Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

The pheromone darcin drives a circuit for innate and reinforced behaviours

Abstract

Organisms have evolved diverse behavioural strategies that enhance the likelihood of encountering and assessing mates1. Many species use pheromones to communicate information about the location, sexual and social status of potential partners2. In mice, the major urinary protein darcin—which is present in the urine of males—provides a component of a scent mark that elicits approach by females and drives learning3,4. Here we show that darcin elicits a complex and variable behavioural repertoire that consists of attraction, ultrasonic vocalization and urinary scent marking, and also serves as a reinforcer in learning paradigms. We identify a genetically determined circuit—extending from the accessory olfactory bulb to the posterior medial amygdala—that is necessary for all behavioural responses to darcin. Moreover, optical activation of darcin-responsive neurons in the medial amygdala induces both the innate and the conditioned behaviours elicited by the pheromone. These neurons define a topographically segregated population that expresses neuronal nitric oxide synthase. We suggest that this darcin-activated neural circuit integrates pheromonal information with internal state to elicit both variable innate behaviours and reinforced behaviours that may promote mate encounters and mate selection.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Fig. 1: Darcin elicits an array of behaviours.
Fig. 2: Optogenetic silencing of the AOB results in the suppression of darcin-evoked behaviours.
Fig. 3: Activation of darcin-responsive neurons in the MeA recapitulates pheromone-induced behaviours.
Fig. 4: nNOS neurons in the MeA are necessary for darcin-mediated behaviours.

Data availability

The datasets generated and/or analysed during the current study are available from the corresponding authors on reasonable request.

References

  1. 1.

    Malte, A. Sexual Selection (Princeton Univ. Press, 1994).

  2. 2.

    Wyatt, D. T. Pheromones and Animal Behavior: Chemical Signals and Signatures (Cambridge Univ. Press, 2014).

  3. 3.

    Roberts, S. A. et al. Darcin: a male pheromone that stimulates female memory and sexual attraction to an individual male’s odour. BMC Biol. 8, 75 (2010).

    Article  Google Scholar 

  4. 4.

    Roberts, S. A., Davidson, A. J., McLean, L., Beynon, R. J. & Hurst, J. L. Pheromonal induction of spatial learning in mice. Science 338, 1462–1465 (2012).

    ADS  CAS  Article  Google Scholar 

  5. 5.

    Roberts, S. A., Davidson, A. J., Beynon, R. J. & Hurst, J. L. Female attraction to male scent and associative learning: The house mouse as a mammalian model. Anim. Behav. 97, 313–321 (2014).

    Article  Google Scholar 

  6. 6.

    Kaur, A. W. et al. Murine pheromone proteins constitute a context-dependent combinatorial code governing multiple social behaviors. Cell 157, 676–688 (2014).

    CAS  Article  Google Scholar 

  7. 7.

    Schultz, W. Neuronal reward and decision signals: from theories to data. Physiol. Rev. 95, 853–951 (2015).

    CAS  Article  Google Scholar 

  8. 8.

    Halpern, M. & Martínez-Marcos, A. Structure and function of the vomeronasal system: an update. Prog. Neurobiol. 70, 245–318 (2003).

    CAS  Article  Google Scholar 

  9. 9.

    Dulac, C. & Wagner, S. Genetic analysis of brain circuits underlying pheromone signaling. Annu. Rev. Genet. 40, 449–467 (2006).

    CAS  Article  Google Scholar 

  10. 10.

    Gradinaru, V., Thompson, K. R. & Deisseroth, K. eNpHR: a Natronomonas halorhodopsin enhanced for optogenetic applications. Brain Cell Biol. 36, 129–139 (2008).

    Article  Google Scholar 

  11. 11.

    Boyden, E. S., Zhang, F., Bamberg, E., Nagel, G. & Deisseroth, K. Millisecond-timescale, genetically targeted optical control of neural activity. Nat. Neurosci. 8, 1263–1268 (2005).

    CAS  Article  Google Scholar 

  12. 12.

    Root, C. M., Denny, C. A., Hen, R. & Axel, R. The participation of cortical amygdala in innate, odour-driven behaviour. Nature 515, 269–273 (2014).

    ADS  CAS  Article  Google Scholar 

  13. 13.

    Martín-Sánchez, A. et al. From sexual attraction to maternal aggression: when pheromones change their behavioural significance. Horm. Behav. 68, 65–76 (2015).

    Article  Google Scholar 

  14. 14.

    Kimoto, H., Haga, S., Sato, K. & Touhara, K. Sex-specific peptides from exocrine glands stimulate mouse vomeronasal sensory neurons. Nature 437, 898–901 (2005).

    ADS  CAS  Article  Google Scholar 

  15. 15.

    Papes, F., Logan, D. W. & Stowers, L. The vomeronasal organ mediates interspecies defensive behaviors through detection of protein pheromone homologs. Cell 141, 692–703 (2010).

    CAS  Article  Google Scholar 

  16. 16.

    National Research Council (US) Committee. Guide for the Care and Use of Laboratory Animals (National Academies Press, 2011).

  17. 17.

    Champlin, A. K. Suppression of oestrus in grouped mice: the effects of various densities and the possible nature of the stimulus. J. Reprod. Fertil. 27, 233–241 (1971).

    CAS  Article  Google Scholar 

  18. 18.

    Byers, S. L., Wiles, M. V., Dunn, S. L. & Taft, R. A. Mouse estrous cycle identification tool and images. PLoS ONE 7, e35538 (2012).

    ADS  CAS  Article  Google Scholar 

  19. 19.

    Tchernichovski, O., Nottebohm, F., Ho, C. E., Pesaran, B. & Mitra, P. P. A procedure for an automated measurement of song similarity. Anim. Behav. 59, 1167–1176 (2000).

    CAS  Article  Google Scholar 

  20. 20.

    Van Segbroeck, M., Knoll, A. T., Levitt, P. & Narayanan, S. MUPET—Mouse Ultrasonic Profile ExTraction: a signal processing tool for rapid and unsupervised analysis of ultrasonic vocalizations. Neuron 94, 465–485.e5 (2017).

    Article  Google Scholar 

  21. 21.

    Sanders, J. I. & Kepecs, A. A low-cost programmable pulse generator for physiology and behavior. Front. Neuroeng. 7, 43 (2014).

    Article  Google Scholar 

  22. 22.

    Grimm, D. et al. In vitro and in vivo gene therapy vector evolution via multispecies interbreeding and retargeting of adeno-associated viruses. J. Virol. 82, 5887–5911 (2008).

    CAS  Article  Google Scholar 

Download references

Acknowledgements

We thank Z. J. Huang, R. Paik, H. Taniguchi, G. Enikopolov, S. P. Ranade and D. Kvitsiani for discussions; L. McLean for assistance with recombinant proteins; R. Eifert, B. Burbach and R. Specht for technical support; S. Brenner-Morton for generating the guinea pig Fos antibody; K. Chatpar and Y. Sun for assistance with experiments; N. Zabello for help with mice; L. Stowers for the cat lipocalin (Fel-D4) plasmid; C. Denny for the gift of the Arc-CreER mouse; D. Hattori, J. Scribner, A. S. Lee and B. Noro for comments on the manuscript; and C. H. Eccard, P. Kisloff, A. Nemes and M. Gutierrez for laboratory support. This work was supported by the Howard Hughes Medical Institute (R.A.), the Biotechnology and Biological Sciences Research Council (J.L.H. and R.J.B.), and The Robert E. Leet and Clara Guthrie Patterson Trust Fellowship (E.D.).

Author information

Affiliations

Authors

Contributions

E.D., R.J.B., J.L.H., A.K. and R.A. discussed the design of experiments and the results, and wrote the manuscript. J.I.S. designed the custom behaviour and stimulation systems. E.D. performed all of the experiments and analysis. K.L. and N.B.-K. helped with the experiments and analysis. The recombinant MUPs were provided by R.J.B.

Corresponding authors

Correspondence to Adam Kepecs or Richard Axel.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature thanks Stephen Liberles and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Darcin and photoactivation of posterior MeA neurons condition scent-marking place preference.

a, Representative frames from videos of the pheromone (1) and photoactivation (2) sessions, and free-range behaviours (3). b, Distance from urinary drop to each of the poke ports during various sessions. Individual frames were analysed using Adobe Photoshop CC to quantify the distance from the centre of a urinary drop to the base of each poke port. Units are scaled from pixels to centimetres. Distances were compared using the two-sided Wilcoxon signed-rank test (***P < 0.0005, *P = 0.01; n = 24 mice, (1); n = 12 mice, (2); n = 20 mice, (3)). c, Area of urinary drops under various conditions. Individual frames were analysed using Adobe Photoshop CC to quantify the area of the urinary marks. Units are scaled from square pixels to square centimetres. Scent-mark area (mean ± s.e.m., cm2): darcin, 5 ± 0.05, n = 24 mice; recall of darcin, 5 ± 0.09, n = 14 mice; photoactivation, 4 ± 0.4, n = 12 mice; recall of photoactivation, 4 ± 0.5, n = 8 mice; free urination, 13 ± 2; n = 20 mice. Areas were compared using the two-sided Mann–Whitney test (***P < 0.0005), adjusted for multiple comparisons. The bounds in the box plots in b, c are defined by the 25th and 75th percentile of the distribution. The lines represent the median and the upper and lower whiskers represent the 75th percentile + 1.5 × IQR and 25th percentile − 1.5 × IQR, respectively.

Extended Data Fig. 2 Darcin and photoactivation of posterior MeA neurons reinforce recall of vocalization and scent-marking behaviours.

ac, Data for individual mice for all unique sessions across the study were pooled. a, Mean (horizontal line; n = 43 mice (darcin group), n = 24 mice (photostimulation group)) and total calls made by individual mice (diamonds) detected during various sessions. Call counts were compared using the two-sided Wilcoxon signed-rank test within the respective groups (***P < 0.0005), adjusted for multiple comparisons. b, Latency from the start of the session to urinary marking and vocalization behaviour (mean ± s.e.m., seconds) during exposure to darcin (3,160 ± 311, n = 24 mice), recall of darcin exposure (956 ± 217, n = 14 mice), photostimulation (4,195 ± 372, n = 12 mice) and subsequent recall (1,315 ± 418, n = 8 mice) sessions. Latencies were compared within groups using the matched-pair two-sided t-test (*P = 0.005, ***P = 0.00009). The bounds in the boxplots are defined by the 25th and 75th percentile of the distribution. The line represents the median and the upper and lower whiskers represent 75th percentile + 1.5 × IQR and 25th percentile − 1.5 × IQR, respectively. c, Probability of urinary scent-marking and vocalization behaviours. Mean probabilities are given for the darcin session (0.6, n = 43 mice), recall of darcin session (0.3, n = 43 mice), photostimulation-evoked urinary marking and vocalization (0.5, n = 24 mice) and recall of photostimulation-evoked behaviours (0.3, n = 24). Probabilities were compared using the two-sided McNemar test (*P < 0.05). d, Probability and mean latency to first urinary scent marking in the different sessions (n = 9 mice). Data from 100-min habituation sessions (mean ± s.e.m., latency for urination, seconds) and after exposure to male-soiled bedding in the home cage (1,411 ± 126), low-darcin urine from BALB/c mice (1,116 ± 232), recall session of BALB/c urine (1,607 ± 268), urine from C57BL6/6J mice containing normal levels of darcin (2,666 ± 337) and recall of C57BL6/6J urine (1,032 ± 198) are shown. Probabilities were compared using the two-sided McNemar test (*P = 0.02), adjusted for multiple comparisons. Latencies were compared within groups using the matched-pair two-sided t-test and across groups using the unpaired two-sided t-test (**P = 0.0008, *P = 0.02), adjusted for multiple comparisons. Scent-marking behaviours in response to low-darcin urine during the subsequent recall sessions were compared (habituation to recall, P = 1, cue to recall session comparison, P = 0.1, two-sided McNemar test).

Extended Data Fig. 3 Activation of darcin-responsive neurons in the posterior MeA recapitulates darcin-induced behaviours.

a, Heat map showing occupancy of the chamber during a habituation, photostimulation and recall session. b, Occupancy plot showing the percentage of time spent in the photostimulation room. Arc-CreER mice were exposed to darcin (magenta), saline (green) or MUP11 (blue). The plot shows the mean ± s.e.m. (n = 5 mice per group, total n = 15 mice) percentage of time spent in stimulation room during habituation, photostimulation and recall sessions. For occupancy time, pairwise comparisons were performed using the two-sided Mann–Whitney test (*P < 0.05) and three-way comparisons were performed using Kruskal–Wallis tests (habituation P = 0.6, light stimulation P = 0.009 and recall sessions P = 0.008). cf, Activation of nNOS neurons in the posterior MeA recapitulates darcin-induced behaviours. c, d, Cumulative poke counts during habituation (laser off; 1), light stimulation (laser on; 2), and recall (laser off; 3) sessions in mice expressing eYFP (c) or ChR2 (d) in nNOS neurons. Light stimulation was performed in one port (red) and not in the second port (blue). During habituation (1) and recall (3) sessions, no light stimulation was given, and red and blue reflect right and left ports, respectively. Mean (bold lines, n = 11 mice for each group) and individual (fine lines) cumulative poke counts are shown. The time-stamps for ultrasonic vocalization and scent-marking behaviours are indicated as arrowheads (d (2, 3)). Poke counts were compared using the two-sided Wilcoxon signed-rank test (***P = 0.0001). Control group (eYFP) port entries (c) are contrasted to the ChR2 group (d) during light stimulation (red entries for ChR2 (d (2)) compared to eYFP (c (2)); P = 0.0002) and recall sessions (red entries for ChR2 (d (3)) compared to eYFP (c (3)); P = 0.0002, two-sided Mann–Whitney test, adjusted for multiple comparisons). e, Occupancy plot showing the mean percentage of time spent in the photostimulation room by all mice during various sessions. nNOS-ires-Cre mice were injected with AAV encoding either eYFP (green) or ChR2–eYFP (purple); plots are colour-coded to their respective groups; n = 6 mice per group, n = 12 mice total. Occupancy times were compared using a two-sided Mann–Whitney test (*P < 0.05). f, Mean (horizontal lines, n = 11 per group, n = 22 total) and total calls made by individual mice (diamonds) detected during the photostimulation (2) sessions in mice expressing eYFP (c (2)) or ChR2 (d (2)) in nNOS neurons. Call counts were compared using the two-sided Mann–Whitney test (*P = 0.007).

Extended Data Fig. 4 In lactating females, darcin activates mitral cells in the AOB but fails to activate MeA neurons.

ac, Representative images showing Fos expression (orange) and NeuroTrace (blue) in sagittal sections of the AOB following exposure to saline (a) or darcin in virgin females (b) and lactating females (c). Experiment was independently repeated on 6 mice for each group. d, Bar plots quantifying Fos-expressing cells in the AOB. Fos counts (mean ± s.e.m.): saline 153 ± 38, darcin in virgin females 378 ± 35, darcin in lactating females 358 ± 45; n = 6 mice per group. Cell counts were compared using the two-sided Mann–Whitney test (*P = 0.02), adjusted for multiple comparisons. e, Bar plots quantifying the mitral/tufted cells in the AOB. Number of cells (mean ± s.e.m.): saline 1,188 ± 167, darcin in virgin females 1,129 ± 93, darcin in lactating females 1,210 ± 163; n = 6 mice per group. Cell counts were compared using the two-sided Mann–Whitney test. f, g, Representative images showing eYFP expression in coronal sections of the posterior MeA of Arc-CreER mice after exposure to darcin in virgin females (f) and lactating females (g). Experiment was repeated on 13 mice and 4 mice in f and g, respectively. h, Bar plots quantifying eYFP-expressing cells in the MeApd and the MeApv. Cell counts were compared using the two-sided Mann–Whitney test, adjusted for multiple comparisons. *P = 0.008, **P = 0.0006, ***P < 0.0005. Mean ± s.e.m. eYFP-expressing cell counts: saline, 16 ± 5 in the MeApd and 23 ± 7 in the MeApv, n = 13 mice; darcin exposure in virgin females, 251 ± 29 in the MeApd and 115 ± 16 in the MeApv, n = 13 mice; darcin exposure in lactating females, 23 ± 11 in the MeApd and 15 ± 12 in the MeApv, n = 4 mice.

Extended Data Fig. 5 Identification of neurons in the posterior MeA that respond to vomeronasal stimuli and their overlap with the genetic marker nNOS.

a, Representative images showing the stimulus-responsive (eYFP, orange) and nNOS-expressing (cyan) neurons in the posterior MeA of Arc-CreER mice exposed to cat salivary lipocalin Fel-D4 (n = 5 mice), saline (n = 8 mice), ESP1 (n = 5 mice), MUP11 (n = 5 mice), female urine (n = 5 mice), male urine with low levels of darcin (n = 4 mice), male urine with normal levels of darcin (n = 9 mice) and darcin (n = 7 mice). b, Corresponding box plots quantifying the percentage overlaps between the stimulus-responsive (eYFP) and nNOS+ neurons in the posterior MeA of mice exposed to the various stimuli. Orange plots represent the percentage of YFP cells that overlap with nNOS; cyan plots represent the percentage of nNOS cells that overlap with YFP. The bounds in box plots are defined by the 25th and 75th percentile of the distribution. The lines represent the median and the upper and lower whiskers represent the 75th percentile + 1.5 × IQR and 25th percentile − 1.5 × IQR, respectively.

Extended Data Fig. 6 The additional effects of silencing nNOS neurons in the posterior MeA.

ac, Functional convergence of both olfactory systems mediated by the posterior MeA is pivotal for male urine reinforcement. a, Timeline of the preference assay. Mice were habituated in the chamber for ten days (1), then exposed to male-soiled bedding for 60 h in their home cage (2), followed by one additional day of habituation before male urine (with normal levels of darcin (1 μg μl−1)) was presented in one of the two ports (3). Urine was removed for the recall session one day later (4). Port preference was quantified from port entries. b, c, Cumulative poke counts during habituation (1), habituation after treatment (2), exposure to male urine (3) and recall (4) sessions for mice expressing eNpHR–eYFP (n = 10) with (b) and without (c) optical silencing of nNOS neurons. Poke counts were recorded on the days indicated by purple arrows in a. Mice were exposed to male urine in one port (red) and a blank filter (blue) in the second port (3). During habituation (1, 2) and recall (4) sessions both ports contained a blank filter. Mean (bold lines) and individual (fine lines) cumulative poke counts are shown. Poke counts were compared using the two-sided Wilcoxon signed-rank test (***P = 0.0002). The effect of silencing nNOS neurons is quantified with matched pair differences (male urine session comparisons, b (3) to c (3), P = 0.002) and recall of male urine with darcin (recall session comparisons, b (4) to c (4), P = 0.002) using the two-sided Wilcoxon signed-rank test, adjusted for multiple comparisons. d, e, Optical silencing of nNOS neurons does not affect recall of darcin memory. Cumulative poke counts during habituation (1), habituation after treatment (2), darcin (3) and recall (4) sessions in mice expressing eNpHR (n = 11) with optical silencing during all sessions (d (1–4)) and with optical silencing during recall sessions only (e (4)). Poke counts were recorded on the days indicated by purple arrows in a. Mice were exposed to darcin in one port (red) and a blank filter (blue) in the second port (3). During habituation (1, 2) and recall (4) sessions both ports contained a blank filter. Mean (bold lines) and individual (fine lines) cumulative poke counts are shown. Poke counts were compared using the two-sided Wilcoxon signed-rank test (**P = 0.001). The effect of silencing nNOS neurons during recall sessions was tested with matched pair differences (c, cue (e (3)) to recall (e (4)) comparisons, laser off (e (3)) and on (e (4)), P = 0.1, using the two-sided Wilcoxon signed-rank test, adjusted for multiple comparisons.

Extended Data Fig. 7 Mice subjected to optical silencing of nNOS neurons retained a motivation to poke.

To establish the primacy of the MeA in mediating darcin-evoked behaviours rather than altering general motivation, mice expressing eNpHR in nNOS neurons were also tested. a, Timeline of the two-port preference assay. bd, Cumulative poke counts during habituation (1), habituation after exposure to male-soiled bedding in the home cage (2), darcin exposure (3) and recall (4) sessions with (b) and without (c) optical silencing of nNOS neurons, and with optical silencing again after 4 weeks (d) (n = 11 mice). Poke counts were recorded on the days indicated by purple arrows in a. Mice were exposed to darcin in one port (red) and a blank filter (blue) in the second port. During habituation (1, 2) and recall (4) sessions both ports contained a blank filter. Mean (bold lines) and individual (fine lines) cumulative poke counts are shown. Poke counts were compared using the two-sided Wilcoxon signed-rank test (**P = 0.001). The effect of silencing nNOS neurons after a learning experience is quantified during habituation sessions after exposure to soiled bedding in the home cage (port entries to the same port (red) with blank filters are compared during habituation after home-cage treatment sessions in b (2) and c (2), laser on and off, P = 0.002, in b (2) and d (2), laser on, P = 0.001, and c (2) and d (2), laser off and on, P = 0.5). The paired count differences (red–blue port) are compared across darcin sessions (b (3) to d (3), laser on, P = 0.5, and c (3) to d (3), laser off and on, P = 0.0001) and recall of darcin (recall session comparison b (4) to d (4), P = 0.9, and c (4) to d (4), P = 0.0001) using the two-sided Wilcoxon signed-rank test, adjusted for multiple comparisons. e, Optical silencing of nNOS neurons in the MeA does not affect non-social reinforcement behaviour. Cumulative poke counts during habituation (1), habituation after treatment (2), and water (3) sessions in mice expressing eNpHR (n = 12) in nNOS neurons in the MeA with silencing. Poke counts were recorded on the days indicated by purple arrows in a. Water-restricted mice were rewarded with a drop of water (5 μl) in one port (red) and a blank filter in the second port (blue). During habituation (1, 2) sessions both ports contained a blank filter. Mean (bold lines) and individual (fine lines) cumulative poke counts are shown. Poke counts were compared using the two-sided Wilcoxon signed-rank test (**P = 0.0005).

Extended Data Fig. 8 Ultrasonic vocalizations that are emitted by mice exposed to darcin or stimulated optogenetically consist of seven unique syllable categories.

a, Representative spectrograms of ultrasonic vocalizations classified into seven categories of call. The heat maps show the intensities of the vocalizations. Descriptive statistics (mean ± s.d., sample sizes) for frequencies are given in Extended Data Table 2 for the locations indicated by the corresponding letters on the spectrograms. b, Percentages of different call categories emitted by mice exposed to darcin (n = 24, in green) and optogenetically stimulated (n = 12, in blue).

Extended Data Table 1 Cell counts for exposure to different cue types, nNOS expression and the overlaps in the posterior MeA
Extended Data Table 2 Syllable categories for darcin and light-evoked ultrasonic vocalizations

Supplementary information

Reporting Summary

Video 1

: Darcin reinforces recall of ultrasonic vocalization and scent marking behaviours A female mouse, previously exposed to darcin, emits nearly synchronous ultrasonic vocalization and scent marking by the prior darcin exposure port.

Video 2

: Activation of darcin-responsive neurons in the medial amygdala reinforces recall of ultrasonic vocalization and scent marking behaviours A female mouse, previously experienced photo-stimulation of neurons expressing ChR2-eYFP induced by darcin exposure, emits nearly synchronous ultrasonic vocalization and scent marking by the prior photo-stimulation port.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Demir, E., Li, K., Bobrowski-Khoury, N. et al. The pheromone darcin drives a circuit for innate and reinforced behaviours. Nature 578, 137–141 (2020). https://doi.org/10.1038/s41586-020-1967-8

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing