Macrophages regulate tumor necrosis factor-α expression in adipocytes through the secretion of matrix metalloproteinase-3

Abstract

Objective:

Adipocytes accumulated in the visceral area change their function to induce tumor necrosis factor-α (TNF-α) secretion with concomitant matrix metalloproteinase (MMP)-3 induction in mice. This study was performed to clarify the role of macrophages (Mφ)-secreted MMP on the functional changes in adipocytes using a culture system.

Design:

Cultures of 3T3-L1 adipocytes with THP-1 Mφ or the Mφ-conditioned medium were used to investigate the role of Mφ-MMP on the TNF-α gene in 3T3-L1 adipocytes by the addition of MMP inhibitors. For animal experiments, male C57BL/6J mice were rendered insulin resistant by feeding a high-fat diet, and the expression of an Mφ marker F4/80, and MMP-3 genes in mesenteric and subcutaneous fat tissue specimens were examined.

Results:

Mφ-conditioned media (Mφ-CM) increased the levels of TNF-α mRNA expression in 3T3-L1 adipocytes, and these adipocyte responses were abolished by treatment with GM6001, a broad-spectrum MMP inhibitor, or NNGH (N-isobutyl-N-(4-methoxyphenylsulfonyl)-glycylhydroxamic acid), an MMP-3 inhibitor. The activated form of MMP-3 enhanced glycerol release as well as TNF-α protein secretion from 3T3-L1 adipocytes. The incubation of adipocytes with MMP-3 inhibited insulin-induced glucose uptake in adipocytes. Furthermore, a high-fat intake increased the expression of MMP-3, decreased the insulin-induced glucose uptake of adipocytes and induced expression of F4/80 in mesenteric fat tissue of C57BL/6 mice.

Conclusion:

Mφ may cause a pathological link with surrounding adipocytes through the secretion of MMP-3 followed by TNF-α expression in adipocytes in visceral fat tissue.

Introduction

Disturbed insulin sensitivity plays an important role in the accumulation of various metabolic disorders, and has been recognized as ‘metabolic syndrome’.1, 2 In accordance with the clinical significance of evaluation of visceral fat accumulation in metabolic syndrome, it has become evident that visceral fat has direct interaction with other tissues, such as muscles, liver or vessel walls, through the secretion of several molecules regulating the insulin sensitivity in tissues.3, 4 The transplantation of cultured cells into the intramesenteric space of mature mice has been established as an adequate mode for the analyses of the interaction between visceral fat and insulin sensitivity.5 The mice with transplanted cultured adipocytes showed that visceral fat, and not subcutaneous fat, secretes the tumor necrosis factor-α (TNF-α), and the secreted molecules actually disturb the insulin sensitivity based on the decreased insulin action in tissues.5 The accumulated visceral fat caused drastic changes in expression of matrix metalloproteinase (MMP) family genes, among which MMP-3 potentiated free fatty acid-induced TNF-α secretion from adipocytes.6 Therefore, the MMP-3 activity in visceral fat seems to be directly linked to cytokine expression in adipocytes.

There is an infiltration of macrophages (Mφ) in the accumulated fat tissues, and active Mφ cause a pathological inter-relationship with surrounding adipocytes in visceral fat, which leads to the progression of insulin resistance.7 A variety of inflammatory bioactive molecules plays an important role in pathological interaction between Mφ and adipocytes in visceral fat.8, 9, 10, 11, 12, 13, 14 An overexpression of monocyte chemoattractant protein (MCP)-1 in adipose tissues causes macrophage recruitment and insulin resistance in mice.10, 11 TNF-α secretion is highly related to the free fatty acid (FFA)-induced inflammatory changes in both adipocytes and Mφ.12, 13, 14, 15 The peroxisome proliferator-activated receptor activation in Mφ is able to regulate the FFA-induced TNF-α secretion from adipocytes.16

The present study was designed to identify the role of MMP-3 in the interaction between Mφ and adipocytes for TNF-α gene induction. Conditioned media from Mφ (Mφ-CM) increased the TNF-α mRNA expression in adipocytes. The induced levels of TNF-α mRNA were largely abolished by treatment with GM6001, a broad-spectrum MMP inhibitor, or N-isobutyl-N-(4-methoxyphenylsulfonyl)-glycylhydroxamic acid (NNGH), an MMP-3 inhibitor. The active form of MMP-3 enhanced release of TNF-α and glycerol from 3T3-L1 adipocytes, and inhibited insulin-induced glucose uptake into the cells. The MMP-3 expression in Mφ, in addition to adipocytes, is potentially important for the development of a pathological link between Mφ and adipocytes through TNF-α secretion in visceral fat tissue.

Methods

Cell culture and preparation of Mφ conditioned media

3T3-L1 cells (American Type Culture Collection, Manassas, VA, USA) were cultured and differentiated into adipocytes as described previously.16 The human monocytic cell line THP-1 (American Type Culture Collection) was cultured in RPMI 1640 supplemented with l-glutamine (GibcoBRL, Tokyo, Japan) penicillin/streptomycin (100 U per 100 mg ml−1; GibcoBRL) and 10% fetal bovine serum (GibcoBRL, medium A). To allow the monocytes to differentiate into adherent macrophages, THP-1 cells were washed in phosphate-buffered saline (calcium- and magnesium-free; GibcoBRL, buffer A) and resuspended in fresh medium A containing phorbol 12-myristate-13-acetate (50 ng ml−1 PMA; Sigma, St Louis, MO, USA) for 3 days (at day 0), and were incubated for 3 more days in Dulbecco's modified Eagle's medium (DMEM) supplemented with 2% bovine serum albumin (BSA). At day 3, the culture media were collected, centrifuged and stored as Mφ-CM. Control CM were prepared by incubating the THP-1 cells with DMEM supplemented with 2% BSA for 3 days (THP1-CM). Mφ-CM and THP1-CM were stored at −80 °C until use. The differentiation of THP1 to mature Mφ was evaluated by the quantification of CD11b and CD68 mRNA levels using real-time PCR. The differentiated macrophages with CD11b and CD68 mRNA levels of more than two fold greater than those in THP-1 were used for further experiments. Co-culture of adipocytes and Mφ was performed using transwell inserts with 0.4-μm porous membrane (Becton Dickinson, Franklin Lakes, NJ, USA) to separate adipocytes from Mφ. To determine the role of Mφ-secreted factors on adipocyte responses, serum-starved 3T3-L1 adipocytes were incubated with Mφ-CM or THP1-CM ranging from 10 to 50% of the final volume, for the indicated time periods. To evaluate the effects of MMP inhibition on Mφ-CM, Mφ-CM was treated with a broad-spectrum MMP inhibitor GM-6001, a specific peptide inhibitor of the gelatinases MMP-2 and -9, CTTHWGFTLC-decapeptide (CTT) or an MMP-3 inhibitor, NNGH (Calbiochem, San Diego, CA, USA) prior to the addition to adipocytes.

RNA preparation and quantitative real-time RT–PCR

Total RNA was isolated from cultured cells, and quantitative real-time reverse transcription (RT)–PCR was performed with an ABI 7000 sequence detection system using TaqMan Universal PCR Master Mix and Assays-on-Demand Gene Expression Assay Mix (PE Applied Biosystems, Foster City, CA, USA) described previously.17 The quantification of a given gene, expressed as relative mRNA level compared with a control, was calculated after normalization to 18S rRNA.

Enzyme-linked immunosorbent assay

Serum-starved 3T3-L1 adipocytes were incubated with 100 μg ml−1 human MMP-3 (Sigma) for 1–3 days, and the culture medium was assayed for mouse TNF-α using commercial enzyme-linked immunosorbent assay (ELISA) kits (BioLegend, San Diego, CA, USA) according to the manufacturer's instructions as described previously.16

Glycerol release measurement

Differentiated 3T3-L1 adipocytes were incubated with DMEM supplemented with 1% FFA-free BSA for 2 days, and then treated with same medium with Mφ-CM at 50% of volume, THP1-CM at 50% of volume or human MMP-3 at 100 μg ml−1, in the absence or presence of 60 μM NNGH for 6 h. The concentrations of glycerol in the media were determined using a free glycerol determination kit (Sigma) following the manufacturer's protocol.

2-Deoxyglucose uptake assay

Differentiated 3T3-L1 adipocytes were preincubated in serum-starved DMEM with 50% Mφ-CM, 50% THP1-CM or human MMP-3 at 100 μg ml−1, in the absence or presence of 60 μM NNGH for 6 h. Single adipocytes were prepared from mesenteric or subcutaneous fat of mice, fed with high-fat or regular diet as described.14 The cells were incubated in DMEM without serum for 2 h at 37 °C, and then either treated or not treated with 100 nM insulin for 15 min at 37 °C, as described previously.18 After stimulation, 10 μM 2-[3H]deoxyglucose was added and incubated for 5 min. Glucose uptake was stopped by the addition of ice-cold Krebs-Ringer HEPES buffer with 5 μM cytochalasin B and 25 mM glucose. The cells were washed three times with ice-cold Krebs-Ringer HEPES buffer with 25 mM glucose, and the 3H-labeled radioactivity was counted using a scintillation counter (LS-6500; Beckman Coulter Inc., Fullerton, CA, USA).

Animals and animal care

Male C57BL/6J mice (Charles River, Wilmington, MA, USA) were rendered insulin resistant by feeding a high-fat diet consisting of 20% protein, 20% carbohydrate and 60% fat (Research Diet, New Brunswick, NJ, USA) starting at 8 weeks of age for 2 weeks as described previously.14 Control mice were fed a standard diet consisting of 4.5% fat (Research Diet). Mesenteric and subcutaneous fat tissue specimens were resected, and total RNA was isolated as described previously.14 All applicable institutional and governmental regulations concerning the ethical use of animal were followed during this research. All animal care and procedures were approved by the Animal Care Committee of Chiba University School of Medicine as described previously.

Western blot analysis

Membranes from fat tissue specimens were prepared and solubilized in solubilization buffer (200 mM Tris-maleate, pH 6.5, 2 mM CaCl2, 0.5 mM PMSF, 2.5 mM leupeptin and 1% Triton X-100) as previously described.19 The protein concentrations were determined using the BCA Protein Assay Reagent (Pierce, Rockford, IL, USA). For immunoblotting, equal amounts of membrane protein, protein extracted from pelleted beads, or concentrated media were separated by 10% SDS–PAGE after heating to 95 °C for 5 min under reducing conditions, and transferred to a nitrocellulose membrane. The blots were incubated with antibody against MMP-3 (SC-6839, 1:100 dilution), followed by peroxidase-conjugated anti-goat IgG, and then they were developed using the ECL detection reagents (Amersham Pharmacia, Piscataway, NJ, USA). The signals were quantified by densitometric scanning using the NIH image software program.

Statistical analysis

Results are presented as mean±s.d. Statistical significance between two groups was evaluated by Student's t-test. Statistical significance among several groups was performed using a one-way ANOVA. A value of P<0.05 was considered to be significant.

Results

Effects of Mφ-CM on TNF-α gene expression in 3T3-L1 adipocytes

3T3-L1 adipocytes were incubated with Mφ in the transwell system to evaluate the interactions between Mφ and adipocytes. A co-culture of adipocytes and Mφ revealed significant induction of TNF-α gene in adipocytes relative to the control culture at 24 h (Figure 1a). The extent of changes in TNF-α mRNA expression was dependent on the number of Mφ (data not shown). The role of Mφ factors on TNF-α gene in adipocytes was investigated by incubating 3T3-L1 adipocytes with Mφ-CM for 4 h. Consistent with the results in the transwell system, Mφ-CM significantly induced expression of TNF-α mRNA in adipocytes (Figure 1b). The induction of mRNA for TNF-α was 1.9-fold after 4 h of incubation with Mφ-CM in comparison to that in control. Mφ-CM dose-dependently increased TNF-α mRNA expression at the concentrations from 10 to 50% (Figure 1c). There were no obvious changes in the morphology of the adipocytes, and there was no apparent toxicity with either Mφ-CM or THP1-CM (data not shown).

Figure 1
figure1

Mφ-secreted factors increased the expression of tumor necrosis factor-α (TNF-α) in 3T3-L1 adipocytes. (a) Time course of the TNF-α expression in 3T3-L1 adipocytes co-cultured either with THP-1 cells (open circle) or Mφ (filled circle). 3T3-L1 adipocytes were first seeded on the well bottom, and then THP-1 cells or THP-1 Mφ were seeded on the permeable membrane of the insert. TNF-α mRNA levels were analyzed using quantitative real-time RT–PCR. Relative ratios of TNF-α mRNA levels in 3T3-L1 adipocytes co-cultured either with THP-1 cells or Mφ to those co-cultured with THP-1 cells for 8 h (control) were presented. Data are expressed as mean±s.d. (n=4). *P<0.05 in comparison to the value with THP1-CM. (b) Effects of the conditioned media of THP-1 (THP1-CM) or Mφ cultures (Mφ-CM) on the TNF-α gene expression in 3T3-L1 adipocytes. Serum-starved 3T3-L1 adipocytes were treated with either 10% THP1-CM or Mφ-CM for 4 h. TNF-α mRNA levels in 3T3-L1 adipocytes were analyzed using quantitative real-time RT–PCR. The relative ratios of TNF-α mRNA levels in 3T3-L1 adipocytes with Mφ-CM to those with THP1-CM (control) were presented. Data are expressed as mean±s.d. (n=4). *P<0.05 in comparison to the value of the control with THP1-CM. (c) Dose-dependent effect of THP1-CM or Mφ-CM on TNF-α gene expression in 3T3-L1 adipocytes. The TNF-α mRNA levels in 3T3-L1 adipocytes were analyzed using quantitative real-time RT–PCR. Relative ratios of TNF-α mRNA levels to those with THP1-CM at 10% of volume (control) were presented. Data are expressed as mean±s.d. (n=4). *P<0.05.

Role of Mφ-derived factors in induction of TNF-α mRNA in adipocytes

The expression of the MMP-3 gene is one of most induced genes in accumulated visceral fat tissues, and MMP-3 induces the TNF-α secretion from adipocytes.6 To explore the molecular mechanisms of the above observed interaction between Mφ and adipocytes, the role of MMP secreted from Mφ in the induction of TNF-α mRNA was investigated in adipocytes. The expression of MMP genes significantly increased in Mφ in comparison to those in THP-1 cells (Figure 2). Among them, MMP-9 was most induced gene in Mφ (199-fold). The expression of MMP-3 and -12 genes was hardly detected in THP-1 cells. These results raise the possibility that Mφ-secreted MMP enhances the expression of the TNF-α gene in 3T3-L1 adipocytes in co-culture system. Mφ-CM treated with various types of MMP inhibitors was added to 3T3-L1 adipocytes to examine the changes of TNF-α gene expression in 3T3-L1 adipocytes (Figure 3). GM6001, a broad-spectrum MMP inhibitor, markedly altered the stimulatory effects of Mφ-CM on the gene expression of TNF-α (−42%). The gelatinases inhibitor, CTT and an MMP-3 inhibitor, NNGH were used to determine the role of the gelatinases (MMP-2 and -9) and MMP-3 on the TNF-α gene expression in 3T3-L1 adipocytes. The stimulatory effect of Mφ-CM on the TNF-α gene expression was not significantly inhibited by CTT treatment. In contrast, the induction of TNF-α by Mφ-CM was markedly inhibited by NNGH treatment (−56%), suggesting an important role for MMP-3 in the adipocyte function. To determine if MMP-3 is the soluble mediator causing TNF-α induction in adipocytes, 3T3-L1 adipocytes were treated with activated MMP-3, and TNF-α mRNA expression and release were measured. MMP-3 treatment significantly increased TNF-α mRNA expression by 3.2-fold (Figure 4a), and the increases were also detected after 50–200 ng ml−1 MMP-3 treatments for 8 h (data not shown). Figure 4b shows that MMP-3 treatment (100 ng ml−1) increased TNF-α secretion in a time-dependent manner.

Figure 2
figure2

The expression of matrix metalloproteinase (MMP) family genes was induced in Mφ. THP-1 cells were differentiated into Mφ by incubating PMA for 72 h. MMP-2, MMP-3, MMP-9, MMP-12 and TIMP-1 mRNA levels were analyzed using quantitative real-time RT–PCR. Relative ratios of mRNA levels in Mφ to those in THP-1 cells (control) or absolute mRNA levels were presented. Data are expressed as mean±s.d. (n=6). *P<0.05 in comparison to the value of the control with THP-1 cells. N.D., not detected.

Figure 3
figure3

Effect of the inhibition of matrix metalloproteinase (MMP) activity on Mφ-CM-induced tumor necrosis factor-α (TNF-α) gene expression in adipocytes. Serum-starved 3T3-L1 adipocytes were treated with THP-1-CM or Mφ-CM in the absence or presence of 10 μM GM6001, 85 μM CTT or 60 μM NNGH for 4 h. the TNF-α mRNA levels were analyzed using quantitative real-time RT–PCR. Relative ratios of the TNF-α mRNA levels in 3T3-L1 adipocytes to those with THP1-CM (control) were presented. Data are expressed as mean±s.d. (n=4). *P<0.05. n.s., not significant.

Figure 4
figure4

Matrix metalloproteinase (MMP)-3 induced the tumor necrosis factor-α (TNF-α) mRNA and protein expression in 3T3-L1 adipocytes. (a) Serum-starved 3T3-L1 adipocytes were treated with the active form of MMP-3 (100 ng ml−1) for 8 h. The TNF-α mRNA levels were analyzed using quantitative real-time RT–PCR. Relative ratios of the TNF-α mRNA levels to those without MMP-3 (control) were presented. Data are expressed as mean±s.d. (n=6). *P<0.05 compared to the value of the control. (b) Serum-starved 3T3-L1 adipocytes were treated with MMP-3 for 1–3 days. TNF-α protein concentrations in conditioned media were analyzed using an ELISA. Relative ratios of TNF-α concentration with those in the absence of MMP-3 for a day (control) were presented. Data are expressed as mean±s.d. (n=6). *P<0.05 in comparison to the value without MMP-3.

Active MMP-3 induces lipolysis, and reduces insulin-induced glucose incorporation in 3T3-L1 adipocytes

In order to determine the role of Mφ-derived MMP-3 in the functional changes of adipocytes to induce the TNF-α mRNA expression in adipocytes, the effect of MMP-3 on the lipolysis of 3T3-L1 adipocyte was analyzed (Figure 5a). The glycerol release was significantly increased in the media of 3T3-L1 adipocytes incubated with Mφ-CM, in comparison to those incubated with THP1-CM. The increase in glycerol release observed in the cells incubated with Mφ-CM was inhibited by 38% in the presence of NNGH. The glycerol release in the media of 3T3-L1 adipocytes incubated with MMP-3 was also significantly increased, in comparison to those incubated with THP1-CM. The increased release was almost abolished by the NNGH treatment. Next, the effect of MMP-3 on the insulin-induced glucose incorporation into 3T3-L1 adipocytes was analyzed (Figure 5b). The glucose uptake was significantly decreased in the media of 3T3-L1 adipocytes incubated with Mφ-CM, in comparison to those incubated with THP1-CM. The decrease in glycerol release by the incubation 3T3-L1 cells with Mφ-CM was recovered by 69% in the presence of NNGH. The glycerol release in the media of 3T3-L1 adipocytes incubated with MMP-3 was significantly decreased, in comparison to those incubated with THP1-CM, and that reduction thereafter almost completely recovered due to the NNGH treatment. Therefore, Mφ-CM induces lipolysis, and reduces insulin-induced glucose uptake in 3T3-L1 adipocytes, possibly in part through the secretion of MMP-3.

Figure 5
figure5

Effect of matrix metalloproteinase (MMP)-3 on lipolysis and insulin-induced glucose incorporation in 3T3-L1 adipocytes. (a) Glycerol release was measured in the media of 3T3-L1 adipocytes after treatment with 50%THP1-CM, 50% Mφ-CM or 100 μg ml−1 human MMP-3, in the absence or presence of 60 μM NNGH for 6 h. The relative ratios of glycerol content with those treated with THP1-CM (control) were presented. Data are expressed as mean±s.d. (n=8). *P<0.05. n.s., not significant. (b) 2-DG uptake was measured in 3T3-L1 adipocytes in the absence (open column) or presence (closed column) of 100 nM insulin for 15 min after treatment with either 50% THP1-CM, 50% Mφ-CM or 100 μg ml−1 human MMP-3, in the absence or presence of 60 μM NNGH, for 6 h. Relative ratios of 2-DG contents with those treated with THP1-CM in the absence of insulin are indicated. Data are expressed as mean±s.d. (n=6). *P<0.05. n.s., not significant.

High-fat intake induced the expression of MMP-3 in mesenteric fat tissues as well as the induction of F4/80 gene

To assess the expression of MMP-3 gene in adipose tissue Mφ, the levels of MMP-3 mRNA were examined in mesenteric fat tissue from mice fed with high-fat diet in relation to the expression of the F4/80 gene, an Mφ-specific antigen15, 20 (Figure 6). High-fat intake for 2 weeks significantly induced the expression level of F4/80 mRNA in mesenteric fat tissue by 2.2-fold in comparison to the level in the control mice. The levels of MMP-3 and TNF-α genes in mesenteric fat tissue were also significantly induced by 2.8- and 2.5-fold in the mice fed the high-fat diet compared in the control mice, respectively. The expression of F4/80, MMP-3 or TNF-α gene in subcutaneous fat tissues was not significantly different between the mice fed with regular chow and high-fat diet. The MMP-3 protein expression was analyzed in either visceral or subcutaneous fat tissue specimens (Figure 7a). The MMP-3 protein levels in visceral fat tissues, but not in subcutaneous fat, were significantly higher in the mice fed with a high-fat diet than those fed with regular chow. The insulin-induced glucose uptake activity in the adipocytes prepared from visceral fat tissues was significantly decreased in the mice fed with high-fat diet in comparison to those consuming regular chow (Figure 7b). These results indicate that high-fat intake causes Mφ recruitment into visceral fat, and possibly leads to the induction of the MMP-3 and TNF-α expression, as well as the inhibition of glucose incorporation of adipocytes.

Figure 6
figure6

High fat induces F4/80, tumor necrosis factor-α (TNF-α) and matrix metalloproteinase (MMP)-3 mRNA expression in mesenteric fat. Mesenteric or subcutaneous fat tissues were prepared from C57BL/6 mice fed regular chow (Reg) or high-fat (HF) diet. F4/80, TNF-α and MMP-3 mRNA levels were analyzed using quantitative real-time RT–PCR. Relative ratios of mRNA levels in mice fed high-fat to those fed regular chow were presented. Data are expressed as mean±s.d. (n=6). *P<0.05 in comparison to the value of the control.

Figure 7
figure7

High fat induces matrix metalloproteinase (MMP)-3 protein expression, and reduces insulin-induced glucose incorporation of adipocytes in mesenteric fat. Mesenteric (Mesent) or subcutaneous (Subcut) fat tissues were prepared from C57BL/6 mice fed regular chow (Reg) or high-fat (HF) diet. (a) The MMP-3 protein expression of tissue extracts were analyzed by western blot analysis using an antibody against MMP-3. Relative ratios of MMP-3 protein levels with those of mesenteric fat tissues of mice fed regular chow (control) are indicated. Data representative of three experiments are shown. Bars are expressed as mean±s.d. (n=6). *P<0.05. n.s., not significant. (b) The 2-DG uptake was measured in single adipocytes prepared from fat tissues in the absence (open column) or presence (closed column) of 100 nM insulin for 15 min. The relative ratios of 2-DG contents with basal of cells prepared from mesenteric fat of mice fed regular chow are indicated. Data are expressed as mean±s.d. (n=6). *P<0.05. n.s., not significant.

Discussion

The current study demonstrated that Mφ-CM influences the expression of TNF-α from 3T3-L1 adipocytes. This induction of TNF-α is attenuated by an MMP-3 inhibitor, NNGH. The active form of MMP-3 showed the capability for the induction of lipolysis and the inhibition of the insulin-induced glucose uptake, as well as for the enhanced secretion of TNF-α. These findings suggest that MMP-3 thus plays a role in the modulation of the adipocyte function from Mφ in adipose tissues.

Recent observations suggested that inflammatory conditions evoked in fat tissues recruit activated Mφ, possibly enhancing and/or continuing the chronic process in fat tissues.8, 9 TNF-α is suspected to be one of the key players among many cytokines in the interactive modification of function in Mφ and adipocytes.4, 7 Based on the results obtained herein using a culture system, infiltrating Mφ may therefore modify the maturation process and secretion level of TNF-α in adipocytes in fat tissues. The expression of TNF-α is observed in 3T3-L1 preadipocytes, and declines gradually after the beginning of maturation in the presence of inducers.21 The mice with transplanted cultured 3T3-L1 cells showed that the transplanted adipocytes in visceral space, and not subcutaneous space, secret TNF-α and the secreted molecules actually disturb the systemic insulin sensitivity, based on the decreased insulin action in tissues.5 The induced expression of TNF-α is also observed in the adipocytes in visceral spaces of subcutaneously lipectomized mice.22 Therefore, the adipocytes that accumulate in visceral space are potentially sensitive to induce the TNF-α gene expression in mice.

Recent studies have indicated that extracellular matrix (ECM) degradation is important for adipogenesis. MMPs are essential for proper matrix remodeling, a process that takes place during adipose tissue formation. Human mature adipocytes secret MMP-2 and -9 and their proteolytic activities are induced during differentiation of murine-cultured adipocytes.23 mRNA levels for MMP-2, MMP-3, MMP-12, MMP-14, MMP-19 and TIMP-1 are strongly induced in obese adipose tissues in a genetic or a diet-induced model of obesity.24 The treatment of cultured preadipocytes with either synthetic MMP inhibitors or neutralizing antibodies decreases differentiation.22 These previous studies using cultured adipocytes suggest that MMP activity is required for adipocyte conversion. The body weight of MMP-3-deficient mice is increased in comparison to that of wild-type mice, as is the weight of the isolated subcutaneous and gonadal fat deposits.25 MMP-11-deficient mice develop adipocyte hypertrophy in comparison to wild-type mice.26 Furthermore, the membrane-anchored metalloproteinase, MT1-MMP, acts as a 3D-specific adipogenic factor that directs the dynamic adipocyte-ECM interactions critical to WAT development.27 These studies using knockout models revealed critical roles of MMPs in fat tissue development and adipogenesis, and possibly also in fat accumulation accompanied with insulin resistance. A recent study reported that the MMP-3 expression levels are negatively correlated with percent body fat, and the MMP-3 gene variants are associated with both BMI and type 2 diabetes in Pima Indians.28

The mice with transplanted cultured 3T3-L1 cells showed that the transplanted adipocytes in the visceral space, and not subcutaneous space, increased TNF-α gene expression.5 A microarray analysis revealed that the MMP-3 gene expression is drastically induced in addition to TNF-α.6 Therefore, the MMP-3 gene expression in visceral fat seems to be directly linked to cytokine expression in adipocytes. The current study showed that the active form of MMP-3 enhanced glycerol release, as well as TNF-α protein secretion, from 3T3-L1 adipocytes. The incubation of adipocytes with MMP-3 inhibited insulin-induced glucose uptake in adipocytes. Therefore, the induction of MMP-3 gene expression may modulate lipid and glucose metabolism in visceral adipocytes, leading to the induction of TNF-α secretion. The treatment of 3T3-L1 preadipocytes with the MMP inhibitor Ilomastat has been shown to prevent their differentiation into adipocytes.29 The subcutaneous administration of MMP inhibitor KB-R7785 reduced the plasma glucose and insulin levels with a concomitant decrease in the TNF-α production in KK-Ay mice.30 These observations indicate that Mφ-MMP may thus play a functional role in the induction of TNF-α gene expression impairing insulin sensitivity in adipocytes.

Recently, MMP-3 has been shown to be a signaling molecule via the ERK pathway, followed by proinflammatory cytokine induction, and induce superoxide generation in microglia.31 Moreover, activated MMP-3 is present in the nuclear compartment of malignant and nontransformed hepatocytes, and is associated with the onset of apoptosis.32 These studies suggested a novel function of MMP-3 as a signaling molecule active for intracellular functions. The current results showed that high-fat intake induced a decrease in insulin-induced glucose incorporation in adipocytes, as well as an increase in Mφ-infiltration and TNF-α expression in visceral fat tissue. Therefore, MMP-3 may affect the lipid metabolism of adipocytes through the ECM degradation and the activation of other extracellular and intracellular molecules leading to the lipolysis and glucose incorporation. Therefore, Mφ-derived MMP-3 may modulate the secretion of TNF-α in adipocytes by modulating the lipid metabolism, which is tightly linked to visceral fat accumulation and systemic insulin resistance.

In conclusion, this study suggests that MMP-3 is important for the function of pathological link between Mφ and adipocytes, which leads to insulin resistance in metabolic syndrome through the regulation of cytokine expression such as TNF-α. The further elucidation of the role of MMP-3 and its secretion from activated Mφ and adipocytes is therefore expected to contribute to the elucidation of the unexpected relationship between chronic inflammation and disturbed insulin sensitivity in humans.

References

  1. 1

    Grundy SM . Obesity, metabolic syndrome, and cardiovascular disease. J Clin Endocrinol Metab 2004; 89: 2595–2600.

  2. 2

    Garg A . Regional adiposity and insulin resistance. J Clin Endocrinol Metab 2004; 89: 4206–4210.

  3. 3

    Matsuzawa Y . Adiponectin: identification, physiology and clinical relevance in metabolic and vascular disease. Atheroscler Suppl 2005; 6: 7–14.

  4. 4

    Pittas AG, Joseph NA, Greenberg AS . Adipocytokines and insulin resistance. J Clin Endocrinol Metab 2004; 89: 447–452.

  5. 5

    Shibasaki M, Takahashi K, Itou T, Miyazawa S, Ito M, Kobayashi J et al. Alterations of insulin sensitivity by the implantation of 3T3-L1 cells in nude mice. A role for TNF-α? Diabetologia 2002; 45: 518–526.

  6. 6

    Unoki H, Bujo H, Shibasaki M, Saito Y . Increased matrix metalloproteinase-3 mRNA expression in visceral fat in mice implanted with cultured preadipocytes. Biochem Biophys Res Commun 2006; 350: 392–398.

  7. 7

    Wellen KE, Hotamisligil GS . Inflammation, stress, and diabetes. J Clin Invest 2005; 115: 1111–1119.

  8. 8

    Weisberg SP, McCann D, Desai M, Rosenbaum M, Leibel RL, Ferrante Jr AW . Obesity is associated with macrophage accumulation in adipose tissue. J Clin Invest 2003; 112: 1796–1808.

  9. 9

    Xu H, Barnes GT, Yang Q, Tan G, Yang D, Chou CJ et al. Chronic inflammation in fat plays a crucial role in the development of obesity-related insulin resistance. J Clin Invest 2003; 112: 1821–1830.

  10. 10

    Kamei N, Tobe K, Suzuki R, Ohsugi M, Watanabe T, Kubota N et al. Overexpression of monocyte chemoattractant protein-1 in adipose tissues causes macrophage recruitment and insulin resistance. J Biol Chem 2006; 281: 26602–26614.

  11. 11

    Kanda H, Tateya S, Tamori Y, Kotani K, Hiasa K, Kitazawa R et al. MCP-1 contributes to macrophage infiltration into adipose tissue, insulin resistance, and hepatic steatosis in obesity. J Clin Invest 2006; 116: 1494–1505.

  12. 12

    Suganami T, Nishida J, Ogawa Y . A paracrine loop between adipocytes and macrophages aggravates inflammatory changes: role of free fatty acids and tumor necrosis factor α. Arterioscler Thromb Vasc Biol 2005; 25: 2062–2068.

  13. 13

    Shi H, Kokoeva MV, Inouye K, Tzameli I, Yin H, Flier JS . TLR4 links innate immunity and fatty acid-induced insulin resistance. J Clin Invest 2006; 116: 3015–3025.

  14. 14

    Murakami K, Bujo H, Unoki H, Saito Y . High fat intake induces a population of adipocytes to co-express TLR2 and TNFα in mice with insulin resistance. Biochem Biophys Res Commun 2007; 354: 727–734.

  15. 15

    Nguyen MT, Favelyukis S, Nguyen AK, Reichart D, Scott PA, Jenn A et al. A subpopulation of macrophages infiltrates hypertrophic adipose tissue and is activated by free fatty acids via Toll-like receptors 2 and 4 and JNK-dependent pathways. J Biol Chem 2007; 282: 35279–35292.

  16. 16

    Murakami K, Bujo H, Unoki H, Saito Y, Murakami K, Bujo H et al. Effect of PPARα activation of macrophages on the secretion of inflammatory cytokines in cultured adipocytes. Eur J Pharmacol 2007; 561: 206–213.

  17. 17

    Hirata T, Unoki H, Bujo H, Ueno K, Saito Y . Activation of diacylglycerol O-acyltransferase 1 gene results in increased tumor necrosis factor-α gene expression in 3T3-L1 adipocytes. FEBS Lett 2006; 580: 5117–5121.

  18. 18

    Shibasaki M, Bujo H, Takahashi K, Murakami K, Unoki H, Saito Y . Catalytically inactive lipoprotein lipase overexpression increases insulin sensitivity in mice. Horm Metab Res 2006; 38: 491–496.

  19. 19

    Ohwaki K, Bujo H, Jiang M, Yamazaki H, Schneider WJ, Saito Y . A secreted soluble form of LR11, specifically expressed in intimal smooth muscle cells, accelerates formation of lipid-laden macrophages. Arterioscler Thromb Vasc Biol 2007; 27: 1050–1056.

  20. 20

    Khazen W, M'bika JP, Tomkiewicz C, Benelli C, Chany C, Achour A et al. Expression of macrophage-selective markers in human and rodent adipocytes. FEBS Lett 2005; 579: 5631–5634.

  21. 21

    Cowherd RM, Lyle RE, McGehee Jr RE . Molecular regulation of adipocyte differentiation. Semin Cell Dev Biol 1999; 10: 3–10.

  22. 22

    Ishikawa K, Takahashi K, Bujo H, Hashimoto N, Yagui K, Saito Y . Subcutaneous fat modulates insulin sensitivity in mice by regulating TNF-α expression in visceral fat. Horm Metab Res 2006; 38: 631–638.

  23. 23

    Bouloumié A, Sengenès C, Portolan G, Galitzky J, Lafontan M . Adipocyte produces matrix metalloproteinases 2 and 9: involvement in adipose differentiation. Diabetes 2001; 50: 2080–2086.

  24. 24

    Chavey C, Mari B, Monthouel MN, Bonnafous S, Anglard P, Van Obberghen E et al. Matrix metalloproteinases are differentially expressed in adipose tissue during obesity and modulate adipocyte differentiation. J Biol Chem 2003; 278: 11888–11899.

  25. 25

    Maquoi E, Demeulemeester D, Voros G, Collen D, Lijnen HR . Enhanced nutritionally induced adipose tissue development in mice with stromelysin-1 gene inactivation. Thromb Haemost 2003; 89: 696–704.

  26. 26

    Lijnen HR, Van HB, Frederix L, Rio MC, Collen D . Adipocyte hypertrophy in stromelysin-3 deficient mice with nutritionally induced obesity. Thromb Haemost 2002; 87: 530–535.

  27. 27

    Chun TH, Hotary KB, Sabeh F, Saltiel AR, Allen ED, Weiss SJ . A pericellular collagenase directs the 3-dimensional development of white adipose tissue. Cell 2006; 125: 577–591.

  28. 28

    Traurig MT, Permana PA, Nair S, Kobes S, Bogardus C, Baier LJ . Differential expression of matrix metalloproteinase 3 (MMP3) in preadipocytes/stromal vascular cells from nonobese nondiabetic versus obese nondiabetic Pima Indians. Diabetes 2006; 55: 3160–3165.

  29. 29

    Croissandeau G, Chretien M, Mbikay M . Involvement of matrix metalloproteinases in the adipose conversion of 3T3-L1 preadipocytes. Biochem J 2002; 364: 739–746.

  30. 30

    Morimoto Y, Nishikawa K, Ohashi M . KB-R7785, a novel matrix metalloproteinase inhibitor, exerts its antidiabetic effect by inhibiting tumor necrosis factor-alpha production. Life Sci 1997; 61: 795–803.

  31. 31

    Kim YS, Choi DH, Block ML, Lorenzl S, Yang L, Kim YJ et al. A pivotal role of matrix metalloproteinase-3 activity in dopaminergic neuronal degeneration via microglial activation. FASEB J 2007; 21: 179–187.

  32. 32

    Si-Tayeb K, Monvoisin A, Mazzocco C, Lepreux S, Decossas M, Cubel G et al. Matrix metalloproteinase 3 is present in the cell nucleus and is involved in apoptosis. Am J Pathol 2006; 169: 1390–1401.

Download references

Acknowledgements

This work was partly supported by Grants-in-Aid for Scientific Research to HB and HU from the Ministry of Education, Culture, Sports, Science and Technology, Japan, and Grants-in-Aid for Research Committee to HB and YS from the Ministry of Health, Labor and Welfare, Japan.

Author information

Correspondence to H Bujo.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Unoki, H., Bujo, H., Jiang, M. et al. Macrophages regulate tumor necrosis factor-α expression in adipocytes through the secretion of matrix metalloproteinase-3. Int J Obes 32, 902–911 (2008). https://doi.org/10.1038/ijo.2008.7

Download citation

Keywords

  • macrophage
  • matrix metalloproteinase-3
  • tumor necrosis factor-α
  • adipocyte
  • insulin resistance

Further reading