Structural determinants of a conserved enantiomer-selective carvone binding pocket in the human odorant receptor OR1A1

Abstract

Chirality is a common phenomenon within odorants. Most pairs of enantiomers show only moderate differences in odor quality. One example for enantiomers that are easily discriminated by their odor quality is the carvones: humans significantly distinguish between the spearmint-like (R)-(-)-carvone and caraway-like (S)-(+)-carvone enantiomers. Moreover, for the (R)-(-)-carvone, an anosmia is observed in about 8% of the population, suggesting enantioselective odorant receptors (ORs). With only about 15% de-orphaned human ORs, the lack of OR crystal structures, and few comprehensive studies combining in silico and experimental approaches to elucidate structure-function relations of ORs, knowledge on cognate odorant/OR interactions is still sparse. An adjusted homology modeling approach considering OR-specific proline-caused conformations, odorant docking studies, single-nucleotide polymorphism (SNP) analysis, site-directed mutagenesis, and subsequent functional studies with recombinant ORs in a cell-based, real-time luminescence assay revealed 11 amino acid positions to constitute an enantioselective binding pocket necessary for a carvone function in human OR1A1 and murine Olfr43, respectively. Here, we identified enantioselective molecular determinants in both ORs that discriminate between minty and caraway odor. Comparison with orthologs from 36 mammalian species demonstrated a hominid-specific carvone binding pocket with about 100% conservation. Moreover, we identified loss-of-function SNPs associated with the carvone binding pocket of OR1A1. Given carvone enantiomer-specific receptor activation patterns including OR1A1, our data suggest OR1A1 as a candidate receptor for constituting a carvone enantioselective phenotype, which may help to explain mechanisms underlying a (R)-(-)-carvone-specific anosmia in humans.

Keywords: GPCR; Molecular modeling; Ortholog; Site-directed mutagenesis; Structure–function study.

Fig. 1

Enantiomer-selective and enantiomer-specific homolog carvone ORs. a Phylogenetic tree of human OR1A1 and homologs. The numbers represent the amino acid identities of respective homologs to human OR1A1. Concentration–response relations of (R)-(−)-carvone and (S)-(+)-carvone for b OR1A1 (Homo sapiens), c OR1A2 (Homo sapiens), d PTOR1A1 (Pan troglodytes), e btOR1A1 (Bos taurus), and f Olfr43 (Mus musculus). Shown are mean ± SD of n = 3–5. Mock control was subtracted. Data were normalized to the OR1A1 maximum amplitude. The red dashed line indicates the normalization level. RLU relative luminescent unit. Subpanels b and d were taken with permission from [40]

Fig. 2

Impact of amino acid residues within a predicted putative binding site of human OR1A1 and its murine ortholog Olfr43 on (R)-(−)-carvone and (S)-(+)-carvone function. Concentration–response relations of (R)-(−)-carvone and (S)-(+)-carvone for a OR1A1-wt (wild type, H. sapiens) and b Olfr43-wt (wild type, M. musculus). c Schematic snake diagram with localization of mutated amino acid positions within OR1A1 and Olfr43. d, e Mutual exchange of amino acids at positions 108^3.36 and 152^4.53. f Quantification of amplitudes of OR1A1 mutant receptors. g, h Mutual exchange of amino acids at positions 108^3.36 and 205^5.46. i, j Mutual exchange of amino acids at positions 152^4.53 and 205^5.46. k Quantification of amplitudes of Olfr43 mutant receptors. l, m Mutual exchange of amino acids at positions 108^3.36, 152^4.53, and 205^5.46. n Replacement of the amino acids at positions 106^3.34, 108^3.36, 152^4.53, and 205^5.46 in Olfr43. Shown are mean ± SD of n = 3–5. Mock control was subtracted. Data were normalized to the maximum response of OR1A1-wt with (R)-(−)-carvone. The red dashed line indicates the normalization level. RLU relative luminescent unit. *Significant differences (p < 0.05) of the (R)-(−)-carvone (black asterisks) and (S)-(+)-carvone (grey asterisks) responses between wild type (wt) and mutant receptors. Concentration–response relations for single amino acid mutations in OR1A1 and Olfr43 are given in Supplemental Fig S5. For didactic reasons, we included the panels of OR1A1-wt and Olfr43-wt from Fig. 1b, f

Fig. 3

Molecular models of differing ligand binding cavities for human OR1A1 (grey, a, b) and murine Olfr43 (sand, c, d) located between transmembrane helices TMH3-7 and the interaction with (R)-(−)-carvone (orange) and (S)-(+)-carvone (cyan). Simplified schemes are shown as insets. Residues defining the binding pocket (calculated grey surface) are shown as sticks (oxygen: red; nitrogen: blue). The small side chain of Gly108^3.36 in OR1A1 provides at TMH3 an enlarged ligand binding pocket (grey surface) compared to Olfr43 where the slightly larger side chain of Ala108^3.36 decreased the binding pocket at TMH3. (R)-(−)-carvone (orange) and (S)-(+)-carvone (cyan) are visualized including their hydrogens. The small 2-methyl and bulky 5-(1-methyl-ethenyl) moiety of the carvones is oriented vice versa within the binding pocket for the two enantiomeric conformations. Hydrogen bonds are indicated by dashed lines in yellow. The essential intermolecular hydrogen bond between the 1-ketone oxygen (red) of (R)-(−)- and (S)-(+)-carvone to the NH₂ group of Asn109^3.37 at TMH3 can be established in OR1A1 by a (R)-(−)-carvone and b (S)-(+)-carvone, since the small Gly108^3.36 side chain provides enough space interacting with both enantiomers. c Restricted binding pocket (due to Ala108^3.36) in Olfr43 allows only interaction with the small 2-methyl group of (R)-(−)-carvone to form an H-bond to Asn109^3.37. d Binding of (S)-(+)-carvone is prevented in Olfr43 due to sterical clashes with pocket borders restricted by Ala108^3.36 and Val205^5.46 (hindrance, red oval). This is indicated by a manual docking pose of (S)-(+)-carvone (cyan), where the bulky 5-(1-methyl-ethenyl) and the 2-methyl moieties are protruding the calculated pocket surface (grey) and cannot properly establish the H-bond towards Asn109^3.37, which explains the diminished effect on Olfr43. Conserved/variant residues are colored in green/yellow

Fig. 4

Molecular models of the ligand binding cavity for the quadruple mutant Olfr43–G₁₀₆A/A₁₀₈G/A₁₅₂G/V₂₀₅I. Simplified schemes are shown as insets, the quadruple mutation changes the position of TMH3 (green circle) slightly compared to Olfr43-wt (dashed black circle). Residues defining the binding pocket (calculated grey surface) are shown as sticks (oxygen: red; nitrogen: blue). (R)-(−)-carvone (orange) and (S)-(+)-carvone (cyan) are visualized including their hydrogens. Compared to Olfr43-wt, the position of TMH3 is slightly altered in Olfr43–G₁₀₆A/A₁₀₈G/A₁₅₂G/V₂₀₅I, which in combination with the shortening of side chain A₁₀₈G enlarges the binding cavity. These changes allow a better binding of a (R)-(−)-carvone (orange) and b enable binding of (S)-(+)-carvone (cyan)

Fig. 5

Indirect stabilization of the OR1A1 binding site. a View onto the interaction between TMH3 and TMH4 indicates an interaction of Ala106^3.34 with Gly152^4.53 and Asn109^3.37 with Asn155^4.56. Changes in either Ala106^3.34 or Gly152^4.53 lead to steric effects between TMH3 and TMH4 and thereby influence the spatial positions of Gly108^3.36 and Asn109^3.37. The side chain of Asn109^3.37 is positioned through interaction with Asn155^4.56, and points into the binding pocket (Fig. 3). Decreased functional effects by the single mutations of b N₁₀₉D, c N₁₅₅S, d G₁₀₈A, e I₂₀₅V, f A₁₀₆G, and g G₁₅₂A support this model. Shown are mean ± SD of n = 3–5. Mock control was subtracted. Data were normalized to the maximum response of OR1A1-wt (wild type) with (R)-(−)-carvone. The red dashed line indicates the normalization level. RLU relative luminescent unit. For didactic reasons, we included the panel of OR1A1-wt (wild type) from Fig. 1b in the background (grey) of each subpanel

Fig. 6

Amino acids directly interacting with carvone within OR1A1 binding site. The OR1A1 binding site model for a (R)-(−)-carvone and b (S)-(+)-carvone shows the residues directly interacting with carvone: Ile205^5.46, Tyr251^6.44, and Tyr276^7.41. The single mutations of c Y₂₅₁F, d Y₂₅₁M, e Y₂₇₆F, and f Y₂₇₆L support this model. g View onto the interactions between TMH3, TMH6, and TMH7, which indicate an interaction of Tyr250^6.43 with Asp111^3.39. The predicted H-bond interaction is directing the side chain orientation of Tyr250^6.43, which defines the binding site flanking (R)-(−)-carvone (orange) between TMH3 and TMH6. Decreased functional effects by single mutations at both sites h Asp111^3.39 and i Tyr250^6.43 support this model. Shown are mean ± SD of n = 3–5. Mock control was subtracted. Data were normalized to the maximum response of OR1A1-wt (wild type) with (R)-(−)-carvone. The red dashed line indicates the normalization level. RLU relative luminescent unit. For didactic reasons, we included the panel of OR1A1-wt (wild type) from Fig. 1b in the background (grey) of each subpanel

Fig. 7

Impact of amino acid residues within a predicted putative binding site of human OR1A1 and its ortholog btOR1A1 on (R)-(−)-carvone and (S)-(+)-carvone function. a Schematic snake diagram with localization of mutated amino acid positions within human and bovine OR1A1. Concentration–response relations of (R)-(−)-carvone and (S)-(+)-carvone for b OR1A1-wt (wild type, H. sapiens), and c btOR1A1-wt (wild type, B. taurus) with btOR1A-L₁₀₅I/I₂₀₅V/I₂₇₇T (red curves). d Quantification of amplitudes of btOR1A1 mutant receptors. e Replacement of the amino acid at position 277^7.42 from threonine (Thr) to methionine (Met) in human OR1A1 (SNP, single-nucleotide polymorphism). f Replacement of the amino acid at position 277^7.42 from isoleucine (Ile) to threonine (Thr) in btOR1A1. This mutant corresponds to human OR1A1. Shown are mean ± SD of n = 3–5. Mock control was subtracted. Data were normalized to the maximum response of human OR1A1-wt (wild type) with (R)-(−)-carvone. The red dashed line indicates the normalization level. RLU relative luminescent unit. *Significant differences (p < 0.05) of the (R)-(−)-carvone (black asterisks) and (S)-(+)-carvone (grey asterisks) responses between wild-type (wt) and mutant receptors. Concentration response relations for further amino acid mutations in btOR1A1 are given in Fig S6. For didactic reasons, we included the panels of OR1A1-wt (wild type) and btOR1A1-wt (wild type) from Fig. 1b, e

Fig. 8

Carvone-binding residues are conserved among OR1A1 orthologs. a Snake diagram of human OR1A1 with putative odorant binding site by [20] (green filled circles) and carvone binding site (red circles). Overlapping amino acid residues are shown as green filled red circles. Carvone binding site residues are Ala106^3.34, Gly108^3.36, Asn109^3.37, Asp111^3.39, Gly152^4.53, Asn155^4.56, Ile205^5.46, Tyr250^6.43, Tyr251^6.44, and Tyr276^7.41. The numbers refer to human OR1A1 amino acid sequence. b Alignments of transmembrane helices (TMH1-7) and extracellular loop 2 (ECL2) of human OR1A1, and its 37 orthologs. Shown are sequence logos, the consensus sequence, and the human OR1A1 sequence with the marked putative odorant binding site by Ref. [20] (green letters) and carvone binding site (red boxes). Overlapping residues are shown as green letters in red boxes. The consensus amino acid refers to the most frequent one, which is determined by letter height and stacking order. The letters of each stack are ordered from the most frequent to the least frequent. Amino acid conservation is measured in bits, and a 100% conservation correlates to 4.32 bits [49]. Basic amino acids (K, R, H) are blue, polar (G, S, T, Y C) are green, hydrophilic (Q, N) are purple, acidic (D, E) are red, and hydrophobic (A, V, L, I, P, W, M, F) are black. A complete phylogenetic tree of all reference sequences investigated, as well as their accession numbers are given in Fig S7 and Tab S1