Cinnamtannin B1 attenuates rosacea-like signs via inhibition of pro-inflammatory cytokine production and down-regulation of the MAPK pathway

Preparation of Cinnamtannin B1 (CB1)

The CB1 used in this study was isolated from the stem of Formosan Cinnamomum validinerve, which was collected at Mudan mountain at Pingtung County, Taiwan. The plant sample was identified by one of the authors, Prof. Ih-Sheng Chen, and the voucher specimen was deposited in the Herbarium of the College of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan. The separation and isolation of the CB1 has been reported (Yang et al., 2020b). Briefly, the stem was dried and extracted with MeOH. The extract was then partitioned with EtOAc: H₂O (1:1) solvent system and then separated by column chromatography. The purity of CB1 in this preparation was over 90% based on ¹H & ¹³C NMR analysis.

Reagents and chemicals

The cleaved human cathelicidin peptide LL-37 (LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES) was customarily synthesized with 95% purity and without any modifications by Genemed Synthesis Inc. (South San Francisco, CA). The chemicals used in the animal model were obtained from Sigma (St Louis, MO, USA). Reagents for reverse transcription-polymerase chain reactions (RT-PCR) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). The enzyme-linked immunosorbent assay (ELISA) sets for inflammatory-related cytokine detections in both animal models and in vitro tests were purchased from BD Biosciences (San Diego, CA). All of the reagents for cell culture were from Hyclone (Logan, UT, USA).

LL-37-induced rosacea-like animal model

The mouse model of rosacea was adopted from Yamasaki et al. (2007). The experimental protocol was approved by the Kaohsiung Medical University Institutional Animal Care and Use Committee (IACUC number 106015). Briefly, 4-5 weeks-old male BALB/c mice (National Laboratory Animal Center, Taipei, Taiwan) were weighed and grouped to minimize group weight differences for naïve control (NA), phosphate-buffered saline (PBS)-treated group (LL-37), and CB1-treated group (LL-37 + CB1) after one-week acclimatization. Each animal was marked 4-6 positions on the dorsal area. Except for the NA group, mice were injected with 40 µL of LL-37 (320 µM) intradermally twice daily for 2 days at the marked positions on the back. The CB1 (4 mg/mL) or PBS were intraperitoneally administered (100 µL/20 g) to mice daily for two days after the first and the third inductions. For each mouse, the marked positions were photographed and harvested a day after the last LL-37 induction. The skin sections were processed for the histological examination, myeloperoxidase (MPO) measurements, MIP-2 measurements, or mRNA extraction. The mice were anesthetized by isoflurane (2%) before harvesting the skin sections and euthanization.

Histological examination of the skin section

The marked mice skin sections were harvested and fixed with 4% neutral formalin buffer for at least 1 day. The skin sections were embedded in paraffin and cut into 3 µm thick slices. The sections were stained with hematoxylin and eosin or processed for immunohistochemistry staining for Gr-1, rat anti-mouse polyclonal antibody (R&D Systems, Minneapolis, MN, USA). Each skin section was separated into 4-6 fields at 200-fold magnification; the numbers of neutrophils were counted manually from each field then normalized by the field number.

RNA isolation and real-time reverse transcription-polymerase chain reactions (RT-PCR)

Total RNA was isolated from the skin samples by isol-RNA lysis reagent (5-Prime, San Francisco, CA, USA). The total RNA of each sample was reverse-transcribed to cDNA by RevertAid RT Kit (Thermo Fisher Scientific, Waltham, MA, USA) using oligo (dT). The expression of the genes was quantified using SYBR Mix (Thermo Fisher Scientific, Waltham, MA, USA) by Real-time PCR 7900 (Applied Biosystems, Foster City, CA, USA). The assay reactions were 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. The fold-changes of the genes between groups were calculated by the ΔΔCt method. The primers used in this study included HPRT (5′-TCAGTCAACGGGGGACATAAA-3′ and 5′-GGGGCTGTACTGCTTAACCAG-3′) (Shi et al., 2010), MIP-2 (5′-AGTGAACTGCGCTGTCAATGC-3′ and 5′-GCCCTTGAGAGTGGCTATGACT-3′) (Mukhopadhyay et al., 2011), and KC (5′-CAGCCACCCGCTCGCTTCTC-3′ and 5′-TCAAGGCAAGCCTCGCGACCAT-3′) (Degos et al., 2015).

Cytokine measurements

The skin samples were harvested and homogenized with 5-fold tissue weight volume of lysis buffer (10 mM pH 8.0 Tris–HCl buffer with 150 mM NaCl, 1% Tween 20, 10% Glycerol, 5 mM EDTA) and 200:1 diluted protease inhibitor cocktail solution (Merck Millipore, Danvers, MA, USA). After centrifugation, the supernatants were quantified for macrophage inflammatory protein 2 (MIP-2) by ELISA kit (R&D Systems, Minneapolis, MN, USA).

Myeloperoxidase (MPO) activity assay

The MPO activity, as an indicator of the total inflammation level of the lesions, was adopted from protocols published previously (Queiroz et al., 2009). Briefly, the skin sections were homogenized in one mL cooled phosphate buffer (pH 4.7) supplemented with a 5 µL Protease inhibitor cocktail set solution (Merck Millipore, Danvers, MA, USA). After centrifuging the homogenized sample, the sample buffer was replaced by 0.05 M sodium phosphate buffer (pH 5.4 , containing 0.5% hexadecyltrimethylammonium bromide (HTAB) solution). After three times frozen-thawed cycles (frozen at −80°C for 8 min and thawed at room temperature for 10 min), the samples were centrifuged and the supernatants were collected for MPO evaluation. For MPO quantification, 12.5 µL of the samples and diluted standards were mixed with 62.5 µL of 3,3′, 5,5′-tetramethylbenzidine (TMB) solution (1.6 mM TMB, 0.0045% H₂O₂, 0.5% HTAB solution) in 96-well plates. The plates were then incubated at room temperature for 5–45 min, and the reaction was stopped by adding 50 µL of the stop solution (4 M H₂SO₄). The OD values at 450 nm/570 nm were recorded. The amount of myeloperoxidase was normalized by the amount of total protein in each sample.

Cell culture

Human keratinocyte HaCaT was from Prof. Yen, Kaohsiung Medical University (Huang et al., 2019), and monocyte THP-1 cell line was obtained from Bioresource Collection and Research Center (BCRC). HaCaT and THP-1 cells were maintained in Dulbecco’s Modified Eagle Medium (DMEM) and RPMI 1640, respectively, supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. Cells were seeded to antibiotics free and low/no FBS (0% in HaCaT cells, and 2% in THP-1 cells) culture medium for the experiments.

In vitro cell experiments

HaCaT cells (7 × 10⁴ cells/mL, 300 µL/well) and THP-1 cells (1 × 10⁶ cells/mL, 300 µL/well) were either left untreated (control) or stimulated with LL-37 (5 µM) in the absence or the presence of CB1 (1, 10, and 20 µM) for 24 h in 48-well plates. After 24 h co-incubation, supernatants were collected for cytokine/chemokine measurements by ELISA (R&D Systems, Minneapolis, MN, USA) as recommended by the manufacturer. The cells were harvested for cell viability by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay following the protocol from a previous study (Cheng et al., 2017).

For Western blotting, HaCaT cell (1×10⁵ cells/mL, 10 mL/10cm dish) and THP-1 cell (1×10⁶ cells/mL, 10 mL/10cm dish) were either untreated (NA) or stimulated with 5 µM LL-37 and CB1 (0, 1, 10 µM). Briefly, the cells were lysed by radioimmunoprecipitation assay buffer (RIPA, 20 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 1 mM ethylene glycol tetra-acetic acid, 1% Triton X-100, and 1 mM phenylmethylsulfonyl fluoride, pH 7.4) and heated with 4X sodium dodecyl sulfate (SDS) sample buffer (8% SDS, 0.2 M Tris (pH 6.8), 0.4 M dithiothreitol, 0.4% coomassie brilliant blue R-250, and 40% glycerol) at 95°C for 5 min. The lysates were resolved by 10% SDS-PAGE and transferred to polyvinylidene fluoride membrane (Millipore, Bedford, MA, USA, 0.45 µM). The membranes were blocked with 5% skim milk in Tris buffer saline containing 0.1% Tween-20 (TBST) under 60 rpm rocking at room temperature for 1 h. Membranes were blotted with the primary antibodies in sequence: Phosphorylation protein (Phospho-ERK or Phospho-p38), total protein (total-ERK or total-p38), and internal control protein (GAPDH). The primary antibodies were diluted in TBST supplemented with 5% bovine serum albumin (BSA) or 5% skim milk. The membranes were washed three times with TBST under 120 rpm rocking at room temperature after overnight incubation with a primary antibody under 60 rpm rocking at 4°C. After blotting with the secondary antibody conjugated with horseradish peroxidase (HRP) in TBST with 5% skim milk for 1 h at room temperature, the activity of HRP was developed and detected using enhanced chemiluminescence (Thermo Fisher Scientific, Waltham, MA, USA). Images were captured by using Chemi-Doc Gel Imaging System (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The antibodies were all purchased from Cell Signaling Technology (Danvers, MA, USA). The phospho-antibodies phospho-ERK and phospho-p38 recognize residues surrounding Thr202/Tyr204 of human p44 MAPK and Thr180/Tyr182 of human p38 MAPK, respectively.

Statistical analysis

The statistical differences between the CB1 treatment group and the LL-37-only group were analyzed by parametric analysis of variance (ANOVA) with Dunnett’s test by Prism 5.0 (GraphPad Software Inc., USA). P <0.05 was defined as statistical significance.

CB1 attenuated LL-37-induced skin redness and local inflammation responses in vivo

LL-37 induced rosacea-like signs, characterized by cutaneous erythema with central necrosis, at the sites of injection, while the CB1 treatment reduced the LL-37-induced rosacea-like signs (Figs. 1A–1C). Histological sections of the lesions showed that LL-37 induced neutrophil infiltration, hemorrhage, thickening and cell necrosis in the epidermal tissues, and these observations appeared to be reduced in the CB1 group (Figs. 1D–1F). Immunohistochemistry staining of the skin sections with Gr-1 for polymorphonuclear neutrophils indicated that LL-37 induced neutrophil recruitment to the skin, and the CB1 treatment decreased the neutrophil recruitment (Figs. 2A–2F). The number of GR-1⁺ cells was significantly higher than the controls after LL-37 induction and the number was decreased by CB1 (Fig. 2G).

CB1 suppressed MPO and MIP-2 production in vivo

The skin sections were homogenized for further examination of inflammatory responses. MPO, a biomarker for overall inflammation, was elevated in the LL-37-induced lesion, and the CB1 treatment reduced the LL-37-induced MPO level (Fig. 3A). MIP-2 and KC are functional homologs of neutrophil chemoattractant IL-8 in mice. The elevated levels of MIP-2 in the LL-37-induced lesions were also attenuated by CB1 treatment (Fig. 3B). MIP-2 and KC gene expression profiles although were not significantly decreased by CB1, they showed similar trends as the MIP-2 levels (Figs. 3C & 3D).

CB1 reversed LL-37-induced IL-8 production by HaCaT and THP-1 cells in vitro

Since the recruitment of neutrophils and production of MIP-2 (mouse functional homologs of IL-8) were both significantly attenuated by administration of CB1 in vivo, the effects of CB1 on human cells were further studied in human keratinocyte HaCaT and human monocyte THP-1 cells in vitro.

LL-37 significantly decreased the viability of HaCaT cells at 5 µM (Fig. 4A), and CB1 at the concentration of 10 µM reversed the LL-37-induced cytotoxicity. By contrast, LL-37 did not affect the viability of THP-1 at 5 µM (Fig. 4C). LL-37 significantly induced IL-8 production both in HaCaT and THP-1 cells, and CB1 reversed the LL-37-induced IL-8 secretion (Fig. 4B and Fig. 4D).

CB1 inhibited LL-37-induced IL-8 production through extracellular signal-regulated kinase (ERK) mitogen-activated protein kinase (MAPK) pathway

LL-37 up-regulated the phosphorylation of ERK and p38 both in HaCaT and THP-1 cells. CB1 at 10 µM reversed the LL-37-induced phosphorylation of ERK but not p38 (Figs. 5B and 5E).