Optimization of Insect Odorant Receptor Trafficking and Functional Expression Via Transient Transfection in HEK293 Cells

Abstract

Insect odorant receptors (ORs) show a limited functional expression in various heterologous expression systems including insect and mammalian cells. This may be in part due to the absence of key components driving the release of these proteins from the endoplasmic reticulum and directing them to the plasma membrane. In order to mitigate this problem, we took advantage of small export signals within the human HCN1 and Rhodopsin that have been shown to promote protein release from the endoplasmic reticulum and the trafficking of post-Golgi vesicles, respectively. Moreover, we designed a new vector based on a bidirectional expression cassette to drive the functional expression of the insect odorant receptor coreceptor (Orco) and an odor-binding OR, simultaneously. We show that this new method can be used to reliably express insect ORs in HEK293 cells via transient transfection and that is highly suitable for downstream applications using automated and high-throughput imaging platforms.

Keywords: Drosophila melanogaster; HCN1; odorant receptors; rhodopsin.

Figure 1.

Optimization of OR trafficking to the plasma membrane in HEK293 cells. (a) The human HCN1 receptor sequence ¹⁰⁶VNKFSL¹¹¹ encoding a minimal endoplasmic reticulum release signal (“mER”, or “E”) and the human Rhodopsin sequence ³⁴⁴QVAPA³⁴⁸ encoding a minimal Rho tag (“mRho”, or “R”) were used to enhance the functional expression of insect ORs by promoting their exit from the endoplasmic reticulum and the transport of post-Golgi vesicles to the plasma membrane, respectively. (b) Schematic representation of the constructs tested. HEK293 cells were cotransfected by electroporation with 2 pcDNA3.1(-) plasmid constructs. The first, with the insertion of a human codon-optimized version of the D. melanogaster Or47a (hOr47a) tagged at the N-terminus with: a peptide composed of the mRho, mER and a V5 tag—to detect the receptor via immunochemistry if necessary—(R.E.hOr47a construct), or tagged only with the mER and V5 peptides (E.hOr47a) or the V5 tag alone (hOr47a). The second construct was constituted by a noncodon-optimized version of D. melanogaster Orco coreceptor. As controls, cells were cotransfected with the Orco coreceptor together with the empty vector backbone (EV) instead of the hOr47a construct, or with the empty vector alone. (c) Changes in intracellular calcium concentration over time (Δ[Ca²⁺]_i) in HEK293 cells transfected with the constructs shown in panel (b), following stimulations with 100 µL of 100 µM of the Or47a agonist pentyl acetate (PA) at 50 s and the synthetic Orco agonist VUAA1 at 350 s. Graphs represent mean ± SD. n = 5 for each panel. Each n = 1 represents the mean of all imaged cells coming from each independent electroporated cuvette (see Methods). (d) Intensity of calcium responses following a pentyl acetate or a VUAA1 stimulation in cells transfected with the constructs described in panel (b). Values were extracted 50 s after a stimulation with 100 µL of 100 µM pentyl acetate, and 20 s after a stimulation with 100 µL of 100 µM VUAA1. Mean ± SD values for pentyl acetate: EV, -0.07 ± 0.63; Orco, 0.11 ± 0.64; hOr47a + Orco, 0.35 ± 0.61; E.hOr47a + Orco, 4.44 ± 0.66; R.E.hOr47a + Orco, 6.89 ± 1.41. Statistical analysis for pentyl acetate: R.E.hOr47a + Orco versus E.hOr47a + Orco: t = 3.52, P = 0.014; R.E.hOr47a + Orco versus hOr47a + Orco: t = 9.52, P < 0.001; R.E.hOr47a + Orco versus Orco: t = 9.81, P < 0.001. Mean ± SD values for VUAA1: EV, −0.68 ± 0.24; Orco, 9.52 ± 4.11; hOr47a + Orco, 15.88 ± 5.29; E.hOr47a + Orco, 45.02 ± 7.02; R.E.hOr47a + Orco, 61.32 ± 13.29. Statistical analysis for VUAA1: R.E.hOr47a + Orco versus E.hOr47a + Orco: t = 2.42, P = 0.051; R.E.hOr47a + Orco versus hOr47a + Orco: t = 7.10, P = 0.0015; R.E.hOr47a + Orco versus Orco: t = 8.32, P = 0.0015. (e) Percentage of responding ROIs following a pentyl acetate or a VUAA1 stimulation. Mean ± SD values for pentyl acetate: EV, 0.18 ± 0.16; Orco, 0.64 ± 0.50; hOr47a + Orco, 2.90 ± 1.05; E.hOr47a + Orco, 21.61 ± 4.16; R.E.hOr47a + Orco, 30.51 ± 6.56. Statistical analysis for pentyl acetate: R.E.hOr47a + Orco versus E.hOr47a + Orco: t = 2.56, P = 0.039; R.E.hOr47a + Orco versus hOr47a + Orco: t = 9.29, P = 0.0015; R.E.hOr47a + Orco versus Orco: t = 10.16, P = 0.0015. Mean ± SD values for VUAA1: EV, 0.06 ± 0.14; Orco, 24.12 ± 9.10; hOr47a + Orco, 30.24 ± 7.21; E.hOr47a + Orco, 61.49± 5.95; R.E.hOr47a + Orco, 71.76 ± 3.29. Statistical analysis for VUAA1: R.E.hOr47a + Orco versus E.hOr47a + Orco: t = 3.38, P = 0.015; R.E.hOr47a + Orco versus hOr47a + Orco: t = 11.71, P < 0.001; R.E.hOr47a + Orco versus Orco: W = 25, P = 0.016 (Wilcoxon rank test). Unless otherwise stated, tests are unpaired 2-tail Welch’t t-test. P values corrected for multiple comparisons using Holm’s correction. Bar plots represent mean ± SD, n = 5 for each treatment. *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant.

Figure 2.

OR cotransfection via a bidirectional expression vector in HEK293 cells. (a,b) Examples of calcium imaging experiments with HEK293 cells cotransfected with the R.E.hOr47a + Orco constructs in 2 different pcDNA3.1(-) vectors (a) and with the pDmelOR-R.E.hOr47a housing both the hOr47a and a codon-optimized version of Orco (hOrco) bearing the β-globin/IgG chimeric intron (c.i.) (b). Panels represent—from left to right—a schematic representation of the constructs tested; transmission light image of the tested HEK293 cells; base level [Ca²⁺]_i of transfected HEK293 cells (0 s); [Ca²⁺]_i 25 s after application of 100 µL of 100 µM pentyl acetate (75 s) and 25 s after application of 100 µL of 100 µM VUAA1 (375 s). Scale bar = 50 µm. (c) Changes in intracellular calcium concentration over time (Δ[Ca²⁺]_i) in HEK293 cells transfected with the construct shown in panels (a-b), following stimulations with 100 µL of 100 µM of the Or47a agonist pentyl acetate (PA) at 50 s and the synthetic Orco agonist VUAA1 at 350 s. Graphs represent mean ± SD. n = 5 for R.E.hOr47a + Orco and n = 3 for pDmelOR-R.E.hOr47a. Each n = 1 represents the mean of all imaged cells coming from independently electroporated cuvettes (see Methods). (d) Intensity of calcium responses following a pentyl acetate or a VUAA1 stimulation in cells transfected with the constructs described in panels (a–b). Δ[Ca²⁺]_i values were extracted 50 s after a stimulation with 100 µL of 100 µM pentyl acetate, and 20 s after a stimulation with 100 µL of 100 µM VUAA1. Mean ± SD values for pentyl acetate: R.E.hOr47a + Orco, 6.89 ± 1.41; pDmelOR-R.E.hOr47a, 38.43 ± 6.04. Statistical analysis for pentyl acetate: pDmelOR-R.E.hOr47a versus R.E.hOr47a + Orco, t = 8.90, P = 0.010, Two-tailed Welch’s t test. Mean ± SD values for VUAA1: R.E.hOr47a + Orco, 61.32 ± 13.29; pDmelOR-R.E.hOr47a, 149.24 ± 20.12. Statistical analysis for VUAA1: pDmelOR-R.E.hOr47a versus R.E.hOr47a + Orco, W = 15, P = 0.036, Wilcoxon rank sum test. Graphs represent mean ± SD. n = 5 for R.E.hOr47a + Orco and n = 3 for pDmelOR-R.E.hOr47a. (e) Percentage of responding ROIs following a pentyl acetate or a VUAA1 stimulation. Mean ± SD values for pentyl acetate: R.E.hOr47a + Orco, 30.51 ± 6.56; pDmelOR-R.E.hOr47a, 76.01± 10.64. Statistical analysis for pentyl acetate: pDmelOR-R.E.hOr47a versus R.E.hOr47a + Orco: t = 6.68, P = 0.0073. Mean ± SD values for VUAA1: R.E.hOr47a + Orco, 71.76 ± 3.29; pDmelOR-R.E.hOr47a, 92.47± 4.03. Statistical analysis for VUAA1: pDmelOR-R.E.hOr47a versus R.E.hOr47a + Orco: t = 7.52, P = 0.0024. Two-tailed Welch’s t tests. Graphs represent mean ± SD. n = 5 for R.E.hOr47a + Orco and n = 3 for pDmelOR-R.E.hOr47a.

Figure 3.

Analysis of insect OR agonist sensitivity and specificity by means of an automated imaging platform. (a) Normalization of the odor response respect to the VUAA1 internal control. After calculating the ratio (R) between the emission light at 340 nm and 380 nm, base levels were subtracted from peak agonist responses to obtain the odor, e.g., pentyl acetate (PA), and the VUAA1 net responses (∆x and ∆y, respectively). The ratio ϱ (rho) between the odor and VUAA1 net responses expressed in percentage (ϱ = [∆x/∆y]*100) can be used to account for the cell-specific OR expression level due to the transient transfection protocol. (b,c) Example showing how normalization of the odor response to the VUAA1 internal control can reduce response variability. (b) Time series of R (340/380 nm) ratios for 2 regions of interests (ROI1 and ROI2) showing different response intensities to the same odor and VUAA1 stimulation. (c) Time series of normalized ϱ values for the odor responses of the same ROI1 and ROI2 as shown in panel (b). ϱ values reveal that both ROI1 and ROI2 show a very similar maximal response to the PA stimulation. (d,e) Dose–response curve for HEK293 cells transfected with pDmelOR-R.E.hOr47a and stimulated with pentyl acetate (d), or with pDmelOR-R.E.hOr56a and stimulated with geosmin (e). Data were fitted to a 3-parameter logistic function using the drc package in R (see Supplementary Code). Values of the data fit (mean ± st. error) for pDmelOR-R.E.hOr47a: slope, −2.89 ± 0.42; upper limit, 30.80 ± 1.28; EC₅₀ (log), −4.61 ± 0.06. EC₅₀ values expressed in Molar (estimate, 2.5% and 97.5% confidence intervals): 2.45·10^–5 (1.83 × 10^–5; 3.29 × 10^–5). Values of the data fit for pDmelOR-R.E.hOr56a: slope, −2.46 ± 0.58; upper limit, 11.93 ± 0.92; EC₅₀ (log), −6.14 ± 0.13. EC₅₀ values expressed in Molar (estimate, 2.5% and 97.5% confidence intervals): 7.19·10^–7 (3.97 × 10^–7; 1.30 × 10^–6). Plot shows estimate ± SE. For 5 ≤ n ≤ 6 (b) and n = 7 (c) for each odor concentration. (f) Odor response profile for Or47a using the pDmelOR-R.E.hOr47a construct. Odors are disposed according to their expected potency as agonists according to the DoOR database (see Supplementary Figure 2). Kruskall-Wallis rank sum test: χ ² = 50.148, df = 9, P < 0.001. Post-hoc Dunnett’s test for comparing each treatment versus DMSO. 3-octanol versus DMSO, P = 0.039; 2-heptanone versus DMSO, P < 0.001; isobutyl acetate versus DMSO, P = 0.013; hexyl acetate versus DMSO, P = 0.002; 3-methylthio-1-propanol (FG) versus DMSO, P = 1; propyl acetate (AG) versus DMSO, P = 0.036; methyl hexanoate versus DMSO, P < 0.001; butyl hexanoate versus DMSO, P < 0.001; pentyl acetate versus DMSO, P < 0.001. 4 ≤ n ≤ 10. The red box highlights 2 odors whose response intensities do not follow the expected trend, i.e., 3-methilthio-1-propanol and pentyl acetate. (g) The effect of propyl acetate is consistent across commercial stocks with different purity. The response to pentyl acetate was significant compared to the DMSO control when working solutions were prepared from a 99% purity stock odor (Sigma-Aldrich, Cat. Nr. 133108), and a ≥99% analytical standard grade (AG) stock odor (Fluka, Cat. Nr. 40858). Kruskall–Wallis rank sum test: χ ² = 10.088, df = 2, P = 0.0064. Post-hoc Dunnett’s test for comparing each treatment versus DMSO. 99% versus DMSO, P = 0.0147; ≥99% AG versus DMSO, P = 0.0147. 4 ≤ n ≤ 9. (h) The effect of 3-methilthio-1-propanol is not consistent across commercial stocks with different purity. Responses were significant compared to the DMSO control when working solutions were prepared from a 98% purity stock odor (Sigma-Aldrich, Cat. Nr. 318396), but not from a ≥98% food/pharmaceutical (FG) grade stock odor (Sigma-Aldrich, Cat. Nr. W341509). Kruskall–Wallis rank sum test: χ ² = 8.681, df = 2, P = 0.013. Post-hoc Dunnett’s test for comparing each treatment versus DMSO. 98% versus DMSO, P = 0.904; ≥98% (FG) versus DMSO: P < 0.001. 4 ≤ n ≤ 9. (i) Responses of HEK293 cells transfected with the pCMV-BI empty vector to odor stimuli (300 µM for all odors, except 30 µM for geosmin) and DMSO. Kruskall–Wallis rank sum test: χ ² = 27.115, df = 12, P = 0.0074. Post-hoc Dunnett’s test for comparing each treatment versus DMSO. 3-octanol versus DMSO, P = 0.00064; 2-heptanone versus DMSO, P = 0.734; isobutyl acetate versus DMSO, P = 0.996; hexyl acetate versus DMSO, P = 0.949; 3-methylthio-1-propanol (98%) versus DMSO, P = 0.762; 3-methylthio-1-propanol (FG) versus DMSO, P = 0.572; propyl acetate (99%) versus DMSO, P = 0.922; propyl acetate (AG) versus DMSO, P > 0.999; methyl hexanoate versus DMSO, P > 0.999; butyl acetate versus DMSO, P > 0.999; pentyl acetate versus DMSO, P = 0.961; geosmin versus DMSO, P = 0.749. For geosmin, one data point (black circle in the bar plot) was considered as an outlier and was omitted from analysis. 2 ≤ n ≤ 11. ***P < 0.001, **P < 0.01, *P < 0.05.