The perfume of reproduction in birds: chemosignaling in avian social life

Abstract

This article is part of a Special Issue "Chemosignals and Reproduction". Chemical cues were probably the first cues ever used to communicate and are still ubiquitous among living organisms. Birds have long been considered an exception: it was believed that birds were anosmic and relied on their acute visual and acoustic capabilities. Birds are however excellent smellers and use odors in various contexts including food searching, orientation, and also breeding. Successful reproduction in most vertebrates involves the exchange of complex social signals between partners. The first evidence for a role of olfaction in reproductive contexts in birds only dates back to the seventies, when ducks were shown to require a functional sense of smell to express normal sexual behaviors. Nowadays, even if the interest for olfaction in birds has largely increased, the role that bodily odors play in reproduction still remains largely understudied. The few available studies suggest that olfaction is involved in many reproductive stages. Odors have been shown to influence the choice and synchronization of partners, the choice of nest-building material or the care for the eggs and offspring. How this chemical information is translated at the physiological level mostly remains to be described, although available evidence suggests that, as in mammals, key reproductive brain areas like the medial preoptic nucleus are activated by relevant olfactory signals. Olfaction in birds receives increasing attention and novel findings are continuously published, but many exciting discoveries are still ahead of us, and could make birds one of the animal classes with the largest panel of developed senses ever described.

Keywords: Avian; Olfaction; Pheromone; Sexual selection recognition.

Figure 1

Number of papers on avian olfaction after the seminal Bang 1960 paper published in Nature, based on ISI Web of Science [search term: (olfaction OR smell OR odo*r) AND (bird* OR avian)]. Panel A presents the absolute number of papers per year, panel B these papers expressed as a percentage of all papers on birds for the given year. This figure illustrates the tremendous increase of scientific research on the topic of avian olfaction, especially within the last 5 years. Open bars (panel A) and open circles (panel B) highlight the 121 papers published before 1999, black bars and circles refer to the 259 papers that were published over the past 15 years.

Figure 2

Differences and similarities in chemical nature of volatile organic compounds present in the secretion of the uropygial (preen) gland in various species of birds as analyzed by gas chromatography–mass spectrometry. A. The first two factors extracted by a principal component analysis of data (presence or absence of 172 tentatively identified chemical compounds) collected in 12 different species (n=1/species) provide a classification of species based on similarities in preen gland oil composition that diverges completely from the species phylogenetic distance (e.g. the 3 mimidae species northern mockingbird, brown trasher and gray catbird are quite distant). The divergent oil composition of closely related species is consistent with the notion that the oil volatile compounds may play a role in avoiding hybridization between species. B. The relative abundance of two specific alkanols (1-tetradecanol and 1-hexadecanol) represents a species characteristic that is also not directly related to their phylogenetic relatedness. Redrawn from data in Soini et al. (2013)

Figure 3

Results of choice test experiments showing that various avian species are able to recognize, based on odors only, their own species (A), the sex of a conspecific (B), their sexual partner (C) and their own odor (D). A. Crested auklet preferentially orient in a Y maze towards the odor of conspecific feathers or of two of its major components, cis-4-decenal and octanal but avoid the musky odor of a mammalian predator. B. Both male and female spotless starlings prefer to approach male scent over female scent in a two way choice test. C–D. Antarctic prions preferentially approach, in a Y maze, the odor of their partner as opposed to another conspecific (C) but avoid their own odor as compared to the conspecific odor (D). In panels A and B the horizontal dotted line indicates the null hypothesis (50% choice of each stimulus). Redrawn from data in Hagelin et al. 2003 (A), Amo et al. 2012a (B) or Bonadona and Nevitt 2004 (C–D)

Figure 4

Mice are able to discriminate individual odors of blue petrels. A. Schematic presentation of the experimental design in which mice were first habituated to the odor of either an adult female, a nestling chick in downy plumage or a fully feathered chick near fledging (the Referent [R] odor; left drawing). After they habituated to this stimulus as evidenced by a decrease in olfactory exploration, they were then exposed to pairs of odors, one of which was related to the referent odor while the other was not (right drawing). Mice were able to discriminate between the odors of a familiar vs. unfamiliar female (B) and between the odors of the parent of referent fledgling chick and of an unrelated adult (D) but not between the odors of the parent of a downy chick and an unrelated adult (C). These data indicate that blue petrels have individual odors and that odors of fledgling chicks match those of the parents but not in the downy chick whose uropygial gland is still poorly or not developed. Data are means ± SEM of percentages sniffing bouts. *= p<0.05. Redrawn from data in Célérier et al. (2011).

Figure 5

The proportions of volatile compounds in the uropygial gland secretion of dark-eyed juncos differ between males and females and predict reproductive success. A. Comparison of the relative abundance of four volatile compounds in the secretions of males and females. Both tetradecanoic and hexadecanoid acids are present in higher concentration in females than in males. B. Relationships between the number of genetic offspring produced (sum of within- and extra-pair offspring) with the relative proportion of diverse volatile compounds present in uropygial gland secretion as represented by the proportion score 3 derived from a principal component analysis. Proportion score 3 largely reflects the abundance of 2-tridecanone, 2-tetradecanone and 2-pentadecanone that are particularly abundant in males. These two variables are positively correlated in males (black regression line) but negatively correlated in females (dotted regression line). Therefore females with a more female-like and males with a more male-like secretion (odor?) have a higher reproductive success. Redrawn from data in Whittaker et al. 2013

Figure 6

Effect of the (hidden) presence or absence of aromatic herbs on the addition of fresh aromatic plants in the nests of blue tits. The figure shows the percentages of nests in which tits added aromatic plant fragments within 24 or 48 hours, according to whether these nests contained hidden aromatic plants (Herb +) or did not contain such material (Herb -). The numbers of nests in each category are indicated at the base of the bars. **= p<0.01, ***= p<0.001. Drawn from data in Petit et al. (2002)

Figure 7

Occlusion of the nostrils in male Japanese quail markedly decreases the induction of c-fos expression in the medial preoptic area (mPOA) and in the rostral bed nucleus of the stria terminalis (BSTM) as well as the induction of zenk expression in the olfactory bulbs following sexual interactions with a female and performance of copulatory behavior. Cells immunoreactive for the Fos or the Zenk protein (Fos-ir and Zenk-ir) were counted in these three brain regions in males who had copulated with a female (Sex), had copulated with a female with their nostril occluded (Sex w. nostril plugged) or males who stayed in their home cage as a control. *=p<0.05 vs Control; #=p<0.05 vs. Sex. Redrawn from data in Taziaux et al. 2008.