In Vitro and In Vivo Screening of Wild Bitter Melon Leaf for Anti-Inflammatory Activity against Cutibacterium acnes

Abstract

Cutibacterium acnes (formerly Propionibacterium acnes) is a key pathogen involved in the development and progression of acne inflammation. The numerous bioactive properties of wild bitter melon (WBM) leaf extract and their medicinal applications have been recognized for many years. In this study, we examined the suppressive effect of a methanolic extract (ME) of WBM leaf and fractionated components thereof on live C. acnes-induced in vitro and in vivo inflammation. Following methanol extraction of WBM leaves, we confirmed anti-inflammatory properties of ME in C. acnes-treated human THP-1 monocyte and mouse ear edema models. Using a bioassay-monitored isolation approach and a combination of liquid-liquid extraction and column chromatography, the ME was then separated into n-hexane, ethyl acetate, n-butanol and water-soluble fractions. The hexane fraction exerted the most potent anti-inflammatory effect, suppressing C. acnes-induced interleukin-8 (IL-8) production by 36%. The ethanol-soluble fraction (ESF), which was separated from the n-hexane fraction, significantly inhibited C. acnes-induced activation of mitogen-activated protein kinase (MAPK)-mediated cellular IL-8 production. Similarly, the ESF protected against C. acnes-stimulated mouse ear swelling, as measured by ear thickness (20%) and biopsy weight (23%). Twenty-four compounds in the ESF were identified using gas chromatograph-mass spectrum (GC/MS) analysis. Using co-cultures of C. acnes and THP-1 cells, β-ionone, a compound of the ESF, reduced the production of IL-1β and IL-8 up to 40% and 18%, respectively. β-ionone also reduced epidermal microabscess, neutrophilic infiltration and IL-1β expression in mouse ear. We also found evidence of the presence of anti-inflammatory substances in an unfractionated phenolic extract of WBM leaf, and demonstrated that the ESF is a potential anti-inflammatory agent for modulating in vitro and in vivo C. acnes-induced inflammatory responses.

Keywords: Cutibacterium acnes; anti-inflammation; wild bitter melon (WBM); β-ionone.

Figure 1

Effects of the crude methanolic extract (ME) of wild bitter melon (WBM) leaf on C. acnes-induced IL-8 production in vitro and on ear swelling in mice. THP-1 cells were cultured with DMSO as the negative control, or co-incubated with C. acnes (M.O.I. = 75) and different concentrations (25, 50 or 100 μg/mL) of ME for 24 h. The culture supernatants were subsequently collected and analyzed for the IL-8 levels (a). In the ear edema mouse model, ME (0.25 or 0.5 mg/site) or vehicle (phosphate-buffered saline; PBS) was intradermally injected, immediately followed by the C. acnes injection. The inhibitory effects of ME on C. acnes-induced ear swelling were evaluated by measuring the ear thickness and ear biopsy weight (b). Each value shows the mean ± SD. Values with different symbols are significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*) and p < 0.001 (***).

Figure 2

Effects of four partitioned fractions from ME of WBM leaf on C. acnes-induced IL-8 production in vitro. Four partitioned fractions as described in Figure 6, including n-hexane (Hex), ethyl acetate (EtA), n-butanol (n-BuOH) and water (H₂O) sub-extracts. THP-1 cells were cultured with DMSO as the negative control, or co-incubated with C. acnes (M.O.I. = 75) and different concentrations of four respective sub-extracts for 24 h. IL-8 level was examined by the method described above. Each value shows the mean ± SD. Values with different symbols are significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***).

Figure 3

Effects of the ethanol-soluble fraction (ESF) of n-hexane layer on C. acnes-induced IL-8 production and mitogen-activated protein kinase (MAPK), p38, extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) activation in THP-1 cells and C. acnes-stimulated ear swelling in mice. THP-1 cells were cultured with DMSO as the negative control, or co-incubated with C. acnes (M.O.I. = 75) and different concentrations of ESF for 24 h (for IL-8 production) or 2 h (for MAPK activation). IL-8 concentration was determined using the method described above (a). MAPK activation was determined by western blot (b). In the ear edema mouse model, PBS, ESF (2, 4 or 6 μg/site) or luteolin (50 μg/site) was intradermally injected, immediately followed by the C. acnes injection. Infiltrated neutrophils were observed in a hematoxylin and eosin-stained cross section of the C. acnes-injected ear (×1000 magnification panel). Arrow (⟶): neutrophil infiltration. Scale bars represent 200 μm. The inhibitory effects of ESF and luteolin on C. acnes-stimulated mouse ear edema was quantified as described above (c). Each value shows the mean ± SD. Values with different symbols are significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***).

Figure 3

Effects of the ethanol-soluble fraction (ESF) of n-hexane layer on C. acnes-induced IL-8 production and mitogen-activated protein kinase (MAPK), p38, extracellular signal-regulated kinase (ERK) and c-Jun N-terminal kinase (JNK) activation in THP-1 cells and C. acnes-stimulated ear swelling in mice. THP-1 cells were cultured with DMSO as the negative control, or co-incubated with C. acnes (M.O.I. = 75) and different concentrations of ESF for 24 h (for IL-8 production) or 2 h (for MAPK activation). IL-8 concentration was determined using the method described above (a). MAPK activation was determined by western blot (b). In the ear edema mouse model, PBS, ESF (2, 4 or 6 μg/site) or luteolin (50 μg/site) was intradermally injected, immediately followed by the C. acnes injection. Infiltrated neutrophils were observed in a hematoxylin and eosin-stained cross section of the C. acnes-injected ear (×1000 magnification panel). Arrow (⟶): neutrophil infiltration. Scale bars represent 200 μm. The inhibitory effects of ESF and luteolin on C. acnes-stimulated mouse ear edema was quantified as described above (c). Each value shows the mean ± SD. Values with different symbols are significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***).

Figure 4

Effects of β-ionone or dihydroactinidiolide on C. acnes-induced inflammation in vitro and in vivo. THP-1 cells were co-incubated with C. acnes (M.O.I. = 75) and different concentrations (10, 20 or 50) of β-ionone or dihydroactinidiolide for 24 h. IL-8 level was examined by the method described above (a). PBS, β-ionone, dihydroactinidiolide or luteolin (50 μg/site) was intradermally injected, immediately followed by the C. acnes injection. Infiltrated neutrophils were observed in a hematoxylin and eosin-stained cross section of the C. acnes-injected ear. The inhibitory effects on C. acnes-stimulated mouse ear edema were evaluated by examined by the method described above (b). Evaluation of C. acnes-induced immune cells by flow cytometry in mouse ear after intradermal injection of PBS, β-ionone or luteolin. Twelve hours after the injection, flow cytometric analysis of the inflammatory cells harvested from C. acnes-induced ear tissues was performed. Cell suspensions were incubated with anti-CD45/PerCP and anti-Ly6G/FITC, and analyzed by flow cytometry (c). Each value shows the mean ± SD. Values with different symbols are considered to be significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***).

Figure 4

Effects of β-ionone or dihydroactinidiolide on C. acnes-induced inflammation in vitro and in vivo. THP-1 cells were co-incubated with C. acnes (M.O.I. = 75) and different concentrations (10, 20 or 50) of β-ionone or dihydroactinidiolide for 24 h. IL-8 level was examined by the method described above (a). PBS, β-ionone, dihydroactinidiolide or luteolin (50 μg/site) was intradermally injected, immediately followed by the C. acnes injection. Infiltrated neutrophils were observed in a hematoxylin and eosin-stained cross section of the C. acnes-injected ear. The inhibitory effects on C. acnes-stimulated mouse ear edema were evaluated by examined by the method described above (b). Evaluation of C. acnes-induced immune cells by flow cytometry in mouse ear after intradermal injection of PBS, β-ionone or luteolin. Twelve hours after the injection, flow cytometric analysis of the inflammatory cells harvested from C. acnes-induced ear tissues was performed. Cell suspensions were incubated with anti-CD45/PerCP and anti-Ly6G/FITC, and analyzed by flow cytometry (c). Each value shows the mean ± SD. Values with different symbols are considered to be significantly different from the C. acnes control (C. acnes alone) at p < 0.05 (*), p < 0.01 (**) and p < 0.001 (***).

Figure 5

Effects of β-ionone on C. acnes-induced cellular IL-1β production and caspase-1 expression. THP-1 cells were cultured with DMSO as the negative control, or co-incubated with C. acnes (M.O.I. = 75) and different concentrations (10, 20 or 50 μM) of β-ionone for 24 h (IL-1β production) or 16 h (caspase-1 expression). The culture supernatants were subsequently collected and analyzed for IL-1β level (a). Caspase-1 expression was determined by western blot (b). Each value shows the mean ± SD. Values with different symbols are considered as significantly different from the C. acnes control (C. acnes alone) at p < 0.01 (**) and p < 0.001 (***).

Figure 6

Flowchart of the extraction methods used to separate the anti-inflammatory components from WBM leaf. Fractions were designated as Hex-1, Hex-2 and Hex-3, as indicated in the figure. The structures of 1 (β-ionone) and 2 (dihydroactinidiolide) isolated from the Hex-2 layer.