Ethanol extract of Cnidii Rhizoma ameliorates dextran sulfate sodium-1 induced ulcerative colitis by inhibiting JNK / NF-κB signaling 2 3

1Department of Oriental Pharmacy, College of Pharmacy, Wonkwang-Oriental Medicines 7 Research Institute, Wonkwang University, 460 Iksandae-ro, Iksan, Jeonbuk 54538, Republic 8 of Korea 9 2 Pharmaceutical Safety Evaluation Division, Pharmaceutical Safety Bureau, Ministry of 10 Food and Drug Safety, Cheongju, Korea. 11 12 13 Correspondence to: Prof. Seung-Heon Hong, Department of Oriental Pharmacy, College of 14 Pharmacy, Wonkwang-Oriental Medicines Research Institute, Wonkwang University, 460 15 Iksandae-ro, Iksan, Jeonbuk 54538, Republic of Korea. Tel: (+82) 63-850-6805, Fax: (+82) 6316 843-3421, E-mail: jooklim@wku.ac.kr 17 18 †These authors contributed equally to this work. 19 20 Running title: Cnidii Rhizoma inhibit colitis via JNK/NF-κB signaling 21 Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 October 2018 doi:10.20944/preprints201810.0144.v1


Introduction
Ulcerative colitis (UC) is a chronic and multifactorial inflammatory bowel disease (IBD), which is mainly occurred in the rectum and colon [1].Nevertheless UC is not lifethreatening in itself, it has been reported that chronic recurrence of UC could increase the risk of colorectal cancer, which is the third most common cancer in the world [2].Several drugs such as aminosalicylic acid drugs, immune suppressor, and hormone agents have been used to treat UC.However, many people are still suffered from high recurrence rates of UC and sideeffects of drugs [3].Although the etiologies of UC and a lot of therapies have been studied, it is very difficult to treat because of recurrence and other unknown reasons.
Among the various causes of UC, pro-inflammatory cytokines, which are mainly secreted from macrophages, T cells and epithelial cells, is certainly thought as a crucial factor in the treatment of UC.Since recurrence of intestinal inflammation is related to an abnormal immune response, cytokines can be an important mediator to control the intestinal immune system [4].In patients with UC, increased level of pro-inflammatory cytokines including IL-1β, IL-6, and TNF-α have observed and expression of these cytokines are regulated by many kinds of intracellular proteins [5,6].Among them, mitogen-activated protein kinases (MAPKs), including extracellular signal-regulated kinases (ERKs), the c-Jun N-terminal kinases (JNKs) and the p38 MAPK, are participated in inflammatory response by producing inflammatory mediators [7].These signaling pathways can regulate inflammation by converting extracellular stimuli into the intracellular signaling, which induce the expression of inflammation related genes [8].
Nuclear factor-kappa B (NF-κB), which is one of the downstream factor of MAPKs, is also crucial factor for treatment of UC.It has been reported that patients with UC show strongly activated NF-κB in colon tissues [9].NF-κB is consisted of p50 and p65, which normally exist in cytoplasm with inhibitor of NF-κB (IκB) [10].By extracellular stimuli, IκB is degraded and phosphorylated through ubiquitin/proteasome pathway, and then NF-κB is activated.Activated NF-κB is subsequently translocated into nucleus to bind with sequence of target genes [11].
Another downstream target of NF-κB is iNOS, which leads vasodilation of inflammatory cells easily [15].iNOS is highly expressed in inflamed intestine mucosa and induces tissue damage by generating nitric oxide (NO) [16].Therefore, suppressing NF-κB activation through MAPKs cascades is crucial for treating colitis.
Intestinal epithelium exhibits barrier function against luminal antigens.The intestinal epithelium is composed of a single epithelial cell layer.This layer is covered by mucus, which protects epithelial layer by distinguishing from mucosal bacteria [17].To protect the intestine from bacterial infection and damage, epithelial cells secrete anti-microbial peptides into mucus [18].Anti-microbial peptides are regulated by trefoil factor 3 (TFF3).TFF3, which is synthesized in the goblet cells of the small and large intestine, plays important roles in the protection and repair of epithelial surfaces, including the gastrointestinal tract [19].
In this study, we used rhizomes of Cnidium officinale Makino (Cinidii Rhizoma; CR).Asia [20].Since CR contains 1-2 % of essential oil and several compounds, it has shown various pharmacological effects such as muscle relaxing, anti-tumor, anti-anemia, and antimetastatic effects [21].In particular, CR has been previously reported that improvement of blood circulation and anti-inflammatory effects [22].However, there is no report on the therapeutic effect of CR on UC.Therefore, this study investigated that improvement effect of CR on colitis in in vitro and in vivo.

EtCR Prevents Symptoms of Colitis in DSS-induced Colitis Mice
To estimate whether EtCR could inhibit progression of colitis, mice were administrated 5% DSS containing water with or without EtCR (50 mg/kg/day) and 5-ASA (50 mg/kg/day).
As shown in Figure 1A, administration of DSS decreased body weight compared to control group.However, administration of EtCR inhibited weight loss similarly with 5-ASA, which is treatment drug of colitis.Also, the severity of colitis was presented by DAI score, which concerned other factors, including bleeding and diarrhea, as well as body weight.As shown in Figure 2B, DAI score of DSS group was steadily increased from day 5 to day 7.However, EtCR and 5-ASA treated groups showed better DAI score compared to only DSS-treated group.
Since Shortening of colon length is closely related to the increased DAI scores [24], we measured colon length after sacrificing the mice.Similarly with DAI scores, EtCR and 5-ASA suppressed the reduction of colon length (Figure 1C,D).Furthermore, enlarged spleen was observed in DSS-treated mice, whereas administration of EtCR and 5-ASA decreased the size of spleen (Figure 1E).These results suggest that EtCR ameliorate DSS-induced colitis similar to 5-ASA.

EtCR Inhibits Colon Tissue Damages Induced by DSS
Since the administration of DSS induces colon tissue damages, including loss of epithelial layer, infiltration of inflammatory cells into the mucosa, and loss of goblet cells [25], we conducted H&E staining and alcian blue staining to confirm colonic protective effect of EtCR from DSS induced damages.As shown in Figure 2A, morphology alterations were observed.Epithelial layer were almost eroded and crypt architecture were not found in DSStreated colon tissue.In addition, infiltration of inflammatory cells was also shown in DSStreated group.However, administration of EtCR and 5-ASA reduced not only damages of colon tissue, but also infiltration of inflammatory cells.To compare the degree of inflammation, we assessed the each group of H&E stained images and assigned scores.The scores indicated that administration of EtCR relieved severity of colitis.(Figure 2B).In addition, administration of EtCR inhibited the expression of COX-2 and iNOS, which are inflammatory factors increased by DSS (Figure 2E,F).Next, we confirmed goblet cells, which secrete a mucosal protective substance [26].To stain goblet cells, alcian blue staining was conducted.As shown in Figure 2C, blue spots were well-shown in colon tissue of control group and it was seldom seen in DSS-treated group.In EtCR and 5-ASA-administrated groups, the blue stained cells were increased in colon tissues (Figure 2D).Furthermore, the TFF3 expression was recovered in EtCR and 5-ASA-administrated groups, compared with DSS-treated group (Figure 2E,G).
Therefore, these results indicated that the EtCR could ameliorate colitis by protecting colon tissue.

EtCR Regulates LPS-stimulated Cytokine and Pro-Inflammatory Mediators in in vitro
To EtCR did not show significant cytotoxicity up to 100 μg/ml of EtCR (Figure 3A), we used 100 μg/ml of EtCR as a highest concentration.As shown in Figure 3B-D, EtCR reduced the productions of IL-1β, TNF-α, and IL-6, compared to LPS treated group.

EtCR Decreases LPS-induced NO and PGE2 by Suppressing iNOS and COX-2 in RAW 264.7 Cells
It has been reported that colon tissue damages induced by colitis is related to inflammatory mediators [28].Thus, reducing the pro-inflammatory mediators as well as cytokines is also crucial for colitis.As shown in Figure 4A,B, pre-treatment of EtCR decreased levels of NO and PGE2, which were elevated by LPS.In addition, we investigated expression levels of iNOS and COX-2, which produce NO and PGE2, respectively.Similarly with the tendency of NO and PGE2, EtCR reduced the expressions of iNOS and COX-2 (Figure 4C,D).

EtCR Reduces Inflammation Through NF-κB and MAPKs in RAW 264.7 Cells
NF-κB pathway regulates various inflammatory mediators.NF-κB exists in the cytoplasm in an inactive form with IκB.After inflammatory stimulation, IκB are phosphorylated and degraded.Continuously, NF-κB is translocated into the nucleus and leads to inflammatory responses [29].As shown in Figure 5A,B, IκB in the cytoplasm was phosphorylated and degraded by LPS stimulation, which is an inflammatory stimulus.And then, NF-κB p65 was translocated into the nucleus.However, the treatment of EtCR suppressed this signaling pathway.In addition, the result of NF-κB binding assay showed that the EtCR treatment effectively reduced the NF-κB binding activity, dose dependently (Figure 5C).It was also reported that the NF-κB signaling is regulated by MAPKs, including ERK, JNK, and p38 [30].As shown in Figure 5D and E, EtCR inhibited LPS-induced phosphorylation of JNK, whereas ERK and p38 phosphorylation was not suppressed by EtCR.These results suggested that EtCR inhibited the inflammatory response by inhibiting JNK and NF-κB signaling pathway.

EtCR Regulates TFF3 in HT 29 cells
To confirm whether EtCR recovered the expression of TFF3, we used HT29 known as colonic epithelial cells.Since HT29 cells show properties of pluripotent intestinal cells, they are commonly used to investigate intestinal epithelial circumstance [31].In our experiment, TFF3 was fully expressed in HT29 cells, but this expression was inhibited by LPS treatment.
However, the treatment of EtCR increased TFF3 in both protein and mRNA level without cytotoxicity (Figure 6).Thus, these results indicate that EtCR could ameliorate colitis by protecting the epithelial cells of the intestine.

Discussion
UC and Crohn's disease (CD) are the two most common forms of IBD, which is characterized by a chronic inflammation of intestine.The etiologies of both diseases are unclear.
However, main cause of these diseases is considered to multifactorial, including environmental factors such as smoking and western diet, the genetic factors and the intestinal flora [32].
Although the symptoms between UC and CD are similar in many parts, there are also slightly different.People suffering from UC have the abdominal pain in the left side of the abdomen, while suffering from CD have abdominal pain in their all abdomen.It has been known that CD is more severe than UC, but the incidence of CD is much lower than UC [33].In addition, the treatment drugs of UC, including 5-ASA and glucocorticoids, occurred side-effects such as arthralgia, myalgia, headache, and abdominal pain [34].Therefore, our study focused on UC rather than CD and suggested the new therapeutic agents of UC using natural products with few side effects.
EtCR is expected that could improve colitis through several research findings about EtCR.
It has been reported that the health functional food containing CR improved TNBS-induced colitis and histological changes such as thickening, dilatation, ulceration, and infiltration [35].
In addition, CR had anti-inflammatory effect by reducing NO production in LPS-stimulated BV-2 microglia cell [36].Despite these possibilities, there was no report on the effect of EtCR on UC.In this study, we confirmed whether EtCR can ameliorate UC, using DSS-induced colitis model.Since the symptoms, including weight loss, bloody stool, damages of colon tissue, colon shortening, and diarrhea, of this model are similar with human UC, it is often used to colitis experiment [37].In our experiment, administration of DSS reduced body weight, which indicates DSS induced colitis.However, the administration of EtCR significantly inhibited weight loss compared to DSS-treated group (Figure 1A).In addition, DAI scores and colon shortening were improved by administration of EtCR (Figure 1B-D).It has been also reported that the weight of spleen, which can produce inflammatory cytokines, was increased in DSStreated mice [38].As shown in Figure 1E, the spleen weight was increased by DSS treatment, whereas EtCR inhibited enlargement of spleen.In particular, EtCR exerted the similar therapeutic effect as 5-ASA, which is a treatment drug of UC (Figure 1).
In normal condition, epithelial cell layer secretes protective substances to colon to maintain colon condition.However, it cannot be secreted because of collapse of cell layer in colitis [39].
Therefore, symptoms such as diarrhea and bleeding are accompanied.Furthermore, the weakened epithelial cell layer allows immune cells to infiltrate into the intestinal lumen [40].
In our experiment, administration of DSS almost eroded epithelial layer and crypt.
Inflammatory cell infiltration also found in colon tissue of DSS-treated group.However, EtCR protected DSS-inducing mucosal erosion, disruption of intestinal crypts as well as inflammatory cell infiltration (Figure 2A,B).We also confirmed goblet cell, which are abundant in colon tissue and secreted protective mediators [41].As shown in Figure 2C, blue spots were well-shown in normal colon tissue, which is seldom seen in DSS-treated group.
However, the blue positive cells were increased in the colon tissue of 5-ASA group as well as EtCR-administrated group (Figure 2C,D).Accordance with result of alcian blue staining, the expression of TFF3, which is a constituent of mucus and is produced from goblet cell in large intestine, was decreased in the colon tissue of DSS-treated group, but it was recovered by EtCR administration (Figure 2E,G).These structural changes are known to be closely associated with inflammatory reactions, and the inflammatory factors such as COX-2 and iNOS are found in almost UC patients [42].Similarly with clinical case, expressions of COX-2 and iNOS were  2E,F).As a result, EtCR improved colitis in in vivo experiment, and the therapeutic effect of EtCR is slightly lower than 5-ASA.
To further investigate the underlying anti-inflammatory mechanisms of EtCR, we confirmed pro-inflammatory mediators in LPS-stimulated RAW 264.7 cells.The stimulated macrophages secrete large amounts of IL-1, IL-6, and TNF-α [43].Since the increased expressions of pro-inflammatory cytokines such as IL-1, IL-6, IL-8, and TNF-α are also detected in UC patients [5,6], this in vitro model was used to investigate the anti-inflammatory mechanism of EtCR.Falcarindiol, 6-hydroxy-7-methoxy-dihydroligustilide, ligustilidiol, and senkyunolide H were isolated from methanol extract of CR and they showed inhibitory effects on iNOS and COX-2 expresion in LPS-stimulated RAW 264.7 cells [44].In our experiment, LPS-stimulated RAW 264.7 cells secreted pro-inflammatory cytokines, but EtCR treatment reduced the production of IL-1β, TNF-α, and IL-6 (Figure 3).Similar to in vivo results, the expressions of iNOS and COX-2 were inhibited by EtCR.In addition, the production of NO and PGE2, which are produced by iNOS and COX-2, were also decreased by EtCR (Figure 4).
Thus, it is quite convincing that therapeutic effect on UC may come from anti-inflammatory properties of EtCR.
Ethanol extract of Cnidium officinale Makino showed anti-inflammatory effects by blocking NF-κB pathway in BV2 microglial cells [45].Therefore, we confirmed whether EtCR can regulate anti-inflammatory factors via NF-κB pathway in RAW 264.7 cells.NF-κB is plays a critical role in the immune system, which controls the expression of enzymes and cytokines during inflammatory responses [46].In normal condition, NF-κB binds to its inhibitor IκB, and could not translocate to the nucleus.Once IκB is phosphorylated by IKK, IκB is degraded and NF-κB undergoes nuclear translocation, and it triggers the expression of pro-inflammatory genes [47].Therefore, we investigated IκB phosphorylation and NF-κB translocation.Based on our results, LPS treatment induced phosphorylation of IκB and nuclear translocation of NF-κB, but EtCR decreased IκB phosphorylation and NF-κB translocation (Figure 5A,B).
Furthermore, the binding NF-κB to the nucleus induced by LPS was inhibited by EtCR (Figure 5C).Moreover, we confirmed expression of MAPKs in EtCR-treated cells, because the activation of MAPKs have been known to regulate NF-κB as an upstream factors [48].In this study, EtCR treatment attenuated LPS-induced phosphorylation of JNK, but not ERK and p38 (Figure 5D,E).These results showed that EtCR inhibited the expression of inflammatory mediators through JNK and NF-κB pathway.From this result, the inhibitory effect of EtCR on DSS-induced colitis was considered to the regulation of JNK and NF-κB signaling pathway.
To confirm expression level of TFF3, LPS-stimulated human epithelial cell line HT29 was used.Once mucosal layer is depleted, epithelial cells are exposed directly to luminal contents such as bacteria and their metabolites [49].In this experiment, LPS was used as an inflammatory stimulator to intestinal epithelial cells.It has been known that LPS reduces TFF3 expression by activating NF-κB in HT29 cells [50].Consistent with this report, LPS inhibited the expression of TFF3, but treatment of EtCR on HT29 cells recovered the expression of TFF3 (Figure 6).TFF3 antibody was purchased from Invitrogen (Carlsbad, CA, USA).

Ethanol Extract of CR (EtCR)
CR was purchased from Omniherb (Uiseong, Republic of Korea).EtCR was obtained by extracting dried 100 g of CR in 1 liter of 70% ethanol for approximately 3 h in heating mantle with reflux condenser.After eliminating solvent with rotary evaporator under reduced pressure (Rotary evaporator Model NE-1, Tokyo, Japan), the remnant was filtered and lyophilized (Freeze dryer FD-1, Japan) at -56°C and 9 mm Torr to acquire herbal sample.The yield from the starting material was about 5%.The obtained sample was diluted using ethanol and then filtered through a 0.22-µm syringe filter.

Induction of DSS-induced Colitis in Mouse Model
Female BALB/c mice (6 weeks old) were purchased from Samtaco Science (Osan,

Disease Activity Index (DAI)
DAI was calculated by scoring weight loss, diarrhea, and rectal bleeding.Each score was assessed as shown in Table 1, which was described by Murthy et al [23].Briefly, weight loss was determined as the difference between initial and final weights.Diarrhea was evaluated to degree of the fecal pellet formation.Rectal bleeding was assessed based on the presence of visible blood.After calculation using the following formula, DAI were recorded: DAI = {(weight loss score) + (diarrhea score) + (rectal bleeding score)}/2.

Colon Tissue Staining and Evaluation of Colitis Severity
All resected colon tissues were fixed in 10% neutral buffered formalin for 24 hours.After paraffin embedding, samples were sectioned to 5 μm.The sections were stained with hematoxylin and eosin (H&E) and Alcian blue/Nuclear Fast Red.Each stained images were captured by Leica Application Suite MicroSoftware (Leica Microsystems Inc, IL, USA).
Histological scores were based on epithelium and infiltration.The damages of epithelium were determined as follows: Normal morphology; 0, loss of goblet cells; 1, loss of goblet cells in large areas; 2, loss of crypts; 3, loss of crypts in large areas; 4. Infiltration of mucosal immune cell were determined as follows: no infiltrate; 0, infiltrate around the crypt basis; 1, infiltrate reaching the L. muscularis mucosae; 2 extensive infiltration reaching the L. muscularis mucosae and thickening of the mucosa with abundant edema; 3, infiltration of the L.
submucosa; 4. The total histological score was given as sum of epithelium and infiltration.Alcian Blue positive cells were counted in ten random areas using microscope (Leica, Wetzlar, Germany).

Cell Culture
RAW 264.7 cells and HT29 cells were purchased in Korean cell line bank (Seoul, Republic of Korea).The cells were cultured at 37°C under 5% CO2 in 10% FBS/DMEM supplemented with 1% penicillin/streptomycin.The cells were pretreated with 1, 10, and 100 μg/mL of EtCR for 1 h, and then stimulated with 200 ng/mL of LPS for 24 h.EtCR was diluted by ethanol whose final concentration was below 0.1%.After 24 h, 5 mg/mL of MTT solution was treated for 4 h in incubator.After washing off the supernatant, the insoluble formazan product was dissolved in DMSO and loaded into 96-well microplate.The absorbance was measured at 540 nm with a VERSA max microplate reader (Molecular Devices, Sunnyvale, CA, USA).To confirm the cytokine production from EtCR and LPS treated cells, 3 × 10 5 cells were seeded into 24-well plates.Various concentrations of EtCR were pretreated for 1 h prior to 24 h treatment with 200 ng/mL LPS.And then, cytokines, including IL-1β, IL-6, and TNF-α, were confirmed using supernatant.The production levels of IL-1β, IL-6, and TNF-α were confirmed using ELISA kits, according to the manufacturer's instructions.

NO Assay
RAW 264.7 cells were seeded (5 × 10 5 cell/well) and incubated with EtCR and LPS as already described in our study.To measure the NO content in the culture supernatant, Griess Reagent (1% sulfanilamide/0.1% N-(1-naphthyl)-ethylenediamine dihydrochloride/2.5% H3PO4) was used.Equal volume of supernatant and Griess reagent were incubated for 10 minutes at room temperature.The NO 2-level in supernatant was determined using sodium nitrite as a standard and the absorbance was measured by a microplate reader at a wavelength of 540 nm.

PGE2 Assay
To measure the amount of PGE2, 5 × 10 5 of RAW 264.7 cells were seeded in 24-well plated and treated with EtCR and LPS.The PGE2 concentration was calculated using a PGE2 ELISA kit (Enzo Life Sciences, USA) according to the manufacturer's directions.RAW 264.7 cells (5 × 10 6 cell/well) and HT29 cells (5 × 10 6 cell/well) were cultured in 6 cm dish.EtCR and LPS were treated as already described in our study.The cells were washed with PBS and then lysed by lysis buffer (iNtRON Biotech, Seongnam, Republic of Korea).

Western Blot Analysis
Nuclear extracts were isolated using the NE-PER Nuclear and Cytoplasmic Extraction Reagents kit (Thermo Fisher Scientific, Waltham, MA, United States) according to the manufacturer's protocols.The cell lysates were mixed with an equal volume of 2 × SDS sample buffer and boiled at 95 °C for 5 min.The samples were loaded and separated by SDS-page gels.
After electrophoresis, the proteins were transferred onto PVDF membranes and divided according to the size of each target proteins.The membranes were blocked by 5% skim milk for 2 hours and incubated with primary antibodies overnight at 4°C.After washing the membrane with 0.1% PBST 5 times, the membranes were incubated with horseradish peroxidase-conjugated secondary Abs for 1 hour.The membranes were washed with PBST 5 more times and the protein bands were visualized by ECL solution.Images were captured using the FluorChem E system (ProteinSimple, San Jose, CA, USA).

NF-κB Binding Assay
The nuclear extract was incubated with an immobilized oligonucleotide on a 96-well plate containing a specific NF-κB binding site and detected by a specific p65 NF-κB subunit ELISA kit (Affymetrix, Santa Clara, CA, USA), according to the manufacturer's instructions.

Conclusions
Our study demonstrated that EtCR ameliorated UC in DSS-induced mice and protected epithelial cell layer by recovering expression of TFF3.EtCR inhibited inflammatory factors in dimethylthiazol-2-yl)-diphenyltetrazoliumbromide (MTT) assay.Briefly, RAW 264.7 and HT29 cells were seeded in 24-well plates at 3 × 10 5 cells/well and the cells were incubated with various concentrations of EtCR.

Figure captions Figure 1 .
Figure captions

Figure 2 .
Figure 2. The Effect of EtCR on colon tissue damage and inflammation in colitis.(A)

Figure 3 .
Figure 3.The Effect of EtCR on cytokines LPS-stimulated RAW 264.7 cells.Cells were pre-

Figure 4 .
Figure 4.The Effect of EtCR on productions of NO and PGE2 and the expressions of iNOS

Figure 5 .
Figure 5.The Effect of EtCR on NF-κB and MAPKs in LPS-stimulated RAW 264.7 cells.

preprints.org) | NOT PEER-REVIEWED | Posted: 8 October 2018 doi:10.20944/preprints201810.0144.v1
).Before inducing colitis, mice were maintained in experimental environment (22 ± 2°C under a 12-h light/dark cycle) for one week.To induce colitis in mice, drinking water containing 5% (w/v) DSS is given for seven days.Mice were randomly divided into four groups (n = 8): the control group were fed drinking water without DSS, the DSS group fed drinking water with DSS, the EtCR group fed drinking water with DSS and oral administration of EtCR (50 mg/kg), and the 5-ASA group fed drinking water with DSS and oral administration of 5-ASA (50 mg/kg; a reference drug).EtCR and 5-ASA were suspended in 0.3% CMC water.EtCR and 5-ASA were orally administered daily during 7 days.Control group and DSS group were administered with the same volume of water.Drugs were Preprints (www.

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 October 2018 doi:10.20944/preprints201810.0144.v1
Total RNA was extracted from the cells using an Easy-Blue total RNA extraction kit (iNtRON Biotech, Republic of Korea) according to the manufacturer's instructions.First- 4.14.Statistical analysisThe results are expressed as the mean ± S.D. of independent experiments.Statistical evaluations were performed by a one-way analysis of variance (ANOVA) with Tukey's multiple comparison test.All statistical analyses were computed by SPSS statistical analysis software version 12.0 (SPSS Inc., Chicago, IL, United States).Values of p < 0.05 were considered to represent significant differences.