Key amino acids alter activity and trafficking of a well-conserved olfactory receptor

Abstract

In this study, we elucidate factors that regulate the trafficking and activity of a well-conserved olfactory receptor (OR), olfactory receptor 558 (Olfr558), and its human ortholog olfactory receptor 51E1 (OR51E1). Results indicate that butyrate activates Olfr558/OR51E1 leading to the production of cAMP, and evokes Ca²⁺ influx. We also find olfactory G protein (Golf) increases cAMP production induced by Olfr558/OR51E1 activation but does not affect trafficking. Given the 93% sequence identity between OR51E1 and Olfr558, it is surprising to note that OR51E1 has significantly more surface expression yet similar total protein expression. We find that replacing the Olfr558 N-terminus with that of OR51E1 significantly increases trafficking; in contrast, there is no change in surface expression conferred by the OR51E1 TM2, TM3, or TM4 domains. A previous analysis of human OR51E1 single nucleotide polymorphisms (SNPs) identified an A156T mutant primarily found in South Asia as the most abundant (albeit still rare). We find that the OR51E1 A156T mutant has reduced surface expression and cAMP production without a change in total protein expression. In sum, this study of a well-conserved olfactory receptor identifies both protein regions and specific amino acid residues that play key roles in protein trafficking and also elucidates common effects of Golf on the regulation of both the human and murine OR.

INTRODUCTION

Olfactory receptors (ORs) are seven-transmembrane G protein-coupled receptors (GPCRs) primarily known for their role as chemosensors in the olfactory epithelium, where they mediate the sense of smell (1). In addition, ORs also play functional roles in nonolfactory tissues including lungs, kidneys, eyes, muscle, prostate, and sperm (2–9). In olfactory sensory neurons, ligand binding to ORs activates the olfactory G protein (Golf), which then stimulates adenylyl cyclase (AC3) to increase intracellular cAMP (10–13). The consequent production of cAMP leads to the opening of cyclic nucleotide-gated (CNG) channels and an influx of Na⁺ and Ca²⁺, ultimately leading to an action potential (14).

There are ∼1,000 murine ORs (15) and 400 human ORs (16), however, only three of these ORs have one-to-one orthology among placental mammals (17). In this study, we elucidate factors that regulate the trafficking and activity of one of these three well-conserved ORs: murine olfactory receptor 558 (Olfr558)/human olfactory receptor 51E1 (OR51E1). Olfr558 is an attractive and promising candidate for study for several reasons: first, Olfr558 has identified ligands (18, 19), and in a recent study we confirmed that the strongest ligand for both Olfr558 and OR51E1 is butyrate (20). Second, Olfr558 has been extraordinarily well-conserved by evolution: Olfr558 is one of only three murine ORs to have a clear ortholog in mice, rats, rabbits, elephants, horses, and five species of primates (17). Moreover, OR51E1 has one of the lowest rates of single nucleotide polymorphism (SNPs) among human ORs (21). Finally, OR51E1 is of particular interest because has the broadest tissue expression of all human ORs, with expression reported in 13 tissues (22). Intriguingly, despite the evolutionary conservation between Olfr558 and OR51E1, we identify differences in how these two receptors traffic in vitro, and in this study, we work to uncover what may govern these differences. In addition, we characterize the effects of Golf coexpression on Olfr558 and OR51E1. Finally, a recent publication noted that human OR51E1 has a single nucleotide polymorphism (SNP) which is most abundant, albeit still rare, in those from South Asia (23). In this study, we report for the first time that this mutant influences OR51E1 trafficking and function.

METHODS

Cloning

Olfr558 and OR51E1 have been previously cloned with Lucy-Flag-Rho tags at their N-terminus, in the pME18S vector (24). Surface trafficking of exogenously expressed ORs is often problematic, and we also previously reported that both Lucy- and Rho-tags promote surface expression of ORs (24). Thus, we used the Lucy-Flag-Rho-tag for all ORs. All Olfr558 mutants and OR51E1 A156T were made by overlap extension PCR using Olfr558 and OR51E1 as a template, respectively (25). All mutants were sequenced to confirm.

Real-Time cAMP Assay

Human embryonic kidney 293 T(HEK 293T) cells were grown in a poly-l-lysine-coated black 96-well plate overnight. Cells were transfected with various olfactory receptor (OR) constructs, as well as with receptor transporting protein 1S (RTP1S) and/or olfactory G protein (Golf), as indicated. After 4 h transfection, the cADDis cAMP sensor (BacMam baculovirus, Cat. No. U0205G, Montana Molecular) was transfected into the cells. Real-time cAMP fluorescence was measured 24 h after cADDis transfection at excitation 490 nm and emission 525 nm. Real-time cAMP production was measured every minute after stimulus for a total 10 min. The stimulus was presented throughout 10 min sampling period. Sodium butyrate (Cat. No. 303410, MilliporeSigma) was used as stimuli to increase cAMP production.

Immunofluorescence

HEK 293 T cells were grown on poly-l-lysine coated coverslips and transfected with Lucy-Flag-Rho tagged OR constructs and RTP1S. Cell surface staining was performed using live cells blocked with diluted superblock (1:4 dilution in PBS with 1 mM MgCl₂ and 0.1 mM CaCl₂, Thermo Fisher Scientific), and incubated with rabbit polyclonal anti-flag antibody (Cat. No. F7435, MilliporeSigma) at 1:100 in diluted superblock at 4°C for 1 h. Cells were washed, fixed with 4% paraformaldehyde (PFA), permeabilized with 0.3% Triton X-100, and blocked with diluted superblock for 1 h at room temperature. Total expression staining was done in fixed, permeabilized cells by staining with mouse monoclonal anti-flag antibody (M2, Cat. No. F1802, MilliporeSigma) at 1:100 in diluted superblock at four overnight. All surface and total staining cells were incubated with secondary antibody at 1:1,000 in diluted superblock with Hoechst 33342 (Cat. No. LSH3570, Invitrogen Molecular Probes) at 1:2,500 in the dark at room temperature for 1 h. Cells were then washed and mounted using VectaShield Hard Set mounting medium (H-1400, Vector Laboratories). Images were acquired using a confocal microscope (Zeiss LSM 700). Images were quantified manually by ImageJ by counting the number of positive cells and total cells. Percentage of positive-to-total cells and signal per cell were quantified by ImageJ software. For signal per cell: pictures were taken at ×20. For each image, the signal per cell for every positive cell within the field of view was quantified, and values for cells within the same field of view were averaged. Thus, each “n” for this measurement represents the average of values for a given field of view.

Ca²⁺ Imaging

Changes in intracellular calcium levels ([Ca²⁺]_i) in response to OR stimuli were measured using calcium imaging. HEK 293 T cells were seeded into poly-l-lysine coated chambers (Cat. No. 155382, Lab-Tek II) and grown overnight, then transfected with OR51E1 and RTP1S, with or without Golf for 24 h. Cells that were nontransfected or transfected with Golf alone were used as controls. Cells were loaded with 5 μM fura-2A (Cat. No. F0888, MilliporeSigma) and 10% Pluronic F127 (Cat. No. P6866, Invitrogen Thermo Fisher Scientific) dissolved in DMSO in HBSS (with 0.5 mM MgCl₂ and 0.5 mM CaCl₂) solution for 1 h in a 37°C incubator with 5% CO₂. The images were visualized with an inverted fluorescence microscope (Zeiss AXI0) and a charge-coupled device (CCD) camera (sCMOS, pco.edge). Cells were stimulated with 2 mM butyrate, and 100 μM ATP was used to stimulate the cells at the end of each experiment to confirm cell viability. Images were captured every 2 s with excitation wavelengths of 340 and 380 nm and an emission wavelength of 510 nm. Cell focusing, defining regions, and imaging acquisition were controlled by Metafluor software (Molecular Devices). One hundred micromolar EGTA was added to HBSS (with 0.5 mM MgCl₂ and 0.5 mM CaCl₂) solution to investigate Ca²⁺ release from endoplasmic reticular (ER) stores. Thapsigargin at 1 μM was used as a positive control at the end of each experiment to evoke robust Ca²⁺ release from the ER stores.

Flow Cytometry

We utilized flow cytometry to evaluate OR cell surface expression, as previously reported (26). HEK 293 T cells were grown in 35-mm cell-culture dishes overnight, then transfected with various ORs and RTP1S for 24 h. Cells were treated with Cellstripper (Cat. No. 25-056-Cl, Corning) and transferred into a round-bottom tube on ice. Cells were stained with rabbit polyclonal anti-flag antibody (M2, Cat. No. F1802, MilliporeSigma) at 1:100 in staining and washing solution (PBS containing 2% FBS and 15 mM NaN₃) for 1 h on ice. Cells were then washed and stained with 1:100 phycoerythrin (PE)-conjugated anti-rabbit IgG (Cat. No. 711-116-152, Jackson ImmunoResearch) for 30 min on ice. The intensity of the PE signal was measured and quantified by flow cytometry (Calibur, BD). Cells without antibody and without OR transfection staining were used as controls to determine the PE fluorescent range and nonspecific PE fluorescence, respectively.

Enzyme-Linked Immunosorbent Assay

OR total protein expression in HEK 293 T cells were measured by enzyme-linked immunosorbent assay (ELISA), as we previously described (24). Briefly, cells were seeded in a poly-l-lysine coated clear 96-well plate and transfected with OR. Cells were then fixed with 4% PFA and permeabilized after 24 h transfection. OR-expressing cells were incubated with mouse monoclonal anti-flag antibody (M2, Cat. No. F1802, MilliporeSigma) at 1:100 dilution, and detected with anti-mouse horseradish peroxidase (HRP)-conjugated secondary antibody (Cat. No. 115-035-146, Jackson) at 1:8,000 dilution. HRP levels were measured with 1-Step Ultra TMB (3,3′,5,5′-tetramethylbenzidine; Cat. No. 34028, Thermo Fisher Scientific) at 450 nm excitation wavelength.

Statistical Analysis

Data are presented as arithmetic means ± standard deviation (SD). Sample size (n) is indicated for each reported value. Statistical significance was determined by t test or ANOVA for multiple groups (P < 0.05).

RESULTS

Butyrate, a Ligand for Olfr558 and OR51E1, Evokes Calcium Responses via Ca²⁺ Influx

Using a luciferase-based cAMP reporter assay, we previously reported that butyrate is the strongest ligand for both Olfr558 and its human ortholog, OR51E1 (20). However, OR activation in the literature is measured by both cAMP and Ca²⁺ increases (26, 27), and canonical OR signaling elevates not only intracellular cAMP but subsequently, intracellular Ca²⁺. Thus, here we investigated real-time changes in both cAMP and Ca²⁺. We previously reported similar activation of Olfr558 to butyric acid and butyrate (20), and thus we only examined butyrate in these studies. Real-time cAMP assay data for both Olfr558 and OR51E1 indicated that dose-dependent responses to butyrate gradually increased during the first 5 min and then plateaued, whereas the trace for no butyrate addition control (0 μM) remained flat (Fig. 1, A and B). cAMP production at the 10 min time point demonstrated that butyrate dose-dependently and significantly activated Olfr558 (Fig. 1C) and OR51E1 (Fig. 1D).

Figure 1.

Butyrate, a ligand for Olfr558 and OR51E1, increases [cAMP]_i and evokes calcium responses via Ca²⁺ influx, consistent with canonical olfactory receptor signaling. A and B: real-time cAMP assays (HEK 293 T cells) demonstrate activation of exogenously expressed Olfr558 and OR51E1 by butyrate. Summary data for butyrate shows cAMP production at the 10 min time point in Olfr558 (C) or OR51E1 (D) transfected cells (n = 4 independent experiments with triplicates in each experiment; means ± SD). E: percentage of butyrate-responsive cells are shown for cells transfected with or without OR51E1 and Golf (Golf is the olfactory G protein, NT = nontransfected). Butyrate evoked Ca²⁺ responses are decreased in calcium-free buffer (W/O Ca²⁺). n = 5–8. F: Ca²⁺ responses are elicited by butyrate (2 mM) in HEK 293 T cells expressing OR51E1 and Golf. ATP (100 μM) is added at the end of each experiment as a positive control. Black trace represents an average of all cells in the field (gray traces). G: the addition of thapsigargin (1 μM) demonstrates that ER calcium stores are intact in calcium-free buffer. For C and D, *P < 0.05, **P < 0.01 control (no butyrate) vs. butyrate using one-way ANOVA. For A–D, data shown as means ± SD. ER, endoplasmic reticulum; Golf, olfactory G protein; HEK 293 T, human embryonic kidney 293 T; Olfr558, olfactory receptor 558; OR51E1, olfactory receptor 51E1.

For Ca²⁺ imaging, OR51E1 was transiently expressed in HEK 293 T cells with or without chimeric olfactory G protein (Golf). Butyrate at 2 mM evoked Ca²⁺ responses in 12.13% of the OR51E1 + Golf cells (160 of 1,319 cells) but failed to activate nontransfected cells or cells transfected with Golf alone (Fig. 1, E and F). Weak activation was seen when OR51E1 alone was transfected. We further investigated whether butyrate acts to invoke Ca²⁺ influx (consistent with the canonical pathway) or release from endoplasmic reticulum (ER) stores. We found that butyrate evokes Ca²⁺ responses only in the presence of extracellular calcium, indicating that butyrate elevates Ca²⁺ via Ca²⁺ influx (Fig. 1, E and G). In Fig. 1G, thapsigargin was utilized to ensure that ER calcium stores remain intact in calcium-free buffer.

Golf Increases the Amount of cAMP Induced by Olfr558 and OR51E1 Activation

The calcium data from Fig. 1 imply that Golf enhances Ca²⁺ influx of OR51E1 activation. Next, we assayed whether this is true for cAMP production. Real-time cAMP assays demonstrated that responses to butyrate dose-dependently and significantly increased with Golf for both Olfr558 (Supplemental Fig. S1, A and B; all Supplemental material is available at https://Doi.org/10.6084/m9.figshare.19604521) and OR51E1 (Supplemental Fig. S1, C and D). For summary data, cAMP production was normalized to its control at the 10 min time point (Fig. 2, A and B). To determine if the effect of Golf is specific to butyrate, we studied isovaleric acid (IVA), another Olfr558/OR51E1 ligand. We found that Golf also significantly increased Olfr558/OR51E1 activation by IVA (Fig. 2, C and D, Supplemental Fig. S1, E and F). We then investigated whether the Olfr558 EC₅₀ for butyrate is altered by Golf. Results indicated that the EC₅₀ for Olfr558 with Golf was 0.2 mM, which is significantly elevated to 0.7 mM in the absence of Golf (Olfr558 and empty vector-transfected cells; Fig. 2E).

Figure 2.

Golf increases the amount of cAMP produced in response to Olfr558 and OR51E1 activation. Real-time cAMP assays indicate that butyrate (A and B) and IVA (C and D) enhance cAMP production at the 10 min time point normalized to control (butyrate/IVA 0 mM). E: the EC₅₀ value of butyrate activation of Olfr558 without and with Golf. For A–D, **P < 0.01, ***P < 0.001 for OR vs. OR+Golf using two-way ANOVA. n = 3–5 independent experiments. Data shown as means ± SD. Golf, olfactory G protein; IVA, isovaleric acid; Olfr558, olfactory receptor 558; OR: olfactory receptor; OR51E1, olfactory receptor 51E1.

In addition, we assayed whether similar changes were seen with one of the other three murine ORs well-conserved by evolution, olfactory receptor 78 (Olfr78; 17), which is a sibling OR for Olfr558. We find that the response of Olfr78 to its ligands (acetate and propionate) significantly and dose-dependently increased when Golf was coexpressed (Supplemental Fig. S2, A–C), suggesting that this effect is not specific to Olfr558/OR51E1. To better understand the mechanism of Golf-increased OR function, we further evaluated whether Golf increased Olfr558/OR51E1 trafficking to the cell surface; however, cell surface immunostaining indicates that Golf did not alter cell surface expression of OR51E1 (Fig. 3, A–C) or Olfr78 (Supplemental Fig. S3, A–C).

Figure 3.

Golf does not alter cell surface expression of OR51E1. A: surface trafficking of OR51E1 or OR51E1+Golf are visualized by immunocytochemical staining. HEK 293 T cells transfected with OR51E1 or OR51E1+Golf are stained (unpermeabilized) with rabbit anti-flag antibody (surface, red in merge). Representative images are shown. Scale bar: 20 μm. B: quantification of OR51E1 cell surface protein expression as assayed by immunofluorescence demonstrates that Golf does not affect the ratio of cells with OR51E1 surface staining to total cells. C: quantification of signal per cell indicates that Golf does not alter surface staining of OR51E1 per cell. Each data point represents a single field of view (cells within the same field of view were averaged). ns: nonsignificant difference by t test. n = 3 independent experiments. Data shown as means ± SD. Golf, olfactory G protein; HEK 293 T, human embryonic kidney 293 T; OR51E1, olfactory receptor 51E1.

Olfr558 N Terminal Mutant Significantly Increases Cell Surface Expression

Olfr558/OR51E1 is exceptionally well-conserved by evolution, with 93% protein identity between the murine and human receptor. Given the high sequence similarity, we were surprised to note that OR51E1 routinely had better cell surface expression (Fig. 4, A–D). We investigated this by asking if specific regions of the protein were responsible. As GPCRs, Olfr558/OR51E1 have seven transmembrane domains (TM1-7). Olfr558 and OR51E1 have the same protein sequence in TM1, TM5, TM6, and TM7, whereas single or several amino acids are different in the N terminus, TM2, TM3, and TM4 (Fig. 4B). We made four Olfr558 mutants: N terminal (G3D, F4P, and S6G), TM2 (L71I and M84L), TM3 (V101L), and TM4 (A144T, M148V, and V156A) to determine if these regions were critical for cell surface trafficking. Cell surface immunofluorescence staining data indicated that Olfr558 N-terminal (Olfr558 but with the OR51E1 sequence in the N-terminus) exhibited significantly enhanced cell surface expression as compared with native Olfr558. Other mutants exhibited similar cell surface expression to Olfr558 (Fig. 4A). Quantification of the ratio of “cells with surface staining to total cells” indicates that OR51E1 has better ratio with 60.8 ± 2.3% versus Olfr558 with 35.1 ± 0.4% (Fig. 4C). Quantification of the amount of signal (surface staining) per cell also demonstrates that OR51E1 has more cell surface protein expression per cell compared with Olfr558 (Fig. 4D). However, the Olfr558 N-terminal mutant exhibited increased (50.1 ± 1.6%) cell surface expression. We then used flow cytometry to confirm changes in cell surface expression (Fig. 5, A and B). For OR51E1, 67.9 ± 1.8% of the cells exhibited cell surface expression, as compared with 46.2 ± 2.7% for Olfr558. Olfr558 N-terminal mutant increased cell surface expression to 60.1 ± 2.2%. Olfr558 TM2 (47.6 ± 2.4%) and TM3 (51.2 ± 2.4%) showed similar cell surface expression to Olfr558, whereas TM4 exhibited a slight decrease in cell surface expression (Fig. 5B). These data suggest that the Olfr558/OR51E1 N-terminus plays a pivotal role in trafficking to the cell surface, or, in increasing total protein expression.

Figure 4.

Olfr558 N terminal mutant significantly increases cell surface expression. A: cell surface staining of Olfr558, OR51E1, and Olfr558 mutants is visualized by immunocytochemical staining in unpermeabilized cells (surface). Representative images are shown. Scale bar: 20 μm. B: Olfr558 and OR51E1 protein sequence alignment showing 93% identity. Protein sequences are the same in transmembrane domain TM1, TM5, TM6, and TM7 (shown in blue), whereas single or several amino acids are different in N terminal, TM2, TM3, and TM4 (shown in red). C: quantification of surface staining shows that OR51E1 has more cell surface expression than Olfr558 and that the N-terminus of OR51E1 improves the ratio of positive to total cells. Quantification of signal per cell is shown in D (each data point represents the average signal/cell for a given field of view). Each data point represents a single field of view. The percentage of total cells which have positive staining is shown. Olfr558 N has significantly increased cell surface expression of Olfr558, but is still lower than OR51E1. ***P < 0.001, ##P < 0.01, ns, nonsignificant difference by one-way ANOVA. n = 3–6 independent experiments. Data are shown as means ± SD. Olfr558, olfactory receptor 558; OR51E1, olfactory receptor 51E1.

Figure 5.

Olfr558 N terminal mutant significantly increases cell surface expression by flow cytometry. A: surface trafficking of Olfr558, OR51E1, and Olfr558 mutants are analyzed by flow cytometry. Representative images of cell-surface flow-cytometry are shown. The intensity of anti-flag phycoerythrin (PE) signal is measured and plotted. Nontransfected (NT) and Olfr558 transfected cells without anti-flag staining are used as control. SSC-H: side scatter histogram; FL2-H: anti-flag-PE. Q1 and Q4: anti-flag negative cells, Q2 and Q3: anti-flag positive cells. B: quantification of cells with surface expression. OR51E1 exhibits more cell surface expression than Olfr558. Olfr558 N significantly increases cell surface expression of Olfr558, but is still lower than OR51E1. **P < 0.01, ***P < 0.001, #P < 0.05 by one-way ANOVA. n = 3–5 independent experiments. Data are shown as means ± SD. Olfr558, olfactory receptor 558; OR51E1, olfactory receptor 51E1.

To distinguish between these two possibilities, we further investigated the total protein expression of Olfr558 mutants by immunofluorescence staining (Fig. 6A). Results demonstrated that cells transfected with Olfr558, OR51E1, and Olfr558 mutants all had a similar percentage of cells expressing each receptor with 62.5 ± 0.46% (Olfr558), 63.1 ± 2.01% (OR51E1), 58.6 ± 1.00% (Olfr558 N terminal), 63.8 ± 1.08% (Olfr558 TM2), 59.6 ± 1.16% (Olfr558 TM3), and 59.6 ± 1.07% (Olfr558 TM4; Fig. 6B). Quantification of the amount of signal per cell indicated that total protein expression per cell also did not differ among Olfr558, OR51E1, and Olfr558 mutants (Fig. 6C). To quantify OR total protein expression, we also performed an ELISA assay and found that Olfr558, OR51E1, and Olfr558 mutants exhibited similar total protein expression (Supplemental Fig. S4A).

Figure 6.

Olfr558, OR51E1, and Olfr558 mutants show similar total protein expression in HEK 293 T cells. A: representative images of total protein staining for Olfr558, OR51E1, and Olfr558 mutants. Total protein expression is visualized by immunocytochemical staining. HEK293T cells, transfected with an OR of interest, are stained by mouse anti-flag antibody after permeabilization (green in merge). Scale bar: 20 μm. B: quantification of positive to total cells. Percentage of positive-to-total cells are shown with SD. C: quantification of signal per cell. Each data point represents a single field of view. n = 3 independent experiments. Data shown as means ± SD. All comparisons are N.S. (not significant) by one-way ANOVA. HEK 293 T, human embryonic kidney 293 T; Olfr558, olfactory receptor 558; OR: olfactory receptor; OR51E1, olfactory receptor 51E1.

OR51E1 A156T Exhibits Less Cell Surface Expression

A previous analysis of human OR51E1 SNPs identified an A156T mutant, most prevalent in South Asia, as being the most abundant OR51E1 SNP (albeit still rare; 23). We made this OR51E1 A156T mutant and evaluated the total protein expression and cell surface trafficking when exogenously expressed in HEK 293 T cells. The A156T mutant did not affect the percentage of cells expressing the receptor (59.1 ± 0.80% vs. 58.2 ± 0.76%; Fig. 7, A and B), or, total protein expression (Fig. 7, A and C; Supplemental Fig. S4B). However, the OR51E1 A156T mutant did lower the percentage of cells with cell surface staining for the receptor, with 32.9 ± 0.75% compared with 56.4 ± 0.47% for OR51E1 (Fig. 7, D and E). Similarly, the A156T mutant has a lower amount of cell surface receptor signal per cell (Fig. 7, D and F). These results imply that this SNP may have functional consequences for OR51E1 function.

Figure 7.

OR51E1 A156T shows less cell surface expression. Representative images and quantification of total protein (A–C) and cell surface (D–F) expression are shown. HEK293T cells, transfected with OR51E1 or OR51E1 A156T, are stained by mouse anti-flag (permeabilized total staining, green in merge) or rabbit anti-flag antibody (unpermeabilized surface staining, red in merge). Scale bar: 20 μm. Quantification of ORs total (B) and surface (E) protein expression is shown with the % of positive/total cells per field of view plotted. Quantification of signal per cell for OR total (C) and surface expression (F), is shown where each data point represents the average signal per cell for a given field of view. ns: nonsignificant difference; **P < 0.01, ***P < 0.001 using t test. n = 3 independent experiments. Data are shown as means ± SD. HEK 293 T, human embryonic kidney 293 T; Olfr558, olfactory receptor 558; OR: olfactory receptor; OR51E1, olfactory receptor 51E1.

Butyrate Activity is Reduced in OR51E1 A156T Transfected Cells

Finally, we evaluated butyrate activation of the OR51E1 A156T mutant. Consistent with the decreased surface expression of A156T, real-time cAMP data demonstrated that butyrate-induced cAMP production was significantly and dose-dependently attenuated in OR51E1 A156T, with or without Golf cotransfection (Fig. 8, A–E). We also found that Ca²⁺ influx induced by butyrate was significantly reduced in OR51E1 A156T and Golf cotransfected cells with 12.1% (97 of 801 cells) compared with 4.4% (30 of 685 cells; Fig. 8F).

Figure 8.

cAMP production is significantly reduced in butyrate activated OR51E1 A156T mutant compared with OR51E1 transfected cells. A–D: real-time cAMP assay indicates that butyrate induced cAMP production is significantly and dose-dependently attenuated in OR51E1 A156T HEK 293 T cells with and without Golf. E: cAMP production is normalized to its DPBS (Dulbecco’s PBS) control at 10 min time point (butyrate was dissolved in DPBS). *P < 0.05, **P < 0.01, ***P < 0.001 OR51E1 vs. OR51E1 A156T, OR51E1+Golf vs. OR51E1 A156T+Golf using two-way ANOVA. ns, nonsignificant difference. F: graph shows the percentage of butyrate (2 mM) induced Ca²⁺ responsive cells in OR51E1+Golf or OR51E1 A156T+ Golf transfected cells. n = 3 independent experiments. Data are shown as means ± SD. Golf, olfactory G protein; HEK 293 T, human embryonic kidney 293 T; OR51E1, olfactory receptor 51E1.

DISCUSSION

In canonical OR transduction, ligand binding stimulates Golf to increase cAMP production by adenylyl cyclase, leading to Ca²⁺ influx and evoking an action potential (14). We previously used a cAMP response element (CRE) that drives a firefly luciferase reporter gene to determine that butyrate, one of the major short-chain fatty acids (SCFAs), is the best ligand for the well-conserved OR Olfr558/OR51E1 (20). In this study, we find that butyrate increases cAMP production and evokes calcium responses via Ca²⁺ influx in Olfr558/OR51E1 transfected HEK 293 T cells and that Golf enhances OR signaling in vitro. Of note, however, the percentage of cells responding with calcium changes was lower than one would have predicted from the transfection efficiency (Fig. 6B). In addition, we find that the N-terminus of OR51E1 increases surface trafficking and that a human SNP influences surface trafficking and signaling of OR51E1. We find that the EC₅₀ for butyrate is in the millimolar range. Although butyrate is unlikely to be present in circulating blood at millimolar levels, in the colonic lumen butyrate levels are ∼26 mM (28), and the human ortholog of Olfr558 (OR51E1) is expressed in the colon (22). In contrast, within the circulating blood baseline butyrate is much lower (∼10–40 μM). However, plasma butyrate levels can vary quite widely with changes in diet, and thus it is possible that plasma levels reach the threshold of detection for butyrate.

In HEK 293 T cells, we presume that ORs couple to a native Gs protein. When Golf and Olfr558/OR51E1 are cotransfected, Golf increases the amount of cAMP generated in response to Olfr558/OR51E1 activation. We saw a similar effect with a different ligand (IVA) and with a different OR (Olfr78) but found that Golf does not affect the cell surface expression. Given that ORs transduce signals through Golf in vivo (13), this is not surprising (29). We suspect that this finding is likely generally true for most/all ORs.

Of the nearly 1,000 murine ORs, Olfr558 is one of the only three which are well-conserved across placental mammals (17), and we previously published that Olfr558 and its human ortholog OR51E1 have similar ligand profiles (20). Olfr558 and OR51E1 share 93% protein sequence identity and given the high sequence similarity, we were surprised to note that OR51E1 traffics to the cell surface better than Olfr558. We found that the OR51E1 N-terminus is sufficient to improve cell surface trafficking of Olfr558. Other regions of the protein may also be important for trafficking given that the Olfr558 N terminal mutant still has less surface expression than OR51E1. In future studies, it would be informative to assay the amount of cAMP generated in response to activation of Olfr558 mutants; similarly, it would be useful to examine these mutants in the presence versus absence of the Lucy-Flag-Rho tags.

Humans have around 400 ORs, which are divided into two main classes, 18 families and more than 150 subfamilies (16). A study examining SNPs in human ORs found that OR51E1 has one of the lowest rates of SNPs among the 378 functional ORs (21) implying that this OR may have strong evolutionary pressure to conserve its sequence. We find that a SNP seen in the coding region of OR51E1 influences OR signaling. OR51E1 A156T exhibits significantly decreased cell surface trafficking without affecting the total protein expression, as well as decreased butyrate response. We hypothesize that the decreased response to butyrate is a result of decreased cell surface trafficking. Finally, OR51E1 is of particular interest because expression profiling of ectopic OR expression in humans found that OR51E1 is the most broadly expressed OR with expression in 13 tissues including kidney, heart, adipose, breast, skeletal muscle, ovary, and prostate (22). This suggests that OR51E1 may play a pivotal role in nonolfactory tissues as well.

In conclusion, we have elucidated factors that modulate the cell surface expression and activity of a well-conserved olfactory receptor, murine Olfr558/human OR51E1. We find that Golf acts to increase cAMP production induced by OR activation independent of trafficking, and, that both protein regions and specific amino acid residues play a pivotal role in protein trafficking.

SUPPLEMENTAL DATA

Supplemental Figs. S1–S4: https://doi.org/10.6084/m9.figshare.19604521.

GRANTS

This work was supported by National Institutes of Health Grant R56DK107726 (to J.L. Pluznick).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.X. and J.L.P. conceived and designed research; J.X. performed experiments; J.X. analyzed data; J.X. and J.L.P. interpreted results of experiments; J.X. prepared figures; J.X. and J.L.P. drafted manuscript; J.X. and J.L.P. edited and revised manuscript; J.X. and J.L.P. approved final version of manuscript.