Luteolin was purified and provided by the Key Laboratory of Bioactive Substances and Resources Utilization of Chinese Herbal Medicine, Ministry of Education (Beijing, China) and the purity of the product was >98%, detected by HPLC (UV). The molecular structure is shown in Fig. 1A. Monoclonal antibodies against protein kinase B (AKT; no. 4691), phosphor-AKT (Ser473; no. 4060) and horseradish peroxidase-conjugated anti-mouse or anti Rabbit IgG (nos. 7076 and 7074) were purchased from Cell Signaling Technology, Inc. (Boston, MA, USA). Monoclonal antibodies against heme oxygenase-1 (HO-1; no. ab13243), B-cell lymphoma 2 (BCL-2; no. ab692), BCL-2-associated X protein (BAX; no. ab32503), activated caspase-3 (no. ab2302) and GAPDH (no. ab9484) were purchased from Abcam (Cambridge, MA, USA). Alexa 488-conjugated goat anti-rabbit secondary antibodies (no. A-11078) were purchased from Thermo Fisher Scientific, Inc. (Waltham, MA, USA). The Pierce BCA Protein Assay kit was purchased from Thermo Fisher Scientific, Inc.

Cultured immortalized HaCaT human keratinocyte cells were obtained from Nanjing KeyGen Biotech Co., Ltd. (Nanjing, China) (11). The cells were grown in DMEM (Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin, and 100 µg/ml streptomycin, cultured at 37°C in 5% CO₂. When 80% confluence was achieved, the HaCaT cells were treated with hydrogen peroxide (100 µM) for 2 h with or without the pretreatment of indicated concentrations (3, 6, 12, 25 µg/ml) of luteolin for 12 h at 37°C in 5% CO₂.

Cell viability was assessed using standard MTS methods. Briefly, the cells were seeded in triplicate at a density of 1×10⁵ cells/ml in a 96-well plate, and cell viability assays were performed using the CellTiter 96 Aqueous One Solution Cell Proliferation Assay kit (Promega Corporation, Madison, WI, USA), the absorbance at 490 nm was measured using a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA).

For the apoptosis assays, a fluorescein-conjugated Annexin V (Annexin V-FITC) staining assay kit (BD Biosciences, San Jose, CA, USA) was used to quantitatively assess the level of induced cell apoptosis. Briefly, the cells were washed with PBS and stained with 5 µl of Annexin V-FITC and 5 µl of propidium iodide (PI) in each sample to quantify the cell number at different stages of cell death. Following incubation at room temperature in the dark for 15 min, the percentages of apoptotic cells were quantified as a percentage of the Annexin V-positive population with a FACSCalibur flow cytometer (BD Biosciences), and the data were analyzed with FlowJo Version 7.6.2 software (Tree Star, Inc., Ashland, OR, USA).

The MMP values were determined using the dual-emission potential-sensitive probe 6,6′-tetrachloro-1,1′,3,3′-tetraethyl-imidacarbocyanineiodide (JC-1; Sigma-Aldrich; Merck KGaA, Darmstadt, Germany). Briefly, the cells were seeded in the camber slides at a density of 5×10⁵ cells/well in 200 µl culture medium and incubated with 10 µM of JC-1 for 20 min at 37°C in the dark. The JC-1 was then removed, and the cells were washed with cold PBS to remove unbound dye. The quantity of JC-1 retained in the cells was assessed with a laser scanning confocal microscope system (Olympus Corporation, Tokyo, Japan).

Skin keratinocytes were incubated for a period of time following hydrogen peroxide exposure, following which cell lysates were collected by lysis in RIPA buffer, and the concentration of protein was detected using a Pierce BCA protein kit purchased from Thermo Fisher Scientific, Inc. Equal quantities of proteins (20 µg) were mixed with loading buffer and subjected to electrophoresis using 10% SDS-polyacrylamide gels. The separated proteins were transferred onto polyvinylidene fluoride membranes and non-specific bindings were blocked with 5% (w/v) skim milk dissolved in tris-buffered saline with Tween. The membranes were then subjected to immunoblot analysis with the appropriate antibodies (1;1,000 dilution for the primary antibodies for 2 h at room temperature; 1:2,000 dilution for the secondary antibodies for 1 h at room temperature). The immune-reactive protein bands were visualized using an enhanced chemiluminescence detection system (GE Healthcare Life Sciences, Chalfont, UK) followed by autoradiography. A G:Box Bioimaging system (Syngene, Frederick, MD, USA) was used to assess autoradiographic signals and bands were quantified using GeneTools Image Analysis Software version 4.3.7 (Syngene).

The animal experiments were performed in the Experimental Animal Laboratory of Nanjing University School of Medicine (Nanjing, China), and approved by the Institutional Animal Care and Use Committee of Nanjing University. A total of 18 male 8-10 weeks-old Sprague-Dawley rats weighing 220-280 g were purchased from Research Institute of Model Organisms at Nanjing University, the rats were housed in separate cages at 25°C and in a 12-h light/dark lighting system. All animals has free access to food and water. The rats were randomly allocated into three groups: Mock control group (Ctl), I/R injury group (I/R), and I/R injury with luteolin treatment group (I/R + Luo).

The rats were anesthetized with an intraperitoneal (i.p.) injection of ketamine (100 mg/ml) and xylazine (20 mg/ml) at a total dose of 0.2 ml/100 g of body weight. Abdominal hair was removed with an electric clipper, and all surgical procedures were performed under sterile conditions. The borders of the flaps were outlined on the abdomen using a template measuring 3×6 cm. The flap was raised with the base at the left inferior epigastric artery, including the skin and the intimately attached panniculus carnosus, as previously described (11,12). Ischemia was induced by applying a single microvascular clamp across the femoral vascular pedicle, and the flap was sutured to the donor bed using a 4/0 polypropylene suture. Following 4 h of ischemia, the clamps were removed. The rats in the mock control group received the flap elevation and vessel dissection but no ischemic insult. The surgical procedure is shown in Fig 1B. To assess the functions of phosphoinositide-3-kinase (PI3K)/AKT and HO-1 in luteolin-mediated tissue protection, HO-1 inhibitor (ZnPP) or PI3K/AKT inhibitor (LY294002) was administrated to the rats on day 1 and day 4 post-I/R injury at doses of 3 and 100 mg/kg body weight (i.p.). The surviving area of the skin flap was then measured.

Following surgery, all animals were sacrificed on post-operative day 7. Gross observation was performed to identify the line of demarcation between the viable and the necrotic tissue. The dorsum of the rat in each group was covered with clear paper in anesthesia to accurately measure the surviving or necrotic areas of the flap, and the flap was cut into two sections: Viable and necrotic. The entire flap and the necrotic and viable regions were measured using two-dimensional planimetry in a blinded-manner. The surviving proportions of the flaps were determined as a percentage of the entire flap area (surviving flap proportion=viable flap area/total area ×100%). Following assessment, the rats were sacrificed with an overdose of sodium pentothal.

The transcriptional expression of interleukin (IL)-1β, tumor necrosis factor (TNF)-α and IL-6 were detected by RT-qPCR analysis. Total RNA was extracted from the skin flap tissue on day 1 post-I/R using an RNAeasy Micro kit (Qiagen GmbH, Hilden, Germany) according to the manufacturer’s protocol. mRNA was reverse transcribed into cDNA with PrimeScript RT Master mix (Takara Bio, Inc., Otsu, Japan). SYBR green qPCR was performed using PCR Master mix (Thermo Fisher Scientific, Inc.). Each cDNA reaction was prepared from 1 µg RNA, diluted to 100 µl of the final volume and 1 µl cDNA was subsequently used for each PCR reaction, and the reaction mixture had a total volume of 20 µl containing 10 µl PCR Master Mix (2X), 0.5 µl PCR forward primer (10 mM), 0.5 µl PCR reverse primer (10 mM) and 8 µl H₂O. The PCR conditions were as follows: 95°C for 30 sec for pre-incubation, 95°C for 5 sec and 60°C for 30 sec for amplification; 95°C for 10 sec and 65°C for 10 sec to melting curve, and 40°C for 30 sec for cooling. The following primer pairs were used: IL-1β, 5′-GGA ACC CGT GTC TTC CTA AAG-3′ (forward) and 5′-CTG ACT TGG CAG AGG ACA AAG-3′ (reverse); TNF-α, 5′-CCA ACA AGG AGG AGA AGT TCC-3′ (forward) and 5′-CTC TGC TTG GTG GTT TGC TAC-3′ (reverse); IL-6, 5′-GAA AGT CAA CTC CAT CTG CC-3′ (forward) and 5′-CAT AGC ACA CTA CGT TTG CC-3′ (reverse); first measure-actin, 5′-AAC CCT AAG GCC AAC CGT GAA AAG-3′ (forward) and 5′-TCA TGA GGT AGT CTG TCA GGT-3′ (reverse).

The relative expression of target genes was determined to β-actin and was calculated using the 2^−ΔΔCq method. The relative mRNA expression was quantified as described previously (13).

The oxidative stress status of the flaps was assessed by measuring the superoxide dismutase (SOD) activity and the content of myeloperoxidase (MPO) and malondialdehyde (MDA) in the skin flap tissue. Tissue samples (1×1 cm) were separated from the central area of the surgical flaps in each group; these samples were weighed, homogenized, and diluted to 10% (v/v) in an ice bath. The homogenate was then centrifuged at 600 × g for 15 min at 4°C and the supernatant solution was collected. The activity of SOD and the levels of MPO and MDA in the homogenate were then determined using a commercial kit following the protocol suggested by the manufacturer (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Tissue specimens were embedded in paraffin following fixation in 10% formalin. The sections (4-6 µm) were deparaffinized in xylene and hydrated with ethanol. The sections were incubated with 1% bovine serum albumin (Sigma-Aldrich; Merck KGaA) in PBS/Tween to block unspecific binding of the antibodies. Following overnight incubation at 4°C, the sections were incubated with primary antibodies (1:500 for P-AKT, HO-1 and active Caspase-3) for 2 h followed by the secondary anti-bodies (1:1,000) for 30 min in a dark room at 25°C. Nuclei were stained using DAPI solution (Molecular Probes, Thermo Fisher Scientific, Inc.). Fluorescent images were captured and visualized using an Olympus fluorescence microscope (Olympus Corporation).

Statistical analyses were performed using GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). One-way analysis of variance or Student’s t-test was used to test for significant difference. P<0.05 was considered to indicate a statistically significant difference.

Prior to investigating the protective properties of luteolin against hydrogen peroxide-induced cell death, the cytotoxic effects of hydrogen peroxide were first examined in the HaCaT cells. The MTT assay indicated that the treatments with various doses of hydrogen peroxide resulted in cytotoxic effects, and cell viability was significantly decreased at a concentration of 100 µM. Therefore, 100 µM of hydrogen peroxide was selected as the optimum concentration for the subsequent in vitro assay. To measure the protective effect of luteolin, the HaCaT cells were sham-exposed or received treatment with various doses of luteolin. MTT assays revealed that the hydrogen peroxide-induced reduction of cell viability was effectively prevented by pretreatment with luteolin (Fig. 2A).

To further evaluate the effect of luteolin on hydrogen peroxide-induced apoptosis, the HaCaT cells were stained with Annexin V and PI, and cell apoptosis was measured by FACS, as shown in Fig. 2B. Cells with a high expression of Annexin V and not expressing PI were considered early apoptotic cells, whereas cells with a high expression of Annexin V and expressing PI were classified as late apoptotic cells. The results showed that HaCaT cells exposed to hydrogen peroxide treatment had a high level of cell apoptosis, and treatment of these cells with luteolin significantly reduced cell apoptosis; the percentage of apoptotic cells was 12.06±2.32% (luteolin treatment group), vs. 18.37±1.92% (untreated group). This suggested that luteolin may inhibit hydrogen peroxide-induced keratinocyte apoptosis.

A change in MMP was examined as mitochondria are major sites of oxidative phosphorylation and ROS production, and they are involved in the initiation of apoptosis through membrane permeabilization (14). As shown in Fig. 2C, compared with the control group, the hydrogen peroxide-treated cells exhibited a loss of MMP; increased fluorescence intensity was observed in the hydrogen peroxide-treated HaCaT cells following JC-1 dye staining, indicative of mitochondrial depolarization. However, pretreatment with luteolin significantly inhibited the loss of MMP in the hydrogen peroxide-treated cells. Altered cell morphology was also observed in response to hydrogen peroxide treatment, however, the cell morphology was maintained and the overall cell shape was maintained following treatment with increased dosed of luteolin, indicating the protective effect of luteolin against hydrogen peroxide-induced cell death.

To assess the mechanism by which luteolin treatment can protect against hydrogen peroxide-induced keratinocyte apoptosis, the present study investigated whether luteolin pretreatment was associated with varied intracellular signaling pathway activation. As shown in Fig. 3, treatment of the keratinocytes with hydrogen peroxide resulted in a significant inhibition of PI3K/AKT pathway activation, which indicated the inhibition of cell growth and differentiation. Consistent with this observation, there was increased expression of BAX and decreased expression of BCL-2, and the BCL-2/BAX ratio were decreased, which suggested that these cells underwent apoptosis once exposed to hydrogen peroxide treatment.

Luteolin pretreatment not only significantly restored the cell viability, but also decreased the apoptotic rate, upregulated the expression of BCL-2, downregulated the expression of BAX and increased the BCL-2/BAX ratio. In addition, luteolin pretreatment increased the phosphorylation of AKT in a dose-dependent manner, as the PI3K/AKT pathway is one of the most important intracellular survival signaling pathways. This result suggested the protective effect of luteolin in I/R injury may be associated with PI3K/AKT pathway activation.

From the observations in vitro, it was hypothesized that luteolin may have a protective effect in I/R injury during skin flap surgery. To evaluate this hypothesis, the present study successfully established a cutaneous I/R injury rat model. To assess the effect of luteolin on I/R injury in the skin flap model, the surviving areas of the flaps were measured 7 days following surgery. It was observed that the I/R injury group exhibited smaller surviving flap areas compared with the mock control groups. The rats that received luteolin pretreatment had larger surviving flap areas than those in the I/R injury group at 7 days following surgery (Fig. 4A).

It was also observed that luteolin treatment suppressed the mRNA levels of pro-inflammatory cytokines and chemokines. The expression of pro-inflammatory cytokines in the biopsied skin samples were examined using an RT-qPCR assay. As shown in Fig. 4B, the expression levels of TNF-α, IL-6 and IL-1β were significantly decreased in the luteolin-treated groups compared with those in the I/R injury groups. The histopathological examination indicated that the severity of tissue injury in the luteolin pretreatment groups was markedly reduced; there was less inflammatory cell infiltration and the damaged skin area was significantly decreased (Fig. 4C).

To assess the I/R injury induced oxidative stress damage, the levels of MPO, MDA, and SOD were examined in the surgical flaps. There was increased production of MPO and MDA in the ischemia group compared with the control group, and luteolin treatment significantly decreased the expression of these enzymes. The decreased expression of SOD following I/R injury was recovered in response to luteolin treatment. These results indicated the oxidative stress scavenging effects of luteolin (Fig. 4D).

Using immunofluorescent staining (Fig. 5A and B), it was observed that there was increased activation of caspase-3 protein in the ischemia group compared with the control group and the luteolin treatment group. Luteolin treatment reduced the level of cell apoptosis induced by I/R injury, as the protein expression of activated caspase-3 was significantly reduced. The levels of phospho-AKT and antioxidant enzyme HO-1 were significantly increase in response to luteolin treatment, indicating that these two molecules may be involved in the protective function of luteolin.

To determine whether luteolin-induced skin protection was mediated by the PI3K/AKT pathway, the PI3K inhibitor LY294002 (14) was used. The inhibition of AKT activity markedly reduced the luteolin-induced protection and inhibition of apoptosis in skin I/R injury (Fig. 5C). These results suggested that luteolin inhibited apoptosis and improved skin flap survival at least partly through the PI3K/AKT pathway in skin flap surgery. By contrast, using the HO-1 inhibitor ZnPP (15), no significant change in luteolin-induced protection was observed; the surviving flap areas and neutrophil infiltration levels were similar to those in the luteolin treatment groups, suggesting that HO-1 may not be a crucial molecule in regulation of skin damage during the I/R injury caused by skin flap surgery.