• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Nature. Author manuscript; available in PMC Jun 6, 2013.
Published in final edited form as:
PMCID: PMC3674497

Loss-of-function mutations in sodium channel Nav1.7 cause anosmia


Loss of function of the gene SCN9A, encoding the voltage-gated sodium channel Nav1.7, causes a congenital inability to experience pain in humans. Here we show that Nav1.7 is not only necessary for pain sensation but is also an essential requirement for odour perception in both mice and humans. We examined human patients with loss-of-function mutations in SCN9A and show that they are unable to sense odours. To establish the essential role of Nav1.7 in odour perception, we generated conditional null mice in which Nav1.7 was removed from all olfactory sensory neurons. In the absence of Nav1.7, these neurons still produce odour-evoked action potentials but fail to initiate synaptic signalling from their axon terminals at the first synapse in the olfactory system. The mutant mice no longer display vital, odour-guided behaviours such as innate odour recognition and avoidance, short-term odour learning, and maternal pup retrieval. Our study creates a mouse model of congenital general anosmia and provides new strategies to explore the genetic basis of the human sense of smell.

The inability to sense odours is known as general anosmia; individuals born with this phenotype are afflicted with congenital general anosmia. Except for some syndromic cases such as Kallmann syndrome, no causative genes for human congenital general anosmia have been identified so far13. Nine mammalian genes encoding voltage-gated sodium channel α-subunits have been cloned and shown to be differentially expressed in the nervous system4,5. Of these, SCN9A, encoding the tetrodotoxin (TTX)-sensitive sodium channel Nav1.7, has received specific attention because of its key role in human pain perception. Individuals carrying loss-of-function mutations in SCN9A are unable to experience pain, and an essential requirement of Nav1.7 function for nociception in humans has been established69. Whether all other sensory modalities are fully preserved in these individuals remained unclear, although an association between congenital inability to experience pain and sense of smell deficits has been suggested7. In this study we examine human patients carrying SCN9A loss-of-function mutations and demonstrate that they fail to sense odours. We establish a mouse model of congenital general anosmia and provide mechanistic insight into the role of Nav1.7 in olfaction. Together with previous findings68,10, our results establish that loss-of-function mutations in a single gene, SCN9A, cause a general loss of two major senses—nociception and smell—thus providing a mechanistic link between these two sensory modalities.

Requirement for Nav1.7 in human olfaction

Three individuals with congenital analgesia were ascertained and studied. All three were in their third decade of life and had never experienced acute pain but had no other neurological, cognitive, growth, appearance or health problems. All had suffered from multiple painless fractures and other injuries. Two had given birth painlessly. A working diagnosis of channelopathy-associated insensitivity to pain (CAIP) was made and in each SCN9A was sequenced6. In the first, who has been the subject of a detailed case report, the mutations c.774_775delGT and c.2488C>T were found10. These mutations, frameshift and nonsense, respectively, would be predicted to lead to a lack of functional Nav1.7 protein. The other two were siblings and had the mutations c.4975A>T and c.3703delATAGCATATGG; again, nonsense and frameshift mutations and predicted to lead to no functional Nav1.7 protein. The mother of the siblings was found to be heterozygous for the 11-base-pair deletion and the father heterozygous for the nonsense mutation. Therefore the diagnosis of CAIP was substantiated. We next assessed their sense of smell; none complained of having no sense of smell, one had been a cigarette smoker, none had chronic nasal problems. In the first woman smell function was assessed by using the University of Pennsylvania Smell Identification Test (UPSIT), a standardized 40-item smell test. The results revealed that she was unable to detect any of the odours (Fig. 1a, black bar). Nine healthy, young individuals served as controls (Fig. 1a, grey bars). In the sibling pair we assessed the parents and their two affected offspring together. All were tested in sequence with cotton wool pads suffused with selected odour stimuli: balsamic vinegar, orange, mint, perfume, water (control) and coffee. Both parents correctly identified all stimuli, including smelling nothing for the water. The siblings detected none of the odours. For the siblings the test was repeated using subjectively unpleasant amounts of balsamic vinegar and perfume: the parents identified the odours correctly and found them unpleasant; the siblings neither identified the odours nor experienced any discomfort.

Figure 1
Nav1.7 in human olfaction

We proposed that these odour-sensing deficits are caused by loss of Nav1.7 function in olfactory sensory neurons (OSNs). Indeed, when we investigated expression of Nav1.7 in normal human olfactory epithelium, we detected messenger RNA for Nav1.7 and the GTP-binding protein Gαolf, a prototypical signature of classical OSNs (Fig. 1b). Immunohistochemistry using an antibody specific to Nav1.7 verified that Nav1.7 is normally expressed in human OSNs (Fig. 1c, d).

Conditional Nav1.7 null mice

To investigate the mechanisms that underlie the essential role of Nav1.7 in odour perception, we first examined Nav1.7 expression in the mouse olfactory system and then used the Cre-loxP system to delete the channel in those cells that express olfactory marker protein (OMP), which includes all classical OSNs11. These mice enabled us to examine the mechanisms underlying Nav1.7-associated anosmia and thebehavioural consequences. Consistent with our findings in human OSNs, OSNs from wild-type mice (C57BL/6, referred to as B6) showed Nav1.7 immunoreactivity at their somata (Supplementary Fig. 1). Of greater interest, coronal sections containing main olfactory epithelium (MOE), olfactory nerves and the two olfactory bulbs revealed the most marked Nav1.7 staining in the superficial olfactory nerve layer (ONL, containing axons from OSNs) as well as the glomerular layer (a complex neuropil that includes the presynaptic OSN boutons) of the olfactory bulb (Fig. 2a–c). Higher magnification of individual glomeruli verified co-expression of Nav1.7 with OMP in the glomerular neuropil (Fig. 2b), whereas olfactory bulb projection neurons (the mitral/tufted or M/T cells) and local interneurons did not show Nav1.7 immunoreactivity (Fig. 2a). Thus, Nav1.7 occupies a critical presynaptic location at the first synapse in the olfactory system.

Figure 2
Nav1.7 expression in the mouse main olfactory system

Nav1.7 is not the sole Nav channel expressed in mouse OSNs. Real-time quantitative polymerase chain reaction with reverse transcription (qRT–PCR) analysis identified Nav1.3 as an additional candidate (Supplementary Fig. 2) and immunohistochemistry verified its expression in OSNs and their axons (Fig. 2d). However, unlike Nav1.7 we did not observe Nav1.3 immunoreactivity in individual glomeruli (Fig. 2d), indicating that Nav1.7 could be the sole Nav channel underlying action potential propagation in olfactory glomeruli and OSN nerve terminals.

To create a conditional knockout mouse model, we crossed ‘floxed’ Nav1.7 mice harbouring a loxP-flanked Scn9a gene12 to homozygous OMP–Cre mice in which the OMP-coding region is replaced by that of Cre recombinase13. Further breeding established offspring that were both homozygous for the floxed Scn9a alleles and heterozygous for cre and Omp. In these mice, Cre-mediated Nav1.7 deletion was restricted to OMP-positive cells (henceforward referred to as cNav1.7−/− mice). These mice lacked Nav1.7 expression in a tissue-specific manner (Fig. 2e, f and Supplementary Fig. 3). Successful matings occurred between cNav1. 7+/− males and cNav1.7−/− zygous knockout females whereas homo-pairs did not produce any offspring. cNav1.7−/− mice showed a reduced body weight during the first three months of postnatal development (Supplementary Fig. 4a). Because both Nav1.7 (refs 14, 15) and OMP16 are also expressed in some neurons mediating hormonal regulation, we assayed insulin-like growth factor (IGF-1, also known as somatomedin C) in cNav1.7−/− (650 ± 94 ng ml−1; n = 4) versus cNav1.7+/− mice (684 ± 27 ng ml−1; n = 4; mean ± s.d.) but found no significant difference between the two genotypes (P = 0.26). Given that newborn cNav1.7−/− mice had very little milk in their stomachs (Supplementary Fig. 4b), the diminished weight gain was probably caused by a deficit to suckle effectively, consistent with results in mice deficient in Gαolf (Gnal) or the cAMP-gated cation channel (Cnga2)17,18.

Loss of synaptic transfer in olfactory glomeruli

To define the function of Nav1.7 in OSNs, we prepared MOE tissue slices19 and recorded sodium currents in voltage-clamped OSNs. Both Nav1. 7+/− and Nav1.7−/− OSNs displayed sizeable, TTX-sensitive sodium currents in response to step depolarizations (Fig. 3a, b). On the basis of its biophysical properties, Nav1.7 has been suggested to transduce generator potentials into action potentials in sensory neurons9. However, peak current densities of voltage-activated sodium currents were reduced only moderately, by about 20%, in Nav1.7−/− OSNs (Fig. 3b). To determine whether Nav1.7−/− OSNs could still produce odour-evoked action potentials, we used extracellular loose-patch recording from visually identified OSN dendritic knobs20 and analysed spike frequency histograms after brief odour exposure (Fig. 3c). There was no obvious difference in odour responsiveness in Nav1.7−/− versus Nav1.7+/− OSNs (Fig. 3c). We obtained similar results when we stimulated the cells with 3-isobutyl-1-methylxanthine (IBMX)21, which raises intracellular cAMP by inhibiting endogenous phosphodiesterase activity (Fig. 3c). Thus, although the initial site of odour-evoked action potential generation in OSNs is unknown, Nav1.7 is not essential for this activity.

Figure 3
Nav1.7 is essential for synaptic transfer in the olfactory glomerulus

Because Nav1.7 is expressed in olfactory bulb glomeruli (Fig. 2), we reasoned that it could be required for action potential conduction in OSN terminals. Olfactory glomeruli are delineated spheres of neuropil containing synapses from the OSN axon terminals onto juxtaglomerular interneurons and M/T projection neurons22,23. To examine whether presynaptic activity of Nav1.7 underlies transmitter release in the olfactory glomerulus, we prepared olfactory bulb tissue slices24 and combined ONL focal electric stimulation with whole-cell patch-clamp recording from visually identified M/T cells. With the chosen protocol, in control cNav1.7+/− mice a single electrical stimulus in the ONL produced a reliable postsynaptic response in M/T cells. Under current clamp, such responses consisted of a prolonged excitation lasting on average for 2.4 ± 0.4 s (Fig. 3d, top; n = 29), with response latencies of 22 ± 4 ms (n = 29). Under voltage clamp, we observed bursts of postsynaptic currents (Fig. 3f; duration, 3.2 ± 0.4 s; n = 26). In stark contrast, in the cNav1.7−/− mice such postsynaptic responses were virtually absent in M/T cells, even when the stimulus strength was increased by several-fold (Fig. 3d–f; n = 49). Importantly, M/T cells in these mice still produced normal action potentials when depolarized via current injection through the patch pipette (Fig. 3d, bottom), consistent with the fact that M/T cells lack both OMP and Nav1.7 expression (Fig. 2) and indicating that the effect of deleting Nav1.7 is presynaptic to the M/T cells. The inability of M/T cells to produce synaptic responses to ONL stimulation was not due to a potential deficit in synapse formation because: (1) immunohistochemistry showed normal expression of the vesicular glutamate transporter 2 (vGluT2, which is selectively expressed in OSN axon terminals)25,26 (Supplementary Fig. 5); and (2) electron microscopy revealed the existence of normal OSN boutons and synapses in the glomeruli of cNav1.7−/− mice (Supplementary Fig. 6). Furthermore, conditional OSN expression of tetanus toxin light chain, which inhibits synaptic release, does not alter the pattern of axonal targeting in olfactory bulb glomeruli during development27.

Tyrosine hydroxylase (TH) expression in juxtaglomerular neurons of the olfactory bulb, a correlate of afferent trans-synaptic activity, requires olfactory nerve input and odour-stimulated glutamate release by OSN terminals28. Consistent with a loss of OSN synaptic release, TH expression was markedly reduced in cNav1.7−/− mice (Fig. 3h; n = 6). The level of TH downregulation was similar to that observed after odour deprivation by naris occlusion29 or after deletion of the Cnga2 cation channel gene30. Thus, we conclude that the presence of Nav1.7 in OSN axons is an essential and non-redundant requirement to initiate information transfer from OSN terminals to neurons in the olfactory bulb.

The absence of odour-guided behaviours

To further validate these results, we investigated several odour-guided behaviours in B6, cNav1.7+/− and cNav1.7−/− mice. First, we performed an odour preference test31 to assess recognition abilities for innate odour qualities (Fig. 4a). Filter papers scented with various cues representing both species-specific and food odours (male and female urine, peanut butter, milk) were presented to the mice and investigation times were analysed. Water was used as a neutral stimulus and 1,8-cineole (eucalyptol), which does not evoke innate attraction, served as the control (n ≥ 7 for each cue and strain, respectively). B6 and cNav1.7+/− mice both showed strong attraction towards con-specific and food odours, whereas cNav1.7−/− mice failed to show any interest in these stimuli.

Figure 4
cNav1.7−/− mice are anosmic

Second, we explored whether Nav1.7 is required for innate avoidance behaviour towards a predator odour, trimethyl-thiazoline (TMT)31, which is normally secreted from the fox anal gland and known to induce aversive behaviour and fear responses in mice. We observed robust avoidance behaviour in both B6 (n = 6) and cNav1.7+/− mice (n = 5) but, notably, cNav1.7−/− mice lacked an innately aversive response in this assay (n = 5; Fig. 4b, c).

Third, we investigated the performance of cNav1.7−/− mice in a habituation–dishabituation assay, which allows for measurement of novel odour investigation, short-term odour learning, and odour discrimination32 (Fig. 4d). Mice of both sexes were each presented three distinct stimuli (water, female urine, male urine), each delivered for three successive trials, and investigation time during each trial (3 min) mice was analysed. Consistent with the results of Fig. 4a, cNav1.7−/− (n = 8) failed to show significant odour investigation, habituation, or discrimination abilities when compared with B6 (n = 8) or cNav1.7+/− mice (n = 8) (Fig. 4d; least significant difference (LSD), P < 0.0001).

Last, we examined pup retrieval ability of female mice, a social behaviour that probably depends on a functional main olfactory system (Fig. 4e). Three pups of a litter were removed from the nest, randomly distributed in the cage, and the time to retrieve each pup into the nest was quantified. In contrast to the performance of B6 (n = 12) or cNav1.7+/− mice (n = 6), cNav1.7−/− mice (n = 5) failed to retrieve any of the three pups during a 10-min trial period (Fig. 4e).

Conclusions and prospects

Our results establish a critical role of the Nav1.7 sodium channel in olfaction. Using conditional Nav1.7 null mice, we demonstrate that, in the absence of Nav1.7, OSNs are still electrically active and generate odour-evoked action potentials but fail to initiate synaptic signalling to the projection neurons in the olfactory bulb. These results provide evidence that Nav1.7 is an essential and non-redundant requirement for action potential propagation in the sections of OSN axons within the olfactory glomerulus. The conditional null mice no longer show a wide range of vital, odour-guided behaviours including innate attraction to food and conspecific odours, odour discrimination and short-term odour learning, innate avoidance towards a predator odour, effective suckling behaviour of newborn pups, and maternal pup retrieval. Within the limits of our anatomical analyses, synapse formation in these mice appears normal, indicating that the behavioural phenotype of the mutant mice is most likely the result of a loss of signalling at the first synapse in the olfactory system. Whether Nav1.7 or other sodium channel subunits such as Nav1.3 are involved in OSN axon pathfinding and activity-dependent neural map formation33 in the mouse olfactory system remains to be seen. Importantly, the phenotype of the mutant mice—the inability to perceive odours—is similar to that observed in human patients with confirmed Nav1.7 loss-of-function mutations. Smell tests in three individuals with congenital analgesia establish that they are unable to sense any of the odours. Systematic olfactory testing of patients carrying Nav1.7 loss-of-function mutations will be required in the future.

The genetic basis of sensory deficits such as blindness, deafness and pain disorders has been extensively studied in recent years. By comparison, relatively little progress has been made in understanding human congenital general anosmia1. Mutations in olfactory signal transduction genes such as CNGA2, GNAL and ADCY3 do not seem to be a major cause of human congenital general anosmia2. The identification of a sodium channel subunit as a causative gene for an inherited form of general anosmia provides new insight into the molecular pathophysiology of olfaction and should stimulate further research aimed at understanding the genetic basis of the human sense of smell.


Human biopsies

Human nasal mucosa was obtained by biopsy during routine nasal surgery with patients under general anaesthesia. Biopsy specimens were obtained from three individuals and snap-frozen in liquid nitrogen for later processing. All samples were obtained under a protocol approved by the Ethics Committee of the University of Saarland School of Medicine. All biopsy tissues were obtained with the informed consent of the patients.

Human psychophysics

The UPSIT was obtained from Sensonics. The test was applied over a period of 25 min. Testing and scoring was done according to standardized operating procedures summarized in the test manual. The reference values have been derived from recorded reference ranges for the UPSIT test based on British individuals.

Olfactory mucosa biopsies and PCR analyses

Human surgical material containing olfactory mucosa collected from three different patients was examined individually. RT–PCRs from human samples were performed on a MyCycler (BIO-RAD) with Herculase (Agilent Technologies) following suppliers’ instructions. To amplify human Gαolf we used the oligonucleotides TGGAAAGAATCGACAGCGTCAGC and GGCCACCAACATCAAACATGTGG. Human Nav1.7 was amplified by CATGAATAACCCACCGGACTG and CCTATGCCCTTCGACACCAAGG. PCR conditions were: 95 °C for 2 min pre-denaturation, followed by 35 cycles (95 °C for 30 s, 60 °C for 30 s (Gαolf) or 1 min (Nav1.7), 72 °C for 30 s), followed by a final extension 72 °C for 5 min. Mouse tissue was pooled from four different B6 mice (4–8 weeks old). RNA was isolated with the InnuPREP RNA isolation kit (Analyticjena). RNA quality was assessed by gel electrophoresis and photometric measurements. cDNA was synthesized from 0.5 μg of total RNA using the Smart cDNA Synthesis technology (Clontech) and Supercript II reverse transcriptase (Invitrogen). qPCR for different mouse Nav subunits were done on a My-iQ-cycler using iQ SYBRGreen Supermix following the supplier’s instructions (BIO-RAD). We used the following oligo-nucleotides: Nav1.1 (AGCCTGGTAGAACTTGGCCTTGC and TGCCAACCA CGGCAAAAATAAAG); Nav1.2 (TGGGATCTTCACCGCAGAAATG and TGGGCCAGGATTTTGCCAAC); (AGCTTGGCCTGGCAAACGTG Nav1.3 and ATGCCGACCACGGCAAAAATG); Nav1.5 (ACAGCCGAGTTTGAG GAGATGC and CGCTGATTCGGTGCCTCA); Nav1.6 (ACGCCACAATTC GAACATGTCC and CCTGGCTGATCTTACAGACGCA); Nav1.7 (ACGGAT GAATTCAAAAATGTACTTGCAG and GTTCTCGTTGATCTTGCAAACA CA). PCR conditions were: 95 °C for 3 min pre-denaturation, followed by 42 cycles 95 °C for 30 s, 64 °C for 20 s, 72 °C for 30 s. Each reaction was performed in three replicates on 96-well plates and analysed with the iQ5 Software (BIO-RAD). Specificity of all PCR products was confirmed by gel electrophoresis and sequencing.


Animal care and experimental procedures were performed in accordance with the guidelines established by the animal welfare committee of the University of Saarland School of Medicine. Mice were kept under a standard light/dark cycle with food and water ad libitum. Tissue-specific, Nav1.7-deficient mice were generated by crossbreeding ‘floxed’ Nav1.7 mice that carry two loxP sites, flanking exons 14 and 15 of Scn9a12 with homozygous OMP–Cre mice (B6;129P2-Omptm4(cre)Mom/MomJ) that express Cre recombinase under the control of the OMP promoter13. Further breeding established offspring that were both homozygous for the floxed Nav1.7 alleles and heterozygous for cre and Omp. In these mice, Cre-mediated Nav1.7 deletion was restricted to OMP-positive cells. Additionally, C57BL/6J (B6) and OMP–GFP (B6;129P2-Omptm3Mom/MomJ) mice were used.


Perfusion of mice and preparation of mouse olfactory tissues for immunohistochemistry followed previously described methods34. Cryosections (10–12 μm) of either human or mouse olfactory tissues were postfixed using 4% paraformaldehyde in PBS, before blocking and antibody administration. Primary antibodies were: mouse-specific anti-Nav1.7 (1:500, rabbit polyclonal; Millipore), human-specific anti-Nav1.7 (1:500, rabbit polyclonal; Abcam), Nav1.3 (1:500, rabbit polyclonal; Millipore), OMP (1:3,000, goat polyclonal; gift of F. Margolis), vGluT2 (1:2,000, rabbit polyclonal; Synaptic Systems), tyrosine hydroxylase (TH, 1:3,000, mouse monoclonal; ImmunoStar). Secondary antibodies and conjugated compounds were: Alexa-Fluor 488 donkey-anti-goat (1:1,000; Invitrogen), Alexa-Fluor 555 donkey-anti-rabbit (1:1,000; Invitrogen), Alexa-Fluor 546 Streptavidin (1:200; Invitrogen). Procedures were conducted at room temperature (21 °C), except for incubation in primary antibodies (4 °C). Expression of Nav1.7 in human was detected by direct immunofluorescence. Expression of Nav1.3 and Nav1.7 in mouse was detected by tyramid signal amplification using manufacturer’s protocol (TSA-Biotin System, Perkin Elmer). Incubation in primary antibody was for 2–3 days, in biotinylated anti-rabbit antibody (1:400; Jackson ImmunoResearch) for 1 h, in streptavidin-HRP (1:100) for 30 min, in biotinylated tryamid (1:100) for 10 min, and visualized using Alexa 546-conjugated steptavidin (Invitrogen, 1:200). OMP colocalization was detected using a Alexa 488-conjugated anti-goat secondary antibody. Detection of vGluT2 was exactly as previously described26. TH was detected in 30-μm free-floating sections using the avidin-biotin method (Vectastain ABC-Elite, Vector). Incubation in primary TH antibody was for 1 day, in biotinylated horse-anti-mouse secondary antibody (1:400, Vector Laboratories) for 1 h, and in avidin/biotin–HRP complex (Vector) for 90 min. Immunoreactivity was visualized with 0.05 g l−1 3,3′-diaminobenzidine and 0.015% H2O2. Fluorescence images were acquired on either a BX71 microscope attached to a DP71 camera (Olympus) or an LSM 710/ConfoCor-3 microscope (Zeiss). Image stacks are presented as maximum intensity projections, assembled and minimally adjusted in brightness using Adobe PhotoShop 6.0.

Electron microscopy

Following routine processing for electron microscopy, as previously described26,35, thin 70–100-nm sections were cut on a Reichert Ultracut E and examined on a JEOL 1200 transmission electron microscope. Images were captured at ×12,000, digitized at 1,200 dots per inch (DPI), and examined for ultrastructural features of the olfactory sensory axons and their synaptic terminals.


Whole-cell patch-clamp recordings from individual OSNs were obtained in acute MOE tissue slices of P1–P5 mice19. The anterior aspect of the head containing olfactory epithelium and bulb was embedded in agarose (4%), placed in oxygenated, ice-cold extracellular solution (95% O2, 5% CO2) containing: 120 mM NaCl, 25 mM NaHCO3, 5 mM KCl, 5 mM BES (N,N-bis[2-hydroxyethyl]-2-aminoethansulphonic acid), 1 mM MgSO4, 1 mM CaCl2, 10 mM glucose, osmolarity adjusted to 300 mOsm, pH 7.3. Coronal slices (250 μm) were cut on a vibratome (Microm HM 650 V), transferred to a recording chamber and kept under continuous flow (2 ml min−1) of oxygenated solution or remained on ice in oxygenated solution until needed (for up to 4 h). Experiments were performed at room temperature. The CsCl-based electrode solution contained: 140 mM CsCl, 1 mM EGTA, 10 mM HEPES, 0.5 mM GTP Na-salt, 2 mM ATP Mg-salt, pH 7.1, 290 mOsm. To assess OSN firing properties under non-invasive conditions, we used extracellular loose-patch recording from OSN knobs20. In this case, the septal epithelium of juvenile (P1–P5) or adult mice was dissected and transferred to a recording chamber. Patch pipettes (9–12 MΩ) were filled with a HEPES-based extracellular solution containing: 140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 10 mM HEPES, pH 7.4, 300 mOsm. IBMX was prepared in 10 mM stock solution containing 5% dimethylsulphoxide (DMSO) (v/v). For M/T cell recordings, brains were rapidly dissected in ice-cold oxygenated (95% O2, 5% CO2) solution containing: 83 mM NaCl, 26.2 mM NaHCO3, 1 mM NaH2PO4, 2.5 mM KCl, 3.3 mM MgSO4, 0.5 mM CaCl2, 70 mM sucrose, pH 7.3, 300 mOsm. Horizontal olfactory bulb slices (300 μm) were cut in this solution. Until use, slices were transferred to oxygenated modified artificial cerebrospinal fluid (ACSF, 95% O2, 5% CO2) containing: 125 mM NaCl, 25 mM NaHCO3, 2.5 mM KCl, 1.25 mM NaH2PO4, 1 mM MgCl2, 2 mM CaCl2 and 25 mM glucose. Recording pipettes had resistances of 4–7 MΩ. M/T cells were identified by size and location of their somata and filled with Lucifer Yellow during patch recording. The intracellular solution contained: 140 mM KCl, 1 mM EGTA, 10 mM HEPES, 1 mM ATP Na-salt, 0.5 mM GTP Mg-salt, 0.1 mM Lucifer Yellow; pH 7.1, 290 mOsm. M/T cells were held at −55 to −60 mV. Input and series resistances were 200–300 MΩ and 15–20 MΩ, respectively. After establishing a whole-cell recording, the ONL was stimulated using a glass electrode (1–1.5 MΩ) filled with HEPES-buffered extracellular solution connected to an electrical stimulator (single stimulus: 20 ms, 40 V, 266–400 μA). The stimulus pipette was placed rostrally to the recorded cell in the ONL. If a given M/T cell showed no postsynaptic response, the position of the stimulus pipette was changed until OSN axon bundles were found that caused M/T cell responses. Ionic currents were analysed using PulseFit 8.54 (HEKA) and IGOR Pro software (Wavemetrics)36. OSNs with leak currents >20 pA and M/T cells with leak currents >100 pA (all measured at −70 mV) were excluded from analysis. Cell capacitance (Cm) was monitored using the automated function of the EPC-9 amplifier. A stable Cm value over time was an important criterion for the quality of an experiment. Spike analysis was done off-line using IGOR Pro software with custom-written macros. Chemicals were purchased from Sigma unless otherwise stated. Drugs used in the electrophysiological experiments were prepared as stock solutions in DMSO or distilled water and diluted to the final concentration in HEPES-based extracellular solution. NaCl, MgCl2, glucose and CaCl2 were from Merck. IBMX (100 μM) and cineole (100 μM) were diluted in a HEPES-buffered extracellular solution (< 0.1% DMSO) and focally ejected using multibarrel stimulation pipettes.

Behavioural tests

The innate olfactory preference test followed previously described procedures31. Briefly, mice were habituated to the test conditions before odour exposure. Mice were individually placed in an empty cage for 30 min and then transferred to a new cage. This habituation was repeated three to four times for each animal. Soon after habituation, mice were transferred to the test cage, and a filter paper scented with a test odorant was introduced. Investigation times of the filter paper during the 3-min test period was recorded and quantified. Odour stimuli were freshly collected male and female B6 mouse urine (5 μl), peanut butter (10% w/v, 15 μl), milk powder (10% w/v, 15 μl), water (15 μl) and cineole (100 μM, 15 μl).

For the innate olfactory avoidance test, following habituation (see innate preference test), a filter paper scented with 5 μl TMT (7.6 mM) was placed in one corner of the test cage. Mouse behaviour was recorded for 30 min. The test cage was subdivided into three equally sized areas. Time spent in area 1 of the cage (farthest distance from the TMT source) was evaluated as avoidance, whereas time spent in area 2 (consisting of the TMT source) was evaluated as attraction37. Animal movements were tracked with SwisTrack (Swarm Intelligent Systems Group, Swiss Federal Institute of Technology).

For the olfactory habituation–dishabituation assay, following habituation (see innate preference test) mice were exposed for 3 min to distilled water (15 μl). This procedure was repeated three times with 1-min intervals, followed by a three-time presentation of female urine (5 μl) and a three-time presentation of male urine (5 μl). Investigation times during the 3-min test periods were measured.

For the pup retrieval test, lactating mice were habituated to the experiment for several minutes. Experiments were performed in the bedded home cages of the dams. Three pups (1–3-days old) were removed from the nest and randomly distributed in the cage. The latency for pup retrieval back into the nest was measured. If a dam had not completed retrieval within 10 min the test was terminated, resulting in a latency of 600 s.

Experiments were performed in empty standard cages (38 ×19 ×12 cm) and test substances were applied on filter paper (~1 ×3 cm). Mouse behaviour was recorded with a digital camera (Sony) for the experimental times indicated. Statistical video analyses were done randomly and blindly. Peanut butter (Barney’s Best) and milk powder (Bio-Anfangsmilch, Hipp) were diluted to 10% (w/v) in water.

IGF-1 assays

IGF-1 levels were measured by sandwich ELISA (ALPCO Diagnostics). IGF-1 was dissociated from the binding proteins by diluting samples with an acidic buffer. The analytical sensitivity of the assay was 0.029 ng ml−1. Inter and intra-assay variability was below 7%. Experiments used plasma of 4–5-weeks-old mice (n = 4, each genotype).


Data were analysed using NCSS 2004 statistical software (NCSS). The Student’s t-test (two-tailed) was used for measuring the significance of difference between two distributions. Multiple groups were compared using a one-way or two-way analysis of variance (ANOVA) with Fisher’s LSD as a post hoc comparison. Unless otherwise stated, results are presented as means ± s.e.m.

Supplementary Material

Supplementary Files


We thank the individuals who participated in this study, P. Mombaerts for supplying OMP–Cre and OMP–GFP mice, F. Margolis for anti-OMP antibodies, J. Epelbaum for supporting the IGF-1 measurements, P. Hammes for assistance with the immunohistochemistry and C. Kaliszewski for assistance with the electron microscopy. This work was supported by grants from the Deutsche Forschungsgemeinschaft (DFG) to F.Z. (SFB 530 and SFB 894) and T.L.-Z. (SFB 894). E.J. was supported by the DFG-fundedInternational GraduateSchool GK 1326. T.L.-Z. is a Lichtenberg Professor of the Volkswagen Foundation. J.N.W. was supported by the Biotechnology and Biological Sciences Research Council, Medical Research Council, Wellcome Trust and grant number R31-2008-000-10103-0 from the World Class University project of the Korean Ministry of Education, Science and Technology and the National Research Foundation of Korea.


Full Methods and any associated references are available in the online version of the paper at www.nature.com/nature.

Supplementary Information is linked to the online version of the paper at www.nature.com/nature.

Author Contributions J.N.W. and F.Z. conceived the study. J.W., M.P., E.J., B.B. and P.Z. performed experiments. B.S. performed human biopsies. C.G.W., S.J.G. and J.N.W. performed human smell tests. J.W., M.P., E.J., B.B., V.W., P.Z., S.J.G., C.A.G., T.L.-Z., C.G.W., J.N.W. and F.Z. analysed results. M.P., T.L.-Z., J.N.W. and F.Z. contributed key reagents. F.Z. wrote the manuscript. All authors edited the manuscript.

Reprints and permissions information is available at www.nature.com/reprints.

The authors declare no competing financial interests.

Readers are welcome to comment on the online version of this article at www.nature.com/nature.


1. Hasin-Brumshtein Y, Lancet D, Olender T. Human olfaction: from genomic variation to phenotypic diversity. Trends Genet. 2009;25:178–184. [PubMed]
2. Feldmesser E, et al. Mutations in olfactory signal transduction genes are not a major cause of human congenital general anosmia. Chem Senses. 2007;32:21–30. [PubMed]
3. Keller A, Vosshall LB. Better smelling through genetics: mammalian odor perception. Curr Opin Neurobiol. 2008;18:364–369. [PMC free article] [PubMed]
4. Goldin AL. Resurgence of sodium channel research. Annu Rev Physiol. 2001;63:871–894. [PubMed]
5. Catterall WA, Goldin AL, Waxman SG. International Union of Pharmacology. XLVII Nomenclature and structure–function relationships of voltage-gated sodium channels. Pharmacol Rev. 2005;57:397–409. [PubMed]
6. Cox JJ, et al. An SCN9A channelopathy causes congenital inability to experience pain. Nature. 2006;444:894–898. [PubMed]
7. Goldberg YP, et al. Loss-of-function mutations in the Nav1.7 gene underlie congenital indifference to pain in multiple human populations. Clin Genet. 2007;71:311–319. [PubMed]
8. Ahmad S, et al. A stop codon mutation in SCN9A causes lack of pain sensation. Hum Mol Genet. 2007;16:2114–2121. [PubMed]
9. Dib-Hajj SD, Cummins TR, Black JA, Waxman SG. From genes to pain: Nav1.7 and human pain disorders. Trends Neurosci. 2007;30:555–563. [PubMed]
10. Nilsen KB, et al. Corrigendum to “Two novel SCN9A mutations causing insensitivity to pain. Pain. 2009;145:264. [PubMed]
11. Munger SD, Leinders-Zufall T, Zufall F. Subsystem organization of the mammalian sense of smell. Annu Rev Physiol. 2009;71:115–140. [PubMed]
12. Nassar MA, et al. Nociceptor-specific gene deletion reveals a major role for Nav1.7 (PN1) in acute and inflammatory pain. Proc Natl Acad Sci USA. 2004;101:12706–12711. [PMC free article] [PubMed]
13. Li J, Ishii T, Feinstein P, Mombaerts P. Odorant receptor gene choice is reset by nuclear transfer from mouse olfactory sensory neurons. Nature. 2004;428:393–399. [PubMed]
14. Klugbauer N, Lacinova L, Flockerzi V, Hofmann F. Structure and functional expression of a new member of the tetrodotoxin-sensitive voltage-activated sodium channel family from human neuroendocrine cells. EMBO J. 1995;14:1084–1090. [PMC free article] [PubMed]
15. Morinville A, et al. Distribution of the voltage-gated sodium channel Nav1.7 in the rat: expression in the autonomic and endocrine systems. J Comp Neurol. 2007;504:680–689. [PubMed]
16. Baker H, Grillo M, Margolis FL. Biochemical and immunocytochemical characterization of olfactory marker protein in the rodent central nervous system. J Comp Neurol. 1989;285:246–261. [PubMed]
17. Belluscio L, Gold GH, Nemes A, Axel R. Mice deficient in Golf are anosmic. Neuron. 1998;20:69–81. [PubMed]
18. Zhao H, Reed RR. X inactivation of the OCNC1 channel gene reveals a role for activity-dependent competition in the olfactory system. Cell. 2001;104:651–660. [PubMed]
19. Spehr M, et al. Essential role of the main olfactory system in social recognition of major histocompatibility complex peptide ligands. J Neurosci. 2006;26:1961–1970. [PubMed]
20. Leinders-Zufall T, et al. Contribution of the receptor guanylyl cyclase GC-D to chemosensory function in the olfactory epithelium. Proc Natl Acad Sci USA. 2007;104:14507–14512. [PMC free article] [PubMed]
21. Munger SD, et al. Central role of the CNGA4 channel subunit in Ca2+-calmodulin-dependent odor adaptation. Science. 2001;294:2172–2175. [PMC free article] [PubMed]
22. Shepherd GM, Chen WR, Greer CA. In: The Synaptic Organization of the Brain. Shepherd GM, editor. Oxford Univ. Press; 2004. pp. 165–216.
23. Wachowiak M, Shipley MT. Coding and synaptic processing of sensory information in the glomerular layer of the olfactory bulb. Semin Cell Dev Biol. 2006;17:411–423. [PubMed]
24. Nickell WT, Shipley MT, Behbehani MM. Orthodromic synaptic activation of rat olfactory bulb mitral cells in isolated slices. Brain Res Bull. 1996;39:57–62. [PubMed]
25. Gabellec MM, Panzanelli P, Sassoe-Pognetto M, Lledo PM. Synapse-specific localization of vesicular glutamate transporters in the rat olfactory bulb. Eur J Neurosci. 2007;25:1373–1383. [PubMed]
26. Richard MB, Taylor SR, Greer CA. Age-induced disruption of selective olfactory bulb synaptic circuits. Proc Natl Acad Sci USA. 2010;107:15613–15618. [PMC free article] [PubMed]
27. Yu CR, et al. Spontaneous neural activity is required for the establishment and maintenance of the olfactory sensory map. Neuron. 2004;42:553–566. [PubMed]
28. Puche AC, Shipley MT. Odor-induced, activity-dependent transneuronal gene induction in vitro: mediation by NMDA receptors. J Neurosci. 1999;19:1359–1370. [PubMed]
29. Cho JY, Min N, Franzen L, Baker H. Rapid down-regulation of tyrosine hydroxylase expression in the olfactory bulb of naris-occluded adult rats. J Comp Neurol. 1996;369:264–276. [PubMed]
30. Baker H, et al. Targeted deletion of a cyclic nucleotide-gated channel subunit (OCNC1): biochemical and morphological consequences in adult mice. J Neurosci. 1999;19:9313–9321. [PubMed]
31. Kobayakawa K, et al. Innate versus learned odour processing in the mouse olfactory bulb. Nature. 2007;450:503–508. [PubMed]
32. Wesson DW, Levy E, Nixon RA, Wilson DA. Olfactory dysfunction correlates with amyloid-β burden in an Alzheimer’s disease mouse model. J Neurosci. 2010;30:505–514. [PMC free article] [PubMed]
33. Sakano H. Neural map formation in the mouse olfactory system. Neuron. 2010;67:530–542. [PubMed]
34. Pyrski M, et al. Sodium/calcium exchanger expression in the mouse and rat olfactory systems. J Comp Neurol. 2007;501:944–958. [PubMed]
35. Au WW, Treloar HB, Greer CA. Sublaminar organization of the mouse olfactory bulb nerve layer. J Comp Neurol. 2002;446:68–80. [PubMed]
36. Ukhanov K, Leinders-Zufall T, Zufall F. Patch-clamp analysis of gene-targeted vomeronasal neurons expressing a defined V1r or V2r receptor:ionic mechanisms underlying persistent firing. J Neurophysiol. 2007;98:2357–2369. [PubMed]
37. Papes F, Logan DW, Stowers L. The vomeronasal organ mediates interspecies defensive behaviors through detection of protein pheromone homologs. Cell. 2010;141:692–703. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...