SEARCH

SEARCH BY CITATION

Keywords:

  • Nicotinic acetylcholine receptor;
  • Insect;
  • Subunit composition;
  • Xenopus oocyte;
  • Functional expression;
  • Immunoprecipitation;
  • α-Bungarotoxin

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

Abstract: Although neuronal nicotinic acetylcholine receptors from insects have been reconstituted in vitro more than a decade ago, our knowledge about the subunit composition of native receptors as well as their functional properties still remains limited. Immunohistochemical evidence has suggested that two α subunits, α-like subunit (ALS) and Drosophilaα2 subunit (Dα2), are colocalized in the synaptic neuropil of the Drosophila CNS and therefore may be subunits of the same receptor complex. To gain further understanding of the composition of these nicotinic receptors, we have examined the possibility that a receptor may imbed more than one α subunit using immunoprecipitations and electrophysiological investigations. Immunoprecipitation experiments of fly head extracts revealed that ALS-specific antibodies coprecipitate Dα2, and vice versa, and thereby suggest that these two α subunits must be contained within the same receptor complex, a result that is supported by investigations of reconstituted receptors in Xenopus oocytes. Discrimination between binary (ALS/β2 or Dα2/β2) and ternary (ALS/Dα2/β2) receptor complexes was made on the basis of their dose-response curve to acetylcholine as well as their sensitivity to α-bungarotoxin or dihydro-β-erythroidine. These data demonstrate that the presence of the two α subunits within a single receptor complex confers new receptor properties that cannot be predicted from knowledge of the binary receptor’s properties.

Nicotinic acetylcholine (ACh) receptors (nAChRs) are oligomeric receptor complexes that participate in fast synaptic transmission. In the insect nervous system, nAChRs are known to play an important role in excitatory neurotransmission (Sattelle, 1980). A first functional reconstitution in lipid bilayers of an insect nAChR has been achieved with protein preparations purified from locust nervous system, using α-bungarotoxin (α-Bgt) as an affinity ligand (Hanke and Breer, 1986). However, the identity of the subunit(s) involved in the reconstituted receptor has not been determined. By molecular cloning of genes and cDNAs, it has been shown for various insect species that multiple α- and β-nAChR subunits are expressed in the CNS (Gundelfinger, 1992; Gundelfinger and Hess, 1992; Eastham et al., 1998; Hermsen et al., 1998; Schulz et al., 1998; Sgard et al., 1998; Gundelfinger and Schulz, 2000). From Drosophila melanogaster five genes encoding neuronal nAChR subunits have been cloned so far (for review, see Gundelfinger, 1992; Gundelfinger and Hess, 1992; Schulz et al., 1998). According to their amino acid sequences, they can be classified into α-type—α-like subunit (ALS; nACRα-96Aa), Drosophilaα2 subunit/second α-like subunit of Drosophila (Dα2/SAD/nACRα-96Ab), and Drosophilaα3 subunit (Dα3; nACRα-7E)—and β-type—ACh receptor protein of Drosophila (ARD; nACRβ-64B) and second β-like subunit of Drosophila (SBD; nACRβ-96A). However, the structure and the subunit composition of Drosophila nAChRs still remain unknown, and functional reconstitution of native receptors is not yet available (Bertrand et al., 1994; Lansdell et al., 1997).

Our knowledge about subunits that coassemble within the same receptor complex in the fruit fly is circumstantial. Immunoprecipitation experiments have suggested that ALS and ARD may be subunits of the same receptor complex that binds the snake toxin α-Bgt with high affinity (Schloss et al., 1991). In vivo promoter studies showed that both ARD and Dα2 genes are expressed in an overlapping if not identical set of neurons (Hess et al., 1994; Jonas et al., 1994). Immunohistochemical studies revealed a codistribution of ALS, ARD, and Dα2 subunits in the synaptic neuropil of the Drosophila CNS, suggesting that either an nAChR complex containing two α subunits exists or that receptor subtypes with different subunit compositions are colocalized (Schuster et al., 1993; Jonas et al., 1994).

Previous electrophysiological characterization of nAChRs reconstituted with either ALS or Dα2 subunits was achieved by coexpression of these subunits with the chick β2 subunit in Xenopus oocytes (Bertrand et al., 1994). However, the observed colocalization in Drosophila of the ALS and Dα2 subunits forced us to test if these α subunits are coassembled in naturally occurring receptor complexes. To gain further insight into this matter, we first designed a set of experiments using immunoprecipitation with specific antibodies directed against either ALS or Dα2 subunits. These determinations were done on native receptors obtained from fly head extracts. In a second set of experiments we have examined if the presence of both ALS and Dα2 subunits can be revealed at the functional level using heterologous expression in Xenopus oocytes.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

Antibodies and immunoprecipitation experiments

The following subunit-specific antibodies and antisera were used in this study: Rabbit polyclonal antibody (pab) CIII-1, rat pab R14, and mouse monoclonal antibody (mab) D4 were generated against a bacterially expressed fusion protein of the large cytoplasmic loop of ALS (Schloss et al., 1991; Schuster et al., 1993). Mouse mab C3/1 is a sibling clone of mab C3 produced against the large cytoplasmic loop of Dα2 (Jonas et al., 1994). Antibodies were purified on protein A-Sepharose (Pharmacia, Bonn, Germany) or protein G-Sepharose (Gamma-Bind plus; Pharmacia) before use.

Detergent extracts of Drosophila head membranes were prepared as previously described (Schloss et al., 1991). GammaBind plus Sepharose was equilibrated in buffer A [10 mM Tris-HCl (pH 8.0), 10 mM NaCl, and 1% Triton X-100]. Fifty microliters of a 1:1 GammaBind plus Sepharose/buffer A suspension was incubated for 2 h at 4°C with 10 μg of antibody. Detergent extract was preabsorbed with GammaBind plus Sepharose to eliminate nonspecific binding. Then the preabsorbed detergent extract (2.5 mg of protein) was incubated overnight with antibody-coupled GammaBind plus Sepharose in a final volume of 1 ml of buffer A. In competition experiments, an ≈1,000-fold molar excess (as compared with antibody molecules) of the ALS fusion protein, which served as antigen for the production of ALS-specific antibodies [CIII (Schloss et al., 1991)], was included in the incubation mixture. Immunoprecipitates were collected by centrifugation, and pellets were washed twice with buffer A, twice with buffer B (buffer A containing 1 M NaCl), and once with buffer C (50 mM Tris-HCl, pH 6.8). Proteins were eluted from the GammaBind plus Sepharose, subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and electroblotted as described (Langnaese et al., 1997). Immunoblots were developed using the ECL detection system (Amersham, Braunschweig, Germany).

Expression and recording in Xenopus oocytes

Functional nAChRs were reconstituted in Xenopus oocytes by nuclear injection of ALS and/or Dα2 subunits from D. melanogaster in combination with the chick β2 subunit. The β2 subunit cDNA has been cloned into the Flip expression vector (Bertrand et al., 1994), whereas ALS and Dα2 cDNAs were recloned into the PMT3 vector (Swick et al., 1992), which yielded higher expression levels for these two subunits. Taking advantage of a HincII restriction site coding for amino acids 365 and 366 of ALS (Bossy et al., 1988), a truncated version of ALS cDNA was cloned into the Flip expression vector (Bertrand et al., 1991). From this construct a C-terminally truncated ALS comprising the N-terminal extracellular domain and the first three transmembrane regions (ALSΔ) was expressed. Unless otherwise stated, the oocytes were injected with 10 nl at 0.1 ng/nl of each of cDNA constructs to be assayed for (Bertrand et al., 1991).

Electrophysiological recordings were made using a dual-electrode voltage clamp (GENECLAMP 500; Axon Instruments, Foster City, CA, U.S.A.) as previously described (Bertrand et al., 1994). In some of the experiments, oocyte injection and voltage-clamp recording were done by mounting the injection pipette or recording electrodes onto an x, y, z positioning device (Isel-Automation, Munich, Germany) controlled by a Macintosh (7600) personal computer. Oocytes of similar size were placed into individual conical wells of a microtiter plate (NUNC type 8 × 12; Lifetechnologies, Basel, Switzerland) and automatically injected and recorded. Drugs and solutions were supplied as appropriate from an automated liquid handler (type XL-222; Gilson, Paris, France).

Solution and toxin applications

Superfusion with OR2 [oocyte Ringer’s solution (see Bertrand et al., 1991)] containing 82.5 mM NaCl, 2.5 mM KCl, 2.5 mM CaCl2, 1 mM MgCl2, and 5 mM HEPES (pH 7.4, adjusted with NaOH) was used in all electrophysiological experiments. ACh stock solution (0.1 M; Fluka, Buchs, Switzerland) was kept frozen and added to OR2 just before the experiment.

Incubations with the snake venom component α-Bgt (Sigma, Buchs) were performed by adding the toxin to the perfusion medium. To prevent adsorption of the toxin to plastic surfaces, 20 μg/ml bovine serum albumin (fraction V; Sigma) was added to the solution. The presence of this concentration of albumin did not affect in a detectable manner the recordings of ACh-evoked currents from oocytes. For the dose-response inhibition with dihydro-β-erythroidine (DHβE; RBI, Buchs), responsive cells were first challenged with ACh and then in the presence of a series of progressively increasing inhibitor concentrations. DHβE was coapplied with ACh.

Curve fitting

Dose-response activation curves were fitted with the empirical Hill equation:

1 y=1/[1+(EC50/x)nH]

where y = amplitude of the evoked current normalized to unity, EC50 = half-maximal activation concentration, x = agonist concentration, and nH = Hill coefficient.

Simulations of dose-response curves were computed with the following equation:

2 y=a1[1/[1+(EC50,1/x)nH,1]] +  a2 [1/[1+(EC50,2/x)nH,2]]

where y = amplitude of the evoked current normalized to unity, a1 = fraction of receptor type 1, EC50,1 = half-maximal activation concentration for receptor type 1, nH,1 = Hill coefficient of receptor type 1, a2 = fraction of receptor type 2, EC50,2 = half-maximal activation concentration for receptor type 2, nH,2 = Hill coefficient of receptor type 2, and x = agonist concentration.

Dose-response inhibition curves were fitted with the empirical Hill equation:

3 y=1/[1+(x/IC50)nH]

where y = amplitude of the evoked current normalized to unity, IC50 = half-maximal inhibition concentration, x = antagonist concentration, and nH = Hill coefficient.

Dual-step dose-response inhibitions were computed with the following equation:

4 y=a1[1/[1+(x/IC50,1)nH,1]] + a2 [1/[1+(x/IC50,2)nH,2]]

where y = amplitude of the evoked current normalized to unity, a1 = fraction of receptor type 1, IC50,1 = half-maximal inhibition concentration for receptor type 1, nH,1 = Hill coefficient of receptor type 1, a2 = fraction of receptor type 2, IC50,2 = half-maximal inhibition concentration for receptor type 2, nH,2 = Hill coefficient of receptor type 2, and x = antagonist concentration.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

Immunoprecipitation of receptor complexes containing two different α subunits

To test whether ALS and Dα2 coassemble in native nAChRs, immunoprecipitation studies with subunit-specific antibodies were performed. The antibodies pab CIII-1, pab R14, and mab D4 were generated against bacterial fusion proteins of the main cytoplasmic loop of ALS (Schloss et al., 1991; Schuster et al., 1993). The antibody mab C3/1 was produced against the corresponding region of Dα2 (Schuster et al., 1993; Jonas et al., 1994). The sequences of the cytoplasmic loop regions of the two proteins have not been conserved at all during evolution (Gundelfinger, 1992), and therefore the respective antibodies are expected to be subunit-specific. Accordingly, anti-ALS antibodies recognize the ALS protein on western blots but do not recognize recombinant fusion proteins of Dα2 (data not shown) or Dα2 itself (Fig. 1A, lane 3; B, lane 1; and E, lane 2), and vice versa, the anti-Dα2 antibody mab C3/1 recognizes Dα2 but not ALS fusion protein (data not shown) or ALS itself (Fig. 1C, lane 2, and D, lane 2). These data indicate that on immunoblots the antibodies favorably detect their cognate antigens.

image

Figure 1. Coimmunoprecipitation of ALS and Dα2 Drosophila nAChR subunits. Immunoprecipitations were performed with anti-ALS pab CIII-1 (A-C) and anti-Dα2 mab C3/1 (D and E), and immunodetections on western blots were done with anti-ALS pab R14 (A, B, and E) or mab C3/1 (C and D). A: ALS antibodies immunoprecipitate ALS protein (asterisk). Detergent extracts of head membranes (50 μg of total protein; lane 1), supernatant (lane 2), and pellet (lane 3) after precipitation, and material absorbed by GammaBind plus Sepharose in the absence of pab CIII-1 (lane 4) are shown. B: Immunoprecipitation of ALS is competed by recombinant ALS fragment. Immunopellet in the absence (lane 1) and presence (lane 2) of excess of recombinant ALS cytoplasmic fragment CIII (arrow) is shown. C: The pab CIII-1 coimmunoprecipitates Dα2 protein (triangle). Supernatant (lane 1) and pellet (lane 2) after immunoprecipitation and the immunopellet in the presence of excess ALS-CIII fragment (lane 3) and material absorbed by GammaBind plus Sepharose in the absence of pab CIII-1 (lane 4) are shown. D: Dα2 antibody precipitates Dα2 protein (triangle). Detergent extracts of head membranes (50 μg of total protein; lane 1), pellet after precipitation (lane 2), immunopellet probed only with secondary anti-mouse antibody, which detects immunoglobulin heavy and light chains (lane 3), and material absorbed by GammaBind plus Sepharose in the absence of mab C3/1 (lane 4) are shown. E: Dα2 antibody coimmunoprecipitates ALS protein (asterisk). Supernatant (lane 1) and pellet (lane 2) after precipitation and material absorbed by GammaBind plus Sepharose in the absence of mab C3/1 (lane 3) are shown. Note that the weak 80K band visible in lanes 1 and 3 probably does not represent ALS immunoreactivity. It is visible also in parts of the gel where no ALS-containing fractions were present (data not shown).

Download figure to PowerPoint

As observed previously (Schuster et al., 1993; Jonas et al., 1994), none of the antibodies used in this study recognized the respective antigen in untreated head membrane detergent extracts (Fig. 1A and D, lanes 1). We assume that this is due to the low amount of antigen present in crude extracts.

First, immunoprecipitation experiments with the rabbit anti-ALS pab CIII-1 were performed (Fig. 1A-C). Immunopellets were analyzed on western blots. Rat anti-ALS pab R14 recognized an antigen of 80 kDa (Fig. 1A, lane 3). A protein of the same size was detected by the mouse anti-ALS mab D4 (data not shown). The 80-kDa band is not detected in the supernatant (Fig. 1A, lane 2) or if the immunoprecipitation procedure is performed without pab CIII-1 (Fig. 1A, lane 4). If immunoprecipitation is performed in the presence of a bacterial fusion protein that includes the cytoplasmic domain of ALS and served as antigen for production of ALS-specific antibodies (Schloss et al., 1991; Schuster et al., 1993), precipitation of the 80-kDa protein is prevented (Fig. 1B). Taken together, these data show that the 80-kDa polypeptide is ALS. To analyze whether Dα2 is coprecipitated with ALS by pab CIII-1, a comparable set of western blots was probed with mab C3/1, which is specific for Dα2. As shown in Fig. 1C, mab C3/1 detects a protein of ≈65 kDa in the CIII-1 immunopellet (lane 2). Precipitation of Dα2 can also be precluded by addition of recombinant ALS fusion protein, suggesting that Dα2 is precipitated because of its physical interaction with ALS (lane 3).

Immunoprecipitation with anti-Dα2 mab C3/1 specifically pulls down the 65-kDa Dα2 protein (Fig. 1D, lane 3). As in this case the same antibody had to be used for precipitation and immunodetection, the heavy and light immunoglobulin chains are detected on the blot (Fig. 1D, lane 3). As shown in Fig. 1E, mab C3/1 coprecipitates the 80-kDa ALS band as detected by rat pab R14. This set of data confirms that ALS and Dα2 coassemble in at least a fraction of nAChR complexes of the fly brain.

Coexpression of ALS and Dα2 subunits in Xenopus oocytes

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

Up to now, electrophysiological studies of insect receptors expressed in a heterologous system have been hampered by the lack of a functional insect β subunit. Despite this limitation, physiological and pharmacological properties of receptors reconstituted with either ALS or Dα2 subunits can readily be obtained when one of these α subunits is expressed with the chick β2 subunit (Bertrand et al., 1994). Biochemical results presented above indicate, however, that insect receptors may contain more than one α subunit and therefore that physiological properties of reconstituted receptors mimicking those of native receptors will be obtained only if the two α subunits are expressed simultaneously. Before examining the electrophysiological data obtained with multiple α subunit injection experiments, it is of value to explore the theoretical outcomes of the binary or ternary receptor complexes.

Modeling of ACh dose-response curves. As depicted in Fig. 2A, expression of two α subunits with a single β subunit may result in at least three possible receptor combinations.

image

Figure 2. Theoretical considerations on coexpression of ALS, Dα2, and β2 subunits. A: Possible receptor subtypes that can assemble with two α and one β subunit. Homomeric receptors were never observed (Bertrand et al., 1994). B: Theoretical agonist dose—response curves resulting from different ratios of ALS/β2 and Dα2/β2 complexes. Continuous thick lines correspond to the dose—response profiles of ALS/β2 and Dα2/β2 receptors, respectively (Bertrand et al., 1994). Dashed lines were simulated assuming three different ALS/β2 and Dα2/β2 receptor ratios. EC50 values given here correspond to the concentration for half-maximal ACh activation of the ALS/β2 and Dα2/β2 receptors. Dashed lines were computed using Eq. 2 (see MATERIALS AND METHODS) using EC50 values, Hill coefficients, and fraction of receptors as indicated.

Download figure to PowerPoint

Considering first that only two types of receptors can be made, e.g., ALS/β2 or Dα2/β2 (binary), this should already be resolved at the level of a given agonist dose-response curve. Accordingly, the overall dose-response curve must correspond to the prediction made by theoretical computation of two independent Hill equations (Eq. 1). As shown in Fig. 2B, the 2 log unit difference observed between ALS/β2 (0.23 μM) and Dα2/β2 (30 μM) ACh sensitivities (Bertrand et al., 1994) results in a clear-cut distinction between the two corresponding dose-response relationships. Equation 2 allows the computation of the possible dose-response curve for an oocyte expressing more than one binary receptor subtype. Thus, if ALS/β2 and Dα2/β2 are equally expressed within a given cell, the overall profile of the dose-response curve must correspond to the dashed line (b) represented in the middle of Fig. 2B. Alternatively, if ALS/β2 and Dα2/β2 are not equally expressed, the dose-response curve will progressively shift toward the preferentially expressed binary receptor (dashed line a or c). As illustrated in Fig. 2, these dose-response curves show an intermediate plateau phase. If, in contrast, the two α subunits coassemble to form a new ternary receptor type (ALS/Dα2/β2, see below), the resulting dose-response curve may display no plateau phase, and its EC50 may not be predicted from data obtained from each of the binary receptor combinations in isolation. Finally, the use of specific pharmacological agents, such as α-Bgt or DHβE, should allow discrimination between independent (binary) or combined (ternary) assembly.

Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

To distinguish between binary and ternary receptor assembly, ACh dose-response relationships of oocytes expressing ALS/β2, Dα2/β2, or the mixture ALS/Dα2/β2 have been determined over a wide range of concentrations.

Figure 3A illustrates typical ACh-evoked currents recorded in such a set of oocytes. Measurements of ACh-evoked currents in several oocytes (n = 10) injected with the mixture ALS, Dα2, and β2 yielded a dose-response curve with an EC50 value of 5.44 ± 0.98 μM (Fig. 3B). Note the difference between this dose-response curve and that obtained in sibling oocytes for ALS/β2 (0.23 ± 0.12 μM, n = 10) or Dα2/β2 (30 ± 10 μM, n = 7). The continuous nature of this curve with the absence of a plateau phase and its EC50 both suggest that it cannot be attributed to a mixture of binary receptors that contain either ALS/β2 or Dα2/β2 but rather to a new type of ternary receptor. Theoretical values, computed as in Fig. 2B, with different putative ratios of ALS/β2 and Dα2/β2 receptors are represented by the dashed lines. Comparison of experimental results with theoretical predictions highlight the differences between possible combinations of several ratios of separate populations of receptors with those obtained with coexpression of the mixture. These data therefore suggest that coinjection of the two α subunits leads to the assembly of a receptor fraction displaying new and distinct features.

image

Figure 3. ACh sensitivity of ALS/β2 and Dα2/β2 and ALS/Dα2β2 receptors. A: Time course of currents evoked by three ACh concentrations of oocytes expressing either a single α subunit or the mixture ALS/Dα2. Agonist applications are indicated by the horizontal bars above traces. Cells were held at -100 mV throughout the experiment. B: ACh dose—response curves of ALS/β2, Dα2/β2, and ALS/dα2/β2 receptors. Currents recorded from seven to 10 cells were normalized with respect to their maximal value recorded at saturating ACh concentrations. Continuous lines correspond to best fits obtained with Hill equations (Eq. 1) with respective values of EC50 = 0.23 ± 0.12 μM, nH = 0.97 ± 0.16 (n = 10) for ALS/β2; EC50 = 5.44 ± 0.98 μM, nH = 0.98 ± 0.13 (n = 10) for ALS/Dα2/β2; and EC50 = 30 ± 10 μM, nH = 1.22 ± 0.2 (n = 7) for Dα2/β2. For comparison, theoretical values computed as in Fig. 2B for nine different ratios of ALS/β2 versus Dα2/β2 receptors are represented by the dashed lines. Receptor ratios (a1 and a2) ranged from 0.1 to 0.9 in increments of 0.1.

Download figure to PowerPoint

Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

Given the differences observed at either the amino acid sequence level or physiological properties of ALS and Dα2, it could be expected that receptors reconstituted with these subunits should display dissimilar pharmacological profiles. Indeed, whereas the ALS/β2 receptors were fully and almost irreversibly blocked by a 30-min preincubation with 100 nMα-Bgt, the Dα2/β2 receptors were not affected by this treatment (Bertrand et al., 1994). To assess further whether ALS and Dα2 can assemble within a single receptor complex, the sensitivity to α-Bgt of oocytes coinjected with ALS/β2, Dα2/β2, or all three subunits was determined. Application of 100 nMα-Bgt for 10 s in the superfusion solution caused a slow and sustained inward current at ALS/β2 receptors (Fig. 4A). In addition, the ACh-evoked current recorded immediately at the end of the α-Bgt pulse presented a marked reduction in amplitude. Very little recovery was observed after washing the cell for 5 min with control medium.

image

Figure 4. Effects of α-Bgt on the ALS/β2 and Dα2/β2 receptors. A: ACh evokes a large inward current in an oocyte expressing ALS/β2 receptor (trace a). Brief preapplication of 100 nMα-Bgt induces an inward current in the same oocyte and subsequent inhibition of ACh-evoked current (thick line, trace b). The response evoked by the same ACh concentration after a 5-min wash with control OR2 medium is also shown (trace c). B: Same protocols as in A applied to an oocyte expressing Dα2/β2 receptor. The holding current remains constant during the α-Bgt preapplication. The ACh-evoked current recorded immediately at the end of the α-Bgt pulse shows a marked inhibition (trace b), and full recovery (trace c) is obtained after a 1-min wash. C: Currents evoked with the same experimental paradigm in an oocyte expressing the ALS/Dα2/β2 mixture. Note the presence of a significant inward current during the α-Bgt preapplication (trace b). Incubation for 10 s with 100 nMα-Bgt inhibits ∼30% of the ACh-evoked response (trace b). Complete recovery was observed after a 1-min wash (trace c). Cells were continuously superfused during all the experiments and held in voltage clamp at -100 mV.

Download figure to PowerPoint

When the same experiment was performed on oocytes expressing Dα2/β2 receptors, no inward current could be detected in response to the α-Bgt application (Fig. 4B). Moreover, a small but consistent inhibition of the ACh-evoked current was observed immediately at the end of the α-Bgt pulse, and complete recovery was achieved within a 60-s wash. Thus, a clear distinction could be made from these pharmacological differences. The ALS/β2 receptors were first activated and then persistently inhibited by α-Bgt, whereas Dα2/β2 receptors remained almost unresponsive to this toxin. As shown in Fig. 4C, oocytes injected with ALS, Dα2, and β2 exhibited a mixed pharmacological profile, with the toxin evoking a consistent current but only a transient blockade. Similar data were obtained in every cell tested (n = 3). Thus, the observation of an α-Bgt-evoked current in oocytes expressing the ALS/Dα2/β2 mixture may reflect the presence of a fraction of ALS/β2 or a distinct property of the ternary receptor. Computation of the amplitude of this response versus the residual ACh-evoked current observed after a 1-min wash supports, however, the latter hypothesis.

Determination of the ACh dose-response curve over a broad range of agonist concentrations before and after a 30-min incubation in 100 nMα-Bgt followed by washing showed an almost complete (85%) recovery of the overall evoked current (Fig. 5A). Because oocytes were removed from the recording chamber for the α-Bgt incubation, a perfect match between pre- and postincubation cannot be expected. If present, ALS/β2 receptors should have been blocked by this α-Bgt exposure, and a modification of the dose-response curve with a displacement of the EC50 toward that of Dα2/β2 receptors would be expected. However, no modification of the apparent agonist sensitivity has been observed (Fig. 5A).

image

Figure 5. α-Bgt does not modify the ACh sensitivity of ALS/Dα2/β2 receptor. A: ACh dose—response curves were determined in three oocytes coinjected with ALS, Dα2, and β2 cDNA expression vectors before and after incubation with 100 nMα-Bgt for 30 min. Currents were normalized with respect to the maximal value recorded at saturating ACh concentration in the control. Continuous lines are the best fits obtained with the empirical Hill equation with identical EC50 and Hill coefficients of 5.5 μM and 1, respectively (Eq. 1). A scaling factor of 0.85 was used with the Hill equation for data recorded after α-Bgt treatment. B: Typical currents evoked by five ACh concentrations on an oocyte expressing the ALS/Dα2/β2 mixture are superimposed. Data obtained from the same cell before and after incubation with 100 nMα-Bgt (30 min) are illustrated.

Download figure to PowerPoint

The EC50 value of the receptor fraction insensitive to α-Bgt (5.5 μM) differs from that of Dα2/β2 receptors (30 μM), suggesting that a novel fraction of receptors should have assembled in oocytes injected with all three subunits. In addition, the time course of the ACh-evoked currents was not affected by the α-Bgt treatment (Fig. 5B).

Dose-response inhibition to a competitive inhibitor

In an attempt to gain a further insight in the putative ALS and Dα2 assembly, we have determined the dose-response inhibition curves of oocytes expressing either a single class of receptors or the mixture. Dose-response inhibition curves measured for DHβE revealed that this compound behaves as a competitive inhibitor on both the ALS/β2- and Dα2/β2-expressing oocytes (data not shown). A marked difference in the DHβE IC50 between ALS/β2 and Dα2/β2 was, however, observed (Fig. 6A). Both dose-response inhibition curves are well fitted with a single Hill equation (Eq. 3).

image

Figure 6. The competitive inhibitor DHβE reveals two phases in the dose-response curve. A: Dose—response inhibition curves of ALS/β2- and Dα2/β2-expressing oocytes. Responses evoked by brief ACh pulses (3 μM, 3 s) are reversibly inhibited by the presence of DHβE. Plots of the peak evoked currents as a function of the DHβE concentration are readily fitted by a single Hill equation: ALS/β2 (left panel; n = 3) and Dα2/β2 (right panel; n = 4) (see Eq. 3, Materials and Methods). B: At 0.3 μM, ACh evokes a current in oocytes expressing either ALS/β2 or ALS/Dα2/β2 mixture but not in those expressing Dα2/β2. C: DHβE dose—response inhibition of oocytes expressing the ALS/β2 mean values of two cells. Inhibition curve was measured as in A but with ACh test pulses at 0.3 μM. The line connecting the data points corresponds to the Hill equation (Eq. 3, Materials and Methods). D: DHβE dose—response inhibition of oocytes expressing the ALS/Dα2/β2 mixture (n = 4). Responses evoked by 0.3 μM ACh recorded in the presence of increasing DHβE concentrations are plotted as the logarithm of the inhibitor concentration. The line through the data points corresponds to the algebraic sum of two Hill equations (Eq. 4, Materials and Methods) with respective IC50 values of 0.0025 and 0.035 μM, Hill coefficients of 1 and 1.4, and amplitudes of 0.4 and 0.6.

Download figure to PowerPoint

As shown in Fig. 6B, a low ACh test pulse will only evoke a current in oocytes expressing ALS/β2 or the mixture (ALS/Dα2/β2) but not in those expressing Dα2/β2. Under these conditions, determination of the DHβE dose-response inhibition profile in oocytes expressing ALS/β2 still yields a single inhibition curve but with a decreased IC50 (0.05 μM ACh vs. 0.3 μM at 3 μM ACh), as predicted for a competitive inhibitor (Fig. 6C, Eq. 3). When the same experiment was done in oocytes expressing the ALS/Dα2/β2 mixture, a biphasic dose-response inhibition profile was observed (Fig. 6D). These data are readily fitted by the sum of two Hill equations (Eq. 4), with respective IC50 values of 0.0025 and 0.035 μM, Hill coefficients of 1 and 1.4, and an amplitude of 0.4 and 0.6, which suggests the presence of a low- and a high-affinity binding site.

Truncated ALS inhibits expression of Dα2/β2

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

C-terminal truncation of nAChR subunits can have dominant negative effects on receptor assembly (Verrall and Hall, 1992; Jonas et al., 1994). Therefore, to confirm that ALS preferentially coassembles with Dα2/β2, competition experiments with a truncated ALS subunit (ALSΔ) were performed. Deletion of a large segment of ALS cDNA ranging from the amino acid located after the third transmembrane segment at position 365 up to the C-terminal (Fig. 7A) is expected to alter the protein function but should still allow its assembly with the Dα2/β2 complex.

image

Figure 7. Truncated ALS inhibits Dα2/β2 expression. A: Schematic representation of the wild-type and truncated ALS proteins. The number and arrow indicate the amino acid position at which deletion starts. B: Typical current traces recorded in oocytes injected with Dα2/β2, ALSΔ/β2, and ALSΔ/Dα2/β2. Cells were held at -100 mV and challenged with short ACh pulses indicated by the horizontal bars. C: ACh-evoked currents recorded in sibling oocytes injected with different subunit combinations. Data are mean ± SD (bars) values (no. of cells tested). ALSΔ cDNA ratios are indicated under the columns.

Download figure to PowerPoint

Indeed, attempts to reconstitute functional receptor with ALSΔ and β2 cDNAs showed no detectable current in any of the oocytes tested (n = 42, 1 mM ACh; Fig. 7B). However, coinjection of ALSΔ together with the wild-type Dα2 and β2 subunits yielded, even when tested at saturation, ACh-evoked currents of lower amplitude than those evoked in sibling oocytes injected only with Dα2/β2 subunits (Fig. 7B). To evaluate further the effects of ALSΔ on the current amplitude, the following experiment was performed: Oocytes were injected either with equal amount of each cDNA (1:1:1) or with a lower ratio of ALSΔ (0.4:1:1). Measurements of the maximal evoked current in a large population of cells expressing these different cDNA ratios yielded results presented in Fig. 7C. To verify that ALSΔ was not simply quenching β2, we increased by 10-fold the β2 cDNA concentration. As this maneuver did not restore the maximal current amplitude to the control level (data not shown), this suggests that ALSΔ inhibition cannot be explained on the basis of a quenching effect.

From all the data presented above, we therefore concluded that oocytes injected with the ALS/Dα2/β2 mixture display a new type of nAChRs, the properties of which can only be explained assuming the formation of ternary receptor complexes containing both ALS and Dα2.

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

In this study, we used different approaches, i.e., immunoprecipitation of receptors from brain membrane extracts and functional expression of nAChR subunits in Xenopus oocytes, to show for one type of neuronal nAChRs from the fruit fly Drosophila that at least two different ACh-binding subunits can be contained within the oligomeric receptor complex.

In immunoprecipitation studies, ALS- and Dα2-specific antibodies coprecipitate the two α subunits from head membrane extracts, although they recognize only their own antigen subunit on immunoblots. The apparent molecular masses of ALS and Dα2 proteins had not been determined because the antibodies do not recognize them on immunoblots of membrane protein preparations. This is probably due to the fact that not enough antigen is present in the membrane preparation and only by immunoprecipitation is the antigen enriched enough to be detected. ALS migrates as an 80-kDa protein (as recognized by three independent antibodies), whereas the core protein has a calculated Mr of 61.941 (Bossy et al., 1988). One explanation for this apparent discrepancy may be extensive N-glycosylation of the protein. Previous chemical cross-linking experiments with 125I-α-Bgt revealed two adducts in Drosophila head membranes of 50 and 90 kDa. Both are immunoprecipitated by ALS antisera (Schloss et al., 1992). We interpreted these data as the 50-kDa adduct being a complex composed of 8-kDa toxin and 42-kDa ALS protein and the 90-kDa adduct being a trimeric complex of two nAChR subunits and α-Bgt. In light of the findings reported here, the 90-kDa product may be the actual dimer of ≈80-kDa ALS and 8-kDa α-Bgt, whereas the 50-kDa labeled product may be a distinct proteolytic degradation product. Dα2, which has a calculated Mr of 60,963 (Baumann et al., 1990), migrates as a 65-kDa band. It will be interesting to see whether Dα2 is identical with the 66-kDa protein recently identified from Drosophila head membranes by photoaffinity labeling to bind 125I-azidonicotinoid (Tomizawa et al., 1996). This would be consistent with the particular sensitivity of Dα2/β2 receptors to the insecticide imidacloprid (Matsuda et al., 1998). Although previous experiments have indicated that ALS and ARD may be components of the same α-Bgt binding site (Schloss et al., 1991), the latter protein has not been detected in immunoprecipitates with ALS pab. At present we do not know whether ARD is absent in ALS/Dα2-containing nAChRs or whether anti-ARD antibodies are unable to detect properly the antigen on western blots.

Attempts in our laboratories to reconstitute functional homomeric receptors following cDNA injections with either ALS or Dα2 alone never resulted in significant ACh-evoked currents. Also, up to now, Drosophilaβ subunits were not found to contribute to functional nAChRs in reconstitution experiments (Sawruk et al., 1990; Bertrand et al., 1994; Lansdell et al., 1997). Therefore, coexpression of Drosophilaα subunits with the chick β2 subunit still constitutes the best available model for studying physiological properties of insect nAChR subunits.

Coinjection of ALS/Dα2/β2 yielded a receptor that displays distinct properties from either ALS/β2 or Dα2/β2. The ACh dose-response profile of the ALS/Dα2/β2 receptor is intermediate between that of ALS/β2 and Dα2/β2 receptors. As shown by computation of Hill equations, these data cannot be explained by any ratio of the latter two populations, and therefore a third fraction of receptors must be postulated. Further determination of the pharmacological profile of ALS/β2 and Dα2/β2 receptors revealed a partial activation of the former by α-Bgt. Although surprising, the partial agonist effect of α-Bgt on the ALS/β2 and ALS/Dα2/β2 receptors can be explained assuming that this toxin stabilizes both the active (open) and resting (closed) state of the receptor with different kinetics. Transient activation of nAChRs by this toxin was already reported for the L247T mutant of α7 homomeric receptors (Bertrand et al., 1992). The partial agonistic activation of the α-Bgt observed for ALS/Dα2/β2 could therefore be attributed either to the presence of a certain percentage of ALS/β2 receptor in the oocyte membrane or to the intrinsic properties of the mixture.

A further evaluation of the ternary receptor complex properties is provided by the ACh dose-response curve measured before and after α-Bgt incubation. At most the 15% reduction in current amplitude (Fig. 5A) observed after the toxin treatment could be attributed to the ALS/β2 receptors. However, the blockade of these receptors, if present, should be noticed on the EC50 value with a shift of the curve toward that of Dα2/β2. The consistency in the EC50 before and after α-Bgt further indicates the presence of ternary receptor complexes and illustrates the difference in their properties from either ALS/β2 or Dα2/β2 receptors.

In the muscle receptor, the presence of two distinct ACh-binding sites was originally deduced from the two-step inhibition profile measured with competitive antagonists (Sine and Taylor, 1980). The use of a comparable experimental paradigm revealed that ALS/β2 and Dα2/β2 receptors display a single inhibition profile for the competitive inhibitor DHβE. Moreover, the apparent affinities of these two receptors differ by more than two orders of magnitude. Taking advantage of these differences it can be proposed that if both the ALS and Dα2 subunits are present within a single receptor complex, a dual-step inhibition profile with a plateau phase might be observed in the DHβE inhibition curve. Measurements done in oocytes expressing the ALS/Dα2/β2 mixture indeed yielded a dose-response inhibition with a plateau phase (see Fig. 6D). Because the ACh concentration used in these experiments was low enough to evoke a current only in oocytes expressing either ALS/β2 or the ALS/Dα2/β2 mixture but not in those expressing Dα2/β2, a putative contribution of this receptor can be excluded. This implies that in ALS/Dα2/β2 receptors, the interface between Dα2 and β2 forms the high-affinity site for DHβE, whereas the ALS with β2 interface constitutes the low-affinity site.

Strong evidence for the presence of multiple α subunits in a single receptor complex was also obtained in vertebrates from immunoprecipitation assays as well as labeling with antibodies. For example, it was shown that the α5 subunit could be copurified with α3-containing receptors (Vernallis et al., 1993). Similarly, α7 and α8 subunit-containing receptors can be purified from the chick optic lobe (Schoepfer et al., 1990; Gotti et al., 1994). When expressed in Xenopus oocytes, addition of mRNA coding for the α5 subunit was reported to modify significantly the properties of α4/β2 vertebrate receptors (Ramirez-Latorre et al., 1996; F. Wang et al., 1996).

It is well documented that the N-terminal and first transmembrane domains play a major role in determining assembly of the muscle nAChR (Verrall and Hall, 1992; Sumikawa and Nishizaki, 1994; Z. Z. Wang et al., 1996). Subunits in which the fourth transmembrane domain was deleted competed with the formation of functional receptors (Sumikawa and Nishizaki, 1994). In agreement with these results we found that oocytes injected with ALSΔ and β2 subunits expressed no detectable current even when exposed to 1 mM ACh. The lower amplitude of the ACh-evoked currents observed in oocytes coinjected with ALSΔ/Dα2/β2 versus those recorded in cells expressing the Dα2/β2 mixture is disclosing a competition between the two α subunits. The competitive nature of this inhibition is further illustrated by its dependence on the injected cDNA ratios. Moreover, because excess injection of β2 failed to rescue the inhibition caused by the truncated ALS, this suggests that the reduction of the ACh-evoked current cannot be attributed to a limited number of β2 proteins. Altogether, these experiments indicate that the formation of ternary complexes in oocytes may be favored over the formation of binary receptors.

In the view of the preferential assembly of two α subunits, e.g., ALS and Dα2, within the same receptor complex, it is tempting to speculate about the possible role of such assembly. The presence of more than one α subunit within a given receptor may provide both physiological and pharmacological advantages. For instance, it is well documented that the ACh-binding site resides at the interface between the α and non-α subunit (Bertrand and Changeux, 1995). The presence of two distinct α subunits may therefore provide a broader spectrum of sensitivity to agonists while increasing the protection against antagonists, as illustrated from the DHβE experiment. Consequently, the presence of multiple subunits could also provide a better resistance to changes in the environmental conditions. For example, it has been shown that although a mutation in the ε subunit of the muscle receptor impairs its function, expression of the γ subunit could restore its normal activity (Milone et al., 1998).

In conclusion, immunoprecipitation studies as well as electrophysiological and pharmacological data support the hypothesis that ALS and Dα2 preferentially coassemble within a single receptor complex. By analogy with vertebrate nAChRs, assembly within the same heteropentamer of two subunits that form the major components of the ligand-binding site can be viewed as a further alternative to increase receptor diversity. Moreover, properties of ternary receptor complexes containing two different α subunits can be revealed by electrophysiological and pharmacological experiments, but the properties of these receptors cannot be predicted from the knowledge of binary receptors containing only one type of α subunit.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References

We are indebted to Prof. M. Ballivet for providing the chick β2 construct. This work was supported by the Land Sachsen-Anhalt, the Deutsche Forschungsgemeinschaft, and the Fonds der Chemischen Industrie to E.D.G. and by grant 31-53638.98 from the Swiss National Foundation and the Office Fédéral de l’Education et des Sciences to D.B.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. Coexpression of ALS and Dα2 subunits in Xenopus oocytes
  6. Characterization of oocytes coexpressing ALS, Dα2, and β2 subunits
  7. Pharmacological signatures of ALS/β2, Dα2/β2, and mixture receptors
  8. Truncated ALS inhibits expression of Dα2/β2
  9. DISCUSSION
  10. Acknowledgements
  11. References
  • 1
    Baumann A., Jonas P., Gundelfinger E.D. (1990) Sequence of Dα2, a novel alpha-like subunit of Drosophila nicotinic acetylcholine receptors. Nucleic Acids Res. 18 3640.
  • 2
    Bertrand D. & Changeux J.P. (1995) Nicotinic receptor: an allosteric protein specialized for intercellular communication. Semin. Neurosci. 7 7590.
  • 3
    Bertrand D., Cooper E., Valera S., Rungger D., Ballivet M. (1991) Electrophysiology of neuronal nicotinic acetylcholine receptors expressed in Xenopus oocytes following nuclear injection of genes or cDNA, in Methods in Neuroscience (Conn M., ed), pp. 174193. Academic Press, San Diego.
  • 4
    Bertrand D., Devillers-Thiéry A., Revah F., Galzi J.L., Hussy N., Mulle C., Bertrand S., Ballivet M., Changeux J.P. (1992) Unconventional pharmacology of a neuronal nicotinic receptor mutated in the channel domain. Proc. Natl. Acad. Sci. USA 89 12611265.
  • 5
    Bertrand D., Ballivet M., Gomez M., Bertrand S., Phannavong B., Gundelfinger E.D. (1994) Physiological properties of neuronal nicotinic receptors reconstituted from the vertebrate β2 subunit and Drosophila alpha subunits. Eur. J. Neurosci. 6 869875.
  • 6
    Bossy B., Ballivet M., Spierer P. (1988) Conservation of neuronal nicotinic acetylcholine receptors from Drosophila to vertebrate central nervous system. EMBO J. 7 611618.
  • 7
    Eastham H.M., Lind R.J., Eastlake J.L., Clarke B.S., Towner P., Reynolds S.E., Wolstenholme A.J., Wonnacott S. (1998) Characterization of a nicotinic acetylcholine receptor from the insect Manduca sexta. Eur. J. Neurosci. 10 879889.
  • 8
    Gotti C., Hanke W., Maury K., Moretti M., Ballivet M., Clementi F., Bertrand D. (1994) Pharmacology and biophysical properties of α7 and α7-α8 α-bungarotoxin receptor subtypes immunopurified from the chick optic lobe. Eur. J. Neurosci. 6 12811291.
  • 9
    Gundelfinger E.D. (1992) How complex is the nicotinic receptor system in insects. Trends Neurosci. 15 206211.
  • 10
    Gundelfinger E.D. & Hess N. (1992) Nicotinic acetylcholine receptors of the central nervous system of Drosophila. Biochim. Biophys. Acta 1137 299308.
  • 11
    Gundelfinger E.D. & Schulz R. (2000) Insect nicotinic acetylcholine receptors: genes, structure, physiological and pharmacological properties, in Handbook of Experimental Pharmacology, Vol. 144: Neuronal Nicotinic Receptors (Clementi F., Gotti C., and Fornasari D., eds), pp. 497521. Springer-Verlag, Heidelberg.
  • 12
    Hanke W. & Breer H. (1986) Channel properties of an insect neuronal acetylcholine receptor protein reconstituted in planar lipid bilayers. Nature 321 171174.
  • 13
    Hermsen B., Stetzer E., Thees R., Heiermann R., Schrattenholz A., Ebbinghaus U., Kretschmer A., Methfessel C., Reinhardt S., Maelicke A. (1998) Neuronal nicotinic receptors in the locust Locusta migratoria. Cloning and expression. J. Biol. Chem. 273 1839418404.
  • 14
    Hess N., Merz B., Gundelfinger E.D. (1994) Acetylcholine receptors of the Drosophila brain: a 900 bp promoter fragment contains the essential information for specific expression of the ARD gene in vivo. FEBS Lett. 346 135140.
  • 15
    Jonas P., Phannavong B., Schuster R., Schroeder C., Gundelfinger E.D. (1994) Expression of the ligand-binding nicotinic acetylcholine receptor subunit Dα2 in Drosophila central nervous system. J. Neurobiol. 25 14941508.
  • 16
    Langnaese K., Beesley P.W., Gundelfinger E.D. (1997) Synaptic membrane glycoproteins gp65 and gp55 are new members of the immunoglobulin superfamily. J. Biol. Chem. 272 821827.
  • 17
    Lansdell S.J., Schmitt B., Betz H., Sattelle D.B., Millar N.S. (1997) Temperature-sensitive expression of Drosophila neuronal nicotinic acetylcholine receptors. J. Neurochem. 68 18121819.
  • 18
    Matsuda K., Buckingham S.D., Freeman J.C., Squire M.D., Baylis H.A., Sattelle D.B. (1998) Effects of the alpha subunit on imidacloprid sensitivity of recombinant nicotinic acetylcholine receptors. Br. J. Pharmacol. 123 518524.
  • 19
    Milone M., Wang H.L., Ohno K., Prince R., Fukudome T., Shen X.M., Brengman J.M., Griggs R.C., Sine S.M., Engel A.G. (1998) Mode switching kinetics produced by a naturally occurring mutation in the cytoplasmic loop of the human acetylcholine receptor epsilon subunit. Neuron 20 575588.
  • 20
    Ramirez-Latorre J., Yu C.R., Qu X., Perin F., Karlin A., Role L. (1996) Functional contributions of α5 subunit to neuronal acetylcholine receptor channels. Nature 380 347351.
  • 21
    Sattelle D.B. (1980) Acetylcholine receptors of insects. Adv. Insect Physiol. 15 215315.
  • 22
    Sawruk E., Schloss P., Betz H., Schmitt B. (1990) Heterogeneity of Drosophila nicotinic acetylcholine receptors: SAD, a novel developmentally regulated α-subunit. EMBO J. 9 26712677.
  • 23
    Schloss P., Betz H., Schröder C., Gundelfinger E.D. (1991) Neuronal nicotinic receptors in Drosophila: antibodies against an α-like and a non-α-subunit recognize the same high-affinity α-bungarotoxin binding complex. J. Neurochem. 57 15561562.
  • 24
    Schloss P., Mayser W., Gundelfinger E.D., Betz H. (1992) Cross-linking of 125I-α-bungarotoxin to Drosophila head membranes identifies a 42 kDa toxin-binding polypeptide. Neurosci. Lett. 145 6366.
  • 25
    Schoepfer R., Conroy W.G., Whiting P., Gore M., Lindstrom J. (1990) Brain α-bungarotoxin binding protein cDNAs and mAbs reveal subtypes of this branch of the ligand-gated ion channel gene superfamily. Neuron 5 3548.
  • 26
    Schulz R., Sawruk E., Mülhardt C., Bertrand S., Baumann A., Phannavong B., Betz H., Bertrand D., Gundelfinger E.D., Schmitt B. (1998) Dα3, a new functional α subunit of nicotinic acetylcholine receptors from Drosophila. J. Neurochem. 71 853862.
  • 27
    Schuster R., Phannavong B., Schroeder C., Gundelfinger E.D. (1993) Immunohistochemical localization of a ligand-binding and a structural subunit in the central nervous system of Drosophila melanogaster. J. Comp. Neurol. 335 149162.
  • 28
    Sgard F., Fraser S.P., Katkowska M.J., Djamgoz M.B., Dunbar S.J., Windass J.D. (1998) Cloning and functional characterisation of two novel nicotinic acetylcholine receptor α subunits from the insect pest Myzus persicae. J. Neurochem. 71 903912.
  • 29
    Sine S.M. & Taylor P. (1980) The relationship between agonist occupation and the permeability response of the cholinergic receptor revealed by bound cobra α-toxin. J. Biol. Chem. 255 1014410156.
  • 30
    Sumikawa K. & Nishizaki T. (1994) The amino acid residues 1-128 in the α subunit of the nicotinic acetylcholine receptor contain assembly signals. Mol. Brain Res. 25 257264.
  • 31
    Swick A.G., Janicot M., Cheneval-Kastelic T., McLenithan J.C., Lane M.D. (1992) Promoter-cDNA-directed heterologous protein expression in Xenopus laevis oocytes. Proc. Natl. Acad. Sci. USA 89 18121816.
  • 32
    Tomizawa M., Latli B., Casida J.E. (1996) Novel neonicotinoidagarose affinity column for Drosophila and Musca nicotinic acetylcholine receptors. J. Neurochem. 67 16691676.
  • 33
    Vernallis A.B., Conroy W.G., Berg D.K. (1993) Neurons assemble acetylcholine receptors with as many as three kinds of subunits and can segregate subunits among receptor subtypes. Neuron 10 451464.
  • 34
    Verrall S. & Hall Z.W. (1992) The N-terminal domains of acetylcholine receptor subunits contain recognition signals for the initial steps of receptor assembly. Cell 68 2331.
  • 35
    Wang F., Gerzanich V., Wells G.B., Anand R., Peng X., Keyser K., Lindstrom J. (1996) Assembly of human neuronal nicotinic receptor α5 subunits with α3, β2, and β4 subunits. J. Biol. Chem. 271 1765617665.
  • 36
    Wang Z.Z., Hardy S.F., Hall Z.W. (1996) Assembly of the nicotinic acetylcholine receptor. J. Biol. Chem. 271 2757527584.