• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Jan 11, 2011; 108(2): 822–827.
Published online Dec 27, 2010. doi:  10.1073/pnas.1017983108
PMCID: PMC3021043

Unitary response of mouse olfactory receptor neurons


The sense of smell begins with odorant molecules binding to membrane receptors on the cilia of olfactory receptor neurons (ORNs), thereby activating a G protein, Golf, and the downstream effector enzyme, an adenylyl cyclase (ACIII). Recently, we have found in amphibian ORNs that an odorant-binding event has a low probability of activating sensory transduction at all; even when successful, the resulting unitary response apparently involves a single active Gαolf–ACIII molecular complex. This low amplification is in contrast to rod phototransduction in vision, the best-quantified G-protein signaling pathway, where each photoisomerized rhodopsin molecule is well known to produce substantial amplification by activating many G-protein, and hence effector-enzyme, molecules. We have now carried out similar experiments on mouse ORNs, which offer, additionally, the advantage of genetics. Indeed, we found the same low probability of transduction, based on the unitary olfactory response having a fairly constant amplitude and similar kinetics across different odorants and randomly encountered ORNs. Also, consistent with our picture, the unitary response of Gαolf+/− ORNs was similar to WT in amplitude, although their Gαolf-protein expression was only half of normal. Finally, from the action potential firing, we estimated that ≤19 odorant-binding events successfully triggering transduction in a WT mouse ORN will lead to signaling to the brain.

Keywords: mammal, olfaction, olfactory transduction

G protein-coupled receptor (GPCR) signaling is ubiquitous in cells and tissues. Textbook dogma about this signaling mechanism has been that an activated GPCR molecule goes on to activate many downstream G-protein molecules, each of which, in turn, typically activates an effector-enzyme molecule. This notion originates from the detailed understanding of visual transduction in retinal rod photoreceptors, in which each rhodopsin (rod pigment) molecule, once excited by light, remains active long enough to activate a substantial number of downstream GαT1 (rod transducin) molecules, each of which, in turn, activates a cGMP-phosphodiesterase molecule (effector enzyme) (1). Several years ago, however, we found that this dogma does not appear to apply to olfactory transduction, based on the observation that the unitary olfactory response is stereotypical in amplitude and kinetics across different odorants and randomly encountered olfactory receptor neurons (ORNs) that presumably expressed different odorant-receptor (OR) species (and indeed had different macroscopic sensitivities to a given odorant) (2). We concluded that each odorant-binding event has only a low probability of successfully triggering transduction to produce a unitary electrical response, apparently reflecting at least a very short lifetime of the odorant–OR complex attributable to rapid unbinding of the odorant (2). This previous work was on frog ORNs. It is important to know whether the same fundamental conclusion applies to mammals. Here, we report experiments on mouse ORNs.

The canonical mechanism of olfactory transduction is now quite well understood, at least in the main olfactory epithelium (37). An odorant binds to an OR (8) on an ORN cilium, which, via Golf, activates the olfactory adenylyl cyclase (ACIII) by the formation of an active Gαolf–ACIII complex that synthesizes cAMP. As a result, the intracellular free cAMP concentration increases, leading to the opening of cyclic-nucleotide-gated (CNG) nonselective cation channels to produce membrane depolarization of the cell. On reaching the firing threshold, the ORN gives rise to action potentials that propagate to the olfactory bulb in the brain. In addition, a Ca2+ influx through the open CNG channels during olfactory transduction amplifies the transduced signal by opening a Ca2+-activated Cl channel to generate an inward Cl current but also triggers adaptation via negative-feedback regulation on transduction involving multiple Ca2+-calmodulin-mediated pathways (37).

In this work, we also asked how many successful odorant-binding events (i.e., successful in triggering transduction, and thus in producing a unitary electrical response) in a mouse ORN are required for signaling to the brain. For an amphibian ORN, our previous work has shown that some low tens of such events are necessary (9). We estimate here that a comparable number is required for a mouse ORN. As in our amphibian work (9), we deal here only with excitation but not inhibition of ORNs by odorants.


Stimulus–Response Relation at Room Temperature.

As in previous work (2, 9), our approach consisted of first finding by trial and error an appropriate odorant concentration for stimulating a responsive cell and then varying the odorant strength by simply changing the odorant-pulse duration to generate the stimulus–response (S-R) relation. We kept this odorant duration within 80 ms to elicit so-called “impulse responses,” such that the odorant concentration and its duration are interchangeable without affecting the response (2, 9). In normal-Ca2+ solution, the overall foot of the S-R relation for mouse ORNs was supralinear (Fig. 1A, collected data from all cells in Fig. 1B), as previously found for amphibian ORNs (2, 9, 10). In Ca2+-free solution (no added Ca2+ plus 1 mM EGTA), the response to a given weak stimulus was considerably larger (Fig. 1C, collected data from all cells in Fig. 1D), presumably attributed to the removal of Ca2+-dependent adaptation, which dominates the intracellular effects of Ca2+ in ORNs (2, 9). More importantly, the foot of the S-R relation was clearly linear in Ca2+-free solution.

Fig. 1.
S-R relations from isolated mouse ORNs in normal-Ca2+ and Ca2+-free solutions. Room temperature. (A and C) Two different cells, with the response family (Upper) and S-R relation (Lower). 1-Heptanol was applied at a dose of 50 μM as a stimulus ...

The parsimonious interpretation of the linear foot of the S-R relation in Ca2+-free solution is that successful odorant-binding events (i.e., successful in triggering transduction) activated spatially restricted unitary transduction domains, such that when few events occur, the domains are spatially segregated and therefore noninteractive, thus producing an overall response given simply by the linear sum of the unitary responses (2, 9). By the same reasoning, we should expect the very initial foot of the S-R relation even in normal-Ca2+ solution to be linear (9). Because the unitary response is much smaller in such a solution, however, this linear segment can be difficult to resolve (2, 9). In some cases, a short linear foot of the stimulus–response relation was nonetheless detectable in mouse ORNs in normal-Ca2+ solution, being generally found in cells with larger receptor currents (see below and Fig. 3) presumably because of longer and/or more cilia.

Fig. 3.
Derivation of the unitary-response amplitude in normal-Ca2+ solution at room temperature based on the macroscopic-response ratio between normal-Ca2+ and Ca2+-free conditions. (A and B) Response families and S-R relations of the same cell to the same odorant ...

For amphibian ORNs, the S-R relation in low-Ca2+ solution beyond the linear segment is typically sublinear (2, 9). This is also often the case for mouse ORNs (Fig. 1C, Lower), although not invariably so (Fig. 1D). Currently, we do not have an explanation for this variability (9).

Unitary Response of Mouse ORNs in Ca2+-Free Solution at Room Temperature.

The relatively large responses in the linear segment of the S-R relation in Ca2+-free solution permitted the use of fluctuation analysis to extract the amplitude of the unitary olfactory response (2). The protocol consisted of delivering a long series of identical weak odorant pulses (usually ≥40 pulses, 20–50 ms in duration; 10–100 μM cineole or 1-heptanol) to a responsive ORN in the linear range of its S-R relation. A partial response series from an ORN to cineole is shown in Fig. 2A. The individual responses had similar time courses (Fig. 2A, Right) but fluctuating amplitudes (Fig. 2B), as evidenced by the good agreement in time course between their ensemble mean square, m2(t), and the stimulus-induced increase in ensemble variance, σ2(t) (Fig. 2C). From the Poisson distribution, the unitary amplitude is given by σ2(t)/m(t) at the response's transient peak, denoted simply as σ2/m and equal to 0.52 pA in this experiment. Altogether, four ORNs responsive to cineole gave a unitary response of 0.66 ± 0.17 pA (mean ± SD). Five other cells responsive to 1-heptanol gave a similar 0.77 ± 0.16 pA (Fig. 2D, Left). Thus, as in amphibian ORNs, the unitary-response amplitude of mammalian ORNs appeared rather constant across cells and for more than one odorant, even though each responsive cell (being randomly encountered) most likely expressed a different OR and had a different macroscopic sensitivity to the tested odorant. The waveform of the weak response was also generally similar across cells; the mildly different kinetics, especially of an occasional cell, may or may not be meaningful in that they may depend on a cell's condition (Fig. 2D, Right). In conclusion, neither the affinity between odorant and OR nor the efficacy of an odorant–OR complex appeared to affect the unitary-response characteristics, even though both parameters should, in principle, be unique for each distinct odorant–OR complex.

Fig. 2.
Fluctuation analysis to extract the unitary olfactory response in Ca2+-free solution at room temperature. A series of forty 50-μM, 25-ms cineole pulses were delivered to an ORN at time 0. (A) (Left) Sample responses. (Right) Superposition of the ...

With ~103 OR species distributed over mouse ORNs (38), the likelihood of encountering an ORN responsive to a given odorant was quite low. For example, cineole and 1-heptanol individually activated only ~5% and ~10%, respectively, of the randomly encountered ORNs (of hundreds of cells tested). The wide-ranging sensitivities of the responsive cells made the experiment even more challenging, because fluctuation analysis had to be carried out in the narrow linear range of the S-R relation (9). Because the probability that more than one odorant in a mixture of a few odorants would activate a given OR with overlapping linear ranges is low (i.e., when the most effective odorant in the mixture activated an ORN in the linear range, the other odorants most likely remained ineffective), however, there is some justification for using an odorant mixture for stimulation (1-heptanol, cineole, acetophenone, hexanal, and (+)citronellal). An odorant mixture substantially enhanced the probability of encountering responsive ORNs (to ~40% for this mixture). Nonetheless, the unitary response had the same amplitude (0.63 ± 0.14 pA, n = 13 cells) and roughly similar kinetics (Fig. 2E) as those triggered by cineole or 1-heptanol alone. Again, the slightly variable kinetics may not be meaningful (see above). We adopted the grand mean, 0.66 pA, for subsequent calculations.

Unitary Response in Normal-Ca2+ Solution at Room Temperature.

To estimate the unitary-response amplitude in normal-Ca2+ solution, we used a strategy previously adopted for amphibian ORNs (9); namely, we scaled the unitary amplitude according to the macroscopic-response ratio between normal-Ca2+ and Ca2+-free solutions for a given cell. This approach assumes that the number of odorant-binding events successfully triggering transduction at a given odorant strength is independent of external Ca2+ concentration. We have previously verified this point for amphibian ORNs (9). For eight mouse ORNs, despite the small response, an initial linear segment was detectable at the foot of the S-R relation even in normal-Ca2+ solution (e.g., Fig. 3A). In conjunction with the linear S-R relation for the same cells in Ca2+-free solution and in the same stimulus range (e.g., Fig. 3B), we obtained a ratio of 0.11 ± 0.06 (range: 0.02–0.22) between the slopes of the linear fits under the two Ca2+ conditions. Multiplying this ratio by the mean unitary-response amplitude of 0.66 pA in Ca2+-free solution (see above), we obtained ~0.073 pA in normal-Ca2+ solution. For another 14 cells, we did the same experiment, although there were insufficient data to provide certainty that the responses were all in the linear range. Nonetheless, these cells led to a very similar unitary-response size of 0.065 ± 0.021 pA. The above value of ~0.073 pA has to be multiplied by ~2 to account for the imperfect collection of transduction current by the suction pipette (Materials and Methods), giving a unitary response of ~0.15 pA in normal-Ca2+ at room temperature after correction.

Effect of Temperature.

The above experiments were all carried out at room temperature. At more physiological temperatures, the ORNs did not last long enough in Ca2+-free solution for fluctuation analysis. Accordingly, we simply measured an ORN's macroscopic response in normal-Ca2+ solution to a fixed weak stimulus at both room temperature and 35 °C (instead of 37 °C, because the cells tended to last longer at 35 °C). We used a stimulus that elicited a response ≤10% of maximum to minimize nonlinearities in the response. The response was larger at 35 °C than at room temperature, and its kinetics were faster (Fig. 4A). Because the response shape did not necessarily stay constant on changing the temperature, we used two parameters to describe the response speed: time to peak (tp, the time elapsed between the stimulus onset and the transient peak of the response) and integration time (tint), given by tint = ∫ r(t)dt/rp, where r(t) is the response function and rp is the response amplitude at transient peak (11). The larger the disparity between tp and tint, the more asymmetrical is the response; also, the longer the tint, the slower are the response kinetics. From 14 cells, rp was 1.9 ± 0.59-fold (range: 1.3–3.0-fold) and tp and tint were 0.69 ± 0.16-fold (range: 0.35–0.88-fold) and 0.71 ± 0.25-fold (range: 0.21–0.97-fold), respectively, as large at 35 °C as at 23 °C (Fig. 4B). Applying this temperature correction to the unitary-response amplitude of ~0.073 pA in normal-Ca2+ solution at room temperature in the previous section, we arrived at ~0.14 pA in normal-Ca2+ solution and at 35 °C. This value again has to be multiplied by ~2 to account for the imperfect collection of transduction current by the suction pipette (Materials and Methods), giving a unitary response of ~0.28 pA in normal-Ca2+ at 35 °C after correction.

Fig. 4.
Effect of temperature on olfactory response in normal-Ca2+ solution. (A) (Upper) Weak response (<10% of maximum) of an ORN to a fixed stimulus (10 μM 1-heptanol, 30 ms) at room temperature (23 °C) and 35 °C. (Lower) Same ...

Effect of Gαolf Level.

So far, our data indicate that, as in amphibians (2, 9), the mouse unitary olfactory-response properties are quite independent of the odorant–OR complex. Our interpretation is that the odorant–OR complex has a low probability of successfully signaling downstream, such that the unitary response, the fundamental stochastic unit in the macroscopic response, corresponds to the action of a single Gαolf–ACIII molecular complex (2, 9). The availability of mouse genetics allows us to check this point by examining the effect of reducing the Gαolf protein level. If each unitary response resulted from the action of many Gαolf–ACIII complexes all triggered by the same odorant–OR complex [i.e., analogous to the many GαT1/phosphodiesterase complexes triggered by one photoactivated rhodopsin molecule in rod phototransduction (1)], a decrease in the Gαolf protein level should then reduce the number of such Gαolf–ACIII complexes; hence, the unitary-response amplitude. Indeed, in rod photoreceptors, the single-photon response was smaller when the rod-transducin protein level was lowered by 20% in the T1+/− mouse (12). In contrast, if an odorant–OR complex has a low probability of successfully activating even one Gαolf–ACIII complex, lowering the Gαolf protein level should only reduce the probability of success further but not affect the unitary-response amplitude because the latter still reflects the action of one Gαolf–ACIII complex.

We used adult olf+/− mice (13) and verified by Western blotting that their olfactory epithelium expresses half of the WT level of Gαolf protein (14) (Fig. 5A). To ensure that this genotype does not affect the expression levels of the other transduction proteins, we also examined ACIII (as an example) and confirmed its similar level in both genotypes (Fig. 5A). With fluctuation analysis, we found that the unitary response of olf+/− ORNs at room temperature was indeed similar to that of WT in amplitude, instead of being smaller (Fig. 5B, Left). The response kinetics were likewise similar in both (Fig. 5B, Right), although this parameter would not necessarily change even if the amplitude did. In conclusion, each odorant-binding event does appear to activate at most one Gαolf–ACIII complex (and, most of the time, none at all). In other words, there is essentially no amplification in olfactory transduction up to the step of effector-enzyme activation. Incidentally, adult olf−/− mice are anosmic, indicating that there is no developmental compensation for the loss of Golf in ORNs (13).

Fig. 5.
Similar unitary olfactory responses from WT and olf+/− ORNs. (A) Western blot analysis on olfactory-epithelium homogenates to indicate that, in olf+/− ORNs, the Gαolf protein level is one-half of that in WT ...

Olfactory Threshold for Signaling to Brain.

How many unitary transduction events triggered by a brief odorant pulse in an ORN are required for the cell to signal to the brain via action potentials? We recorded the responses of an ORN to a series of weak identical odorant pulses in normal-Ca2+ solution at 35 °C, using low-pass filtering to measure the transduction current (Fig. 6A) and band-pass filtering to record action potentials (Fig. 6B). The procedure was repeated at several stimulus strengths in random order. The firing frequency was measured in 100-ms intervals, and a peristimulus time histogram was plotted (Fig. 6C). We define the threshold for signaling to the brain conservatively as the minimal transduction current that elevated the firing frequency (averaged over 0.5 s after the stimulus, roughly corresponding to the duration of the receptor current at 35 °C) above the mean basal rate by at least 2 SDs of the basal rate. For the experiment in Fig. 6 AC (see legend), this change in firing rate was produced by a receptor current of ≤4 pA (with the “≤” sign reflecting the fact that the number of stimulus strengths was too few to identify exactly the threshold as defined above) (Fig. 6C, second panel from top). To convert this current into unitary events, we extrapolated the initial linear foot of the S-R relation for this cell to the odorant stimulus that gave a 4-pA response and read off the response value if the relation were linear, giving ≤1.9 pA (Fig. 6D). Dividing this 1.9 pA by the unitary response of 0.14 pA (both values being uncorrected for imperfect current collection) under the recording conditions gave ≤14 events. From nine cells, the receptor-current threshold was ≤2.9–4.9 pA [mean ± SD = 3.7 ± 1.0 pA, which, as in Fig. 6C above, was typically already beyond the initial linear S-R segment (mostly confined to <2 pA)]. By the same analysis as in Fig. 6D, we obtained a threshold of ≤19 ± 5 unitary events averaged over all cells. The spontaneous firing rate of these cells was 0.1–2.0 Hz, with no apparent correlation to the olfactory threshold.

Fig. 6.
Threshold for ORN signaling to brain in normal-Ca2+ solution at 35 °C. (A) Superposed individual responses (black) to 30 stimulus trials of the same weak odorant pulse (10 μM 1-heptanol for 50 ms) delivered at time 0 (arrow) as well as ...

Just for comparison with our previous measurements on amphibian ORNs, we also carried out the experiment at room temperature [see example in Fig. S1, giving a threshold receptor current of ≤2.4 ± 0.69 pA (n = 5 cells)], corresponding to ≤24 ± 8.4 unitary events from a similar analysis as above.


As in amphibian ORNs, an odorant-binding event in a mammalian ORN appears to have a low probability of successfully triggering olfactory transduction. The experiment on mouse olf+/− ORNs is consistent with this notion. This low probability of transduction may come from a low probability of an odorant–OR complex to activate a Golf molecule, a low probability of an activated Gαolf to activate an ACIII molecule, or both. Based on our previously published amphibian data, the simplest explanation would be that the odorant–OR complex has a short lifetime because of the short dwell-time of the odorant on the OR (2). There is currently no information about the densities of Golf and ACIII on the ORN ciliary membrane or any cytoarchitecture linking the various transduction components in the cilium, unlike the detailed quantitative information available about rod phototransduction. Until such information is available, it is impossible to dissect this low probability further. It is also presently unclear how prevalent this low probability of transduction is among various odorants and ORs, but it will not be too surprising if this turns out to be the norm rather than the exception. Possibly, this notion of low probability can also be generalized to many other ligand-triggered GPCR signaling pathways. In other words, a high amplification may be fairly unique to rod phototransduction, perhaps to compensate for the physical disappearance of a photon after absorption and also the permanent inactivation of a pigment molecule after bleaching (2). One subtle but important question emerging from the above picture is that if the unitary response indeed represents the action of a single Gαolf–ACIII molecular complex, how does it manage to have a constant amplitude and time course, features that we found (2) to be especially pronounced in amphibian ORNs? Single-molecule lifetime is stochastic. In rod phototransduction, the constancy of the amplitude and time course of the single-photon response apparently results from averaging over the multiple phosphorylations on a single active rhodopsin molecule required for initiating its inactivation and also from averaging over many GT1 molecules activated downstream of an active rhodopsin molecule (1517). No regulatory mechanisms are currently known to eliminate the stochasticity of the lifetime of a single Gαolf–ACIII complex, but they may exist.

The unitary olfactory response of the mouse ORN in normal-Ca2+ solution at room temperature is ~0.073 pA × 2 = 0.15 pA (after correction for imperfect current collection), which is larger but of the same order of magnitude as the 0.034 pA that we previously estimated for frog ORNs (9). The ORN response kinetics are also similar for the two species, both with tp ≤ ~0.5 s and tint ≤ ~1 s at room temperature. Thus, overall, the amplification and dynamics of olfactory transduction are comparable between amphibians and mammals. At the more physiological 35 °C for mammals, the mouse unitary response is ~0.28 pA and is faster, with tp ≤ ~0.25 s and tint ≤ ~0.5 s. The unitary-response ratio between low-Ca2+ and normal-Ca2+ conditions appears to be larger for frog ORNs (45-fold) (2, 9) than for mouse ORNs (10-fold). The actual difference may be even larger because, for the low-Ca2+ condition, we used 100 nM free-Ca2+ for frog ORNs (2, 9), but literally Ca2+-free condition for mouse ORNs (0Ca2+-EGTA; Materials and Methods). In the normal-Ca2+ condition, the Ca2+ influx through the CNG channels during olfactory transduction tends to enhance the receptor current by activating an inward Cl current but also to reduce the receptor current by adaptation via negative feedback. These effects disappear partially or completely in the low-Ca2+ condition. Thus, the difference in ratio may result from the Cl current being larger in mammalian than amphibian ORNs. There is indeed suggestion for this from experiments in the literature (1821), but the measurements [except for those by Lowe and Gold (19)] came mostly from large responses and may not apply to the small responses that we deal with here. A different explanation is also possible, such as a genuine quantitative difference between the mouse and frog in the effect of removing external Ca2+ on the Cl current and adaptation.

How many CNG channels open in a unitary response? For frog ORNs, the single-CNG-channel current at −50 mV (near the resting membrane potential) with normal external concentrations of divalent cations has been reported to be 0.028 pA (22), which is rather similar to the unitary response of 0.034 pA that we have previously estimated for the same species (9). Thus, the unitary response at transient peak possibly corresponds to the simultaneous opening of perhaps one or a small number of CNG channels. This does not mean that the physical domain in the cilium affected by the unitary response contains only a few CNG channels. Rather, in response to the cAMP concentration at transduction peak, only a small number of them out of many randomly open at a given time instant. The situation may be similar in the mouse. How much of the unitary-response current at transient peak, if any, is Cl current remains to be examined.

An ORN signals to the brain via action potentials in its axon. The signaling threshold for a mouse ORN at 35 °C is perhaps ≤19 unitary events (i.e., odorant-binding events that succeed in triggering transduction). We define threshold here as an ORN's mean firing frequency (in a time window of 0.5 s after the odorant pulse) that reaches or exceeds the mean basal rate by at least 2 SDs. Without knowing how the brain processes impulse information from ORNs, this threshold criterion is necessarily somewhat arbitrary. At room temperature, the threshold is not very different, being ≤24 unitary events. Previously, we defined threshold for amphibian ORNs as a firing probability of 0.5 during a 1-s time interval after the odorant pulse (essentially spanning the time course of the receptor current at room temperature) and arrived at a threshold of ~35 events for frog ORNs (9). This definition of threshold works for frog ORNs, which are generally quiescent at room temperature without odorant stimulus, but is not generally useful for neurons that are not quiescent at rest, such as with mouse ORNs. It is possible, however, to convert the frog data roughly into a form more relatable to the mouse data. From the Poisson distribution, P≥1 = 1 − P0 = 1 − e−m, where P≥1 is the probability of one or more action potentials during the 1-s period, P0 is the probability of no action potentials, and m is the mean number of action potentials. Frog ORNs have a basal probability of firing over 1 s of 0.2 or less (9); thus, P≥1 ≤ 0.2, giving m ≤ 0.22, or a firing rate of 0.22 Hz. At P≥1 = 0.5, the previously defined threshold, we get m = 0.69, or a firing rate of 0.69 Hz. In this previous work, we have not calculated the SD of basal firing, but 0.69 Hz may well be more than 2 SDs above 0.22 Hz (i.e., beyond the threshold defined in the current work). If so, the frog ORN's signaling threshold according to our definition here could be significantly less than 35 unitary events, thus getting closer to the value for mouse ORNs at room temperature. At any rate, the signaling thresholds for frog and mouse ORNs appear comparable.

At this point, we do not know whether the signaling by one ORN to the brain will trigger the animal's conscious detection of an odorant pulse because the details of signal transmission at the glomerulus in the olfactory bulb, and also further up in the olfactory cortex, are largely unknown, with information only emerging recently and gradually [e.g., (23, 24)]. As a comparison, the single-photon response of a mouse rod photoreceptor at 35 °C is ~0.5 pA (25, 26), thus remarkably similar to the unitary olfactory response that we have estimated here for mouse ORNs. On the other hand, the graded signaling of rods allows the single-photon response to be faithfully signaled to the postsynaptic cell (27). Even at the threshold of human consciousness, only a handful of photons need to be absorbed by the rods in a small locality of the retina (27), and the number of retinal ganglion cells conveying this information to the brain may only be a few (28, 29). Possibly, the olfactory threshold at consciousness also involves the impulse activity of only one or a few ORNs.

Materials and Methods

Tissue Preparation.

The olfactory turbinates of a euthanized animal were removed from the nasal cavity and stored in a mammalian solution (30) at 4 °C until use. When needed, a small piece of epithelium was peeled from a turbinate, placed in an Eppendorf tube containing 200 μL of mammalian solution, and briefly vortexed. The dissociated cells were allowed to settle for 30 min in a recording chamber and identified based on their typical morphology.

Solutions and Odorant Application.

Mammalian solution contained 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 0.01 mM EDTA, and 10 mM Hepes. The Ca2+-free mammalian solution contained 140 mM NaCl, 5 mM KCl, 3.2 mM MgCl2, 1 mM EGTA, and 10 mM Hepes (3 mM free Mg2+, calculated according to the Web-based program MaxChelator, http://www.stanford.edu/~cpatton/maxc.html) (31). In all solutions, the pH was adjusted to 7.5 with NaOH. Glucose (10 mM) was added to the mammalian solution used for storing the intact epithelium or superfusing the dissociated cells in the experimental chamber before odorant application. Odorant solutions were made fresh daily by dissolving the odorants in the mammalian solution. The odorant mixture contained 1-heptanol, cineole, acetophenone, hexanal, and (+)citronellal (all from Sigma). The concentration of each odorant was 10–300 μM as indicated, whether singly or in a mixture. The cells in the chamber were continuously superfused with mammalian solution containing Ca2+. To apply odorant under the normal- or Ca2+-free condition, the recorded cell held at the pipette tip was first transferred to a stream of mammalian solution with or without Ca2+ on a Perfusion Fast-Step SF-77B Solution Changer (Warner Instrument) and then moved across the interface between this solution and an adjacent stream of the same solution containing odorant for a defined duration before being returned to the base solution. From junction-current measurements with 90% (vol/vol) mammalian solution to monitor the time course of solution change, a change was essentially complete within 15 ms. For Ca2+-free experiments, the cell was returned to normal-Ca2+ solution for several seconds between stimulus trials. Experiments were performed at room temperature (23 °C) unless stated otherwise. For experiments at 35 °C, the solutions were heated before entering the solution changer (30). The temperature near the pipette tip was continuously monitored with a telethermometer.

Recording and Western blot details are provided in SI Materials and Methods.

Supplementary Material

Supporting Information:


We thank D. Hervé for the Gαolf antibody, R. Reed for the Gαolf+/− mouse line, and M. T. Do for the spike-detection program. We also thank S. J. Kleene and T.-Y. Chen for helpful comments and reviews. In particular, we thank S. J. Kleene for pointing out the similar amplitudes of the unitary olfactory response and the current of a single CNG channel. Finally, we thank M. T. Do, C.-C. Lin, D.-G. Luo, T. Xue, and other members of the K.-W.Y. laboratory as well as V. Bhandawat, G. Lowe, and F. Rieke for discussions and comments. This work was supported by National Institutes of Health Grant DC06904 and a Human Frontier Science Program Organization Long-Term Fellowship to Y.B.-C.


The authors declare no conflict of interest.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1017983108/-/DCSupplemental.


1. Luo D-G, Xue T, Yau K-W. How vision begins: An odyssey. Proc Natl Acad Sci USA. 2008;105:9855–9862. [PMC free article] [PubMed]
2. Bhandawat V, Reisert J, Yau K-W. Elementary response of olfactory receptor neurons to odorants. Science. 2005;308:1931–1934. [PMC free article] [PubMed]
3. Schild D, Restrepo D. Transduction mechanisms in vertebrate olfactory receptor cells. Physiol Rev. 1998;78:429–466. [PubMed]
4. Matthews HR, Reisert J. Calcium, the two-faced messenger of olfactory transduction and adaptation. Curr Opin Neurobiol. 2003;13:469–475. [PubMed]
5. Kleene SJ. The electrochemical basis of odor transduction in vertebrate olfactory cilia. Chem Senses. 2008;33:839–859. [PubMed]
6. Kato A, Touhara K. Mammalian olfactory receptors: Pharmacology, G protein coupling and desensitization. Cell Mol Life Sci. 2009;66:3743–3753. [PubMed]
7. Munger SD, Leinders-Zufall T, Zufall F. Subsystem organization of the mammalian sense of smell. Annu Rev Physiol. 2009;71:115–140. [PubMed]
8. Buck LB. The molecular architecture of odor and pheromone sensing in mammals. Cell. 2000;100:611–618. [PubMed]
9. Bhandawat V, Reisert J, Yau K-W. Signaling by olfactory receptor neurons near threshold. Proc Natl Acad Sci USA. 2010;107:18682–18687. [PMC free article] [PubMed]
10. Firestein S, Picco C, Menini A. The relation between stimulus and response in olfactory receptor cells of the tiger salamander. J Physiol. 1993;468:1–10. [PMC free article] [PubMed]
11. Baylor DA, Hodgkin AL, Lamb TD. The electrical response of turtle cones to flashes and steps of light. J Physiol. 1974;242:685–727. [PMC free article] [PubMed]
12. Calvert PD, et al. Phototransduction in transgenic mice after targeted deletion of the rod transducin alpha-subunit. Proc Natl Acad Sci USA. 2000;97:13913–13918. [PMC free article] [PubMed]
13. Belluscio L, Gold GH, Nemes A, Axel R. Mice deficient in G(olf) are anosmic. Neuron. 1998;20:69–81. [PubMed]
14. Corvol JC, Studler JM, Schonn JS, Girault JA, Hervé D. Galpha(olf) is necessary for coupling D1 and A2a receptors to adenylyl cyclase in the striatum. J Neurochem. 2001;76:1585–1588. [PubMed]
15. Mendez A, et al. Rapid and reproducible deactivation of rhodopsin requires multiple phosphorylation sites. Neuron. 2000;28:153–164. [PubMed]
16. Krispel CM, et al. RGS expression rate-limits recovery of rod photoresponses. Neuron. 2006;51:409–416. [PubMed]
17. Doan T, Mendez A, Detwiler PB, Chen J, Rieke F. Multiple phosphorylation sites confer reproducibility of the rod's single-photon responses. Science. 2006;313:530–533. [PubMed]
18. Kurahashi T, Yau K-W. Co-existence of cationic and chloride components in odorant-induced current of vertebrate olfactory receptor cells. Nature. 1993;363:71–74. [PubMed]
19. Lowe G, Gold GH. Nonlinear amplification by calcium-dependent chloride channels in olfactory receptor cells. Nature. 1993;366:283–286. [PubMed]
20. Zhainazarov AB, Ache BW. Odor-induced currents in Xenopus olfactory receptor cells measured with perforated-patch recording. J Neurophysiol. 1995;74:479–483. [PubMed]
21. Boccaccio A, Menini A. Temporal development of cyclic nucleotide-gated and Ca2+-activated Cl currents in isolated mouse olfactory sensory neurons. J Neurophysiol. 2007;98:153–160. [PubMed]
22. Kleene SJ. High-gain, low-noise amplification in olfactory transduction. Biophys J. 1997;73:1110–1117. [PMC free article] [PubMed]
23. Murphy GJ, Glickfeld LL, Balsen Z, Isaacson JS. Sensory neuron signaling to the brain: Properties of transmitter release from olfactory nerve terminals. J Neurosci. 2004;24:3023–3030. [PubMed]
24. Franks KM, Isaacson JS. Strong single-fiber sensory inputs to olfactory cortex: Implications for olfactory coding. Neuron. 2006;49:357–363. [PubMed]
25. Xu J, et al. Prolonged photoresponses in transgenic mouse rods lacking arrestin. Nature. 1997;389:505–509. [PubMed]
26. Chen CK, et al. Abnormal photoresponses and light-induced apoptosis in rods lacking rhodopsin kinase. Proc Natl Acad Sci USA. 1999;96:3718–3722. [PMC free article] [PubMed]
27. Hecht S, Shlaer S, Pirenne MH. Energy, quanta, and vision. J Gen Physiol. 1942;25:819–840. [PMC free article] [PubMed]
28. Barlow HB, Levick WR, Yoon M. Responses to single quanta of light in retinal ganglion cells of the cat. Vision Res. 1971;11:87–101. [PubMed]
29. Mastronarde DN. Correlated firing of cat retinal ganglion cells. II. Responses of X- and Y-cells to single quantal events. J Neurophysiol. 1983;49:325–349. [PubMed]
30. Reisert J, Matthews HR. Response properties of isolated mouse olfactory receptor cells. J Physiol. 2001;530:113–122. [PMC free article] [PubMed]
31. Reisert J, Lai J, Yau KW, Bradley J. Mechanism of the excitatory Cl response in mouse olfactory receptor neurons. Neuron. 2005;45:553–561. [PMC free article] [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...