In Vivo Anti-Influenza Virus Activity of Japanese Herbal (Kampo) Medicine, “Shahakusan,” and Its Possible Mode of Action

Background. A Kampo medicine, Shahakusan (SHS), has been prescribed in late phase of infection that causes inflammations in the lung. But effect of SHS on viral infection in respiratory tract has never been reported. Objectives. To evaluate anti-influenza virus activity of SHS and its mode of actions through immune systems. Methods. SHS (0.3 g/kg/day) was orally administered to BALB/c miceforupper (URI) or lower respiratory tract infection (LRI) of influenza virus A/PR/8/34. The virus titer of nasal lavage fluid (NLF) at 5 or 2 day postinfection (p.i.) and cytokine mRNA expressions in mandibular lymph node or lung at 5 or 4 day p.i. were evaluated for URI or LRI, respectively. The histopathological examinations of lung tissue and NK cell activity in the splenocytes were also evaluated at 4 day p.i. on LRI. Results. When SHS was administered from 7 days before to 4 days p.i. for URI, the virus titer was significantly decreased in comparison with water-treated control, and IL-4, IL-1β, and IL-10 mRNA expression was decreased, but IL-12A mRNA expression was increased. Administration of SHS from one day before to one day p.i. for LRI significantly decreased the virus titer. SHS also decreased infiltration of inflammatory cells in the bronchoalveolar spaces and damage of desquamated mucosal epithelia of bronchiole, decreased IP-10 mRNA expression, and increased NK cell activity. Conclusion. SHS has no direct effect on influenza virus infection but exerts antiviral effect in mice by its immunomodulating activity through action of NK cells and anti-inflammatory activity in the lung.


Introduction
Influenza is a viral infection that affects mainly upper respiratory tract. Influenza infection usually lasts for about one to two weeks and is characterized by sudden onset of high fever, followed by myalgia, headache, and severe malaise, nonproductive cough, sore throat, and rhinitis [1]. Most infected people recover within one to two weeks without requiring medical treatment. However, the high risk of complications occurs among children younger than age two, adults older than age 65, and people of any age with underlying diseases such as chronic heart, lung (bronchial asthma, etc.), kidney, liver, blood, and metabolic diseases (diabetes, etc.), or compromised immune systems (AIDS, etc.) [1]. It has been reported that around 80% of mortality by influenza virus infection is caused by pneumonia [2,3].
Anti-influenza virus drugs can be classified into M2 protein inhibitors such as amantadine and rimantadine, which are effective for only influenza A virus, and inhibitors for influenza virus neuraminidase such as oseltamivir and zanamivir, which have been used for the treatment of influenza A and B viruses. However, occurrence of drugresistant strains and several side effects associated with these drugs have been reported [3,4].
Traditional herbal medicines in Far Eastern countries have played an important role in health care of this area, 2 Evidence-Based Complementary and Alternative Medicine especially in Japan, China, and Korea [5]. Among Japanese herbal (Kampo) medicines, Maoto (Ma-huang-tang in Chinese) is one of the formulas which has traditionally been prescribed to patients with influenza in Japan, and it has reported that Maoto shows an antipyretic effect clinically [6][7][8]. However, based on traditional Kampo medicine theory, Maoto containing Ephedrae Herba as one of component herbs is not suitable to use for patients with sweating or general debility and is suitable to use for them in early phase of infection.
A Kampo medicine, Shahakusan (SHS, Xie-bai-san in Chinese, Sabaeksan in Korean) was originally described in a classical Chinese book for treatments of infantile diseases, Xiao'er Yaozheng Zhijue (Key to Differentiation and Therapeutics of Children's Diseases) [9] and has been also used in Japan [10] although it is not so popular. SHS has been prescribed for the treatments of infantile pneumonia, bronchitis, and early stage of measles. SHS is composed of four herbs: Mori Cortex, Lycii Cortex, Glycyrrhizae Radix, and Oryzae Fructus, and has been used in late phase of infection that reveals inflammations due to heat accumulating in the lung [9]. The four component herbs show mild actions, thus are considered suitable for patients with general debility such as children and elderly people. Although in vivo anti-inflammatory effect of SHS on the lipopolysaccharide (LPS) induced murine acute lung injury model [11], and a few in vitro anti-inflammatory activity have been reported [12,13], effect of SHS on viral infection in respiratory tract has never been reported.
In present study, because SHS had no direct in vitro antiinfluenza virus effect, we examined in vivo its oral antiviral effect in mice. If SHS shows in vivo anti-influenza virus activity, it has possibilities to show host-mediated effects such as enhancements of humoral immunity and/or cytotoxic NK cell activity, and/or suppression of inflammation caused by the viral infection. In order to evaluate these hypotheses, the effects of SHS on the immune functions were also studied.  University in Tokyo, Japan. SHS extract was prepared as follows: a mixture of Mori Cortex (4.0 g), Lycii Cortex (4.0 g), Glycyrrhizae Radix (2.0 g), and Oryzae Fructus (2.0 g) was extracted with 600 mL of distilled water until the volume to half. The aqueous extract was filtered (GFP; Kiriyama glass Co., Tokyo, Japan) and the filtrate was lyophilized (yield: 14.4%). The lyophilized material was suspended in distilled water and used for oral administration to mice (0.3 g/kg/day).

Three-Dimensional High Performance Liquid
Chromatography (3D-HPLC) Analysis. SHS extract (10 mg) was dissolved in water (1 mL) with Vortex mixier and sonication. The solution was filtered using a DISMIC-13HP PTFE membrane filter (0.45 μm, Advantec, Tokyo, Japan). Analysis of the filtrate was performed with an Agilent 1100 3D-HPLC system (Agilent Technologies, Inc., Santa Clara, CA, USA) equipped with a photodiode-array detector. A TSK gel ODS-80Ts column (4.6 × 250 mm) (Tosoh Co., Tokyo, Japan) was kept at 40 • C. Elution of the mobile phase was carried out by a linear gradient of 10 mM phosphoric acid-acetonitrile (95 : 5 → 5 : 95) for 60 min at a flow rate of 0.8 mL/min. A 3D-HPLC profile of an aqueous SHS extract was shown in Figure 1. The analysis based on ultraviolet (UV) absorption clearly showed the presence of the following major constituents in SHS: liquiritin, liquiritigenin, and glycyrrhizin (originating from Glycyrrhizae Radix), and other constituents were identified their origins of the component herbs by the comparison with 3D-HPLC profile of each aqueous extract from four component herbs. The 3D-HPLC profile clearly showed the major constituents derived from three component herbs except Oryzae Fructus ( Figure 1).    Chemicals Inc., Toronto, Canada) (50 μL) and phosphate buffered saline (PBS; 50 μL) were added to the culture supernatant (50 μL), and incubated at 37 • C for 10 min in Labsystems Fluoroskan II (Dainippon Sumitomo Pharma, Osaka, Japan). The monolayers of MDCK cells in the culture plate were washed with PBS, and the viable cells were determined by a colorimetric method, which is based on the in situ reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide (MTT) by viable cells [15] according to the procedure of Watanabe et al. [16]. One hundred μL of MTT (1 mg/mL) in PBS was added to each well of the monolayers, and the contents were incubated for 2 hours at 37 • C. The resulting formazan precipitate was dissolved with isopropanol containing 0.04 M HCl (100 μL/well), and the absorbance of the solution was determined spectrophotometrically at 570 nm with Infinite 200 M microplate reader (Tecan, Männedorf, Switzerland). Oseltamivir (Tamiflu, Roche, Basel, Switzerland) (20 μL/well, final concentration of 10 μg/mL) was used as a positive control of an antiviral agent for influenza virus A/PR/8/34. For cytotoxicity assay, test sample solutions prepared in the sterilized water, were added to the confluent MDCK cells without virus as described above. The cells were cultured at 37 • C with 5% CO 2 for 3 days, and the number of viable cells was determined using the MTT assay as described above.  (Figure 2(a)). At 5 day p.i., serum specimens were prepared from mice by drawing blood from the infra-axillary artery. A nasal lavage fluid (NLF) was obtained by washing the nasal cavity of the head with 2 mL of PBS containing 0.1% BSA. A bronchoalveolar lavage fluid (BALF) was obtained by injecting 2 mL of the washing solution twice into the trachea and lungs, which were separated from the body. A mandibular lymph node was harvested and stored at −80 • C. Lower respiratory tract infection of influenza virus: mice were anesthetized by intraperitoneal injection of Somnopentyl, and then infected intranasally by dropping 20 μL of influenza virus suspension (2 LD 50 ) in PBS containing 0.1% BSA into one nostril. SHS extract was administered orally to the mouse from one day before to one day p.i. or 3 days p.i. of the virus (Figure 2(b)). At 2 day p.i. and 4 day p.i., BALF, NLF, and serum samples were prepared as previously described. The lung specimens were stored at −80 • C.

Plaque Assay.
Influenza virus titers were determined with plaque assay according to the method described by Kiyohara et al. [18]. Briefly, duplicate cultures of MDCK cells in a 6-well plastic plate (BD Falcon, Franklin Lakes, NJ, USA) were exposed to 100 μL of BALF and NLF samples for 30 minutes at 37 • C. The cells were overlaid with 2 mL of medium containing 3 μg/mL acetyltrypsin, 0.01% DEAEdextran hydrochloride (Sigma-Aldrich Co.), 0.1% glucose, and 0.8% agar (Ina Food Industry Co., Nagano, Japan) followed by cultivation at 37 • C for 2 days. The cells were  overlaid with 1 mL of the above medium containing 0.01% neutral red (Wako Pure Chemical Industries, Osaka, Japan) and cultured at 37 • C for one day. The number of plaques was counted.

Determination of Anti-Influenza Virus
Antibodies. The titers of IgA, IgG 1 and IgM antibodies specific for influenza virus were determined using a reverse-phase fluorescence ELISA [19] as follows. Each well of 96-well Immulon 4 EIA plates (Thermo Fisher Scientific, Waltham, MA, USA) was coated with 100 μL of capture antibody (BD Biosciences Pharmingen, San Jose, CA, USA) (1 μg/mL purified anti-mouse IgA, IgG 1 , or IgM antibody in 0.01 M sodium carbonate-bicarbonate buffer [pH 9.5] containing BSA 10 μg/mL) and incubated at 37 • C for 3 hours. Then, the capture antibody was discarded and 300 μL of blocking reagent (1% fat-free milk in PBS) was added to each well and incubated for one hour at 37 • C. After incubation, all the 2.9. NK Cell Activity. The NK cell activity of mouse splenocytes was assessed using flow cytometry [20] as described previously [21]. The level of target cell lysis was determined using a Cytomics FC500 flow cytometer (Beckman Coulter, Brea, CA, USA), and the NK cell activity was expressed as the percentage of effector cell-specific lysis. tissue paraffin sections stained with hematoxylin and eosin were ordered to Mitsubishi Chemical Medience Co. (Tokyo, Japan).

Quantitation of mRNA Expression in Mandibular Lymph
Node and Lung. Total RNA was extracted using Sepasol-RNA I Super G (nacalai tesque, Kyoto, Japan) from the mandibular lymph node and lung. Total RNA (0.5 or 2.0 μg) was then reverse transcribed using ReverTra Ace (TOYOBO, Osaka, Japan) according to the manufacturer's protocol.

Statistical Analysis.
The significance of differences between three experimental groups was analyzed with oneway ANOVA followed by post hoc multiple comparison. The software used was KaleidaGraph Ver.4.0 (Synergy Software, HULINKS Inc., Tokyo, Japan). The significance of differences between two experimental groups was analyzed with an independent t-test. A probability (P) value <0.05 was considered significant.

Effects of SHS on Influenza Virus Infection in MDCK Cells.
To test the effect of SHS on the proliferation of influenza virus and its cytotoxic activity, the MDCK cells were infected with A/PR/8/34 (H1N1) virus or mock-infected and cultured in the presence of various concentrations of SHS (0.39 to 100 μg/mL) or under a drug free condition. Oseltamivir showed a potent antiviral activity against the viral infection at 10 μg/mL. However, the negligible effects of SHS were shown on the proliferation of virus (Figure 3(a)) and survival of infected MDCK cells (Figure 3(b)), and SHS showed very low cytotoxicity against mock-infected MDCK cells (Figure 3(b)).

Effect of SHS on the Proliferation of Influenza Virus in the Nasal Cavity of Mice.
When SHS was administered orally to BALB/c mice at a dose of 0.3 g/kg/day from 7 days before to 4 days post infection (p.i.) of influenza virus, the infectious virus titer of the NLF at 5 day p.i. was decreased significantly compared to the water-treated group (Figure 4). Oseltamivir showed a potent antiviral effect under the same condition. When SHS was administered to the mice at the same dose from one day before to 4 days p.i., the virus titer of the NLF observed a tendency of decrease compared with the watertreated control. However, when SHS was administered orally to the mice at the same dose from 2 hours p.i. to 4 days p.i., no antiviral effect was observed (Figure 4). titers in NLF and BALF, IgG 1 antibody titers in serum and BALF, and IgM antibody titers in NLF were not enhanced on SHS-treated mice in comparison with those of water-treated group (data not shown).

Effects of SHS on Cytokine mRNA Expressions in the Mandibular Lymph Nodes of Mice.
To understand the responses to SHS treatment on cytokine mRNA level, realtime quantitative PCR analyses of the mandibular lymph nodes were performed. At 5 day p.i., IL-4, IL-1β, IFN-γ, and IL-10 mRNA expressions on influenza virus-infected mice were significantly increased in comparison with those of mock-infected groups, but the IL-12A mRNA expression was decreased significantly ( Figure 5).
When SHS was administered orally from 2 hours p.i. to 4 days p.i., IL-4 mRNA expression was decreased significantly in comparison with that of water-treated group. IL-1β mRNA expression showed a tendency of decrease by the administration of SHS. Whereas oral administration of SHS 10 Evidence-Based Complementary and Alternative Medicine for 5 days (same duration from 2 hours p.i. to 4 days p.i.) did not affect IL-4 and IL-1β mRNA expressions in mock-infected mice ( Figure 5). IL-10 mRNA expressions also showed a tendency of decrease by the administration of SHS from 2 hours p.i. to 4 days p.i. On the contrary, IL-12A mRNA expression showed a tendency of increase when SHS was administrated from one day before to 4 days p.i. compared with the water-treated group ( Figure 5). However, IFN-γ and IL-12B mRNA expressions on SHS-treated mice were not changed in comparison with those of water-treated group.

Effects of SHS on Lower Respiratory Tract Influenza Virus
Infection in Mice

Effects of SHS on the Proliferation of Influenza Virus in the Nasal and Bronchoalveolar Cavities of Mice.
When SHS was administered orally to the mice at a dose of 0.3 g/kg/day from one day before to one day p.i., the infectious virus titer of the NLF was significantly reduced than that of water-treated group at 2 day p.i. (Figure 6). When SHS was administered orally to the mice at the same dose from one day before to one day p.i. or 3 days p.i. (Figure 2(b)), the virus titer of the BALF showed a tendency of decrease compared to those of water-treated group both at 2 day p.i. and 4 day p.i. (Figure 6). However, the virus titer of NLF from SHS-treated mice at 4 day p.i was seemed to decrease in comparison with that of water-treated group, but it was not significant.

Effects of SHS on Anti-Influenza Virus Antibody Titers of
Mice. At 4 day p.i., SHS did not enhance the anti-influenza virus IgA antibody titers in NLF and BALF, IgG 1 antibody titers in serum and BALF, and IgM antibody titers in BALF of the virus infected mice compared to water as control (data not shown).

Effect of SHS Treatment on the Lung Tissue of Mice.
At 4 day p.i., representative histological sections of lungs were harvested and stained with hematoxylin and eosin. The histological image of the control lung exhibited normal pulmonary and bronchiolar architectures and resident cells (Figure 7(a)). Influenza virus infection induced marked infiltration of inflammatory cells on the perivascular and peribronchiolar areas (Figure 7(b), thin arrow), bronchial dilatation, and desquamated mucosal epithelia of bronchioles in the lung tissue (Figure 7(b), open arrow). However, these influenza virus induced pathological findings in lung were attenuated by the administration of SHS (Figure 7(c)) as similar as these of control mock-infected mice. water-treatment at 2 day p.i. and 4 day p.i., but significant decrease was shown only at 4 day p.i. (Figure ??). IP-10 protein level in BALF was also seemed to decrease by the administration of SHS compared to the water-treatment only at 2 day p.i., but the decrease was not significant (Figure ??).

Effects of SHS on the NK Cell Activity in Splenocytes of Mice.
When SHS was administered orally for 5 days without influenza virus infection, a tendency of increase of NK cell activity in splenocytes was observed in comparison with that of the water-administered control ( Figure 9). When influenza virus was infected to the mice, the NK cell activity in SHS-treated mice was enhanced significantly compared to that of water-treated group at 4 day p.i. (Figure 9).

Discussion
Some of Kampo medicines have been reported their in vivo effects for influenza. Antipyretic potency of Maoto (Mahuang-tang in Chinese) in human [8,22] and Gingyosan (Yin-qiao-san in Chinese) in mice [23,24] have been reported, Kakkonto (Ge-gen-tang in Chinese) has increased IL-12 production in the BALF of mice [25], and Shoseiryuto (Xiao-qing-long-tang in Chinese) has augmented anti-influenza virus IgA antibody followed by alleviation of the proliferation of influenza virus in the respiratory tract of mice [26][27][28]. Based on Kampo medicine theory, these formulas have been used in early phase of infection. It has also been reported that Hochuekkito (Bu-zhong-yiqi-tang in Chinese), which has been used for the recovery of debilitating condition, has anti-influenza virus activity in mice [29], and Unpito (Wen-pi-tang in Chinese) for the recovery of fatigue condition alleviates inflammation in the lung of influenza virus infected mice by inhibition of xanthine oxidase (XOD) [30].
Shahakusan (SHS) has been used in late phase of infection that caused inflammations due to heat accumulating in the lung [9], and this formula shows mild actions, thus is considered suitable for patients with general debility such as children and elderly people. However, effect of SHS for influenza virus infection has never been reported.
Present study clarified that SHS had a negligible effect on the proliferation of virus and survival of infected MDCK cells (Figure 3), suggesting that SHS has no direct effect for influenza virus infection. However, present study clearly showed that the prior oral administration of SHS can protect the proliferation of influenza virus A/PR/8/34 (H1N1) in the respiratory tract of mice (Figures 4 and 6), but SHS did not enhance antiviral antibodies in NLF, BALF and serum of the infected mice both on the upper and lower respiratory tract infections. These results suggest that SHS shows hostmediated anti-influenza virus activity through the action mechanism which is different from that of Shoseiryuto.
Present study also showed that IL-1β mRNA expression in mandibular lymph node (one of regional lymph node of nasopharynx) was increased by influenza virus infection but downregulated by the administration of SHS from 2 hours p.i. to 4 days p.i. on upper respiratory tract infection ( Figure 5). A proinflammatory cytokine IL-1β is an important mediator of the inflammatory response, and is involved in a variety of cellular activities, including cell proliferation, differentiation, and apoptosis [31]. IL-1β is also readily produced by influenza A virus-infected leukocytes [32]. The report for anti-inflammatory effect of SHS on the LPS induced murine lung injury model also showed decrease of IL-1β mRNA expression in lung and protein level in BALF [11]. In present study, SHS alone did not affect the IL-1β mRNA expression in mandibular lymph node of mock-infected mice ( Figure 5). These results suggest that decrease of IL-1β mRNA expression by the action of SHS in present study may occur as a result of suppression of inflammation in nasal cavity but not direct suppression by SHS. With these results, SHS may have the potential to suppress inflammation due to influenza virus infection. IL-4 is a cytokine that induces differentiation of naive helper T cells (Th0 cells) to Th2 cells [33]. IL-10 is produced primarily by monocytes and a lesser extent by lymphocytes, and downregulates the expression of Th1 cytokines [34]. It has been reported that when the influenza virus was infected to the mice, increases of IL-4 and IL-10 mRNA expressions were observed in the lung [35]. In present study, IL-4 and IL-10 mRNA expressions also increased in the mandibular lymph node of mice by influenza virus infection, but SHS did not affect the IL-4 mRNA expression in mockinfected mice ( Figure 5). IL-12 is a heterodimeric cytokine encoded by two separate genes, IL-12A (p35) and IL-12B (p40). It is known that IL-12 activates on T and NK cells, and has broad biological activities. This cytokine has been found to be important for sustaining a sufficient number of memory/effector Th1 cells to mediate long-term protection to an intracellular pathogen [36] and induces a cell-mediated immune response [37]. Present study showed that IL-4 and IL-10 mRNA expressions were downregulated, and IL-12A mRNA expression was upregulated in mandibular lymph node by SHS administration to the influenza virus-infected mice ( Figure 5). These results suggest that SHS measurably influenced Th1/Th2 balance and activated Th1 cells in nasopharynx. SHS may inhibit the proliferation of influenza virus in nasal cavity by the activation of cell-mediated immune response through the activation of Th1 cells. In present study, administration of SHS from one day before the virus infection inhibited the proliferation of influenza virus in nasal cavity at 5 day p.i., but the administration from 2 hours p.i. did not inhibit on upper respiratory tract infection ( Figure 4). IL-12A mRNA expression was increased when SHS was administered from one day before to 4 days p.i. on upper respiratory tract infection ( Figure 5). Therefore, these results suggest that the increased IL-12A mRNA expression involves in the antiviral effect of SHS as a possible action. It has been reported that when another Kampo medicine, Kakkonto was administered orally to the mice from one day before influenza virus infection, the level of IL-12 in the BALF was increased on 2 day p.i. and virus yield in BALF was reduced on 3 day p.i. [25]. These observations suggest that at least a part of the anti-influenza virus activities of SHS and Kakkonto cause by upregulation of IL-12 on respiratory tract. Further, bronchial asthmatic patients are one of groups at high risk of influenza virus infection, and IL-4 is a typical Th2-type cytokine that plays a key role in human allergic asthmatic responses [38]. The decrease of IL-4 mRNA expression by the action of SHS suggests that SHS also has a potential to protect and/or treat diseases such as severe asthma.
The histological experiment suggests that SHS has an anti-inflammatory effect in the lung of mouse on the lower respiratory tract infection (Figure 7). IP-10 (CXCL10) is a Th1 cell-related chemokine which is important in the recruitment of Th1 cells involved in host immune defense against intracellular pathogens such as viral infection [39]. IP-10 expression level is known to associate with inflammatory diseases including infectious diseases [40]. The decrease of this chemokine mRNA expression was observed in the lung of SHS-treated mice on lower respiratory infection ( Figure 8). Therefore these results suggest that SHS shows anti-inflammatory effect through the decrease of IP-10 in the lung.
Present study also showed that oral administration of SHS augmented NK cell activity in splenocyte of the lower respiratory tract infected mice and also non-infected mice ( Figure 9). In this respect, SHS might augment cellular immunity and protect influenza virus infection, because NK cells are lymphocytes of the innate immune system that play a crucial role in the early host defense against various virus infections [41,42]. We could not measure NK cell activity in the lung of influenza virus-infected mice after SHS administration in present study, because it was hard to collect enough amounts of lymphocytes from lung of mouse for NK cell assay. Elucidation of effect of SHS on NK cell activity in the lung must await further study.
Glycyrrhizin is a major component of Glycyrrhizae Radix, one of component herbs of SHS, and its in vivo anti-influenza virus activity has been reported as a result of i.p. administration to the mice [43]. However, in vivo anti-influenza virus activities of the other component herbs of SHS have never been reported. Although oral effect of glycyrrhizin has not been reported, it has a possibility to be one of active substances in SHS. Elucidation of orally active anti-influenza substances in SHS and further action mechanism of anti-influenza virus activity of SHS are now in progress by our group.

Conclusions
Consequently, this study shows that SHS has in vivo antiinfluenza virus activity for the first time. In addition, SHS exerts its anti-influenza virus effects in mice through its immunomodulating activity such as NK cell activity in splenocytes and anti-inflammatory activity in lung.