Expression pattern and response profile of three human testicular odorant receptors. A, testis-specific expression of candidate sperm ORs (OR1D2, OR4D1, and OR7A5) is confirmed by quantitative TaqMan real-time PCR analysis of various human tissues (2–3 different samples/tissue). The dashed line indicates relative expression below noise level (cut-off threshold of ≤40). Conventional RT-PCR from human tissue samples (inset) substantiates testicular expression of OR1D2, OR4D1, and OR7A5. Two OR genes that had not been attributed ectopic expression before (OR5D16 and OR2M4) served as negative controls. B and C, characterization of OR7A5 and OR4D1. Receptor activation in response to focal odor application (50 μm; 5 s) is monitored by ratiometric Ca²⁺ signals in fura-2-loaded HEK293 cells (3 μm) transfected with OR-encoding expression vectors (pcDNA3). The original traces depict the integrated fluorescence ratio F₃₄₀/F₃₈₀ of representative odor-sensitive cells (dashed circles) as a function of time. Pseudocolor images illustrate the relative Ca²⁺ concentration at time points 1–4 (rainbow 256 color map; blue, low Ca²⁺; red, high Ca²⁺). OR7A5 is selectively activated by Myrac, whereas OR4D1 responds to PI-23472. ATP-induced activation of endogenous P2Y receptors (PLC-mediated Ca²⁺ release from internal storage organelles) controls for cell viability. Both ORs accommodate structurally similar ligands (active; red background color), whereas more profound changes in chemical structure are not tolerated (inactive; gray background color). Scale bars, F₃₄₀/F₃₈₀ = 0.1 (vertical); time = 30 s (horizontal).

Human sperm display distinct and agonist-specific Ca²⁺ responses. A, heterogeneous distribution of Ca²⁺ signal onsets in different subcellular compartments (as illustrated in F). The moment a Ca²⁺ response is detected in the proximal flagellum is set as a temporal reference (t = 0; dotted line), and both region- and agonist-dependent onset intervals are calculated (n = 23 (bourgeonal), 18 (Myrac), and 25 (PI-23472)). Along a virtual longitudinal sperm axis, Ca²⁺ signal onsets are significantly different between adjacent compartments (p < 0.01; note the following exceptions: (a) Myrac responses in the proximal flagellum, midpiece, and posterior head; (b) PI-23472 responses in the proximal and intermediate flagellum). In non-flagellar compartments, onset intervals are significantly shorter for Myrac responses as compared with PI-23472 and bourgeonal signals (p < 0.01). B, representative traces (F₃₄₀/F₃₈₀ versus time) of Ca²⁺ transients recorded from different subcellular compartments (regions of interest are color-coded) of a single fura-2-loaded sperm challenged with PI-23472 (50 μm; 10 s; dotted line). On expanded coordinates (dotted square region; left), the faster response onset in the proximal and intermediate flagellum becomes apparent. C, Ca²⁺ extrusion in different subcellular compartments is odor-independent. Time constants (τ) of odor-specific Ca²⁺ signal decays derived from monoexponential fits (red dashed line; inset). Base Ca²⁺ levels are reconstituted significantly faster in flagellar compartments than in the acrosomal and posterior head region (p < 0.01; n = 20 (bourgeonal), 12 (Myrac), and 35 (PI-23472)). D, merged Ca²⁺ traces recorded from the head of the same cell in response to consecutive 10-s applications of Myrac (blue), PI-23472 (black), and bourgeonal (red). On expanded coordinates (dotted square regions), both the faster onset of the Myrac-induced signal (left) and the characteristic differences in response amplitudes (right; PI-23472 > bourgeonal > Myrac) are readily identified. E, stimulus identity defines characteristic response parameters. In a two-dimensional plot (amplitude versus rise time), response profiles derived from different sperm compartments group in odor-specific clusters. Generally, Myrac (blue; n = 18) triggers slowly rising Ca²⁺ signals of small amplitudes, whereas bourgeonal responses (red; n = 21) rise fast without reaching peak Ca²⁺ levels as induced by PI-23472 (black; n = 34). F, representative traces (F₃₄₀/F₃₈₀ versus time) showing compartment-specific Ca²⁺ signals (color-coded) of a single spermatozoon in response to bourgeonal. The graph depicts the faster onset (see A), steeper slope (see E), and faster decay (see C) of flagellar responses as compared with Ca²⁺ signals in the acrosomal and posterior head region. Highlighted time points 1–5 (dotted lines) correspond to individual pseudocolor images (bottom) illustrating relative compartment-dependent changes in Ca²⁺ concentrations (rainbow 256 color map; blue, low Ca²⁺; red, high Ca²⁺). B, D, and F, original traces have been mathematically smoothed using a binomial algorithm.

Agonists of OR7A5 and OR4D1 boost sperm swimming speed, whereas OR1D2 signaling mediates both chemokinesis and chemotaxis. A and B, swimming speed (A) and accumulation (B) assays of human sperm in HTF and an ascending gradient of undecanal (u), respectively (controls) or ascending gradients of bourgeonal (b), PI-23472 (PI), and Myrac (m), respectively (10⁻⁷ m ± equimolar undecanal). OR agonists all significantly elevate sperm velocity relative to HTF and undecanal (p < 0.03, all comparisons). In contrast, sperm density inside microcapillaries is significantly increased by bourgeonal and Myrac gradients but not by PI-23472 gradients. Undecanal blocks effects mediated by bourgeonal, whereas behavioral responses to PI-23472 or Myrac are unaffected. C, sperm accumulation in microcapillary tubes presenting an ascending gradient, uniform concentrations, or a descending chemical gradient (Myrac, bourgeonal, PI-23472 at 10⁻⁷ m; HTF as the control). Significantly elevated densities relative to HTF (p < 0.05) are observed for both bourgeonal (ascending gradient) and Myrac (ascending gradient and uniform concentration). D, representative randomly chosen paths of sperm in ascending gradients of Myrac, PI-23472, or bourgeonal. Open circles correspond to the location of sperm heads in video images captured at 30 Hz; arrowheads indicate directions of travel for individual cells. Analyzed motility parameters (ω, θ, r, and p) are defined as described under “Experimental Procedures.” Error bars, S.E.

Agonists of OR1D2, OR7A5, and OR4D1 trigger Ca²⁺ transients in human sperm. A, representative Ca²⁺ signals recorded from the head regions of single fura-2-loaded cells. The integrated fluorescence ratio F₃₄₀/F₃₈₀ is shown as a function of time. Brief focal application of bourgeonal, Myrac, and PI-23472 (50 μm; 10 s) triggers transient increases in Ca²⁺ (n ≥ 41). As shown for recombinant ORs (supplemental Fig. S1), preincubation with equimolar undecanal (black horizontal bar) blocks bourgeonal responses, whereas Myrac and PI-23472 signaling is unaffected (n = 11). B, a given spermatozoon consecutively responds to ligands that selectively activate different ORs. Shown is a representative original Ca²⁺ recording (F₃₄₀/F₃₈₀ versus time; scaling as shown in A) of a single fura-2-loaded cell (depicted in C). Highlighted time points 1–4 correspond to individual pseudocolor images (right) that illustrate relative changes in Ca²⁺ concentrations (rainbow 256 color map; blue, low Ca²⁺; red, high Ca²⁺). D, bar chart illustrating the relative occurrence of distinct odor sensitivity profiles in individual spermatozoa consecutively challenged with bourgeonal (b), Myrac (m), and PI-23472 (PI) in random sequence (50 μm; 10 s; n = 311). E, odor specificity of human sperm Ca²⁺ signals as assessed by population response imaging in microtiter plates (n = 6). Helional, vertoprenal, and mustard oil essentially fail to activate sperm. Peak ratiometric Ca²⁺ signals (F₃₄₀/F₃₈₀) are normalized to bourgeonal-induced response amplitudes (each odor = 50 μm). F, normalized quasi-dose-response curves derived from single cell Ca²⁺ imaging experiments (n = 52 (bourgeonal), 181 (Myrac), and 309 (PI-23472)). Sigmoid curves were calculated from averaged data points using the Hill equation. Response thresholds are observed in the nanomolar range. Although Myrac and PI-23472 sensitivity are similar with respect to the overall percentage of responding cells, the dose-response curve for bourgeonal stimulation is significantly shifted to lower odor concentrations (p < 0.05).

Distinct agonist-dependent Ca²⁺ response phenotypes emerge during prolonged stimulus exposure. A, Ca²⁺ signaling during constant agonist exposure (≥10 min). Shown are representative original traces (F₃₄₀/F₃₈₀ versus time) in response to prolonged presence (horizontal bars) of bourgeonal (red), PI-23472 (black), and Myrac (blue), respectively. B, bar graph illustrating agonist-specific sperm response patterns during continuous stimulation (n = 209 (bourgeonal), 341 (Myrac), and 211 (PI-23472)). Bourgeonal and PI-23472 induce immediate, rapidly rising Ca²⁺ signals, whereas most Myrac-sensitive cells show a delayed, slowly developing response (see A). C and D, in a cross-adaptation paradigm, Myrac responses are affected by PI-23472 but not bourgeonal pre-exposure. Although Ca²⁺ signals triggered by prolonged Myrac and bourgeonal stimulation appear mutually independent (data not shown), prior activation by PI-23472 converts oscillatory Myrac signals to a single, transient response. C and E, representative original Ca²⁺ signals (F₃₄₀/F₃₈₀ versus time) mediated by consecutive prolonged application (horizontal bars) of PI-23472 (black) and Myrac (blue) (C) or PI-23472 (black) and progesterone (green; n = 120), respectively (E). F, representative original control recording (F₃₄₀/F₃₈₀ versus time). During constant superfusion with bath solution (control), a constant resting Ca²⁺ level is maintained (n = 30).

Stimulus-evoked changes in flagellar beating, escape laser power, and curvilinear velocity. A, scatter plot (escape power versus curvilinear velocity) of individual laser tweezer experiments in presence (black filled circles) or absence (gray filled diamonds) of bourgeonal. B, box plot illustrating the median values of curvilinear velocity and escape laser power of laser-trapped human sperm under control conditions (gray; n = 95 (bourgeonal), n = 84 (Myrac), and n = 125 (PI-23472)) or challenged with ascending gradients of bourgeonal, PI-23472, or Myrac (black; n = 89 (bourgeonal), n = 64 (Myrac), n = 99 (PI-23472). Notches in the box represent an estimate of the uncertainty about the median value. The box plot medians are different at the 5% significance level if their notches do not overlap. Error bars, S.E.