European Journal of Neuroscience European Journal of Neuroscience, Vol. 37, pp. 572–582, 2013 doi:10.1111/ejn.12066 NEUROSYSTEMS Neuropeptide receptors provide a signalling pathway for trigeminal modulation of olfactory transduction Philipp Daiber,1 Federica Genovese,1 Valentin A. Schriever,2 Thomas Hummel,2 Frank Mohrlen€ 1 and Stephan Frings1 1Department of Molecular Physiology, Centre for Organismal Studies, Heidelberg University, Heidelberg, Germany 2Interdisciplinary Centre for Smell & Taste, University Hospital Dresden, Dresden, Germany Keywords: bimodal, CGRP, irritants, odorants Abstract The mammalian olfactory epithelium contains olfactory receptor neurons and trigeminal sensory endings. The former mediate odor detection, the latter the detection of irritants. The two apparently parallel chemosensory systems are in reality interdependent in various well-documented ways. Psychophysical studies have shown that virtually all odorants can act as irritants, and that most irritants have an odor. Thus, the sensory perception of odorants and irritants is based on simultaneous input from the two sys- tems. Moreover, functional interactions between the olfactory system and the trigeminal system exist on both peripheral and cen- tral levels. Here we examine the impact of trigeminal stimulation on the odor response of olfactory receptor neurons. Using an odorant with low trigeminal potency (phenylethyl alcohol) and a non-odorous irritant (CO2), we have explored this interaction in psychophysical experiments with human subjects and in electroolfactogram (EOG) recordings from rats. We have demonstrated that simultaneous activation of the trigeminal system attenuates the perception of odor intensity and distorts the EOG response. On the molecular level, we have identified a route for this cross-modal interaction. The neuropeptide calcitonin-gene related pep- tide (CGRP), which is released from trigeminal sensory fibres upon irritant stimulation, inhibits the odor response of olfactory receptor neurons. CGRP receptors expressed by these neurons mediate this neuromodulatory effect. This study demonstrates a site of trigeminal–olfactory interaction in the periphery. It reveals a pathway for trigeminal impact on olfactory signal processing that influences odor perception. Introduction The mammalian nose harbours a set of distinct chemosensory organs uli with low, if any, trigeminal potency (examples are vanillin, H2S that differ in their chemosensory receptors, their specific wiring in and phenylethyl alcohol), and the only known irritant with little or the brain and their role in the animal’s responses to environmental no detectable olfactory percept in humans is CO2 (Bensafi et al., cues. In rodents, five such organs can be distinguished: the main 2008). Thus it is highly likely that each sniff elicits a bimodal neu- olfactory epithelium, the trigeminal system, the vomeronasal organ, ronal response: activation of olfactory receptor neurons and, at the the septal organ and the Grueneberg ganglion (Ma, 2007, 2010; same time, activation of trigeminal sensory fibres. Importantly, the Munger et al., 2009; Dauner et al., 2012). The human nose, how- concomitant trigeminal activity detectably alters odorant perception ever, appears to operate with only two systems, the main olfactory (Kobal & Hummel, 1988; Cometto-Muniz & Hernandez, 1990; epithelium and the trigeminal system. Although these two systems Brand, 2006). Key features of olfactory performance, including sen- are generally held to be responsible for the separate sensory modali- sitivity and odor discrimination, can be modified by co-stimulation ties, one mediating olfaction and the other nociception, psychophysi- of trigeminal sensory fibres (Jacquot et al., 2004). Interaction cal and electrophysiological studies have clearly demonstrated that between the two sensory systems is thus a relevant factor in olfac- the intimate intertwining of the two systems precludes this distinc- tory performance. tion. In fact, practically every odorant co-stimulates the trigeminal At which sites and through which pathways trigeminal activity system and virtually all irritants co-stimulate the main olfactory epi- impacts on the olfactory system is currently not understood. There thelium (Cain, 1977; Doty et al., 1978; Cain & Murphy, 1980; Liv- is evidence that interaction occurs both in the olfactory epithelium ermore et al., 1992; Silver, 1992; Hummel & Livermore, 2002; (Kratskin et al., 2000) and in various brain regions including the Brand, 2006). Only a very few odorants are selective olfactory stim- olfactory bulb (Schaefer et al., 2002) and, possibly, in the medio- dorsal thalamic nucleus, where the two systems converge (Inokuchi et al., 1993; Brand, 2006). Earlier recordings from the frog olfactory epithelium revealed that odor-induced field potentials [electroolfacto- Correspondence: Professor Dr S. Frings, as above. grams (EOGs)] were altered when the ophthalmic branch of the tri- E-mail: [email protected] geminal nerve was stimulated antidromically, an effect that was Received 31 July 2012, revised 23 October 2012, accepted 24 October 2012 partially recapitulated by application of the trigeminal neuropeptide © 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd Trigeminal modulation of olfactory receptor cells 573 substance P to the epithelium (Bouvet et al., 1987, 1988). This Table 1. Air-phase concentrations of odorants in the odor tubes observation suggests a neuromodulatory role for trigeminal fibres that appears to be mediated by neuropeptides and targeted at the Molarity (M) PEA conc. (ppm) IAA conc. (ppm) AIC conc. (ppm) olfactory receptor neurons. Release of trigeminal neuropeptides dur- À À À 0.0003 0.0228 9 10 3 2.54 9 10 3 2.37 9 10 3 ing a sniff may, therefore, alter the response characteristics of olfac- 0.001 0.0759 9 10À3 8.46 9 10À3 7.91 9 10À3 tory receptor neurons. 0.003 0.228 9 10À3 0.0254 0.0237 Here we have examined this hypothesis by investigating the sen- 0.01 0.760 9 10À3 0.0847 0.0791 À3 sory response to 2-phenylethyl alcohol (PEA), a floral odorant with 0.03 2.28 9 10 0.254 0.228 9 À3 particularly weak trigeminal potency. We have also studied its mod- 0.1 7.64 10 0.852 0.795 0.3 0.0232 2.59 2.41 ulation by the trigeminal agonist CO2 in human subjects. For com- 1 0.0815 9.09 8.33 parison, we measured PEA responses in rat olfactory epithelium. 3 0.288 32.1 28.0 Because of the pronounced CO2 sensitivity of the rat chemosensory 6 0.782 87.2 68.1 system we used the trigeminal neuropeptide calcitonin-gene related Undiluted 1.61 180 168 peptide (CGRP) to examine trigeminal–olfactory interactions. We Calculated concentrations (in ppm) of phenylethyl alcohol (PEA), isoamylac- explored the trigeminal innervation of the olfactory epithelium and etate (IAA) and allylisothiocyanate (AIC) in the gas phase of odor tubes. the expression of CGRP receptors in this tissue. Our studies provide Stock solutions, prepared at the indicated concentrations (in M) in mineral a concept for trigeminal modulation of the primary olfactory signal oil, were added to filter paper within the odor tube at room temperature. Val- and its consequence for odor perception. ues of ppm were calculated for 20 °C from the vapor pressure and the molar fraction of the two fluid components. Materials and methods 5; CaCl2, 1; MgCl2, 1; HEPES, 10; glucose, 10; and pyruvate, 1; pH All experiments with human subjects followed the Declaration of adjusted to 7.4 with NaOH. The reference air flow was deodorised, Helsinki on biomedical research involving human subjects and was humidified and adjusted to 0.1 L/min. Recording electrodes were approved by the Ethics Committee from the University of Dresden pulled to a tip aperture of 20–25 lm from borosilicate glass capillaries Medical School (EK332092011, EK118072003). All participants (OD 1.5 mm, ID 0.87 mm) using a Flaming–Brown puller (Sutter provided written informed consent. Experiments with animal tissues Instruments, Novato, CA, USA) and filled with Ringer’s solution. The were performed in accordance with the Animal Protection Law and local surface potential of the olfactory epithelium was amplified (DP- the guidelines and permissions of Heidelberg University. 301; Warner Instruments), digitised (BNC 2120; National Instru- ments) and processed using the WINWCP software provided by Strath- clyde University, UK. Odorants were dissolved in mineral oil Psychophysical investigations (BioUltra, Sigma 697934; Sigma) at the concentrations listed in Forty healthy subjects, 29 female and 11 male, age range 20– Table 1 in M. Forty microlitres of each solution was placed inside an 32 years (mean Æ SD, 24.95 Æ 3.13 years) participated in this odor tube from which the odorant could be injected under computer study. All subjects were normosmic as ascertained using the 16-item control into the air stream using a pneumatic pico pump (PV830; odor identification test from the Sniffin’ Sticks test kit (mean score, WPI, Sarasota, FL, USA). For dose–response experiments, increasing 14.06 Æ 1.08; score range, 13–16; Hummel et al., 1997). Stimuli concentrations were applied at 2-min intervals. Odor concentrations in were presented monorhinally using a computer-controlled olfactome- the air space of the odor tubes were calculated from the respective ter (Olfactometer OM6b; Burghart, Wedel, Germany) with an air- vapour pressures according to Table 1. No attempt was made to esti- flow of 12 L/min. PEA was used for olfactory stimulation and CO2 mate odor concentrations at the sensory surface. for trigeminal stimulation. The concentrations for both stimuli ran- Submerged EOGs were recorded from similar preparations. A ged from 5 to 20% v/v. The stimulus duration for PEA was set to constant flow of Ringer’s solution (in mM: NaCl, 120; NaHCO3, 25; 200 ms. For CO2 stimulus, durations of 200, 1000, 2000 and KCl, 5; BES, 5; MgSO4, 1; CaCl2, 1; and glucose, 10; pH 7.4) was 3000 ms were used. The inter-stimulus interval ranged between 27 applied flowing over all olfactory turbinates in the caudal direction. and 33 s. The two stimuli were presented simultaneously. The study Recording electrodes were pulled from capillaries (GB150-10; Sci- was divided into sessions of 12–15 min each.