Potentiation of Neuronal Nicotinic Receptors by 17β-Estradiol: Roles of the Carboxy-Terminal and the Amino-Terminal Extracellular Domains

Abstract

The endogenous steroid 17β-estradiol (βEST) potentiates activation of neuronal nicotinic receptors containing α4 subunits. Previous work has shown that the final 4 residues of the α4 subunit are required for potentiation. However, receptors containing the α2 subunit are not potentiated although it has these 4 residues, and only one amino acid difference in the C-terminal tail (FLAGMI vs. WLAGMI). Previous work had indicated that the tryptophan residue was involved in binding an analog of βEST, but not in potentiation by βEST. To determine the structural basis for the loss of potentiation we analyzed data from chimeric subunits, which indicated that the major factor underlying the difference between α2 and α4 is the tryptophan/phenylalanine difference, while the N-terminal extracellular domain is a less significant factor. When the tryptophan in α4 was mutated, both phenylalanine and tyrosine conferred lower potentiation while lysine and leucine did not. The reduction reflected a reduced maximal magnitude of potentiation, indicating that the tryptophan is involved in transduction of steroid effects. The regions of the α4 N-terminal extracellular domain involved in potentiation lie near the agonist-binding pocket, rather than close to the membrane or the C-terminal tail, and appear to be involved in transduction rather than binding. These observations indicate that the C-terminal region is involved in both steroid binding (AGMI residues) and transduction (W). The role of the N-terminus appears to be independent of the C-terminal tryptophan and likely reflects an influence on conformational changes caused during channel activation by agonist and potentiation by estradiol.

Fig 1. Representative responses of oocytes injected with the indicated subunits.

In each case the lower bar above the traces indicates the time of application of ACh while the upper bar shows the time of application of the same concentration of ACh plus 10 μM βEST. The concentration of ACh was chosen to produce a response of about 10% of maximal. The horizontal scale bar in the upper left panel shows 10 sec for all panels. For the trace obtained with α4&β2 the ACh concentration was 3 μM, for α4&β4 3 μM, for α4(F)&β4 3 μM, for α2&β2 10 μM, for α2&β4 20 μM and for α2(W)&β4 20 μM. For the figure the data were smoothed by averaging 10 points.

Fig 2. Chimeric subunits used to determine regions contributing to potentiation by βEST.

Panel A shows a linear cartoon of a nicotinic subunit, with the amino-terminus (“N”) at the left. Some major structural features are indicated: in the amino-terminal extracellular domain there are 6 domains involved in agonist binding, the A, B, C loops contributed by the α subunit and the D, E, F loops contributed by the β subunit. The 4 transmembrane regions are indicated by the enlarged boxes; the channel is formed by TM2 helices contributed by the 5 subunits in the receptor. The 4 regions studied were the N-terminal domain (blue; extending from the amino-terminus to V(210)IRRLP sequence at the start of TM1 (the amino acid residue numbering is for the mature α4 subunit; NCBI Reference Sequence: NP_000735.1). The TM1-TM3 regions (gray with vertical hatching) are identical in the mouse α2 and human α4 subunits. The region called the cytoplasmic loop (red with forward hatching) extends from N(301)VHHRSP at the end of TM3 to D(571)RIFL at the start of TM4. TM4 (green with backwards hatching) is the region from D(571)RIFL to FLP(592)P. The tryptophan in α4 is in the carboxy-terminal region, W(594)LAGMI. The cytoplasmic loop is highly divergent in nicotinic subunits, and contains between about 150 and 270 amino acids. Panel B shows the mean potentiation ratio for the chimeras studies. Chimeras are identified by which subunit contributed the relevant portion of the construct, so α(2,2,4,W) indicates an α subunit with α2 sequence in the amino-terminus, α2 sequence in the cytoplasmic loop, α4 sequence in the TM4 region and W in the C-terminal tail. The figure shows the mean ± SE potentiation ratio produced by βEST (see Table 2 for numbers of observations).

Fig 3. The effects of mutations on the concentration-effect relationship for potentiation.

Data are shown for mutations of the tryptophan residue (WLAGMI; panels A and B) and the methionine residue (WLAGMI; panel C). ACh was applied at a concentration producing a response that produced a response of about 5 to 7% of maximal in the absence of steroid. The α subunit was expressed with β4 at an injection ratio of 20:1. The concentration-response relationship for potentiation was characterized by fitting a 3 parameter Hill equation to the ratios (Z[drug] = 1 + Zmax (1 / (1 + (EC₅₀/[drug])^nHill), where Z is the response to a concentration of ACh in the presence of drug relative to that in the absence, Zmax is the maximal effect, EC₅₀ is the concentration producing half-maximal potentiation, and nHill is the Hill coefficient. The fits did not include points that appeared to demonstrate block (e.g. data with 30 μM 17α-vinylestradiol). The data from each cell were fit separately, giving the following mean parameter estimates for βEST: α4: Zmax = 1.12 ± 0.12 (11 cells) and EC₅₀ = 4.92 ± 0.48 μM; for α4(F): 0.41 ± 0.06 (3) P = 0.0002 and 10.43 ± 2.00 μM P = 0.4; for α4(G): 0.56 ± 0.03 (4) P = 0.001 and 4.90 ± 0.38 μM P = 0.98; and for α4(MtoF): 1.00 ± 0.03 (8) P = 0.34 and 11.76 ± 0.39 μM P < 0.0001, with values given as mean ± SE (number of cells) and significance of difference to α4 by t-test. The best-fitting estimates for 17α-vinylestradiol are: α4: 1.93 ± 0.06 (3) and 3.84 ± 0.27 μM and α4(F): 1.41 ± 0.08 (4) P = 0.004 and 7.32 ± 0.68 μM P = 0.010. The points show mean ± SE while the lines show the relationship predicted by the mean fit parameters.

Fig 4. The effects of N-terminal chimeras on potentiation.

Panel A shows a homology model of the α4 subunit (see Methods). A single subunit is shown, with the ion channel to the left and the extracellular solution at the top. The approximate location of the membrane is shown by the dashed box, note that the membrane spanning region is truncated in this view. The D to A loop region (MMT(57) to IVL(97)) is shown in red with amino acid side chains shown in stick representation, while the E loop (VTH(109) to QWT(124)) is shown in yellow. The carboxyl terminal region is shown in magenta and indicated by the red oval. In aligning the α4 subunit to sequence in the GluCl structure, the last amino acid in the structure aligned with the leucine in α4, so the AGMI residues are not shown. Note that the regions exchanged are far from the membrane and the C-terminal domain, but include portions of the ACh-binding site. Panel B shows potentiation for the different chimeras. The chimeric regions interchanged are indicated along the abscissa; chimeras were constructed on the α4 (filled triangles) and α4(F) (filled circles) subunits inserting sequence from the α2 subunit. Homologous transfers were made by transferring sequence from the α4 subunit into the α2 (hollow circles) and α2(W) (hollow triangles) subunits. The lines through the points for points show the lines predicted by linear regression on the source of the inserted region (see Methods). The slopes are -0.072 for chimeras on α4 and -0.095 on α4(F), and 0.008 for α2 and 0.066 for α2(W). None of the slopes differ significantly from 0 (P > 0.09). However, as shown in Table 5 in all cases the potentiation of receptors containing α4(DtoA+E) regions was significantly larger than those containing the homologous α subunits containing α2(DtoA+E) regions. Data values are given in Table 5.