The Japanese herbal medicine Hangeshashinto enhances oral keratinocyte migration to facilitate healing of chemotherapy-induced oral ulcerative mucositis

Chemotherapy often induces oral ulcerative mucositis (OUM) in patients with cancer, characterized by severe painful inflammation. Mouth-washing with the Japanese herbal medicine hangeshashinto (HST) ameliorates chemotherapy-induced OUM in patients with colorectal cancer. Previously, we demonstrated that HST decreased interleukin 1β-induced prostaglandin E2 production in human oral keratinocytes (HOKs) and OUM-induced mechanical or spontaneous pain in rats. However, HST effects on tissue repair functions in HOKs remain unclear. Here, we examined the effects of HST on scratch-induced wound healing in vitro and in vivo. In vitro, HST enhanced wound healing mainly through scratch-induced HOK migration. Screening of the seven constituent medicinal herbs and their major components revealed that Scutellaria root, processed ginger, and Glycyrrhiza components mainly induced the scratch-induced HOK migration. Pharmacokinetic analyses indicated that the active ingredient concentrations in rat plasma following oral HST administration were below the effective doses for HOK migration, suggesting direct effects of HST in OUM. Mitogen-activated protein kinase and C-X-C chemokine receptor 4 inhibitors significantly suppressed HST-induced HOK migration. Moreover, HST enhanced tissue repair in our OUM rat model. Thus, HST likely enhanced OUM tissue repair through oral keratinocyte migration upon MAPK and CXCR4 activation and may be useful in patients with cancer-associated OUM.

Hangeshashinto (HST), a traditional Japanese medicine (Kampo medicine), contains the extracted ingredients of seven medicinal herbs and has been approved by Japan's Ministry of Health, Labour and Welfare as a prescribed medicine 15,21 . From the 16th century to the present, HST has been used in Japan to treat inflammatory diarrhoea, gastritis, and oral mucositis. Recently, double-blind, placebo-controlled, randomized studies have reported that repetitive mouth-washing with HST was effective for the improvement of chemotherapy-induced OUM in patients with colorectal or gastric cancer 22 . Several in vitro studies also reported that HST induced antibacterial and antioxidant effects, suggesting that these effects were involved in the mechanisms of HST with respect to the improvement of OUM [23][24][25] . Moreover, our previous studies in vitro and in vivo revealed additional mechanisms of HST in OUM, including both anti-inflammatory and analgesic effects [26][27][28] . Nevertheless, an important process during the restoration of ulcer sites is rapid repair of the ulcer site by migration and covering of the epithelium cells. However, the effects of HST on the wound-induced migration of oral keratinocytes or OUM healing (tissue repair) remain unclear.
In the present study, we therefore examined the effects of HST on wound healing using both in vitro and vivo experiments. First, we analysed the effects of HST on scratch-induced wound healing using human oral keratinocytes (HOKs) and identified the active medicinal herbs in HST and their ingredients by screening the seven constituents and their major components. Second, to reveal the actions of these active ingredients via the blood, we evaluated their pharmacokinetics rat plasma along with the effects of several inhibitors, such as against intracellular signalling molecules, on HST-mediated scratch-induced wound healing. Finally, we examined the effects of HST on OUM healing using an OUM rat model.

HSt enhances scratch-induced wound healing mainly by facilitating HoK migration in vitro.
We first examined the effects of HST on scratch-induced wound healing using HOKs in vitro. Compared with the vehicle, HOKs treated with HST (100 µg/mL) covered the wound area in a time-dependent manner ( Fig. 1a and Supplementary Movie S1). The extent of the wound healing area was then calculated as the percentage of confluence of HOKs in the wound area (Fig. 1b). HST significantly promoted scratch-induced wound healing in a time and a dose-dependent manner (Fig. 1b). Treatment with 10 and 100 µg/mL HST for 72 h significantly enhanced the scratch-induced wound healing (Fig. 1c), with the effect of 100 µg/mL HST being approximately 4-fold higher than that of the vehicle [ Fig. 1c; vehicle vs. 100 µg/mL HST (mean ± S.E.M.); 1.0 ± 0.06 vs. 3.89 ± 0.20].
In addition, we investigated the effects of HST on cell growth in HOKs. HST at a concentration of 100 µg/mL but not 10 µg/mL significantly increased the cell growth ( Fig. 1d), with this effect being approximately 1.5-fold higher than that of the vehicle [ Fig. 1d; vehicle vs. 100 µg/mL HST (mean ± S.E.M.): 1.0 ± 0.10 vs. 1.65 ± 0.09], whose growth action was relatively smaller than action of wound healing. In addition, animations of the scratched area in HOKs treated with vehicle or 100 µg/mL HST also revealed that HOKs treated with HST covered the wound area via HOK migration ( Supplementary Fig. S1). Taken together, these data suggested that HST enhanced the scratch-induced wound healing mainly by facilitating the migration of HOKs. components of scutellaria root, processed ginger, and Glycyrrhiza enhance scratch-induced HoK migration in vitro. HST (1). Thus, even Pinellia tuber constitutes at most 30% of HST. We therefore analysed the effects of 1, 10, or 30 µg/mL of the seven individual medicinal herbs on scratch-induced HOK migration to reveal which medicinal herbs in HST were involved in facilitating scratch-induced wound healing. As shown in Fig. 2, Scutellaria root, processed ginger, and Glycyrrhiza markedly facilitated the scratch-induced HOK migration in a dose-dependent manner. Pinellia tuber and jujube also significantly induced HOK migration, although to a lesser degree. Neither ginseng nor Coptis rhizome affected the scratch-induced migration (Fig. 2). Therefore, Scutellaria root, processed ginger, and Glycyrrhiza were used in subsequent analyses.
HST mitigates the severity of OUM development. No difference in body weight was observed between the control and HST-administered rats prior to and following 5-fluorouracil administration (before acetic acid www.nature.com/scientificreports www.nature.com/scientificreports/ treatment) (HST, 250.9 ± 9.3 g; control, 246.1 ± 5.3 g, n = 7 in each group). On day 6 after acetic acid treatment, the body weight was significantly increased in HST rats compared with that in control rats (HST, 274.3 ± 7.7 g; control, 260.7 ± 6.7 g). Ulcerations developed on day 2 after acetic acid treatment in both control and HST rats. The OUM score of HST-administered rats was significantly lower than that of control rats on day 6 and 7 following acetic acid treatment (Fig. 5).

HSt does not enhance the cell growth or scratch-induced migration of cancer cells in vitro.
To determine whether the stimulatory effects of HST were specific to normal (i.e., HOK) cells, we examined the effects of HST on cancer cell growth using cell lines derived from human squamous cell carcinoma of the tongue (HSC-4, SCC-25), colorectal cancer (DLD-1), and human gastric cancer (MKN-45). As shown in Fig. 6a and b, HST (1, 10, and 100 µg/mL) did not significantly enhance the cell viability of HSC-4 or SCC-25 cells, compared with the effects of vehicle treatment. In contrast, HST significantly decreased the cell growth of DLD-1 and MKN-45 cell lines (Fig. 6c, d).
In addition, we investigated the effects of HST on scratch-induced wound healing using HSC-4 and SCC-25 cell lines in vitro. Compared with the effects of vehicle treatment, HST (1, 10, and 100 µg/mL) did not significantly enhance the scratch-induced cell migration of these cell lines (Fig. 6e, f).

Discussion
Our present study, for the first time, revealed that HST enhances scratch-induced wound healing. Although HST also enhanced the growth of HOKs, this effect was smaller than the wound healing effect (Fig. 1). In addition, the Supplementary Movie. S1 showed that the scratch-induced wound healing was induced by HOK migration. Taken together, these findings suggested that the effect of HST on scratch-induced wound healing might be due to migration rather than growth of HOKs. Moreover, we identified Scutellaria root, processed ginger, and Glycyrrhiza as active constituents among the seven medicinal herbs comprising HST (Fig. 2). We also found that the active ingredients of these herbs consisted of baicalein in Scutellaria root; [  www.nature.com/scientificreports www.nature.com/scientificreports/  (Table 1). These results suggested that these ingredients could cooperatively enhance scratch-induced HOK migration.
In addition, we also confirmed that orally administered HST reduced the severity of OUM in vivo using a chemotherapy-administered OUM rat model. Consistent with this effect, topical HST (100 mg/mL) application significantly enhanced the healing of cutaneous wounds on day 14 following wound surgery ( Supplementary  Fig. S1). Notably, although oral HST administration increased body weight at day 6, at which time healing of OUM was enhanced (Fig. 5), the healing action of HST toward the cutaneous tissues was considered to be independent of body weight because of the equal effects of control (0 µg/mL) and HST (1, 10, 100 µg/ mL)-administrated rats ( Supplementary Fig. S1). These data suggested that the HST-evoked weight gain in the chemotherapy-administered OUM model might have been induced by an increase in food intake consequent to reduced OUM severity/OUM healing, which reduced the OUM-induced pain.
For the treatment of patients with cancer presenting with OUM, an important consideration would be to ensure that HST enhances neither the cell growth nor the migration of cancer cells. We found that, at the concentrations tested, HST did not increase and in some cases decreased the viability of various cancer cell lines including two derived from human squamous cell carcinoma of the tongue (HSC-4, SCC-25) human colorectal cancer (Fig. 6), nor did it increase HSC-4 and SCC-25 scratch-induced migration (Fig. 6). In addition, Miyashita T et al. have reported that HST reduced the incidence of reflux-induced esophageal cancer and the infiltration www.nature.com/scientificreports www.nature.com/scientificreports/ of macrophages in a surgical rat model 43 . Taken together, these findings indicated that HST does not promote the cancer itself, suggesting that HST may be useful for patients with cancer suffering from OUM. HST might not only ameliorate oral mucositis but also suppress growth of cancer cells, although further study is absolutely required.
In clinical practice for OUM treatment, HST was dissolved in hot water and administered by mouthwashing 22,44 . Therefore, HST may have two modes of action: directly affecting OUM, or affecting OUM following absorption in the blood. As shown in our plasma pharmacokinetic study, the doses of compounds derived from Scutellaria root, processed ginger, and Glycyrrhiza shown to be effective in the HST-induced HOK migration assay were higher than those found in the plasma ( Fig. 3 and Table 1). In addition, our previous study has shown that the doses of components of the HST solution administered in clinical practice were 1-100 µM 26 . In comparison, the dose of baicalein required for HST-induced HOK migration was 10 µM and that of the ingredients of processed ginger was 0.1-10 µM (Table 1). Therefore, the doses in the HST-induced HOK migration were the same or lower than that of HST solution in clinical practice. Taken together, these finding suggested that the wound healing function afforded by HST is evoked by the direct action of HST on OUM.
In addition, several reports have demonstrated relationships between MAPKs and CXCR4 in cell migration, which have been classified into two pathways: 1) MAPK activation increases the expression of CXCR4 and/or its ligands such as CXCL11 and CXCL12, and 2) MAPKs are involved in the cell migration induced by activation of CXCR4. With regard to the former, it has been reported that CXCR4 expression is induce d by ERK activation, with expression of CXCL-11 and CXCL-12, both of which are ligands for CXCR4, also being increased through activation of ERK and p38 31,[33][34][35] . For the latter, some reports have shown that CXCL12 evokes T-cell migration through ERK activation, with the CXCR4-activated cell migration being associated with activation of ERK and/ or p38 29,30,36,41 . It has also been reported that cell migration induced by the CXCL12/CXCR4 axis is due to the activation of JNK and p38 38 . To clarify the pathways underlying the beneficial effects of HST, studies investigating www.nature.com/scientificreports www.nature.com/scientificreports/ the role of HST and its constituents on the mRNA expression of CXCR4, CXCL11, and CXCL12 in HOK are now on-going in our laboratory.
Although chemotherapy-induced OUM is associated with the use of various anti-cancer drugs, there are not so many effective prevention methods or therapeutic modalities. HST thus constitutes a promising therapy, the efficacy of which has been demonstrated via a clinical study with a high evidence level 22,44 . Basic research findings have indicated that HST enhances OUM healing through multiple pharmacological actions (e.g., anti-oxidant, anti-inflammatory, anti-bacterial, and analgesic activities) [23][24][25] . With regard to anti-inflammatory effects, we have shown that various ingredients in HST decrease the interleukin 1β-induced prostaglandin E2 (PGE2) production www.nature.com/scientificreports www.nature.com/scientificreports/ in HOKs with multi-targeting effects, such as dual suppression of cyclooxygenase-2 expression and PGE2 metabolic activity 26 . For analgesic effects, the swab application of HST on an OUM rat model suppressed spontaneous and mechanical pain through Na + channel blockage, which was mediated by several ingredients in HST 27,28 . We now add a new pharmacological action by demonstrating that HST exhibits wound healing activity, with its pharmacological actions being regulated by different the constituent medicinal herbs (Supplementary Table 1). Taken together, these data suggested that the seven medicinal herbs comprising HST exert different pharmacological actions in the treatment of cancer therapy-induced OUM. www.nature.com/scientificreports www.nature.com/scientificreports/ Irinotecan, an anti-cancer drug, exerts an inhibitory effect on nucleic acid synthesis by inhibition of topoisomerase I 45 . One of the dose-limiting toxicities is diarrhea (beginning more than 24 h after infusion) that leads to dehydration and electrolyte imbalance, making it a life-threatening condition 46 . The mechanism of irinotecan-induced diarrhea is direct damage to the intestinal mucosa induced by SN-38, which is a metabolite and an active form of irinotecan 47 . In an animal experiment, HST exhibited protective effects against intestinal toxicity caused by irinotecan 48 . In addition, a randomized controlled trial had revealed that the grade of diarrhea significantly improved after treatment with HST in patients with non-small cell lung cancer treated with cisplatin and irinotecan 49 . HST significantly reduced the frequency of severe diarrhea (grade 3 or 4). Although further studies is needed, one of the mechanism of HST on the improvement of irinotecan-induced diarrhea could be HST-induced migration of intestinal epithelial cells.
In conclusion, the findings of the present study suggested that HST, specifically the active ingredients in its constituents Scutellaria root, Processed ginger, and Glycyrrhiza, has direct beneficial effects on OUM and enhances tissue repair through oral keratinocyte migration as likely induced by activation of MAPKs and CXCR4. This study provides scientific evidence supporting the use of HST in patients with chemotherapy-induced OUM. Further studies should be performed on the efficacy, safety and mechanisms of this promising agent for the management of mucositis in cancer patients treated with other treatment modalities. www.nature.com/scientificreports www.nature.com/scientificreports/ animals were allowed free access to water and standard laboratory food, and housed at 20-26 °C, 30-70% relative humidity, and 12 h light:12 h dark cycle (07:00 to 19:00 lights on). All experimental procedures were performed according to the Tsumura & Co. 'Guidelines for the Core and Use of Laboratory Animals' and approved by the Tsumura & Co. Laboratory Animal Committee for the ethical and experimental procedures.
Plasma pharmacokinetic analysis of HST-derived ingredients. For pharmacokinetic analysis, we focused on the main ingredients in each crude HST component and their metabolites (Table 2). To quantify the compounds, the plasma samples were analysed using liquid chromatography-tandem mass spectrometry (LC-MS/MS) following pretreatment as described below. The LC-MS/MS system consisted of an API4000 triple quadrupole mass spectrometer (SCIEX, Framingham, MA, USA) equipped with an Agilent 1100 system (Santa Clara, CA, USA).
For glycyrrhetinic acid and isoliquiritigenin quantification, 200 µL plasma was mixed with 25 µL acetonitrile, an equal volume of niflumic acid (100 ng/mL) as an internal standard, and 750 µL methanol. To prepare the calibration curve, the same volumes of various concentrations of working solution were used instead of the acetonitrile. The mixture was centrifuged (3, were housed in pairs in clear cages with wood chips under specific-pathogen-free conditions, and maintained on a 12:12 h light-dark cycle in a temperature-and humidity-controlled room (21-23 °C and 40-60%, respectively) with food pellets and water provided ad libitum, except during the experimental period. All in vivo experiments were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and approved by the Animal Experiment Committee of Kyushu Dental University (approval nos. 18-014 and 18-025). Rats were randomly selected for each experiment. Evaluation of OUM severity and cutaneous wound healing was performed in a manner blind to the experimental conditions. Generation of chemotherapy-administered OUM. According to our previous study 50 , the representative chemotherapy drug 5-fluorouracil (40 mg/kg/day in saline) was intraperitoneally administered 3 times at 2-day intervals in rats. Two days after the final administration, the labial fornix region of the inferior incisors was treated with a filter paper (3 mm × 3 mm, Whatman, Maidstone, UK) soaked in 50% acetic acid soaked for 30 s under anaesthesia (mixture of medetomidine (0.375 mg/kg), midazolam (2 mg/kg), and butorphanol (2.5 mg/kg) by intraperitoneal administration.
HST application and OUM healing analysis. For acclimation, rats were fed a powder diet instead of pellets from 1 week prior to the OUM healing experiments. From the day of acetic acid treatment, rats were fed the powder diet containing 1% HST or normal powder diet (control) until the end of the experiment.
To evaluate OUM severity, we used the following visual oral mucositis score, modified from that for a radiation-induced OUM model 50,51 : 0, normal; 0.5, possible presence of redness; 1, slight but definite redness; 2, severe redness; 3, focal pseudomembrane, without a break in the epithelium; 4, broad pseudomembrane, with a break in the epithelium within the acetic acid-treated mucosal area; 5, virtual loss of epithelial and keratinized layers over the acetic acid-treated mucosal area; and 6, severe swelling of the lower lip with OUM at a score of 5. The evaluation was performed every day under 2% isoflurane anaesthesia. Statistical analysis. The data are presented as the means ± S.E.M. Statistical analysis was performed using one-or two-way analysis of variance (ANOVA) followed by Tukey's or Sidak's multiple comparisons tests (GraphPad Prism 6, San Diego, CA, USA). p < 0.05 was considered statistically significant.