Equine metabolic syndrome is characterized by obesity and increased adiposity, insulin resistance, and a predisposition toward laminitis. Although our understanding of EMS is expanding, the pathological mechanisms leading to insulin resistance in horses are still under investigation.1 Formation of a metabolic profile in horses with EMS would be a key step to understanding the progression of the clinical syndrome.
Adipose tissue functions as an active endocrine organ and secretes proteins referred to as adipokines that have key roles in lipid and carbohydrate metabolism. Adiponectin is an adipokine of particular interest because it possesses profound insulin sensitizing, anti-inflammatory and antiatherosclerotic functions in multiple species. Total adiponectin concentrations are typically low in humans with obesity or type 2 diabetes, and research in horses reveals decreased total adiponectin concentrations in obese and unfit horses.2–4
Adiponectin is synthesized as a 30-kDa monomer primarily in white adipose tissue and is secreted into the peripheral circulation at high concentrations in humans (0.5 to 30 μg/mL).5 The adiponectin transcript is comprised of 3 structural domains: an N-terminal variable region that is the most diverse among species, a collagenous domain, and a C-terminal globular head region that is the receptor-binding domain. Formation of complexes in multiples of 30 kDa occurs in the endoplasmic reticulum and Golgi apparatus of adipocytes, and only the higher order complexes are secreted.6,7 In most species, circulating plasma adiponectin is composed of 3 higher order complexes: 90- to 100-kDa trimers, 180-kDa LMW hexamers, and > 360-kDa HMW multimers.7
High–molecular weight adiponectin multimers possess insulin-sensitizing, metabolically active functions in humans and rodents,8 which may be attributable to better affinity of the adiponectin receptors for the HMW complex.7,9,10 Studies evaluating adiponectin multimerization that use genetically modified mice and samples from lean and obese humans reveal that normal adipocytes secrete more of the HMW form and retain more of the LMW and trimeric forms and that impaired multimerization and secretion of adiponectin are among the causes of a diabetic phenotype.7,8,11 Thus, concentrations of HMW adiponectin decrease first, and the total adiponectin may or may not be initially affected because more LMW adiponectin may be secreted.7 Evidence in rodent studies and prospective epidemiological studies in humans reveal that HMW adiponectin concentrations correlate better than total adiponectin concentrations with a metabolic phenotype,8,12,13 but both total and HMW values correlated with metabolic disease in other studies.14,15 In humans, an increase in the ratio of HMW to total adiponectin correlates with increased insulin sensitivity after treatment with thiazolidinediones, but the total adiponectin does not correlate.8 Diabetic mice have a decreased HMW adiponectin-to-total adiponectin ratio but similar concentrations of total serum adiponectin.8 Multiple studies7,11,16–30 in humans have identified HMW adiponectin concentrations to be lower but LMW concentrations to be increased or the same in various metabolic conditions. These studies stress the importance of evaluating HMW adiponectin in addition to the total concentrations of adiponectin when evaluating metabolic phenotypes.
Enzyme-linked immunosorbent assays have been developed to specifically measure HMW adiponectin. The HMW complex is differentiated from the other complexes in ELISAs by serial protease digestion,31 a monoclonal antibody that only recognizes HMW adiponectin,32 or a combination of the 2 methods.33 Serum concentrations of total adiponectin have been assessed in horses by use of a human RIA,2–4,34,35 and a studya found that HMW adiponectin in horses was > 669 kDa and was reduced in horses with EMS versus lean horses as measured with native immunoblots. There are no peer-reviewed publications evaluating ELISAs to measure total or HMW adiponectin, to our knowledge. The objectives of the study reported here were to characterize circulating adiponectin protein complexes in serum from healthy adult horses by use of denaturing, reducing, and native gel electrophoresis; to validate total and HMW adiponectin ELISAs in horses; to compare total and HMW adiponectin concentrations (measured via ELISA) between lean and obese horses; and to evaluate the relationships of HMW adiponectin with a subjective obesity index (BCS) and serum insulin activity.
Materials and Methods
Study population characteristics—Serum samples were collected from 26 healthy lean adult horses (12 mares and 14 geldings) between 3 and 15 years of age and from 18 obese horses (6 mares and 12 geldings) between 8 and 21 years of age that were either client-owned horses (obese and some lean) or were owned by Auburn University for use on the equestrian team (lean horses). All horses were healthy as determined on the basis of physical examination, and BCS and cresty neck scores were determined by trained observers using standard scoring systems.36,37 Blood samples were collected by jugular venipuncture into evacuated tubes containing no additive and allowed to clot at room temperature (22° to 25°C) for 30 minutes. Serum was removed after centrifugation, aliquoted, and stored at −80°C for batched analysis. Insulin activity was measured by the Auburn University Endocrine Diagnostic Laboratory with a common commercial RIA validated for horses.38,39,b Food was not withheld from horses before samples were obtained, and all were collected between 10 am and 2 pm. The study protocol was approved by the Institutional Animal Care and Use Committee at Auburn University.
Standard SDS-PAGE immunoblotting for adiponectin—Serum samples from lean horses were subjected to various denaturing and reducing conditions and evaluated via standard SDS-PAGE and immunoblot analysis. Sodium dodecyl sulfate is contained in the gels and the running buffers, so the immunoblot technique is denaturing overall, but sample treatment differed in denaturing conditions and reducing agents. Higher order complexes of adiponectin are formed by noncovalent hydrogen bonding in the collagenous region of 3 monomers to form trimers, so denaturing conditions will resolve trimers into dimers and monomers. Dimers and monomers are never seen in circulation because they are not secreted from adipocytes. The LMW and HMW forms are formed by covalent disulfide bonds between trimers, also at the collagenous stalk of the protein. The HMW complexes are further modified by hydroxylation and glycosylation at the globular head region of the protein.40 Reducing agents disrupt the disulfide bonds, allowing further resolution of LMW complexes and trimers. A combination of heat denaturing and reducing is the only way to biochemically resolve adiponectin to its monomer form. High–molecular weight and LMW complexes and trimers circulate to various degrees in all mammalian species tested to date.41,42
Five microliters of diluted serum from each horse (1:10 in deionized H2O) was solubilized 1:2 in Laemmli sample buffer in the absence (nonreducing) or presence (reducing) of 5% 2-mercaptoethanol. Samples were boiled for 10 minutes (denatured) or loaded immediately (nondenatured) onto the gel. Lysate from tailhead adipose tissue from a horse (collected from a necropsy specimen) and lysate from differentiated mouse 3T3-L1 cells were used as positive controls. Equine adipose lysate was created by homogenization of 0.2 g of adipose tissue in 2 mL of lysis buffer containing 10mM Tris-HCl, 120mM NaCl (pH, 8.0), 1% tergitol-type nonyl phenoxypolyethoxylethanol, 0.2% SDS, and a protease inhibitor mixture. The lysate was passed through a 22-gauge needle, shaken at 1,000 oscillations/min for 30 minutes at 4°C, and clarified by centrifugation at 14,000 × g for 15 minutes at 4°C. Protein concentration was determined with a commercial protein assay,c and 20 μg of adipose lysate solubilized 1:2 in Laemmli buffer was loaded on the gel. Protein bands were separated on 4% to 15% tris-glycine gelsd at 200 V for 35 to 45 minutes and transferred to PVDF membranes at 100 V for 1 hour. Membranes were stained with Ponceau S to confirm transfer. After blocking (5% nonfat dry milk in Tris-buffered saline [0.9% NaCl] solution with 0.05% Tween 20) for 1 hour, immunoblot analysis with a commercial rabbit polyclonal antiadiponectin antibodye (1:1,000 diluted in 5% nonfat milk 12 to 16 hours at 4°C) was performed. This antibody was directed against a synthetic peptide (amino acids 225 to 244 of human adiponectin). This corresponds to the globular head region of the protein and is similar among species. Enhanced chemiluminescent signal was elicited with a goat anti-rabbit IgG horseradish peroxidase–conjugated secondary antibodyf (1:5,000 diluted in 5% nonfat milk for 1.5 hours at room temperature) and a standard film developer. Band intensities were measured via densitometry.
Total and HMW adiponectin ELISA—Serum concentrations of HMW adiponectin were analyzed in duplicate from batched frozen (—80°C) samples by use of a commercially available human sandwich ELISA kith according to the manufacturer's instructions except where specifically indicated. Briefly, 20 μL of equine serum or digestion quality control were added to 170 μL of proprietary digestion buffer (from the ELISA kit), mixed on an orbital microtiter plate shaker at 700 oscillations/min for 10 minutes (5 minutes longer than instructions), and incubated at 37°C for 3 hours (1 hour longer than instructions). After 2 hours, 10 μL of the digested serum and buffer solution were added to 190 μL of sample dilution buffer, resulting in a final sample dilution of 1:200 used for the assay procedure. Ten microliters of the diluted and digested samples and digestion control and 10 μL of preprepared HMW adiponectin standard dilutions (1.56 to 200 ng/mL) were added to 90 μL of assay buffer on the ELISA plate coated with a monoclonal anti-HMW adiponectin antibody. Samples and standards were incubated on the ELISA plate with shaking at 450 oscillations/min for 2 hours.i Standard sandwich ELISA procedures of washing and detection with a second polyclonal biotinylated antiadiponectin antibody were performed. Colorimetric signal was detected with a streptavidin–horseradish peroxidase–conjugated secondary antibody and substrate. A blue color change indicated success of the reaction, which was then stopped with acidification. A yellow color change was then detectable. Absorbance was read at 450 nm in a microtiter plate reader within 5 minutes after stopping the reaction. The absorbance of the highest HMW standard (200 ng/mL) did not exceed an optical density of 2.8 units. After the absorbance was read, a concentration (ng/mL) was reported, which was the concentration range of the standards, and the 1:200 dilution of the equine samples is in this range. A correction for the 1:200 dilution was then made, bringing sample concentrations into the μg/mL range. The sensitivity of this assay was 0.5 ng of HMW adiponectin/mL, and all samples and standards were measured in duplicate.
Linearity and parallelism of the assay for equine samples were assessed by recovery calculation after serial dilution or spiked standard adiponectin in samples from 4 lean horses. A standard curve and digestion quality control samples were evaluated with every ELISA performed, including the validation. To evaluate parallelism, pooled equine serum was digested and diluted as described, except 5 μL of serum was used instead of 20 μL so that when the standards were spiked with the diluted and digested equine serum, the optical density would not read > 2.8 units. Then, 10 μL of the digested and diluted sample was added to each of the standard concentrations and used in the ELISA. The correction for the dilution factor was not used when reporting the expected concentrations so that the standards and the equine serum would be on the same scale. For the linearity calculation, 2.5-, 5-, 10-, and 20-μL aliquots of equine serum were digested and diluted and used in the ELISA as described. The intraassay CV was calculated from 20 replicates of a pooled high-concentration sample (9 to 21 μg/mL), a pooled medium-concentration sample (3 to 5 μg/mL), and a pooled low-concentration sample (0.7 to 1.5 μg/mL) analyzed in the same assay. Interassay CV was calculated by repeating the ELISA 5 times on pooled low-, pooled high-, and pooled medium-concentration samples on different days by 2 operators. Attempts to measure total adiponectin were made by use of 3 commercial human and rodent ELISAs.j–l These ELISAs did not detect the equine protein, so no results were reported.
Native immunoblot for HMW adiponectin—Western blot analysis of equine HMW adiponectin under native conditions was performed with 0.5 μL of equine serum/lane by use of a described method with tris-acetate gels.43 Human serum (1 μL) was used as a positive control. Equine and human serum samples were diluted 1:10 in deionized H2O, and 5 μL of the diluted horse serum or 10 μL of the diluted human serum was solubilized in native sample buffer,m separated by electrophoresis by use of native buffersn on 3% to 8% tris-acetate gelso at 150 V for 2 hours, and transferred to PVDF for 1.5 hours at 100 V. Membranes were stained with Ponceau S to confirm transfer. There were slight differences in the protocol depending on the primary antibody used. For the native immunblots performed with the monoclonal human antiadiponectin antibody from the human HMW ELISA, membranes were blocked in 5% nonfat milk for 1 hour and then immunoblotted with the primary antibody (1:1,000 in 5% nonfat milk for 12 to 16 hours at 4°C). This antibody is not commercially available and was obtained by our specific request to the company to validate this ELISA for horses. To confirm specificity of the monoclonal adiponectin antibody, an identical membrane was blotted with anti-human HMW adiponectin antibody that was preadsorbed with 40 μg of purified human HMW adiponectin protein (the standard from the ELISA was used) for 1 hour before the 2-hour incubation with the membrane. Enhanced chemiluminescent signal was elicited with a goat anti-mouse IgG horseradish peroxidase–conjugated secondary antibodyp (1:10,000 in 5% nonfat milk for 1 hour at room temperature) and a standard film developer.
For native immunoblots performed with the commercial polyclonal adiponectin antibody (previously used for the denaturing gels), PVDF membranes were blocked in commercial blocking bufferq and immunoblotted with the commercial polyclonal antibody (1:1,000 in blocking buffer for 3 hours at 4°C). Fluorescent signal was obtained after incubating with a goat-anti-rabbit IgG secondary antibodyr (1:10,000 in blocking buffer with 0.1% Tween-20 and 0.01% SDS for 1 hour at room temperature) and imaging with an infrared scanner.s The different protocols were used because of the need for different secondary antibodies to avoid secondary cross-reactivity with higher molecular weight equine proteins, as was identified in preliminary experiments.
Data analysis—All data are reported as mean ± SD values, with 95% confidence intervals where appropriate. All data were evaluated for normality by use of the D'Agostino and Pearson statistic. Non-normal data were natural log transformed where indicated to obtain a normal distribution. Differences in HMW adiponectin between lean and obese horses were determined on natural log-transformed data via a 1-way ANOVA. If the overall difference between groups was significant (P < 0.05), Tukey post hoc comparison was used to determine significance between individual groups, maintaining an experiment-wise error rate of P < 0.05. Insulin data were not normally distributed even after transformation, so analysis between groups was performed by use of the Kruskal-Wallis test and Dunn multiple comparison test if overall significance (P < 0.05) was detected between groups. Correlations between BCS, HMW adiponectin, and insulin were examined by calculating Spearman correlation coefficients. Difference in densitometry percentages of the various adiponectin complexes between lean and obese horses were identified via 2-way ANOVA and Bonferonni post hoc analysis. All analyses were performed with a commercial software package.t For all comparisons, a value of P < 0.05 was considered significant.
Results
Study population characteristics—Breeds represented by the lean horses were Quarter Horses (10/26), warmbloods (7/26), Thoroughbreds (3/26), Tennessee Walking Horses (2/26), and grades (4/26). Horses included both client-owned horses and Auburn University–owned horses. All lean horses had a BCS between 4 and 6 on a 9-point scale. Lean horses had a cresty neck score < 3 on a 5-point scale. All lean horses were used for athletic events (western pleasure, reining, English pleasure, and show jumping) and were ridden approximately 5 times/wk. Serum samples from client-owned obese horses were collected at time of routine examinations by the Auburn University Ambulatory service. The BCS of the obese horses ranged from 7 to 9 of 9, and all horses had cresty neck scores ≥ 3 of 5. Breeds of obese horses included Tennessee Walking Horses (6/18), Spanish breeds (2/18), Quarter Horses (2/18), pony breeds (3/18), Arabians (2/18), American Saddlebred (1/18), and grades (2/18).
Evaluation of adiponectin complexes with standard SDS-PAGE—Western blot analysis by use of standard tris-glycine SDS-PAGE was used to evaluate adiponectin complexes in equine serum and adipose tissue (Figure 1). The molecular weight profile obtained in untreated samples (nondenaturing and nonreducing) via SDS-PAGE revealed a higher molecular weight band at > 250 kDa, bands between 120 and 250 kDa, and bands between 90 and 120 kDa. The bands between 120 and 250 kDa were consistent with LMW adiponectin, and the band at approximately 90 kDa was consistent with trimers. Denaturing or reducing individually produced more distinct protein bands at lower (90-kDa trimer and 60-kDa dimer) molecular weights. The 30-kDa monomer and 28-kDa globular head region were only identified under both denaturing and reducing conditions, and the higher order complexes were not observed when both denaturing and reducing conditions were applied. The positive control of lysate from 3T3-L1 cells had a prominent monomer band and faint dimer and trimer bands, which is typical of the adiponectin secreted from this cell line.44 The intensity of bands was lower in obese horses versus lean horses, suggesting less total adiponectin in the serum of the obese horses used in this experiment. A similar pattern of complex breakdown was evident in equine adipose tissue lysate under the various denaturing and reducing conditions.
Results of western blot analysis performed by use of standard tris-glycine SDS-PAGE performed under nondenaturing (ND) and nonreducing (NR), denaturing (D) and NR, ND and reducing (R), and D and R conditions to evaluate adiponectin complexes in serum and adipose tissue of horses. A—Notice that bands from 120 to 250 kDa (lower molecular weight), 90 to 120 kDa (trimer), 60 kDa (dimer), 30 kDa (monomer), and 28 kDa (globular [Glob.] head region) became apparent with DN and R conditions. A higher–molecular weight band at > 250 kDa was identified under ND and NR conditions and to a lesser extent under ND and R conditions. Denatured and reduced media from 3T3L1 adipocytes were used as a positive control for the monomer and globular head. Band intensity is reduced overall in the serum from the obese horse, compared with the lean horse. B—Twenty micrograms of lysate from equine tail-head adipose tissue was separated by SDS-PAGE with similar bands to panel A becoming apparent with D and R conditions.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Results of western blot analysis performed by use of standard tris-glycine SDS-PAGE performed under nondenaturing (ND) and nonreducing (NR), denaturing (D) and NR, ND and reducing (R), and D and R conditions to evaluate adiponectin complexes in serum and adipose tissue of horses. A—Notice that bands from 120 to 250 kDa (lower molecular weight), 90 to 120 kDa (trimer), 60 kDa (dimer), 30 kDa (monomer), and 28 kDa (globular [Glob.] head region) became apparent with DN and R conditions. A higher–molecular weight band at > 250 kDa was identified under ND and NR conditions and to a lesser extent under ND and R conditions. Denatured and reduced media from 3T3L1 adipocytes were used as a positive control for the monomer and globular head. Band intensity is reduced overall in the serum from the obese horse, compared with the lean horse. B—Twenty micrograms of lysate from equine tail-head adipose tissue was separated by SDS-PAGE with similar bands to panel A becoming apparent with D and R conditions.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Results of western blot analysis performed by use of standard tris-glycine SDS-PAGE performed under nondenaturing (ND) and nonreducing (NR), denaturing (D) and NR, ND and reducing (R), and D and R conditions to evaluate adiponectin complexes in serum and adipose tissue of horses. A—Notice that bands from 120 to 250 kDa (lower molecular weight), 90 to 120 kDa (trimer), 60 kDa (dimer), 30 kDa (monomer), and 28 kDa (globular [Glob.] head region) became apparent with DN and R conditions. A higher–molecular weight band at > 250 kDa was identified under ND and NR conditions and to a lesser extent under ND and R conditions. Denatured and reduced media from 3T3L1 adipocytes were used as a positive control for the monomer and globular head. Band intensity is reduced overall in the serum from the obese horse, compared with the lean horse. B—Twenty micrograms of lysate from equine tail-head adipose tissue was separated by SDS-PAGE with similar bands to panel A becoming apparent with D and R conditions.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Measurement of HMW adiponectin by ELISA—Total adiponectin was not detectable by use of any of the commercially available ELISAs. The human HMW adiponectin ELISA detected HMW adiponectin in horse serum. Recovery analysis by use of pooled equine serum spiked with HMW adiponectin standards was 94% to 109% (Table 1; Figure 2). Recovery after serial dilution of equine serum (linearity) was 72% to 109% with an R2 value of 0.99. The intraassay CV of this ELISA for equine samples was 9% for the pooled high-concentration sample, 11% for the pooled medium-concentration sample, and 11% for the pooled low-concentration sample. The interassay CV was 13% for the pooled high-concentration sample, 11% for the pooled medium-concentration sample, and 23% for the pooled low-concentration sample. The IQR of HMW adiponectin in lean horses ranged from 4.4 to 9.2 μg/mL and in obese horses from 0.7 to 4.9 μg/mL (Table 2), but there were horses with higher HMW values in both the lean (24.2 μg/mL) and obese (14.2 μg/mL) groups. Mean ± SD values and IQR for HMW adiponectin and insulin in the different groups are reported. Mean HMW adiponectin concentration was significantly higher in lean (8.0 ± 4.6 μg/mL) versus obese (3.6 ± 3.9 μg/mL) horses, and this difference was not influenced by sex. There was no difference in HMW adiponectin concentration between lean mares and geldings or obese mares and geldings. High–molecular weight adiponectin concentration was negatively correlated with BCS and serum insulin activity but not with age (ρ = −0.147; P = 0.474; Figure 3). Serum insulin activity was positively correlated with BCS (ρ = 0.678; P = 4.1 × 10−-6) as expected, and insulin activity was significantly higher in obese horses (85 ± 118 μU/mL) versus lean horses (5.8 ± 6.2 μU/mL).
Percentage recovery of HMW adiponectin determined with an HMW adiponectin ELISA.
| HMW adiponectin standard (ng/mL) | Expected value (ng/mL) | Observed value (ng/mL) | Recovery (%) |
|---|---|---|---|
| 0 | 14.6 | 14.6 | — |
| 1.563 | 16.1 | 15.2 | 94 |
| 3.125 | 17.7 | 17.2 | 97 |
| 6.25 | 20.8 | 21.2 | 102 |
| 12.5 | 27.1 | 27.9 | 103 |
| 25 | 39.6 | 43.1 | 109 |
| 50 | 64.6 | 63.6 | 99 |
| 100 | 114.6 | 112.9 | 99 |
The expected value is the amount that the addition of equine serum would be expected to increase the standard value (calculated from the measurement of the equine serum without any standard added), and the observed value is what was actually measured with the ELISA.
— = Not applicable.
High–molecular weight adiponectin ELISA performance characteristics. A—Graph of mean optical density (OD) values versus increasing concentrations of HMW adiponectin standards, revealing parallelism through spike recovery analysis. B—Notice linearity under dilution (1:2, 1:4, 1:8, and 1:10 dilutions) in equine serum.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
High–molecular weight adiponectin ELISA performance characteristics. A—Graph of mean optical density (OD) values versus increasing concentrations of HMW adiponectin standards, revealing parallelism through spike recovery analysis. B—Notice linearity under dilution (1:2, 1:4, 1:8, and 1:10 dilutions) in equine serum.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
High–molecular weight adiponectin ELISA performance characteristics. A—Graph of mean optical density (OD) values versus increasing concentrations of HMW adiponectin standards, revealing parallelism through spike recovery analysis. B—Notice linearity under dilution (1:2, 1:4, 1:8, and 1:10 dilutions) in equine serum.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Mean serum HMW adiponectin concentration and insulin activity in lean and obese horses.
| Group | HMW adiponectin (μg/mL) | Insulin (μU/mL) |
|---|---|---|
| Lean mares (n = 12) | 8.8 ± 5.9a (4.1–9.4) | 5.1 ± 6.6a (2.3–4.3) |
| Lean geldings (n = 14) | 7.4 ± 3.2a (4.9–9.2) | 6.4 ± 5.9a (2.3–9.9) |
| All lean (n = 26) | 8.0 ± 4.6a (4.4–9.2) | 5.8 ± 6.2a (2.3–5.7) |
| Obese mares (n = 6) | 4.4 ± 4.4b (0.3–9.4) | 85 ± 151b (6.3–127) |
| Obese geldings (n = 12) | 3.2 ± 3.8b (0.8–4.4) | 86 ± 105b (20–124) |
| All obese (n = 18) | 3.6 ± 3.9b (0.7–4.9) | 85 ± 118b (18–107) |
Data are mean ± SD (IQR).
Different superscript letters indicate groups are significantly (P < 0.05) different within a column.
Evaluation of serum HMW adiponectin concentration in lean versus obese horses, and BCS and serum insulin activity in relation to HMW adiponectin concentration. A—High–molecular weight adiponectin concentration in 26 lean and 18 obese adult horses. * Value is significantly (P < 0.001) different from that of lean horses. Each box represents the 25th to 75th percentile, the line through the box is the median value, the short line in the box is the mean value, the whiskers are the 90th and 10th percentiles, and circles indicate outliers. B—Correlation between BCS and HMW adiponectin concentration. C—Correlation between insulin activity and HMW adiponectin concentration. For panels B and C, Spearman correlation coefficients are reported as ρ values and triangles represent values of individual horses.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Evaluation of serum HMW adiponectin concentration in lean versus obese horses, and BCS and serum insulin activity in relation to HMW adiponectin concentration. A—High–molecular weight adiponectin concentration in 26 lean and 18 obese adult horses. * Value is significantly (P < 0.001) different from that of lean horses. Each box represents the 25th to 75th percentile, the line through the box is the median value, the short line in the box is the mean value, the whiskers are the 90th and 10th percentiles, and circles indicate outliers. B—Correlation between BCS and HMW adiponectin concentration. C—Correlation between insulin activity and HMW adiponectin concentration. For panels B and C, Spearman correlation coefficients are reported as ρ values and triangles represent values of individual horses.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Evaluation of serum HMW adiponectin concentration in lean versus obese horses, and BCS and serum insulin activity in relation to HMW adiponectin concentration. A—High–molecular weight adiponectin concentration in 26 lean and 18 obese adult horses. * Value is significantly (P < 0.001) different from that of lean horses. Each box represents the 25th to 75th percentile, the line through the box is the median value, the short line in the box is the mean value, the whiskers are the 90th and 10th percentiles, and circles indicate outliers. B—Correlation between BCS and HMW adiponectin concentration. C—Correlation between insulin activity and HMW adiponectin concentration. For panels B and C, Spearman correlation coefficients are reported as ρ values and triangles represent values of individual horses.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Native immunoblot analysis of HMW adiponectin—To further ensure specificity of the human HMW ELISA for equine samples, western blot analysis of equine serum under native (nonreducing and nondenaturing) conditions was performed with the anti-human HMW adiponectin monoclonal antibody from the human HMW ELISA kit (Figure 4). Native gel electrophoresis allows resolution of HMW proteins and protein complexes. A distinct HMW band was detected in samples from lean horses at approximately 750 to 800 kDa, and fainter bands were noted at approximately 540 and 300 kDa in some samples. The enhanced chemiluminescent signal for these bands was blocked when the antibody was preadsorbed with purified HMW adiponectin from the ELISA kit, thus blocking the antibody-binding sites for the equine HMW adiponectin on the membrane. The antibody did not detect LMW complexes, trimers, dimers, or monomers on denaturing gels, and the LMW complexes and trimers were not evident on the native immunoblots run with resolution of lower molecular weight proteins.
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions. A—L1 and L2 represent lean horses, ob1 and ob2 represent obese horses with low HMW adiponectin concentration (< 1.5 μg/mL), and Hu represents human serum. Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 and 300 kDa in some samples. B—PL represents diluted pooled serum of 2 lean horses, Pob represents diluted pooled serum of 2 obese horses. Anti-human HMW adiponectin antibody was either preadsorbed with purified human HMW adiponectin protein (+) or not preadsorbed (−). Notice that band density is decreased in the blocked PL lane and completely absent in the blocked Pob lane. For panels A and B, a representative band of protein from the ponceau S–stained membrane is included to demonstrate equal loading of samples. C—L1, L2, and L3 reptesent lean horses. To determine antibody specificity for HMW adiponectin, the gel was allowed to run such that the 66-kDa marker was still present. Notice that no bands < 300 kDa are apparent, confirming specificity of the antibody from the ELISA kit.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions. A—L1 and L2 represent lean horses, ob1 and ob2 represent obese horses with low HMW adiponectin concentration (< 1.5 μg/mL), and Hu represents human serum. Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 and 300 kDa in some samples. B—PL represents diluted pooled serum of 2 lean horses, Pob represents diluted pooled serum of 2 obese horses. Anti-human HMW adiponectin antibody was either preadsorbed with purified human HMW adiponectin protein (+) or not preadsorbed (−). Notice that band density is decreased in the blocked PL lane and completely absent in the blocked Pob lane. For panels A and B, a representative band of protein from the ponceau S–stained membrane is included to demonstrate equal loading of samples. C—L1, L2, and L3 reptesent lean horses. To determine antibody specificity for HMW adiponectin, the gel was allowed to run such that the 66-kDa marker was still present. Notice that no bands < 300 kDa are apparent, confirming specificity of the antibody from the ELISA kit.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions. A—L1 and L2 represent lean horses, ob1 and ob2 represent obese horses with low HMW adiponectin concentration (< 1.5 μg/mL), and Hu represents human serum. Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 and 300 kDa in some samples. B—PL represents diluted pooled serum of 2 lean horses, Pob represents diluted pooled serum of 2 obese horses. Anti-human HMW adiponectin antibody was either preadsorbed with purified human HMW adiponectin protein (+) or not preadsorbed (−). Notice that band density is decreased in the blocked PL lane and completely absent in the blocked Pob lane. For panels A and B, a representative band of protein from the ponceau S–stained membrane is included to demonstrate equal loading of samples. C—L1, L2, and L3 reptesent lean horses. To determine antibody specificity for HMW adiponectin, the gel was allowed to run such that the 66-kDa marker was still present. Notice that no bands < 300 kDa are apparent, confirming specificity of the antibody from the ELISA kit.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
The commercial polyclonal antiadiponectin antibody was also used on native immunoblots to confirm the results from the ELISA kit with a different antibody in serum from horses with high and low concentrations of HMW complexes (Figure 5). These 18 horses were chosen so that the maximum number of samples could be run on the same gel and so that horses with a range of HMW adiponectin concentrations were depicted. Obese horses with high and low adiponectin concentrations, and lean horses with high concentrations of HMW adiponectin were specifically selected. Band intensity was consistent with the amount of HMW adiponectin measured by ELISA in all samples and revealed that not all obese horses had low concentrations of HMW adiponectin.
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions by use of a commercial polyclonal antiadiponectin antibody. A—Serum from lean horses (n = 8) and obese horses (10). Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 kDa. B—The HMW ELISA results and the insulin activities for each immunoblotted sample are depicted.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions by use of a commercial polyclonal antiadiponectin antibody. A—Serum from lean horses (n = 8) and obese horses (10). Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 kDa. B—The HMW ELISA results and the insulin activities for each immunoblotted sample are depicted.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis of equine HMW adiponectin under native (nonreducing and nondenaturing) conditions by use of a commercial polyclonal antiadiponectin antibody. A—Serum from lean horses (n = 8) and obese horses (10). Notice the strongest HMW band at approximately 800 kDa and fainter bands at approximately 540 kDa. B—The HMW ELISA results and the insulin activities for each immunoblotted sample are depicted.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Distribution of adiponectin by molecular weight in lean versus obese horses—To semiquantitatively compare complex distribution, band intensities for the different complexes were compared (Figure 6) for samples from the same 18 lean and obese horses (Figure 5). Samples were all untreated (no denaturing or reducing), so no bands < 90 kDa were identified on the gel. Band intensity was calculated as a percentage of the total density for each of the bands represented on the membrane. Comparisons were made among all lean horses, all obese horses, obese horses with HMW adiponectin concentrations < 5 μg/mL as determined via ELISA, and obese horses with HMW adiponectin concentrations > 5 μg/mL as determined via ELISA. Higher molecular weight complexes > 250 kDa were 32% of the total density in lean horses, 22% in all obese horses, 29% in obese horses with HMW adiponectin concentrations > 5 μg/mL, and 15% in obese horses with HMW adiponectin concentrations < 5 μg/mL. The percentage of complexes > 250 kDa was significantly higher in lean versus obese horses with HMW adiponectin concentrations < 5 μg/mL and in obese horses with HMW adiponectin concentrations > 5 μg/mL versus those with HMW adiponectin concentrations < 5 μg/mL. The lower molecular weight complexes between 120 and 250 kDa did not significantly differ among any groups and were approximately 50% of the total. The trimer percentage was significantly higher in obese horses with HMW adiponectin concentrations < 5 μg/mL (33% of the total) versus those with HMW adiponectin concentrations > 5 μg/mL (18% of the total). The concentrations of trimers (18% of the total) were also significantly lower in lean horses than in obese horses with HMW adiponectin concentrations < 5 μg/mL (33% of the total). The total of all bands < 250 kDa was significantly higher in obese horses with HMW adiponectin concentrations < 5 μg/mL (85%) versus lean horses (68%) and obese horses with HMW adiponectin concentrations > 5 μg/mL (70%). Total adiponectin density was significantly lower in obese horses with HMW adiponectin concentrations < 5 μg/mL versus lean horses or obese horses with HMW adiponectin concentrations > 5 μg/mL.
Western blot analysis by use of standard Tris-glycine SDS-PAGE to evaluate the percentage distribution of different molecular weight categories of adiponectin in lean and obese horses. A—Sera from lean (n = 8) or obese (10) horses were separated by electrophoresis under nondenaturing and nonreducing conditions. The plus symbol indicates that an obese horse had an HMW concentration > 5 μg/mL; an obese horse without a symbol had a concentration < 5 μg/mL. Bands from 120 to 200 kDa (lower molecular weight), 90 to 120 kDa (trimer), and > 250 kDa (higher-molecular weight complexes) were compared by use of densitometry. B—Comparison among all groups of mean ± SD total band densities (%) for each molecular weight category. *Value is significantly (P < 0.05) different, compared with lean horses. †Value is significantly (P < 0.05) different, compared with obese horses with HMW concentration > 5 μg/mL. C—Comparison of mean ± SD total adiponectin band density (all bands on gel) among groups.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis by use of standard Tris-glycine SDS-PAGE to evaluate the percentage distribution of different molecular weight categories of adiponectin in lean and obese horses. A—Sera from lean (n = 8) or obese (10) horses were separated by electrophoresis under nondenaturing and nonreducing conditions. The plus symbol indicates that an obese horse had an HMW concentration > 5 μg/mL; an obese horse without a symbol had a concentration < 5 μg/mL. Bands from 120 to 200 kDa (lower molecular weight), 90 to 120 kDa (trimer), and > 250 kDa (higher-molecular weight complexes) were compared by use of densitometry. B—Comparison among all groups of mean ± SD total band densities (%) for each molecular weight category. *Value is significantly (P < 0.05) different, compared with lean horses. †Value is significantly (P < 0.05) different, compared with obese horses with HMW concentration > 5 μg/mL. C—Comparison of mean ± SD total adiponectin band density (all bands on gel) among groups.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Western blot analysis by use of standard Tris-glycine SDS-PAGE to evaluate the percentage distribution of different molecular weight categories of adiponectin in lean and obese horses. A—Sera from lean (n = 8) or obese (10) horses were separated by electrophoresis under nondenaturing and nonreducing conditions. The plus symbol indicates that an obese horse had an HMW concentration > 5 μg/mL; an obese horse without a symbol had a concentration < 5 μg/mL. Bands from 120 to 200 kDa (lower molecular weight), 90 to 120 kDa (trimer), and > 250 kDa (higher-molecular weight complexes) were compared by use of densitometry. B—Comparison among all groups of mean ± SD total band densities (%) for each molecular weight category. *Value is significantly (P < 0.05) different, compared with lean horses. †Value is significantly (P < 0.05) different, compared with obese horses with HMW concentration > 5 μg/mL. C—Comparison of mean ± SD total adiponectin band density (all bands on gel) among groups.
Citation: American Journal of Veterinary Research 73, 8; 10.2460/ajvr.73.8.1230
Discussion
Circulating adiponectin complexes account for approximately 0.01% to 0.03% of total plasma proteins in mammalian species, circulating at concentrations 3 times those of most hormones in humans.5 Total adiponectin concentrations have been extensively characterized in a number of species and range from 5 to 30 μg/mL in humans, 9 to 17.4 μg/mL in mice, and 0.052 to 33.8 μg/mL in dogs.5,41,45 Compared with these findings, the total range of results of the ELISA for HMW adiponectin concentration in lean horses in the present study (2.8 to 24.2 μg/mL) is reasonable and expected. Previous studies2–4,34,35,41 evaluating adiponectin in horses found much lower concentrations of total adiponectin (1.3 to 4.9 μg/mL) by use of a human RIA. After identifying much higher amounts of HMW adiponectin via the ELISA used in the present study than the total amounts reported for horses, we attempted to use different ELISAs to measure total adiponectin. Unfortunately, no ELISA tested was able to recognize the equine protein. Antibodies for commercial assays are often made with peptide sequences in the N-terminal variable region of the protein to achieve species specificity, and the peptide sequences used to create the antibodies are often proprietary (as was the case for assays used in the present study), so comparisons are difficult. Considering that use of RIA and ELISA has produced significantly different results for total adiponectin in other species,41 our results suggest that the adiponectin antibody used in the reported RIA recognizes the equine protein but with low affinity.
In the present study, the complex distribution of adiponectin in comparison to total adiponectin was evaluated via western blot analysis. Several obese horses had HMW adiponectin concentrations measured by ELISA that were on par with concentrations in lean horses (> 5 μg/mL). Interestingly, those horses were not different from the lean horses in total adiponectin density or in the percentage of the different molecular weight categories. The obese horses with HMW adiponectin concentrations < 5 μg/mL had significantly lower total adiponectin and a lower percentage of higher order complexes but an increase in the percentage of trimers and in the percentage of all complexes < 250 kDa. This change in ratio is consistent with the literature reported on adiponectin complexes in other species, although no difference was identified in the 120- to 250-kDa molecular weight bands among any groups. This suggests that obese horses have more changes in the HMW complexes and in the trimers versus the LMW complexes. However, we interpret these data with some caution because although use of this method has been described in the literature,11 the biochemical separation of the complexes is somewhat undefined and highly susceptible to slight changes in temperature or denaturing state, so the ratio may not be completely accurate.7 We were only able to visualize all 3 complexes on a partially denaturing and reducing gel, and the higher molecular weight complexes visible on these gels were only slightly > 250 kDa. The native gel revealed that equine HMW adiponectin is > 720 kDa. That would suggest that some of the HMW complex is breaking down to smaller complexes and not completely entering the denaturing SDS-PAGE gel, which would make measurement of the HMW complex on this gel potentially inaccurate.
Despite the inability to quantitatively measure total adiponectin in horses, the success of the ELISA for HMW adiponectin measurement is a crucial step for understanding the role of adiponectin in equine metabolic disease. Adiponectin (HMW adiponectin in particular) not only correlates with metabolic disorders but also directly affects glucose and lipid insulin metabolism.46–50 There is still controversy about the predictive value of adiponectin in human cardiovascular disease51 and type 1 diabetes52 and about the precise role of adiponectin in human insulin resistance.53 However, a strong correlation between adiponectin and metabolic syndrome, type 2 diabetes, and obesity and the importance of the HMW complex in this correlation has been found repeatedly 7,8,11,16–30,54 Measurement of both HMW and total adiponectin is typically performed in human and rodent studies, but 1 report14 suggests that measurement of total adiponectin is just as useful for predicting metabolic disease, whereas other studies55,56 have measured HMW adiponectin only. Optimization of a total adiponectin assay that allows high throughput analysis of both total and HMW adiponectin in horses would be ideal, but the HMW adiponectin ELISA is an essential tool to begin these analyses until such an assay is developed.
The higher order multimers of adiponectin are joined through tertiary and quaternary structures, including hydrophobic interactions and covalent disulfide bonds. Even with denaturing SDS-PAGE, additional denaturing and reducing are required to break the complexes into the smallest 30-kDa monomer. It has been suggested that the HMW complex is primarily held by disulfide bonds and interactions between hydroxylated and glycosylated residues and that the smaller complexes (< 60 kDa) are held by noncovalent interactions.6 High–molecular weight adiponectin in horses was determined to be > 720 kDa by use of native electrophoresis. Serum adiponectin has been resolved at 900 kDa in humans and up to 540 kDa in cattle and mice.40 These results suggest that the higher order adiponectin complexes in horses are more similar to those found in human serum, compared with other species.
A consistent sexual dimorphism exists for HMW adiponectin, with females having higher amounts of the HMW complex than do males in humans and rodents.57 In the present study, a difference in HMW adiponectin between mares and geldings was not detected. In other species, the sexual dimorphism is thought to be caused by the influence of testosterone on adipocyte metabolism, so the most likely reason the present study did not find a difference in HMW adiponectin was the relative absence or near absence of testosterone in the geldings. Although it is plausible that the study did not have adequate power to detect a difference between geldings and mares, these results are consistent with previous findings that adiponectin synthesis and secretion is regulated by testosterone.58
Serum HMW adiponectin concentrations were significantly higher in lean horses, compared with obese horses. In fact, BCS accounted for approximately 40% of the variation in HMW adiponectin concentrations. Although insulin sensitivity was not examined in the present study, a significant positive correlation was observed between insulin activity and BCS and a significant negative correlation was observed between HMW adiponectin and serum insulin activity. The baseline insulin activity ranges in the present study were consistent with what would be expected for lean and obese animals, but food was not withheld before samples were obtained, so they must be interpreted with caution. Samples were all obtained in the middle of the day several hours after provision of grain in the morning, and insulin activities in lean horses were all between 2.3 and 12 μU/mL, except for that in 1 horse, for which insulin activity was 25 μU/mL. The insulin activities in obese horses (which were sampled under the same conditions) ranged from 5.2 to 392 μU/mL and were significantly higher than in the lean horses.
There was no correlation between HMW adiponectin concentration and age, but that association may be less robust and require many more samples over a wide range of ages. The lean horses were all < 15 years old in the present study. Despite the fairly strong relationship between HMW adiponectin, BCS, and serum insulin activity, we did observe high concentrations of HMW adiponectin in obese animals and somewhat lower concentrations of HMW adiponectin in lean animals (although no lean horses in the present study had an HMW adiponectin concentration < 2.8 μg/mL). It is tempting to speculate that horses that are able to sustain adiponectin concentrations despite the development of metabolic derangements may have a reduced incidence or progression of disease, but studies with larger numbers of horses are required to test this hypothesis. There were no obvious differences in diet, breed, or sex in the obese horses with high concentrations of HMW adiponectin. No mares were pregnant or lactating. Insulin activities were determined for the individual horses, but the number of horses was still too low to speculate on whether obese horses with high insulin activities had lower concentrations of HMW adiponectin, especially in light of the fact that food was not withheld before the insulin samples were obtained and samples were obtained at various times during the year. As an example, 1 obese horse with a high concentration of HMW adiponectin (9.7 μg/mL) had the highest insulin activity measured in the study (392 μU/mL). Again, future studies with large numbers of horses correlating insulin sensitivity, season, breed, diet, and incidence of laminitis with adiponectin are needed to fully characterize the role of adiponectin in equine metabolic disorders. A high-throughput tool with which to measure HMW adiponectin concentration, such as the ELISA described in the present report, will make these studies possible.
The ability to successfully and consistently measure HMW adiponectin in horses will greatly enhance the ability of researchers to form metabolic profiles of horses and open new avenues of research. The metabolic profiles may aid in identification of specific horses at risk of developing complications of EMS, such as laminitis, so that appropriate preventative measures can be implemented sooner. Adiponectin is of increasing interest as a biomarker and a potential drug target in human medicine, so new treatments and diagnostic tests may be developed for horses.42,59,60
The HMW adiponectin concentration can be measured in horses by use of a commercially available human ELISA kit and was negatively correlated with obesity and serum insulin activity. Adding HMW adiponectin concentration to the variables evaluated in research studies of EMS should further elucidate the role of this hormone in equine metabolic disease.
ABBREVIATIONS
| BCS | Body condition score |
| CV | Coefficient of variation |
| EMS | Equine metabolic syndrome |
| HMW | High molecular weight |
| IQR | Interquartile range |
| LMW | Low molecular weight |
| PVDF | Polyvinylidene difluoride |
| RIA | Radioimmunoassay |
Chameroy KA. Diagnosis and management of horses with equine metabolic syndrome. PhD dissertation, Graduate School, University of Tennessee, Knoxville, Tenn, 2010.
Coat-A-Count Insulin RIA, Siemans Medical Solutions Diagnostics, Los Angeles, Calif.
Bio-Rad DC Protein Assay, Hercules, Calif.
Bio-Rad Ready Gel, Hercules, Calif.
Antiadiponectin, Sigma-Aldrich, St Louis, Mo.
Goat anti-rabbit IgG, HRP conjugated, Santa Cruz Biotechnology Inc, Santa Cruz, Calif.
ImageJ, version 1.43u, National Institutes of Health, Bethesda, Md.
Millipore Human HMW adiponectin (EZHMWA-64K) ELISA, Billerica, Mass.
Microplate orbital shaker, Eppendorf, Hamburg, Germany.
Millipore mouse (EZMADP-60K) total adiponectin ELISA, Billerica, Mass.
Millipore human (EZHADP-61K) total adiponectin ELISA, Billerica, Mass.
Millipore rat (EZRADP-62K) total adiponectin ELISA, Billerica, Mass.
Novex Tris-Glycine native sample buffer, Invitrogen, Carlsbad, Calif.
Novex Tris-Glycine native running buffer, Invitrogen, Carlsbad, Calif.
NuPage Tris-Acetate gels, Invitrogen, Carlsbad, Calif.
Goat anti-mouse, HRP conjugated, Stressgen, Enzo Life Sciences, Farmindale, NY.
Odyssey blocking buffer, LI-COR Biosciences, Lincoln, Ne.
IRDye 680LT Goat anti-rabbit IgG, LI-COR Biosciences, Lincoln, Ne.
Odyssey infrared imaging system, LI-COR Biosciences, Lincoln, Ne.
Prism, version 5, GraphPad Software Inc, San Diego, Calif.
References
1 Frank N, Geor RJ, Bailey SR, et al. Equine metabolic syndrome. J Vet Intern Med 2010; 24: 467–475.
2 Gordon ME, McKeever KH, Betros CL, et al. Plasma leptin, ghrelin and adiponectin concentrations in young fit racehorses versus mature unfit Standardbreds. Vet J 2007; 173: 91–100.
3 Gordon ME, McKeever KH, Betros CL, et al. Exercise-induced alterations in plasma concentrations of ghrelin, adiponectin, leptin, glucose, insulin, and cortisol in horses. Vet J 2007; 173: 532–540.
4 Kearns CF, McKeever KH, Roegner V, et al. Adiponectin and leptin are related to fat mass in horses. Vet J 2006; 172: 460–465.
5 Gil-Campos M, Canete RR, Gil A. Adiponectin, the missing link in insulin resistance and obesity. Clin Nutr 2004; 23: 963–974.
6 Richards AA, Stephens T, Charlton HK, et al. Adiponectin multimerization is dependent on conserved lysines in the collagenous domain: evidence for regulation of multimerization by alterations in posttranslational modifications. Mol Endocrinol 2006; 20: 1673–1687.
7 Schraw T, Wang ZV, Halberg N, et al. Plasma adiponectin complexes have distinct biochemical characteristics. Endocrinology 2008; 149: 2270–2282.
8 Pajvani UB, Hawkins M, Combs TP, et al. Complex distribution, not absolute amount of adiponectin, correlates with thiazolidinedione-mediated improvement in insulin sensitivity. J Biol Chem 2004; 279: 12152–12162.
9 Hug C, Wang J, Ahmad NS, et al. T-cadherin is a receptor for hexameric and high-molecular-weight forms of Acrp30/adiponectin. Proc Natl Acad Sci US A 2004; 101: 10308–10313.
10 Radjainia M, Wang Y, Mitra AK. Structural polymorphism of oligomeric adiponectin visualized by electron microscopy. J Mol Biol 2008; 381: 419–430.
11 Waki H, Yamauchi T, Kamon J, et al. Impaired multimerization of human adiponectin mutants associated with diabetes. Molecular structure and multimer formation of adiponectin. J Biol Chem 2003; 278: 40352–40363.
12 Araki S, Dobashi K, Kubo K, et al. High molecular weight, rather than total, adiponectin levels better reflect metabolic abnormalities associated with childhood obesity. J Clin Endocrinol Metab 2006; 91: 5113–5116.
13 Wang AY, Hickman IJ, Richards AA, et al. High molecular weight adiponectin correlates with insulin sensitivity in patients with hepatitis C genotype 3, but not genotype 1 infection. Am J Gastroenterol 2005; 100: 2717–2723.
14 Almeda-Valdes P, Cuevas-Ramos D, Mehta R, et al. Total and high molecular weight adiponectin have similar utility for the identification of insulin resistance. Cardiovasc Diabetol 2010; 9: 26.
15 Liu Y, Retnakaran R, Hanley A, et al. Total and high molecular weight but not trimeric or hexameric forms of adiponectin correlate with markers of the metabolic syndrome and liver injury in Thai subjects. J Clin Endocrinol Metab 2007; 92: 4313–4318.
16 Bobbert T, Rochlitz H, Wegewitz U, et al. Changes of adiponectin oligomer composition by moderate weight reduction. Diabetes 2005; 54: 2712–2719.
17 Hara K, Horikoshi M, Yamauchi T, et al. Measurement of the high-molecular weight form of adiponectin in plasma is useful for the prediction of insulin resistance and metabolic syndrome. Diabetes Care 2006; 29: 1357–1362.
18 Inoue T, Kotooka N, Morooka T, et al. High molecular weight adiponectin as a predictor of long-term clinical outcome in patients with coronary artery disease. Am J Cardiol 2007; 100: 569–574.
19 Kobayashi H, Ouchi N, Kihara S, et al. Selective suppression of endothelial cell apoptosis by the high molecular weight form of adiponectin. Circ Res 2004; 94: e27–e31.
20 Salani B, Briatore L, Andraghetti G, et al. High-molecular weight adiponectin isoforms increase after biliopancreatic diversion in obese subjects. Obesity (Silver Spring) 2006; 14: 1511–1514.
21 Swarbrick MM, Austrheim-Smith IT, Stanhope KL, et al. Circulating concentrations of high-molecular-weight adiponectin are increased following Roux-en-Y gastric bypass surgery. Diabetologia 2006; 49: 2552–2558.
22 Torigoe M, Matsui H, Ogawa Y, et al. Impact of the high-molecular-weight form of adiponectin on endothelial function in healthy young men. Clin Endocrinol (Oxf) 2007; 67: 276–281.
23 Wang ZV, Schraw TD, Kim JY, et al. Secretion of the adipocyte-specific secretory protein adiponectin critically depends on thiol-mediated protein retention. Mol Cell Biol 2007; 27: 3716–3731.
24 Aso Y, Yamamoto R, Wakabayashi S, et al. Comparison of serum high-molecular weight (HMW) adiponectin with total adiponectin concentrations in type 2 diabetic patients with coronary artery disease using a novel enzyme-linked immunosorbent assay to detect HMW adiponectin. Diabetes 2006; 55: 1954–1960.
25 Basu R, Pajvani UB, Rizza RA, et al. Selective downregulation of the high molecular weight form of adiponectin in hyperinsulinemia and in type 2 diabetes: differential regulation from nondiabetic subjects. Diabetes 2007; 56: 2174–2177.
26 Halperin F, Beckman JA, Patti ME, et al. The role of total and high-molecular-weight complex of adiponectin in vascular function in offspring whose parents both had type 2 diabetes. Diabetologia 2005; 48: 2147–2154.
27 Hara K, Horikoshi M, Yamauchi T, et al. Measurement of the high-molecular weight form of adiponectin in plasma is useful for the prediction of insulin resistance and metabolic syndrome. Diabetes Care 2006; 29: 1357–1362.
28 Lara-Castro C, Doud EC, Tapia PC, et al. Adiponectin multimers and metabolic syndrome traits: relative adiponectin resistance in African Americans. Obesity (Silver Spring) 2008; 16: 2616–2623.
29 Lara-Castro C, Luo N, Wallace P, et al. Adiponectin multimeric complexes and the metabolic syndrome trait cluster. Diabetes 2006; 55: 249–259.
30 Murdolo G, Nowotny B, Celi F, et al. Inflammatory adipokines, high molecular weight adiponectin, and insulin resistance: a population-based survey in prepubertal schoolchildren. PLoS One 2011; 6:e17264.
31 Ebinuma H, Miyazaki O, Yago H, et al. A novel ELISA system for selective measurement of human adiponectin multimers by using proteases. Clin Chim Acta 2006; 372: 47–53.
32 Nakano Y, Tajima S, Yoshimi A, et al. A novel enzyme-linked immunosorbent assay specific for high-molecular-weight adiponectin. J Lipid Res 2006; 47: 1572–1582.
33 Sinha MK, Songer T, Xiao Q, et al. Analytical validation and biological evaluation of a high molecular-weight adiponectin ELISA. Clin Chem 2007; 53: 2144–2151.
34 Gordon ME, McKeever KH. Diurnal variation of ghrelin, leptin, and adiponectin in Standardbred mares. J Anim Sci 2005; 83: 2365–2371.
35 Gordon ME, McKeever KH. Oral and intravenous carbohydrate challenges decrease active ghrelin concentrations and alter hormones related to control of energy metabolism in horses. J Anim Sci 2006; 84: 1682–1690.
36 Carter RA, Geor RJ, Burton Staniar W, et al. Apparent adiposity assessed by standardised scoring systems and morphometric measurements in horses and ponies. Vet J 2009; 179: 204–210.
37 Henneke DR, Potter GD, Kreider JL, et al. Relationship between condition score, physical measurements and body fat percentage in mares. Equine Vet J 1983; 15: 371–372.
38 Frank N. Equine metabolic syndrome. Vet Clin North Am Equine Pract 2011; 27: 73–92.
39 Freestone JF, Wolfsheimer KJ, Kamerling SG, et al. Exercise induced hormonal and metabolic changes in Thoroughbred horses: effects of conditioning and acepromazine. Equine Vet J 1991; 23: 219–223.
40 Wang Y, Lam KS, Yau MH, et al. Post-translational modifications of adiponectin: mechanisms and functional implications. Biochem J 2008; 409: 623–633.
41 Radin MJ, Sharkey LC, Holycross BJ. Adipokines: a review of biological and analytical principles and an update in dogs, cats, and horses. Vet Clin Pathol 2009; 38: 136–156.
42 Kusminski CM, Scherer PE. The road from discovery to clinic: adiponectin as a biomarker of metabolic status. Clin Pharmacol Ther 2009; 86: 592–595.
43 Murdolo G, Hammarstedt A, Schmelz M, et al. Acute hyperinsulinemia differentially regulates interstitial and circulating adiponectin oligomeric pattern in lean and insulin-resistant, obese individuals. J Clin Endocrinol Metab 2009; 94: 4508–4516.
44 Bogan JS, Lodish HF. Two compartments for insulin-stimulated exocytosis in 3t3-L1 adipocytes defined by endogenous Acrp30 and Glut4. J Cell Biol 1999; 146: 609–620.
45 Shetty S, Kusminski CM, Scherer PE. Adiponectin in health and disease: evaluation of adiponectin-targeted drug development strategies. Trends Pharmacol Sci 2009; 30: 234–239.
46 Kubota N, Terauchi Y, Yamauchi T, et al. Disruption of adiponectin causes insulin resistance and neointimal formation. J Biol Chem 2002; 277: 25863–25866.
47 Maeda N, Shimomura I, Kishida K, et al. Diet-induced insulin resistance in mice lacking adiponectiNACRP30. Nat Med 2002; 8: 731–737.
48 Berg AH, Combs TP, Scherer PE. ACRP30/adiponectin: an adipokine regulating glucose and lipid metabolism. Trends Endocrinol Metab 2002; 13: 84–89.
49 Wang Y, Lam KS, Chan L, et al. Post-translational modifications of the four conserved lysine residues within the collagenous domain of adiponectin are required for the formation of its high molecular weight oligomeric complex. J Biol Chem 2006; 281: 16391–16400.
50 Wang Y, Xu A, Knight C, et al. Hydroxylation and glycosylation of the four conserved lysine residues in the collagenous domain of adiponectin. Potential role in the modulation of its insulin-sensitizing activity. J Biol Chem 2002; 277: 19521–19529.
51 Empana JP. Adiponectin isoforms and cardiovascular disease: the epidemiological evidence has just begun. Eur Heart J 2008; 29: 1221–1223.
52 Leth H, Andersen KK, Frystyk J, et al. Elevated levels of high-molecular-weight adiponectin in type 1 diabetes. J Clin Endocrinol Metab 2008; 93: 3186–3191.
53 Cook JR, Semple RK. Hypoadiponectinemia—cause or consequence of human “insulin resistance”? J Clin Endocrinol Metab 2010; 95: 1544–1554.
54 Pajvani UB, Du X, Combs TP, et al. Structure-function studies of the adipocyte-secreted hormone Acrp30/adiponectin. Implications for metabolic regulation and bioactivity. J Biol Chem 2003; 278: 9073–9085.
55 Sun L, Hu FB, Yu Z, et al. Lean body mass, interleukin 18, and metabolic syndrome in apparently healthy Chinese. PLoS One 2011; 6:e18104.
56 Yu D, Yu Z, Sun Q, et al. Effects of body fat on the associations of high-molecular-weight adiponectin, leptin and soluble leptin receptor with metabolic syndrome in Chinese. PLoS One 2011; 6:e16818.
57 Combs TP, Berg AH, Rajala MW, et al. Sexual differentiation, pregnancy, calorie restriction, and aging affect the adipocyte-specific secretory protein adiponectin. Diabetes 2003; 52: 268–276.
58 Xu A, Chan KW, Hoo RL, et al. Testosterone selectively reduces the high molecular weight form of adiponectin by inhibiting its secretion from adipocytes. J Biol Chem 2005; 280: 18073–18080.
59 Wagner JA, Wright EC, Ennis MM, et al. Utility of adiponectin as a biomarker predictive of glycemic efficacy is demonstrated by collaborative pooling of data from clinical trials conducted by multiple sponsors. Clin Pharmacol Ther 2009; 86: 619–625.
60 Nawrocki AR, Scherer PE. Keynote review: the adipocyte as a drug discovery target. Drug Discov Today 2005; 10: 1219–1230.