Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jan 26;168(3):377-389.e12.
doi: 10.1016/j.cell.2016.12.033.

Crystal Structure of an LSD-Bound Human Serotonin Receptor

Affiliations
Free PMC article

Crystal Structure of an LSD-Bound Human Serotonin Receptor

Daniel Wacker et al. Cell. .
Free PMC article

Abstract

The prototypical hallucinogen LSD acts via serotonin receptors, and here we describe the crystal structure of LSD in complex with the human serotonin receptor 5-HT2B. The complex reveals conformational rearrangements to accommodate LSD, providing a structural explanation for the conformational selectivity of LSD's key diethylamide moiety. LSD dissociates exceptionally slow from both 5-HT2BR and 5-HT2AR-a major target for its psychoactivity. Molecular dynamics (MD) simulations suggest that LSD's slow binding kinetics may be due to a "lid" formed by extracellular loop 2 (EL2) at the entrance to the binding pocket. A mutation predicted to increase the mobility of this lid greatly accelerates LSD's binding kinetics and selectively dampens LSD-mediated β-arrestin2 recruitment. This study thus reveals an unexpected binding mode of LSD; illuminates key features of its kinetics, stereochemistry, and signaling; and provides a molecular explanation for LSD's actions at human serotonin receptors. PAPERCLIP.

Keywords: GPCR; crystallography; hallucinogens; serotonin receptor; structure-function.

Figures

Figure 1
Figure 1. Architecture and ligand-receptor interactions of the LSD-bound human 5-HT2B receptor
(A) 5-HT2BR cartoon representation (light blue) with helices labeled according to GPCR nomenclature. LSD is shown as a stick model with carbons, nitrogens, and oxygens colored in magenta, blue, and red, respectively. The LSD stick model is overlaid with a semi-transparent surface representation of the compound. (B) Close up view of LSD and the orthosteric binding site of the receptor from the membrane. (C) Close up view of LSD and the orthosteric binding site of the receptor from the extracellular space (D) 2D representation of LSD, Ergotamine (ERG), and 5-hydroxytryptamine (5-HT/serotonin). LSD belongs to the class of ergolines like ERG, and contains a diethylamide substituent (highlighted in light blue) connected to the ergoline scaffold (highlighted in yellow). Ergolines contain a tryptamine core scaffold (dark blue) like the endogenous ligand 5-hydroxytryptamine (5-HT/serotonin). Diagram of interactions between LSD and the receptor in the ligand binding pocket is shown, with the hydrogen bonds between D1353.32 and the LSD basic nitrogen in yellow, and G2215.42 and the LSD indole nitrogen indicated by red dashes, respectively. Residues are labeled according to Ballesteros-Weinstein nomenclature. Residues highlighted in red show significant changes between LSD- and ERG-occupied 5-HT2BR while residues highlighted in green show a significant interaction with ERG but not LSD. See also Figure S1.
Figure 2
Figure 2. Conformational differences in the ligand binding pockets of LSD- and ERG-bound 5-HT2BR
(A) Close-up view of the orthosteric pockets of 5-HT2BR (light blue) bound to LSD (magenta) superposed with 5-HT2BR (green) bound to ERG (dark green). Compounds and relevant sidechains are shown as sticks and residues are labeled according to Ballesteros-Weinstein nomenclature. Surface representation of M2185.39 illustrates how ERG binding requires a conformational change to accommodate the phenyl ring of ERG. Insert shows schematic illustrating that different ergoline substituents (R, red circle) and their interactions with the receptor likely determine the orientation of the ergoline scaffold, which seems to be able to rotate around the hydrogen bond to the conserved aspartate D3.32. (B) View of the 5-HT2BR ligand binding pocket from the extracellular space highlighting conformational differences in helix and loop positions in response to binding of ERG (green) vs LSD (light blue). Distances were measured between the Cα atoms of T1142.64, L209EL2, L3476.58, N354EL3, and T3567.29. (C–D) Surface representation illustrating shape of orthosteric binding pocket in the 5-HT2BR/LSD complex (C) and the 5-HT2BR/ERG complex (D). This particular cross-section cuts through M2185.39 in such a way that the extended binding site appears smaller in the presence of ERG than in the presence of LSD, although calculation of binding pocket volume with CASTp shows a 28.6% decrease in overall volume of LSD vs ERG. See also Figure S2 and S3.
Figure 3
Figure 3. Diethylamide configuration determines LSD pharmacology at 5-HT2BR and 5-HT2AR
(A) Side and top view of LSD (magenta) bound 5-HT2BR (light blue) crystal structure overlaid with small molecule crystal structure of unbound LSD (yellow) highlight differences in LSD’s diethylamide conformation. (B) Snapshots of LSD (magenta) from a 5-HT2BR-bound MD simulation show that LSD maintains its 5-HT2BR-bound crystallographic conformation, with substantial fluctuation only in the terminal methyl groups. Snapshots are aligned on the ergoline ring system. (C) Chemical structures of LSD (purple) and diethyl constrained lysergamides, (S,S)-Azetidide (SSAz, green), (R,R)-Azetidide (RRAz, red), and lysergic acid amide (LSA, orange) indicating 5-HT2BR-bound LSD diethyl conformation resembles the conformation of (S,S)-Azetidide. (D) Lysergamide-mediated β-arrestin2 recruitment at 5-HT2BR and 5-HT2AR (n =3) highlights the importance of diethylamide conformation for LSD’s function. (E) Lysergamide-mediated Gq-calcium flux at 5-HT2BR and 5-HT2AR (n =3) indicates lack of stereospecific preference for LSD azetidides in this pathway. See also Figure S4.
Figure 4
Figure 4. Mutation of an EL2 “lid” decreases LSD’s long residence time at 5-HT2BR, which affects functional selectivity
(A) Comparison of LSD dissociation from wild type 5-HT2BR and L209AEL2 mutant (n =3) at 37°C shows a slow LSD off-rate at the wild type and a faster off-rate at the mutant. (B) (Left and center) In the 5-HT2BR crystal structure, EL2 residues 207–214 form a lid (dark blue, with other nearby residues in light blue) that covers the binding pocket. (Right) In MD simulations of the wild type, this lid occludes access to the binding pocket most of the time, but occasionally moves aside. (C) (Left) LSD (magenta) bound to 5-HT2BR (light blue) orthosteric pocket viewed from membrane and extracellular space, with EL2 residue L209EL2 highlighted in red. (Right) The L209EL2 sidechain (now viewed from roughly the opposite direction) forms extensive hydrophobic contacts with residues in TMs III, IV, and V. (D) Mutating the sidechain L209EL2 to alanine does not lead to a more exposed binding pocket in the crystal structure (left); exposure of the binding pocket still depends on motion of the lid, as seen in simulation (right). (E) The lid is more mobile in simulations of the mutant (red) than in simulations of the wild type (gray). Each image shows six representative snapshots of the lid from simulation, with the remainder of the receptor in light gray. The alpha carbon atom of residue 209 is shown as a sphere. (F) Root mean square fluctuation (RMSF) of the alpha carbon of each lid residue provides a quantitative measure of mobility, demonstrating that the lid fluctuates more in simulations of the L209AEL2 mutant than in simulations of the wild-type receptor, whether or not LSD is bound. RMSF measures the fluctuations of an atom around its average position during simulation. Error bars show standard error of the mean across 3–9 simulations per condition. *represent significant differences between WT and L209AEL2 simulations (p < 0.01, as measured by a two-sided Welch’s t-test). (G) Arrestin and Gq functional activities indicate that 5-HT2BR L209AEL2 mutant selectivity disrupts β-arrestin2 recruitment activity, leaving Gq function intact (n=3). See also Figure S3 and S5.
Figure 5
Figure 5. Accelerated ligand kinetics also affect pharmacological profile of LSD at 5-HT2AR
(A) Docking pose of LSD (magenta) in 5-HT2AR model (beige) illustrates similar binding mode compared to 5-HT2BR/LSD complex structure. L229EL2 (red) is in the same position as L209EL2 in the 5-HT2BR/LSD complex structure. (B) Comparison of LSD dissociation from wild type 5-HT2AR and L229AEL2 mutant (n =3), shows increased LSD off-rate at the mutant receptor. (C) Arrestin and Gq functional activities indicate that 5-HT2AR L229AEL2 mutant selectivity disrupts β-arrestin2 recruitment activity, leaving Gq function intact (n=3). (D) Kinetic measurement of LSD-mediated β-arrestin2 recruitment at wild type 5-HT2AR and L229AEL2 mutant using bioluminescence resonance energy transfer (BRET). At the wild type receptor, LSD’s potency and efficacy increase with longer compound stimulation, whereas LSD exhibits weak potency and efficacy at the L229AEL2 mutant that does not change over time. (E) Kinetic measurement of LSD-mediated β-arrestin2 recruitment at wild type 5-HT2BR and the L209AEL2 mutant using bioluminescence resonance energy transfer (BRET). At the wild type receptor, LSD’s potency and efficacy increase with longer compound stimulation, whereas LSD exhibits weak potency and efficacy at the L209AEL2 mutant that does not change over time. (F) Heat map of time-dependent alterations in signaling for WT and mutant 5-HT2AR and 5-HT2BR. The heat map [see Supplementary Methods for details] illustrates how signaling as quantified by calculating transduction coefficients are altered in a time-dependent fashion at WT and mutant 5-HT2AR and 5-HT2BR. Time is measured in minutes. See also Figure S5.

Comment in

  • A Receptor on Acid.
    Chen Q, Tesmer JJG. Chen Q, et al. Cell. 2017 Jan 26;168(3):339-341. doi: 10.1016/j.cell.2017.01.012. Cell. 2017. PMID: 28129534 Free PMC article.

Similar articles

See all similar articles

Cited by 72 articles

See all "Cited by" articles

Publication types

LinkOut - more resources

Feedback