The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Department of Genetics, Southwest Foundation for Biomedical Research, P.O. Box 760549, San Antonio, TX, 78245-0549. E-mail: email@example.com
Objectives: The hormones adiponectin and resistin have been associated with insulin resistance. This paper analyzed the potential relationship between adiponectin and resistin and insulin resistance-related phenotypes in baboons.
Research Methods and Procedures: One hundred eight adult baboons (84 female and 24 male) were studied. Weight was measured, and a blood sample was collected under fasting conditions for plasma and monocyte isolation. Fasting glucose, insulin, C-peptide, and adiponectin levels in plasma were measured by standard methods. Insulin resistance was calculated by the homeostasis model assessment index. Resistin mRNA abundance in monocytes was determined by real-time quantitative reverse transcription-polymerase chain reaction. Data were clustered by weight tertiles for statistical analysis.
Results: As observed in humans, the insulin resistance-related phenotypes were related to weight, plasma levels of adiponectin, and C-peptide. No significant relationship between resistin circulating levels or expression in monocytes and insulin resistance-related phenotypes was found in baboons.
Discussion: These findings suggest that resistin is not associated with insulin resistance. However, previous observations of relationships among weight, adiponectin, and insulin resistance are confirmed.
The adipose tissue-derived peptides adiponectin and resistin have been proposed as possible links between adiposity and insulin resistance (1, 2). It is believed that adiponectin improves insulin resistance, whereas resistin antagonizes insulin effects (1).
Adiponectin is an adipose tissue-derived protein that is abundant in the plasma and affects insulin resistance (1, 2, 3). Low levels of plasma adiponectin are associated with obesity and type 2 diabetes in humans (2, 3) and rhesus monkeys (4) and with coronary heart disease (5) and risk factors for the metabolic syndrome in humans (1, 6). Yang et al. (6) and Esposito et al. (7) observed a significant increase of adiponectin levels after weight reduction in humans, with concomitant improvement of insulin resistance. Studies from Bruun et al. (8) and Engeli et al. (9) reported that adiponectin expression in subcutaneous adipose tissue is inversely related to cytokine levels. Tumor necrosis factor-α seems to reduce the adiponectin mRNA levels in adipose tissue (8). This mechanism may contribute to the insulin resistance observed under inflammatory conditions. Yamauchi et al. reported the presence of adiponectin receptors in muscle and liver. This finding supports the hypothesis that this protein has an endocrine activity (10). Collectively, these studies indicate that circulating levels of adiponectin influence insulin sensitivity.
Resistin belongs to the resistin-like molecule family of proteins. These proteins are detected in the inflammatory response to allergic pulmonary inflammation (11, 12). The human form of murine resistin has been mapped to chromosome 19 and is produced by white (1, 13, 14) and brown (15) adipose tissue, bone marrow, lung, and monocytes in humans (16, 17).
Initial observations from Steppan et al. (13) reported that circulating levels of resistin protein and mRNA in adipose tissue were increased in obese mice. Administration of intraperitoneal recombinant resistin protein to mice diminished glucose tolerance and raised plasma insulin levels. In addition, blocking with specific murine antibodies to inactivate resistin improved insulin resistance. These observations imply that resistin reduces insulin sensitivity and is a possible link between adipose tissue and insulin resistance. In 2002, McTernan et al. (18) reported detectable levels of resistin mRNA in human adipose tissue. Higher resistin expression was found in both abdominal subcutaneous and omental depots, as compared with those in the thigh (19). These higher values validate the proposed association between omental body fat abundance and insulin resistance. Also, Patel et al. (16) detected significant amounts of resistin in human macrophages and other tissues. They concluded that thiazolidinediones, peroxisome proliferator-activated receptor-γ activators, reduce resistin expression at the RNA and protein level in these cells. A recent study by McTernan et al. found higher levels of resistin among type 2 diabetic subjects as compared with healthy controls (20). Degawa et al. (21) reported significantly higher values of this protein in obese subjects. Thus, these studies support conclusions from the original paper by Steppan that resistin antagonizes insulin action.
In contrast, Way et al. (22) reported that treatment with thiazolidinediones enhanced resistin expression and that obese rodents had lower resistin mRNA levels in adipose tissue as compared with lean pairs. Nagaev and Smith (17) reported that resistin expression in subcutaneous adipose tissue was not related with insulin resistance, as measured by euglycemic clamp in insulin-sensitive, insulin-resistant, or type 2 diabetic patients. Although little or no resistin expression in human adipose tissue was found, the expression in monocytes was measurable. A possible connection between resistin expression in these cells and insulin resistance was not explored. In 2002, Janke et al. (23) analyzed the association between resistin mRNA abundance in cultured human subcutaneous preadipocytes and adipocytes and insulin resistance using the homeostasis model assessment (HOMA-IR)1 utilizing fasting glucose and insulin levels. No association was found between resistin expression and weight or insulin sensitivity. The study of Lee et al. found no association between resistin and insulin resistance in cross-sectional and intervention studies in humans (24). Thus, the conflicting reports on the biological role of resistin in insulin sensitivity are puzzling.
The baboon is an excellent animal model for the study of insulin resistance-related phenotypes because there is a close correspondence to humans in the interaction of genes involved in obesity and type 2 diabetes (25, 26, 27, 28). Approximately 10% of a captive colony of adult baboons at the Southwest Foundation for Biomedical Research (SFBR) (San Antonio, TX) developed spontaneous obesity, and 4% become hyperglycemic or diabetic, even though they were housed under similar conditions and ate the same diet (A.G. Comuzzie, unpublished data). Recently, variations in plasma adiponectin and resistin (A.G. Comuzzie and M. Lazar, unpublished data) across different baboons were detected in plasma using a human antibody. These observations confirmed the presence of circulating levels of these hormones in baboons and a high structural similarity with the human proteins. The purpose of this study was to analyze the association of plasma adiponectin levels, resistin protein in plasma, and mRNA expression in monocytes and insulin resistance-related phenotypes in adult baboons.
Research Methods and Procedures
A cross-sectional study was carried out at the primate colony at the SFBR. When the baboons were sedated for their annual medical check, weight was recorded and blood samples were drawn for separation into monocytes and plasma. Glucose, insulin, C-peptide, and adiponectin were assayed in plasma. A baboon resistin fragment was cloned from omental adipose tissue and a quantitative, real-time reverse transcription (RT) followed by polymerase chain reaction (PCR) method was developed. The mRNA abundance of resistin in monocytes was assayed.
The 108 baboons consisted of 24 male and 84 nonpregnant or lactating females from the pedigreed colony at the SFBR. This colony was founded in San Antonio, TX with 400 feral animals (360 females and 40 males) from a mixture of two species, Papio hamadryas anubis and Papio hamadryas cynocephalus. All animals are gang-housed and fed ad libitum on a standard chow diet (Harlan Tecklad 15% Monkey Diet, 8715, Indianapolis, IN).
All samples were collected after an overnight fast (12 hours), with the animals under sedation with ketamine. Weight was measured on a calibrated electronic scale (GSE, Chicago, IL). A 20-mL sample of blood was drawn from the antecubital vein and divided as follows: an 8-mL sample was collected in cell preparation tubes (BD vacutainer CPT mononuclear cell preparation, BD Biosciences, San Jose, CA) containing a density gradient polymer gel and sodium citrate (Becton Dickinson, Franklin Lakes, NJ) for monocyte isolation, 4 mL in sodium fluoride tubes for glucose analysis, and 7 mL in EDTA tubes for analysis of adiponectin, insulin, and C-peptide. All samples were centrifuged for 10 minutes at 2000g. The resultant plasma from the sodium fluoride and EDTA tubes was collected and frozen at −80 °C for subsequent analysis.
Cloning of a Baboon Resistin cDNA
A 1-g omental adipose tissue biopsy was collected from a healthy male baboon while under sedation with ketamine. This sample was frozen for further RNA isolation with Trizol Reagent (Molecular Research Center, Inc., Gaithersburg, MD). A 445-bp fragment of baboon resistin was cloned from this RNA sample by a two-step RT-PCR method using the THERMOSCRIPT RT-PCR System (Invitrogen, Carlsbad, CA). Primers for cloning were designed based on the human sequence (GenBank accession no. AF352730). The amplified product was cloned using the CloneAmp pAMP1 kit (Invitrogen), according to their protocol. The cloned baboon resistin cDNA fragment was sequenced on an ABI 377 automated DNA sequencer using the Big Dye Terminator kit (Applied Biosystems, Foster City, CA). This fragment was 95% identical to the human mRNA sequence.
Assay for Resistin mRNA Expression
Monocytes were collected by centrifugation in cell preparation tubes (BD Biosciences), washed twice with 10 mL of sterile phosphate-buffered saline, and frozen at −80 °C for subsequent RNA extraction. RNA was isolated using a 4 PCR RNAquous system from Ambion (Austin, TX). RNA integrity was verified on 1% agarose gel using ethidium bromide stain. The RNA yield and purity were analyzed by UV spectrophotometry. Resistin expression in monocytes was measured by real-time, quantitative RT-PCR (Taq Man, Applied Biosystems). The primers and probe sequences were designed with the Primer Express Software Version 1 (Applied Biosystems) using the baboon resistin clone. The sequences of forward and reverse primers were 5′ TCCTCCTGCCTGTCCTGG 3′ and 5′ CGCCCTCCTGAATCTTCTCAT 3′, respectively. The sequence for the resistin probe was 5′TCTAGCCAGACCCTGTGCTCCATGG 3′.
Ribosomal 18S RNA (rRNA) was used as an internal control and measured by the Universal 18S system from Ambion. The master mix primer-to-competimer ratio was 4:6. The probe for the 18S mRNA was r RNA Ambiprobe from Applied Biosystems.
A sample of 50 ng of total RNA was used per assay. RT-PCR conditions were 48 °C for 50 minutes for RT and 40 cycles of 60 °C for 1 minute, followed by 90 °C for 15 seconds. Data were obtained as Ct values (the number of cycles at which logarithmic plots of PCR product accumulation cross a specific threshold line), according to the manufacturer's specifications. Resistin expression was corrected for sample-to-sample measurement error, calculated as resistin Ct divided by the 18S rRNA Ct, and corrected by the 18S rRNA Ct mean in any given run. Inter- and intra-assay coefficients of variation for resistin expression were 6% and 4% and 5% and 7% for 18S rRNA, respectively.
Assays for Glucose, Adiponectin, Insulin, and C-Peptide
Glucose was analyzed by the glucose oxidase method on an Analox spectrophotometer. Adiponectin levels were measured by radioimmunoassay (Linco Research, Inc., St. Charles, MO). Insulin and C-peptide were analyzed by quimioluminescence in a Luminex100 using the Endocrine Multiplex Immunoassay (Linco Research, Inc.). Resistin protein was assayed using a commercial enzyme-linked immunosorbent assay kit (Bio-Vendor Laboratory Medicine, Inc., Brno, Czech Republic). Parameters with variations >5% were reanalyzed. All samples were analyzed in duplicate and compared with standard curves. The HOMA-IR index was calculated as indicated by Matthews (29).
Descriptive values and analytical tests were calculated using SPSS (SPSS Inc., Chicago, IL). Comparisons between male and female baboons were performed by Student's t tests for independent samples. Partial correlations controlling for sex were calculated among variables. A step-wise regression analysis was conducted to identify the model with the highest predictive value for the HOMA-IR index. Baboons were classified in tertiles by weight, and the resulting groups were compared using one-way ANOVA and least significant difference (LSD) as post hoc tests.
In this study, the female-to-male ratio was 3 to 1, which represents the sex ratio at the baboon colony at the SFBR. The weight values for baboons ranged from 15 to 32 kg for females and 20 to 40 kg for males. Three females and one male were hyperglycemic, as indicated by fasting glucose levels >125 mg/dL. The HOMA index for insulin resistance ranged from 0.1 to 4.5 in females and 0.8 to 3.2 in males. Values > 3.0 are considered as elevated insulin resistance in humans (30).
Table 1 compares the analyzed traits between female and male baboons. Females had lower body weights and were older than the males. Fasting glucose and insulin levels, as well as the glucose/insulin index, did not differ according to sex. Higher concentrations of C-peptide and the HOMA-IR index values were observed in females as compared with males, probably indicating higher insulin resistance. Adiponectin and resistin circulating protein levels, as well as the resistin mRNA expression in monocytes, were not different between male and female baboons.
Table 1. Mean (SD) of characteristics of female and male baboons
Female (n = 84)
Male (n = 24)
Resistin expression (Ct)
Table 2 shows the analyzed phenotypic mean values within each weight tertile in female baboons. Insulin, C-peptide levels, and HOMA index values increased with weight. The reduction in adiponectin plasma levels with weight increments was not significant (p = 0.15); however, the log-transformed values of this hormone were correlated significantly to plasma insulin levels (r = −0.31, p < 0.05), the HOMA-IR index (r = −0.27, p < 0.05), and the glucose-to-insulin ratio (r = 0.31, p < 0.05) in baboons (Table 4). The glucose-to-insulin ratio varied across tertiles, but only values in the highest tertile were significantly lower than the other groups. Data on male baboons are presented in Table 3. Males showed a similar trend, with increasing insulin and C-peptide circulating levels according to weight increments, but differences among tertiles were not significant, presumably due to the smaller sample size. Levels of adiponectin decreased with higher body weight, with no significant difference as observed in females. The glucose-to-insulin ratio declined significantly as body weight increased. The resistin protein showed a nonsignificant decreasing trend across the body weight tertiles in female and male baboons. Resistin mRNA abundance in monocytes was not associated with weight in any of the studied phenotypes in either sex group. Table 4 shows a matrix of bivariate correlations among the studied traits.
Table 2. Mean (SD) of characteristics of female baboons according to weight tertiles
The step-wise regression analysis model included the C-peptide, weight, age, sex, adiponectin and resistin protein, and resistin monocyte mRNA levels. The C-peptide and adiponectin levels were the only variables selected in the model and accounted for 65% of the variation of the HOMA-IR index (Table 5).
Table 5. Multiple stepwise regression model for HOMA-IR index in adult baboons (n = 108)
R = 0.80; R2 = 0.65.
7.02 × 30 − 3
A human study has reported higher circulating adiponectin levels in females than in males (9). In our study, levels of this protein were not different between sex groups, and circulating levels in baboons were within the range observed in humans (8, 9, 32). Adiponectin was not significantly associated with weight in this study. However, a larger sample of baboons was analyzed in a previous study from our laboratory, and negative associations between weight and adiponectin log-transformed values were found for females (r = −0.2, p = 0.008, n = 174) and males (r = −0.32, p = 0.002, n = 97) (A.G. Comuzzie, unpublished data). This association followed the same trend observed in humans, with lower adiponectin levels in plasma in heavier animals. Plasma adiponectin was related to plasma insulin levels, HOMA-IR, and glucose/insulin indices in female baboons. Similar findings have been observed in a human study by Matsubara et al. (31), who found an inverse association between adiponectin levels and indicators of insulin resistance in Japanese nondiabetic women. Plasma log-transformed adiponectin levels were significantly correlated with insulin levels (r = −0.39, p < 0.001) and the HOMA-R index (r = −0.37, p < 0.001).
Since its discovery in 2001, the biological role of resistin has been under extensive investigation. Most of the published work has focused on resistin production in adipose tissue, and there are few studies on resistin expression in monocytes and its potential contribution to insulin resistance. Recently, Patel et al. (16) reported a down-regulating effect of peroxisome proliferator-activated receptor-γ activators on resistin expression in cultured monocyte-derived macrophages in vitro. These findings support initial observations from Steppan et al. in mice (13) and suggest a possible contribution of monocyte resistin to insulin sensitivity. The biological function of resistin and its relationship with insulin resistance are still obscure. Recent human studies have contradictory findings (20, 21, 24). The circulating levels of resistin in baboons are very similar to those found in humans (20, 21), with no difference between sex groups in the protein and mRNA monocyte levels, as opposed to findings from Yanakoulia et al. (32), who reported higher values of resistin protein in females than in males.
In our study, resistin levels in plasma and mRNA expression in monocytes were not associated with the HOMA-IR, C-peptide, insulin, and the glucose/insulin indices. These observations agree with reports from Degawa (21) and Lee (24). A slight but significant association was found between resistin mRNA expression in monocytes and circulating levels of the protein (r = −0.26, p < 0.05).
Although the HOMA-IR index has not been validated in baboons as an indicator for insulin resistance, findings from our group and observations of baboons in the wild (27, 28) support the idea that the variations in fasting glucose and insulin are related to body weight and diet and may be reflective of insulin resistance.
In the present study, a profound sexual dimorphism in size existed between female and male baboons, with females having a wider range in weight and age. Although female baboons were older than males, both groups were considered as adults, with females at childbearing stage. Previously Comuzzie et al. (26) reported that percentage of body fat measured by DXA and weight are significantly correlated in baboons (r = 0.76, p = 2 × 10−5). The levels of insulin and C-peptide in baboons in the present study were close to those reported for humans (33). The increase in these parameters of insulin resistance with escalating body weights was supported by the parallel increase in C-peptide levels and the HOMA-IR index, as illustrated in Figure 1. The effect of weight on insulin resistance-related phenotypes was more pronounced in the female baboons, with significant differences across the weight tertiles. The male baboons showed a similar trend, but the smaller sample size and narrow weight interval in the male sample presumably precluded significance. A previous study on body composition of the baboon found that weights of 20 kg in females and 38 kg in males are associated with ∼20% body fat (26). In the present study, 35 females and 1 male exceeded this reference weight. It should be noted that standards for overweight and obesity have not been established in baboons.
In summary, similar to findings in humans, the levels of plasma adiponectin were associated significantly with insulin and the HOMA-IR index. Body weight was associated with variations in insulin resistance-related traits, as indicated by levels of fasting insulin and C-peptide levels, and the HOMA index. Resistin protein and mRNA levels were not related to any parameter measured in this study. The observations above support the value of the baboon as a model for the study of obesity-related conditions.
This work was supported by grants from the Southwest Foundation Forum, NIH Grant HL28972, and NIH Grant 251RR013986. M.E. Tejero was funded by a Fulbright-CONACyT grant.
Nonstandard abbreviations: HOMA-IR, homeostasis model assessment; SFBR, Southwest Foundation for Biomedical Research; RT, reverse transcription; PCR, polymerase chain reaction; rRNA, ribosomal 18S RNA; LSD, least significant difference.