1Our Japanese-French collaboration began in 2006, when I (C. Garcia) started my postdoctoral research with Michael A Huffman and Keiko Shimizu at the Primate Research Institute (PRI) of Kyoto University.
2The PRI is one of the most prestigious primate centers in the world and is a nationally supported institute equipped with all supporting facilities and staff, including enclosures with individuals or social groups of Japanese macaques, chimpanzees, and other primates, and laboratories for biochemical, genetic, and pathological analyses. The PRI offers unique opportunities to conduct multi-disciplinary projects combining ecology, behavior, cognition, and physiology. With the creation of a new center (Center for International Collaboration and Advanced Studies in Primatology, CICASP) in 2009, that has the explicit goal of facilitating international collaborations, the Primate Research Institute has expanded its ambitions to become an international center in primatology.
3My postdoctoral research on "Social stress, nutritional state and reproduction in female Japanese macaques" was funded by a Lavoisier grant (2006 - 2007) from the French Ministry of Foreign Affairs and by a JSPS Postdoctoral Fellowship (2008 / follow-up analysis). Since that date, I have been coming regularly to PRI and have been collaborating with Huffman and Shimizu, among others. Huffman has conducted field studies on wild populations of primates at several study sites throughout the world, and is a specialist in mating behavior of Japanese macaques. Shimizu (now at Okayama University of Science) has developed a method to quantify reproductive hormones in fecal and urinary samples from different primate species. Her team is one of the first in the world to describe and validate hormonal assays shown to reflect female ovarian function accurately in primates. I have benefited from this unique “savoir-faire” and I have learned these specific skills in PRI and then developed and applied them in our laboratory in Paris. Since my recruitment at the National Center for Scientific Research (CNRS), I have pursued my collaboration with Japanese researchers at PRI and we have obtained two consecutive PICS France - Japan (International Projects for Scientific Cooperation) funded by the CNRS. A PICS is a research project involving two teams, one in a CNRS-affiliated laboratory and the other abroad. These PICS, awarded for a 3-year period (2013-2015), are aimed at forming sustained networks and consolidating our on-going collaboration. Over the last ten years, our two teams have produced 7 joint publications and more than 10 communications in international and national congresses. Communication between the two institutes is continuous. Collaboration between CNRS and Japanese teams has also allowed the training of students in the lab and in the field (sample preparation and hormonal assays for monitoring reproductive hormones; parasitological analysis; behavioral observations of wild primates), and I am now co-supervising with Furuichi (PRI) a PhD student (Rigaill) (affiliated with Kyoto University) working on "Multimodal sexual signaling and mating strategies in Japanese macaques". Moreover, I am a member of the Leading Graduate Program in Primatology and Wildlife Science (Kyoto University), dedicated to training future professionals in conservation and animal welfare. In the framework of this program, I was invited to give a talk at the Kick-off Symposium held in March 2014 in Kyoto.
4Our research interests are focused on the evolution of mating and social systems in primates and on the adaptation of primate reproductive function to environmental constraints, especially energetic constraints. We examine the evolution of life-history traits, reproductive strategies, and sexual signaling in wild and captive populations of primates.
5In primates with diluted milk and long lactation relative to body size, the interaction of energetics and reproductive rates remains the subject of considerable debate as to mechanisms, controls and costs (Key and Ross 1999). Despite the long-held view that reproduction is a costly activity, there have been only a few studies directly assessing energy metabolism in free-ranging primates (Emery Thompson and Wrangham 2008a; Simmen et al., 2010). Anthropologists have long viewed humans and most other primates as organisms with particularly high reproductive allocation costs due to expanded brain size, long periods of infant and juvenile dependency, and socially mediated learning. This mixture of energy costs (growth to weaning) with time costs (such as the prolonged period to reach first reproduction), has confounded the assessment of reproductive effort in primates.
6Primate life histories can be formalized as strategies to optimize reproductive effort, or the allocation of resources to reproduction. Necessary constraints on the allocation of resources generate unavoidable trade-offs between investment in self and investment in offspring. Thus, the optimization of reproductive effort can be viewed as the cornerstone of successful life history strategies.
7Our research explores the question of how costly are infants and aims at highlighting the links between energetics (energy intake, energy expenditure, and energy availability), health status and reproduction in female primates. The physiological regulation of reproductive effort is manifested primarily in the allocation of metabolic energy to reproduction. Both the amount of energy allocated and the temporal pattern of allocation have profound consequences for ultimate reproductive success (Rigaill et al., 2015). The key questions tackled in our projects are: a) What is the role of energetics in primate fertility, e.g. the capacity to conceive, gestate, and lactate? ; b) How costly is lactation among primates?; c) How do female primates minimise these costs?
8Primate social systems depend on signals that convey information between individuals. These signals often comprise complex cues that can communicate information ranging from identity to current social or reproductive status. Although some sexual signals appear based on a single main element, theoretical models often emphasize the importance of multiple elements of sexual signals (i.e. multimodal signaling) when evaluating their role in mate preferences and mating decisions (Candolin 2003).
9In many primate species, males can discriminate within female reproductive cycles, as they are more likely to copulate on days when conception is most probable (Deschner et al., 2004; Emery Thompson and Wrangham 2008b; Engelhardt et al., 2004; Garcia et al., 2009; Higham et al., 2009). However, the cues and reliability of the various signals that are available to males making mating decisions are still poorly understood. Potential cues include: female behaviors (Carosi and Visalberghi 2002; Engelhardt et al., 2005), auditory cues (Pradhan et al., 2006; Semple et al., 2002), visual cues (e.g. sexual swellings and redness of sexual skin: Dubuc et al., 2009; Higham et al., 2008, 2009, 2010), and olfactory cues (Clarke et al., 2009; Michael and Keverne 1970; Rigaill et al., 2013). Males of virtually any human and non-human primate species can probably detect the fertility status of females through some combination of cues. Visual cues are suggested to be particularly important. There is variation over the ovarian cycle in the sexual skin of females, i.e. an increase in size (sexual swelling) and/or change in color. Several primate species, such as olive baboons and Barbary macaques, express skin color changes in the anogenital region, while some also express color changes in ventral anatomical areas such as the chest and abdomen (gelada baboons), as well as the face, e.g. rhesus macaques (Dubuc et al., 2009) and Japanese macaques. As vision is the dominant sensory modality for diurnal primates, sexual skin color is therefore posited as the primary cue in fertility signaling. Nevertheless, few studies have investigated the potential link between female fertility, mating decisions and red skin coloration in catarrhine species (Bradley and Mundy 2008; Dubuc et al., 2009; Higham et al., 2008, 2010; Setchell et al., 2006). More data are needed to better understand the interaction.
10In addition to visual cues, it is also known that olfactory cues are associated with reproductive status and ovulation in a number of mammalian species. However, few attempts have been made to identify their use as a sexual signal by diurnal primates, and our knowledge of chemical communication in primates is relatively poor compared to our understanding of both visual and auditory communication (Heymann 2006). This may be because primates are traditionally regarded as microsmatic so that olfaction is generally considered to play only a vestigial role in the mediation of inter-sexual interactions. However, an increasing body of evidence suggests that olfactory cues play a much greater role in primate communication than previously assumed (Stevenson 2010). Although some research has been carried on olfactory communication in strephsirrhines and New World primates, very little information exists for catarrhines. Several studies have reported sniffing behaviors in a sexual context (baboons: Clarke et al., 2009; Rigaill et al., 2013) and olfactory signals are known to advertise reproductive state, dominance rank, and/or individual identity in strepsirrhines (Crawford and Drea 2015; Palagi and Dapporto 2006; Scordato and Drea 2007), callitrichids (Ziegler et al., 1993), and catarrhines (Setchell et al., 2010). Female odors are also suggested to elicit both behavioral and endocrinological responses in male primates (Cerda-Molina et al., 2006; Keverne and Michael 1971; Michael and Keverne 1970; Michael and Zumpe 1982), and studies of macaques (Engelhardt et al., 2004) and baboons (Clarke et al., 2009) have suggested that olfactory cues may be critical elements in primate fertility signaling systems.
11The objective of our past and current research is then to provide an assessment of the possible role of olfactory and visual cues in the multimodal signaling of reproductive status in a species of catarrhines lacking sexual swellings (i.e. an obvious visual cue), the Japanese macaque (Macaca fuscata).
12Japanese macaques are a good model for exploring multimodal reproductive signaling (i.e. ovulation and pregnancy signaling), since females exhibit color changes in the face, utter copulation calls during mating (Garcia et al., 2009), and produce vaginal secretions that could potentially offer information concerning the reproductive status and the timing of ovulation. Studies investigating the simultaneous effect of visual, auditory, behavioral and olfactory cues as signals of reproductive state have only been carried out in olive baboons (Rigaill et al., 2013) and crested macaques (Higham et al., 2012), species with clear advertisements of the reproductive status (e.g. exaggerated swelling of the sexual skin). Our project in Japanese macaques is therefore the first one to investigate simultaneously the different sexual signals that could influence male mating decisions, in a species lacking sexual swellings. Moreover, it is the first study to assess if female sexual signals are still present after conception in Japanese macaques, and to analyze the occurrence of post-conceptive sexual behaviors.
13We hypothesize that females undergo subtle changes in color and luminance of the face that could be used as a reproductive cue, and we predict that these two parameters will be more intense during the fertile phase. This visual cue - visible at considerable distances - may be of primary importance for males making decisions about association and consortship; additionally, males who have close access to a female and thus able to inspect and sniff her anogenital region, i.e. consorting males, may use other estrogen-dependent cues, such as tactile and olfactory cues (vaginal secretions of volatile aliphatic acids), to assess her reproductive state more precisely (Garcia et al., 2009; Higham et al., 2009; Rigaill et al., 2013). Based on current knowledge of Japanese macaque behavioral ecology and olfactory communication in other primate species, we predict that vaginal secretions will encode information concerning the timing of ovulation, and will undergo changes in quality and/or quantity across the ovarian cycle that males could detect.
14Our collaborative research is conducted both in captive settings (Primate Research Institute) and in the wild (Yakushima and Koshima islands). We combine behavioral observations (sexual interactions), data on energetics (morphometric measurements and isotopic data), hormonal data (fecal hormones), parasitological data (parasite stages voided in macaque feces), chemical data (chemical compounds of vaginal secretions), and digital photography (visual scoring of the sexual skin color) in wild and captive Japanese macaques (Macaca fuscata).
15During behavioral observations, we used focal animal and ad libitum sampling (Altmann 1974). We recorded changes over the reproductive cycle (non-fertile phase, fertile phase, gestation) in the following behaviors: agonistic interactions, female behavior towards males, and male behavior towards females. Any female-specific pre-copulation sexual behavior directed towards males was recorded: approaches, holds, presentations of anogenital area, estrus and copulation calls, mounts and grooming (Garcia et al., 2009). Concerning male behavior towards females, we compared consorting and non-consorting males, recording the following behaviors: approaches, holds, grooming, tactile and olfactory inspections, mounts with and without ejaculation.
16We determined female nutritional status and body composition using morphometric measurements and isotope-labeled water (deuterium). In brief, we monitored body mass, body length, arm and calf circumferences, and skinfold thickness at three different sites (abdominal, supra-iliac, subscapular). Female physical condition was assessed using the Quetelet Index (Garcia et al., 2010).
17We also conducted isotopic experiments by injecting deuterium subcutaneously and following the kinetics of deuterium equilibration throughout the body. The isotopic ratio was measured using mass spectrometry. With this method, we can determine body composition (total body water, total body fat and fat-free mass).
18We compared female nutritional status and body composition between the beginning and the end of the mating season and linked those energetic variables to reproductive outcomes.
19We collected fecal samples to assess ovarian activity in females and to determine ovulation and conception dates. The fecal samples were analyzed for estradiol and progesterone using previously described and validated enzyme immunoassays shown to reflect female ovarian function accurately in the Japanese macaque (Garcia et al., 2009). We also used these fecal samples to analyze adrenal hormones, e.g. glucocorticoids and dehydroepiandrosterone-sulfate (DHEAS), and to explore the role of social and environmental factors on these hormones (Takeshita et al., 2014).
20To explore the relationship between sociality and individual health, we also used the fecal samples for examination of intestinal parasite stages voided by the macaques, via the methods described in MacIntosh et al., (2010). Parasite stages were identified and quantified using a modified sedimentation protocol.
21We determined the chemical components of female vaginal and urinary secretions to investigate if the composition of the secretions undergoes changes across the ovarian cycle. After training the monkeys in a non-stressful method to collect urine and vaginal samples (i.e. rubbing sterile cotton swabs around the vulva), we analyzed the samples by gas chromatography–mass spectrometry (GC-MS) to identify chemical volatile compounds.
22Digital images of females’ faces were used to measure color (chromatic) and luminance (achromatic) (across the ovarian cycle and gestation), which were then assessed for variation within the visual system of Japanese macaques (following methods in Rigaill et al., 2013). To standardize images of face color, we used a Gretag X-Rite Color Checker (24 colored squares of known and varying reflectance; Figure 1). This technique allows images to be standardized such that color measurements are comparable across images in different lighting conditions. It allows us to compare changes in facial redness and luminance across time (intra-individual variability) and females (inter-individual variability). We converted our camera sensor measures to estimates of Japanese macaque retinal receptor stimulation using previously published information (Higham et al., 2010; Rigaill et al., 2013; Stevens et al., 2009). We assessed how a female face varied chromatically, and we also analyzed facial luminance to measure how light or dark the face is according to our model of Japanese macaque visual perception.
Image of a female face with Gretag color standard and showing the area (black line) used for color analyses.
Image de la face d'une femelle avec un standard de couleur Gretag et montrant l'aire (ligne noire) utilisée pour l'analyse de la coloration.
23In a previous research project conducted in collaboration with researchers at the Primate Research Institute, we examined the relationship between nutritional status, seasonality, parasitic infection and reproductive output in female Japanese macaques.
24Female Japanese macaques are characterized by physiological adaptations and unique fat deposition mechanisms that facilitate their survival through the sometimes-harsh seasonal conditions of temperate climates. Our project aimed at exploring the relationship between nutritional status, body composition, seasonality and reproductive outcomes using isotope-labeled water and morphometric measurements from 14 captive females (Primate Research Institute). We showed that females gained weight and accumulated energy reserves (i.e body fat) in Fall, which supported the hypothesis that individual females needed to attain a sufficient physical condition prior to the long mating season and to survive the severe ecological conditions of winter with high thermoregulatory costs. We also found a relationship between conception rates and energetic condition, with females that conceived during one mating season being in better condition (i.e. fatter) at the end of their previous mating season. Together, these results suggested that, even in captive settings with constant food availability, seasonal breeding entails relatively high energy costs, and that females with higher energy status could invest more in reproductive activities and could afford to reproduce more rapidly (Garcia et al., 2010, 2011).
25We also explored socially-mediated parasite transmission (MacIntosh et al., 2012). The relationship between social interactions and infectious disease transmission produces an evolutionary tradeoff between sociality and individual fitness. Social structure can affect disease risk; however, which factor better explains observed infection phenotype is not always clear. For example, the individual's position in the dominance hierarchy can have an impact on exposure to parasites (e.g. a central individual will be more exposed and at higher risks of disease transmission), and hierarchies can also create asymmetries in susceptibility through variation in stress hormones (e.g. a subordinate may be more stressed and thus more susceptible to infection). We examined the relationship between dominance and infection by directly-transmitted parasitic nematodes in 18 wild female Japanese macaques (Macaca fuscata yakui). Our results suggest that high-ranking females were infected by a greater diversity of parasites than those of lower rank. Estimated infection intensity was also highest in dominant females for the most common parasite (Oesophagostomum aculeatum). Social network analysis showed positive relationships between indices of parasitic infection and network position. Network centrality was positively associated with dominance, suggesting the importance of rank-mediated exposure in parasite transmission. By contrast, fecal cortisol was not positively associated with rank, reducing the likelihood that stress-induced immunosuppression caused the observed dominance-related effect. Untangling these relationships is critical to a mechanistic understanding of disease-related constraints on sociality.
26To better understand the relationship between sociality, well-being and health or reproductive status, we then explored the role of environmental and social factors on fecal adrenal steroid concentrations (glucocorticoids i.e. fGC, and DHEAS) in 39 captive female Japanese macaques (Takeshita et al., 2014) housed in social groups and in single cages. Our results revealed that both fGC and fecal DHEAS concentrations were higher in females housed indoors in single cages than in those living outdoors in social groups. We also found that fGC concentrations were higher in cycling females during the mating (winter) season than lactating females in the birth (spring) season. We did not observe any association between dominance rank and either fecal DHEAS or fGC. This study showed that measurement of fecal DHEAS and fGC can be a good method to assess stress and thus welfare in Japanese macaques. These findings provide insights about the physiology of these two adrenal hormones in female Japanese macaques, which can be applied to wild populations and is fundamental for captive management and conservation biology.
27We also investigated how Japanese macaques perceived the “quality” and reproductive status of other individuals. This research aimed at understanding the links between physiology and signal perception. In this study, we specifically investigated the information that male Japanese macaques received from females that might enable them to discern periods with higher probabilities of conception. We combined behavioral and endocrine data (fecal reproductive hormones) on semi-free ranging Japanese macaques living at the Primate Research Institute to investigate the extent to which adult males rely on female behavior to make their mating decisions as well as the effect of social rank on mating success. We found no increase in female proceptive behaviors during the fertile phase, suggesting that female behaviors did not clearly signal the probability of conception or the timing of ovulation. Despite this, the frequencies of ejaculatory copulations were highest during this phase, indicating that the attractivity of females increased significantly during the period with higher probability of conception. Males, especially the highest-ranking male, were able to discriminate females nearing ovulation and to concentrate their mating effort. The alpha male monopolized most matings, probably due in part to the low number of females simultaneously ovulating and the limited number of hiding places where lower ranking males could mate with females while avoiding alpha male aggression. These results suggested that different males may have access to different ovulatory signals and/or are differentially able to act on that information (Garcia et al., 2009). However, the exact nature of the estrogen-related cues male Japanese macaques used to recognize female reproductive status, and to what extent males used them, still remains to be investigated.
28We have therefore taken this further and developed a new research program on multimodal sexual signaling which aims to investigate potential variation in female sexual signals (behavioral, visual, auditory and olfactory) between fertile and non-fertile phases that might influence male mating strategies. A pilot study (2011-2012) validated a new method for assessing skin coloration in female Japanese macaques. For facial coloration, we used an innovative method of color analysis developed by Higham (New York University), never previously used in this species. We showed that this method is both appropriate and useful for studying skin coloration in Japanese macaques. These data provided a basis on which to conduct our ongoing project, which aims to investigate changes in skin coloration that occur during the reproductive cycle and the extent to which Japanese macaques have evolved a system of multimodal signaling of reproductive status. Recently we published a study on pregnancy signals and how they might modulate post-conception mating decisions in wild male Japanese macaques (Rigaill et al., 2015). The question was whether males use changes in female sexual signals (behavioral, visual and auditory) to discriminate pregnancy and adjust their socio-sexual behaviors. Combining behavioral observations, digital photography and endocrinological (progestogen and estrogen) measures, we collected data during three one-month periods: the pre-conceptive period, the 1st month of pregnancy and the 2nd month of pregnancy. We analyzed variation in the probability of male and female socio-sexual behaviors and estrus calls, as well as changes in female facial color parameters, in relation to female reproductive state. Results showed that males did not copulate during the pregnancy period, and that female socio-sexual behaviors generally decreased from the pre-conceptive to post-conceptive periods. Female facial luminance decreased from the pre-conceptive month to the pregnancy period whereas facial color varied only between the 1st and 2nd month of gestation. These results suggest that female Japanese macaques display sexual cues of pregnancy that males use to avoid wasting energy on non-reproductive copulations. We hypothesize that females advertize their pregnancy through changes in behavioral, visual and potentially auditory signals, and that males use these signals to adjust their mating behaviors.
29The long-term interactions between the Japanese teams (Center for International Collaboration and Advanced Studies in Primatology and Department of Ecology and Social Behavior of PRI, Kyoto University, Inuyama) and our French team (UPR 2147, UMR 7206 – CNRS, Paris) have led to the creation of a sustainable network. In addition to publications and communications already resulting from the collaboration, the co-supervision of a PhD student (2013 - 2017) at Kyoto University will continue this tradition.
30We have recently got a funding for another International Project for Scientific Cooperation (CNRS) between France and Japan (2016-2018). This research project is a collaborative effort that brings together faculty, Ph.D. students of different departments and institutions (Japan: PRI, Okayama University; France: CNRS; Italy: National Research Council; UK: Durham University; USA: New York University) as well as veterinary staff and technicians from the Center for Human Evolution Modeling Research (PRI). This project aims to study multimodal advertisement of reproductive status in Japanese macaques, with a focus on olfactory cues i.e. the chemical components of female vaginal secretions and their effects on male sexual behaviors.