NMR studies in dodecylphosphocholine of a fragment containing the seventh transmembrane helix of a G-protein-coupled receptor from Saccharomyces cerevisiae

Abstract

The structure and dynamics of a large segment of Ste2p, the G-protein-coupled alpha-factor receptor from yeast, were studied in dodecylphosphocholine (DPC) micelles using solution NMR spectroscopy. We investigated the 73-residue peptide EL3-TM7-CT40 consisting of the third extracellular loop 3 (EL3), the seventh transmembrane helix (TM7), and 40 residues from the cytosolic C-terminal domain (CT40). The structure reveals the presence of an alpha-helix in the segment encompassing residues 10-30, which is perturbed around the internal Pro-24 residue. Root mean-square deviation values of individually superimposed helical segments 10-20 and 25-30 were 0.91 +/- 0.33 A and 0.76 +/- 0.37 A, respectively. 15N-relaxation and residual dipolar coupling data support a rather stable fold for the TM7 part of EL3-TM7-CT40, whereas the EL3 and CT40 segments are more flexible. Spin-label data indicate that the TM7 helix integrates into DPC micelles but is flexible around the internal Pro-24 site, exposing residues 22-26 to solution and reveal a second site of interaction with the micelle within a region comprising residues 43-58, which forms part of a less well-defined nascent helix. These findings are discussed in light of previous studies in organic-aqueous solvent systems.

FIGURE 1

Cartoon of the region of Ste2p examined in this investigation. EL3, third extracellular loop. TM7, seventh transmembrane helix. CT40, 40 residues of the cytosolic tail.

FIGURE 2

[¹⁵N, ¹H]-HSQC spectrum recorded at 700 MHz, 310 K, on a 0.5-mM sample of the peptide in 300 mM DPC. The sequence-specific assignment has been annotated.

FIGURE 3

Strips from the 3D CBCA(CO)NH and the HNCACB spectra taking at various amide proton positions displaying the assignment process of the ¹⁵N,¹³C, and ¹H chemical shifts of backbone and C_β atoms. The ¹⁵N chemical shift, at which the strip was extracted, is displayed above the strips.

FIGURE 4

[¹⁵N, ¹H]-HSQC spectra of selectively labeled constructs. The reference spectrum of ¹⁵N-uniformly labeled peptide is shown in panel A. Panels B–D display spectra from Ala, Ser, and Leu residues, respectively.

FIGURE 5

(Top) Sequence plot displaying characteristic upper distance restraints along the sequence derived from NOEs. For those residues for which dihedral angle restraints derived from ¹³C chemical shifts were applied, a solid circle is placed under the residue number when restraining φ in the range of [−120.0.. −20.0°] and ψ to [−120.0.. −20.0°] and for open circles when restraining to the much looser bounds of [−120.0.. 80.0°] for φ and [−100.0.. 60.0°] for ψ. (Bottom) ³J_HNα scalar coupling constants as extracted by the INFIT method from the HSQC spectrum. The region containing reduced scalar couplings representative of helical tendencies is shaded.

FIGURE 6

Presentation of the peptide backbone of the 20 lowest energy calculated conformers of E3-TM7-T40 in DPC micelles. Backbone atoms of residues 10–22 (A) and 26–31 (B) have been used to superimpose the structures, and the corresponding bonds are coded in black. Panels C and D display individual conformers with different dihedrals in the segment comprising Leu-23-Pro-24-Leu-25.

FIGURE 7

RDCs of E3-TM7-T40 measured at 700-MHz proton frequency.

FIGURE 8

Relative peak volumes of signals computed from HSQC spectra recorded in the presence of low-power presaturation on the water resonance during the relaxation delay relative to a reference experiment without presaturation. In this figure and in Figs. 10 and 11, the putative TM region is shaded.

FIGURE 9

Values of the ¹⁵N{¹H}-NOE recorded at 700-MHz proton frequency.

FIGURE 10

Relative HSQC signal intensities measured on EL3-TM7-CT40 in the presence of various spin labels: 5-doxylstearate (A), 16-doxylstearate (B), and Gd-DOTA (C). The chemical structures of the spin labels are indicated in D.

FIGURE 11

Possible insertion modes of the TM portion of the peptide into the DPC micelle. The areas of largest influence for the two spin labels 5-doxylstearate (top, A and B) and 16-doxylstearate (bottom, C and D) have been shaded. Below the micelles expected residual signal intensities for the various insertion topologies are depicted. In the center, the experimental signal attenuations are shown for a segment containing the putative TM segment (see text).

FIGURE 12

Plot of the free energies for transferring whole amino acids as determined by Wimley and White into the membrane interior (left) and the membrane-water interface (right) for the sequence of the investigated peptide (72,73). The area of favorable energies is shaded.