| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
The Medical Research Laboratories (K.K.A., J.F., A.F.), Clinical Institute and Medical Department M (Diabetes and Endocrinology), Aarhus University Hospital, DK-8000 Aarhus C, Denmark; Childrens Clinic Randers (O.D.W., C.H.), DK-8900 Randers, Denmark; and Department of Pediatrics (C.H.), Skejby Sygehus, Aarhus University Hospital, DK-8200 Skejby, Denmark
Address all correspondence and requests for reprints to: Professor Allan Flyvbjerg, M.D., The Medical Research Laboratories, Clinical Institute, Medical Department M (Diabetes and Endocrinology), Aarhus University Hospital, DK-8000 Aarhus C, Denmark. E-mail: allan.flyvbjerg{at}dadlnet.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: The goal of this study was to investigate whether pubertal stages and testosterone levels influenced total serum ADPN levels and the distribution of ADPN isomers.
Design: This is a cross-sectional study.
Patients: The study included 859 children and adolescents (396 males) aged 6–20 yr.
Main Outcome Measures: Total ADPN and ADPN isomers were measured using a validated in-house immunofluorometric assay. Fractioning of the ADPN into the three major molecular fractions was performed in representative subgroups of pre- and postpubertal males and females (n = 40, 10 in each group) using a validated fast protein liquid chromatography method.
Results: Total ADPN levels before puberty were 13.4 (11.1–15.9) mg/liter (median and interquartile range) and 14.7 (12.3–18.1) mg/liter (P = not significant), in males and females, respectively. After puberty, ADPN levels were significantly reduced in males, 9.7 (8.2–12.0) mg/liter but remained unchanged in females, 12.1 (9.7–15.3) mg/liter (P < 0.0001). Concomitantly, a reduction was seen in the ratio of high-molecular-weight (HMW) isomers to total ADPN (HMW ratio) when comparing prepubertal and postpubertal males. Also, postpubertal males had lower HMW ratios than corresponding females (P = 0.038). Finally, a negative correlation was seen between HMW ratio and testosterone (r = –0.430, P = 0.007).
Conclusion: Serum total ADPN levels decrease through puberty in males. Also, a reduced HMW ratio is seen in males at the onset of puberty. We speculate that the suppression of HMW ADPN may be caused by testosterone.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
An increasing number of children experience health problems related to obesity and diabetes, and low ADPN levels have been shown to predict type 2 diabetes. Only a few studies have investigated circulating levels of total ADPN during childhood and adolescence. A recent study by Böttner et al. (4) showed a temporal relationship between the decrease in ADPN levels and the progression from childhood to adolescence, which was most pronounced in males. In the same study, several negative correlations between various endocrine parameters (e.g. testosterone) and ADPN were found. This is consistent with results obtained in preclinical studies, in which an inhibitory effect of testosterone on serum ADPN levels was shown in mice (5). Furthermore, in the same study, lower levels of total ADPN were observed in male than female mice.
The aim of the present study was to measure serum total ADPN in a population of normal school children aged 6–20 yr. Furthermore, because the impact of puberty on the distribution of ADPN complexes has not previously been elucidated, we also investigated the complex distribution of serum ADPN in a subpopulation of pre- and postpubertal subjects using a validated in-house size exclusion gel fast protein liquid chromatography (FPLC) method.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
ADPN isomer subgroup
A subset of 40 subjects [10 males and 10 females from Tanner stages (TS) 1 and 5] was used for assessment of ADPN isomers. Subjects were selected so their serum ADPN was near the median of each TS group. None of the individuals were smokers, used medicine (including contraceptives), or had any medical conditions.
ADPN immunoassay
ADPN was measured using a validated in-house time-resolved immunofluorometric assay, as previously described (6).
FPLC fractioning of ADPN and method validation
Materials. A HiLoad 16/60 Superdex 200 prep grade column (GE Healthcare, Amersham Biosciences, Buckinghamshire, UK), a Smartline 1000 pump with a 10-ml titanium pump head, an Autosampler 3800 (Knaur, Berlin, Germany), and a fraction collector CHF122SB (Advantec, Dublin, CA) were used. As running buffer, PBS (50 mM phosphate, 150 mM NaCl, 0.2% vol/vol Tween 20, 0.2% wt/vol BSA, 0.02% wt/vol Na-azide, pH 7.2) was used.
FPLC running condition.
Separation was performed at ambient temperature. A sample size of 500 µl, consisting of 50 µl of serum and 450 µl of running buffer, and a flow of 0.8 ml/min was used. Fractions were collected in one of two ways: either 30 fractions of 1 ml or 17 fractions of 1–3 ml (determined on the basis of prior runs). The fraction sizes were chosen to preserve the discrimination between the three peaks (see Fig. 1
), while allowing a higher sample output. The fractions did not undergo any treatment before being assayed. None of the fractions were kept at a temperature below 4 C and all assays were performed less than 36 h after fractioning.
|
Day-to-day variation. On average, every sixth sample was an internal control. This was done to determine day-to-day variation. In each series of samples, controls were placed in random order. The day-to-day CVs of the three different isomers were less than 6, 12, and 7% for the HMV, MMW, and LMW fractions, respectively (n = 7).
Fractions of ADPN to different elution volumes.
The FPLC fractioning yielded three major peaks apparently representing HMW, MMW, and LMW ADPN (peaks 1, 2, and 3; Fig. 1
). Western blotting was performed on the fractions corresponding to the peaks, e.g. eluent volumes 50, 58, and 65 (1 ml each). A nonheated and nonreduced approach was used as described by Waki et al. (2) using a 6% acrylamide gel. Monoclonal mouse anti-ADPN (catalog no. BAM 1065; R & D Research, Abingdon, UK) and goat-antimouse IgG (catalog no. 31430; Pierce, Rockford, IL) were used. The samples were applied to the acrylamide gel immediately after fractioning to minimize any degradation of the proteins. The fractions of the presumed HMW, MMW, and LMW peaks showed different bands on the Western blot (Fig. 1). In the LMW fraction, a wide 70-kDa band was found. Also, a less specific band of 110 kDa was seen. Three bands of 135, 127, and 120 kDa were seen in the MMW fraction, with the latter being the most dense. The bands in the HMW fraction consisted of two bands above 250 kDa and a weaker band of 135 kDa.
Dose response. To determine the linearity of the method, the FPLC column was loaded with 75 and 100 µl of serum, i.e. 150 and 200% of the routine volume, respectively. However, we observed no differences with regards to the distribution of the isomers when compared with the routine sample size of 50 µl (data not shown).
Recovery. The concentrations of total ADPN in the samples loaded onto the column were determined in the same assays as the corresponding fractions. The total amount calculated on the basis of the concentration was compared with the sum of ADPN found in the three fractions, hereby determining the recovery of the method, which averaged 83% (range 73–101%, n = 7 control sera).
Freezing and thawing of serum. To investigate the effects on the different isomers of ADPN of repetitive freeze-thaw cycles, serum that had undergone 10 cycles was subjected to fractioning and compared with serum that had only been frozen and thawed once. Furthermore, serum that had been frozen and thawed once was compared with fresh serum. Serum (n = 5) that had undergone 10 cycles of repetitive freeze-thawing showed no difference in the relative HMW fractions when compared with controls (P = not significant). In the MMW fractions, there was a small but significant difference with means of 30.1 ± 0.6 and 27.2 ± 0.4% for the freeze-thawed and fresh serum, respectively (P < 0.001). Accordingly, in the LMW fractions we found means of 13.0 ± 0.8% in the freeze-thawed and 15.3 ± 0.6% in the fresh serum (P = 0.008). Fresh serum did not differ from serum frozen once (data not shown).
Handling of the fractions. Because the time between FPLC fractioning and the fractions being assayed could vary because of practical considerations, the same fractions were assayed twice with 24 h in between to investigate whether any difference occurred because of this procedural step. Fractions were kept at 4 C in between assays. The relative distribution of HMW and MMW fractions did not change significantly between two assays performed 24 h apart. However, the relative LMW fraction did change significantly (P = 0.019) with relative means of 15.9 ± 2.9 and 15.2 ± 2.7% for the first and second assay, respectively.
Other assays used
Testosterone was determined in the ADPN isomer subgroup using a commercially available kit B050–201 (PerkinElmer, Turku, Finland) according to the manufacturer’s instructions. The CVs were 4.2 and 6.9%, within and in-between assay, respectively. The detection limit was 0.3 nmol/liter.
Calculations
Isomer distribution.
The distribution of HMW, MMW, and LMW ADPN in serum was calculated on the basis of the different concentrations in each fraction. When the ADPN concentration of the fractions was plotted against the elution volume, a graph with three peaks was produced (see Fig. 1). The lowest value between two peaks was included in the highest fraction.
BMI Z-scores. Using data on Danish BMI related to gender and age (7), BMI Z-scores were obtained for the participants.
Statistical analysis
The statistical analysis was performed using SigmaStat (version 3.11; Systat, Point Richmond, CA). Absolute ADPN was found to be nonnormally distributed, and ADPN levels were therefore given as medians ± interquartile ranges. All other data were given as means ± SD. In the statistical analysis, ADPN data were log transformed. Student’s t test was used for parametric data; otherwise Mann-Whitney rank sum test was used. One-way ANOVA was used to compare multiple groups. Spearman rank order correlation was used to describe the relationship between total ADPN and testosterone as the criteria for constant variance was not met. Significance was defined as P 
0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
ADPN and pubertal stages.
Anthropometric data and total ADPN levels for the entire cohort are given in Table 1. Serum ADPN differed significantly with TS in males as well as in females (ANOVA, P < 0.03). In males, serum ADPN was higher in TS 1 than in TS 3, 4, and 5, whereas TS 2 was higher than TS 5. In females, Serum ADPN was higher in TS 1 than in TS 5; otherwise no difference was seen. When males and females were compared within the same TS, no difference was seen in TS 1 and 2, whereas serum total ADPN was higher in females than males in TS 3 to 5 (P < 0.05).
|
0.003; see Fig. 2).
|
ADPN isomers and pubertal stages. As mentioned above, a total of 40 subjects were selected for measurements of ADPN isoforms. The power of this part of the study was 0.78, which is close the desired level of 0.80.
A shift in the distribution of the isomers was seen from the prepubertal TS 1 to the postpubertal stage of TS 5 (Fig. 3). In the male group, the mean HMW ratio before puberty averaged 50 ± 5%, which decreased to 45 ± 3% after puberty (P = 0.023). In the female pre- and postpubertal groups, the mean HMW ratios were 50 ± 4% and 48 ± 3% (ns). Before puberty, there was no difference between the male and female group, but after puberty the HMW ratios of the two groups were different (P = 0.038). The relative fractions of MMW changed accordingly, whereas the relative LMW fraction on large remained constant (data not shown).
|
The ADPN isomer subgroups were representative of the total group with regards to BMI and ADPN except in the male TS 1 subgroup. In this group BMI was found to differ significantly (see Table 2). A negative correlation was found in the whole subgroup between HMW ratio and the serum level of testosterone (r = –0.43, P = 0.007, n = 38; see Fig. 4). Also, total ADPN and testosterone showed a negative correlation (r = –0.473, P = 0.003, Spearman). Two subjects were not included in the analysis because their testosterone levels were undetectable and, thus, could not be log transformed.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our data on total ADPN are consistent with previous reports. Accordingly, we found a decrease in total ADPN through puberty in males similar to what Böttner et al. (4) reported in their cohort of normal-weight children. In that study a close relationship was found between endocrine measures of male androgen status (e.g. testosterone) and total ADPN. As a new observation in the present study, we found a negative relationship between the HMW ratio and testosterone in the subgroup of ADPN isomer subjects. This, we speculate, is further evidence that the decrease in total ADPN is predominantly due to a decrease in the HMW fraction of ADPN. For the ADPN isomer, subgroup subjects were selected around the mean of the respective TS groups to make the subgroups representative. Because the means in the pre- and postpubertal male and female groups differed, we were unable to directly compare the absolute concentrations of the HMW, MMW, and LMW fractions. The decrease in total ADPN levels and the HMW ratio suggests that the pubertal decrease in ADPN in males may be primarily a reduction in the HMW isomers of ADPN.
In the published literature there seems to be an consistent relationship between BMI and total ADPN (12, 13). We used BMI Z-scores to adjust for gender and age because it was a pediatric cohort. Interestingly, we found no relationship between BMI Z-scores and total ADPN in the prepubertal groups, whereas the relationship was seen in postpubertal (TS 5) males and females. This finding could further indicate that the complex endocrine regulation in which adipocytes play a part changes from childhood to adulthood (14).
A decreased secretion of the HMW isoforms from adipocytes in vitro is seen when adipocytes are exposed to male androgens (5). Likewise, in vivo castrated male mice show a markedly elevated HMW fraction when compared with controls (5). To our knowledge, the influence of male androgens, which in the present study was assessed indirectly as pubertal stages and by measurements of serum testosterone levels, on ADPN isomers in children has not been investigated before. The results in the isomer subgroup suggest that the change in androgen status alters the isomer composition during puberty. In the same groups, however, neither the pre- and postpubertal BMI means nor ADPN levels were fully comparable, and this may have influenced the results. The postpubertal groups had higher BMI, which is known to be associated to lower total ADPN levels. The importance of BMI on ADPN isomer distribution deserves further investigation. In the present study, we compared relative isomer distributions in groups with different total ADPN medians. Recent work by Nakano et al. (15) shows that the HMW ratio is positively associated to total ADPN levels in healthy adults. The median ADPN levels in the isomer subgroups were very similar to each other when compared with the whole cohort. They were selected to have ADPN serum levels close to the median of their respective TS groups. Therefore, it is likely that the difference in the medians between the groups was of less or no importance.
The FPLC method is laborious and has a limited capacity. Therefore, it has only been used to fraction ADPN isomers in a few studies and in small numbers of subjects. On the other hand, all ADPN isomers are visualized and the technique enables a quantitative measurement. Importantly, there is a good agreement between the elution profile and the Western blot of the different peaks (see Fig. 1).
Several other methods of fractioning the isomers of ADPN have been used. Western blotting performed under nonreducing and nonheating conditions was elegantly shown by Waki et al. (2) to effectively separate the various ADPN isomers. Later on, others have used the same technique, and by densitometry measurements calculated a HMW ratio on a semiquantitative basis (16, 17). However, the resulting HMW ratios varied considerably between the different studies, ranging from 6.2–62%. Velocity gradients have been used by several researchers (3, 18). Very recently, sandwich assays have emerged using proteases to separate the isoforms. Although some of the isomer fractions were calculated indirectly, it seems to be a valid way of determining the HMW ratio. Compared with size exclusion gel chromatography, it still requires more handling of the samples in the form of enzymatic digestion. The only handling of the samples using the FPLC method occurs when they are loaded onto the column. Another method includes sandwich assays based on monoclonal antibodies (MABs) that have been raised specifically against the HMW isomers of ADPN in combination with MABs with affinity for all isomers (15). This assay gives high HMW ratios, and it has to be noted that differences in MABs make comparison of results between methods difficult. Although the high affinity for the HMW fraction is an advantage, it is also a drawback when investigating relative distributions of ADPN isomers, because this requires MABs with affinity for all isomers.
In conclusion, total ADPN as well as the HMW ratio are decreased in males after completed pubertal development. The difference in HMW ratio between males before and after puberty is most likely due to testosterone’s suppression of the HMW isomers, but the exact mechanism requires further investigation. The roles of the different ADPN isomers and their relative distributions are only starting to be appreciated. Methods, such as ours, to effectively separate and quantify the isomers enable future investigations of the roles of ADPN isomers.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Statement: K.K.A. and C.H. have nothing to declare. J.F. consults for Hoffmann–La Roche. O.D.W. consults for Chiesi Pharmaceutici SpA and has received lecture fees from Merck Sharp & Dohme. A.F. consults for Hoffmann–La Roche and Taisho and has received lecture fees from GlaxoSmithKline and Novo Nordisk.
First Published Online February 6, 2007
Abbreviations: ADPN, Adiponectin; BMI, body mass index; CI, confidence interval; CV, constant of variance; FPLC, fast protein liquid chromatography; HMW, high molecular weight; LMW, low molecular weight; MABs, monoclonal antibodies; MMW, medium molecular weight; TS, Tanner stage.
Received October 23, 2006.
Accepted January 30, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Kozakova, E. Muscelli, A. Flyvbjerg, J. Frystyk, C. Morizzo, C. Palombo, and E. Ferrannini Adiponectin and Left Ventricular Structure and Function in Healthy Adults J. Clin. Endocrinol. Metab., July 1, 2008; 93(7): 2811 - 2818. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Caminos, R. Nogueiras, F. Gaytan, R. Pineda, C. R. Gonzalez, M. L. Barreiro, J. P. Castano, M. M. Malagon, L. Pinilla, J. Toppari, et al. Novel Expression and Direct Effects of Adiponectin in the Rat Testis Endocrinology, July 1, 2008; 149(7): 3390 - 3402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ibanez, A. Lopez-Bermejo, L. Suarez, M. V. Marcos, M. Diaz, and F. de Zegher Visceral Adiposity without Overweight in Children Born Small for Gestational Age J. Clin. Endocrinol. Metab., June 1, 2008; 93(6): 2079 - 2083. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Peterlin, G. Alexander, D. Tabby, and E. Reichenberger Oligomerization state-dependent elevations of adiponectin in chronic daily headache Neurology, May 13, 2008; 70(20): 1905 - 1911. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
