| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Technical Briefs |
1 Diagnostics Research Laboratories, Daiichi Pure Chemicals Co. Ltd., Ibaraki, Japan; 2 Division of Clinical Preventive Medicine, Department of Community Preventive Medicine, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan;3 Department of Metabolic Diseases, Graduate School of Medicine, University of Tokyo, Tokyo, Japan;
aaddress correspondence to this author at: Division of Clinical Preventive Medicine, Department of Community Preventive Medicine, Niigata University Graduate School of Medical and Dental Sciences, Asahimachi 1-757, Chuo-ku, Niigata, Niigata 951-8510, Japan; fax 81-25-223-0996, e-mail: miida{at}med.niigata-u.ac.jp
Abstract
Background: Human serum adiponectin exists in 3 multimer forms: high molecular weight (HMW), middle molecular weight, and low molecular weight (LMW), with some of the latter bound to albumin (Alb)-LMW. Some studies have suggested that adiponectin crosses the blood–brain barrier and plays a central role in energy homeostasis.
Methods: To determine cerebrospinal fluid (CSF) adiponectin at extremely low concentrations, we modified the protocol of the ELISA system used to assay serum adiponectin. The 3 multimers of adiponectin were measured separately by pretreating CSF with 2 proteases. We measured the CSF adiponectin concentrations in anonymous human samples (n = 19). The molecular sizes of adiponectin in CSF pretreated with proteases or untreated were determined by use of native PAGE and immunoblotting.
Results: The ELISA system measured adiponectin in the range of 1.0–167 µg/L. The between-assay imprecision estimates (CVs) were 6%–17% for the 3 forms. The mean total CSF adiponectin concentration (7.2 µg/L) was 
1/1000 of the mean concentration in serum. Unlike serum adiponectin, the LMW and Alb-LMW forms predominated in all of the CSF samples. Immunoblotting analysis revealed that most LMW forms were bound to Alb, although the HMW form was detected in some samples.
Conclusions: The modified ELISA system measures the 3 multimers separately and is sufficiently sensitive to measure adiponectin in CSF.
Adiponectin is an adipocyte-derived adipokine (1) with multiple functions, including antidiabetic (2), antiatherogenic (3), and antiinflammatory actions. Although the adiponectin target organs are the liver, muscles, and blood vessels, some studies have suggested that adiponectin has central effects on energy homeostasis. The intracerebroventricular administration of adiponectin in normal mice led to dose-dependent decreases in the body weight without substantial inhibition of food intake. Furthermore, intravenous adiponectin injection induced a >3-fold increase in cerebrospinal fluid (CSF) adiponectin concentration (4). In contrast, a study of humans revealed that the adiponectin concentration in CSF made up only 0.1% of the serum adiponectin concentration, suggesting that adiponectin does not cross the blood–brain barrier (5). Hence, controversy remains as to whether adiponectin can cross the blood–brain barrier under physiological conditions.
Adiponectin in human blood exists as 3 multimers with distinct molecular sizes: trimeric low molecular weight (LMW), hexameric middle molecular weight (MMW), and high molecular weight (HMW) forms (6). Some of the LMW adiponectin exists as albumin (Alb)-bound forms (Alb-LMW) (7). The molecular weights of adiponectin affect the strength of their metabolic actions. Several studies have suggested that HMW adiponectin and the ratio of HMW adiponectin to total adiponectin are more closely associated with insulin sensitivity and metabolic syndrome than is total adiponectin (8)(9). ELISAs for measuring the 3 multimers separately (10) are available in Asia (Daiichi Pure Chemicals), the US, and the European Union (ALPCO Diagnostics).
We developed and tested a highly sensitive ELISA system for measuring CSF adiponectin and examined whether HMW adiponectin is detectable in human CSF. To measure extremely low concentrations of adiponectin in CSF, we modified the protocol of the ELISA used to measure serum adiponectin (10). Human CSF was obtained from residual clinical samples that had normal cell counts and CSF concentrations of total protein, Alb, IgG, β2-macroglobulin, glucose, lactate dehydrogenase, and chloride. We excluded samples in which erythrocyte contamination was detected by microscopic examination. After ensuring confidentiality of the identities of the sample donors, their CSF samples were labeled only with age and sex and were sent to our laboratory. To ensure that the samples were not contaminated with erythrocytes, we examined all of the CSF samples after low-speed centrifugation. Finally, we obtained 19 CSF samples [9 males and 10 females; mean (SD) age 51.6 (17.2) years] and stored them in a deep-freezer until use. Informed consent was obtained from all of the study participants, and the procedures were strictly in accordance with the statement of the Japanese Society of Laboratory Medicine on the use of residual clinical samples (11).
This modified ELISA system enabled us to measure the 3 multimers separately by pretreating the samples with 2 proteases. A total of 150 µL of CSF is required to measure the concentrations of total adiponectin and all 3 multimers. To determine total adiponectin, we mixed 50 µL of CSF with 50 µL of pretreatment buffer containing no protease [50 mmol/L Tris-HCl (pH 8.0)]. The mixture was added to 50 µL of sample buffer [100 mmol/L sodium citrate (pH 3.0) containing 20 g/L sodium dodecyl sulfate (SDS)]. For selective MMW + HMW adiponectin determination, 50 µL of CSF was incubated with the pretreatment buffer containing 1.0 g/L protease A Amano (Amano Enzyme) for 20 min at 37 °C. The mixture was then added to 50 µL of the sample buffer. For selective HMW adiponectin determination, we incubated 50 µL of CSF with the pretreatment buffer containing 3.75 kU/L of proteinase K (Roche Diagnostics) for 20 min at 37 °C. The mixture was then added to 50 µL of the sample buffer.
These pretreated samples were further diluted 31-fold with PBS (15 mmol/L, pH 7.5) containing 10 g/L BSA and 0.5 g/L Tween 20 (BSA-PBST). Each well of the polystyrene microtiter plates (Nunc) was coated with 50 µL of antihuman adiponectin monoclonal antibody (No. 64405; 5 mg/L in PBS) (10) and incubated overnight at 4 °C. After PBST rinsing, the wells were blocked with 100 µL of BSA-PBST at room temperature for 2 h. The calibrators (0–1.8 µg/L dimeric adiponectin from human serum) (10) or diluted samples (50 µL each) were placed in the wells and incubated at room temperature for 2 h. After rinsing with PBST, 50 µL of antihuman adiponectin biotinylated monoclonal antibody (No. 64404) (10) was added to each well, and the plate was incubated at room temperature for 1 h. After PBST rinsing, the biotinylated antibody was allowed to react with horseradish peroxidase–conjugated streptavidin (Pierce) at room temperature for 30 min. The trapped adiponectin–antibody complexes were then washed extensively and incubated with substrate solution (o-phenylenediamine in citrate buffer, pH 5.0, containing hydrogen peroxide) at room temperature for 20 min; the absorbance was measured at 492 nm. The concentrations of MMW and LMW (including Alb-LMW) adiponectin were obtained by subtracting the HMW concentration from the MMW + HMW concentrations and the MMW + HMW concentrations from the total adiponectin concentration, respectively. In the modified ELISA, the analytical limit of detection (12) was 1.0 µg/L, and the calibration curve was linear up to 167 µg/L (y = 1.10x + 0.14, r = 0.999). Decreasing the sample dilution and increasing the sample incubation and substrate reaction times rendered this modified system 200 times more sensitive than the original system used to measure serum adiponectin. We confirmed that the curve of measured values of serially diluted CSF samples paralleled the calibration curve (Fig. 1A
). The intraassay CVs (n = 8) were 7.3% and 2.7% at adiponectin concentrations of 7.1 and 17.8 µg/L, respectively. The interassay CVs (n = 4) were 6.3%, 17.2%, and 5.8% for the total adiponectin, MMW + HMW, and HMW assays, respectively, at adiponectin concentrations of 19.0 µg/L.
|
Despite the extremely low CSF-adiponectin concentrations, we measured adiponectin successfully in 19 clinical samples (Table 1
). The mean (SD) total CSF adiponectin concentration was 7.2 (7.2) µg/L, which is 1000 times lower than the serum adiponectin concentration. We also used electrophoresis to quantify the total adiponectin concentrations. The adiponectin multimers were converted into the dimer by heat denaturation and subjected to SDS-PAGE. The dimeric adiponectin bands were visualized by Western blotting with goat antiadiponectin antibody (R&D Systems) and quantified by densitometric intensity, which increased linearly from 1.0 to 25 µg/L. We detected a strong positive correlation (n = 19, r = 0.940) between the total adiponectin values determined by the 2 methods.
|
In all 19 samples, the LMW adiponectin was the dominant form of the 3 multimers (Table 1
). In 16 of the 19 samples, the MMW or HMW concentrations were below the detection limits. In 3 samples, all 3 adiponectin multimers were detected. For these 3 samples, the mean (SD) ratios of HMW, MMW, and LMW (including Alb-LMW) to total adiponectin were significantly different from the reported values for the serum samples (n = 47): 0.18 (0.04) vs 0.39 (0.13), P <0.05; 0.15 (0.01) vs 0.27 (0.05), P <0.005; 0.69 (0.05) vs 0.34 (0.10), P <0.005, respectively, according to the Mann–Whitney U-test (10).
To confirm these results, we separated the same CSF samples using 2%–15% native PAGE followed by Western blotting against adiponectin using goat antiadiponectin antibody (10). The adiponectin multimers in the CSF samples were at the same positions as those of the serum adiponectin multimers (Fig. 1B
). As expected, the CSF samples showed considerably weaker MMW and HMW bands and clearer Alb-LMW (double-stained) bands than those observed in the serum control (7). Such double staining might result from hydroxylation or glycosylation of the Alb-LMW form (13). Because the LMW form cannot be detected clearly in native PAGE analysis, we used nonheating SDS-PAGE analysis to separate the LMW form from CSF (6) and thus confirmed the existence of the LMW form in human CSF (Fig. 1C
).
Finally, we examined whether protease A Amano and proteinase K selectively digest the CSF adiponectin multimers. We used native PAGE and Western blotting to analyze the digested products of CSF (10). Consequently, we obtained HMW and MMW adiponectin after protease A Amano digestion and HMW adiponectin after proteinase K digestion (Fig. 1D
). As with serum samples, these selective digestions enabled us to measure the 3 multimers in CSF samples.
The predominance of the smaller molecular forms in human CSF is supported by a recent report (14). In men, but not in women, the CSF adiponectin concentrations correlated positively with serum adiponectin and negatively with body mass index. Furthermore, adiponectin receptors were detected in the brain (4)(5). These results suggest that the presence of CSF adiponectin did not result from contamination with peripheral blood during CSF sampling.
In conclusion, our modified ELISA system is sufficiently sensitive for measuring CSF adiponectin, and it can measure the 3 multimers. Although LMW (including Alb-LMW) adiponectin was found to be the dominant form in this set of human CSF samples, HMW and MMW adiponectin multimers were also detected in some samples. Additional studies are required to clarify how the adiponectin multimers are transported across the blood–brain barrier and whether they play central roles in energy homeostasis. Our modified ELISA system will serve as a useful tool for future studies using CSF samples.
Acknowledgments
Grant/funding support: None declared.
Financial disclosures: None declared.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  Z. Kovacova, M. Vitkova, M. Kovacikova, E. Klimcakova, M. Bajzova, Z. Hnevkovska, L. Rossmeislova, V. Stich, D. Langin, and J. Polak Secretion of adiponectin multimeric complexes from adipose tissue explants is not modified by very low calorie diet Eur. J. Endocrinol., April 1, 2009; 160(4): 585 - 592. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pini, M. E. Gove, J. A. Sennello, J. W. P. M. van Baal, L. Chan, and G. Fantuzzi Role and Regulation of Adipokines during Zymosan-Induced Peritoneal Inflammation in Mice Endocrinology, August 1, 2008; 149(8): 4080 - 4085. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Ahima and M. A. Lazar Adipokines and the Peripheral and Neural Control of Energy Balance Mol. Endocrinol., May 1, 2008; 22(5): 1023 - 1031. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
