Effects of C-reactive protein on the expression of matrix metalloproteinases and their inhibitors via Fcγ receptors on 3T3-L1 adipocytes.

The association between obesity and inflammation is well documented in epidemiological studies. Proteolysis of extracellular matrix (ECM) proteins is involved in adipose tissue enlargement, and matrix metalloproteinases (MMPs) collectively cleave all ECM proteins. Here, we examined the effects of C-reactive protein (CRP), an inflammatory biomarker, on the expression of MMPs and tissue inhibitors of metalloproteinases (TIMPs), which are natural inhibitors of MMPs, in adipocyte-differentiated 3T3-L1 cells. We analyzed the expression of Fcγ receptor (FcγR) IIb and FcγRIII, which are candidates for CRP receptors, and the effects of anti-CD16/CD32 antibodies, which can act as FcγRII and FcγRIII blockers on CRP-induced alteration of MMP and TIMP expression. Moreover, we examined the effects of CRP on the activation of mitogen-activated protein kinase (MAPK) signaling, which is involved in MMP and TIMP expression, in the presence or absence of anti-CD16/CD32 antibodies. Stimulation with CRP increased MMP-1, MMP-3, MMP-9, MMP-11, MMP-14, and TIMP-1 expression but did not affect MMP-2, TIMP-2, and TIMP-4 expression; TIMP-3 expression was not detected. Adipocyte-differentiated 3T3-L1cells expressed FcγRIIb and FcγRIII; this expression was upregulated on stimulation with CRP. Anti-CD16/CD32 antibodies inhibited CRP-induced expression of MMPs, except MMP-11, and TIMP-1. CRP induced the phosphorylation of extracellular signal-regulated kinase (ERK) 1/2 and p38 MAPK but did not affect SAPK/JNK phosphorylation, and Anti-CD16/CD32 attenuated the CRP-induced phosphorylation of p38 MAPK, but not that of ERK1/2. These results suggest that CRP facilitates ECM turnover in adipose tissue by increasing the production of multiple MMPs and TIMP-1 in adipocytes. Moreover, FcγRIIb and FcγRIII are involved in the CRP-induced expression of MMPs and TIMP-1 and the CRP-induced phosphorylation of p38, whereas the FcγR-independent pathway may regulate the CRP-induced MMP-11 expression and the CRP-induced ERK1/2 phosphorylation.

Introduction

Obesity is frequently associated with hyperglycemia, hyperinsulinemia, hypertension, and dyslipidemia 1, 2; this cluster of metabolic disorders comprises metabolic syndrome, which is a known risk factor for cardiovascular disease 3,4 and type 2 diabetes 5,6. Obesity onset and exacerbation arise from adipose tissue enlargement involving adipogenesis, angiogenesis, and proteolysis of extracellular matrix (ECM) proteins 7-10.

The matrix metalloproteinase (MMP) family comprises over 20 neutral endopeptidase that can collectively cleave all ECM and non-ECM proteins 11,12. MMP activity depends on interactions between MMPs and tissue inhibitors of metalloproteinases (TIMPs), which are natural inhibitors of MMPs 11,12. Changes in MMP and TIMP levels were observed in an obesity mouse model 13-15, suggesting that the MMP and TIMP system has a potential role in obesity development following adipose tissue hypertrophy and hyperplasia.

C-reactive protein (CRP) is the most extensively studied inflammatory biomarker. The association of elevated CRP levels with obesity, cardiovascular disease, and diabetes development has been well documented in epidemiological and in vivo experimental studies 16-21. Fcγ receptors (FcγRs), a family of glycoproteins, bind to extracellular IgGs and to CRP or serum amyloid P, which are involved in the innate immune system 22,23. Thus, CRP can act as an FcγR ligand 21,24. In human cells, three FcγR classes have been identified: FcγRI (CD64), FcγRII (CD32), and FcγRIII (CD16) 22,23. A study using human histiocytes indicated that CRP induced MMP-1 expression via FcγRII 25. The effects of CRP on adipocytes acting as endocrine cells, secreting various adipokines, were reported by Yuan et al. 26,27; their in vitro studies using 3T3-L1 murine adipocytes revealed that CRP suppresses adiponectin and leptin expression but induces interleukin (IL)-6 and tumor necrosis factor (TNF)-α expression. Less information is available about the effect of CRP on MMP and TIMP expression, and activation of mitogen-activated protein kinase (MAPK) signaling, which regulates MMP and TIMP expression 28-29, in adipocytes. Here, we focused on the degradation of ECM, which is involved in adipose tissue enlargement, and conducted an in vitro study to examine the effects of CRP on MMP and TIMP expression in adipocytes. We investigated the effects of anti-CD16/CD32 antibodies (Abs), which can act as blockers of FcγRs, on CRP-induced alteration of MMP and TIMP expression. Moreover, we examined the effect of CRP on the phosphorylation of extracellular signal-regulated kinase (ERK) 1/2, p38 MAPK, and stress-activated protein kinases/c-jun N-terminal kinases (SAPK/JNK) in the presence or absence of anti-CD16/CD32 Abs.

Material and methods

Cell culture and differentiation

We used cells of the mouse embryo cell line 3T3-L1 as model preadipocytes. 3T3-L1 (Riken BioResource Center, Tsukuba, Japan) cells were cultured at 37°C in 5% CO2 in Dulbecco's modified Eagle's medium (DMEM; Gibco-BRL, Rockville, MD, USA) containing 25 mM glucose, 10% heat-inactivated fetal bovine serum (FBS; Gibco-BRL), and 1% (v/v) penicillin/streptomycin (Sigma-Aldrich, St. Louis, MO, USA). At confluence, 3T3-L1 cells were cultured for 2 days in DMEM further supplemented with 1 μM insulin, 0.5 μM isobutylmethylxanthine, and 0.1 μM dexamethasone (AdipoInducer Reagent; Takara Bio, Shiga, Japan). On day 2 and thereafter, DMEM containing 10% FBS, 1% (v/v) penicillin/streptomycin, and 1 μM insulin was subsequently replaced every 2 days. By day 8, 90% of the preadipocytes differentiated into adipocytes, as determined by lipid accumulation visualized with Oil Red O staining.

Stimulation with CRP

Adipocytes were starved for 6 h in FBS-free medium and then stimulated with 0, 25, or 50 μg/mL human recombinant CRP (Calbiochem, La Jolla, CA, USA) for 12 h. The CRP concentration range was chosen based on previous studies 26,27. To investigate the role of FcγRs in CRP-induced alteration of MMP and TIMP expression in adipocytes, the cells were cultured in the presence or absence of 1.0 µg/mL anti-CD16/CD32 Abs (Abcam, Cambridge, MA, USA) for 1 h before stimulation with CRP. The Ab concentrations used were based on manufacturer instructions. CRP and anti-CD16/CD32 Abs did not apparently affect cellular lipid accumulation or architecture (Fig. 1).

Figure 1

Lipid accumulation in CRP-stimulated and unstimulated 3T3-L1 cells. Differentiated 3T3-L1 cells were cultured with 0 (control) or 50 µg/mL CRP in the presence or absence of anti-CD16/32 Abs for 12 h; cells were stained with Oil Red O.

Real-time reverse transcription (RT)-polymerase chain reaction (PCR)

Total RNA was isolated using NucleoSpin RNA (Takara Bio) and treated with DNase. mRNA was converted into complementary DNA (cDNA) with an RNA PCR kit (PrimeScript; Takara Bio). The resulting cDNA mixture was diluted 1:2 in sterile distilled water, and 2 µL diluted cDNA was subjected to real-time polymerase chain reaction (PCR) with SYBR Green I. The reactions were performed in 25 µL SYBR premixed Ex Taq solution (Takara Bio) containing 10 µM sense and antisense primers (Table 1). The PCRs were performed using a Thermal Cycler Dice Real Time System (Takara Bio) and analyzed using the instrument's software. The protocol for MMPs, TIMPs, and FcγRs was 40 cycles at 95°C for 5 s and 60°C for 30 s. All real-time PCR experiments were performed in triplicate; product specificity was verified through melting curve analysis. Calculated gene expression levels were normalized to 36B4 mRNA levels.

Table 1

PCR primers used in the experiments.

Target	Forward primer	Reverse primer	Genbank acc no.
MMP-1	5'-CACATTGATGCTGCTGTTGTGA-3'	5'-TCTGCTGTTAATCTGGGATAACCTG-3'	NM_032006.3
MMP-2	5'-GATAACCTGGATGCCGTCGTG-3'	5'-GGTGTGCAGCGATGAAGATGATA-3'	NM_008610.2
MMP-3	5'-CTCATGCCTATGCACCTGGAC-3'	5'-TCATGAGCAGCAACCAGGAA-3'	NM_010809.1
MMP-9	5'-GCCCTGGAACTCACACGACA-3'	5'-TTGGAAACTCACACGCCAGAAG-3'	NM_013599
MMP-11	5'-TGGAGACTATTGGCGTTTCCAC -3'	5'-TTCACGGGATCAAACTTCCAG -3'	NM_008606
MMP-13	5'-TCCCTGGAATTGGCAACAAAG-3'	5'-GCATGACTCTCACAATGCGATTAC-3'	NM_008607.2
MMP-14	5'-GCAGTGGACAGCGAGTACCCTA-3'	5'-AGTCCCGCAGAGCTGACTTG-3'	NM_008608.3
TIMP-1	5'-CTATAGTGCTGGCTGTGGGGTGTG-3'	5'-TTCCGTGGCAGGCAAGCAAAGT-3'	NM_001044384.1
TIMP-2	5'-GGCCTCCCTCCCTTACTCC-3'	5'-GACTTCATATTCCAGCACGCACAT-3'	NM_011594.3
TIMP-3	5'-CTAAGTCGGCTGTTTGGGTTGA-3'	5'-CAGCACAGCTGGCTTGCTAGA-3'	NM_011595.2
TIMP-4	5'-GCCTGAATCATCACTACCACCAGA-3'	5'-TGAGATGGTACATGGCACTGCATA-3'	NM_080639.3
FcγIIb	5'-ACTTTGTGCCATATGCTACTGTGGA -3'	5'-GAGTTTGACCACAGCCTTTGGAA -3'	NM_001077189
FcγIII	5'-GCCAATGGCTACTTCCACCAC -3'	5'-GTCCAGTTTCACCACAGCCTTC -3'	NM_010188
36B4	5'- AAGCGCGTCCTGGCATTGTCT-3'	5'-CCGCAGGGGCAGCAGTGGT -3'	NM_007475

SDS-PAGE and western blotting

Cells were lysed with extraction buffer containing 0.05% Triton X-100, 10 mM β-mercaptoethanol, 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM ethylenediaminetetraacetic acid, and 25 mM Tris-HCl (pH 7.4). Cell membranes were disrupted by sonication, and the samples were clarified by centrifugation. Supernatants containing 20 µg intracellular protein were dissolved in 10 µL sample buffer containing 1% sodium dodecyl sulfate (SDS), 2 M urea, 15 mg/mL dithiothreitol, and bromophenol blue and heated at 95°C for 5 min before loading. The proteins were resolved by 4-20% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) with a discontinuous Tris-glycine buffer system 30, transferred to a polyvinylidene fluoride membrane by using a semidry transfer apparatus, and probed with Abs. The polyclonal or monoclonal IgG primary Abs used included the following: rabbit anti-MMP-2, anti-MMP-13, anti-TIMP-1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and anti-MMP-14 (Assay Biotech, Sunnyvale, CA, USA) Abs; goat anti-MMP-1, anti-MMP-3, anti-MMP-9, and anti-MMP-11 Abs (Santa Cruz Biotechnology); mouse anti-β-tubulin Abs (Santa Cruz Biotechnology); and rabbit anti-ERK1/2, anti-phospho-ERK1/2, anti-p38 MAPK, anti-phospho-p38 MAPK, anti-SAPK/JNK, and anti-phospho-SAPK/JNK (Cell Signaling Technology, Danvers, MA, USA) Abs. They were used with the appropriate biotin-conjugated donkey anti-goat IgG (Chemicon International, Temecula, CA, USA), goat anti-rabbit IgG (Zymed, San Francisco, CA, USA), or goat anti-mouse IgG (Abcam plc, Cambridge, UK) secondary Abs. The membranes were labeled with streptavidin-horseradish peroxidase (streptavidin-HRP) and visualized using a commercial chemiluminescence kit (Amersham Life Sciences, Little Chalfont, Buckinghamshire, UK). For reprobing with different Abs, the membrane was stripped with Restore PLUS Western blot stripping buffer (Thermo Scientific, Rockford, IL, USA) at room temperature for 15 min.

Statistical analysis

Values have been reported in terms of mean ± standard deviation (SD). Significant differences were determined using one-way analysis of variance (ANOVA) followed by Tukey's multiple comparison test. Differences with p value <0.05 were considered statistically significant.

Results

Effect of CRP on MMP and TIMP mRNA expression

MMP and TIMP mRNA expression was determined by real-time PCR using 3T3-L1 cells cultured for 12 h with or without CRP. MMP-1, MMP-11, and MMP-13 expression significantly increased by 2-2.5, 1.5-3.0, and 2.0-2.6 fold, respectively, in cells stimulated with 25 and 50 μg/mL CRP, as compared to levels in unstimulated control cells (Fig. 2A, E, and F). MMP-2, MMP-3, MMP-9, and MMP-14 expression was significantly, i.e., 1.2, 1.8, 1.5, and 1.5-fold higher, respectively, in cells stimulated with 50 μg/mL CRP than in unstimulated control cells (Fig. 2B-D and G).

Figure 2

Effect of CRP on MMP and TIMP mRNA expression. Differentiated 3T3-L1 cells were cultured with 0 (control), 25, or 50 µg/mL CRP for 12 h and the mRNA expression of seven MMPs (A-G) and four TIMPs (H-J) was determined by real-time PCR. Each bar indicates the mean ± standard deviation (SD) of three independent experiments. *p < 0.05, **p < 0.01 (stimulation with CRP vs. control).

TIMP-1 mRNA expression was significantly, i.e., 1.3-1.5 fold, higher in cells stimulated with 25 or 50 μg/mL CRP than in unstimulated control cells, whereas TIMP-2 and TIMP-4 mRNA expression was unaffected by CRP (Fig. 2H-J). TIMP-3 mRNA expression was not detected in CRP-stimulated and unstimulated control cells (data not shown).

Effect of anti-CD16/CD32 Abs on CRP-induced MMP and TIMP-1 expression

Before performing the inhibitory assay with anti-CD16/CD32 Abs, FcγRIIb (isoform of FcγRII) and FcγRIII mRNA expression in 3T3-L1 adipocytes was analyzed. CRP-stimulated and unstimulated control cells expressed FcγRIIb and FcγRIII mRNA; the expression of both receptors in the presence of 50 μg/mL CRP was significantly higher than that in the control (Fig. 3A and B). Next, 3T3-L1 adipocytes were stimulated with 50 μg/mL CRP in the presence or absence of anti-CD16/CD32 Abs, and MMP-1, MMP-2, MMP-3, MMP-9, MMP-11, MMP-13, MMP-14, and TIMP-1 expression was determined by real-time PCR and western blotting. MMP-1, MMP-2, MMP-3, MMP-9, MMP-13, MMP-14, and TIMP-1 mRNA expression in cells stimulated with CRP in the presence of Abs was significantly lower than that in cells stimulated with CRP in the absence of Abs (Fig. 4A-D, F-H). Thus, the anti-CD16/CD32 Abs blocked induction of MMPs and TIMP-1 mRNA expression by CRP. The effect of anti-CD16/CD32 Abs on CRP-induced MMP-11 expression was unexpected. MMP-11 expression was slightly higher in cells stimulated with CRP in the presence of Abs than in cells stimulated with CRP in the absence of Abs; this difference was statistically significant. Thus, anti-CD16/CD32 Abs slightly enhanced the effects of CRP on MMP-11 expression (Fig. 4E). The effects of stimulation with CRP on MMP and TIMP-1 protein expression in the presence or absence of anti-CD16/CD32 Abs were similar to those on mRNA expression, except in the case of MMP-2. CRP significantly induced MMP-1, MMP-3 MMP-9, MMP-13, MMP-14, and TIMP-1 protein expression, which was decreased by anti-CD16/CD32 Abs (Fig. 5A, B, D, E, G-I). MMP-11 protein expression was also increased by stimulation with CRP, and anti-CD16/CD32 Abs enhanced this increase (Fig. 5A and F). In contrast, MMP-2 protein expression levels in cells stimulated with CRP in both presence and absence of anti-CD16/CD32 Abs did not significantly differ from those in unstimulated control cells (Fig. 5A and C).

Figure 3

Effect of CRP on FcγRIIb and FcγRIII mRNA expression. Differentiated 3T3-L1 cells were cultured with 0 (control), 25, or 50 µg/mL CRP for 12 h; FcγRIIb (A) and FcγRIII (B) mRNA levels were determined by real-time PCR. Each bar indicates the mean ± standard deviation (SD) of three independent experiments. **p < 0.01 (stimulation with CRP vs. control)

Figure 4

Effect of anti-CD16/CD32 Abs on CRP-induced MMP and TIMP-1 mRNA expression. Differentiated 3T3-L1 cells were cultured with 0 (control) or 50 µg/mL CRP in the presence or absence of anti-CD16/CD32 Abs for 12 h; the mRNA expression of seven MMPs (A-G) and TIMP-1 (H) was determined by real-time PCR. Each bar indicates the mean ± standard deviation (SD) of three independent experiments. **p < 0.01 (stimulation with CRP vs. control), †p < 0.05, p < 0.01 (stimulation with CRP vs. anti-CD16/32 Ab + stimulation with CRP).

Figure 5

Effect of anti-CD16/CD32 Abs on CRP-induced MMP and TIMP-1 protein expression. Differentiated 3T3-L1 cells were cultured with 0 (control) or 50 µg/mL CRP in the presence or absence of anti-CD16/32 Abs for 12 h; the expression of seven MMPs and TIMP-1 protein was detected by western blotting (A). The blot intensities of MMPs (B-H) and TIMP-1 (I) were analyzed using a densitometer. Each bar indicates the mean ± standard deviation (SD) of three independent experiments. **p < 0.01 (stimulation with CRP vs. control), †p < 0.05, ††p < 0.01 (stimulation vs. anti-CD16/32 Ab + stimulation).

Effect of CRP and anti-CD16/CD32 Abs on the phosphorylation of ERK1/2, p38 MAPK, or SAPK/JNK

To determine the effects of CRP via FcγR on the activation of MAPK, the phosphorylation statuses of ERK1/2, p38 MAPK, or SAPK/JNK were determined after stimulating the cells with CRP in the presence or absence of anti-CD16/CD32 Abs. CRP induced the phosphorylation of ERK1/2 and p38 MAPK (Fig. 6A, B and C), but had no effect on SAPK/JNK phosphorylation (Fig. 6A and D). Anti-CD16/CD32 attenuated the CRP-induced phosphorylation of p38 MAPK, but not that of ERK1/2 (Fig. 6A, B, and C).

Figure 6

Effect of CRP and anti-CD16/CD32 on the phosphorylation of ERK1/2, p38 MAPK, and SAPK/JNK. Differentiated 3T3-L1 cells were cultured with 0 (control) or 50 µg/mL CRP in the presence or absence of anti-CD16/32 Abs for 12 h; the phosphorylation of ERK1/2, p38 MAPK, and SAPK/JNK was examined by western blotting (A). The blot intensities of the phosphorylated ERK1/2 (B), p38 MAPK (C), and SAPK/JNK (D) were analyzed using a densitometer. Each bar indicates the mean ± standard deviation (SD) of three independent experiments. **p < 0.01 (stimulation with CRP vs. control), ††p < 0.01 (stimulation vs. anti-CD16/32 Ab + stimulation).

Discussion

ECM proteins in adipose tissue undergo constant turnover, and MMPs are involved in the degradation of collagenous and noncollagenous proteins 7-12,31,32. MMP-1 and MMP-13 (classified as collagenase-1 and collagenase-3, respectively) cleave the triple helix of fibrillar collagen (e.g., collagen I) into two fragments at the three-quarters position from the N terminus 11,12. Subsequently, MMP-2 and MMP-9 (gelatinase-A and gelatinase-B, respectively) can degrade collagen fragments. In the current study, CRP induced MMP-1, MMP-13, and MMP-9 mRNA and protein expression 11,12. CRP also upregulated MMP-2 mRNA but not protein expression. The increase in MMP-2 mRNA expression on stimulation with CRP (1.2 fold relative to control) was lower than that for other MMPs (1.5-3.0 fold relative to control); therefore, we considered that CRP had only a marginal effect on MMP-2 expression. These findings and our results indicated that CRP facilitated the overall process of fibrillar collagen degradation in adipose tissue via upregulation of collagenase (MMP-1 and MMP-13) and gelatinase (MMP-9) expression in 3T3-L1 adipocytes.

Mature adipocytes are surrounded by a basement membrane consisting of nonfibrillar collagen IV, laminin, and proteoglycans 8,9,31,33. Collagen VI binds various other ECM components, including collagens I and IV 31,34. Our in vitro study revealed that CRP induced MMP-3, MMP-11, and MMP-14 expression. MMP-3 and MMP-11 are classified as stromelysin-1 and stromelysin-3, respectively; the former degrades collagen IV, laminin, and proteoglycan, and the other mainly degrades collagen VI. MMP-14 (MT1-MMP) is expressed on the surface of cellular membranes and has broad substrate specificity 11,12. Most MMPs are secreted as inactive proMMPs, whereas MMP-14 is expressed as an active enzyme on the cell surface and degrades ECM proteins, including collagen I, gelatin, laminin, and fibronectin, and activates proMMPs 11,35. These findings and our results suggest that CRP facilitates proteolysis of both fibrillar and nonfibrillar ECM proteins in adipose tissue via stromelysin (MMP-3 and MMP-11) and MT-MMP (MMP-14) upregulation.

Posttranslational regulation of MMP activity depends on interactions between MMPs and TIMPs. TIMP-1 and TIMP-4 act as inhibitors against most types of MMPs, whereas they do not fully work as MMP-14 inhibitors; MMP-14 activity is suppressed by TIMP-2 and TIMP-3 9,11,12. Here, CRP increased TIMP-1 expression in adipocytes, whereas TIMP-2 and TIMP-4 expression was unaffected by stimulation with CRP. Degradation of ECM components in adipose tissue, including interstitial fibers and the basement membrane, allow adipocyte hypertrophy and hyperplasia in obesity 7-10. The ECM acts as a frame maintaining the 3D structure of adipose tissue 9,31,33. Thus, CRP might facilitate ECM degradation via upregulation of expression of multiple MMPs, whereas CRP-induced TIMP-1 might contribute to proteolysis regulation to maintain the frame of adipose tissue. Nutritional excess is one of the most important risk factors for obesity; association of obesity with systemic or local infection with organisms like Chlamydia pneumoniae or Porphyromonas gingivalis (periodontitis-related pathogen) has been documented 36-39. These findings and our results suggest that elevation of CRP levels caused by these infections may act as a modification factor for exacerbating obesity via facilitation of ECM protein turnover in adipose tissue.

Here, we did not detect TIMP-3 mRNA expression in 3T3-L1 adipocytes. Bernot et al. 40 reported that mRNA expression of all four TIMPs in 3T3-L1 cells was markedly lower in adipocyte-differentiated 3T3-L1 than in undifferentiated cells. They found that TIMP-1, TIMP-2, and TIMP-4 expression remained at detectable levels during adipocyte conversion, whereas the decrease in TIMP-3 mRNA was substantial as compared to that of other TIMPs 40. Here, we determined the mRNA expression levels of four TIMPs in adipocyte-differentiated 3T3-L1 cells; our result that TIMP-3 mRNA expression was not detected agreed with the findings of Bernot et al. 41.

FcγRIIa, one of the three FCγRII isoforms, has been identified as a CRP receptor in human immune cells 41,42. Murine immune cells express FCγRIIb but not FCγRIIa and FCγRIIc 22,23. Studies using mice reported that CRP induced insulin resistance thorough FCγRIIb 21,43. This finding suggests that CRP can act as an FCγRIIb ligand in mice. Moreover, murine cells express FcγRIII, which is more closely related to human FCγIIa; these two receptors show high sequence homology in their extracellular domains 22,23. To our knowledge, no study has reported FCγR expression in adipocytes. Here, we confirmed that adipocyte-differentiated 3T3-L1cells expressed both FCγRIIb and FcγRIII. Therefore, we focused on these two receptors and examined the effects of anti-CD16/CD32 Abs, which can act as FcγIII and FcγRII blockers, on CRP-induced MMP and TIMP-1 expression. Anti-CD16/CD34 Abs inhibited CRP-induced expression of MMPs, except MMP-11, and TIMP-1. These results suggested that CRP increased these MMPs and TIMP-1 via FcγRIIb and/or FcγIII in 3T3-L1 mouse adipocytes. Studies using adipocyte-differentiated 3T3-L1 cells reported that CRP altered the expression of adipokines such as adiponectin, leptin, or TNF-α via phosphatidylinositol-3 kinase, which is located in the downstream signaling pathway of FCγRs 26,27. However, they did not focus on the effects of CRP on MMP and TIMPs expression and did not examine FCγR expression in adipocytes. Our findings, i.e., the involvement of FCγRs in CRP-induced MMPs and TIMP-1 expression, represent the first report of this phenomenon, although we did not determine which signaling pathways located directly downstream of FCγRIIb or FCγRIII were closely related to CRP-induced MMP and TIMP-1 expression. Montero et al. previously indicated that CRP induced MMP-1 and MMP-10 expression in human umbilical vein endothelial cells (HUVECs) and human endothelial cells; however, HUVECs did not express FcγRII (CD32) or FcγRIII (CD16) 44. Here, the increase in MMP-11 expression in CRP-stimulated cells was higher in the presence than absence of anti-CD16/CD32 Abs. The findings of Montero et al. 44 and our results indicated that there is FcγR-independent induction of MMP expression. Anti-CD16/CD32 Abs inhibited CRP binding to FcγRII and FcγRIII; thus, CRP-induced MMP-11 expression via an FcγR-independent pathway might be facilitated.

MMP and TIMP expression is regulated by the MAPK pathway in many cell types, including fibroblast-like synoviocytes 28 and osteoblasts 29. Here, CRP had a stimulatory effect on ERK1/2 and p38 MAPK phosphorylation and had no effect on SAPK/JNK phosphorylation. These results suggest that CRP-induced phosphorylation of ERK1/2 and p38 MAPK might be involved in CRP-induced MMP and TIMP-1 expression. In addition, anti-CD16/CD34 Abs attenuated CRP-induced p38 phosphorylation, but did not affect CRP-induced ERK1/2 phosphorylation. These results suggest that CRP induced p38 MAPK phosphorylation via FcγRIIb and/or FcγIII, whereas CRP-induced ERK1/2 phosphorylation might be mediated via other CRP receptors, and not via FcγR. Further research is required to clarify the receptor and its downstream pathway that regulate the effects of CRP on MMP expression and the MAPK signaling pathway in adipocytes.

In conclusion, our results suggest that CRP facilitates ECM turnover in adipose tissue by increasing the production of multiple MMPs and TIMP-1 in adipocytes. Moreover, FcγRIIb and FcγRIII are involved in CRP-induced expression of MMPs and TIMP-1, and CRP-induced phosphorylation of p38, whereas the FcγR-independent pathway may regulate CRP-induced MMP-11 and CRP-induced ERK1/2 phosphorylation.