GGCTACAAACATGAAATCAAGTACAACTGC specific for the wild-type allele. The same reverse primer positioned 400 bp downstream of the targeted region CCGTAGAAGTTCCTCGTCTTTGGG was used in both reactions. “Triple-Master PCR System” (Eppendorf, Hamburg, DE) was used for all PCR-based screens.
The second PCR-based method for genotyping the α3-containing allele utilized a BstZ17I restriction site in the mutated region. The digestion-based PCR screen amplified a ~650 bp product with an internal restriction site. A forward primer GTGACCTACTTCCCGTTTGACTACC was designed to anneal ~250 bp upstream of the mutated region. The reverse primer CCGTAGAAGTTCCTCGTCTTTGGG was placed ~400 bp downstream of the targeted region. The PCR products were subjected to restriction digestion using BstZ17I (NE Biolabs, Ipswich, MA) and agarose gel electrophoresis (2% gels). Reaction products from wild-type alleles were resistant to digestion, whereas PCR products from heterozygous or homozygous mutants were partially or completely digested with BstZ17I, respectively.
Primers GGATCTCCTGTCATCTCACCTTGCT and ATCCTGATCGACAAGACCGGCTTC were used to identify mice positive for the Neo cassette. Primers CTTGTCCATCGTCATCACAGTCT TTGTG and GAGTCTAATTTTCTAACCTCTGCCCTATGC were used to identify mice negative for Neo.
Primers TGCTGTTTCACTGGTT GTGCGGCG and TGCCTTCTCTACACCTGCGGTGCT were used to identify mice positive for the Cre-recombinase expression cassette. Primers CCTGATTTTATTTCTATAGGACTGAAAGAC and TAAGTAATTATACTTACACAGTAGCTC TTC were used to identify mice negative for Cre.
The Chrna7tm1Bay allele in α7 knockout mice (B6.129S7-Chrna7tm1Bay/J) was detected by PCR as recommended by the Jackson Laboratory (Bar Harbor, ME) using primers IMR1002, IMR1003 and IMR1004.
SCG tissue from wild-type and knock-in mice was dissected and placed immediately into ice-cold RNA-later solution (Ambion, Austin, TX) and stored at 4° overnight. RNA isolation was performed using RNAqueous-4PCR (Ambion) according to the manufacturer’s protocol. In brief, SCG tissue was disrupted in 500 µl of lysis/binding buffer using a 1 ml Potter-Elvehjem tissue grinder (Kontes, Vineland, NJ). DNase I treatment was performed by adding 7 µl of 10X DNase buffer to 1 µl of DNase. To precipitate the RNA, 2.5 volumes of 100% EtOH were added, and the mixture was stored overnight at −20°C. The pellet was resuspended in 12 µl of RNase-free H2O. Total RNA concentration was measured by UV absorbance using a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). Reverse transcriptase (RT) reactions were performed using the RETROscript Kit (Ambion, Austin, TX) according to manufacturer’s protocol for 2-step RT-PCR without heat denaturation of RNA. 0.4–1 µg of total RNA was used per RT reaction. Random primers were used for all RT reactions. PCR primer pairs were designed to generate ~480 bp products from wild-type and mutant cDNA templates. If genomic DNA was present and amplified, ~8500 bp PCR products would be expected. A wild-type forward primer (CATCCAGTTTGAGGTGTCTATGTCTCAG) was designed to anneal within exon II of mouse Chrna3. Mutant and wild-type allele-specific reverse primers were designed to anneal within the targeted region of exon V (wild-type, GCAGTTGTACTTGATTTCATGTTTGTAGC and α3, CAGGTGTAGTATACCCAATGTTTCCAGC). TripleMaster PCR System (Eppendorf, Hamburg, DE) was used for all amplification reactions, which were assembled using 600 ng of total cDNA per reaction. Positive control primers designed to amplify the rig/S15 “housekeeping” gene sequence were supplied by Ambion and negative controls included reagent-only and RNA-only reactions.
To label recombinant α3-containing receptors, freshly dissected ganglia from α3-homozygous mice (>2 months of age) asphyxiated with CO2 were placed in L15 Air solution containing 100 nM rhodamine conjugated αBTX (rhoBTX, Sigma Chemical) for 1 h. Next, ganglia were placed in 0.5% paraformaldehyde (PFA) in 0.1 M phosphate buffer (PB) for 1 h, transferred into a solution of 3% Triton-X in phosphate buffered saline (PBS) for 3 h, and then incubated overnight with primary antibodies (polyclonal goat anti-VAChT (1:750, Chemicon, Temecula, CA) and monoclonal mouse anti-neurofilament (NFM; 1/10000, Sternberger) or polyclonal goat anti-post synaptic density 93 (PSD93; 1/100, Synaptic Systems) dissolved in PBS containing 10% normal donkey serum at 4°C. The next day ganglia were rinsed twice for 15 min in PBS and placed in secondary antibodies Alexa 488 donkey anti-goat antibodies (1/1000, Molecular Probes) or Alexa 402 donkey anti-mouse antibodies (1/1000, Molecular Probes) in PBS containing 10% normal donkey serum for 2.5 h at room temperature. The ganglia were then rinsed twice for 15 min with PBS and mounted with Vectashield (Vector Laboratories, Burlingame, CA). Nonspecific staining was assessed by processing the sections without primary antibody.
A series of z stacks (0.3 µm/plane) was obtained with a confocal microscope (LSM 510; Zeiss) and a 63x, 1.4 numerical aperture Plan Neofluor oil-immersion objective. In experiments with denervated ganglia, we set values for detector gain, detector offset, and amplifier offset by imaging control ganglia from the same mouse stained with rhoBTX and anti-VAChT and found values that captured the distribution of intensities within an image depth of 8 bits. We then used these acquisition parameters to image denervated ganglia. For colocalization experiments where the intensities of the signals were less important, we adjusted the acquisition settings to optimize the staining for synaptic proteins.
Images were quantified off-line with MetaMorph (Universal Imaging Corporation Downington, PA, USA). The representative images in figures of rhoBTX stained and VAChT-stained SCG were created by averaging three consecutive image planes from a z stack, cropping the averaged image, splitting the color components of the cropped image, thresholding each color channel individually to 2x the value of background intensity, and adding the two component images back together. Representative images of NFM/VAChT/rhoBTX triple stained ganglia were processed in a similar way. Occasionally, we used the low pass filter tool with a 2x2 convolution kernel to remove single pixel noise from our images.
Ganglia from mice greater than 2 months of age were pinned down in a small recording chamber (1.5 ml volume) with minutia pins, perfused continuously at 3–4 ml/min with oxygenated modified Tyrode's solution supplemented with glucose (5.6 mM) and choline (0.01 mM) (pH = 7.3–7.4) at 36–37°C, and viewed through a dissecting microscope (SMZ-10; Nikon, Tokyo, Japan). The preganglionic nerve in the cervical sympathetic trunk was connected to a stimulator (Warner instruments) with a suction electrode, and the postganglionic trunk was connected to an alternating current differential amplifier (DP-301; Warner Instruments, Hamden, CT) with a suction electrode. The postganglionic compound action potentials were amplified (1000x), filtered at 10 Hz (low-pass cutoff) and 3 kHz (high-pass cutoff), digitized, displayed, and stored on a Pentium II-based personal computer with Patchkit (Alembic Software, Montreal, Quebec, Canada). The data were analyzed off-line with Igor (WaveMetrics, Lake Oswego, OR). All drugs were added directly to the oxygenated Ringer's solution.
Glass microelectrodes with a resistance of 40–80 MΩ (G150F-4; Warner Instruments) made with a DMZ universal puller (Zeitz Instruments, Munich, Germany) and filled with 1 M potassium acetate (KAc) were used for intracellular recordings from ganglion cells. Stable intracellular recordings were achieved with a high inertial precision microdrive (Inchworm 8200; EXFO, Vanier, Quebec, Canada) attached to a micromanipulator (SM11; Narishige, Tokyo, Japan) to drive the electrode through the ganglion. The recording electrode was connected by a thin silver chlorided wire to the head stage of an Axoclamp 2A amplifier (Axon Instruments, Union City, CA) used in current-clamp mode. Depolarizing or hyperpolarizing constant current pulses were applied through the recording electrode, and membrane potentials were filtered at 10 kHz, sampled, displayed, and stored on a Pentium II-based personal computer. Stimulation and acquisition was done with Patchkit software (Alembic Software), and the data were analyzed with IGOR. The preganglionic nerve was stimulated with brief (0.1–0.2 ms) voltage pulses applied to the cervical sympathetic trunk through the suction electrode. αBTX (100 nM) and methyllycaconitine (MLA; 50 nM) were dissolved in oxygenated Tyrode's solution modified as above. Only neurons with membrane potentials more negative than −40 mV were included in this study.
SCG ganglia were removed under sterile conditions from neonatal mice (P12–P14) following asphyxiation with CO2. The ganglia were incubated at 37° C in a trypsin solution (1 mg/ml of trypsin from Worthington Biochemical Corp. Lakewood, NJ, USA, 3X crystallized TRL3) in Hank’s Balanced Salt Solution (HBSS) without Ca+2 or Mg+2. After 1 hour the ganglia were triturated gently with a fire polished glass pipette. The resultant solution was centrifuged for 5 minutes and rinsed with growth media (L-15 media supplemented with vitamins, cofactors, penicillin-streptomycin, 5% rat serum and NGF (25–50 ng/ml). The cells were then plated on laminin-coated coverslips and incubated at 37° (5% CO2) until needed. The neurons were transferred to 30°C for 24 h prior to recording.
Whole-cell patch-clamp recordings were made with an Axiopatch 200B integrated patch clamp. Patch electrodes had a resistance of between 2–5 MΩ. Data were recorded with Axon Clampex 9.0 and visualized on Axon Clampfit 9.0. Axon Patch electrodes were filled with 65 mM KF, 55 mM KAc, 5 mM NaCl, 0.2 mM CaCl2, 1 mM MgCl2, 10 mM EGTA, and 10 mM HEPES, pH 7.4. The cells were continuously perfused with extracellular recording media (140 mM NaCl, 5.4 mM KCl, 0.33 mM NaH2PO4 0.44 mM KH2PO4, 2.8 mM CaCl2, 0.18 mM MgCl2, 10 mM HEPES, 5.6 mM glucose, 2 mM glutamine, 5 µg/ml phenol red and 1–2 µM atropine, pH 7.4). The agonist ACh was dissolved in extracellular recording media and delivered through an Eppendorf Femtojet pressure perfusion system at 50 kPa for 1 sec through a glass capillary drawn to a diameter of 5–10 µm. Alternatively, in some experiments a 2 sec pulse of ACh was applied to cells through a gravity-fed multibarrel micropipette system.
For denervation experiments, we modified procedures described by McFarlane and Cooper (1992) for neonatal rats. Briefly, mice older than 2 months of age were anesthetized with Ketamine-Xylazine by intraperitoneal injection. The preganglionic cervical sympathetic trunk on one side was exposed and transected approximately halfway between the SCG and the first rib. After the surgery, the mice were returned to their cages for different times up to 2 wk. We observed that all denervated animals exhibited pronounced ptosis of the eye on the operated side. In most animals, we confirmed that the denervation was complete by electrophysiologically recording from the postganglionic nerve with extracellular electrodes while stimulating the preganglionic stump. All denervation procedures were carried out at McGill University in accordance with the guidelines established by the Canadian Council on Animal Care.
The statistical test applied to the analysis of genotype distribution was the one-way Chi-square test. The statistical test applied to analysis of electrophysiological measurements was the unpaired t-test using standard tables and online software provided through The Free Statistics Calculator Website (http://www.danielsioper.com/statcalc).
Mice with αBTX-sensitive α3-containing nAChRs (Chrna3tm1.0Hwrt) were generated by introducing mutations to the α3 gene (Chrna3) located on chromosome 9. Following electroporation of the targeting DNA into mouse ES cells, a total of 192 G418-resistant candidates were collected and screened. Six recombinant ES cell clones were successfully identified using Southern blot analysis (Fig. 1B). The 3' probe detected a single band of ~17.4 kb in the wild-type sample, and a ~17.4 kb band along with a band of ~10.5 kb in the correctly targeted clones. With the 5' probe, a ~17.5 kb band was detected from the wild-type allele and an ~8.9 kb band from the targeted allele.
Genomic DNAs from the correctly targeted ES cell clones were further analyzed by the KpnI restriction site-based PCR detection assay and allele-specific PCR to confirm the presence of the α3 mutation (data not shown). Although the neomycin resistance cassette was detected by Southern blot and PCR in all six of the homologous recombinants, the α3 mutations could only be identified in three of six ES cell clones (A6, D4, and D12). These clones were used to generate chimeras and F1and F2 Chrna3tm1.0Hwrt/+ animals.
Next, we removed the neomycin cassette by mating Chrna3tm1.0Hwrt/+ males and 129S1-Hprt1tm1(cre)Mnn/J females. The 129S1-Hprt1tm1(cre)Mnn/J strain carries an X-linked Cre-recombinase expression cassette that mediates the excision of DNA segments flanked by loxP sites (Tang et al., 2002). The Neo-deleted offspring (Chrna3tm1.1Hwrt/+) were intercrossed to generate homozygous Chrna3tm1.1Hwrt/tm1.1Hwrt pups, which were identified by Southern blot analysis (Fig. 1C). Initial breeding attempts indicated that ~87% of Chrna3tm1.1Hwrt/tm1.1Hwrt animals died within two weeks of birth (data not shown) and exhibited a phenotype resembling that previously reported for α3-null (−/−) mutants (Xu et al., 1999), with the exception that the ptosis in α3 homozygotes appeared less severe.
To ensure that α3 mRNA was being produced in knock-in animals, we performed RT-PCR using RNA isolated from the SCG of Chrna3tm1.1Hwrt/+, Chrna3tm1.1Hwrt/tm1.1Hwrt, and wild-type mice. With the wild-type allele-specific primers, an expected ~480 bp product was detected from Chrna3tm1.1Hwrt/+ (α3/+) and wild-type cDNAs, but not from Chrna3tm1.1Hwrt/tm1.1Hwrt (α3/α3) cDNA. The α3 allele-specific primers amplified a 480 bp product from both mutant samples, but not from the wild-type template (Fig. 1D). These results indicated that the mutated allele was actively transcribed and mRNA produced in the SCG of Chrna3tm1.1Hwrt/+ and Chrna3tm1.1Hwrt/tm1.1Hwrt mice.
Our α3 knock-in mice were maintained initially on a mixed S129S7/C57BL/6 background. After removing the neomycin cassette, we backcrossed them >10 times into C57BL/6 strain to generate congenic Chrna3tm1.1Hwrt/+ animals. The transmissibility of the mutant allele was then characterized in these congenic mice (Table I). Matings of Chrna3tm1.1Hwrt/+ animals with wild-type mice produced the expected Mendelian distribution of genotypes (1:1). As Chrna3tm1.1Hwrt/+ animals were fertile and had normal appearance and behavior for at least the first 12 months of life, no obvious dominant effects resulted from the presence of the mutant allele. Intercrosses of Chrna3tm1.1Hwrt/+ males and females, however, failed to produce the expected proportion of Chrna3tm1.1Hwrt/tm1.1Hwrt pups, indicating that about 50% of homozygous mutants died before genotyping was conducted (at 1–2 weeks of age). The surviving Chrna3tm1.1Hwrt/tm1.1Hwrt animals in the C57BL/6 background typically died within three months of birth. Pathologic examination performed on a deceased Chrna3tm1.1Hwrt/tm1.1Hwrt animal at the age of three months identified mild hydronephrosis of the kidney. However, other major organs were determined to be within normal limits, and the cause of premature death remains unknown. All together, our mating analyses indicate that the five amino acid substitution in the α3 subunit of nAChR constitutes a recessive, hypomorphic mutation that reduces lifespan in the C57BL/6 inbred background.
To increase viability and lifespan of mutant mice, it is sometimes advantageous to transfer the generated mutation on an outbred background. Outbred mice maintain maximum heterozygosity and can offer numerous advantages over inbred strains, such as longer life spans, disease resistance, and more robust breeding (Silver, 1995). To test whether outbred genetic backgrounds would influence the observed phenotypes, we outcrossed Chrna3tm1.1Hwrt/+ mice with two outbred strains: ICR (Taconic, Hudson, NY) and CD-1 (Charles River Canada, St. Constainte, Qc). Intercrosses between ICR outbred Chrna3tm1.1Hwrt/+ animals are summarized in Table II. Although in the sample matings tested the proportion of homozygous Chrna3tm1.1Hwrt/tm1.1Hwrt mice was somewhat lower than expected, the differences were not statistically significant (P>0.05). Moreover, lifespan was increased significantly to an average of five months following outcrosses with the ICR strain. Similar results were obtained with mice outcrossed to the CD-1 strain. For these reasons, outbred-derived α3 homozygotes were chosen for our initial characterizations of synaptic function in intact SCG.
In murine sympathetic ganglia, α3-containing nAChRs are essential for fast synaptic transmission (Rassadi et al., 2005; Krishnaswamy and Cooper, 2009). Therefore, to determine whether α3 subunits were surface-expressed at synapses, we stained living SCG with rhodamine-conjugated αBTX (rhoBTX). The ganglia were then post-fixed and stained with anti-neurofilament (NFM-H) to locate the preganglionic axons, and with anti-vesicle ACh transporter (VAChT) to locate the presynaptic terminals. Sympathetic neurons in SCG obtained from mature (2–3 month) Chrna3tm1.1Hwrt/tm1.1Hwrt mice yielded bright αBTX-labeled clusters that colocalized with presynaptic varicosities on the soma and dendrites of sympathetic neurons (Fig. 2A,C,D,E). Moreover, these αBTX-labeled clusters colocalized with PSD93 (Fig. 2B,D,E), indicating that α3-containing receptors are clustered at synapses. At many of the PSD-93 and VAChT sites, the signal from the clusters of α3-containing receptors was too low to be resolved from background, leading to less than complete correlation (Fig. 2E). In all cases, fluorescent labeling by αBTX was completely abolished by preincubation with 100 nM unlabeled αBTX (Fig. 2C).
Rodent sympathetic neurons express α7 mRNA (Mandelzys et al., 1995; De Koninck and Cooper, 1995) and have 125I-αBTX-binding sites (Smolen 1983; De Koninck and Cooper, 1995; Del Signore et al., 2004). This raised the possibility that the αBTX labeling in α3-containing ganglia from Chrna3tm1.1Hwrt/tm1.1Hwrt mice could be due to α7 nAChRs rather than α3-containing receptors. To rule out this possibility, we crossed α3 mice with α7 null mice and generated Chrna3tm1.1Hwrt/tm1.1HwrtChrna7tm1Bay/tm1Bay double mutants (i.e., α3-homozygous and α7(−/−)). Punctate αBTX labeling on sympathetic neurons from these α3-homozygous α7(−/−) ganglia was similar in appearance and density to that on neurons from α3-homozygous α7(+/+) ganglia (Fig. 2 F), indicating that α7 nAChRs are not contributing to the observed fluorescent signal (Fig. 2C). Moreover, preincubation with 50 nM MLA, a specific α7 antagonist, had no effect on rhoBTX labeling in α3-homozygous ganglia from mice in the α7(+/+) background (Fig. 2C). Mice with wild-type α3 and α7 failed to give rise to any detectable rhoBTX signal (Fig. 2F).
The results above indicate that α3-containing postsynaptic receptors are located at synapses in the SCG of Chrna3tm1.1Hwrt/tm1.1Hwrt mice. Next, we asked whether αBTX blocked synaptic activity in the SCG of these mice. To measure synaptic transmission, we stimulated the preganglionic nerve and recorded the nerve-evoked compound action potentials (CAP) extracellularly from the postganglionic trunk. In wild-type ganglia, stimulating the preganglionic nerve produced a 1.5–3 mV CAP (n=12). In line with previous studies, the CAP in wild-type ganglia was unaffected by αBTX (100 nM for 20–30 min; n=6) (Fig. 3A). The nerve-evoked CAPs recorded from Chrna3tm1.1Hwrt/tm1.1Hwrt mice were smaller (0.2–0.3 mV; n=15) than those from wild-type ganglia. However, unlike CAPs from wild-type ganglia, the CAPs from α3-homozygous ganglia were completely blocked by αBTX (Fig. 3A). Similar levels of block were observed in all preparations tested (n=12).
To further investigate synaptic transmission of intact SCG, we recorded excitatory postsynaptic potentials (EPSPs) intracellularly from these sympathetic neurons. With SCG from wild-type mice, the nerve-evoked EPSPs on sympathetic neurons were large (20–30 mV; n=20) and suprathreshold (Fig. 3B). As with the CAPs, none of the EPSPs recorded from wild-type ganglia were blocked by αBTX (100 nM for 20–30 min; n=6). In contrast, the majority (>80%) of nerve-evoked EPSPs on sympathetic neurons in Chrna3tm1.1Hwrt/tm1.1Hwrt SCG were subthreshold (n=25) with peak amplitudes of only 3.6 ±0.8 mV (n=25). Most relevantly, however, all EPSPs (10/10) on α3-homozygous neurons were blocked by αBTX (Fig. 3B).
To ensure that the nerve-evoked EPSPs on α3-homozygous neurons were not due to resident α7-containing nAChRs (Cuevas et al., 2000; Severance et al., 2004), we examined nerve-evoked EPSPs in ganglia from Chrna3tm1.1Hwrt/tm1.1HwrtChrna7tm1Bay/tm1Bay (α3-homozygous and α7(−/−)) mice. All nerve-evoked EPSPs on sympathetic neurons in the SCG of these homozygous double mutants were similarly blocked by αBTX (n=12), indicating that the αBTX-sensitive EPSPs were mediated by α3-containing receptors (Fig. 3B).
Although we observed reproducible labeling of α3 subunits at synapses, we wondered why the nerve-evoked EPSPs on α3-homozygous sympathetic neurons were smaller than those on wild-type neurons. Conceivably, replacing 5 amino acids in the α3 subunit with corresponding residues from α1 of Torpedo californica may have interfered with the function or surface expression of α3-containing receptors. To examine the possibility that the α3 mutation significantly altered the EC50 for ACh, we carried out a dose response study on cultured α3-homozygous SCG neurons (Fig. 4). We obtained an EC50 of 48±2 µM using cultured SCG neurons isolated from Chrna3tm1.1Hwrt/tm1.1Hwrt mice, and SCG neurons from wild-type mice yielded an EC50 of 64±3 µM, which is in line with published values (Kristufek et al., 1999). Thus, the smaller evoked responses observed in intact SCG are not due to a decrease in affinity or efficacy of the α3-containing nAChRs. In order to define the association and dissociation characteristics of the observed αBTX block in SCG, we carried out a study with cultured dissociated SCG neurons to follow the time course of onset of block, and the time course of recovery of function following a washout of the αBTX (Fig. 5). Full block was achieved within 3–5 minutes upon application of 100 nM αBTX, and full recovery was obtained 15 minutes after the αBTX was removed. The dissociation of αBTX is relatively rapid, certainly with respect to αBTX binding to the neuromuscular junction, but complex formation is empirically stable enough to allow fluorescence labeling under most conditions using live cells (e.g., Fig. 2).
To examine the possibility that the α3 mutation adversely affects surface expression, we carried out additional electrophysiological studies measuring ACh-evoked currents from cultured SCG neurons isolated from neonatal Chrna3tm1.1Hwrt/tm1.1Hwrt mice that were fully congenic (N10) on the C57BL/6 inbred background. No ACh-evoked currents could be detected from α3-homozygous neurons under standard conditions of culture at 37°C (Fig. 6B). However, ACh-evoked currents were detectable following a shift in culture incubation temperature from 37°C to 30°C for 24 h (Fig. 6B), consistent with studies showing that reduced temperature increases the surface expression of some nAChR subtypes (Ross et al., 1991; Cooper et al., 1999; Wanamaker & Green, 2007). The ACh-evoked current densities recorded from cultured Chrna3tm1.1Hwrt/tm1.1Hwrt neurons at 30°C were approximately 6% of those from control neurons and were blocked by αBTX (100 nM) (Fig. 6B). Our analysis of SCG neurons from double homozygous mutant mice (α3-homozygous α7(−/−)) yielded similar results (Fig. 6B), further indicating that the observed ACh-evoked currents were not mediated by α7-containing nAChRs. In comparison, a 24 hr incubation at 30°C of dissociated SCG neurons obtained from wild-type mice did not result in any significant change in current densities nor in the expected insensitivity to αBTX block. These results suggest that the amino acid substitutions in α3 likely have a negative impact on some critical aspect of receptor subunit assembly and/or surface expression, which can be relieved in part by a decrease in incubation temperature.
Next, considering the unexpected differences in surface expression of α3-containing nAChRs in C57BL/6 (Fig. 6B) versus outbred ICR/CD-1 backgrounds (Fig. 2), we sought to more fully explore the influence of genetic background on surface expression of functional α3-containing nAChRs. To do this, we crossed fully congenic C57BL/6 Chrna3tm1.1Hwrt/+ mice with wild-type α3 mice from the following strains: C3H, DBA, BALB/cJ, and ICR. Heterozygous F1 offspring from each of the above-mentioned strains were obtained, from which F2 Chrna3tm1.1Hwrt/tm1.1Hwrt mice were generated through intercrosses. We then prepared primary cultures of SCG neurons from neonatal Chrna3tm1.1Hwrt/tm1.1Hwrt mice in these various backgrounds and measured ACh responses following culture incubation at 30°C. The ACh-evoked current densities of SCG neurons from Chrna3tm1.1Hwrt/tm1.1Hwrt mice derived from the outbred ICR background were dramatically increased compared to neurons from congenic C57BL/6 mice (Fig. 6C). Of several inbred strains tested in this fashion, the BALB/cJ background appears to support a signifcant increase in ACh response amplitude. In addition, we observe that lifespan is greatly extended in Chrna3tm1.1Hwrt/tm1.1Hwrt mice on the BALB/cJ background. Whereas <2 % of Chrna3tm1.1Hwrt/tm1.1Hwrt mice on the C57BL/6 or S129 genetic background survive beyond 70 days of age, more than 60% of the homozygous mice on the BALB/cJ background survive to that point. In addition, none of the Chrna3tm1.1Hwrt/tm1.1Hwrt mice on the BALB/cJ background are born runted, whereas all such homozygotes are born runted on the C57BL/6 background. In all outward appearances, the Chrna3tm1.1Hwrt/tm1.1Hwrt mice on the BALB/cJ background are indistinguishable from mice with wild-type α3. Furthermore, in preliminary studies, cultured SCG neurons from α3 homozygotes produced from early backcrosses onto the C57BL/6-BALB/cJ background and incubated from 24 to 48 hours at 30°C have produced ACh responses approaching near wild-type levels (data not shown). Moreover, in contrast to the situation with C57BL/6, we have observed ACh-evoked whole-cell currents in SCG neurons from older BALB/cJ- and ICR-derived α3-homozygous mice without the necessity of a 30°C incubation period.
The colocalization of α3-containing receptor clusters with presynaptic terminals at ganglionic synapses, as demonstrated in Fig. 2, prompted us to ask whether the presynaptic nerve has a direct role in clustering receptors, similar to the role of presynaptic terminals at the neuromuscular junction (Sanes and Lichtman, 1999; Gingras et al., 2002). Therefore, we cut the preganglionic nerve to the SCG in Chrna3tm1.1Hwrt/tm1.1Hwrt mice and examined denervated neurons at different times up to 2 weeks. Within 2 days of denervation, we observed a significant increase in the size of the αBTX-labeled clusters as well as an increase in the density of clusters compared to control (Fig. 7A,B,E). These increases likely reflect a response similar to denervation supersensitivity, a process observed in many other synaptic systems (Sharpless, 1975; Cangiano, 1985). Interestingly, these changes in clustering were transient, and, by one week after denervation, the receptor clusters were not significantly different from those on neurons in the contralateral control ganglia (Fig. 7C,E). Over the following week, the clusters continued to decrease in size and density (Fig. 7D,E). We did not determine whether this decrease in cluster-labeling, as summarized in Fig. 7E, reflects a dispersal of receptors similar to what is observed on autonomic neurons in PSD93-null mice after denervation (Parker et al., 2004), or whether it reflects a decrease in the total number of receptors at the surface. This analysis confirmed the postsynaptic localization of the rhoBTX-labeled clusters and revealed new insights into the architecture and dynamics of nAChR clustering following disruption of normal preganglionic input.
Efforts to understand fully the physiological roles of specific heteromeric nAChR subtypes in the central nervous system are limited by lack of reliable subtype-selective ligands. Antibodies against neuronal nAChR subunits are often problematic for isolating receptors from intact tissue or neurons, as most endogenous antigenic determinants are localized to the cytoplasmic portion of the subunits. Moreover, studies with knock-out mice have raised serious questions concerning the specificity of several types of neuronal nAChR antibodies (Herber et al., 2004; Moser et al., 2007). Although some α-conotoxins have demonstrated remarkable selectivity among particular nAChR subunit interfaces, additional strategies are still needed to address the critical questions concerning subunit identification in the physiological setting (Millard et al., 2009).
The α1 subunit from the muscle-type nAChR contains the major determinants for binding the classic nicotinic antagonist, αBTX, while most neuronal nAChRs, apart from the α7 subgroup, are completely insensitive to αBTX. Given the considerable homologies shared by the genes encoding the muscle and neuronal nAChR subunits, we hypothesized that functional sensitivity to αBTX could be conferred with minimally mutated neuronal nAChR subunits. This approach has been validated in principle with the neuronal rat α3 subunit as studied in Xenopus oocytes (Levandoski et al., 1999).For decades, αBTX has been used extensively for the study of nicotinic receptors. Fluorescent-labeling with αBTX has many practical advantages over staining with antibodies: αBTX is a relatively small, cell-impermeant protein with numerous commercially available conjugates. Antibodies, on the other hand, are relatively large, divalent molecules which may cap or cluster surface antigens and compromise surface trafficking (Sekine-Aizawa and Huganir, 2004). Efforts in our lab have been focused on extending the experimental advantages of αBTX to the study of αBTX-insensitive neuronal nAChRs. Recently, we introduced an 11 amino acid α1-derived, αBTX-binding sequence into the loop C region of the β4 neuronal nAChR subunit. Co-injection of cRNAs from wild-type α3 together with the chimeric β4/α1 resulted in functional channels that were allosterically inhibited by αBTX (Sanders & Hawrot, 2004). Similar results have been achieved using an analogous chimeric β2 subunit (Sanders & Hawrot, manuscript in preparation) as well as with an analogous chimeric α5 subunit (Chew, Sanders & Hawrot, manuscript in preparation). In such a context, we refer to these ectopic αBTX-binding sequences as “pharmatopes,” as the pharmacological block is localized to subunit-interfaces not normally involved in agonist binding (Sanders & Hawrot, 2004).
High-affinity αBTX-binding sequences have been used in a variety of trafficking studies involving non-nicotinic receptor proteins. Sekine-Aizawa and Huganir (2004) inserted a 13-mer peptide sequence known as HAP-1 (Kasher et al., 2001, Fuchs et al., 2003) at the N-terminus of the GluR2, AMPA receptor subunit. This strategy facilitated the monitoring of surface expression and trafficking dynamics of the αBTX-tagged AMPA receptor subunits expressed in HEK cells using radioactive, fluorescent, and biotinylated αBTX-conjugates. Bogdanov et al. (2006) utilized a comparable approach for investigating the membrane-trafficking of GABAA receptors in hippocampal neurons. Guo et al. (2006) engineered a 13 amino acid αBTX-binding sequence into the N-terminus of a metabotropic glutamate receptor, mGluR8a, and visualized surface-expressed receptors in rat sympathetic neurons using fluorescein-conjugated αBTX. More recently, Wilkins et al. (2008) monitored the cell-surface mobility of αBTX-tagged GABAB subunits (R1a) in real time in transfected hippocampal neurons.
Our present study expands on the simple tagging approach by introducing pharmacological sensitivity to is αBTX in vivo. We provide the first example of a gain-of-αBTX-sensitivity knock-in mouse in which the primary determinants for αBTX-binding are incorporated into the homologous region of a native nAChR subunit that is normally insensitive to αBTX. Our studies show that αBTX-sensitive α3 subunits are incorporated into functional nicotinic receptors in the SCG neurons of homozygous knock-in mice when the appropriate genetic background is used (see discussion below). Live surface-staining with rhodamine-conjugated αBTX revealed high-density receptor clusters in ganglia from α3-homozygous animals (Fig. 2). All αBTX-labeled clusters colocalized with VAChT (Fig, 2D), a marker for presynaptic varicosities and with PSD-93 (Fig. 2D), a marker for the postsynaptic membrane shown previously to be associated with nAChR clusters in the mouse SCG (Parker et al., 2004).
Our novel knock-in mouse allowed us to take advantage of the αBTX binding ability of the α3-containing nAChRs to carry out a quantitative analysis of nAChR clusters in the postsynaptic membrane (Fig. 2D,E,F). In addition, we were able to determine the dynamics of nAChR cluster distribution and intensity following preganglionic denervation at a resolution and level of analysis not previously possible in the SCG (Fig. 7). Our denervation experiments (Fig. 7) demonstrate that presynaptic terminals play a role in clustering α3-containing receptors to synapses in SCG neurons, similar to their role at the neuromuscular junction (Sanes and Lichtman, 1999; Gingras et al., 2002). We found that cutting the presynaptic nerve leads to an initial increase in αBTX-labeled clusters within 2 days, followed by cluster dispersal over the next 1–2 weeks. Previous studies on rat SCG indicated that following preganglionic nerve crush there was no significant effect on the number of nAChRs in intact SCG as measured by total radioactive epibatidine binding and by radioactive αBTX binding (Del Signore et al., 2004). This rat study measured binding on intact SCG and would not have detected changes at the level of postsynaptic nAChR cluster size or density. In a denervation study using mouse submandibular ganglia (Parker et al., 2004), antibody-labeled nAChR clusters appeared to be disassembling two days after denervation and cluster dispersal continued to the end of the study, five days later. At that point, only ~23% of the postganglionic neurons contained nAChR clusters. Our findings reported here are more similar to the mouse submandibular ganglia study than the rat SCG study, but with important differences. In SCG examined two days post denervation, we see a significant increase in cluster size and in puncta density, together with a ~21% decrease in mean puncta intensity. These results would suggest that postganglionic nAChR clusters are initially spreading in size along the lines of what has been reported for muscle type nAChRs following motor nerve denervation (Sharpless, 1975; Cangiano, 1985). Beyond two days, nAChR cluster size returns to the control level, and puncta density decreases over the next two weeks to arrive at a level ~40% of the control (Fig. 7E). Denervation of autonomic neurons in amphibians and chick has been reported to lead to a similar dispersal of receptor clusters (Jacob and Berg, 1987; Sargent and Pang, 1988; McEachern et al., 1989; Levey and Jacob, 1996). Importantly, the unique features of our knock-in mouse have permitted a detailed examination of nAChR cluster architecture and dynamics not previously possible with existing approaches.
Consistent with our prediction that α3-containing nAChRs in SCG neurons would be sensitive to blockade by αBTX, we found that the amplitudes of both the CAPs and fast EPSPs in SCG neurons containing exclusively α3 subunit, although decreased to approximately 10–20% of wild-type control levels, were completely blocked by αBTX and with a rapid rate of block onset (Fig. 4). We were able to rule out the possibility that the decrease in amplitude of the nerve-evoked EPSPs might be due to a decrease in the EC50 of the α3-containing nAChRs in SCG neurons (Fig. 3) by showing that the EC50 for ACh of α3-containing nAChRs in cultured SCG neurons was very similar to the EC50 of WT controls. Based on studies with SCG neurons in culture (Fig. 5), we now believe that the number of functional, surface-expressed α3-containing nAChRs is greatly decreased, in most genetic backgrounds (see below), as compared to nAChRs in wild-type ganglia. A decreased level of surface expression could help explain our observations that not all VAChT or PSD93-labeled puncta in SCG were labeled with RhoBTX (Fig. 2E), although another possible explanation for this observation would be incomplete labeling of all alpha3(5)-containing receptors due to the relatively rapid off-rate of αBTX binding (Figure 5). It is possible that the α3 mutations may have affected the integrity of pentamer assembly or membrane trafficking, or both, possibly leading to retention in the endoplasmic reticulum (Millar & Harkness, 2008). Previous studies have suggested that the N-terminal extracellular region can regulate the association of nAChR subunits (Sumikawa, 1992). It also is possible that the loop C mutations could affect the surface stability of fully assembled receptors, leading to enhanced receptor turnover once receptors reach the surface. Additional investigations will be required to determine which of these possible mechanisms explains the observed decrease in functional responsiveness of α3-containing nAChRs in SCG.
The replacement of α3 with α3 within a pure C57BL/6 genetic background resulted in SCG neurons that did not respond to ACh application when cultured at 37°C (Fig. 6B). Following a shift to 30°C, however, we observed a significant increase in the number of functional receptors, with ACh-responses reaching a level of about 6% of that seen with wild-type neurons cultured under similar conditions. These results are in line with previous studies demonstrating that reduced temperature can increase the surface-expression of neuronal nAChRs in certain situations (Cooper et al., 1999; Nelson et al., 2003). Interestingly, we discovered that the observed effects of lower incubation temperature appear to vary greatly among different strains of mice. For example, whole-cell recordings of 30°C-treated SCG neurons prepared from outbred ICR-derived mice show considerable increases in their ACh-evoked current densities compared to neurons from mice in the congenic C57BL/6 background (Fig. 6C), but with great individual to individual variability as might be expected in an outbred line that is bred for maximal levels of heterozygosity in its strain-delimited gene pool. The results of our most recent observations with early generation backcrosses into the BALB/cJ strain show some ACh responses approaching near wild-type levels in 30°C-treated SCG neurons prepared from several older backcrossed mice (>P20). With these early generation backcrossed mice, we have occasionally observed significant ACh responses in SCG neurons maintained in culture at 37°C with no shift to lower temperature (data not shown). The prospect that the expression of functional α3-containing nAChRs varies among different inbred genetic backgrounds, as observed between the C57BL/6 and BALB/cJ strains, is exciting, as this would provide a powerful new strategy for future genetic studies aimed at dissecting the molecular mechanisms underlying the assembly and trafficking of neuronal nAChRs.
The present study has demonstrated that it is possible to transform an αBTX-insensitive neuronal nAChR subunit into an αBTX-sensitive variant through a knock-in strategy. This was accomplished by replacing five amino acids in the Loop C region of the murine α3 subunit with the corresponding residues from the α1 subunit of Torpedo californica. The work described here establishes a new mammalian model for the study of α3 nAChRs in the adult nervous system as few α3 knock-out mice survive past weaning in standard inbred lines (Xu et al., 1999). Our studies also demonstrate the great potential of the “pharmatope” approach as a general strategy for investigating neuronal nAChRs in an in vivo setting.
We would like to acknowledge Jim Boulter, UCLA, for his assistance with our targeting strategies. We thank Brigitte Pie for excellent technical assistance. We would also like to thank Geoffrey Chew and Suzanne Whitman for their contributions. P.C. and E.H. have been supported by National Institutes of Health grants #GM32629 and PHS P20 RR15578. A.K. and E.C. receive support from the CIHR and HSFC. J.K. is supported by 5P20RR015578.