AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 137:441–448 (2008) Ecological and Reproductive Variance in Serum Leptin in Wild Vervet Monkeys Patricia L. Whitten1* and Trudy R. Turner2,3 1 Department of Anthropology, Emory University, Atlanta, GA 30322 Department of Anthropology, University of Wisconsin-Milwaukee, Milwaukee, WI 53201 3 Mammal Research Institute, University of Pretoria, 0002 Pretoria, Republic of South Africa 2 KEY WORDS adiposity; steroids; energy balance; primates ABSTRACT Leptin was originally thought to be an antiobesity hormone, but increasing evidence suggests that its ancestral role was to mobilize neuroendocrine responses to starvation. Research on wild primates is critical for interpreting the high leptin values seen in Western human populations and captive animals. This study examined natural variation in serum leptin in wild vervet monkeys (Chlorocebus aethiops), testing the hypothesis that serum leptin in vervets varies with sex, adiposity, ecology, and reproductive state. Analyses made use of a unique dataset comprised of serum and morphometric measurements obtained from vervet monkeys in four Kenyan sites differing in altitude, temperature, rainfall, and access to human foods. Leptin and gonadal steroid concentrations were analyzed in serum samples from 116 adults. Low leptin levels in males and acyclic females support the contention that levels seen in captivity are not typical for wild primates. Measures of adiposity were not correlated with serum leptin, reﬂecting the extremely low fat storage in wild cercopithecine primates. Associations with habitat and season, however, indicate that leptin does register ecological variation in energy balance. Leptin levels were higher in sites and seasons with higher rainfall. Moreover, leptin varied signiﬁcantly with reproductive state, with higher levels in pregnant than in acyclic females. Changes in leptin with gestation stage and duration of lactation suggest that transitory and reversible elevations were an important part of its ancestral role. These data show that in this wild primate population leptin is a sensitive index of natural variation in habitat and seasonally ﬂuctuating reproductive state. Am J Phys Anthropol 137:441–448, 2008. V 2008 Wiley-Liss, Inc. Leptin is a protein produced by adipocytes that regulates food intake and energy expenditure (Zhang et al., 1994; Friedman and Halaas, 1998; Jéquier, 2002). Leptin was originally identiﬁed as a lipostat and antiobesity hormone, but subsequent research has cast doubt upon this characterization (Ahima and Osei, 2004). Although leptin has profound negative effects on appetite and fat mass in the rare cases of human mutations in leptin or leptin receptor genes, it is ineffective in most cases of the human obesity (Ahima and Osei, 2004; Gale et al., 2004). This phenomenon, termed leptin resistance, may result in part from a deﬁciency in its transport across the blood-brain barrier (Banks, 2003). The transporter is saturated at the high serum levels typical of obesity (30 ng/ml) and is partially saturated even at leptin levels considered normal in humans (Banks et al., 2000). Only at leptin levels (1 ng/ml) characteristic of anorexic humans is the transporter rapid and effective (Banks et al., 2000). These paradoxical results have been explained by an evolutionary model positing that the leptin feedback loop evolved at much lower levels of serum leptin and adiposity than those considered normal in Western human populations (Banks, 2003; Banks et al., 2006). In this ‘‘Absence of Protection’’ model (Schwartz and Niswender, 2004), an energy dense diet results in obesity because no mechanisms have evolved to confront a situation (high leptin levels) that almost never occurred. For example, anthropologists have shown that the serum leptin levels are very low and unrelated to adiposity in lean males from nonWesternized human populations (Bribiescas, 2001; Lindgärde et al., 2004; Kuzawa et al., 2007; Sharrock et al., 2008). U.S. Pima Indians following a Western lifestyle have four-fold higher leptin concentrations than Mexican Pima Indians living a more traditional lifestyle (Fox et al., 1999). On the other hand, only modest increases (34%) in serum leptin are associated with a shift from hunting and gathering to agriculture over the last 30 years in one community of Shuar Amerindian women (Lindgärde et al., 2004). The broad range of functions affected by leptin suggest that leptin may have evolved as a signal of the adequacy of energy reserves to support costly activities such as reproduction, pubertal onset, and immune function (Banks, 2003; Banks et al., 2006). In the language of life history theory, leptin is a mechanism for enacting optimal resource allocation rules (Alonso-Alvarez et al., 2007). In the case of reproduction, this allocation appears to be accomplished via leptin action on GnRH neurons in the arcuate nucleus (Chan and Mantzoros, 2001). However, this model neglects an important piece of the leptin puzzle. In humans and many other animal species, energy balance and adiposity are not maintained at a constant set point but rather vary with season, habitat, or reproductive state (Casabiell et al., 2001; Rousseau et al., 2003). The cyclical patterns of energy C 2008 V WILEY-LISS, INC. C Grant sponsor: NSF; Grant numbers: BCS 0116465, BNS 7703322, BCS 0115993. *Correspondence to: Patricia L. Whitten, Department of Anthropology, Emory University, 1557 Dickey Drive, Atlanta, GA 30322, USA. E-mail: email@example.com Received 18 February 2008; accepted 19 May 2008 DOI 10.1002/ajpa.20885 Published online 25 July 2008 in Wiley InterScience (www.interscience.wiley.com). 442 P.L. WHITTEN AND T.R. TURNER TABLE 1. Vervet monkey life history traitsa Trait Infant mortality (proportion/1st year) Adult female mortality (3–10 year of age) Fecundity Age at ﬁrst birth (years) Reproductive seasonality (East Africa) Mating Births Weaning age (months) Interbirth interval (months) Lifetime reproductive success (number of offspring born) a Captive Wild 0.4 0.0—0.1 0.4–1.1 4.1 (2.5–5) 0.5 0.1–0.2 0.8 4.4–5.7 Aseasonal Aseasonal 6 10.7 (5.9–24) 0–9 April–Aug Oct–Dec 12–21 14–21 3.3 Captive data from Fairbanks and McGuire, 1984, 1986; Wild data from Cheney et al., 1988; Whitten, 1982. TABLE 2. Sites of vervet populations sampled in Kenya Site Location Rainfall (mm) Elevation (m) Vegetation A B C D Samburu Mosiro Naivasha Kimana 350–400 400–600 >600 200–400 Lowland (650) Highland (1,500) Highland (>2,000) Lowland (1,000) Dry grassland and thornscrub Heavily grazed Acacia woodland and savannah Lakeside grassland, heavily cultivated Dry grassland and thornscrub demand and fat storage in reproductively active females occur without apparent impairment of leptin regulation and provide an alternative window onto leptin physiology not previously explored outside of the laboratory and the clinic. Although some studies have examined serum leptin in nonWestern human populations (Bribiescas, 2001; 2005; Lindgärde et al., 2004; Kuzawa et al., 2007), none of these studies has examined the inﬂuence of the reproductive state. Research on wild primates provides a broader context for investigating the ancestral role of leptin. Levels of serum leptin measured in wild baboons were below 1 ng/ ml, 3–10-fold below the levels seen in captive baboons and in Western men (Banks et al., 2001, 2003). Leptin was not correlated with measures of adiposity in wild baboons, although correlations with BMI and abdominal fat have been reported in captive macaques (Banks et al., 2003; Muehlenbein et al., 2005). These ﬁndings support the hypothesis that leptin evolved to be effective at much lower levels, but other data suggest that variation in energy availability can alter leptin levels in a natural habitat. Access to a garbage dump was associated with markedly elevated leptin levels in baboons but only in a subset of males that became grossly overweight (Banks et al., 2003). Whether these leptin levels are speciﬁc to a human-enriched environment or ever occur in response to variation in natural habitats is unclear. Similarly, leptin varies with reproductive state in captive primates, where leptin may rise as much as 25-fold in gestation (Castracane et al., 2005), but again it is uncertain whether comparable levels are achieved in wild populations. The study tested the hypotheses that serum leptin varies with sex, adiposity, reproductive state, and ecology in wild primates using a unique dataset comprised of blood samples and morphometrics from four ecologically distinct vervet monkey populations. MATERIALS AND METHODS Study site and subjects The subjects of this study were natural populations of vervet monkeys (Chlorocebus aethiops) in Kenya, East American Journal of Physical Anthropology Africa. In classic life history terms, vervets exemplify a fast life history (Rowell and Richards, 1979; Fairbanks and McGuire, 1984; see Table 1). Birth rates are high when compared with other cercopithecines, and pregnancy and lactation often overlap (Fairbanks and McGuire, 1984; Horrocks, 1986). Data and blood samples were obtained in 1978–1979 from four distinct populations of vervet monkeys in widely separated sites in Kenya, East Africa. The sites differed in altitude, temperature, and rainfall (Turner et al., 1997; see Table 2). Site C was surrounded by cultivation, providing access to human food. Wild vervets from 30 troops were trapped as part of a genetic survey and blood, hair, dental casts, and morphological measurements were obtained. Hormone immunoassays were carried out on the serum samples from the adult males and females for whom sufﬁcient sample volume remained for assay of leptin and sex steroids. Subsequently, two of the female samples were excluded from calculation of means and statistical analyses. One anomalous sample with estradiol and progesterone levels typical of the luteal phase and exceptionally high leptin (85 ng/ml) belonged to a female with a 2–3 day old infant. This proﬁle may have represented either residual hormones of gestation (Hess et al., 1979; Hardie et al., 1997; Henson and Castracane, 2002) or a postpartum cycle (Fairbanks and McGuire, 1984) and was excluded as an extreme outlier. A second sample was excluded because the female was markedly dehydrated at capture and lacked a measure of body mass. Exclusion of these two samples left hormonal and morphometric data from 60 adult females and 56 adult males available for statistical analyses. Female reproductive state Pregnancy and stage of gestation were identiﬁed by: (1) palpation during the physical exam; (2) the presence of a cystl aminopeptidase variant in a subsample of females (Turner et al., 1987); (3) serum estradiol and progesterone (Hess et al., 1979; Eley et al., 1989; Rapkin et al., 1995). Females also were classiﬁed as lactating or nonlactating by the presence or absence of expressible milk. 443 LEPTIN VARIATION IN WILD VERVETS The duration of lactation in individual females was estimated by the ages of their presumed infants. Infant ages were determined by dental eruption patterns (McNamara et al., 1977; Turner et al., 1987). Twenty-six percent of lactating females were trapped with a clinging infant who was highly likely to be an offspring. Although vervet females sometimes carry the infants of other mothers, mothers retrieve their infants in competitive feeding situations like that presented by the baited traps. Because most infants within a vervet group fell within the same age range, the stage of lactation in the remaining lactating females was estimated from the modal age of infants within each group, using age groups of \1 week, 1–6 months, and 6–12 months. Measures of adiposity Morphometric measurements were performed by a single member of the ﬁeld team to ensure consistency. Body mass was measured on a Sohne baby scale to the nearest 0.5 gm. Trunk length was measured from the external occipital protuberance to the base of the tail along the curve of the body (Turner et al., 1997). Adiposity was estimated using a primate version of the body mass index, calculated as (kg body mass)/(trunk length m)2. This measure differs from the human BMI in excluding the lower limbs from the measurement of body length. Nevertheless, estimates of BMI in rhesus macaques have been shown to be highly correlated (r 0.9) with body fat estimates obtained by isotope dilution (Kemnitz and Francken, 1986) and dual-energy X-ray absorptiometry (Eisner et al., 2003). Hormone assays Serum samples from the four populations have been stored together at 2208C to 2808C since they were collected in 1978. A number of studies have demonstrated the stability of gonadal steroids and protein hormones like leptin in serum or plasma samples stored at 2208C for periods as long as 15–29 years (Kley et al., 1985; Bolelli et al., 1995; Lissner et al., 1999; Laughlin et al., 2000). Leptin levels in baboon serum stored for 5 years at 2708C did not differ from the levels in freshly obtained serum (Banks et al., 2001). Leptin assays of human serum samples frozen for 29 years yielded the same correlations with relative weight as in a contemporary sample (Lissner et al., 1999). Gonadal steroids. Ether extracts of vervet serum were analyzed for testosterone, estradiol, and progesterone using modiﬁcations of commercial radioimmunoassays as described earlier (Whitten and Russell, 1996; Whitten and Turner, 2004). Intra- and inter-assay coefﬁcients of variation were 5 and 13% for testosterone, 8% and 9– 12% for estradiol, and 9% and 8–12% for progesterone. Leptin. Vervet serum was assayed for leptin using a primate-speciﬁc radioimmunoassay (Linco Research, St. Charles, MO) that uses I125 human leptin and an antibody to human leptin with 100% cross-reactivity with nonhuman primate leptin. The assay has a sensitivity of 0.5 ng/ml with 0.1 ml of sample. Samples were diluted 1:2 with diluent and assayed according to the manufacturer’s instructions. Intra- and inter-assay coefﬁcients of variation were 4–7% and 4–5%. TABLE 3. Statistics for regression analyses of somatic, hormonal, and behavioral inﬂuences on leptin Sex and condition Males (N 5 56) Body mass BMI Order Testosterone-basal Testosterone-suppressed Testosterone-stimulated Acyclic females (N 5 33) Body mass BMI 1 lactation Order Estradiol Duration of lactation Pregnant females (N 5 25) Body mass BMI 1 lactation Order Estradiol Duration of lactation a b Slope 6 SE R2 P 20.303 20.008 20.187 20.067 0.001 0.001 6 6 6 6 6 6 0.472 0.025 0.097 0.239 0.001 0.000 0.008 0.002 0.064 0.002 0.018 0.230 0.523 0.740 0.059 0.897 0.534 0.032a 0.263 0.025 0.008 20.016 20.101 6 6 6 6 6 0.581 0.028 0.027 0.008 0.022 0.007 0.114 0.003 0.003 0.531 0.654 0.349 0.758 0.758 0.000b 21.743 0.609 0.201 0.037 0.259 6 6 6 6 6 3.057 0.276 0.169 0.014 0.414 0.014 0.242 0.058 0.236 0.061 0.574 0.039a 0.245 0.014a 0.555 P \ 0.05. P \ 0.001. Hormonal responsiveness to challenge In previous work (Whitten and Turner, 2004), we have shown that the male testosterone responses to anesthesia varied with the sampling order, providing an index of basal, suppressed, and stimulated testosterone responses to challenge comparable to the responsiveness measure introduced by Sapolsky (1991). In contrast to the pattern of testosterone response, leptin did not show a transient increase following the anesthesia and did not vary significantly with sampling order (see Table 3). However, its correlation with testosterone did vary with sample condition. Leptin levels were not correlated with serum testosterone in basal or suppressed conditions, but were positively correlated with the testosterone under the conditions in which the testosterone levels were elevated (see Table 3). Sampling order had no effect on leptin in acyclic or pregnant females (see Table 3). There were no effects of sampling order on estradiol. Inclusion of sample order in analyses of other factors did not affect the results. Statistical analyses Statistical analyses of hormone data were performed using SYSTAT, Version 11. All variables were tested for normality and homogeneity of variance. Skewed variables were subjected to square root or log transformation before analysis. In these exploratory analyses, a number of univariate analyses were performed. The relative contributions of factors identiﬁed as having signiﬁcant inﬂuences on leptin were investigated further using multivariate analyses. However, sample sizes limited some comparisons. RESULTS Sex differences in serum leptin and body size Males were larger and heavier than females, and females varied substantially in body mass across populations, but both the males and females were below the American Journal of Physical Anthropology 444 P.L. WHITTEN AND T.R. TURNER TABLE 4. Measures of size and leptin concentrations in adult vervet monkeysa Trunk length (cm) Category Sex (M 5 56; F 5 60) Reproductive status Acyclic (N 5 33) Cycling (N 5 2) Pregnant (N 5 25) Population All females A (M: N 5 26; F:N 5 21) B (M: N 5 4; F: N 5 6) C (M: N 5 10; F: N 5 22) D (M: N 5 16; F: N 5 11) Acyclic females A (F: N 5 10) B (F: N 5 6) C (F: N 5 17) D (F: N 5 0) Pregnant females A (F: N 5 11) B (F: N 5 0) C (F: N 5 4) D (F: N 5 10) Habitat: F: acyclic Dry (M: N 5 42; F: N 5 10) Wet (M: N 5 14; F: N 5 23) a b c Male Female Male Female 41.3 6 0.4b 37.1 6 0.4 4.3 6 0.1b 2.9 6 0.1 37.2 6 0.6 39.0 6 3.0 36.8 6 0.5 2.9 6 0.1 3.5 6 0.1 2.9 6 0.1 39.8 42.0 42.4 42.8 6 6 6 6 0.4c 1.1 1.2 0.8 34.3 37.0 38.9 38.8 6 6 6 6 0.4b 0.9 0.6 0.5 BMI (kg/m2) Body mass (kg) 4.1 4.2 4.7 4.3 6 6 6 6 0.1 0.4 0.2 0.2 2.5 2.6 3.3 3.1 6 6 6 6 0.1 0.1 0.1b 0.1b Male Leptin (ng/ml) Female Male 25.0 6 0.4b 21.2 6 0.4 1.1 6 0.2 20.9 6 0.5 23.7 6 4.5 21.4 6 0.5 25.8 23.9 26.2 23.4 6 6 6 6 0.6 1.1 0.9 0.6 21.4 19.2 21.8 20.8 6 6 6 6 0.5 0.5 0.7 1/0 Female 6.2 6 1.3b 1.8 6 0.2 0.9 6 0.1 12.4 6 2.7b 0.7 2.6 2.5 0.5 6 6 6 6 0.1 10.3 6 2.8 1.3b 2.5 6 0.6 0.4b 1.6 6 0.2c 0.1 9.5 6 4.0 1.2 6 0/3 2.5 6 0.6 1.8 6 0.3 18.5 6 3.9 0.7 6 0.4 10.4 6 4.3 0.6 6 0.1 2.5 6 0.4c 1.2 6 0.3 2.0 6 2.6 Values are mean 6 SEM. Subgroup within category differs signiﬁcantly from other subgroups in Tukey HSD pairwise comparisons, P \ 0.001. Subgroup within category differs signiﬁcantly from other subgroups in Tukey HSD pairwise comparisons, P \ 0.05. averages reported for captive vervet monkeys (see Table 4). Body mass measurements for acyclic females (2.9 6 0.1 kg) and males (4.3 6 0.1 kg) were signiﬁcantly (P \ 0.001) below the means reported for captive vervet monkeys (females: 5.3 6 0.07 kg; males: 7.2 6 0.1 kg; Kavanagh et al., 2007) and for free-ranging adults in Barbados (females: 3.3 kg; males: 5.3 kg; Horrocks, 1986). BMI, on the other hand, was higher in adult males (range: 19–34) than in acyclic females (range: 16–28), suggesting that it might be measuring lean mass, rather than fat, in vervet monkey males. BMI levels were substantially lower than the levels reported for captive, nonobese macaque females (28–50: Eisner et al., 2003) or males (30–40: Kemnitz and Francken, 1986; 46–50: Muehlenbein et al., 2005) or wild baboons (40–60: Banks et al., 2003). As in humans (Casabiell et al., 2001), adult vervets displayed a clear sex difference in serum leptin (see Fig. 1 and Table 4). Serum leptin concentrations were highly variable in females, and means were signiﬁcantly higher than in males although the range of values was similar in males (0–6 ng/ml) and acyclic females (0–5 ng/ml). Leptin was not correlated with BMI in either males or acyclic females (see Table 3). Nor were any signiﬁcant correlations observed between leptin and body mass. However, in pregnant females, leptin was signiﬁcantly correlated with BMI when lactation status was included in the regression (see Table 3). Leptin and gonadal steroids The relationship of circulating leptin to gonadal steroids is of interest because leptin both regulates and is regulated by the hypothalamic-pituitary-gonadal axis. In this data set, leptin concentration varied with reproductive and hormonal state. Leptin was positively correlated with serum estradiol in samples from pregnant females American Journal of Physical Anthropology Fig. 1. Sex differences in serum leptin levels in wild vervet monkeys. The plots show median (solid line), mean (dashed line), 25 and 75 percentiles (box), 10 and 90% (whiskers), and outliers (points) in male and female vervets. but was not correlated with estradiol in acyclic females (see Fig. 2). Inclusion of lactation status did not alter these relationships. A similar dissociation of leptin from serum estrogen accords with the data from other lowestrogen subjects such as postmenopausal women (Hong et al., 2007) and athletes with low body fat (Thong et al., 2000). A correlation of leptin with estradiol in pregnant females is not unexpected because both are produced by the placenta in rising quantities as gestation proceeds. Acyclic females are in, by deﬁnition in this study, a low estrogen state in which reproduction is not occurring. It also is a low leptin status that may represent leptin suppression of reproduction in an energy-deﬁcient state. 445 LEPTIN VARIATION IN WILD VERVETS TABLE 5. Serum leptin concentrations by reproductive and lactation status Reproductive Status mean leptin (ng/ml) 6 SEM (N) Acyclic Gestation month 1 2 3 Lactation status Not lactating Lactating Modal lactation duration months (N) 0.2 6.0 12.0 Fig. 2. Correlation of serum leptin with estradiol concentrations in acyclic and pregnant female vervets. Leptin and reproductive state Estradiol and progesterone levels identiﬁed 33 females as acyclic and 25 as pregnant but only two females as cycling (follicular phase). The scarcity of cycling females may be due to the absence of sampling in April and May, primary mating months for East African vervets (Struhsaker, 1967; Whitten, 1982). Leptin varied signiﬁcantly with the female reproductive state with the higher levels in pregnant than in acyclic females (see Table 4). Like other cercopithecines (Henson and Castracane, 2002), vervets exhibited marked increases in leptin early in gestation (F2,22 5 7.410, R2 5 0.402, P 5 0.003), rising to levels as much as 40-fold higher in the second month. Lactation The incidence of lactation varied with reproductive state (v21 5 4.607, P 5 0.032). Expressible milk was present in 62% of acyclic females in contrast to 38% of pregnant females. Leptin declined over the course of lactation. Serum leptin concentration was negatively correlated (b 5 20.101 6 0.081, R2 5 0.518, P 5 0.029) with the duration of lactation, as estimated by the age of infants trapped with their mothers. When modal infant age per group was used as the estimate of the duration of lactation, leptin was negatively correlated among the acyclic mothers but not among the pregnant mothers (see Tables 3 and 5). Unlike many primate species in which lactation ceases before conception, 43% of lactating female vervets were pregnant. ANOVA of reproductive state and the presence or absence of lactation showed an effect only of reproductive state (F2,55 5 7.852, R2 5 0.271, P 5 0.001). Site differences Morphometric measurements showed consistent differences across sites (see Table 4). Turner et al. (1997) have previously reported site differences in size dimensions for adult and juvenile males and females using a nested analysis of variance. Analyses described here were limited to examining size differences in the subset of adults for which leptin measurements were available. These analyses conﬁrmed that vervets were larger and heavier Pregnant 1.2 6 0.5 (8) 17.7 6 3.7 (15)a 16.8 6 0.6 (2) 1.7 6 0.3 (11) 1.8 6 0.2 (11) 13.8 6 3.3 (17) 9.4 6 4.9 (8) 4.6 (1) 2.5 6 0.4 (9) 1.0 6 0.1 (7)a 0.8 (1) 10.6 6 5.4 (7) a Subgroup within category differs signiﬁcantly from other subgroups in Tukey HSD pairwise comparisons, P \ 0.05. in populations B and C. Trunk length varied signiﬁcantly across the populations in both the adult males (P 5 0.011) and females (P 5 0.000), with signiﬁcantly shorter trunks in population A (male: P 5 0.013; female: P \ 0.05). Females also differed in body mass across the populations (P 5 0.000) with heavier animals in the population C (P 0.001). No signiﬁcant population differences in BMI were observed in males (P 5 0.052) or females (P 5 0.254). Male serum leptin varied signiﬁcantly across the populations with the highest levels in the higher rainfall sites B (P 0.006) and C (P 5 0.000; see Table 4). No site differences in leptin were observed in acyclic females. When sites and sexes were combined to compare wet (B and C) to dry (A and D) habitats, ANOVA showed a signiﬁcant effect of habitat (F1,85 5 30.781, R2 5 0.409, P 5 0.000) and a signiﬁcant interactive effect of sex by habitat (F1,85 5 5.915, P 5 0.017) with more marked effects in males than in acyclic females (see Table 4). In acyclic females, there also was a signiﬁcant interactive effect of lactation status with site (F2,27 5 5.322, R2 5 0.391, P 5 0.011). Sites A and B displayed the predicted lower leptin levels in lactating females, but lactating females in Site C had higher leptin levels than nonlactating females (see Fig. 3), perhaps because better nutrition offered by Site C permitted a more positive energy balance in lactating mothers. Seasonal differences Site differences in leptin might also be attributable to the seasonal changes in energy availability. The four populations were sampled at different times of the year, although there was some overlap of seasons of sampling. Leptin varied signiﬁcantly with the month of sampling in both adult males (F8,42 5 6.297, R2 5 0.588, P 5 0.000) and acyclic females (F6,29 5 2.835, R2 5 0.370, P 5 0.027), varying 4–5-fold over the year. In East Africa, the rainy seasons are generally in November–December and March–May, and June–October is the long dry season. Vervet leptin levels were lowest in the long dry season months of June, August, and September and highest in the rainy season months of November-March, with a peak in December (see Fig. 4). American Journal of Physical Anthropology 446 P.L. WHITTEN AND T.R. TURNER Fig. 3. Site differences in Mean 6 serum leptin in lactating and nonlactating females. DISCUSSION These data support the contention that low leptin levels are more typical for primates than the high levels observed in captivity and are likely to represent ancestral levels for humans. Serum leptin levels in vervet males were similar to levels reported for wild baboons (Banks et al., 2001, 2003) and lean men in nonindustrialized or marginally nourished human populations (Bribiescas, 2001; Kuzawa et al., 2007). Leptin levels were higher in acyclic vervet monkey females than in males but still were substantially lower than the levels reported for captive Old World monkeys (Castracane et al., 2005) or women in industrialized or Westernized populations (Bribiescas, 2001; Kuzawa et al., 2007). Weak correlations of leptin with adiposity in lean human males, in some marginally nourished populations, have raised questions about leptin’s presumed role in the regulation of adiposity (Bribiescas, 2001; Kuzawa et al., 2007). For example, extremely low leptin levels have been reported for the Tsimané of lowland Bolivia (Sharrock et al., 2008). Leptin becomes increasingly coupled to fatness in girls in this population. Adiposity increases with age, reaching Frisch’s (Frisch and McArthur, 1974) ‘‘critical fatness’’ threshold of 22% by 15 years of age when leptin increases markedly. Boys, on the other hand, retain a fat level of about 10% and an extremely low leptin concentration throughout adolescence, suggesting that leptin is a more important signal of energy availability in females than in males in this population (Sharrock et al., 2008). In this study, leptin was not correlated with BMI in male or acyclic female vervets. Weak correlations of leptin with BMI also have been reported for wild and captive baboon males (Banks et al., 2003; Muehlenbein et al., 2003). On the other hand, signiﬁcant correlations of BMI, abdominal skinfold, and body mass with serum leptin have been reported for captive male rhesus and pigtailed macaques (Muehlenbein et al., 2005). The differences between baboons and macaques in the relation of leptin to body composition may reﬂect species differences in the fat deposition (Muehlenbein et al., 2003). Percent body fat in wild baboons, estimated from stable isotope American Journal of Physical Anthropology Fig. 4. Monthly variation in Mean 6 SEM serum leptin in males and acyclic females. dilution, is extremely low (males: 0.8–9.3%; females: 1.9%), reﬂecting both the leanness of wild baboons and the very small subcutaneous fat deposits in cercopithecine primates (Altmann et al., 1993). Consequently, individual differences may be below the level of resolution of the available measures of adiposity (Altmann et al., 1993). In marginally nourished populations, factors other than adiposity play an important role in the regulation of leptin concentrations (Chan and Mantzoros, 2005). For example, leptin is a more sensitive and rapid signal of negative energy balance than of declining adiposity (Jéquier, 2002). The leptin response to changes in energy balance is disproportionate to its effect on adiposity (Ahima and Osei, 2004) and may result in low correlations of leptin with measures of adiposity. For example, in an intervention study of nine healthy men and women, a 25% increase in energy intake over energy expense resulted in a 31% increase in plasma leptin but an increase in the body mass of only 1% (Hagobian et al., 2008). Responsiveness to energy balance allows more immediate energy-saving responses such as increased feeding, reduced metabolic rate, and amenorrhea (Ahima and Osei, 2004). In this sample of vervet monkeys, leptin reﬂected habitat and seasonal variation in spite of its low levels. Crop raiding may have contributed to higher leptin levels in Site C. However, the 4- to 5-fold higher leptin levels in both sites and seasons with higher rainfall suggest that leptin also registers natural ecological variation. As a starvation meter, leptin exerts a permissive inﬂuence on reproduction. Falling leptin concentrations with energy deﬁcit diminish its normal stimulation of GnRH pulsatility and gonadal steroid secretion (Chan and Mantzoros, 2005). Recombinant human leptin reverses the starvation response, restoring GnRH and androgen and estrogen secretion (Blüher and Mantzoros, 2004). These effects on the HPG axis in energy deprivation states may explain the varying relationship between leptin and estradiol observed for pregnant and acyclic females. Leptin variation with reproductive state in females shows that very high leptin levels occur in states other than obesity and are reversible. The highest leptin levels occur in pregnancy, when leptin rises in maternal circulation due to placental production (Hardie et al., 1997; LEPTIN VARIATION IN WILD VERVETS Kratzsch et al., 2000). Placental hormone production reﬂects the evolution of maternal-fetal conﬂict over resource allocation, a process in which maternal resistance to placental hormones selects for their increased release (Haig, 1993). These ongoing conﬂicts have fostered the rapid evolution of growth hormone-related and prolactin-related genes in the anthropoid primates (Haig, 2008). For example, the placenta pours out enormous quantities of placental lactogen, independent of maternal regulation (Haig, 1993). Very high concentrations of human placental lactogen produce leptin resistance by reducing the availability of hypothalamic leptin receptors (Augustine et al., 2008). Leptin resistance induces a starvation-like state that stimulates food intake in spite of increasing fat storage (Kratzsch et al., 2000; Henson and Castracane, 2002). After parturition, the high energetic demands of lactation are met by increased appetite and mobilization of fat reserves (Butte et al., 1997; Dufour and Sauther, 2002; Vernon et al., 2002). The resultant negative energy balance suppresses leptin, stimulating increased food intake and reduced energy expenditure. These responses to reproductive state were evident in wild vervet females, who displayed dramatic elevations in circulating leptin during gestation and declines over the period of lactation. In wild populations, high leptin levels appear to be unique to females, but more work will be needed to determine if males undergo similar changes in leptin regulation in other reproductive contexts. Since leptin was identiﬁed in 1994, our view of its role and action has changed radically. Once thought to be an antiobesity hormone, leptin is now known to mobilize key neuroendocrine responses to starvation. It impacts a broad range of physiological systems: assessing energy balance, suppressing reproduction and metabolic expense, mobilizing energy stores, and increasing lipolysis (Chan and Matzoros, 2005). Increasing evidence shows that leptin indexes not only adiposity but also energy balance, reproductive activity, and ontogeny. Leptin resistance and responses to energy deﬁcient states are providing new models for the treatment of reproductive and neuroendocrine abnormalities (Chan and Matzoros, 2005) and have enriched our understanding of the neuroendocrinology of resource allocation. CONCLUSIONS This study addressed the evolutionary model that leptin physiology evolved at much lower levels of leptin and adiposity than levels found in Western human populations. The model was supported by the low levels of leptin in adult males and females but was contradicted by signiﬁcant variation in leptin with habitat, season, and female reproductive state. This variation shows that predictable elevations in serum leptin were as much a feature of ancestral leptin as low basal levels. An expanded view of leptin action has demonstrated its involvement with a range of neuroendocrine factors and axes. A better understanding of the regulation of this variation may provide insights into the control of obesity and amenorrhea today. ACKNOWLEDGMENTS We gratefully acknowledge Dr. N.C. Dracopoli, Dr. J.G. Else, Institute of Primate Research, and Dr. C.J. Jolly for their assistance in the early stages of this pro- 447 ject. We are grateful to Mr. Fred Brett for sharing the serum samples. We thank the Ofﬁce of the President of the Republic of Kenya for permission to conduct the original ﬁeld research in Kenya. We thank Betsy Russell for her excellent laboratory work. The investigations described herein comply with the laws of the United States and Kenya. LITERATURE CITED Ahima RS, Osei SY. 2004. Leptin signaling. Physiol Behav 81:223–241. Alonso-Alvarez C, Bertrand S, Sorci G. 2007. Energetic reserves, leptin and testosterone: a reﬁnement of the immunocompetence handicap hypothesis. Biol Lett 3:271–274. Altmann J, Schoeller D, Altmann SA, Murthi P, Sapolsky RM. 1993. Body size and fatness of free-living baboons reﬂect food availability and activity levels. Am J Primatol 30:149–161. Augustine RA, Ladyman SR, Grattan DR. 2008. From feeding one to feeding many: hormone-induced changes in bodyweight homeostasis during pregnancy. J Physiol 586:387–397. Banks WA. 2003. Is obesity a disease of the blood-brain barrier? Physiological pathological, and evolutionary considerations. Curr Pharm Des 9:801–809. Banks WA, Altmann J, Sapolsky RM, Phillips-Conroy JE, Morley JE. 2003. Serum leptin levels as a marker for a syndrome X-like condition in wild baboons. J Clin Endocrinol Metab 88:1234–1240. Banks WA, Clever CM, Farrell CL. 2000. Partial saturation and regional variation in the blood-to-brain transport of leptin in normal weight mice. Am J Physiol Endocrinol Metab 278:E1158–E1165. Banks WA, Farr SA, Morely JE. 2006. The effects of high fat diets on the blood-brain barrier transport of leptin: failure or adaptation. Physiol Behav 88:244–248. Banks WA, Phillips-Conroy JE, Jolly CJ, Morley JE. 2001. Serum leptin levels in wild and captive populations of baboons (Papio): implications for the ancestral role of leptin. J Clin Endocrinol Metab 86:4315–4320. Blüher S, Mantzoros CS. 2004. The role of leptin in regulating neuroendocrine function in humans. J Nutr 134:2469S–2474S. Bolelli G, Muti P, Micheli A, Sciajno R, Franceschetti F, Krogh V, Pisani P, Berrino F. 1995. Validity for epidemiological studies of long-term cryoconservation of steroid and protein hormones in serum and plasma. Cancer Epidemiol Biomarkers Prev 4:509–513. Bribiescas RG. 2001. Serum leptin levels and anthropometric correlates in Ache Amerindians of eastern Paraguay. Am J Phys Anthropol 115:297–303. Bribiescas RG. 2005. Serum leptin levels in Ache Amerindian females with normal adiposity are not signiﬁcantly different from anorexia nervosa patients. Am J Hum Biol 17:207–210. Butte NF, Hopkinson JM, Nicolson MA. 1997. Leptin in human reproduction: serum leptin levels in pregnant and lactating women. J Clin Endocrinol Metab 82:585–589. Casabiell X, Piñeiro V, Vegal F, De la Cruz L, Diéguez C, Casanueva FF. 2001. Leptin, reproduction and sex steroids. Pituitary 4:93–99. Castracane VD, Hendrickx AG, Henson MC. 2005. Serum leptin in nonpregnant and pregnant women and in Old and New World nonhuman primates. Exp Biol Med 230:251–254. Chan JL, Mantzoros CS. 2001. Leptin and the hypothalamic-pituitary regulation of the gonadotropin-gonadal axis. Pituitary 4:87–92. Chan JL, Mantzoros CS. 2005. Role of leptin in energy-deprivation states: normal human physiology and clinical implications for hypothalamic amenorrhoea and anorexia nervosa. Lancet 366:74–85. Cheney DL, Seyfarth RM, Andelman SJ, Lee PC. 1988. Reproductive success in vervet monkeys. In: Clutton Brock TH, editor. Reproductive success. Chicago: University of Chicago Press. p 384–402. American Journal of Physical Anthropology 448 P.L. WHITTEN AND T.R. TURNER Dufour DL, Sauther ML. 2002. Comparative and evolutionary dimensions of the energetics of human pregnancy and lactation. Am J Hum Biol 14:584–602. Eley RM, Tarara RP, Worthman CM. 1989. Reproduction in the vervet monkey (Cercopithecus aethiops). III. The menstrual cycle. Am J Primatol 17:1–10. Eisner JR, Dumesic DA, Kemnitz JW, Colman RJ, Abbott DH. 2003. Increased adiposity in female rhesus monkeys exposed to androgen excess during early gestation. Obes Res 11:279– 286. Fairbanks LA, McGuire MT. 1984. Determinants of fecundity and reproductive success in captive vervet monkeys. Am J Primatol 7:27–38. Fairbanks LA, McGuire MT. 1986. Age, reproductive value, and dominance-related behaviour in vervet monkey females: crossgenerational inﬂuences on social relationships and reproduction. Anim Behav 34:1710–1721. Fox C, Esparza J, Nicolson M, Bennett PH, Schulz LO, Valencia ME, Ravussin E. 1999. Plasma leptin concentrations in Pima Indians living in drastically different environments. Diabetes Care 22:413–417. Friedman JM, Halaas JL. 1998. Leptin and the regulation of body weight in mammals. Nature 395:763–770. Frisch RE, McArthur JW. 1974. Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 185:949–951. Gale SM, Castracane VD, Mantzoros CS. 2004. Energy homeostasis, obesity and eating disorders: recent advances in endocrinology. J Nutr 134:295–298. Hagobian TA, Sharoff CG, Braun B. 2008. Effects of short-term exercise and energy surplus on hormones related to the regulation of energy balance. Metabolism 57:393–398. Haig D. 1993. Genetic conﬂicts in human pregnancy. Q Rev Biol 68:495–532. Haig D. 2008. Placental growth hormone-related proteins and prolactin-related proteins. Placenta 29, Supplement A, Trophoblast Research, Vol. 22:S36–S41. Hardie L, Trayhurn P, Abramovich D, Fowler P. 1997. Circulating leptin in women: a longitudinal study in the menstrual cycle and during pregnancy. Clin Endocrinol 47:101–106. Henson MC, Castracane VD. 2002. Leptin: roles and regulation in primate pregnancy. Semin Reprod Med 20:113–121. Hess DL, Hendrickx AG, Stabenfeldt GH. 1979. Reproductive and hormonal proﬁles in the African green monkey (Cercopithecus aethiops). J Med Primatol 8:273–281. Hong SC, Yoo SW, Cho GJ, Kim T, Hur JY, Park YK, Lee KW, Kim SH. 2007. Correlation between estrogens and serum adipocytokines in premenopausal and postmenopausal women. Menopause 14:835–840. Horrocks JA. 1986. Life-history characteristics of a wild population of vervets (Cercopithecus aethiops sabaeus) in Barbados. West Indies. Int J Primatol 7:31–47. Jéquier E. 2002. Leptin signaling, adiposity, and energy balance. Ann NY Acad Sci 967:379–388. Kavanagh K, Fairbanks LA, Bailey JN, Jorgensen MJ, Wilson M, Zhang L, Rudel LL, Wagner JD. 2007. Charcterization and heritability of obesity and associated risk factors in vervet monkeys. Obesity 15:1666–1674. Kemnitz JW, Francken GA. 1986. Characteristics of spontaneous obesity in male rhesus monkeys. Physiol Behav 38:477– 483. Kley HK, Schlaghecke R, Krüskemper HL. 1985. Stabilität von Steroiden im Plasma über einen Zeitraum von 10 Jahren. [Stability of steroids in plasma over a 10-year period]. J Clin Chem Clin Biochem 23:875–878. Kratzsch J, Hckel M, Kiess W. 2000. Leptin and pregnancy outcome. Curr Opin Obstet Gynecol 12:501–505. Kuzawa CW, Quinn EA, Adair LS. 2007. Leptin in a lean population of Filipino adolescents. Am J Phys Anthropol 132:642– 649. American Journal of Physical Anthropology Laughlin GA, Barrett-Connor E, Kritz-Silverstein D, von Muhlen D. 2000. Hysterectomy, oophorectomy, and endogenous sex hormone levels in older women: the Rancho Bernardo Study. J Clin Endocrinol Metab 85:645–651. Lindgärde F, Widén I, Gebb M, Ahre’n B. 2004. Traditional versus agricultural lifestyle among Shuar women of the Ecuadorian Amazon: effects on leptin levels. Metabolism 53:1355– 1358. Lissner L, Karlsson C, Lindroos AK, Sjostrom L, Carlsson B, Carlsson L, Bengtsson C. 1999. Birth weight, adulthood BMI, and subsequent weight gain in relation to leptin levels in Swedish women. Obes Res 7:150–154. McNamara JA, Foster DL, Rosenstein BD. 1977. Eruption of the deciduous dentition in the rhesus monkey. J Dent Res 56:701. Muehlenbein MP, Campbell BC, Richards RJ, Svec F, Falkenstein KP, Murchison MA, Myer L. 2003. Leptin, body composition, adrenal and gonadal hormones among captive male baboons. J Med Primatol 32:30–324. Muehlenbein MP, Campbell BC, Richards RJ, Watts DP, Svec F, Falkenstein KP, Murchison MA, Myer L. 2005. Leptin, adiposity, and testosterone in captive male macaques. Am J Phys Anthropol 127:335–341. Rapkin AJ, Pollack DB, Raleigh MJ, Stone B, McGuire MT. 1995. Menstrual-cycle and social-behavior in vervet monkeys. Psychoneuroendocrinology 20:289–297. Rousseau K, Atcha Z, Loudon ASI. 2003. Leptin and seasonal mammals. J Neuroendocrinol 15:409–414. Rowell TE, Richards SM. 1979. Reproductive strategies of some African monkeys. J Mammal 60:58–69. Sapolsky R. 1991. Testicular function, social rank and personality among wild baboons. Psychoneuroendocrinology 16:281– 301. Schwartz MW, Niswender KD. 2004. Adiposity signaling and biological defense against weight gain: absence of protection or central hormone resistance? J Clin Endocrinol Metab 89:5889–5897. Sharrock KCB, Kuzawa CW, Leonard WR, Tanner S, Reyes-Garcia VE, Vadez V, Huanca T, McDade T. 2008. Developmental changes in the relationship between leptin and adiposity among Tsimané children and adolescents. Am J Hum Biol 20:392–398. Struhsaker TT. 1967. Ecology of vervet monkeys (Cercopithecus aethiops) in the Masai-Amboseli Game Reserve. Kenya. Ecology 48:891–904. Thong FS, McLean C, Graham TE. 2000. Plasma leptin in female athletes: relationship with body fat, reproductive, nutritional, and endocrine factors. J Appl Physiol 88:2037–2044. Turner TR, Anapol F, Jolly CJ. 1997. Growth, development, and sexual dimorphism in vervet monkeys (Cercopithecus aethiops) at four sites in Kenya. Am J Phys Anthropol 103:37–68. Turner TR, Whitten PL, Jolly CJ, Else JG. 1987. Pregnancy outcome in free-ranging vervet monkeys. Am J Primatol 12:197– 203. Vernon RG, Denis RGP, Sorensen A, Williams G. 2002. Leptin and the adaptations of lactation in rodents and ruminants. Horm Metab Res 34:678–685. Whitten PL. 1982. Female reproductive strategies among vervet monkeys. Ph.D. dissertation, Harvard University, Cambridge, MA. Whitten PL, Russell E. 1996. Information content of sexual swellings and fecal steroids in sooty mangabeys (Cercocebus torquatus atys). Am J Primatol 40:67–82. Whitten PL, Turner TR. 2004. Male residence and the patterning of serum testosterone in vervet monkeys (Cercopithecus aethiops). Behav Ecol Sociobiol 56:565–578. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM. 1994. Positional cloning of the mouse obese gene and its human homologue. Nature 372:425–432.