Department of Primatology
Max Planck Institute for Evolutionary Anthropology
Deutscher Platz 6
phone: +49 (341) 3550 - 200
fax: +49 (341) 3550 - 299
The behavioral endocrinology group aims to understand how ecological and social parameters shape the diverse endocrinological patterns in great ape species and humans. While we are mainly interested in the investigation of great ape populations in their natural habitat, which necessitates the collection and measurement of non-invasively collected samples, we make use of the opportunities to collect samples in zoos and sanctuaries to develop and validate new methods. Some of the questions we are currently interested in answering using endocrinological approaches are:
Male reproductive strategies
How do males maximize their reproductive success? How are modes of competition related to androgen and glucocorticoid levels?
While in chimpanzees males are dominant over females and reproductive success follows a priority of access pattern (Boesch et al. 2006), in bonobos males are not dominant over females and therefore have to rely on alternative mating strategies. Together with Gottfried Hohmann we investigated male reproductive strategies, dominance, and aggression in relation to testosterone levels in a wild bonobos at Lui Kotal, DRC (Surbeck et al 2012a). We found that although high ranking bonobo males are more aggressive than low ranking males, there was no correlation between rank and testosterone levels. Furthermore, in presence of a receptive female, testosterone levels of high ranking males dropped while those of low ranking males increased (Surbeck et al 2012a). In general it seems that bonobo males invest more in close relationships to adult females than into aggressive interactions to secure mating opportunities. The proximity to receptive females however could lead to increased stress in these high ranking males since they experience elevated cortisol levels during their association with these females (Surbeck et al. 2012b).
How male rank and demographic and ecological parameters are associated with androgen and glucocorticoid levels in male white-faced capuchins is topic of another project in our group. Franka Schaebs found in her PhD research that alpha males hadsignificantly higher fecal androgen levels than subordinates while other factors such as group instability or presence of fertile females did not affect androgen levels (Schaebs et al. in press). This project is conducted in cooperation with Susan Perry at the Lomas Barbudal Forest Reserve, Costa Rica.
Female reproductive strategies
How do females distribute their matings in relation to the fertile period of their cycle? How reliable are the cues of potential fertility available to males? Are females able to exert mate choice?
Based on the difference in dominance relationships between chimpanzees and bonobos, we can predict that these patterns might also be reflected in female reproductive strategies. In Tai chimpanzees, changes in swelling size indicate the probability of ovulation and males can use this information to allocate their mating efforts accordingly (Deschner et al 2003, Deschner et al 2004). Interestingly, although males are dominant over females, females are successful in exerting their choice of mating partners to a large extent (Stumpf and Boesch 2005, 2006, 2010). Fertile periods can not only be advertised by visual but also acoustic signals. We recently investigated the function of copulation calls in relation to ovulation in cooperation with Simon Townsend and Klaus Zuberbühler in Budongo chimpanzees (Townsend et al. 2008, Townsend et al. 2011). Copulation call rates in Budongo females are not dependent on the reproductive state of the female but females suppressed their calls if high-ranking females were nearby. Furthermore, while we did not find any acoustic differences in calls given by females in fertile and non-fertile periods, as assessed by their hormonal profiles, the calls’ acoustic structure did reliably encode individual identity. We are now complementing our studies on chimpanzees with studies on other African great ape species. In a study on mating patterns in relation to cyclic activity in mountain gorillas in collaboration with Martha Robbins (PhD Sosthène Habumuremyi), we found that parous females have a higher frequency of ovarian cycles than nulliparous females. Furthermore changes in fecal progestagen patterns indicated a considerably higher rate of abortions as previously estimated by observations alone (Habumuremyi et al 2016). In a study on sexual swelling patterns in female bonobos in collaboration with Gottfried Hohmann (PhD Heidi Douglas) we found that the timing of ovulation in relation to the maximum swelling period is much more variable than in chimpanzees and that a considerable proportion of ovulations happened even outside the period of maximum tumescence (Douglas et al 2016).
Social bonds and cooperation in chimpanzees and bonobos
How does the identity of partners influence oxytocin and glucocorticoid excretion after cooperative interactions?
Recently, the adaptive value of strong social bonds has been a topic of a number of studies. How such bonds are established and the physiological mechanisms underlying the establishment of bonding is currently unknown. Oxytocin has been implicated in the formation and maintenance of social bonds between relatives and monogamous mates, but its role in other types of social relationships remains largely unexplored. By exploring the behavioral endocrinology of social relationships among unrelated chimpanzees and bonobos in natural social contexts, we aim at gaining novel insights into the adaptations related to maintaining affiliative and cooperative social bonds outside of kinship. In our first study in cooperation with Toni Ziegler, Kevin Langergraber and Klaus Zuberbühler we investigated how relationship qualities influence the oxytocin secretion after grooming (Crockford and Wittig et al. 2013) and how food sharing influences oxytocin secretion (Wittig and Crockford et al. 2014) in Budongo chimpanzees. We could show that single aggressive interactions increase urinary glucocorticoid levels in male Budongo chimpanzees (Wittig and Crockford et al. 2015) and that the presence of a bond partner has an impact on glucocorticoid excretion after grooming and intergroup encounters (Wittig and Crockford et al. 2016).
Furthermore, in the context of a PhD project of Liran Samuni in cooperation with Roman Wittig and Cathy Crockford we found that oxytocin levels increase in Taï chimpanzees in the context of intergroup encounters (Samuni et al 2017). In cooperation with Gottfried Hohmann, a study on relationship quality and cooperation in relation to oxytocin and cortisol levels in male and female bonobos is currently under way in a post doc project of Liza Moscovice and Martin Surbeck. In cooperation with Roman Wittig and Catherine Crockford, we furthermore aim at deepening our understanding of the influence of post conflict behavior on oxytocin excretion and cortisol levels in Tai chimpanzees (PhD Anna Preis).
Population differences in female reproductive hormones
How do human populations differ in their female sex steroid levels and what causes this variation?
Despite four decades of evidence that ovarian steroid levels vary as much as three-fold among healthy pre-menopausal women within a population, and as much as two-fold across populations, the causes of this variation remain obscure. To test several hypotheses regarding the sources of this variation we undertook, in collaboration with Virginia J. Vitzthum (Indiana University, Bloomington), two “natural experiments”.
Our first study is of nomadic Central Asians, who are predominantly dependent on their herd animals for subsistence. From early childhood they consume at least 40% animal fat, a level comparable to or greater than typical U.S., European, and Asian diets, but have higher physical workloads and lower energy intake than are common in industrialized populations. Despite these energetic demands, we found that these nomads had higher progesterone levels than did Germans (identically assayed), suggesting that life-long high fat intake may influence adult hormone levels.
To assess environmental effects independent of possible genetic differences, our second study compared women born and raised in East and West Germany, prior to reunification. These genetically comparable populations experienced divergent living conditions which affected growth and other aspects of their biology and health. Although West Germans have taller heights, indicative of greater food intake and/or less energetically demanding activities, East Germans had higher progesterone levels, arguably because of the relatively higher proportion of animal fats in their diets.
Energy balance in African great apes
Are there rank dependent differences in energy balance in African great apes? How is seasonality related to variation in energy balance? Are limits of the chimpanzee distribution range related to energetic constraints?
A major task in the daily life of wild animals is to consume enough food to obtain the energy that is needed to maintain basal body function and meet travel costs and thermal requirements. In cooperation with Jürgen Kratzsch, Michael Richards, Ben Fuller and Sylvia Ortmann, we have established methods to measure energy balance in great ape urine via C-peptide levels (Deschner et al. 2008) and stable isotope signatures of δ13C, δ15N (Deschner et al. 2012). We used the method of measuring urinary C-peptide to monitor variation in energy balance in relation to female rank and seasonality in mountain gorillas in cooperation with Martha Robbins (Grueter et al. 2014), in relation to rank and mate competition in male bonobos (Surbeck et al. 2015). Additional projects on energy balance in relation to seasonality are underway in female chimpanzees at Taï in cooperation with Christophe Boesch (PhD Livia Wittiger) and at Fongoli, Senegal in cooperation with Hjalmar Kühl and Jill Pruetz (PhD Erin Wessling and female bonobos (PhD Niina Nurmi) in cooperation with Gottfried Hohmann and Oliver Schuelke.
Ontogenetic variation in great apes and humans
How do patterns of endocrinological changes during ontogeny vary between humans, bonobos and chimpanzees?
A number of differences between humans, chimpanzees and bonobos might be related to differences in development. It is therefore of great importance to monitor changes of physiological parameters during ontogeny. One specific event that until now has been only described for humans and chimpanzees is the occurrence of adrenarche characterized by the onset of adrenal secretions of increasing amounts of dehydroepiandrosterone-sulfate (DHEA-S). In a recent study in cooperation with Gottfried Hohmann, we could first validate an EIA for the use of bonobo urine and then show that apart from humans and chimpanzees, bonobos also show the typical pattern of increased DHEA-S secretion around the same age as it is observed in these species and therefore also have adrenarche (Behringer et al. 2012b).
Do bonobos and chimpanzees differ in their onset of maturation? Puberty is characterized by biological events, such as changes in growth rate and the onset of sexual maturation. Rising levels of testosterone are an indirect marker for reproductive maturation of testes and ovaries. In humans the increase of testosterone happens after the juvenile pause with around ten years of age. By analyzing urine samples of bonobos and chimpanzees with an LC-MS we were able to show that in both species in males the testosterone increase happens with around eight years of age. In female chimpanzees the increase of urinary testosterone levels is on year later, but in female bonobos the levels increased already around five years of age. The low urinary testosterone levels in younger individuals indicate that bonobos and chimpanzees experience a juvenile pause during their early development (Behringer et al., 2014a). During ontogeny, growth and the maturation of skeleton and the brain is thyroid hormone (THs) dependent. In humans, THs are high during early life and decrease after the adolescent growth spurt. After having validated an EIA for the use of bonobo and chimpanzee urine we could show that while in both species, immatures had higher values than adults, chimpanzees experienced a significant decrease in TH levels years earlier, but bonobos several years later than humans (Behringer et al., 2014b). This result is in line with the hypothesis that bonobos are delayed in development.
Method development and validation
We have developed an LC-MS method that allows us to simultaneously measure 23 different steroid metabolites in only 200µl of urine (Hauser et al., 2008a). Using this method we detected methodological problems in the extraction and measurement of testosterone in great ape urine in all existing studies so far and we were able to propose better alternatives (Hauser et al 2008b). Furthermore, by measuring a number of androgens and their metabolites in serum and urine of chimpanzees from the Ngamba Island sanctuary in Uganda in cooperation with Lawrence Mugisha, we were able to identify which of these were good indicators of testicular activity in urine and which of these were mainly produced by the adrenal glands. Interestingly, although testosterone constitutes only a minor fraction of the overall mass of androgens in urine, it still turned out to be the best indicator of testicular activity (Hauser et al. 2011). LC-MS analysis proved to be a potent tool for the validation of enzyme immuno assays for endocrinological measurements in matrices such as urine and feces. For example, we found that a testosterone assay formerly considered to be inappropriate for measuring gonadal activity in chimpanzee urine actually delivers very reliable results, and that existing inconsistencies were not related to cross-reactivities of the antibody but to inappropriate extraction methods (Preis et al. 2011). A comparison of female sex steroids in urine and fecal samples of mountain gorillas with LC-MS and EIA revealed that fecal analysis of iPdG and pregnanediol is suitable for detecting ovulation in female mountain gorillas. However, urinary measurements using both EIA and LC–MS appeared to be uninformative for monitoring ovarian activity in this primate (Habumuremyi et al. 2014). In an additional project in cooperation with Susan Perry, we developed a LC-MS method that allows for the measurements of androgens, glucocorticoids, and estrogens in fecal extracts of white-faced capuchins (Weltring et al. 2012).
In order to better monitor stress responses that are mainly triggered via the sympathetic nervous system, we validated in cooperation with Erich Möstl the use of an assay to measure alpha-amylase in the saliva of bonobos (Behringer et al. 2012a). We furthermore validated in cooperation with Fumihiro Kano and Josep Call a thermo-imaging technique to measure the facial skin temperature by simultaneously measuring salivary cortisol levels (Kano et al. 2016).
To enable the monitoring of energy balance via changes in fecal thyroid hormone levels, we validated a commercial thyroid hormone assay for fecal samples of capuchin monkeys (Schaebs et al. 2016).
Determining individual health status is of great importance for a better understanding of life history trade-offs between growth, reproduction, and maintenance. To be able to non-invasively assessactivation patterns of the cellular immune response we validated a commercial EIA for the measurement of neotperin in urine samples of chimpanzees and bonobos in cooperation with Fabian Leendertz (Behringer et al. 2017).
In cooperation with Stefan Wudy and Werner Blum we were able to validate an EIA for the measurement of urinary insulin-like growth factor-binding protein 3 to investigate aging patterns which are related to cell proliferation and apoptosis (Behringer et al. 2016).
Sociality and Health
How are social relationships related to patterns of stress and health in wild chimpanzees?
The interactions between sociality and health in wild chimpanzees are currently unknown. One proposed mediator between sociality and health is stress, which is generally known to increase the secretion of a number of hormones, including cortisol. While long-term stress leads to an increased susceptibility to a variety of diseases, including infectious diseases, and cortisol is known to have a negative impact on the immune system, infections themselves can also increase cortisol excretion. Therefore, changes in cortisol levels prior and during an infection with pathogens, as well as variation in the degree of sickness behaviour shown, could be indicative of the severity of an infection. Recent methodological advances allow now for the monitoring of cellular immune system activity via urinary neopterin levels in chimpanzees and bonobos (Behringer et al. 2017).
In cooperation with Fabian Leendertz form the Robert Koch-Institute, we investigate in a PhD project of Doris Wu patterns of stress and immune system activation during respiratory disease outbreaks in Taï chimpanzees. How sickness behavior is related to stress, immune response and sociality in immature chimpanzees is the topic of the PhD of Therese Löhrich.