Coloraciones cosméticas y bacterias en contextos de selección sexual de abubillas y calaos

Abstract

Bacteria are common symbionts of macro-organisms frequently establishing mutual associations, in which both the symbiont and the host get benefits. One of the benefits that the hosts can obtain from the symbiotic bacteria is the protection against a wide range of pathogens, by means of the contribution of defensive substances that the bacteria synthesize for their competition with other microorganisms. The microbial community associated with the host may be different between individuals of the same species and, consequently, be a property that causes differences in the abilities of those individuals. These differences could be of interest to potential partners and, in this case, the production of signals associated with that type of quality can be selected. In this way, the quality of an individual's microbiome could be the information transmitted by some existing sexual signals in hosts of beneficial bacteria. The evolution and maintenance of the signals depends on the reliability of the information transmitted . In this scenario, it would be expected the evolution of signals dependent on the symbiotic bacteria hosted, through the colours or odours caused by the metabolites of the bacteria An example could be the birds of the order Upupiformes, such as the European hoopoe (Upupa epops) and the African woodhoopoe (Phoeniculus purpureus). These species have special properties in the secretions of their uropygial glands. It is known that they harbor bacteria producing antimicrobial substances in the uropygial gland and it has been shown experimentally that these symbionts are responsible for some of the properties of the secretion, such as its colour and the presence of volatile chemical compounds in their composition. The existence of symbionts in the uropygial secretion of birds could be a more widespread phenomenon. For example, in the order Bucerotiformes, a sister clade of the hoopoes, formed by 61 species of hornbills, some species have coloured (yellowish) uropygial secretions, which they use to stain different parts of their body, possibly with an ornamental function. However, it is unknown if the pigmented uropygial secretions of the hornbills are linked to the presence of microbial communities in the uropygial gland, as described in hoopoes. In this thesis, we study the hypothesis that coloured secretions, both of the European hoopoe and of the hornbills, function as quality signals that they use to transmit information about the community of symbionts associated to their uropygial glands. Sexual signals can act before (to get a partner) or after mating (to get a greater parental investment of the partner). Both, hoopoes and hornbills, have a strict distribution of roles between the two sexes. Only females incubate and take care of the nestlings while they are small, and the males provide all the food that the females and the nestlings need in that period. Therefore, the reproductive success of both members of the couple is closely linked and it is expected that the reproductive investment of both depend on the quality of their partner. In hornbills, stained secretions are used by both sexes throughout the year, and therefore they could act as pre-mating signals. However, in the hoopoes, the coloured and secretions loaded with bacteria are only present in the females, and they only develop them after pairing, during the period they are inside the nest. Therefore, in this case they could function as post-mating signals of females to obtain a greater parental investment of males. In addition, the fact that the association with bacteria in hoopoes is temporary and is linked to the nest habitat, raises the question of how they obtain their bacterioma in each season. One possibility would be the acquisition of symbionts that may remain in reservoirs in the nests used by other hoopoes the previous years. In such a case, it would be beneficial for the hoopoes to be able to detect those nests and select them for nesting. The three general objectives of this thesis intend to test three predictions of the hypothesis that the uropygial glands of hoopoes and hornbills harbour communities of different symbiotic bacteria between individuals and that signals have been selected to show their quality: (I) Hoopoes can incorporate into their gland beneficial bacteria from the material of reused nests, and thus they would select nests previously used by other hoopoes. (II) The properties of the secretion the female hoopoes are used by the males to infer the quality of the females. Therefore males will make a greater reproductive effort when they are paired with females that have higher quality secretions. (III) The pigmented uropygial secretions of hornbills will show evidence of presence of bacteria more frequently than the non-pigmented ones. Background, methods used, and main results obtained, are indicated below for each prediction separately. I. Chapter I. Selection of nesting site in the hoopoe. Effects on uropygial bacterioma. One of the key factors for the establishment and evolution of symbioses, is the mode of transmission of symbionts. Several mechanisms have been proposed to explain the complex bacterial community of the uropygial secretion of the hoopoe. It is known that some of the symbiotic bacteria present in their glands are transmitted vertically and others seem to be horizontally transmitted. Enterococci present in the secretion seem to come from the female gastro-intestinal tract, so they could also be transmitted through faeces. Since faeces are accumulated in the material remaining in used nests, one possibility would be that such material serves as a reservoir of symbiotic bacteria. Thus, females could increase the bacterial diversity of their gland or obtain more competitive strains by selecting previously used nests compared to nests without those bacteria. To test this prediction, an experiment was carried out manipulating the presence of material from nests previously used by hoopoes. It was tested: (1) if there was a preference for nest boxes containing old hoopoe nest material, (2) the consequences of nest choice on the bacterial load of the secretion and the eggshells, in contact with that material, and (3) possible effect on reproductive success. The hoopoes showed a preference for nest boxes containing soft material added to the boxes versus empty ones, regardless if the added material came from previously used hoopoe nests. The characteristics of the experimental material did not affect reproductive success, but they affected bacterial loads of the eggshells and the composition of the bacterial community of the uropygial gland. This is the first time that is shown an effect of the nest material on the bacteriome found in the uropygial secretion in hoopoes. In addition, several OTUs (Operational Taxonomic Units) of female uropygial secretions were positively associated with hatching success. These results agree with previous ones that showed that a main function of the uropygial secretion of females of hoopoes is to protect the eggs from microbial infections. They also support the prediction that the reuse of nests affects the gland's bacterial community. However, the prediction that reused nests should be preferred based on the presence of old material was not fulfilled. II. Chapters II and III. Hoopoe males perform a differential reproductive investment based on the properties of females' uropygial secretions. It is known that the dark colour of the uropygial secretion of hoopoes is associated with the presence of symbiotic bacteria, being reddish when they are eliminated with antibiotics. In addition, secretion colour saturation is negatively related to its antimicrobial capacity. A unique behavior in the hoopoe is that the females actively stain the eggshells with their uropygial secretion, changing their colour from an initial blue tone to a final green-brown. The secretion that covers the eggshells reduces the entry of pathogenic bacteria into the eggs, and that capacity depends on the enterococci it harbors. In this way, the cosmetic colour of the eggshells of hoopoes could have evolved as a post-mating sexual signal of the female's antimicrobial capacity. Numerous studies in other bird species have shown how the colour of the eggs reflects the quality of the female and how this coloration influences male investment. It is the sexually selected eggshell colouration hypothesis (SSEC hypothesis). It has been tested with intrinsic colorations produced by pigments, such as biliverdin and protoporphyrin, but never with cosmetic colorations caused by uropygial secretion. In this thesis it is investigated for the first time if the cosmetic coloration of the eggshells of hoopoes influences the amount of food provided by males (Chapters II and III). In Chapter II a descriptive study of the relationship between the eggshells colour of females and males investment was carried out. In Chapter III, a cross-fostering experiment of clutches between pairs of females was carried out to see if changes in the eggshells colour caused changes in male investment. In both studies hoopoe males invested less in female nests that contained eggshells with more saturated colouration. In addition, eggshell colour saturation was negatively related to the abundance of symbiotic bacteria in the female's secretion (Chapter II). These results are consistent with the hypothesis that the cosmetic coloration of the eggshell due to uropygial secretion would be acting as a post-mating sexual signal of females. This signal would show the antimicrobial potential of female uropygial secretions, which males would use to adjust their parental investment. This is the first experimental demonstration (Chapter III) of the benefits associated with the female cosmetic signal. III. Chapters IV and V. The coloured secretions of hornbills contain symbiotic bacteria and the ornaments of these birds have evolved to signal their quality in contexts of defense against pathogens. Until now, it has not been studied whether the uropygial secretions of hornbills are inhabited by bacteria as is the case with those of the hoopoe. In Chapter IV of this thesis, we test this hypothesis for the first time with 13 species of hornbills housed in different zoos in Spain, Portugal and France. For this, samples were collected from different surfaces of their body, including their uropygial gland tuft and their uropygial secretion. Samples were inoculated in culture media, and DNA was extracted and sequenced in order to study the composition of the bacterial communities present. The results showed bacterial growth in the secretion samples in all species of African hornbills and in one Asian species. In addition, several of these species of hornbills had secretions densely populated of bacteria when viewed under microscope. A clear association of particular bacterial taxa with uropygial glands or body areas covered with secretion in several species was also observed. Several of these bacterial taxa are producers of antibiotic substances and are commonly associated with various groups of animals in mutualistic relationships. In addition, in an Asian hornbill species, a group of pigmented carotenoid-producing bacteria was associated with the uropygial gland, and with the stained areas of the body. This evidence suggests that hornbills are a group of birds involved in coevolutive interactions with bacteria that live in or around their uropygial secretions and that some of those symbionts may be responsible for the special properties of their secretions. The coloured secretions of some species of hornbills could also be dependent on carotenoids obtained in the diet. Carotenoids are responsible for the red, yellow and orange colours. They are important for several biological functions and they are a limited and scarce resource in nature. Thus, carotenoid-based coloration may indicate the access and the ability to obtain nutritional resources, but also other quality traits, such as body condition, antioxidant capacity, parasitic load and its ability to fight against microbial infections. In this way, differences in carotenoid-based coloration among hornbill species could indicate differences in their efficiency in the use of carotenoids and in the importance of signals that reflect their immune response capacity. Many species of hornbills have, in addition, a conspicuous plumage with wide white areas that contrast with a general black color. These white patches could play an important role in sexual signaling, since, in birds, they are usually dependent on physical condition and phenotypic quality predictors. Hornbills also have a unique keratinized structure located on the beak called casque, which could act as an indicator of sexual maturity, having evolved through sexual selection. The coloration of the casque and beak is very variable among different hornbill species. Since the beak and casque covers are made of keratin, their coloration could indicate the individual quality related to their ability to cope with keratinolytic microorganisms. There are numerous differences between hornbill species in all these ornamental traits. Therefore, they are a group in which a comparative approach among species could allow to study the life history traits or the selective pressures that have been associated with the evolution of these signals. Since all of these traits can be related to selective pressures exerted by bacteria, in Chapter V of this thesis, it is analyzed if their presence and/or size are linked to the load of different bacterial groups. The main findings of this comparative analysis are, first, that the degree of exaggeration of two ornaments are positively correlated throughout hornbill phylogeny. These are: the surface of white feathers on the body and the coloured surface of the beak (yellow, orange, or / and red). In addition, variation in ornamentation between species is related to bacterial loads of different parts of the body. Finally, the prevalence in the uropygial gland of a particular group of bacteria known for their production of bacteriocins (enterococci) was related to the abundance of possible pathogenic (keratinolytic) bacteria in the casque. These results suggest that bacteria have played an important role in the evolution of the complex colour designs of hornbills. All the results of the thesis support the general hypothesis that in hoopoes and hornbills, cosmetic colorations of the uropygial secretion may be selected as quality signals associated with the possession of symbiotic bacterial communities. In the case of hoopoes, males responded to the changes in the eggshells colour by adjusting their parental investment. In the case of honbills, it is confirmed for the first time, that some species also maintain symbiosis with bacteria in their secretions. Contrary to expectations in our initial hypothesis, bacteria in this group were not present only in species with coloured secretions. However, some symbionts in these species may be responsible for the special properties of the secretion for their particular biosynthetic abilities. In addition, the comparative study of the ornaments of these species shows associations with the pressure of groups of bacteria, both pathogenic and producing defensive substances, and interactions between these bacterial groups. Together, these findings suggest that hornbills are a new study model of great interest to understand the evolution of mutualistic symbiosis with bacteria in birds, as well as of signals associated with that relationship.