Abstract
Measuring energetic condition of wild animals is of major importance in ecological research, as it is profoundly linked to fitness. However, noninvasive monitoring of energetic condition in wild-living animals is methodologically challenging. Measuring urinary C-peptide levels is a suitable method to noninvasively assess energy balance in wild-living animals. As collecting urine is not always feasible in the wild, it is essential to establish alternative biomarkers for other sample types to assess energy balance. Thyroid hormones (TH) are potential candidates as they are involved in the regulation of metabolic processes. During periods of low energy intake, serum TH levels are reduced, leading to a decrease in metabolic activity. To investigate whether fecal TH can serve as a biomarker for energy balance, we validated a total T3 ELISA to measure immunoreactive T3 (iT3) in fecal samples of yellow-breasted capuchins. We restricted caloric intake of seven males, assessed daily group caloric intake and determined daily individual fecal iT3 levels. Analytical validation of the assay showed that fecal iT3 levels can be reliably measured; however, proper storage conditions must be implemented and possible degradation to be accounted for. IT3 levels were significantly higher on days with high group caloric intake. However, individual iT3 levels varied substantially, resulting in an overlap across individuals between conditions. Our results indicate that fecal iT3 levels can serve as a useful biomarker to detect changes in energy intake of yellow-breasted capuchins. Overall, measuring fecal iT3 levels may present a suitable method for monitoring energy balance when urine collection is impossible.
Introduction
The situation-dependent allocation of energy to phenotypic traits throughout an animal’s life affects the physical appearance of the animal and has an influence on traits such as body size, age of maturation and number of offspring, thereby ultimately determining lifetime reproductive success (Stearns 1992, Roff 2002). The energy allocation model assumes that energy can either be allocated to reproduction, somatic growth and maintenance, or storage (Perrin & Sibly 1993, Roff & Fairbairn 2007, French et al. 2009). Monitoring an individual’s energy balance (defined as energy intake minus energy expenditure) is, therefore, essential to understanding how individuals allocate their energy between competing demands in relation to their environment. The investigation of energy balance in primates is particularly interesting, as they have exceptionally slow rates of growth, reproduction, aging and a low metabolic rate (Pontzer et al. 2014).
Assessing energetic condition of wild-living animals is a valuable tool in behavioral ecology when studying the effects of changing environmental conditions on the energy budget, behavior and life history of an individual. When food resources are limited, individuals have to allocate energetic resources to processes essential for survival to protect future reproductive possibilities (Stearns & Hoekstra 2005). Such a trade-off can be important in energetically costly contexts as, for example, competition over mates, mate guarding and the quest for high dominance rank (Bercovitch & Nurnberg 1996, Janson & Vogel 2006, Higham et al. 2011, Surbeck et al. 2015). The assessment of individual energetic condition allows for the ability to trace patterns of such trade-offs.
Noninvasive quantification of energy balance in wild-living animals remains a major methodological challenge. Up until the last decade, energy intake, energy expenditure and energy balance were assessed by collecting a variety of data, such as the amount of consumed food, time spent on processing food, analyzing nutritional values of food items, travel distance and time spent on finding food resources (e.g., Coelho 1974, Eley et al. 1989, Chivers 1998, Knott 1998). Such detailed measures require intensive observation times and data collection.
An alternative is the quantification of biomarkers of energy balance in noninvasively collected samples. Recent studies have validated the use of urinary C-peptide (UCP) as a physiological marker to assess energy balance in primates (Sherry & Ellison 2007, Deschner et al. 2008, Emery Thompson & Knott 2008, Girard-Buttoz et al. 2011). This method allows for the monitoring of energy balance in wild-living animal populations and thereby for the investigation of trade-offs in relation to competition for rank and reproduction, and food abundance. Dominant chimpanzee males have consistently lower C-peptide levels than low-ranking males (Emery Thompson et al. 2009), supporting previous findings that maintaining high dominance rank is associated with metabolic costs (Muller & Wrangham 2004). In wild bonobos and long-tailed macaques, mate guarding or mate competition does not affect male energy balance (Girard-Buttoz et al. 2014, Surbeck et al. 2015). Male rhesus macaques engage in competitive endurance rivalry leading to high-ranking males, who heavily invested into mating activities during the mating season, being in worse energetic condition than low-ranking males (Higham et al. 2011). Periods of low food availability or lower food quality led to a negative energy balance as measured with UCP levels in female orangutans (Emery Thompson & Knott 2008), mountain gorillas (Grueter et al. 2014) and black-and-white colobus (Harris et al. 2010). Furthermore, female chimpanzee mothers inhabiting lower quality foraging areas were found to be in worse energetic condition than mothers in food-rich areas, resulting in longer periods before resuming ovarian cyclic activity after lactational amenorrhea (Emery Thompson et al. 2012). Therefore, such noninvasive methods for the assessment of energetic condition open the door to investigate costs and benefits of different life history strategies.
However, in many wild-living animal species, such as capuchin monkeys, collecting urine samples is not regularly feasible. Fecal samples might present an alternative matrix for establishing a biomarker for energy balance as they have been successfully used to measure a variety of steroid hormones in animal species (Palme et al. 2005, Heistermann 2010, Goymann 2012). As C-peptide levels cannot be measured in feces, there is a compelling need to establish a biomarker to reliably assess the changes in energy balance in noninvasively collected fecal samples. Researchers have recently measured thyroid hormones (THs) in fecal samples in a validation study across several mammalian species (Wasser et al. 2010), in response to reduced food availability (Ingbar & Galton 1975, Abdullah & Falconer 1977, Ayres et al. 2012, Gobush et al. 2014, Joly et al. 2015) and in relation to ecological parameters (Cristóbal-Azkarate et al. 2016).
THs are crucially involved in the regulation of the metabolic activity of organisms (Kaack et al. 1979) and, therefore, represent suitable candidates for monitoring energy balance. In humans, during periods of acute or chronic starvation and hence restricted caloric intake, serum TH levels decrease and as a consequence, basal metabolism is reduced in order to conserve energy (Merimee & Fineberg 1976, Eales 1988, Blake et al. 1991, Flier et al. 2000, Fontana et al. 2006, Holloszy & Fontana 2007); in contrast, during high caloric intake, serum TH levels are elevated (Danforth et al. 1979, Galofré et al. 2010). Additionally, in several mammal species, changes in energy balance are associated with changes in serum TH levels, as for example, TH levels decreased during fasting in badgers (Harlow & Seal 1981), sheep (Blum et al. 1980) and rabbits (Menchetti et al. 2015). TH levels increase as well with increasing energy expenditure, for example, as a reaction to low temperatures (Silva 1995, Cristóbal-Azkarate et al. 2016) and in relation to the thermic effect of physical activity (Kim 2008). Consequently, TH are well-suited candidates for serving as a physiological marker for energy balance.
The two main TH are thyroxine (3,3′,5,5′-tetraiodothyronine, T4) and triiodothyronine (3′,3,5-triiodothyronine, T3) (Köhrle 1999). Although more than 99% of T3 and T4 are bound to serum proteins, it is mainly the free form of T3 that is biologically active (Danforth & Burger 1989). The sum of free and bound T3 is named total T3 (TT3) (Köhrle 1999). Radiometabolism studies on rats have revealed that while a majority of TH is excreted into the urine (Shakespear & Burke 1976), a considerable amount of TH can be recovered in feces (DiStefano & Sapin 1987). Additionally, in domestic dogs, the majority of thyroid metabolites in urine are excreted as T3 and only a small fraction as T4 (Wasser et al. 2010).
Recently, TH were measured in fecal samples of a variety of animal species, including nonhuman primates (Wasser et al. 2010). In howler monkeys, reduced food intake led to decreased fecal T3 levels in females, but not in males (Wasser et al. 2010). Unfortunately, as no data on caloric intake were collected, an interpretation of a causal relationship between reduced energy intake and fecal T3 levels was difficult to establish. However, while these results indicate that fecal T3 is correlated with energy balance, increased precision in the tracking of energy intake is still needed to validate this hypothesis.
Thus, the objective of this study is to validate a method for the noninvasive assessment of energy intake in fecal samples of yellow-breasted capuchins. Our approach was (1) to investigate whether T3 can be reliably measured in fecal samples of this species with a commercial TT3 ELISA (analytical validation) and (2) to test whether fecal T3 levels are affected by caloric restriction (biological validation). Therefore, we assessed daily caloric intake of two captive yellow-breasted capuchin monkey groups over a 3-week period during which the caloric intake was manipulated to simulate a period of restricted energy availability.
Methods
Subjects
We conducted a feeding experiment during a 3-week period between March and April 2014 on male yellow-breasted capuchins (Sapajus xanthosternos) from two different social groups kept in Zoo Frankfurt, Germany. Age ranged from 1 to 28years. One group consisted of five males and one female (group A), and the other group consisted of three males (group B). We excluded one male from group A from further analysis, as sample size for the male was too low. We also excluded the only female because statistical analysis of one individual would not be possible and would not be representative of the female sex. All individuals were in good health at the time of data collection. The capuchins were fed with a mixture of fruits and vegetables and water was available ad libitum during the whole study period.
Fecal sample collection
All fecal samples were collected throughout the day (09:00h–17:00h) from the ground when explicitly assignable to the defecator and when not contaminated with urine. Fresh fecal samples were collected in plastic tubes and immediately stored at −20°C. After the termination of the experiment, samples were transported frozen to the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany (MPI-EVA).
Sample preparation and TT3 assay
In the lab at MPI-EVA, we freeze-dried all samples for 24h at −40°C, ground and then stored them in whirl packs at −20°C. Seventy-two samples from seven males underwent an ethanol–water extraction (70%) following Wasser et al. (2010) with the following modifications: For each sample, 7.5mL ethanol–water mix (70%) were added to 0.05g freeze-dried fecal powder, vortexed for 30min and then centrifuged for 20min at 1040g. The supernatant was decanted into a fresh vial, the procedure was repeated and the resulting supernatant was combined with the first.
We took 2400µL of ethanol extract, evaporated the extract under nitrogen at 45°C and reconstituted the extract in 120µL TT3 null standard (a buffer provided by IBL International GmbH, Hamburg, Germany, consisting of human serum and stabilizers). Immunoreactive T3 (iT3) was measured with a commercial, competitive TT3 ELISA (Ref. RE55251, IBL International GmbH) for human serum. Samples, standards and quality controls (50µL per well) were run in duplicates, following the instructions from the supplier. Results were calculated as ng per g dried feces.
Out of 72 samples, 15 were above the linear range of the assay and, therefore, needed re-measurement at a concentration of 10:1 with 1200µL EtOH extract reconstituted in 120µL TT3 null standard. Values for two of the samples were discarded from the data set because one was unrealistically high (more than three standard deviations above the mean), indicating a potential unobserved contamination with urine, and the other because we could not achieve consistent values despite repeated measurements.
Analytical validation
Extraction efficiency, parallelism, assay accuracy, inter- and intra-assay coefficients of variance, stability
For the analytical validation of the TT3 assay, we conducted the following steps:
To assess extraction efficiency, four samples were spiked with two different concentrations of standard (high: 0.25ng/mL, low: 0.1ng/mL) before the evaporation step and subsequently measured with the assay. It is not feasible to reliably spike dry fecal powder in a homogeneous manner. Therefore, and to at least assess the extraction efficiency for part of the extraction process, we spiked the ethanol extract and referred to this as extraction efficiency throughout the article. Both the assay recovery and the extraction efficiency were calculated following Behringer et al. (2012).
To test for parallelism, a pool sample (210µL with a concentration of 2.67ng/mL extract) consisting of two male samples was spiked with 40µL of 30ng/mL standard (3,3′,5-triiodo-l-thyronine (T3) purchased from Sigma-Aldrich, CAS number 6893-02-3, batch number T2877) and diluted serially. As the concentration of T3 in fecal samples of this species is relatively low, spiking the pool sample with a standard enabled us to obtain a reasonable dilution curve, which allowed us to estimate relevant matrix effects around the sample concentrations at which the samples are routinely measured.
To assess the accuracy of the assay, a recovery experiment was conducted. After samples underwent extraction, four extracts with known concentration were spiked with high: 2.5ng/mL and low: 1ng/mL standard concentrations, respectively.
Inter-assay coefficients of variance (CV) of high and low concentrations were 3.41% and 6.26%, respectively (n=5). Intra-assay variance was determined by running one high- and one low concentrated pool sample (concentration high pool sample: 61.6ng/g dried feces; concentration low pool sample: 28.9ng/g dried feces), each consisting of six samples with known concentration, equally distributed over one assay plate. Intra-assay CVs were 4.86% for high and 4.08% low concentrations.
To assess whether TH concentrations degrade over time, we tested a subset of 15 fecal samples 17months after the initial measurements were conducted. We tested two potential storage methods, fecal powder and ethanol extract, which are two successive steps in the sample preparation (as explained in detail above). Both the fecal powder and the ethanol extract were stored at −20°C during the 17-month period.
Biological validation: feeding experiment
During days of regular feeding, the two capuchin groups were fed with the usual amount of food (normal caloric intake); during days of reduced caloric intake, the groups received half of the usual amount of calories (reduced caloric intake). To keep overall food mass constant, during the period of reduced caloric intake, we replaced the amount of calorie-rich food items with low-calorie food items. Therefore, during both feeding regimes, the same food items were fed, however, in relatively different quantities. For each group, all fruits and vegetables were weighed before being fed. On the following day, the remaining food was collected and weighed again. We calculated the daily energy intake per group by subtracting the energy content of the food recovered from the energy content of the food provided for each day. The mean caloric intake in group A was 2562.3kcal (s.d.±425; range: 2080.3–3257kcal) and 1637.5kcal (s.d.±244; range: 1471.8–2229.5kcal) in group B during the periods of normal caloric intake. This corresponded to an estimated mean caloric intake per individual of 427.05kcal in group A and 545.8kcal in group B. In contrast, during the periods of reduced feeding, mean caloric intake was 1603.7kcal (s.d.±343; range: 1234.9–1979.5kcal) in group A and 928.8kcal (s.d.±84.5; range: 838.1–1031.7kcal) in group B, which corresponded to an estimated mean individual caloric intake of 267.3kcal in group A and 309.6kcal in group B.
Statistical analysis
To assess the impact of changes in caloric intake on fecal iT3 levels, we ran a general linear mixed model (Baayen 2008), with iT3 levels being the response variable and daily caloric intake being the test predictor with fixed effect. The model was fitted in R (R Core Team 2014) using the function lmer of the R-package lme4 (Bates et al. 2014). We included age of the individual, sampling time and group identity as control predictors with fixed effects. Individual ID was included as a random effect, and we included random slopes of caloric intake and time of sampling within individual ID to keep type I error rate at the nominal level of 5% (Schielzeth & Forstmeier 2009, Barr et al. 2013). We determined model stability by excluding individuals one at a time (see Results section for details) and checked for problems with collinearity by determining variance inflation factors (VIF, Field 2009) based on a standard linear model excluding the random effects, both of which were revealed not to be an issue (maximum VIF: 6.31). To meet the criterion of normal distribution, we subtracted the minimum iT3 value from all iT3 values and then square root transformed them. All quantitative fixed effects were z-transformed to a mean of zero and a standard deviation of 1 (Schielzeth 2010). Control predictors and random effects were kept in the reduced model. To establish the significance of the individual predictors, we used likelihood ratio tests comparing the full with a respective reduced model lacking the particular predictor (Barr et al. 2013).
To test for parallelism, we ran a linear model with the interaction between the respective sample type (standard curve vs pool sample) and the concentration of the standard, with the percent binding as the response. The model was fitted in R (R Core Team 2014) using the function lm. We checked for the assumptions of normality and homogeneity of the residuals by inspecting a qq-plot of the residuals and residuals plotted against fitted values (Field 2009), which did not indicate any problems. We further assessed model stability by means of DFBeta (Field 2009), which indicated the model to be moderately stable. To test for the stability of TH across two potential storage methods, we used an exact Wilcoxon signed-rank test using the R-package exactRankTests (Hothorn & Hornik 2015) to assess whether TH concentrations significantly degrade during storage. In addition, to assess whether potential degradation processes affect relative TH levels, we used a Spearman’s correlation.
Results
Analytical validation
Extraction efficiency, parallelism, assay accuracy, inter- and intra-assay coefficients of variance, stability
The mean extraction efficiency was 102.94% (s.d.=11.17, n=4) for the high-spiked extracts and 75.49% (s.d.=14.16, n=4) for the low-spiked extracts. The serially diluted pool sample was parallel to the standard curve (t(6)=−0.795, P=0.4568; confirmed by visual inspection, Supplementary Fig. 1, see section on supplementary data given at the end of this article). The assay accuracy was on average of 85.3% (s.d.=9.08, n=4) for the high-spiked samples and 93.14% (s.d.=3.3, n=4) for the low-spiked samples. IT3 levels were significantly lower compared with initial measurements when repeated after 17months (original measurement compared with re-measurement of fecal powder: Wilcoxon signed-rank test: T+=120, n=15, P<0.001; original measurement compared with re-measurement of ethanol extract: T+=120, n=15, P<0.001). The average proportion of iT3 loss when re-extracting samples from fecal powder after 17months was 42.8% (s.d.=10.8; range: 26.19–60.43%) and 27.66% (s.d.=12.26; range: 9.77–52.43%) from ethanol extracts. However, the measurements at both time points were strongly and significantly correlated for both conditions (Fig. 1; original measurement and re-measurement of fecal powder: rs=0.92, n=15, P<0.001; original measurement and re-measurement of ethanol extract: rs=0.975, n=15, P<0.001).
Loss of immunoreactive T3 in (A) fecal powder and (B) ethanol extract after a storage period of 17months at −20°C. Depicted are regression lines for each condition.
Citation: Journal of Endocrinology 231, 1; 10.1530/JOE-16-0152
Biological validation: feeding experiment
The mean fecal iT3 level across individuals during days of normal caloric intake was 31.03ng/g dried feces (n=43; s.d.±19.36) and 18.59ng/g dried feces (n=27; s.d.±12.39) during the periods of restricted caloric intake (Fig. 2). The full model was highly significant as compared with the reduced model (χ2=8.34, df=1, P=0.004), indicating that caloric intake per day was a significant predictor of iT3 levels (Figs 2, 3 and Table 1). Levels of iT3 were significantly lower on days when caloric intake was restricted, as indicated by the positive estimate derived by the model.
Change in fecal immunoreactive T3 levels in male yellow-breasted capuchins during high (normal) and low (restricted) caloric intake periods. Each point represents average values of one individual. The dashed lines connect points referring to the same male (n=7 males with an average of ten samples per male).
Citation: Journal of Endocrinology 231, 1; 10.1530/JOE-16-0152
Individual fecal immunoreactive T3 levels in male yellow-breasted capuchins changing with caloric intake per group and day. Depicted are regression lines for each individual.
Citation: Journal of Endocrinology 231, 1; 10.1530/JOE-16-0152
Statistical summary of the impact of caloric intake on fecal iT3 levels of male yellow-breasted capuchins; significant effects are indicated in bold.
| Term | Estimate | s.e.m. | χ2 | Df | P | Minb | Maxb |
|---|---|---|---|---|---|---|---|
| Intercept | 3.467 | 0.513 | a | a | a | 2.951 | 3.659 |
| Intake calories per day | 0.736 | 0.245 | 8.336 | 1 | 0.004 | 0.680 | 0.796 |
| Age at sampling | −0.130 | 0.225 | 0.317 | 1 | 0.573 | −0.426 | −0.049 |
| Sampling time | −0.504 | 0.486 | 1.053 | 1 | 0.305 | −0.831 | −0.353 |
| Group | 1.927 | 1.006 | 3.511 | 1 | 0.061 | 1.275 | 2.791 |
Not indicated because of having no meaningful interpretation
Minimum and maximum of the model stability.
Additionally, we found a considerable overlap of iT3 levels across individuals between days of low and high caloric intake (Fig. 2). Several individuals had lower iT3 levels during days of normal caloric intake than the highest average level achieved by an individual during days of restricted caloric intake.
Discussion
With this study, we were able to successfully validate the measurement of iT3 in fecal samples of male yellow-breasted capuchin monkeys. By conducting experiments to test for extraction efficiency, parallelism and assay accuracy, as well as providing information about the reproducibility of measurements within and between plates, we successfully conducted an analytical validation of the measurement of iT3 levels in fecal samples of this species. Furthermore, by conducting a feeding restriction experiment and monitoring changes in individual iT3 levels of seven males while at the same time linking them to group levels of daily caloric intake, we showed that individual fecal iT3 levels were significantly correlated with the caloric intake of the entire group. Due to the noninvasive design of this study, we were restricted to linking iT3 levels solely to energy intake. However, previous studies found that TH concentrations reflect energy balance (Blum et al. 1980, Harlow & Seal 1981, Menchetti et al. 2015). Therefore, although in this study we could only show strong correlations between energy intake and fecal iT3 levels, measuring iT3 levels may have the potential to be a powerful alternative method for assessing energy balance in animals in which serum or urine collection is impossible.
Detailed studies on degradation patterns of TH in fecal samples during storage are rare; however, degradation of TH was reported for fecal samples of African elephants (Wasser et al. 2010). Our results indicate a significant degree of degradation of TH for both storage methods. Comparing the degree of degradation of the two storage methods suggests that storing fecal samples in the form of ethanol extracts is highly preferable to storing them as dried fecal powder. Taken together, this suggests that within one project, the time between sampling and measuring should be kept short in order to avoid hormone degradation. Alternatively, if that is not possible, the delay should be kept as constant as possible so that measurements are comparable. In cases where this is not possible, a detailed analysis of the degree of degradation is essential, and if this reveals to be important, the time from sampling to measurement should be included as a control variable in the statistical analysis.
In our study, changes in caloric intake were reflected in fecal iT3 levels on the same day. In carnivores, serum T3 levels also reflect reduced caloric intake when measured within the same day (Harlow & Seal 1981, Delgiudice et al. 1987). Due to the time delay associated with excretion via the gut, it would be expected that changes in fecal iT3 levels are detected later than those in serum. However, smaller bodied primates, to which capuchins and most of the other new world primates belong to, have a very fast gut passage time (e.g., 8–24h in common marmosets: Bahr et al. 2000 and 2.1–3.0h in tufted capuchins: Wheeler et al. 2013 as estimated by ACTH challenge). This fast response of hormone excretion into feces would not be expected for larger bodied primates because they are characterized by a rather long gut passage time (e.g., 2–3days in western lowland gorillas, Shutt et al. (2012), and around 29h in chimpanzees, Murray et al. (2013), as estimated by ACTH challenge). Therefore, when adapting this methodology to other animals, species-specific information on gut passage times should be taken into account to properly detect the endocrinological signal of iT3 excretion in feces in relation to fluctuations in caloric intake.
Our results revealed considerable intra- and inter-individual differences in iT3 levels under both feeding conditions and an overlap in levels between the conditions across individuals, for example, some animals had higher iT3 levels during the days of caloric restriction than others had during the days of normal feeding regime. In this study, we were limited to measures of daily caloric intake on a group rather than on an individual level. Therefore, the variability in individual levels within a feeding regime could be due to the differences in individual energy intake and expenditure. To explain additional variance in individual iT3 levels within and between individuals, in future studies, efforts should be made to include more control variables, such as individual caloric intake, energy expenditure and body mass.
TH are involved in developmental processes (for review, see e.g., Crockford 2006), with levels remaining high after birth and during development and then characteristically decreasing after reaching adulthood as shown in humans (Oliner et al. 1957, Beckers et al. 1966, Ryness 1972, Hesch et al. 1977), bonobos and chimpanzees (Behringer et al. 2014). In our study, one subadult individual of approximately 3years of age did not show the expected decline in iT3 levels when caloric intake was restricted. It is possible that individuals during growth phases are less sensitive to environmental stimuli and, therefore, keep metabolic rates constant, even if caloric intake is restricted. This suggests that individuals in different life history stages (e.g., development or pregnancy) can show different energy allocation patterns according to their respective needs. However, as we did not collect individual intake rates, another possibility that would explain this lack of reactivity in iT3 levels is that this individual had a higher energy intake rate than the rest of the group and hence its iT3 levels did not change across the conditions.
This novel method to assess energetic condition by measuring iT3 in noninvasively collected fecal samples provides the possibility to investigate research questions about trade-offs in relation to somatic growth, maintenance and reproduction. For example, mate guarding, a behavior many male primates conduct to monopolize the fertilization of females, potentially comes along with metabolic costs to the males caused by higher rates of agonistic and sexual behaviors and less time spent on feeding (e.g., Alberts et al. 1996, Matsubara 2003, Ancona et al. 2010, Girard-Buttoz et al. 2014). This suggests that energetic condition might be negatively affected by mate-guarding events. Future studies using iT3 levels to monitor energetic condition can now focus on energy allocation during energetically constraining phases, for example, mating season or pregnancy.
To conclude with this study, we have shown that fecal iT3 levels of a nonhuman primate species can be used as a reliable physiological marker of energy intake in a noninvasively collected medium. In the future, fecal iT3 measurements might be used to explore to what extent social, developmental and physiological factors impact individual decisions on energy allocation into somatic growth, maintenance or storage and reproduction. However, detailed validations as exemplified in this study are necessary when this method is applied to a different species or a different matrix such as urine. Importantly, it has to be kept in mind that iT3 levels varied considerably within and between individuals, which means that defining a general threshold of high vs low iT3 levels to assign an individual to a category of energetic condition is not feasible. Rather, this inter-individual variability in iT3 levels necessitates the implementation of within-subject comparisons in which each individual serves as its own control. In addition, to account for within-individual variation, several samples per individual and nutritional state should be analyzed. Finally, future studies should aim at including detailed data on individual intake rates of calories as well as individual rates of energy expenditure to enable more precise insights into patterns of energy allocation in the study species.
Supplementary data
This is linked to the online version of the paper at http://dx.doi.org/10.1530/JOE-16-0152.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research was supported by the Max Planck Society.
Acknowledgements
The support from the curator (Dr Wilms), the veterinaries (Drs Geiger and Schauerte) and the keepers (Harald Thomas, Annie Fuchs and Richard Weigl) of Frankfurt Zoo is gratefully acknowledged. The authors are grateful to Roisin Murtagh for laboratory assistance and advice. The authors are also grateful to Roger Mundry for help in statistical analysis and to Mimi Arandjelovic for editing the manuscript. The authors also thank two anonymous reviewers for their helpful and constructive comments, which significantly improved the quality of this article. The protocol for the restriction study, including the collection of fecal samples, was approved by the regional council of the city of Darmstadt, number: V54 – 19c 20/15, and the authorities of Frankfurt Zoo.
References
Abdullah R & Falconer IR 1977 Responses of thyroid activity to feed restriction in the goat. Australian Journal of Biological Sciences 30 207–216. (doi:10.1071/BI9770207)
Alberts SC, Altmann J & Wilson ML 1996 Mate guarding constrains foraging activity of male baboons. Animal Behaviour 51 1269–1277. (doi:10.1006/anbe.1996.0131)
Ancona S, Drummond H & Zaldívar-Rae J 2010 Male whiptail lizards adjust energetically costly mate guarding to male–male competition and female reproductive value. Animal Behaviour 79 75–82. (doi:10.1016/j.anbehav.2009.10.005)
Ayres KL, Booth RK, Hempelmann JA, Koski KL, Emmons CK, Baird RW, Balcomb-Bartok K, Hanson MB, Ford MJ & Wasser SK 2012 Distinguishing the impacts of inadequate prey and vessel traffic on an endangered killer whale (Orcinus orca) population. PLoS ONE 7 e36842. (doi:10.1371/journal.pone.0036842)
Baayen RH 2008 Analyzing Linguistic Data: A Practical Introduction to Statistics Using R. Cambridge, UK; New York: Cambridge University Press.
Bahr NI, Palme R, Möhle U, Hodges JK & Heistermann M 2000 Comparative aspects of the metabolism and excretion of cortisol in three individual nonhuman primates. General and Comparative Endocrinology 117 427–438. (doi:10.1006/gcen.1999.7431)
Barr DJ, Levy R, Scheepers C & Tily HJ 2013 Random effects structure for confirmatory hypothesis testing: keep it maximal. Journal of Memory and Language 68 255–278. (doi:10.1016/j.jml.2012.11.001)
Bates D, Meachler M, Bolker B & Walker S 2014 lme4: linear mixed-effects models using Eigen and S4. R package version 1.1-7. (available at: https://cran.r-project.org/web/packages/lme4/index.html)
Beckers C, Malvaux P & De Visscher M 1966 Quantitative aspects of the secretion and degradation of thyroid hormones during adolescence. Journal of Clinical Endocrinology & Metabolism 26 202–206. (doi:10.1210/jcem-26-2-202)
Behringer V, Hohmann G, Stevens JMG, Weltring A & Deschner T 2012 Adrenarche in bonobos (Pan paniscus): evidence from ontogenetic changes in urinary dehydroepiandrosterone-sulfate levels. Journal of Endocrinology 214 55–65. (doi:10.1530/JOE-12-0103)
Behringer V, Deschner T, Murtagh R, Stevens JMG & Hohmann G 2014 Age-related changes in thyroid hormone levels of bonobos and chimpanzees indicate heterochrony in development. Journal of Human Evolution 66 83–88. (doi:10.1016/j.jhevol.2013.09.008)
Bercovitch FB & Nurnberg P 1996 Socioendocrine and morphological correlates of paternity in rhesus macaques (Macaca mulatta). Reproduction 107 59–68. (doi:10.1530/jrf.0.1070059)
Blake NG, Eckland DJA, Foster OJF & Lightman SL 1991 Inhibition of hypothalamic thyrotropin-releasing hormone messenger ribonucleic acid during food deprivation. Endocrinology 129 2714–2718. (doi:10.1210/endo-129-5-2714)
Blum JW, Gingins M, Vitins P & Bickel H 1980 Thyroid hormone levels related to energy and nitrogen balance during weight loss and regain in adult sheep. European Journal of Endocrinology 93 440–447. (doi:10.1530/acta.0.0930440)
Chivers DJ 1998 Measuring food intake in wild animals: primates. Proceedings of the Nutrition Society 57 321–332. (doi:10.1079/PNS19980047)
Coelho AM 1974 Socio-bioenergetics and sexual dimorphism in primates. Primates 15 263–269. (doi:10.1007/BF01742287)
Cristóbal-Azkarate J, Maréchal L, Semple S, Majolo B & MacLarnon A 2016 Metabolic strategies in wild male Barbary macaques: evidence from faecal measurement of thyroid hormone. Biology Letters 12 20160168. (doi:10.1098/rsbl.2016.0168)
Crockford SJ 2006 Rhythms of Life: Thyroid Hormone & the Origin of Species. Victoria, BC, Canada: Trafford.
Danforth E & Burger AG 1989 The impact of nutrition on thyroid hormone physiology and action. Annual Review of Nutrition 9 201–227. (doi:10.1146/annurev.nu.09.070189.001221)
Danforth E, Horton ES, O’Connell M, Sims EA, Burger AG, Ingbar SH, Braverman L & Vagenakis AG 1979 Dietary-induced alterations in thyroid hormone metabolism during overnutrition. Journal of Clinical Investigation 64 1336–1347. (doi:10.1172/JCI109590)
Delgiudice GD, Seal US & Mech D 1987 Effects of feeding and fasting on wolf blood and urine characteristics. Journal of Wildlife Management 51 1–10. (doi:10.2307/3801619)
Deschner T, Kratzsch J & Hohmann G 2008 Urinary C-peptide as a method for monitoring body mass changes in captive bonobos (Pan paniscus). Hormones and Behavior 54 620–626. (doi:10.1016/j.yhbeh.2008.06.005)
DiStefano JJ & Sapin V 1987 Fecal and urinary excretion of six iodothyronines in the rat. Endocrinology 121 1742–1750. (doi:10.1210/endo-121-5-1742)
Eales JG 1988 The influence of nutritional state on thyroid function in various vertebrates. Integrative and Comparative Biology 28 351–362. (doi:10.1093/icb/28.2.351)
Eley RM, Strum SC, Muchemi G & Reid GDF 1989 Nutrition, body condition, activity patterns, and parasitism of free-ranging troops of olive baboons (Papio anubis) in Kenya. American Journal of Primatology 18 209–219. (doi:10.1002/ajp.1350180304)
Emery Thompson M & Knott CD 2008 Urinary C-peptide of insulin as a non-invasive marker of energy balance in wild orangutans. Hormones and Behavior 53 526–535. (doi:10.1016/j.yhbeh.2007.12.005)
Emery Thompson M, Muller MN, Wrangham RW, Lwanga JS & Potts KB 2009 Urinary C-peptide tracks seasonal and individual variation in energy balance in wild chimpanzees. Hormones and Behavior 55 299–305. (doi:10.1016/j.yhbeh.2008.11.005)
Emery Thompson M, Muller MN & Wrangham RW 2012 The energetics of lactation and the return to fecundity in wild chimpanzees. Behavioral Ecology 23 1234–1241. (doi:10.1093/beheco/ars107)
Field AP 2009 Discovering Statistics Using SPSS: (and Sex, Drugs and Rock “N” Roll). Los Angeles, CA, USA: SAGE Publications.
Flier JS, Harris M & Hollenberg AN 2000 Leptin, nutrition, and the thyroid: the why, the wherefore, and the wiring. Journal of Clinical Investigation 105 859–861. (doi:10.1172/JCI9725)
Fontana L, Klein S, Holloszy JO & Premachandra BN 2006 Effect of long-term calorie restriction with adequate protein and micronutrients on thyroid hormones. Journal of Clinical Endocrinology & Metabolism 91 3232–3235. (doi:10.1210/jc.2006-0328)
French SS, Moore MC & Demas GE 2009 Ecological immunology: the organism in context. Integrative and Comparative Biology 49 246–253. (doi:10.1093/icb/icp032)
Galofré JC, Frühbeck G & Salvador J 2010 Obesity and thyroid function: pathophysiological and therapeutic implications. Hot Thyroidology 6 2075–2202.
Girard-Buttoz C, Higham JP, Heistermann M, Wedegärtner S, Maestripieri D & Engelhardt A 2011 Urinary c-peptide measurement as a marker of nutritional status in macaques. PLoS ONE 6 e18042. (doi:10.1371/journal.pone.0018042)
Girard-Buttoz C, Heistermann M, Rahmi E, Marzec A, Agil M, Fauzan PA & Engelhardt A 2014 Mate-guarding constrains feeding activity but not energetic status of wild male long-tailed macaques (Macaca fascicularis). Behavioral Ecology and Sociobiology 68 583–595. (doi:10.1007/s00265-013-1673-8)
Gobush KS, Booth RK & Wasser SK 2014 Validation and application of noninvasive glucocorticoid and thyroid hormone measures in free-ranging Hawaiian monk seals. General and Comparative Endocrinology 195 174–182. (doi:10.1016/j.ygcen.2013.10.020)
Goymann W 2012 On the use of non-invasive hormone research in uncontrolled, natural environments: the problem with sex, diet, metabolic rate and the individual: non-invasive hormone research. Methods in Ecology and Evolution 3 757–765. (doi:10.1111/j.2041-210X.2012.00203.x)
Grueter CC, Deschner T, Behringer V, Fawcett K & Robbins MM 2014 Socioecological correlates of energy balance using urinary C-peptide measurements in wild female mountain gorillas. Physiology & Behavior 127 13–19. (doi:10.1016/j.physbeh.2014.01.009)
Harlow HJ & Seal US 1981 Changes in hematology and metabolites in the serum and urine of the badger, Taxidea taxus, during food deprivation. Canadian Journal of Zoology 59 2123–2128. (doi:10.1139/z81-289)
Harris TR, Chapman CA & Monfort SL 2010 Small folivorous primate groups exhibit behavioral and physiological effects of food scarcity. Behavioral Ecology 21 46–56. (doi:10.1093/beheco/arp150)
Heistermann M 2010 Non-invasive monitoring of endocrine status in laboratory primates: methods, guidelines and applications. Advances in Science and Research 5 1–9. (doi:10.5194/asr-5-1-2010)
Hesch RD, Gatz J, Jüppner H & Stubbe P 1977 TBG-dependency of age related variations of thyroxine and triiodothyronine. Hormone and Metabolic Research 9 141–146. (doi:10.1055/s-0028-1093563)
Higham JP, Heistermann M & Maestripieri D 2011 The energetics of male–male endurance rivalry in free-ranging rhesus macaques, Macaca mulatta. Animal Behaviour 81 1001–1007. (doi:10.1016/j.anbehav.2011.02.001)
Holloszy JO & Fontana L 2007 Caloric restriction in humans. Experimental Gerontology 42 709–712. (doi:10.1016/j.exger.2007.03.009)
Hothorn T & Hornik K 2015 exactRankTests: Exact Distributions for Rank and Permutation Tests. R package version 0.8-28. (available at: http://CRAN.R-project.org/package=exactRankTests).
Ingbar DH & Galton AV 1975 The effect of food deprivation on the peripheral metabolism of thyroxine in rats. Endocrinology 96 1525–1532. (doi:10.1210/endo-96-6-1525)
Janson C & Vogel E 2006 Hunger and aggression in capuchin monkeys. In Feeding Ecology in Apes and Other Primates, pp 285–312. Eds Hohmann G, Robbins MM, Boesch C. Cambridge; New York: Cambridge University Press.
Joly K, Wasser SK & Booth R 2015 Non-invasive assessment of the interrelationships of diet, pregnancy rate, group composition, and physiological and nutritional stress of barren-ground caribou in late winter. PLoS ONE 10 e0127586. (doi:10.1371/journal.pone.0127586)
Kaack B, Walker L, Brizzee KR & Wolf RH 1979 Comparative normal levels of serum triiodothyronine and thyroxine in nonhuman primates. Laboratory Animal Science 29 191–194.
Kim B 2008 Thyroid hormone as a determinant of energy expenditure and the basal metabolic rate. Thyroid 18 141–144. (doi:10.1089/thy.2007.0266)
Knott CD 1998 Changes in orangutan caloric intake, energy balance, and ketones in response to fluctuating fruit availability. International Journal of Primatology 19 1061–1079. (doi:10.1023/A:1020330404983)
Köhrle J 1999 Local activation and inactivation of thyroid hormones: the deiodinase family. Molecular and Cellular Endocrinology 151 103–119. (doi:10.1016/S0303-7207(99)00040-4)
Matsubara M 2003 Costs of mate guarding and opportunistic mating among wild male Japanese macaques. International Journal of Primatology 24 1057–1075. (doi:10.1023/A:1026228312706)
Menchetti L, Brecchia G, Canali C, Cardinali R, Polisca A, Zerani M & Boiti C 2015 Food restriction during pregnancy in rabbits: effects on hormones and metabolites involved in energy homeostasis and metabolic programming. Research in Veterinary Science 98 7–12. (doi:10.1016/j.rvsc.2014.11.017)
Merimee TJ & Fineberg ES 1976 Starvation-induced alterations of circulating thyroid hormone concentrations in man. Metabolism 25 79–83. (doi:10.1016/0026-0495(76)90162-1)
Muller MN & Wrangham RW 2004 Dominance, cortisol and stress in wild chimpanzees (Pan troglodytes schweinfurthii). Behavioral Ecology and Sociobiology 55 332–340. (doi:10.1007/s00265-003-0713-1)
Murray CM, Heintz MR, Lonsdorf EV, Parr LA & Santymire RM 2013 Validation of a field technique and characterization of fecal glucocorticoid metabolite analysis in wild chimpanzees (Pan troglodytes): chimpanzee fecal glucocorticoid metabolite method. American Journal of Primatology 75 57–64. (doi:10.1002/ajp.22078)
Oliner L, Kohlenbrener RM, Fields T, Kunstadter RH & Goldstein DR 1957 Thyroid function studies in children: normal values for thyroidal i 131 uptake and pbi 131 levels up to the age of 18*. Journal of Clinical Endocrinology & Metabolism 17 61–75. (doi:10.1210/jcem-17-1-61)
Palme R, Rettenbacher S, Touma C, El-Bahr SM & Möstl E 2005 Stress hormones in mammals and birds: comparative aspects regarding metabolism, excretion, and noninvasive measurement in fecal samples. Annals of the New York Academy of Sciences 1040 162–171. (doi:10.1196/annals.1327.021)
Perrin N & Sibly RM 1993 Dynamic models of energy allocation and investment. Annual Review of Ecology and Systematics 24 379–410. (doi:10.1146/annurev.es.24.110193.002115)
Pontzer H, Raichlen DA, Gordon AD, Schroepfer-Walker KK, Hare B, O’Neill MC, Muldoon KM, Dunsworth HM, Wood BM & Isler K et al. 2014 Primate energy expenditure and life history. PNAS 111 1433–1437. (doi:10.1073/pnas.1316940111)
Core Team R 2014 R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing.
Roff DA 2002 Life History Evolution. Sunderland, MA, USA: Sinauer Associates.
Roff DA & Fairbairn DJ 2007 The evolution of trade-offs: where are we? Journal of Evolutionary Biology 20 433–447. (doi:10.1111/j.1420-9101.2006.01255.x)
Ryness J 1972 The measurement of serum thyroxine in children. Journal of Clinical Pathology 25 726–729. (doi:10.1136/jcp.25.8.726)
Schielzeth H 2010 Simple means to improve the interpretability of regression coefficients: interpretation of regression coefficients. Methods in Ecology and Evolution 1 103–113. (doi:10.1111/j.2041-210X.2010.00012.x)
Schielzeth H & Forstmeier W 2009 Conclusions beyond support: overconfident estimates in mixed models. Behavioral Ecology 20 416–420. (doi:10.1093/beheco/arn145)
Shakespear RA & Burke CW 1976 Triiodothyronine and thyroxine in urine. I. Measurement and application. Journal of Clinical Endocrinology & Metabolism 42 494–503. (doi:10.1210/jcem-42-3-494)
Sherry DS & Ellison PT 2007 Potential applications of urinary C-peptide of insulin for comparative energetics research. American Journal of Physical Anthropology 133 771–778. (doi:10.1002/ajpa.20562)
Shutt K, Setchell JM & Heistermann M 2012 Non-invasive monitoring of physiological stress in the Western lowland gorilla (Gorilla gorilla gorilla): validation of a fecal glucocorticoid assay and methods for practical application in the field. General and Comparative Endocrinology 179 167–177. (doi:10.1016/j.ygcen.2012.08.008)
Silva JE 1995 Thyroid hormone control of thermogenesis and energy balance. Thyroid 5 481–492. (doi:10.1089/thy.1995.5.481)
Stearns SC 1992 The Evolution of Life Histories. Oxford; New York: Oxford University Press.
Stearns SC & Hoekstra RF 2005 Evolution: An Introduction. Oxford, UK; New York: Oxford University Press.
Surbeck M, Deschner T, Behringer V & Hohmann G 2015 Urinary C-peptide levels in male bonobos (Pan paniscus) are related to party size and rank but not to mate competition. Hormones and Behavior 71 22–30. (doi:10.1016/j.yhbeh.2015.03.007)
Wasser SK, Azkarate JC, Booth RK, Hayward L, Hunt K, Ayres K, Vynne C, Gobush K, Canales-Espinosa D & Rodríguez-Luna E 2010 Non-invasive measurement of thyroid hormone in feces of a diverse array of avian and mammalian species. General and Comparative Endocrinology 168 1–7. (doi:10.1016/j.ygcen.2010.04.004)
Wheeler BC, Tiddi B, Kalbitzer U, Visalberghi E & Heistermann M 2013 Methodological considerations in the analysis of fecal glucocorticoid metabolites in tufted capuchins (Cebus apella). International Journal of Primatology 34 879–898. (doi:10.1007/s10764-013-9703-y)
