Introduction

Top Summary Introduction Materials and methods Results Discussion References

Steroid hormones, including endogenously occurring metabolites of progesterone, testosterone, and deoxycorticosterone, as well as synthetic derivatives, act as allosteric modulators of the nicotinic acetylcholine (ACh) receptor (AChR) family of ligand-gated ion channels (reviewed in Franks and Lieb, 1994; MacDonald and Olsen, 1994; Lambert et al., 1995). Progesterone is a weak allosteric potentiator of type A -aminobutyric acid (GABA_A) receptors (Wu et al., 1990), whereas endogenous and synthetic 3-hydroxy derivatives act at submicromolar concentrations (Harrison et al., 1987; Gee et al., 1995). In contrast, at micromolar concentrations progesterone and its derivatives inhibit ACh-activated responses of both neuronal and muscle-type AChRs (Gillo and Lass, 1984; Valera et al., 1992; Ke and Lukas, 1996; Bullock et al., 1997) as well as glycine receptor-activated currents in spinal cord neurons (Wu et al., 1990). For GABA_A receptors electrophysiological studies show that potentiation by steroids is not associated with changes in channel conductance but with an increase in open channel probability at low agonist concentrations and an alteration of the kinetics of desensitization at high agonist concentrations (Twyman and MacDonald, 1992; Lambert et al., 1996; Zhu and Vicini, 1997). For muscle AChR, single-channel analyses of the acute effects of hydrocortisone and corticosterone indicate a dose-dependent decrease in the mean channel open time (Bouzat and Barrantes, 1996; Nurowska and Ruzzier, 1996). Although a mutation within the M2 ion channel domain that increases channel lifetime also increases hydrocortisone potency, there was no evidence for hydrocortisone competition with QX-222, an open channel blocker (Bouzat and Barrantes, 1996). Because substitutions of amino acids within AChR M4 hydrophobic segments at the lipid-protein interface can result in alteration of the duration of the open channel state (Bouzat et al., 1994; Lee et al., 1994; Lasalde et al., 1996), it is possible that steroid interactions with the AChR at the lipid interface result in functional antagonism.

To identify binding sites for progestin steroids in AChRs, we have characterized interactions of 17,21-dimethyl-19-nor-pregn-4,9-diene-3,20-dione (promegestone) with AChR-rich membranes isolated from the electric organ of the marine elasmobranch Torpedo californica. UV irradiation of promegestone results in the formation of a triplet-state biradical at the 3' ketone allowing a covalent bond to be formed between the steroid and an adjacent polypeptide (Benisek, 1977). Photoincorporation of [³H]promegestone has been used to identify several different progesterone receptors (Sadler and Maller, 1982; Stromstedt et al., 1990). The Torpedo electric organ is an extremely rich source of muscle-type AChR containing upwards of 100 mg of receptor protein/kg electroplacque tissue, thus facilitating the use of protein chemistry techniques to identify sites of photoincorporation of competitive and noncompetitive antagonists (reviewed in Hucho et al., 1996).

We report here that promegestone reversibly inhibits ACh-evoked currents of Torpedo AChR expressed in Xenopus oocytes and that it interacts with both the resting and desensitized states of the AChR, inhibiting the specific binding of radiolabeled noncompetitive antagonists to Torpedo AChR-rich membranes in the absence and in the presence of carbamylcholine, respectively. [³H]Promegestone photoincorporates into each of the AChR subunits with the extent of incorporation insensitive to AChR conformational state. No evidence of [³H]promegestone incorporation was detected in any of the channel lining M2 segments, but labeled residues were identified in the M4 hydrophobic segments at positions shown previously to be at the AChR-lipid interface. These results support the hypothesis that inhibition of AChRs by progestin steroids results from direct interactions at the AChR-lipid interface.

	Materials and Methods

Top Summary Introduction Materials and methods Results Discussion References

Materials. [³H]Promegestone (86.7 Ci/mmol; Fig. 1), nonradioactive promegestone, and [³H]phencyclidine ([³H]PCP; 52 Ci/mmol) were obtained from New England Nuclear (Boston, MA). [³H]Tetracaine (36 Ci/mmol) and [³H]histrionicotoxin ([³H]HTX, 60 Ci/mmol) were prepared at New England Nuclear by the tritium gas catalytic reduction of 3,5 dibromotetracaine and dl-decahydro(pentyl)histrionicotoxin, respectively. [³H]HTX was diluted with dl-perhydrohistrionicotoxin (HTX) to a radiochemical specific activity of 3 Ci/mmol for binding assays. 1-Azidopyrene (1-AP) was purchased from Molecular Probes (Eugene, OR). L-1-tosylamido-2-phenylethyl chloromethyl ketone-treated trypsin was purchased from Worthington Biochemical Corp. (Freehold, NJ), endoproteinase Lys-C (EKC) from Boehringer Mannheim (Indianapolis, IN) and Genapol C-100 (10%) from Calbiochem (La Jolla, CA). Prestained low molecular weight gel standards were purchased from Life Technologies, Inc. (Gaithersburg, MD)

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Structure of [3H]promegestone [17-methyl-3H]

AChR-Rich Membranes. AChR-rich membranes were isolated from the electric organ of T. californica (Marinus, Inc., Westchester, CA) as described previously (Pedersen et al., 1986). The final membrane suspensions in 36% sucrose/0.02% NaN₃ were stored at 80°C under argon and contained 1-1.5 nmol of ACh binding sites/mg of protein as measured by a direct [³H]AChR binding assay (Pedersen et al., 1986).

Electrophysiological Recordings. Plasmid cDNAs (pMXT) encoding wild-type -,-, and -subunits of Torpedo AChR were gifts from Dr. Michael White (Allegheny Health Sciences University, Philadelphia, PA), and the -subunit cDNA (pSP64 vector) was from Dr. Henry Lester (California Institute of Technology, Pasadena, CA). cDNAs were linearized with either XbaI (for -, -, and -subunits) or FspI (for -subunit). Linear DNAs were transcribed in vitro with SP6 RNA polymerase (Promega, Madison, WI). Isolated follicle-free oocytes were microinjected with 10 ng of subunit-specific RNAs in a molar ratio of ₂. Oocytes were maintained in low-Ca⁺⁺ ND96 solution containing 96 mM NaCl, 2 mM KCl, 0.3 mM CaCl₂, 1 mM MgCl₂, 5 mM HEPES ,and 50 µg/ml gentamicin (pH 7.6) for at least 48 h before use. Currents elicited by ACh were measured with a standard two-electrode voltage-clamp (OC-725B, Warner Instrument Corp., Hamden, CT) at a holding potential of 70 mV. Electrodes were filled with 3 M KCl and had resistance of 0.5-3.0 M. The recording chamber (~150 µl in volume) was continually perfused by gravity with low-Ca⁺⁺ ND96 (+1 µM atropine, pH 7.6). Stock solutions of steroid were prepared in ethanol and then diluted in low-Ca⁺⁺ ND96 so that the oocytes were exposed to ethanol concentrations <1% (v/v). Various concentrations of steroid were applied simultaneously with 3 µM ACh for 5 s through solenoid valves to the oocyte in the recording chamber. In some experiments, progesterone was perfused continuously through the chamber for up to 5 min, with responses to ACh tested periodically.

Binding of Cholinergic Ligands to AChR-Rich Membranes. Centrifugation assays were used to determine the equilibrium binding of [³H]ACh, [³H]HTX, [³H]PCP, and [³H]tetracaine to Torpedo AChR-rich membranes in Torpedo physiological saline (TPS; 250 mM NaCl, 5 mM KCl, 3 mM CaCl₂, 2 mM MgCl₂, 5 mM sodium phosphate, pH 7.0.). The final concentration of ethanol was <0.3% (v/v) for [³H]ACh or equal to 1% (v/v) for the ³H noncompetitive antagonists. For [³H]ACh, membrane suspensions (50 µg/ml of protein, 50 nM ACh binding sites) were pretreated with 0.3 mM diisopropylphosphofluoridate to inhibit acetylcholinesterase. Membranes were equilibrated at 4°C with [³H]ACh (50 nM) and varying concentrations of noncompetitive antagonist for 30 min, and 0.5 ml aliquots were centrifuged for 45 min at 10,000 rpm in a Sorvall SA-600 rotor. After complete removal of the supernatant, membrane pellets were resuspended in 0.1 ml of 10% SDS and pellet ³H was determined by liquid scintillation counting. For the ³H noncompetitive antagonists, membrane suspensions (0.5 mg of protein/ml, 0.6 µM ACh binding sites) were equilibrated with radioligand and appropriate concentrations of modulatory drug for 2-3 h at room temperature ([³H]HTX, 10 nM) or at 4°C ([³H]PCP, 6 nM; [³H]tetracaine, 2 nM) before centrifugation. Nonspecific binding of [³H]ACh was measured in the presence of 0.1 mM carbamylcholine, that of [³H]HTX and [³H]PCP in the presence of 0.2 mM meproadifen, and that of [³H]tetracaine in the presence of 0.2 mM tetracaine.

Photolabeling AChR-Rich Membranes with [³H]Promegestone. Membranes in TPS (2 mg of protein/ml) were incubated with [³H]promegestone at a final concentration 80-110 nM in the absence or presence of 0.2 mM carbamylcholine. After a 1-h incubation, suspensions were irradiated for 7 min at a distance of <1 cm with a 312-nm lamp (EB-Spectroline, Spectronics, Westbury, NY). Each sample was then pelleted (15,000g), and for analytical labelings the pellets were then solubilized in sample loading buffer (Laemmli, 1970) and submitted to SDS-polyacrylamide gel electrophoresis (PAGE). For preparative scale labelings (12-15 mg of AChR-rich membranes), the pellets were then resuspended in TPS at 2 mg/ml of protein and 1-AP was added to a final concentration of 350 µM. After a 30-min incubation, membranes were once again irradiated (365 nm for 5 min, at <1 cm; EN-Spectroline,) and each sample was pelleted. Samples were then solubilized in electrophoresis sample loading buffer and submitted to preparative scale SDS-PAGE. Labeling with the fluorescent hydrophobic probe 1-AP was used to assist in the initial identification and isolation of AChR subunits and in the subsequent isolation of hydrophobic segments of these subunits (Blanton and Cohen, 1994).

SDS-PAGE. SDS-PAGE was performed as described (Laemmli, 1970) with either 1.0-mm (analytical) or 1.5-mm thick (preparative scale) 8% T (total acrylamide concentration), 4% bis-acrylamide cross-linker concentration relative to total acrylamide concentration (C). For analytical gels, polypeptides were visualized by staining with Coomassie blue R-250 (0.25% w/v in 45% methanol and 10% acetic acid) and destaining in 25% methanol and 10% acetic acid. The gels were then impregnated with fluor (Amplify; Amersham, Arlington Heights, IL) for 20 min with rapid shaking, dried, and exposed at 80°C to X-OMAT LS film (Eastman Kodak Co., Rochester, NY) for various times (4-16 weeks). For preparative scale gels, polypeptides incorporating 1-AP were visualized from their associated fluorescence when the gels were illuminated at 365 nm on a UV-light box. Bands corresponding to AChR subunits were excised either for polypeptide isolation or in the case of the AChR -subunit for proteolytic digestion once the gel piece was transferred to the well of a 15% mapping gel (Cleveland et al., 1977; Pedersen et al., 1986). Mapping gels were composed of a 4.5% T, 2.6% C stacking gel and a 15% T, 2.6% C separating gel. The -subunit gel piece was overlaid with 350 µl of buffer (5% sucrose, 125 mM Tris-HCl, 0.1% SDS, pH 6.8) containing 250 µg of Staphylococcus aureus V8 protease. Electrophoresis was carried out overnight at 15 mA constant current.

Purification of Proteolytic Digests of [³H]Promegestone/1-AP-Labeled AChR Subunits. For EKC digestion, the acetone-precipitated 20-kDa V8 protease fragment of the -subunit (V8-20; Ser-173-Glu-338) was resuspended in 15 mM Tris-HCl, 0.1% SDS (pH 8.1), at 1-2 mg/ml protein. Approximately 1.5 U of EKC were added and incubated at room temperature for 6 days. The - and -subunits and the 10-kDa fragment of the -subunit (V8-10) (Asn-339-Gly-437) were digested with trypsin exactly as described (Blanton et al., 1998). Both trypsin and EKC digests were separated on individual 1.5-mm thick 16.5% T, 6% C tricine SDS-PAGE gels (Schagger and von Jagow, 1987; Blanton et al., 1998).

Sequence Analysis. Amino-terminal sequence analysis was performed on a ABI model 477A (Applied Biosystems, Foster City, CA) protein sequencer with gas phase cycles. Pooled HPLC samples were dried by vacuum centrifugation, resuspended in a small volume of 0.05% SDS (~20 µl), and immobilized on chemically modified glass fiber disks (Beckman Instruments). Approximately 30% of the released phenylthiohydantoin (PTH)-amino acids were separated by an on-line model 120A PTH-amino acid analyzer, and ~60% was collected for determination of released ³H by scintillation counting of each sample for three 5-min intervals. Initial yield (I_o) and repetitive yield (R) were calculated by nonlinear least-squares regression of the observed release (M) for each cycle (n): M = I_oRⁿ (PTH-derivatives of Ser, Thr, Cys, and His were omitted from the fit).

	Results

Top Summary Introduction Materials and methods Results Discussion References

To determine whether promegestone was an antagonist of Torpedo AChRs, we tested it as an inhibitor of AChRs expressed in Xenopus oocytes. When coapplied for 5 s with ACh, promegestone produced a dose-dependent and reversible inhibition of the ACh-induced currents (Fig. 2A). In the absence of promegestone, the ACh concentration-response relation was characterized by K_ap = 20 µM and a Hill coefficient of 1.8 (data not shown). When applied in the presence of 3 µM ACh, 30 µM promegestone produced ~50% reduction of the peak current response, whereas that same concentration resulted in a 90% reduction of the current seen after 5 s exposure. After a 1-min wash after exposure to 30 µM promegestone, there was a 50% recovery of the ACh peak current and a full recovery within 5 min. The potency of promegestone as well as its kinetics of inhibition were similar to that of progesterone (Fig. 2B). Although the time-dependent decrease of the ACh-response could indicate that both steroids increase the rate of agonist-dependent desensitization, it was also possible that neither steroid equilibrated with its binding site during the 5 s exposure. To determine whether kinetic factors were important determinants of the extent of inhibition seen for these drugs, ACh responses were determined after preincubation with progesterone. When oocytes were perfused with 1 µM progesterone for 5 min, there was a dramatic, time-dependent increase in progesterone potency, with 50% inhibition of the peak current seen after a 60-s preincubation (Fig. 2C). After 5 min, the peak current was reduced by 65%, and after a 2.5-min wash, there was a 40% recovery of the peak response (not shown). Although further experimentation is required to analyze the time and concentration dependence of the observed inhibition, these first experiments established that promegestone was in fact an AChR antagonist similar in potency and action to progesterone.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Inhibition of ACh-evoked currents of Torpedo AChRs expressed in Xenopus oocytes by promegestone and progesterone. Currents elicited by ACh from Torpedo AChRs expressed in Xenopus oocytes were measured with a standard two-electrode voltage clamp at a holding potential of 70 mV. A, once a stable response was observed for 3 µM ACh, responses were measured when ACh was applied simultaneously for 5 s with 10 and then 30 µM promegestone, or B, with 1, 3, and then 10 µM progesterone. Inhibition by promegestone and progesterone was reversible, because peak responses to ACh returned to control levels after exposure to 30 µM promegestone or to 10 µM progesterone (not shown). C, current response was recorded when 3 µM ACh was applied for 5 sec (Control), and then progesterone (1 µM) was applied continuously for 5 min, with test pulses of 3 µM ACh (and progesterone) at the indicated times. After 5 min exposure, the peak current was reduced to 35% of control, and after a 2.5 min wash there was a recovery to 60% of control (not shown).

Effects of Promegestone on [³H]ACh and ³H-Noncompetitive Antagonist Binding to Torpedo Membranes. Equilibrium binding of [³H]ACh was assayed at a concentration sufficient to occupy ~20% of sites. At this subsaturating concentration, the assay is sensitive to drugs that either increase or decrease ACh binding affinity (Boyd and Cohen, 1984). Promegestone at concentrations up to 100 µM did not inhibit the equilibrium binding of [³H]ACh, and, in fact, at concentrations between 1 and 30 µM it increased binding by 50% (Fig. 3A). In comparison, proadifen, an aromatic amine noncompetitive antagonist known to stabilize the desensitized state of the AChR (Boyd and Cohen, 1984), increased [³H]ACh binding by 300%.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effects of promegestone on the equilibrium binding of [3H]ACh, [3H]histrionicotoxin, [3H]phencyclidine, and [3H]tetracaine. A, AChR-rich membranes suspensions (50 µg protein/ml, 50 nM ACh binding sites), pretreated with diisopropylphosphofluoridate, were equilibrated at 4°C for 30 min with [3H]ACh (50 nM) and promegestone () or proadifen (), and bound [3H]ACh was determined by a centrifugation assay. B, AChR membranes (0.5 mg/ml, 0.6 µM ACh binding sites) were equilibrated at room temperature for 2 to 3 h with 10 nM [3H]HTX in the absence () or presence () of carbamylcholine, with 6 nM [3H]PCP in the presence of carbamylcholine (), or with 2 nM [3H]tetracaine () and the indicated concentrations of promegestone. Specific binding was determined as the difference between the total binding and the nonspecific binding in the presence of either 200 µM tetracaine (for [3H]tetracaine) or 200 µM meproadifen. In the absence of promegestone, [3H]HTX total binding was 4000 3H cpm in the absence and 4200 in the presence of carbamylcholine, and 350 3H cpm were bound nonspecifically. For [3H]PCP, total and nonspecific binding were 15,000 and 2,500 3H cpm, respectively, and for [3H]tetracaine they were 69,200 and 22,500 cpm.

Photoincorporation of [³H]Promegestone in the AChR. Experiments were first designed to characterize the general pattern of [³H]promegestone photoincorporation into Torpedo AChR-rich membranes, as well as to test the sensitivity of the photoincorporation to cholinergic ligands. Membranes (2 mg/ml) were equilibrated with 80 nM [³H]promegestone in the absence and in the presence of 200 µM carbamylcholine. This promegestone concentration provided a sufficient level of radioactivity to detect AChR subunit labeling even though only a small amount of the total ³H (~2%) was incorporated. After irradiation, membrane suspensions were pelleted and resuspended in electrophoresis sample buffer, and the pattern of incorporation was monitored by SDS-PAGE followed by fluorography. As is evident in the fluorograph of an 8% polyacrylamide gel (Fig. 4), there was incorporation of [³H]promegestone into each of the AChR subunits. Neither the pattern of incorporation into individual AChR subunits nor the overall labeling pattern was affected by the inclusion of 200 µM carbamylcholine (Fig. 4, lane 2). The additional presence of nonradioactive progesterone (1.2 µM, 12 µM, and 100 µM) had no observable effect on the extent of incorporation into any of the AChR subunits either in the presence or absence of agonist (data not shown). Based upon quantification of ³H incorporation in excised gel slices, ~0.1% of -subunits were labeled with the relative incorporation within subunits: ///:1.6/1.1/1.2/1 in the absence of agonist. Similar ratios were seen for photolabeling in the presence of agonist as well as from the amount of ³H present in subunits isolated from preparative scale labelings.

View larger version (48K): [in this window] [in a new window]   Fig. 4.   Photoincorporation of [3H]promegestone into AChR-rich membranes in the absence and in the presence of carbamylcholine. AChR-rich membranes were equilibrated with 80 nM [3H]promegestone in the absence (lane 1) or presence (lane 2) of 200 µM carbamylcholine and irradiated at 312 nm for 7 min. After photolysis, polypeptides were resolved on a 8% polyacrylamide gel and processed for fluorography (5-week exposure). The bulk of the labeled lipid and free photolysis products were electrophoresed from the gel with the tracking dye. AChR subunits and the Na+/K+ ATPase -subunit (NK) are indicated on the left.

Identification of the Sites of [³H]Promegestone Incorporation in the M4 Segments of the , , and  AChR Subunits. The M4 segments from AChRs labeled with [³H]promegestone in the presence of 200 µM carbamylcholine were isolated from tryptic digests of either the intact subunit (, , ~250 µg) or V8-10 (~125 µg). When the tryptic digest of -V8-10 was purified by reverse-phase HPLC, ³H counts eluted in a peak centered at ~82% solvent B (Fig. 5A) along with the peak of 1-AP fluorescence (data not shown). Amino-terminal sequence analysis (Fig. 6A) of the pool of HPLC fractions 31-35 revealed the presence of a primary sequence beginning at Tyr-401 (initial yield, 129 pmol) present at 15-fold higher level than a secondary sequence that began at Ser-388. In the pattern of ³H release shown in Fig. 6A, there were 620 cpm released in the first cycle, equivalent to 1.5% of loaded cpm and similar in magnitude to the 1.3% of ³H cpm washed off the filter by acid treatment before the first cycle of Edman degradation (data not shown). The progressively declining ³H release seen in the first 4 cycles of Edman degradation was likely to result either from release of peptide poorly adsorbed on the glass support or chemical instability of some incorporated ³H. A stable baseline of ³H release was achieved by cycle 5, and, in addition to the prominent ³H release in cycle 12, there was also clear, but lower level, release in cycles 8 and 18. Comparison of the pattern of release with the corresponding amino acids in the primary sequence indicates that the labeled amino acids include His-408 (0.7 cpm/pmol), Cys-412 (9 cpm/pmol), and Cys-418 (2 cpm/pmol). These same three residues were found to have incorporated [³H]promegestone when the M4 region was isolated from membranes labeled in the absence of agonist (data not shown). Cys-412 is the residue in M4 at the lipid interface that is labeled most efficiently by the hydrophobic probes 3-(trifluoromethyl)-3(m-[¹²⁵I]iodophenyl)diazirine[¹²⁵I](TID) and [³H]diazofluorene; Cys-418 is labeled by both probes, and His-408 is labeled only by [³H]diazofluorene (Blanton and Cohen, 1994; Blanton et al., 1998).

View larger version (43K): [in this window] [in a new window]   Fig. 5.   Reverse-phase HPLC purification of proteolytic fragments of [3H]promegestone-labeled AChR subunits containing M4 and M2 hydrophobic segments. AChR subunits were isolated from AChR-rich membranes (12 mg protein) labeled with [3H]promegestone in the presence of carbamylcholine and in the absence () or presence () of 60 µM phencyclidine. Proteolytic digests of subunits were purified by SDS-PAGE and reverse-phase HPLC as described under Materials and Methods. A, trypsin digests of V8-10 (Asn-339-Gly-437) were purified directly; B, purification of a 10-kDa band (-K-10K) isolated by tricine-SDS-PAGE from Endoproteinase-Lys C digests of V8-20 (Ser-173-Glu-338). Trypsin digests of -, -, and -subunits were fractionated by tricine-SDS-PAGE to isolate from -subunit 5 kDa (-T-5K) and 7-kDa (-T-7K) bands, from -subunit a 5-kDa band (-T-5K), and from -subunit a 5 kDa band (-T-5K). C and D, reversed-phase HPLC purification of -T-5K and -T-7K, respectively. E and F, purification of -T-5K and -T-5K respectively.

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Radioactivity and mass release upon amino-terminal sequence analysis of the M4 region of the -, -, and -subunits. For AChRs labeled with [3H]promegestone in the presence of agonist, the M4 regions of the - -, and -subunits were isolated from tryptic digests of V8-10 (Asn-339-Gly-437) and intact - and -subunits. M4 containing proteolytic fragments -T-4K, -T-4K, and -T-5K were purified by reverse-phase HPLC (Fig. 5, A, C, and E) and the fractions containing the peaks of 3H were pooled (fractions 28-32, 28-32, and 31-35, respectively) and sequenced. For each sample 60% of each cycle of Edman degradation was analyzed for released 3H () and 30% for PTH-amino acids () with the dashed lines corresponding to the exponential decay fit of the amount of detected PTH-amino acids for the peptides containing M4. The amino acid sequence of the sequenced peptide containing the M4 region is shown above each panel. A, 3H release from -T-4K [Io = 129 pmol; r = 89%; 44,000 cpm loaded/5800 remaining on filter]. B, 3H release from -T-4K [Io = 129 pmol; r = 95%; 7700 cpm loaded/2100 remaining on filter]. C, 3H release from -T-5K [Io = 116 pmol; r = 88%; 7800 cpm loaded/1950 remaining on filter].

Radiochemical Sequence Analysis of M2 Segments of AChR -, -, and -subunits. Subunit proteolytic fragments beginning at the amino termini of the M2 segments of AChR -, -, and -subunits were isolated from AChRs labeled with [³H]promegestone in the presence of carbamylcholine. For -subunit, an EKC digest of V8-20 (~280 µg) was fractionated by tricine SDS-PAGE, and an ~10- kDa fragment (-K-10K), previously shown to contain the M2-M3 region (Pedersen et al., 1992), was isolated from the gel (see Materials and Methods). When the material eluted from the -K-10K fragment was further purified by reversed-phase HPLC (Fig. 5B), the majority of ³H counts eluted in a peak centered at 84% solvent B. HPLC fractions 32-35 were pooled and sequenced (Fig. 7A). The primary sequence, which began at Met-243, the amino terminus of the M2 region (85 pmol), was present at 10-fold higher level than a secondary sequence beginning at the amino terminus of M3 (Tyr-277, 7 pmol). No significant ³H release above background was observed in any of the 21 sequencing cycles.

View larger version (27K): [in this window] [in a new window]   Fig. 7.   Radioactivity and mass release upon amino-terminal sequence analysis of the M2 regions of -, -, and -subunits. For AChRs labeled with [3H]promegestone in the presence of agonist, proteolytic fragments beginning at the amino termini of the M2 regions of the - -, and -subunits were isolated from either an EKC digest of V8-20 (Ser-173-Glu-338) or tryptic digests of intact - and -subunits. The -K-10K, -T-7K, and T-5K fragments containing M2 segments isolated by tricine SDS-PAGE were further purified by reverse-phase HPLC (Figs. 5B, D, and F, respectively) and the fractions containing the peaks of 3H were pooled (fractions 32-35, 30-34, and 30-33, respectively) and sequenced. For each sample 60% of each cycle of Edman degradation was analyzed for released 3H() and 30% for PTH-amino acids () with the dashed lines corresponding to the exponential decay fit of the amount of detected PTH-amino acids for the each of the M2 containing peptides. The amino acid sequence of the sequenced peptides containing the M2 region is shown above each panel. A, 3H release from -K-10K [Io = 85 pmol; r = 85.8%; 6200 cpm loaded/2000 cpm remaining on filter]. B, 3H release from -T-7K [Io = 86 pmol; r = 90%; 3,600 cpm loaded/1000 cpm remaining on filter]. C, 3H release from -T-5K [Io = 90 pmol; r = 91%; 7,700 cpm loaded/1,200 cpm remaining on filter].

	Discussion

Top Summary Introduction Materials and methods Results Discussion References

We have shown that the synthetic progestin, promegestone, is a NCA of the Torpedo AChR. Promegestone reversibly inhibits ACh-elicited currents for Torpedo AChRs expressed in Xenopus oocytes with the same potency as progesterone. Because promegestone concentrations up to 100 µM increase the [³H]ACh equilibrium binding affinity to AChR-rich membranes, promegestone is a desensitizing NCA. Interestingly, although it is only a weak inhibitor (IC₅₀ ~ 100 µM) of [³H]HTX binding to AChRs in the desensitized state, it is a potent inhibitor (IC₅₀ = 9 µM) of [³H]PCP binding. Because binding of HTX and PCP appear formally competitive (Heidmann et al., 1983; White et al., 1991), this result appears at first surprising. However, studies with TID, an uncharged, photoactivatable NCA, provide clear evidence that the binding sites for PCP and HTX can be differentiated (White et al., 1991; White and Cohen, 1992): TID is a potent, competitive inhibitor of [³H]PCP binding to AChRs in the desensitized state, whereas it is an allosteric inhibitor of [³H]HTX binding. Although HTX and PCP are presumed to bind within the lumen of the ion channel, their site(s) of high-affinity binding have not been mapped directly. The high-affinity binding site for [³H]tetracaine in the AChR in the resting state has been shown by direct photoaffinity labeling to overlap with the [¹²⁵I]TID site in the M2 ion channel domain (White and Cohen, 1992; Gallagher and Cohen, 1994; Gallagher, 1996). Although promegestone at high concentrations (IC₅₀ ~75 µM) inhibits [³H]tetracaine binding, that inhibition probably occurs because promegestone stabilizes the AChR desensitized state that binds [³H]tetracaine only weakly. Although promegestone clearly can be a potent inhibitor of NCA binding, the results of the radioligand assays provide no clear indication of the nature of the promegestone binding site(s) in the AChR.

When [³H]promegestone was photoincorporated into AChR-rich membranes, its incorporation was readily detected in residues in the M4 segments of the -, -, and -subunits that previous work has shown are exposed to lipid (Fig. 8; Blanton and Cohen, 1994; Blanton et al., 1998). In contrast, no incorporation was detected in the sequence analysis of the -, -, or -M2 segments isolated from AChRs labeled in either the absence or presence of agonist. The simplest interpretation of the photolabeling data is that, consistent with its lipophilicity, the steroid promegestone binds to and interacts with the AChR at the lipid-protein interface. The inability of increasing doses of progesterone to affect the photoincorporation of [³H]promegestone, at least at the subunit level, is also consistent with "nonspecific" binding at the lipid-protein interface. Although promegestone interaction at the AChR-lipid protein interface does not meet the classical criteria for a specific drug-receptor binding site, this does not preclude this promegestone-AChR interaction from affecting AChR function. Additional studies are required to determine whether promegestone interactions at the AChR-lipid interface result in the observed inhibition of NCA binding or of ACh-induced ion currents. Also, additional studies are required to determine whether promegestone is interacting at sites on the AChR surface normally occupied by cholesterol, which is present at 15% (w/w) in the AChR-rich membranes and at an effective concentration of 1 mM. There is strong evidence that cholesterol interacts directly with AChRs and that neutral lipids are required to maintain the AChR in a state responsive to agonist (Rankin et al., 1997 and references therein). Inhibition by promegestone (or progesterone) might occur because it displaces cholesterol from a functionally important site.

View larger version (42K): [in this window] [in a new window]   Fig. 8.   Corey-Pauling-Koltun models of the -M4 helix. The amino acids labeled by [3H]promegestone (His-408, Cys-412, and Cys-418) are also labeled by the hydrophobic probe [3H]diazofluorene (as is Met-415 (Blanton et al., 1998)) and by [125I]TID [which also reacts with Thr-422 and Val-425 (Blanton and Cohen, 1994)].

Although our studies have found no evidence that [³H]promegestone binds within the M2 ion channel domain, it remains possible that it binds within the channel but is incapable of photoincorporating into the predominantly aliphatic residues contained within each of the M2 segments. Promegestone photoincorporation proceeds through a "triplet state" biradical, which should readily incorporate into C-H bonds (Benisek, 1977). However, although very limited, the residues that have been shown to react with [³H]promegestone [Met-759 and Met-909 (human progestin receptor); Met-622 and Cys-754 (rat glucocorticoid receptor) (Stromstedt et al., 1990); His-408, Cys-412, and Cys-418 (M4); Tyr-441, Cys-447 (M4); and Cys-451, Trp-453 (M4) AChR, this report] do not include any aliphatic residues. In addition, it is also possible that there is an extracellular binding site for promegestone, but that the efficiency of [³H]promegestone photoincorporation into this region was too low for us to detect. Finally, it also remains to be determined whether [³H]promegestone is photoincorporated in the M1 or M3 hydrophobic segments (or in the AChR extracellular domain) because the subunit fragmentation strategies used in this study were chosen specifically to identify possible sites of incorporation in M2 and M4 segments.

By interacting with residues situated at the lipid-protein interface of the AChR, promegestone could allosterically alter the structure of the ion channel, effecting both ion conductance and binding of NCAs. Several recent reports provide support for this conclusion. Mutations at lipid-exposed residues in the M4 segment dramatically affect open channel lifetimes (Bouzat et al., 1994, 1998; Lee et al., 1994; Lasalde et al., 1996), and a mutation within the M2 ion channel domain that increases channel lifetime also increases hydrocortisone potency (Bouzat and Barrantes, 1996).

Finally, progesterone inhibits AChR activity but potentiates GABA_A receptor currents in the same concentration range (EC₅₀ = 26 µM, Wu et al., 1990). Given the emerging evidence of structural homology between these two receptors (reviewed in Sigel and Buhr, 1997), this raises the possibility that steroids may differentially modulate the activities of these two receptors (as well as other members of the AChR family) by interacting with residues at a common site, potentially at the lipid-protein interface. On the other hand the ACh and GABA_A receptors have different steroid structure-activity relationships (Wu et al., 1990). Although these differences may simply reflect the different modulatory affects progesterone and other steroids have on these two receptors, it may also indicate separate steroid binding domains. In this respect, future studies will be aimed at identifying the site(s) of [³H]promegestone incorporation in GABA_A receptor subunits.