Archives of Insect Biochemistry and Physiology 10:273-279 (1 989) Pheromone Receptor Cells in the Male Moth Manduca sexta Karl-Ernst Kaissling, John G. Hildebrand, and James H. Tumlinson Max-Planck-lnstitut fur Verhaltensphysiologie,Seewiesen, Federal Republic of Germany K - E . K . ) ; Arizona Research Laboratories, Division of Neurobiology, University of Arizona, Tucson (J.G.H.); lnsect Attractants, Behavior, and Basic Biology Research Laboratoy, Agricultural Research Service, U.S. Department of Agriculture, Gainesville, Florida (1.H . T.) Three types of pheromone receptor cells have been identified by electrophysiological recording from single antennal sensilla trichodea of the male sphinx moth Manduca sexta. These cells responded best to the pheromone components (€,Z)-IO,I2-hexadecadienal(type A receptor cell), (€,€,Z)-10,12,14-hexadecatrienal (type B), and (€,€,€)-10,12,14-hexadecatrienal (type C). Cell type B also responded to (€,Z)-ll,13-pentadecadienal, which has been used experimentally as a pheromone substitute. In recordings from 20 trichoid hairs, 17 were found to be innervated by one cell of type A and one of type B; 3 trichoid hairs had cell types A and C. Key words: olfaction, sex pheromone, electrophysiology, antenna INTRODUCTION Several aspects of the development and function of the olfactory system, particularly with regard to perception of sex pheromone, have been studied extensively in the moth Munducu sexta (for reviews see [1-3])/ while only one component of the female’s sex-pheromone blend, (E,Z)-10,12-hexadecadienal (ElO,Z12-16:AL or bombykal), has been identified .Because ultrastructural observations had shown that the hairs of the sexually dimorphic sensillu tvichodeu on the antennal flagellum of male M. sexta contain two receptor-cell dendrites Acknowledgments: We thank Dr. R.E. Doolittle, Ms. A. Brabham, and Ms. M.M. Brennan of the USDA (Gainesville, FL) for chemical synthesis and preparation of several compounds used in this study; Drs. 1. Buckner and J. Svoboda of the USDA for generously supplying M. sexta eggs; W. Knauf of Hoechst AC (Frankfurt/Main, FRC) for kindly providing M. sexta pupae for some of these experiments; and Prof. H.J. Bestmann (Erlangen, FRC) for supplying EIO,Z12-16:AL and Ell,Z13-15:AL used in some of these studies. We are also grateful to R. Montague, P. Randolph, and D. Sakiestewa for assistance with insect rearing and to Drs. T.A. Christensen, E. Priesner, and T.R. Tobin for comments on the manuscript. This research was supported in part by NIH grants (to J.C.H.) Al-17711, Al-23253, and NS-23405. Received December 2,1988; accepted April 5,1989. Address reprint requests to Dr. JohnG. Hildebrand, ARL Division of Neurobiology, 611 CouldSimpson Building, University of Arizona, Tucson, AZ 85721. 0 1989 Alan R. Liss, Inc. 274 Kaissling et al. [5,6], it seemed likely that the female's pheromone blend would contain more than one physiologically active component. Moreover, behavioral observations also pointed to at least a second pheromone component in addition to E10,Z1216:AL [4,7]. Preliminary physiological experiments on trichoid hairs of the male antennal flagellum (Kaissling, unpublished) revealed one type of receptor cell that responded to ElO,Z12-16:AL and a second type sensitive to (E,Z)-11,13-pentadecadienal (Ell,Z13-15:AL). Although this pentadecadienal was much less effective than expected for an actual pheromone, both ElO,Z12-16:AL and Ell,Z13-15:AL (previously called "C1.5" and regarded as a pheromone mimic [8,9])have been used in studies of olfactory interneurons in the deutocerebrum of adult male M.sexta [8-lo]. A reinvestigation of the mixture of volatile substances in solvent rinses of the intact pheromone gland of the female moth revealed twelve CI6and CI8aldehydes (Table l), several or all of which might have pheromonal functions . The aim of our study was to test ElO,Z12-16:AL and two of the newly identified pheromone components, as well as Ell,Z13-15:AL, by a single-sensillum recording method in order to begin to characterize pheromone receptor cells in the antennae. Such information, together with knowledge about the roles of pheromone components in controlling the behavior of the male moth, is needed for ongoing studies of central processing of olfactory information in M.sexta [8-111. MATERIALS AND METHODS Manduca sexta (Lepidoptera, Sphingidae) were raised at the University of Arizona as described previously  or obtained as pupae from W. Knauf, Hoechst AG (Frankfurt/Main). A male moth (1-3 days posteclosion) was mounted with one antenna fixed TABLE 1. Compounds Identified in Solvent Rinses of Intact Pheromone Glands of Calling Virgin-FemaleMoths 171 Full name Blend component" Abbreviation (Z)-9-hexadecenal *(Z)-1I-hexadecenal (E)-11-hexadecenal (S)-16-hexadecanal *( E, Z)-lO,l2-hexadecadienal (E,E)-IO,12-hexadecadienal *(E, E, Z)-10,12,14-hexadecatrienal *( E,E,E)-10,12,14-hexadecatrienal (Z)-11-octadecenal *(Z)-13-octadecenal (S)-18-octadecanal (Z,Z)-ll,13-octadecadienal Z9-16:AL Zll-16:AL E11-16:AL S-16AL ElO,Z12-16:AL ElO,E12-16:AL ElO,El2,Z14-16:AL ElO,El2,E14-16:AL Z11-18:AL Z13-18:AL S-18:AL Zll,Z13-18:AL Approx. nglgland 0.8 13.4 6.8 15.7 23.8 3.9 ca.llb ca. 1 . 2 ~ 6.2 2.2 4.8 1.4 'Compounds tested in this study are indicated with an asterisk ("). bAlthough clearly present in the gland rinses, the trienals could not be reliably quantified owing to their instability during gas-liquid chromatography. The values given here are rough estimates . Pheromone Receptors in Manduca sexta 275 to a wire by bridges of paraffin wax. The tips of several trichoid hairs on the anterior half of one or more antennal annuli (i.e., closer to the leading edge of the annulus than to its trailing, scale-covered side as the antenna is held during flight) were clipped off. Electrical recording from an individual opened hair was accomplished with a glass capillary filled with sensillum lymph saline solution [ 121. The stimulus compounds (Table 1) were loaded on 1 cm2 pieces of filter paper in glass cartridges (7 mm diameter). The odor stimulus was delivered by passing an air current of 100 ml/s through the cartridge, which was positioned 5 cm from the antennal flagellum. To compare the effectiveness of the compounds, we used stimulus loads necessary to produce a standard response of 10-20 action potentials within the second following the first nerve impulse. The stimulus loads were prepared in decadic steps between lop4pg and 1pg, with intermediate steps with factors of 2or 3 in some cases. The stimulus compounds, summarized in Table 1, were prepared by J.H.T., R.E. Doolittle, A. Brabham, and M.M. Brennan (USDA, Gainesville, FL) [7,13]. ElO,Z12-16:AL and Ell,Z13-15:AL used in early experiments were kindly provided by H. Bestmann (Erlangen). RESULTS In each of 50 sensiZla trichodea tested, one receptor cell responded to E10,Z1216:AL at cartridge loads of 1ng. In one out of a total of eight animals, this cell produced action potentials with a much larger recorded amplitude than those of the second cell in the sensillum, which responded to Ell,Z13-15:AL (Fig. 1). In the other moths in this study, the receptor cells of a trichoid hair could not be distinguished on the basis of impulse amplitude or shape. b. E 10,E 12 , E 14- 16:A L , 2x 10 -3 y g ----LA-c. - , d. E10, E12,Zlh-l6:AL. Z X ~ O - ~ J J ~ I I,, -4- E l l , Z13-15:AL, lO-’pg I 1 mvl 1s Fig. 1. DC recordings from two receptor cellsof onesensillurn trichodeurn. Cell type A (larger nerve impulses) responds best to EIO,Z12-16:AL (a). Cell type B (smaller nerve impulses) responds best to E10,E12,Z14-16:AL (c) and i s also sensitive to Ell,Z13-15:AL (d). Neither receptor cell responds to E10,E12,E14-16:AL (b). Loads of stimulant substances on the filter papers are given in pg. Kaissling et al. 276 Experiments intended to discriminate between the two cells innervating the hair under study used the method of selective adaptation of one of the receptor cells within the sensillum . In this procedure, one of the two receptor cells was adapted in order to determine whether the other cell innervating the hair responded to a test compound. Such an experiment, involving pairs of successive stimuli each lasting 1 s, is illustrated in Figure 2. Records a, b, and c demonstrate that the nerve-impulse response of a cell adapted within about 1 s to a low level or to zero. Stimulation during the ensuing second with another pulse of the same stimulant evoked little or no response. A a . (1) EIO.Zl2-16:AL ( 2 ) same b. (1) ElO,E12,Z14-16:AL (21 same .-p----<*/- C. (1) Ell.Zl3-15:AL d. (1) ElO.Zl2-16:AL (2) ElO,E12,Z14-16:AL e. (1) f. (1) Ell, Z13-15:AL ( 2 ) E 10,E12,ZlL-l6:AL ( 2 ) Ell,Z13-15.AL 1. s t i m u l u s 1s 2. s t i m u l u s 1s Fig. 2. DC recordings from two receptor cells of o n e sensillum trichodeum demonstrating selective adaptation. The nerve impulses of both cells have the same amplitude. Two stimuli, each lasting 1 s, were presented in immediate succession (as indicated by the markers below the records). As described in Materials and Methods, the stimulus load for each compound was selected to produce a standard response, so that all stimuli were comparably effective. a-c: Double stimulation with t h e same compound reveals a reduced (adapted) response (i.e., fewer action potentials) during the second stimulus. The stimuli were a) EIO,Z12-16:AL, 1 0 ng, twice; b) E10,E12,Z14-16:AL, 1 0 ng, twice; c) Ell,Zl3-15:AL, 100 ng, twice. d,e: Unadapted response to the second stimulus suggests that each of the 2 compounds excites a separate receptor cell. The stimuli were d ) EIO,Z12-16:AL followed by E10,E12,Z14-16:AL; e) EIO,E12,Z1416:AL followed by EIO,Z12-16:AL. f,g: Reduced response to the second stimulus suggests that both compounds act o n t h e same receptor cell. The stimuli were f ) Ell,Z13-15:AL followed by E1O,E12,Z14-16: AL; g) E10,E12,Z14-16:AL followed by E l 1 ,Z13-I5 :AL. Pheromone Receptors in Manduca sexta 277 response evoked by the second stimulus of a pair (Fig. 2, records d and e) must have been generated by a separate cell that was not excited (and not adapted) by the first stimulus. Therefore, ElO,Z12-16:AL and ElO,E12,Z14-16:AL acted on different receptor cells. The second stimulus elicited no response if the two stimuli were ElO,E12,Z14-16:AL and EllrZ13-15:AL (Fig. 2, records f and g). Both of these compounds therefore must have acted on the same receptor cell. The theoretical possibility that we were recording from one sensory cell with two types of receptor sites seems to us most unlikely in view of the cited morphological findings and our observation of different impulse amplitudes in sensilla of one moth. In recordings from 20 trichoid hairs, three types of receptor cells could be distinguished, two per hair (Table 2). Each hair was innervated by a receptor cell that responded to E10rZ12-16:AL (cell type A). The second cell in 17 trichoid hairs responded to Ell,Z13-15:AL and, with much higher sensitivity, to ElO,El2,Z14-16:AL; the (E,E,E)-isomer(ElO,E12,E14-16:AL) was less effective (cell type B). In three of the 20 trichoid sensilla, the second receptor cell did not respond to Ell1Z13-15:AL and clearly responded better to E10,E12, E14-16:AL than to the (E,E,Z)-isomer (cell type C). Cells of types B and C appeared to be somewhat less sensitive than those of type A, as judged from the stimulus loads necessary to elicit the standard responses. DISCUSSION Each of the long olfactory hairs (sensilla trichodea) on the antenna1 flagella of adult male M . sexta contains the distal dendritic segments of two receptor cells [5,6]. In this study, we found that about 50% of these receptor cells are sensitive to ElO,Z12-16:AL (type A cells), and a majority of the rest are tuned to ElO,El2,Z14-16:AL (type B cells). Both of these substances are major components of the blend found in solvent rinses of the female pheromone gland, and they are the most important components for eliciting characteristic pheromone-dependent behavior in male moths . In fact, in wind-tunnel experiments a mixture of E10jZ12-16:AL and E10,E12,Z14-16:ALr at a concentration of 0.02 female gland equivalents, had the same effects as the complete synthetic blend or the gland rinse . Because receptors tuned to E1OTE12,El4-16:AL are present in the antenna, however, and in view of the fact that intracellular TABLE 2. Types of Receptors Cells in Trichoid Sensilla Stimulus compounds ElO,Z12-16:AL ElO,E12,Zl4-16:AL ElO,El2,E14-16:AL Ell,Z13-15:AL A (n = 20) 0.001 nr nr nrc Receptor cell types" B (n = 17) nrb o.003 0.03 0.1 C (n = 3) nr 0.03 o.003 nrc "Stimulus loads (pg) in cartridges eliciting a standard response of 10-20 nerve impulses within 1 s following the first impulse. The figures for the most effective compounds for these cell types are underlined. bNo response was observed up to cartridge loads of 0.1 pg (nr) or 1 pg (nrc). 278 Kaisslinget al. recordings from central olfactory interneurons reveal responses to that substance [lo], it seems likely that the (E,E,E) isomer has subtle effects on behavior that have not yet been recognized. The most effective stimulants for the central neurons were ElO,Z12-16:AL, ElO,E12,Z14-16:AL, and E10rE12,E14-16:AL. It remains to be shown whether other components of the female’s pheromone blend are perceived by the male moth. Preliminary tests showed no effects of (Z)-11-hexadecenaland (2)-13octadecenal on the three types of receptor cells described in this paper. A recent study of central processing of pheromonal information in M. sexta using all 12 components of the blend indicated that only eight of those substances elicited recognizable responses in neurons in the antennal lobe of the brain [lo]. The findings reported here raise challenging questions about the development of sensilla, as have similar observations in other species of moths. One wonders how the established pattern of differentiative cell divisions that give rise to the trichoid sensillum  generates different sets of receptor cells in different trichoid sensilla. Thus, it would be interesting to learn what developmental mechanism accounts for the formation of the relatively few sensilla in the male antenna that contain a receptor cell (type C cell) tuned to the minor pheromone ElO,E12,E14-16:AL. Approximate correspondence between the proportion of a pheromone in the female’s blend and the frequency of occurrence in the male’s antenna of receptor cells specialized to detect that substance has been described in other species, for example for the third pheromone component of Antherueu pernyi . Meng et al.  discussed the idea that the relative rate of release of a pheromone compound from the female gland, the relative number of corresponding receptor cells, and the sensitivity of those cells determine the potential distance from the female over which the compound can be detected by the male. It can be expected from our results that this potential threshold volume [161 differs for the various pheromone components of M. sextu as well. LITERATURE CITED 1. Hildebrand JG: Metamorphosis of the insect nervous system: Influences of the periphery on the postembryonic development of the antennal sensory pathway in the brain of Manduca sexta. In: Model Neural Networks and Behavior. Selverston A, ed. Plenum, New York, pp 129-148 (1985). 2. Christensen TA, Hildebrand JG: Functions, organization, and physiology of the olfactory pathways in the lepidopteran brain. In: Arthropod Brain: Its Evolution, Development, Structure and Functions. Gupta AP, ed. John Wiley, New York, pp 457-484 (1987). 3. Homberg U, Christensen TA, Hildebrand JG: Structure and function of the deutocerebrum in insects. Annu Rev Entomol34,477 (1989). 4. Starratt AM, Dahm KH, Allen N, Hildebrand JG, Payne TL, Roller H: Bombykal, a sex pheromone of the sphinx moth Manduca sexta. Z Naturforsch 34C, 9 (1979). 5. Sanes JR, Hildebrand JG: Origin and morphogenesis of sensory neurons in an insect antenna. Dev Biol51,300 (1976). 6. Keil T: Fine structure of the pheromone-sensitive sensilla on the antenna of the hawkmoth, Manduca sexta. Tissue Cell, 21,139 (1989). 7. Tumlinson JH, Brennan MM, Doolittle RE, Mitchell ER, Brabham A, Mazomenos BE, Baumhover AH, Jackson DM: Identification of a pheromone blend attractive to Manduca sexta (L.) males in a wind tunnel. Arch Insect Biochem Physiol10,255 (1989). Pheromone Receptors in Manduca sexta 279 8. Christensen TA, Hildebrand JG: Male-specific, sex pheromone-selective projection neurons in the antennal lobes of the moth Munducu sextu. J Comp Physiol [A] 160,553 (1987). 9. Christensen TA, Hildebrand JG: Frequency coding by central olfactory neurons in the sphinx moth Munducu sexta. Chem Senses 13, 123 (1988). 10. Christensen TA, Hildebrand JG, Tumlinson JH, Doolittle RE: Sex pheromone blend of Mundum sextu: Responses of central olfactory interneurons to antennal stimulation in male moths. Arch Insect Biochem Physiol 10,281 (1989). 11. Waldrop B, Christensen TA, Hildebrand JG: GABA-mediated synaptic inhibition of projection neurons in the antennal lobes of the sphinx moth Munducu sextu. J Comp Physiol [A] 261,23 (1987). 12. Kaissling K-E, Thorson J: Insect olfactory sensilla: Structural, chemical and electrical aspects of the functional organization. In: Receptors for Neurotransmitters, Hormones and Pheromones in Insects. Sattelle DB, Hall LM, Hildebrand JG, eds. ElsevieriNorth Holland, Amsterdam, pp 261-282 (1980). 13. Doolittle RE, Brabham A, Tumlinson JH: Sex pheromone of Munducu sextu (L.): Stereoselective synthesis of (E,E,Z)-10,12,14-hexadecatrienal and isomers. J Chem Ecol in press, (1989). 14. Kaissling K-E, Zack Strausfeld C, Rumbo ER: Adaptation processes in insect olfactory receptors: Mechanisms and behavioral significance. Ann NY Acad Sci 510,104 (1987). 15. Meng LZ, Wu CH, Wicklein M, Kaissling K-E, Bestmann HJ: Number and sensitivity of three types of pheromone receptor cells in Antherueu pernyi and A. polyphemus. J Comp Physiol in press, (1989). 16. Bossert WH, Wilson EO: The analysis of olfactory communication among animals. J Theor Biol5,443 (1963).