Introduction

Acne is a chronic inflammatory disease of pilosebaceous follicles that leads to the remodeling of the extracellular matrix in perifollicular areas resulting in scar formation. Acne inflammation has been known to be associated with several factors, including increased production of sebum, abnormal cornification of the pilosebaceous duct, and proliferation of commensal bacteria, especially Propionibacterium acnes (P. acnes).

P. acnes stimulates keratinocytes to produce proinflammatory cytokines such as IL-1, IL-1, IL-8, GM-CSF, and tumor necrosis factor- (TNF-) (Vowels et al., 1995; Chen et al., 2002; Basal et al., 2004; Schaller et al., 2005). Cytokine response is a T helper-1 type immune response (Bialecka et al., 2005), and is mediated by Toll-like receptor (TLR)-2 and TLR-4, the expression of which is increased by P. acnes (Jugeau et al., 2005; Kim, 2005; Nagy et al., 2005).

TNF-, one of the proinflammatory cytokines produced by P. acnes in acne inflammation, plays a critical role in modulating matrix metalloproteinase (MMP) activity in the dermis (Kourbeti and Boumpas, 2005; Numerof and Asadullah, 2006). TNF- stimulates activation of pro-matrix metalloproteinase (proMMP)-2 through the activated NF-B pathway in human dermal fibroblasts (hDF) (Han et al., 2001). In the skin, major sources of TNF- are keratinocytes and inflammatory cells that include mast cells, monocytes, and macrophages (Kock et al., 1990). By contrast, hDF is known to have a receptor for TNF-, indicating that hDF is a target rather than a source of TNF-. A previous report, however, showed that UV irradiation induces TNF- in hDF, suggesting that hDF could be a source of TNF- in response to external stimuli (de Kossode et al., 1995).

MMPs are a group of zinc-dependent endopeptidases that selectively degrade the components of various extracellular matrix as well as non-matrix proteins (Chakraborti et al., 2003). Thus, MMPs are implicated in the remodeling of the extracellular matrix in both physiological and pathological conditions such as wound healing, inflammation, and tumor metastasis (Kleiner and Stetler-Stevenson, 1999; Cook et al., 2000; Kim et al., 2005b). MMPs are synthesized in a variety of cells and secreted as a zymogen or proMMP, which becomes activated by either proteolytic cleavage of its amino-terminal domain or conformational change (Woessner, 1991; Massova et al., 1998). Membrane type MMPs (MT-MMPs) are expressed in many cells and are responsible for activating MMP-2 by proteolytic cleavage of proMMP-2. Membrane type 1 (MT1)-MMP is the most predominant form regulated by cytokines and growth factors, but MT2- or MT3-MMPs share the ability to activate proMMP-2 with MT1-MMP (Chakraborti et al., 2003).

The sebum of acne patients contains several kinds of MMPs, such as MMP-1, MMP-9, and MMP-13, which are secreted from keratinocytes and sebocytes (Papakonstantinou et al., 2005). Moreover, recent microarray analysis showed that MMP-1 and MMP-3 are upregulated in skin lesions from acne patients (Trivedi et al., 2006). However, there is no study on the role of MMP-2 from dermal fibroblasts in acne pathogenesis. As for acne treatment, tetracycline derivatives were revealed to have anti-inflammatory action by modulating proinflammatory cytokines as well as anti-MMP activity (Golub et al., 1991; Jain et al., 2002) besides their antibiotic role.

Inflammation and matrix remodeling takes place mainly in the dermis of acne lesions. However, cytokine response of hDF to P. acnes is yet to be elucidated. In this study, we showed that P. acnes enhanced proMMP-2 expression through TNF- activation in hDF. These results indicate that P. acnes may contribute to the pathogenesis of tissue remodeling in acne by modulating proMMP-2 expression.

Results

P. acnes induces proMMP-2 expression via NF-B signaling pathway

The proliferation of P. acnes is associated with inflammation of pilosebaceous follicles in the dermal skin, and MMPs play an important role in inflammatory matrix remodeling. In a previous report, it was noted that all components from P. acnes, including supernatant, crude cell lysate, and heat-killed P. acnes whole cells, could induce IL-8 and TNF- in human monocytes (Basal et al., 2004). In our preliminary experiments, we also found that the supernatant and pellet of P. acnes lysate, as well as whole P. acnes, could induce proMMP-2 expression in hDF (Figure 1a). From the result that the pellet of P. acnes lysate at a concentration of 1 mg ml^-1 had the strongest potential to induce proMMP-2, we used the pellet fraction as P. acnes extract for the following experiments. Next, we checked proMMP-2 and proMMP-9 expression by gelatin zymography, which was more sensitive than western immunoblot, in detecting gelatinase MMPs. In gelatin zymography, only proMMP-2 expression was detected as a gelatinase MMP, and proMMP-9 expression was not detected in hDF (Figure 1b). In contrast, proMMP-2 and proMMP-9 expressions were detected in fibrosarcoma as a control sample, which is known to produce both isotypes among MMPs (Figure 1b). With these results, the following experiments were focused on the modulation of proMMP-2 expression in hDF.

Figure 1.

Effect of P. acnes on proMMP-2 expression. (a) Western blot analysis for proMMP-2 expression was performed with hDF samples, which were treated with (lanes 2–4) the pellet of P. acnes lysates (0.1–5 mg ml^-1), (lanes 5–7) supernatant of P. acnes lysates (1–100 g ml^-1), and (lanes 9 and 10) whole P. acnes (10⁸–10⁹) for 24 hours. (Lanes 1 and 8) C indicates control hDF samples that were not treated with the P. acnes components. (b) Gelatin zymography to detect proMMP-9 and proMMP-2 expression was performed with cell extract from fibrosarcoma and hDF. (c, f) Western blot analysis and (d, g) gelatin zymography for proMMP-2 expression in culture media of hDF. (c, d) hDF was treated with P. acnes extract (1 mg ml^-1) for varying time periods or (f, g) with different extract concentrations of 0.1 and 1.0 mg ml^-1, for 24 hours, and 30 g of the extract was resolved on 10% SDS-PAGE. (e, h) Densitometric analysis of the expression level of proMMP-2 protein in the P. acnes-treated hDF. The graph represents the mean value and SD based on three independent experiments. ^*P<0.05, ^**P<0.01.

Full figure and legend (109K)

To test whether P. acnes affects the expression of MMPs in hDF, we treated hDF cells with P. acnes extract (1 mg ml^-1) and harvested the cells after various time periods. The treatment with P. acnes extract significantly raised the proMMP-2 level after 12 hours, and about 2.5-fold after 24 hours (Figure 1c and e). Moreover, proMMP-2 expression was detectable in a culture medium by gelatin zymography after 6 hours, and correlated with the level of proMMP-2 bands of western blot analysis (Figure 1d). To confirm the effect of P. acnes extract on the expression of proMMP-2, we treated the hDF cells with different concentrations of P. acnes extract (0.1 and 1.0 mg ml^-1). Western blot analysis (Figure 1f) and gelatin zymography (Figure 1g) of the samples obtained after 24 hours showed that P. acnes extract raised the proMMP-2 expression in a dose-dependent manner (Figure 1h).

It was previously reported that the activation of proMMP-2 is mediated by the NF-B signaling pathway through the induction of MT1-MMP (Han et al., 2001). To determine whether P. acnes activates the NF-B pathway in hDF, we performed electrophoretic mobility shift assay for NF-B-binding ability, p65 NF-B subunit translocation, and IB phosphorylation, which are shown in Figure 2. Treatment of cells with P. acnes extract induced nuclear NF-B binding, reaching its peak at 15 minutes, and then decreased gradually up to 60 minutes (Figure 2a). Immunocytochemical staining showed that p65 NF-B subunit translocation to the nucleus was induced at 15 minutes after the cells were treated with P. acnes extract (Figure 2b). P. acnes induced the decrease of total IB levels from 1 minute after treatment, reaching the lowest level at 30 minutes, and a gradual increase after 60 minutes (Figure 2c).

Figure 2.

P. acnes induces the activation of NF-B. hDF was treated with P. acnes extract (1 mg ml^-1) for varying time periods ranging from 1 minute to 60 minutes, and (a) electrophoretic mobility shift assay for NF-B-binding ability with nuclear extract and (c) western blot analysis of IB were performed. (b) Immunocytochemical staining for p65 NF-B subunit (right panel) was performed in cells that were treated with P. acnes extract (1 mg ml^-1, middle) or TNF- (100 pg ml^-1, lower), and non-treated hDF was used as a control (upper). The samples were also stained with anti-DAPI antibody for nuclear staining (left panel). Bar=100 m. (d) To inhibit NF-B activation, cells were pretreated with SN50 (100 g ml^-1) for 30 minutes, and then treated with P. acnes extract for 3 hours. RT-PCR for proMMP-2 and MT1-MMP was performed. DAPI, 4',6-diamidino-2-phenylindole.

Full figure and legend (129K)

Blockage of NF-B induces the decrease of proMMP-2 and MT1 expression

To verify the above results that P. acnes-induced proMMP-2 expression is mediated by the NF-B pathway, a blocking experiment with SN50, a cell-permeable inhibitory peptide of NF-B translocation, was performed. To inhibit NF-B activation, cells were pretreated with SN50 (100 g ml^-1) for 30 minutes and then further treated with P. acnes extract (1 mg ml^-1) for 3 hours. In reverse transcriptase (RT)-PCR, P. acnes treatment induced mRNA expression of proMMP-2 and MT1-MMP. The P. acnes-induced mRNA expression of proMMP-2 and MT1-MMP was suppressed by SN50 pretreatment (Figure 2d), indicating that the NF-B pathway plays a crucial role in proMMP-2 expression in P. acnes-treated hDF.

TNF- mediates the induction of proMMP-2 expression

P. acnes stimulates keratinocytes to produce various proinflammatory cytokines, including IL-1, IL-1, IL-8, GM-CSF, and TNF- (Kim, 2005). As TNF- is known to induce the activation of proMMP-2 (Köhler et al., 2001), we investigated whether P. acnes also induces TNF- production in hDF cells. The cells were treated with P. acnes extract (1 mg ml^-1), and TNF- mRNA expression level was evaluated by semiquantitative RT-PCR method. It was observed that the TNF- mRNA level began to increase after 1 hour, in a dose-dependent manner, and reached the maximum level after 3 hours (Figure 3a and b). In the same condition, when we measured TNF- protein expression by ELISA, the TNF- protein level began to increase after 3 hours and reached the maximum level after 6 hours (Figure 3c). The result indicates that P. acnes stimulates hDF to produce TNF-, and that TNF- can be a good candidate for cytokine to modulate proMMP-2 expression in P. acnes-treated hDF. To test our assumption, we compared the modulation pattern of proMMP-2 expression in hDF by P. acnes extract or TNF- treatment. When cells were treated with P. acnes extract (1 mg ml^-1) or TNF- (100 pg ml^-1) for 24 hours, proMMP-2 mRNA expression increased in both cases compared with that of the control (Figure 4a and b). Next, we evaluated the modulation of proMMP-2 protein expression following a treatment of hDF cells with TNF-. Western blot analysis of the extracts of hDF cells treated with TNF- (100 pg ml^-1) showed an increase of the proMMP-2 level after 12 hours (Figure 4c). When proMMP-2 expression in culture medium was assayed by gelatin zymography, an increase of proMMP-2 expression was observable within 6 hours (Figure 4d). Treatment with 10 or 100 ng ml^-1 of TNF- for 24 hours appeared to upregulate proMMP-2 expression in a dose-dependent manner (Figure 4f and g), indicating that TNF- modulates the proMMP-2 expression in the same manner as P. acnes does in hDF cells. To confirm the above results, a blocking experiment with anti-TNF- antibody was performed. Cells were pretreated overnight with a blocking anti-TNF- antibody (MAB 1021, Chemicon, Temecula, CA) at a concentration of 100–1,000 pg ml^-1, and then treated with P. acnes extract for 24 hours. The anti-TNF- antibody could suppress P. acnes-induced proMMP-2 expression in a dose-dependent manner (Figure 4i). In immunocytochemical staining, the ubiquitous expression of MMP-2 was detected in the cytoplasm of non-treated control hDF (Figure 5b). When cells were treated with P. acnes extract (1 mg ml^-1) or TNF- (100 pg ml^-1) for 24 hours, the cytoplasmic expression of MMP-2 was further accentuated in both cases (Figure 5c and d).

Figure 3.

P. acnes induces TNF- expression in a time- and dose-dependent manner. Semiquantitative RT-PCR of TNF- using total RNA from the hDF treated (a) with P. acnes extract (1 mg ml^-1) for varying time periods ranging from 30 minutes to 6 hours, or (b) with various concentrations of the extract between 0.1 and 2.0 mg ml^-1 for 3 hours. GAPDH was used as an internal standard. (c) Time-course change of TNF- protein levels, which was measured by ELISA, in the supernatant of P. acnes-treated (1 mg ml^-1) or non-treated control (0 hour) hDF. ^*P<0.05.

Full figure and legend (87K)

Figure 4.

TNF- induces proMMP-2 expression. Semiquantitative RT-PCR of proMMP-2 using total RNA from hDF treated (a) with P. acnes extract (1 mg ml^-1) or (b) TNF- (100 pg ml^-1) for 3–12 hours. Western blot analysis for proMMP-2 using the cell extract of the hDF treated either (c) with TNF- (100 pg ml^-1) for 1–24 hours or (f) with TNF- (10 or 100 pg ml^-1) for 24 hours. (d, g) Gelatin zymography for proMMP-2 expression was performed with culture medium. (e, h) Densitometric analysis for proMMP-2 expression is based on three independent experiments. The graph represents the mean value and SD. ^*P<0.05, ^**P<0.01. (i) Cells were pretreated with a blocking anti-TNF- antibody at concentration of 100–1,000 pg ml^-1 overnight, and then treated with P. acnes extract (1 mg ml^-1) for 24 hours. Gelatin zymography for proMMP-2 expression was performed with culture media.

Full figure and legend (139K)

Figure 5.

Immunocytochemistry for proMMP-2 expression in hDF. Subconfluent hDF cells were treated either with P. acnes extract (1 mg ml^-1) or with TNF- (100 pg ml^-1) for 24 hours. Immunostaining was performed with anti-MMP-2 antibody. DAPI-stained DNA was used to identify nuclei of hDF. (a) DAPI control, (b) untreated hDF, (c) P. acnes-treated hDF, (d) TNF--treated hDF. Bar=100 m. DAPI, 4',6-diamidino-2-phenylindole.

Full figure and legend (112K)

Doxycycline inhibits P. acnes- or TNF--induced proMMP-2 expression

Doxycycline has been known to inhibit MMP-2 expression in human tissues such as brain and heart (Shen et al., 2006; Burggraf et al., 2007). The reports prompted us to evaluate whether doxycycline could inhibit proMMP-2 expression in hDF from the skin, as the agent is one of the standard therapeutic drugs to treat acne. Subconfluent cells were pretreated with doxycycline (0.1–20 g ml^-1) overnight, which was in the range of therapeutic concentration of acne in humans (Dotevall and Hagberg, 1989), prior to the treatment with P. acnes (1 mg ml^-1) or TNF- (100 pg ml^-1) for 24 hours. Western blot analysis for proMMP-2 showed that doxycycline suppressed the expression of proMMP-2, induced either by P. acnes or TNF-, in a dose-dependent manner (Figure 6).

Figure 6.

Effect of doxycycline on proMMP-2 expression induced by P. acnes or TNF-. Subconfluent hDF cells were pretreated with doxycycline (0.1–20 g ml^-1) overnight, and then the cells were further treated either (a) with P. acnes extract (1 mg ml^-1) or (b) with TNF- (100 pg ml^-1) in the presence of doxycycline for 24 hours. The levels of proMMP-2 were assessed by western blot analysis.

Full figure and legend (50K)

MMP-2 is expressed in the epidermis and dermis

To confirm our results, MMP-2 expression was evaluated in the human samples from acne lesion and normal skin by immunohistochemical staining with the anti-MMP-2 antibody, which could recognize reduced and native MMP-2. In normal skin, MMP-2 expression was found in the epidermis, in which the positive signal was prominent in the basal layer and gradually attenuated in the upper epidermal layers, supporting the previous report by Gunduz et al. (2006). In the dermis, MMP-2 expression was weakly detected from dermal connective tissue (Figure 7a). In acne lesion, accentuated MMP-2 expression was detected in dermal fibroblasts as well as infiltrated inflammatory cells (Figure 7b). The positive MMP-2 signals were not detected in the negative control sample, in which primary anti-MMP-2 antibody was omitted during an immunostaining procedure (Figure 7c).

Figure 7.

Immunohistochemical staining for proMMP-2 expression in the skin. Human skin samples from normal control and acne lesions were stained. Frozen sections of skin samples were stained with anti-MMP-2 antibody. MMP-2 expression in (a) normal skin, (b) acne skin, and (c) negative control sample in which the anti-MMP-2 antibody was omitted during immunostaining. Bar=100 m.

Full figure and legend (96K)

Discussion

P. acnes, which is present in sebum-rich skin, has been implicated in acne pathogenesis. Although the initial event of acne starts from hair follicles, substances released from pilosebaceous units are thought to eventually cause inflammation in the dermis. However, the specific role of P. acnes in this process has not been identified. P. acnes is distinct from other Gram-positive bacteria, partly due to its unique features of cell wall or outer envelope having phosphatidylinositol and peptidoglycan, which contribute to induction of an inflammatory response via TLR-2 or TLR-4(Kim, 2005). In this study, we showed that P. acnes extract causes the increase of TNF- expression in hDF cells, which in turn induces proMMP-2 expression. We could not detect proMMP-9 expression in hDF. As for the expression of gelatinase MMPs, dermal fibroblasts were reported to express only MMP-2 activity (Han et al., 2001; Sawicki et al., 2005), as shown in our study, but other studies have been reported to express both MMP-2 and MMP-9 (Kobayashi et al., 2003). This finding indicates that P. acnes is not only a commensal bacterium but also plays an important role in the pathogenesis of acne inflammation in the dermis.

In fact, P. acnes is not supposed to contact directly with hDF in vivo, but hDF reacted directly with P. acnes extract in our experiments due to technical limitations of in vitro experiments. Therefore, we speculate that P. acnes produces one of the proinflammatory cytokines to activate the NF-B pathway, which is a well-known pathway to mediate various inflammatory conditions, including acne. Our results that P. acnes induces TNF- mRNA/protein expression in hDF, and that TNF- treatment induces proMMP-2 expression in a similar way to P. acnes, support our assumption that P. acnes-induced proMMP-2 expression in our experiments is mediated by TNF- as an in vivo system. The P. acnes-induced NF-B activation was confirmed from western immunoblot for IB protein expression and NF-B gel shift.

A number of proinflammatory cytokines, including IL-1, IL-1, IL-8, GM-CSF, and TNF-, have been implicated in the inflammatory process of acne (Ingham et al., 1992; Graham et al., 2004; Kim, 2005). TNF-, in particular, is regarded as a crucial cytokine in acne inflammation by involving a coordinated expression of adhesion molecules (Vowels et al., 1995; Chen et al., 2002; Basal et al., 2004). P. acnes and its derivatives induce TNF- production in epidermal keratinocytes or dermal inflammatory cells, such as lymphocytes, monocytes, mast cell, and macrophages (Graham et al., 2004; Marta Guarna et al., 2006). However, hDF is known to be a target of TNF- that is secreted from other cells. Interestingly, dermal fibroblasts may be another source of TNF-, in response to external stimuli as reported by de Kossode et al. (1995). They observed that chloramphenicol acetyltransferase reporter activity for TNF gene was markedly increased by UVB irradiation in the dermis and dermal fibroblasts. Our result showing that P. acnes upregulates TNF- expression in a time-dependent manner in hDF illustrates another case in which TNF- can be produced by external stimuli in hDF. Since inflammatory reaction takes place in the dermis rather than the epidermis or hair follicles, these findings suggest that hDF might be a source as well as a target of TNF- in response to P. acnes.

The balanced control of MMP expression and MMP activity is essential to prevent permanent dermal damage or scarring after acne inflammation. The levels of MMP-1 and MMP-3 were upregulated in skin samples of acne lesion (Trivedi et al., 2006). In this study, we have shown that MMP-2 is a newly identified enzyme that is induced by P. acnes in hDF. Both MMP-2 activity and MMP expression in hDF were raised in coculture with keratinocytes (Sawicki et al., 2005), suggesting the importance of interaction among keratinocyte, hDF, and sebocyte for modulating MMP-2 in acne skin. Although both proMMP-2 and proMMP-9 are known to be secreted from hDF (Kobayashi et al., 2003), the expression and activity of proMMP-9 were not increased by P. acnes, indicating that P. acnes induces MMPs expression in a type-specific manner, probably through TLRs (Jugeau et al., 2005). Moreover, TNF- induces proMMP-2 expression in various cells, such as nasal fibroblasts, retinal pigment epithelial cells, and cardiac fibroblasts (Siwik et al., 2000; Eichler et al., 2002; Kanai et al., 2004). By contrast, TNF- had no effect on MMP-2 expression in glioma, suggesting the cell-specific regulation of MMP-2 expression by TNF- (Esteve et al., 1998; Qin et al., 1998).

Tetracycline derivatives have been reported to have an anti-acne effect due to anti-MMP action besides their antibiotic activity to kill P. acnes (Golub et al., 1991; Pirila et al., 2001; Jain et al., 2002; Ahuja, 2003; Kim et al., 2005a). Our results showed that doxycycline has an inhibitory effect on proMMP-2 expression in hDF, providing a rationale for using tetracycline derivatives in acne treatment, and helping to understand the role of P. acnes in eliciting dermal inflammation and tissue remodeling in acne skin.

Our result that proMMP-2 is expressed in fibroblasts besides keratinocytes (Figure 7) implies the complexity of cell-to-cell interaction in MMP system in vivo. Interestingly, fibroblasts further activated MMP-2 expression and gelatinase activity in the presence of keratinocytes (Sawicki et al., 2005). Further studies are warranted to evaluate the role of P. acnes in the modulation of MMP-2 via TNF- in a coculture system of epidermal keratinocytes and dermal fibroblasts. For clinical application, the development of new anti-MMP agents, which are effective and safe in blocking MMP activities in skin cells, might open a new field in acne treatment.

Materials and Methods

Primary culture of human dermal fibroblasts

Human dermal tissue was obtained from foreskin. Informed consent was obtained from donors according to the guidelines of the local ethics committee and the Declaration of Helsinki Principles. The skin was cut into small pieces and then treated with 2.4 U ml^-1 dispase (Boehringer Mannheim, Ingelheim, Germany) for 30 minutes at 37 °C. The dermis was separated from the epidermis, and hDF was cultured in DMEM, supplemented with 10% fetal bovine serum and antibiotics (100 U ml^-1 penicillin and 100 g ml^-1 streptomycin), in a 5% CO₂ incubator. The third to seventh passages of hDF were used for all the experiments.

Preparation of P. acnes extract

P. acnes (ATCC 11828) was grown in BBL actinomyces broth (Difco, Sparks, MD) at 37 °C for 24 hours under anaerobic condition (85% N₂, 10% H₂, 5% CO₂). The strain was kept in 20% glycerol stock at -80 °C. In each experiment, the strain was propagated twice in the appropriate broth. P. acnes was harvested by centrifugation (8,000 g, 10 minutes) and resuspended in phosphate-buffered saline (PBS). Cells were sonicated on ice for 30 seconds, and then centrifuged (14,000 g, 10 minutes). The pellet was dissolved in PBS to make 1 mg ml^-1 of P. acnes extract and stored at -80 °C until use.

Gelatin zymography for proMMP-2 expression

Gelatin zymography was performed to measure gelatinase expression in a culture supernatant of hDF. A 20–25 l portion of supernatant samples (containing 2 g of protein) was subjected to 8% SDS-PAGE containing 0.25% gelatin. The gel was washed with renaturation buffer (2.5% Triton X-100) for 1 hour at room temperature (RT), and then incubated overnight in the development buffer (50 mM Tris–HCl (pH 7.5), 200 mM NaCl, 5 mM CaCl₂, and 0.02% Brij-35) at 37 °C. Subsequently, the gel was stained with 2.5% Coomassie blue R250 in 50% methanol and 10% glacial acetic acid for 30 minutes, followed by destaining in 30% methanol and 10% glacial acetic acid. ProMMP-2 expression was visualized as clear bands against a blue background of stained gelatin.

Western blot analysis

Whole-cell extract was prepared by sonication in a buffer (1% Triton X-100, 20 mM Tris–HCl (pH 7.4), 150 mM NaCl, 5% glycerol, 1 mM EDTA (pH 8.0), and protease inhibitor cocktail). After centrifugation (12,000 g, 15 minutes at 4 °C), the protein concentration of the supernatant was measured by using a Coomassie plus protein assay reagent kit (Pierce, Rockford, IL). The cell extract (30 g) in an SDS sample buffer (50 mM Tris–HCl (pH 6.8), 10% glycerol, 2% SDS, 0.1% bromophenol blue, and 5% -mercaptoethanol) was separated in 10% SDS-PAGE and transferred onto a polyvinylidene difluoride membrane (0.45 m, Millipore, Billerica, MA). After incubation with blocking solution (5% skim milk in Tris-buffered saline with 0.1% Tween-20) for 1 hour at RT, the protein bands were probed with rabbit anti-MMP-2 (1:1,000, AB-809; Chemicon), rabbit anti-MMP-9 (1:1,000, AB-19047; Chemicon), and mouse monoclonal -actin (1:1,000, ab-6276; Abcam, Cambridgeshire, UK) for 4 hours at RT, or with polyclonal anti-IB (1:1,000, SC-9242; Cell Signaling, Danvers, MA) for overnight at 4 °C. After incubation with HRP-conjugated goat anti-rabbit IgG (1:3,000) or anti-mouse IgG (1:2,000), immune complexes were visualized using the ECL kit (Santa Cruz Biotechnology, Santa Cruz, CA).

Electrophoretic mobility shift assay

Binding activities of NF-B were examined by electrophoretic mobility shift assay. Nuclear extract of cultured hDF was prepared as described by Yan et al. (2002) with modification. Cells were scraped from the culture, suspended in ice-cold buffer A (10 mM HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 7.9), 10 mM KCl, 0.1 mM EDTA (pH 8.0), 0.1 mM EGTA, 1 mM dithiothreitol, and 0.5 mM PMSF (phenylmethylsulfonyl fluoride)), and incubated on ice for 15 minutes. Then 10% of Nonidet P-40 was added and nuclei were pelleted down by centrifugation at 14,000 r.p.m. for 5 minutes at 4 °C. The pellet was resuspended in ice-cold buffer B (20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EGTA, 1 mM EDTA (pH 8.0), 1 mM dithiothreitol, and 1 mM PMSF) for 30 minutes on ice with shaking. Nuclear extract was collected by centrifugation at 14,000 r.p.m. for 5 minutes at 4 °C.

The double-stranded oligonucleotide probe containing the consensus binding sequence for NF-B (5'-AGTTGAGGGGACTTTCCCAGGC-3') (Promega, Madison, WI) was end-labeled with (-³²P) dATP using T4-polynucleotide kinase buffer for 20 minutes at RT. A 5 g portion of nuclear extract with 1 g of polydeoxyinosinic deoxycytidylic acid, 3 l of 5 DNA binding buffer (10 mM Tris-HCL (pH 7.3), 50 mM NaCl, 10 mM MgCl₂ 1 mM DTT, 25% glycerol), and labeled probe was incubated for 20 minutes at RT. After addition of a gel-loading buffer, the mixture was electrophoresed on 6% non-denaturing polyacrylamide gel in Tris-buffered EDTA (pH 8.0). The gel was vacuum-dried for 1 hour and exposed to X-ray film for 3 days.

Semiquantitative RT-PCR

Total RNA was isolated by using an RNeasy mini kit (Qiagen, Fremont, CA) according to the manufacturer's instructions. cDNA was generated from 1 g of total RNA using the Omniscript RT kit (Qiagen). For semiquantitative PCR, PCR-premixture kit (ELPIS, Taejeon, Korea) was used. PCR experiments for proMMP-2 (Asano et al., 2004), MT-MMP (Nakada et al., 1999), TNF- (Kuwahara et al., 2005), and GAPDH (glyceraldehyde-3-phosphate dehydrogenase) (Minamiya et al., 2005) were conducted by methods previously described. PCR products were resolved by 1.5% agarose gel electrophoresis and visualized by ethidium bromide staining.

ELISA

Medium supernatants treated with P. acnes extract (1 mg ml^-1) were checked by ELISA using TNF--coated 96-well ELISA plates (Quantikine, R&D systems, Minneapolis, MN), as recommended by the manufacturer. Briefly, after adding 25 l assay diluent RD1F to each well, 100 l supernatant was added to each well and the plates were incubated for 2 hours at RT. After washing four times with 400 l wash buffer, 200 l of substrate solution were added to each well with protection from light. After 20 minutes, the reaction was stopped by adding 50 l stop solution to each well, and optical density (OD) at 450 nm was measured using a microplate reader (Molecular Device, Hercules, CA). Each time group was assayed in triplicate.

Immunohistochemical staining

Frozen sections of acne samples of 6–8 m thickness were placed in 3% hydrogen peroxide solution for 5 minutes at RT to block endogenous peroxidase activity. The acetone-fixed frozen sections were stained with the anti-MMP-2 antibody (1:100); the antibody was applied for 40 minutes at RT. After rinsing in PBS, the sections were stained with an LSAB kit (Dako, Carpinteria, CA) that employed a biotinylated secondary antibody and streptavidin–biotin conjugate. After staining with 3-amino-9-ethylcarbazole substrate-chromogen, the sections were counterstained briefly in Mayer's hematoxylin. As a negative control for immunostaining, PBS was used in place of the primary antibodies and then the secondary antibody was incubated under the same conditions as described above. Informed consent was obtained from acne patients for using their biopsy samples.

Immunofluorescence

For immunocytochemical staining, hDF was seeded in each well of the slide chamber at a density of 1 10³ cells per well. hDF was treated with P. acnes (1 mg ml^-1) or TNF- (100 pg ml^-1) for 24 hours. After washing with PBS, the cells were fixed with 4% paraformaldehyde in PBS for 10 minutes and permeabilized with PBS-T (PBS containing 0.5% Triton X-100) for 15 minutes. The cells were incubated with PBS-T containing 1% BSA for 1 hour and stained with the anti-MMP-2 antibody (1:100) for 3 hours, or with phospho-NF-B p65 (Ser536) antibody (Cell Signaling, 1:100) overnight at 4 °C, followed by Alexa Fluor 488 goat anti-rabbit IgG (H+L) (1:100, A11034; Invitrogen, Carlsbad, CA) for 1 hour. Finally, the cells were visualized and photographed.

Statistical analysis

All experiments were performed at least three times, and the statistical significance of the data was determined by Student's t-test (meanSD), based on the P-values of 0.05.

Conflict of Interest

The authors state no conflict of interest.

References

1. Ahuja TS (2003) Doxycycline decreases proteinuria in glomerulonephritis. Am J Kidney Dis 42:376–380 | Article | PubMed |
2. Asano K, Kanai KI, Suzaki H (2004) Suppressive activity of fexofenadine hydrochloride on metalloproteinase production from nasal fibroblast in vitro. Clin Exp Allergy 34:1890–1898 | Article | PubMed | ChemPort |
3. Basal E, Jain A, Kaushal GP (2004) Antibody response to crude cell lysate of Propionibacterium acnes and induction of pro-inflammatory cytokines in patients with acne and normal healthy subjects. J Microbiol 42:117–125 | PubMed | ChemPort |
4. Bialecka A, Mak M, Biedron R, Bobek M, Kasprowicz A, Marcinkiewicz J (2005) Different pro-inflammatory and immunogenic potentials of Propionibacterium acnes and Staphylococcus epidermidis: implications for chronic inflammatory acne. Arch Immunol Ther Exp 53:79–85 | ChemPort |
5. Burggraf D, Trinkl A, Dichgans M, Hamann GF (2007) Doxycycline inhibits MMPs via modulation of plasminogen activators in focal cerebral ischemia. Neurobiol Dis 25:506–513 | Article | PubMed | ChemPort |
6. Chakraborti S, Mandal M, Das S, Mandal A, Chakraborti T (2003) Regulation of matrix metalloproteinases: an overview. Mol Cell Biochem 253:269–285 | Article | PubMed | ISI | ChemPort |
7. Chen Q, Koga T, Uchi H, Hara H, Terao H, Moroi Y et al. (2002) Propionibacterium acnes-induced IL-8 production may be mediated by NFB activation in human monocytes. J Dermatol Sci 29:97–103 | Article | PubMed | ISI | ChemPort |
8. Cook H, Davies KJ, Harding KG, Thomas DW (2000) Defective extracellular matrix reorganization by chronic wound fibroblasts is associated with alterations in TIMP-1, TIMP-2, and MMP-2 activity. J Invest Dermatol 115:225–233 | Article | PubMed | ISI | ChemPort |
9. de Kossode S, Cruz PD Jr, Dougherty I, Thompson P, Silva-Valdez M, Beutler B (1995) Expression of the tumor necrosis factor gene by dermal fibroblasts in response to ultraviolet irradiation or lipopolysaccharide. J Invest Dermatol 104:318–322 | Article | PubMed | ChemPort |
10. Dotevall L, Hagberg L (1989) Penetration of doxycycline into cerebrospinal fluid in patients treated for suspected Lyme neuroborreliosis. Antimicrob Agents Chemother 33:1078–1080 | PubMed | ChemPort |
11. Eichler W, Friedrichs U, Thies A, Tratz C, Wiedemann P (2002) Modulation of matrix metalloproteinase and TIMP-1 expression by cytokines in human RPE cells. Invest Ophthalmol Vis Sci 43:2767–2773 | PubMed |
12. Esteve PO, Tremblay P, Houde M, St-Pierre Y, Mandeville R (1998) In vitro expression of MMP-2 and MMP-9 in glioma cells following exposure to inflammatory mediators. Biochim Biophys Acta 1403:85–96 | Article | PubMed | ChemPort |
13. Golub LM, Ramamurthy NS, McNamara TF, Greenwald RA, Rifkin BR (1991) Tetracyclines inhibit connective tissue breakdown: new therapeutic implications for an old family of drugs. Crit Rev Oral Biol Med 2:297–321 | PubMed | ISI | ChemPort |
14. Graham GM, Farrar MD, Cruse-Sawyer JE, Holland KT, Ingham E (2004) Proinflammatory cytokine production by human keratinocytes stimulated with Propionibacterium acnes and P. acnes GroEL. Br J Dermatol 150:421–428 | Article | PubMed | ISI | ChemPort |
15. Gunduz K, Demireli P, Inanir I, Nese N (2006) Expression of matrix metalloproteinases (MMP-2, MMP-3, and MMP-9) and fibronectin in lichen planus. J Cutan Pathol 33:545–550 | Article | PubMed |
16. Han YP, Tuan TL, Wu H, Hughes M, Garner WL (2001) TNF- stimulates activation of pro-MMP2 in human skin through NF-B mediated induction of MT1-MMP. J Cell Sci 114:131–139 | PubMed | ISI | ChemPort |
17. Ingham E, Eady EA, Goodwin CE, Cove JH, Cunliffe WJ (1992) Pro-inflammatory levels of interleukin-1-like bioactivity are present in the majority of open comedones in acne vulgaris. J Invest Dermatol 98:895–901 | Article | PubMed | ISI | ChemPort |
18. Jain A, Sangal L, Basal E, Kaushal GP, Agarwal SK (2002) Anti-inflammatory effects of erythromycin and tetracycline on Propionibacterium acnes induced production of chemotactic factors and reactive oxygen species by human neutrophils. Dermatol Online J 8:2 | PubMed | ChemPort |
19. Jugeau S, Tenaud I, Knol AC, Jarrousse V, Quereux G, Khammari A et al. (2005) Induction of Toll-like receptors by Propionibacterium acnes. Br J Dermatol 153:1105–1113 | Article | PubMed | ChemPort |
20. Kanai K, Asano K, Hisamitsu T, Suzaki H (2004) Suppression of matrix metalloproteinase production from nasal fibroblasts by macrolide antibiotics in vitro. Eur Respir J 23:671–678 | Article | PubMed | ChemPort |
21. Kim J (2005) Review of the innate immune response in acne vulgaris: activation of Toll-like receptor 2 in acne triggers inflammatory cytokine responses. Dermatology 211:193–198 | Article | PubMed | ChemPort |
22. Kim HS, Luo L, Pflugfelder SC, Li DQ (2005a) Doxycycline inhibits TGF-1-induced MMP-9 via Smad and MAPK pathways in human corneal epithelial cells. Invest Ophthalmol Vis Sci 46:840–848 | Article | PubMed | ISI |
23. Kim WU, Min SY, Cho ML, Hong KH, Shin YJ, Park SH et al. (2005b) Elevated matrix metalloproteinase-9 in patients with systemic sclerosis. Arthritis Res Ther 7:R71–R79 | Article | ChemPort |
24. Kleiner DE, Stetler-Stevenson WG (1999) Matrix metalloproteinases and metastasis. Cancer Chemother Pharmacol 43:S42–S51 | Article | PubMed | ISI | ChemPort |
25. Kobayashi T, Hattori S, Shinkai H (2003) Matrix metalloproteinases-2 and -9 are secreted from human fibroblasts. Acta Derm Venereol 83:105–107 | Article | PubMed | ISI | ChemPort |
26. Kock A, Schwarz T, Kirnbauer R, Urbanski A, Perry P, Ansel JC et al. (1990) Human keratinocytes are a source for tumor necrosis factor : evidence for synthesis and release upon stimulation with endotoxin or ultraviolet light. J Exp Med 17:1609–1614 | Article |
27. Köhler HB, Knop J, Martin M, de Bruin A, Huchzermeyer B, Lehmann H et al. (2001) Involvement of reactive oxygen species in TNF- mediated activation of the transcriptional factor NF-B in canine dermal fibroblasts. Vet Immunol Immmnopathol 71:125–142
28. Kourbeti IS, Boumpas DT (2005) Biological therapies of autoimmune diseases. Curr Drug Targets Inflamm Allergy 4:41–46 | Article | PubMed | ChemPort |
29. Kuwahara K, Kitazawa T, Kitagaki H, Tsukamoto T, Kikuchi M (2005) Nadifloxacin, an antiacne quinolone antimicrobial, inhibits the production of proinflammatory cytokines by human peripheral blood mononuclear cells and normal human keratinocytes. J Dermatol Sci 38:47–55 | Article | PubMed | ChemPort |
30. Marta Guarna M, Coulson R, Rubinchik E (2006) Anti-inflammatory activity of cationic peptides: application to the treatment of acne vulgaris. FEMS Microbial Lett 257:1–6 | Article | ChemPort |
31. Massova I, Kotra LP, Fridman R, Mobashery S (1998) Matrix metalloproteinases: structure evolution and diversification. FASEB J 12:1075–1095 | PubMed | ISI | ChemPort |
32. Minamiya Y, Nakagawa T, Saito H, Matsuzaki I (2005) Increased expression of myosin light chain kinase mRNA is related to metastasis in non-small cell lung cancer. Tumour Biol 26:153–157 | Article | PubMed | ChemPort |
33. Nagy I, Pivarcsi A, Koreck A, Szell M, Urban E, Kemeny L (2005) Distinct strains of Propionibacterium acnes induce selective human -defensin-2 and interleukin-8 expression in human keratinocyte through Toll-like receptors. J Invest Dermatol 124:931–938 | Article | PubMed | ISI | ChemPort |
34. Nakada M, Nakamura H, Ikeda E, Fujimoto N, Yamashita J, Sato H et al. (1999) Expression and tissue localization of membrane-type 1,2, and 3 matrix metalloproteinases in human astrocytic tumors. Am J Pathol 154:417–428 | PubMed | ISI | ChemPort |
35. Numerof RP, Asadullah K (2006) Cytokine and anti-cytokine therapies for psoriasis and atopic dermatitis. BioDrugs 20:93–103 | Article | PubMed | ChemPort |
36. Papakonstantinou E, Aletras AJ, Glass E, Tsogas P, Dionyssopoulos A, Adjaye J et al. (2005) Matrix metalloproteinases of epithelial origin in facial sebum of patients with acne and their regulation by isotretinoin. J Invest Dermatol 125:673–684 | Article | PubMed | ChemPort |
37. Pirila E, Ramamurthy N, Maisi P, Maisi P, McClain S, Kucine A et al. (2001) Wound healing in ovariectomized rats: effects of chemically modified tetracycline (CMT-8) and estrogen on matrix metalloproteinases-8, -13 and type I collagen expression. Curr Med Chem 8:281–294 | PubMed | ISI | ChemPort |
38. Qin H, Moellinger JD, Wells A, Windsor LJ, Sun Y, Benveniste EN (1998) Transcriptional suppression of matrix metalloproteinase-2 gene expression in human astroglioma cells by TNF- and IFN-. J Immunol 161:6664–6673 | PubMed | ISI | ChemPort |
39. Sawicki G, Marcoux Y, Sarkhosh K, Tredget EE, Ghahary A (2005) Interaction of keratinocytes and fibroblasts modulates the expression of matrix metalloproteinases-2 and -9 and their inhibitors. Mol Cell Biochem 269:209–216 | Article | PubMed | ChemPort |
40. Schaller M, Loewenstein M, Borelli C, Jacob K, Vogeser M, Burgdorf WH et al. (2005) Induction of chemoattractive proinflammatory cytokine response after stimulation of keratinocytes with Propionibacterium acnes and coproporphyrin III. Br J Dermtol 153:66–71 | Article | ChemPort |
41. Shen J, O'Brien D, Xu Y (2006) Matrix metalloproteinase-2 contributes to tumor necrosis factor alpha induced apoptosis in cultured rat cardiac myocytes. Biochem Biophys Res Commun 347:1011–1020 | Article | PubMed | ChemPort |
42. Siwik DA, Chang DL, Colucci WS (2000) Interleukin-1 and tumor necrosis factor- decrease collagen synthesis and increase matrix metalloproteinase activity in cardiac fibroblasts in vitro. Circ Res 86:1259–1265 | PubMed | ISI | ChemPort |
43. Trivedi NR, Gilliland KL, Zhao W, Liu W, Thiboutot DM (2006) Gene array expression profiling in acne lesions reveals marked upregulation of genes involved in inflammation and matrix remodeling. J Invest Dermatol 126:1071–1079 | Article | PubMed | ChemPort |
44. Vowels BR, Yang S, Leyden JJ (1995) Induction of proinflammtory cytokines by a soluble factor of Propionibacterium acnes: implications for chronic inflammatory acne. Infect Immun 63:3158–3165 | PubMed | ISI | ChemPort |
45. Woessner JF Jr (1991) Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASAB J 5:2145–2154 | ChemPort |
46. Yan X, Wu Xiao C, Sun M, Tsang BK, Gibb W (2002) Nuclear factor kappa B activation and regulation of cyclooxygenase type-2 expression in human amnion mesenchymal cells by interleukin-1. Biol Reprod 66:1667–1671 | Article | PubMed | ChemPort |

Acknowledgments

We thank Dr Y.D. Kim for critical comments on this paper. This work was supported by the Brain Korea 21 Project, Center for Biomedical Human Resources at Chonnam National University (2007), by the Korea Science and Engineering Foundation (R11-2002-097-07003-0 and R01-2005-000-10364-0), and by the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (AD50174).