| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Division of Endocrinology, Diabetes, and Metabolism, State University of New York at Buffalo and Kaleida Health, Buffalo, New York 14209
Address all correspondence and requests for reprints to: Paresh Dandona, M.D., D.Phil.(Oxon), F.R.C.P., F.A.C.P., F.A.C.C., Director, Diabetes-Endocrinology Center of Western New York, Chief of Endocrinology, State University of New York at Buffalo, 3 Gates Circle, Buffalo, New York 14209. E-mail: pdandona{at}kaleidahealth.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Forty obese subjects [body mass index (BMI), 37.5 ± 4.9 kg/m2] and 40 nonobese healthy subjects (BMI, 22.6 ± 3.4 kg/m2) had their plasma MIF, glucose, insulin, free fatty acids (FFAs) and C-reactive protein (CRP) concentrations measured. Sixteen obese patients and 16 nonobese healthy subjects had RNA prepared from MNCs. Eight obese subjects with normal glucose concentration were treated with metformin 1 g (Glucophage XR; 1000 mg twice daily) twice daily for 6 wk. Eight obese subjects were used as controls. Plasma concentration of glucose, insulin, FFAs, and MIF was measured by appropriate assays. mRNA for MIF was measured by real-time PCR.
Forty obese subjects had a fasting concentration of MIF of 2.8 ± 2.0 ng/ml, whereas 40 nonobese subjects had a fasting MIF concentration of 1.2 ± 0.6 ng/ml (P < 0.001). Plasma MIF concentrations were significantly related to BMI (r = 0.52; P < 0.001). mRNA for MIF was correlated to plasma FFAs (r = 0.40; P < 0.05) and plasma CRP (r = 0.42; P < 0.05) concentrations. Eight obese subjects had their fasting blood samples taken before and after taking a slow-release preparation of metformin at 1, 2, 4, and 6 wk. The mean plasma concentration fell from 2.3 ± 1.4 to 1.6 ± 1.2 ng/ml at 6 wk (P < 0.05). Obese subjects not on treatment with metformin showed no change. During the period of treatment with metformin, the body weight did not change and the plasma concentration of glucose, insulin, and FFAs did not alter.
We conclude that: 1) plasma MIF concentrations and MIF mRNA expression in the MNCs are elevated in the obese, consistent with a proinflammatory state in obesity; 2) these increases in MIF are related to BMI, FFA concentrations, and CRP; 3) metformin suppresses plasma MIF concentrations in the obese, suggestive of an antiinflammatory effect of this drug; and 4) this action of metformin may contribute to a potential antiatherogenic effect, which may have implications for the reduced cardiovascular mortality observed with metformin therapy in type 2 diabetes mellitus.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Migration inhibitor factor (MIF) is a proinflammatory cytokine whose plasma concentration increases after endotoxin injection in experimental animals, and thus it participates in the proinflammatory cascade that follows an endotoxin administration (6, 7, 8). The concentration of endotoxin required to stimulate MIF is 10–100 times lower than that required for TNF
secretion (9). It plays an important role in both innate and adaptive immunity. MIF is secreted by macrophages and adrenocorticotrophic cells in the anterior pituitary gland. It is the only cytokine that is stored in the secretory cells and is thus secreted rapidly on stimulation. In addition, it is also secreted by the cells after de novo synthesis in response to a stimulus (6, 7, 8, 10). MIF may also have a role to play in atherogenesis through macrophage/foam cell stimulation in the atherosclerotic plaque (9, 11, 12). We also recently demonstrated that thiazolidinediones, which are insulin sensitizers, exert a comprehensive suppressive effect on reactive oxygen generation and inflammatory mechanism (13, 14, 15). Because metformin is also an insulin sensitizer and because it reduces cardiovascular mortality and morbidity, it is possible that metformin may also have an antiinflammatory effect that may impede the progression of atherosclerosis and may thus provide one potential mechanism underlying the beneficial effect of metformin in reducing cardiovascular morbidity and mortality.
In view of the relationship of obesity, inflammation, and atherosclerosis, on the one hand, and the role of MIF in inflammation on the other hand, we hypothesized that obesity is associated with an increase in plasma MIF concentration and that metformin, an insulin sensitizer, causes a reduction in MIF. It is noteworthy that MIF is expressed in and secreted by adipose tissue and thus would be expected to be elevated in obesity in a manner similar to TNF and IL-6 (16, 17). MIF has also been shown to be increased in patients with type 2 diabetes (18, 19). However, there is hitherto no study on MIF in obesity, nor are there any data on a potential effect of metformin on inflammation.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Forty obese subjects [19 males and 21 females; age 47 ± 11 yr; body mass index (BMI), 37.5 ± 4.9 kg/m2] and 40 nonobese healthy subjects (26 males and 14 females; age 37 ± 8 yr; BMI 22.6 ± 3.4 kg/m2) participated in the study (Table 1
). mRNA was isolated from only 16 obese (BMI, 37.7 ± 5.0 kg/m2) and 16 nonobese healthy subjects (BMI, 23.8 ± 1.9 kg/m2). Eight obese subjects (three males and five females; BMI, 36.8 ± 3.6 kg/m2) had their fasting blood samples taken before and after taking a slow-release preparation of metformin (Glucophage XR) 1000 mg twice daily for 6 wk at 1, 2, 4, 6, and 12 wk. Subjects were advised to continue their usual eating and exercise habits and other concomitant medications. Eight obese subjects were used as controls (six females and two males; BMI, 35.5 ± 3.8 kg/m2). They had blood samples drawn at 0, 1, 2, 4, 6, and 12 wk. They were given no metformin. The patients’ body weight did not change during the period of the study in both groups. The blood samples were centrifuged, and plasma was separated and frozen at –70 C. All patients gave their written, informed consent, and the study protocol was approved by the Institutional Review Board of the State University of New York at Buffalo.
|
Plasma insulin concentrations were determined using an ELISA kit from Diagnostic Systems Laboratories Inc. (Webster, TX). The sensitivity of this assay is 0.26 µU/ml, whereas the intraassay precision is 2.6% and interassay precision is 5.8%. Plasma glucose levels were measured in plasma by YSI 2300 STAT Plus glucose analyzer (Yellow Springs Instruments, Yellow Springs, OH). Plasma FFA concentrations were measured by a colorimetric assay (Roche Applied Science, Indianapolis, IN). Fasting insulin and HOMA-IR were used as surrogate markers of insulin resistance. HOMA-IR was calculated using the formula (20, 21): HOMA-IR = fasting insulin (microunits per milliliter) x fasting glucose (micromoles per liter)/22.5.
Plasma MIF and C-reactive protein (CRP) measurements
Plasma MIF was assayed with the DuoSet ELISA development system for human MIF (R&D Systems, Minneapolis, MN). The DuoSet for MIF has been evaluated in our laboratory. The intraassay precision for this assay is 3.5%, and the interassay precision is 12%. The calculated sensitivity of this assay is 5 pg/ml, and the lowest standard concentration measured in this assay was 30 pg/ml. The mean for normal range for healthy nonobese subjects was 1.2 ± 0.6 ng/ml (range, 0–2.3 ng/ml). There was no cross-reactivity with seven recombinant proteins tested (macrophage chemoattractant protein-1, TNF, matrix metalloproteinase-9, IL-4, interferon-
, tissue factor, and soluble intercellular adhesion molecule-1). All samples were assayed together in one large batch. Plasm CRP ELISA kit was purchased from Diagnostic Systems Laboratories Inc.
Total RNA isolation and real time RT-PCR
Total RNA was isolated using the acid guanidinium thiocyanate-phenol-chloroform method. The quality and quantity of the isolated RNA was determined before using the RNA. One microgram of total RNA was reverse transcribed using Advantage RT-for-PCR kit (BD Biosciences Clontech, Palo Alto, CA). Real-time RT-PCR was done using Smart Cycler (Cepheid, Sunnyvale, CA) in which 2 µl cDNA, 10 µl Sybergreen Master mix (Qiagen, Valencia, CA), and 0.5 µl of 20 µM gene-specific primers were used (MIF sense primer: 5'-CTCTCCGAGCTCACCCAGCAG-3'; antisense primer: 5'-CGCGTTCATGTCGTAATAGTT-3'). The specificity and size of the PCR products were tested by adding a melt curve at the end of the amplifications and running it on a 2% agarose gel. All values were normalized to 18S expression.
Statistical analysis
Statistical analysis was carried out using SigmaStat software (version 2.03, Jandel Scientific, San Rafael, CA). All data are expressed as mean ± SE. Statistical analysis was carried out using unpaired t test between nonobese controls and obese subjects. Correlation analysis was performed using Spearman rank order correlation among BMI, HOMA, and the proinflammatory markers. Analysis was carried out with one-factor ANOVA for the repeated measures using Dunnett’s test for comparisons against the baseline for normally distributed data. Dunn’s test was used for the nonparametric data. Two-factor ANOVA was used to evaluate the interaction between treatment (metformin vs. control) and time.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
There was no significant difference in fasting plasma glucose concentrations between the obese and lean subjects (Table 1
). Plasma insulin concentrations were significantly higher in the obese as was HOMA-IR when compared with the lean subjects (Table 1). Glucose concentrations did not change significantly after metformin treatment (Table 2). Fasting plasma insulin concentrations were elevated at baseline and did not change after metformin treatment. Plasma glucose and insulin concentrations did not change in controls. HOMA-IR, as a surrogate marker of insulin resistance, did not alter significantly after metformin treatment for 6 wk (Table 2). Plasma FFA concentrations were elevated significantly (P < 0.005) in the obese (0.50 ± 0.20 mM) when compared with the lean subjects (0.26 ± 0.10 mM) and were related significantly to BMI (r = 0.58; P < 0.005) but not to HOMA-IR (r = 0.31; P = 0.12). Plasma FFA concentration did not change significantly after metformin intake (Table 2). Similarly, plasma CRP concentrations were elevated significantly (P < 0.005) in the obese (3.0 ± 0.21 µg/ml) when compared with the lean subjects (0.7 ± 0.17 µg/ml). Plasma CRP concentration was significantly related to BMI (r = 0.81; P < 0.0001), HOMA-IR (r = 0.57; P < 0.005), and FFA concentration (r = 0.42; P < 0.05). Plasma CRP did not change significantly after metformin treatment for 6 wk.
|
Plasma MIF concentration in obese patients was 2.8 ± 2.0 ng/ml. This was significantly greater than that observed in nonobese healthy subjects (1.2 ± 0.6 ng/ml, P < 0.001; Fig. 1). There was a highly significant correlation between MIF concentrations and BMI (r = 0.52, P < 0.001; Fig. 2) and to HOMA-IR (r = 0.46, P < 0.05). There was a trend to increased plasma MIF concentration (P = 0.065) in males (2.2 ± 0.27 ng/ml) when compared with females (1.8 ± 0.31 ng/ml). MIF concentration in lean males (1.4 ± 0.17 ng/ml) was higher than the lean females but not significantly (1.0 ± 0.11 ng/ml; P = 0.16). Similarly, obese male MIF concentration (3.4 ± 0.54 ng/ml) was not significantly higher than obese females (2.4 ± 0.51 ng/ml; P = 0.12).
|
|
MIF mRNA expression in MNCs was significantly greater by 60% (P < 0.05) in the obese when compared with nonobese healthy subjects (Fig. 3). MIF mRNA was significantly related to plasma FFA concentrations (r = 0.40; P < 0.05) and BMI (r = 0.59; P < 0.001; Fig. 4) but not to plasma MIF concentrations (P = 0.513) or HOMA-IR index (P = 0.271).
|
|
Plasma glucose, insulin, HOMA-IR, and FFA did not change after metformin treatment. Metformin treatment resulted in a significant fall in MIF when compared with controls (Fig. 5; P < 0.01; two-factor ANOVA) and the baseline. The mean plasma concentration fell from 2.3 ± 1.4 to 1.6 ± 1.2 ng/ml at 6 wk (P < 0.05). After the withdrawal of the drug for 6 wk, MIF increased back toward the baseline. Control subjects not on treatment with metformin showed no change.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
, IL-6, CRP; increased level of lipid peroxidation; and oxidative damage to proteins in plasma (1, 2, 3, 22). An increase in plasma MIF concentrations may thus contribute to the accelerated atherosclerotic process in the obese (9, 11, 12). The increase in MIF mRNA expression in MNCs also demonstrate the inflammatory activity of this cell in the obese. This is of interest because MNC fraction contains monocytes that in the arterial wall are known to become macrophages and foam cells to form atherosclerotic plaques. Because proinflammatory mechanisms may contribute to the pathogenesis of insulin resistance through interference with insulin signaling, it is possible that increased concentrations of MIF may contribute to insulin resistance. Although plasma MIF concentration was not related to HOMA-IR index, it was related to plasma FFAs, which are known to induce both inflammation (23) and insulin resistance (24). The second important observation in our study is the reduction in MIF after treatment with metformin in obese patients, whereas it did not change in controls. The fall in MIF after metformin was significant when compared with the baseline and untreated controls. Metformin is an antidiabetic drug that reduces hepatic glucose production and enhances insulin sensitivity. In the U.K. Prospective Diabetes Study, it was also shown to reduce cardiovascular complications (25). A recent retrospective analysis on the use of oral hypoglycemic drugs from Saskatchewan has revealed that metformin usage is associated with a reduction in all-cause and cardiovascular mortality by approximately 40% (26). Because cardiovascular complications in diabetes are secondary to atherosclerosis and atherosclerosis is an inflammatory condition, it is of interest that metformin suppresses MIF, a key mediator of innate and adaptive immune mechanisms, especially those mediated by the monocyte/macrophage. The macrophage loaded with oxidized low-density lipoprotein forms the foam cell; these cells collectively form the fatty streak. Lesions with abundant foam cells and a thin fibrous cap are the ones likely to rupture and trigger thrombosis related to proinflammatory effects (27, 28, 29). MIF is a secretory product of both the macrophage and its stimulator after secretion; this is suggestive of an autocrine and paracrine relationship and may therefore be responsible for maintaining the activity of foam cells in the atherosclerotic plaque. MIF may also intensify and prolong inflammation by inhibiting apoptosis of foam cells. The antiinflammatory effect may also contribute to the all-cause mortality reduction described in the Saskatchewan study (26).
Another potentially intriguing role for MIF is the fact that it stimulates the secretion of insulin from ß-cells in pancreatic islets (30). Whether this effect contributes to hyperinsulinemia in proinflammatory insulin-resistant states like obesity and type 2 diabetes is an issue that requires further investigation. MIF also functions as an enzyme that reduces sulfhydryl linkages and breaks them (9). This action could potentially reduce the biological activity of insulin and the efficiency of the insulin receptor, which too has sulfhydryl linkages. This would contribute to insulin resistance and increased requirements of insulin secretion. Thus, metformin-induced reduction in MIF may potentially contribute to diminution in insulin breakdown.
We recently demonstrated that an increase in FFA concentration to a level comparable with that in the obese induced by triglyceride (Liposyn; Abbott Laboratories, North Chicago, IL) and heparin infusion in normal subjects results in an acute proinflammatory response including an increase in MIF (23). Thus, it is of interest that plasma FFA concentrations were significantly related to both MNC mRNA for MIF and plasma MIF concentration. Because obesity is associated with an increased FFA concentration, the increased FFA concentrations may potentially contribute to the elevation of MIF at the cellular and plasma level. Metformin induced a reduction in MIF without a concomitant change in FFA concentrations. It therefore follows that the fall in MIF after metformin observed by us was independent of any change in FFA concentration during the 6-wk course of treatment with metformin. It is possible that metformin may exert a direct effect on the monocyte-macrophage. One previous study showed a fall in FFA concentrations in the obese treated with metformin over a long period. However, in this study, the subjects had been prescribed dietary restriction and had weight loss of 6.5% of their body weight (31). In our study, there was no change in body weight and metformin treatment was over a short period. This may account for the lack of an effect of metformin on FFAs in our study.
One weakness of our study was that the controls were not given a placebo. Because none of our end points were clinical other than body weight, it is unlikely that the lack of placebo would affect our results. The laboratory staff had no knowledge of the study for which the assays were being carried out. Furthermore, in the group treated with metformin, the suppressed concentrations of MIF at 6 wk reversed to baseline 6 wk after the withdrawal of the drug.
We formed the following conclusions: 1) obesity is associated with an increase in plasma MIF concentration and MIF mRNA expression in MNCs; 2) BMI and MIF concentrations are significantly related to CRP and each other; 3) BMI and FFA concentrations are related to mRNA for MIF in MNCs; and 4) metformin treatment in the obese causes a reduction in plasma MIF independently of a change in plasma FFA concentrations. These novel observations further establish the fact that obesity is a proinflammatory state and that metformin may have a suppressive effect on inflammatory and potential atherosclerotic processes.
|
Footnotes |
|---|
Received March 3, 2004.
Accepted July 9, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
in sera of obese patients: fall with weight loss. J Clin Endocrinol Metab 83:2907–2910
B and stimulation of inhibitor B by troglitazone: evidence for an anti-inflammatory effect and a potential antiatherosclerotic effect in the obese. J Clin Endocrinol Metab 86:1306–1312. Biochem Biophys Res Commun 235:94–98[CrossRef][Medline]
and insulin sensitivity in elderly men with non-insulin-dependent diabetes mellitus. Arterioscler Thromb Vasc Biol 18:1199–1202This article has been cited by other articles:
 |
|
 |
|
  A. Fleischman, S. E. Shoelson, R. Bernier, and A. B. Goldfine Salsalate Improves Glycemia and Inflammatory Parameters in Obese Young Adults Diabetes Care, February 1, 2008; 31(2): 289 - 294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Heutling, H. Schulz, I. Nickel, J. Kleinstein, P. Kaltwasser, S. Westphal, F. Mittermayer, M. Wolzt, K. Krzyzanowska, H. Randeva, et al. Asymmetrical Dimethylarginine, Inflammatory and Metabolic Parameters in Women with Polycystic Ovary Syndrome before and after Metformin Treatment J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 82 - 90. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Orio, F. Manguso, S. Di Biase, A. Falbo, F. Giallauria, D. Labella, A. Tolino, G. Lombardi, A. Colao, and S. Palomba Metformin administration improves leukocyte count in women with polycystic ovary syndrome: a 6-month prospective study Eur. J. Endocrinol., July 1, 2007; 157(1): 69 - 73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Zicari, C. Ticconi, F. Ietta, A. Belmonte, N. Bechi, M. Realacci, M. Di Vito, F. Arcuri, M. Russo, E. Piccione, et al. Macrophage Migration Inhibitory Factor-Nitric Oxide Interaction in Human Fetal Membranes at Term Pregnancy Reproductive Sciences, May 1, 2006; 13(4): 263 - 270. [Abstract] [PDF]
|
|
|
|
 |
|
 K. Isoda, J. L. Young, A. Zirlik, L. A. MacFarlane, N. Tsuboi, N. Gerdes, U. Schonbeck, and P. Libby Metformin Inhibits Proinflammatory Responses and Nuclear Factor-{kappa}B in Human Vascular Wall Cells Arterioscler. Thromb. Vasc. Biol., March 1, 2006; 26(3): 611 - 617. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Herder, H. Kolb, W. Koenig, B. Haastert, S. Muller-Scholze, W. Rathmann, R. Holle, B. Thorand, and H.-E. Wichmann Association of Systemic Concentrations of Macrophage Migration Inhibitory Factor With Impaired Glucose Tolerance and Type 2 Diabetes: Results from the Cooperative Health Research in the Region of Augsburg, Survey 4 (KORA S4) Diabetes Care, February 1, 2006; 29(2): 368 - 371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Modugno, R. B. Ness, C. Chen, and N. S. Weiss Inflammation and Endometrial Cancer: A Hypothesis Cancer Epidemiol. Biomarkers Prev., December 1, 2005; 14(12): 2840 - 2847. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Bruun, A. S. Lihn, S. B. Pedersen, and B. Richelsen Monocyte Chemoattractant Protein-1 Release Is Higher in Visceral than Subcutaneous Human Adipose Tissue (AT): Implication of Macrophages Resident in the AT J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2282 - 2289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Dandona, A. Aljada, A. Chaudhuri, P. Mohanty, and R. Garg Metabolic Syndrome: A Comprehensive Perspective Based on Interactions Between Obesity, Diabetes, and Inflammation Circulation, March 22, 2005; 111(11): 1448 - 1454. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Trayhurn Adipose Tissue in Obesity--An Inflammatory Issue Endocrinology, March 1, 2005; 146(3): 1003 - 1005. [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 |
