Protective Effect of Hainosankyuto, a Traditional Japanese Medicine, on Streptococcus pyogenes Infection in Murine Model

Background Streptococcus pyogenes (S. pyogenes) causes various serious diseases including necrotizing fasciitis and streptococcal toxic shock syndrome. One serious problem observed recently with S. pyogenes therapy is attenuation of the antibiotic effect, especially penicillin treatment failure and macrolide resistance. Hainosankyuto, a traditional Japanese medicine based on ancient Chinese medicine, has been used for treatment of infectious purulent diseases in Japan. In this study, we investigated the protective and therapeutic efficacy of Hainosankyuto against S. pyogenes-skin infection. Methodology/Principal Findings A broth microdilution method revealed that Hainosankyuto did not show a direct anti-bacterial effect against S. pyogenes. Force-feeding Hainosankyuto to infected mice for 4 consecutive days increased the survival rate and reduced the size of local skin lesions compared with mice fed PBS. Although we did not find the significant recovery of survival rate in Hainosankyuto administration only after S. pyogenes infection, the sizes of ulcer lesion were significant smaller after Hainosankyuto administration compared with mice fed PBS. No difference was observed in the anti-bacterial effect of Hainosankyuto between macrolide-susceptible and -resistant strains. Blood bactericidal assay showed that the survival rate of S. pyogenes using the blood from Hainosankyuto -treated mice was lower than that using the blood from untreated mice. We also found increased levels of IL-12, IFN-γ and a decreased level of TNF-α in the serum of S. pyogenes-infected mice treated with Hainosankyuto. Mouse peritoneal macrophage from Hainosankyuto-treated mice had significant phagocytic activity and increased mRNA levels of IL-12, IFN-γ and decreased mRNA level of TNF-α compared with control macrophage. Conclusions/Significance Hainosankyuto increased survival rate after S. pyogenes infection and up-regulated both blood bactericidal activity and macrophage phagocytic activity through modulation of inflammatory cytokines. Our data also suggest Hainosankyuto may be useful for the treatment of S. pyogenes infection more prophylactically than therapeutically.


Introduction
The gram-positive bacterium Streptococcus pyogenes (S. pyogenes) is the most common agent causing upper respiratory tract and skin infections [1]. It is also responsible for post-streptococcal diseases, such as rheumatic fever and glomerulonephritis, in addition to increasing the incidence of invasive infections such as streptococcal toxic shock syndrome [1].
One serious problem with S. pyogenes therapy is attenuation of the antibiotic effect. Despite the fact that S. pyogenes is always susceptible to penicillin, S. pyogenes infects up to 20% of penicillintreated patients, and half of these cases are also clinical failures [2]. In recent years, macrolide-and lincosamide-resistant S. pyogenes isolates have gradually spread worldwide [3]. Novel anti-microbial agents other than antibiotics are now being sought.
Traditional herbal medicines are being re-evaluated in the clinical field because of their relatively few side effects and suitability for long-term administration compared with synthetic medicines. Recently, several types of these herbal medicines have been used clinically for various diseases in Japan [4]. Previous reports have described the biological actions of herbal medicines on immune responses in bacterial infection [5]. Herbal medicine research determining the role of entomopathogenic fungus Cordyceps sinensis (C. sinensis) in anti-S. pyogenes therapy has been reported [6]. Production of cytokines, such as interleukin (IL)-12, interferon (IFN)-c and tumour necrosis factor (TNF)-a , and upregulation of phagocytic activity by the C. sinensis fraction might play roles in anti-tumour and anti-bacterial activities.
Hainosankyuto, a traditional herbal Japanese medicine based on ancient Chinese medicine, has been used for the treatment of infectious purulent diseases and has been suggested to have various functions. However, to date Hainosankyuto has not been investigated for its role in anti-bacterial therapy.
In the present study, we evaluated the protective efficacy of Hainosankyuto against S. pyogenes using a murine skin infection model. We pre-treated ICR mice with either Hainosankyuto or PBS and administrated Hainosankyuto or PBS after S. pyogenes infection for 2 days. We then examined serum cytokine levels, blood bactericidal activity and mortality in both groups of mice at each day. Furthermore we also evaluated the therapeutic effect of Hainosankyuto against S. pyogenes infection and the phagocytic activity of Hainosankyuto-treated mouse macrophage as the mechanism of protection.

MIC determination
MIC of Hainosankyuto was above 256 mg/mL. No antibacterial effect of Hainosankyuto against S. pyogenes was observed.

Murine infection model
In the present study, an animal model was used to test whether Hainosankyuto would provide the host with a protective effect against S. pyogenes infection. Mice in the control group started to die on day 2 after inoculation with S. pyogenes 1529, and the survival rate dropped to 0% after day 4. Mice in the Hainosankyuto-treated group started to die on day 4 after inoculation with S. pyogenes 1529, and the survival rate remained at 80% after day 7 ( Figure 1A). In addition, mice in the control group started to die on day 2 after inoculation with S. pyogenes D2TY, and the survival rate dropped to 0% after day 4. Mice in the Hainosankyuto-treated group started to die on day 5 after inoculation with S. pyogenes D2TY, and the survival rate remained at 60% after day 7 ( Figure 1B). Furthermore, as shown in Figure 2A, lesion size on day 3 after injection was smaller in the Hainosankyuto-treated mice injected with S. pyogenes 1529 compared to the control mice (p,0.05). Lesion size in the S. pyogenes-infected mice was measured, and Figure 2B represented the significant difference observed in lesion size between Hainosankyuto-treated and untreated mice (p,0.05).
We next investigated whether Hainosankyuto was useful therapeutically. Mice in the Hainosankyuto-treated group started to die on day 2 after inoculation with S. pyogenes 1529, and the survival rate remained at 33% after day 7 (p = 0.0589)( Figure 3A). In addition, mice in the control group started to die on day 2 after inoculation with S. pyogenes D2TY, and the survival rate dropped to 0% after day 4. Mice in the Hainosankyuto-treated group started to die on day 2 after inoculation with S. pyogenes D2TY, and the survival rate remained at 33% after day 7 (p = 0.0569) ( Figure 3B).Although the survival rates of Hainosankyuto-treated group were higher than those of control group, there was no significant differences between these 2 groups. However, lesion size in the S. pyogenes-infected mice was measured, and Figure 4 demonstrated the significant difference observed in lesion size between Hainosankyuto-treated and untreated mice (p,0.05).

Blood bactericidal assays
To determine whether leukocytes in Hainosankyuto-treated mice show elevated bactericidal activity, we performed S. pyogenes killing assays using mouse blood. As shown in Figure 5A, blood from untreated mice killed approximately 70% of the bacterial inoculum after a 30-min incubation at 37uC. Similar levels of S. pyogenes killing were observed in the blood from Balb/C mice (data not shown). The blood from Hainosankyuto-treated mice killed approximately 90% of S. pyogenes inoculum after a 30-min incubation at 37uC. Significant differences were observed in S. pyogenes 1529 CFU reduction between Hainosankyuto-treated and untreated mice (p,0.05) ( Figure 5A). Significant differences were observed in S. pyogenes D2TY CFU reduction between Hainosankyuto-treated and untreated mice (p,0.05) ( Figure 5B).

Analysis of serum cytokines during infection
The kinetics of appearance of cytokines, such as IL-12, IFN-c and TNF-a, the serum of infected mice with or without Hainosankyuto treatment was investigated. Mice in both groups showed significantly elevated serum IL-12 levels when compared to those before injection (p,0.05) ( Figure 6A). Mice in the Hainosankyuto-treated group showed a significantly elevated serum IL-12 and IFN-c level compared to that in control mice (p,0.05) ( Figure 6B). A significant decrease in the TNF-a level was observed in the Hainosankyuto-treated group at day 1 (p,0.05) ( Figure 6C). However, no significant difference was

Effect of Hainosankyuto on phagocytic activity
We next focussed on macrophage activity from leukocytes because macrophage plays a major role in phagocytic activity in murine infection model. To determine whether peritoneal macrophage from Hainosankyuto-treated mice show elevated phagocytic activity, we performed phagocytic assays on S. pyogenes. As shown in Figure 7A, macrophage from untreated mice killed approximately 60% of the S. pyogenes 1529 inoculum after a 30-min incubation at 37uC. The macrophage from Hainosankyuto-treated mice killed more than 99% of S. pyogenes 1529 inoculum after a 30min incubation at 37uC. Figure 7B showed that macrophage from untreated mice killed approximately 90% of the S. pyogenes D2TY inoculum after a 30-min incubation at 37uC. The macrophage from Hainosankyuto-treated mice killed more than 99% of S. pyogenes D2TY inoculum after a 30-min incubation at 37uC. Significant differences were observed in both S. pyogenes 1529 and D2TY CFU reduction between Hainosankyuto-treated and untreated mice (p,0.05) ( Figure 7A, Figure 7B).

Cytokine expression of perotineal macrophage in Hainosankyuto-treated mice
We examined the cytokine expression induced by Hainosankyuto. A group of three ICR mice were force-fed Hainosankyuto and controls were force-fed PBS. After 24 hour, peritoneal macrophage was extracted from two groups of mice and three cytokine expressions were measured by northern blotting analysis ( Figure 8A) and quantitative real time RT-PCR ( Figure 8B).  Results showed an increase in both IL-12 and INF-c mRNA expression in mouse macrophage after Hainosankyuto treatment compared with PBS control. Conversely, the level of TNF-a mRNA expression was decreasing in mouse macrophage after Hainosankyuto treatment compared with PBS control.

Discussion
To our knowledge, this is the first study on anti-S. pyogenes therapy with Hainosankyuto.
After S. pyogenes infection, Hainosankyuto-treated mice showed increased survival rate, blood bactericidal activity, serum levels of inflammatory cytokines including IL-12 and IFN-c, macrophage phagocytic activity and macrophage mRNA levels of inflammatory cytokines including IL-12 and IFN-c. These results suggest that Hainosankyuto can play an important role in protection against S. pyogenes.
Our data also potentially postulate that Hainosankyuto may be useful in treatment of clindamycin-resistant S. pyogenes. Because Hainosankyuto is not per se an antibiotic and has no anti-bacterial effect against S. pyogenes, the occurrence of Hainosankyuto-resistant S. pyogenes in future seems impossible. We speculate that Hainosankyuto may be efficacious when used against new strains of antibiotic-resistant S. pyogenes.
We also investigated the effect of Hainosankyuto against other emm-type S. pyogenes isolates (M4, M12, M28 and M49) and confirmed that Hainosankyuto had the same inhibitory effect on those isolates (data not shown). Inhibition of the growth of specific emm-type isolates and effects of Hainosankyuto may not be related. Our results imply that Hainosankyuto directly affects the host's immune system with regard to S. pyogenes infection.
Hainosankyuto is a traditional Japanese herbal medicine first prepared by Dr. Todo Yoshimasu, a 17th century Japanese medical doctor, by combining Hainoto and Hainosan, which are also traditional Japanese medicines. However, Hainosankyuto is now popular and widely prescribed for therapy in Japan. Because neither Hainoto nor Hainosan is easily available commercially, we chose Hainosankyuto for our study. Although previous studies on Hainosankyuto have not been found, one report showed the antiinflammatory effect of both Hainoto and Hainosan [7]. Both    B pathway [8]. Glycyrrhizin from Glycyeehiza is a sweet triterpenoid glycoside possessing anti-inflammatory and antiallergic activities [9,10]. Paeoniae radix from Peony root has anti-inflammatory effects in mouse [11]. Ginger root extract, Immature Orange extract, and Jujube extract have superoxide diumutase like activities [12]. Although each component has antiinflammation effect, this mechanism in detail has been unknown. Traditional Japanese herbal medicine is generally composed of several components and the interaction of them may enhance the effect of drugs. As further investigation from this perspective is needed, Hainosankyuto may have pronounced anti-inflammatory effects.
We tried to clarify the mechanism of this protective effect of Hainosankyuto. Our data suggest that the protection provided was not due to any direct anti-bacterial effect of Hainosankyuto. Augmentation of blood bactericidal activity might help the host eliminate bacteria at the infection site, and innate immunity is important in this elimination. IL-12-induced IFN-c expression provided protection against S. pyogenes infection in a murine model by enhancing innate immunity [13,14]. TNF-a suppression also plays a crucial protective role in severe S. pyogenes infection [15]. Not only blood bactericidal activity but also mouse peritoneal macrophage phagocytic activity, was enhanced in the Hainosankyuto-treated group. Hainosankyuto-treated mice showed significantly higher IL-12 and IFN-c levels and a significantly lower TNF-a level than untreated mice. Taken together, we suggested that Hainosankyuto-induced cytokine production augments bactericidal activity including phagocytic activity which in turn causes bacterial elimination.
In addition, we next tried to clarify whether Hainosankyuto would provide protection if given only after bacterial infection. Although we did not find the significant recovery of survival rate in direct Hainosankyuto administration only after S. pyogenes infection, our results showed that the sizes of ulcer lesion were significant smaller after Hainosankyuto administration compared to PBS administration. Hainosankyuto may be useful for therapy of local lesion after S. pyogenes infection. Further clinical study involving human subjects about S. pyogenes therapy using Hainosankyuto is desired. Also whether a similar effect occurs in other bacterial infections warrants further investigation.
In summary, Hainosankyuto was more effective in preventing S.pyogenes infection than in administration only after S. pyogenes infection. We suggest Hainosankyuto as a candidate for novel anti-S. pyogenes therapy.

Bacterial strains
S. pyogenes 1529 and D2TY were clinical isolates from severe invasive disease in Japan [16,17]. S. pyogenes D2TY in particular has a macrolide-resistant gene, ermB. A fresh colony was inoculated overnight on blood agar (Nihon Becton Dickinson, Tokyo, Japan) and cultured for 12 hour at 37uC. The bacteria were harvested by centrifugation and re-suspended in sterile PBS. Bacterial density was determined by measuring absorbance at 660 nm (A660). The bacterial suspension was then diluted with PBS to 10 9 CFU (colony forming unit)/mL using a standard growth curve to relate measured A660 to bacterial concentration.

Compound
Hainosankyuto was provided as a generous gift from the Tsumura Company Limited (Tokyo, Japan) (Table 1), and it was suspended in distilled water to prepare the stock solution at a concentration of 0.1 g/mL.  Table 1. Crude composition of hainosankyuto extract. Susceptibility testing MIC of Hainosankyuto was determined using the microdilution method in accordance with the guidelines of the Clinical and Laboratory Standards Institute [18]. MIC was defined as the lowest concentration showing no visible growth after 24 hour of incubation at 35uC compared to the control.

Murine model of invasive skin and soft tissue infection
The ability of S. pyogenes to cause local skin lesions in mice after subcutaneous inoculation was assessed using a procedure described elsewhere [19]. In brief, S. pyogenes was harvested after 16-hour growth on brain heart infusion agar (Eiken Chemical Co., Tokyo, Japan) containing 0.3% yeast extract (BHY agar) with appropriate antibiotics mixed in 1 mL of PBS and then centrifuged at 2000 g for 2 min. The pellets were diluted in 100 ml PBS to 1610 8 CFU and then injected under the skin surface of inbred 3-week-old female Slc:ICR mice (Japan SLC, Inc., Shizuoka, Japan) using a 25-gauge needle. The number of CFU injected was verified for each experiment by plating the bacteria on BHY agar and counting CFU. Mice were observed daily for lesion and necrosis formation. Lesion sizes (length6width) were measured with length determined as the longest dimension of the lesion. In the Hainosankyuto-treated group, mice were force-fed Hainosankyuto (0.1 mL/10 g body weight (bw)) on days 21, 0, 1 and 2 after S. pyogenes inoculation. Mice in the control group were given an equal volume of PBS and were infected using the same method. We also performed another protocol to clarify the effects of Hainosankyuto in animals developing signs. In the Hainosankyuto-treated group, mice were force-fed Hainosankyuto (0.1 mL/10 g body weight (bw)) on days 1, 2, 3 and 4 after S. pyogenes inoculation. We evaluated the mortality and the size of necrosis using same methods as prophylactic assay.

Blood bactericidal assay
Blood bactericidal assays were performed as an evaluation of whole blood bactericidal effect with some modification [19]. In the Hainosankyuto-treated group, mice were force-fed Hainosankyuto (0.1 mL/10 g body weight (bw)) on 1 day before collection. Mice in the control group were given an equal volume of PBS. Heparinised blood was collected from Hainosankyuto-treated and untreated mice not infected with S. pyogenes. Blood bactericidal assays were performed as follows; approximately 1000 CFU of strain was added to 1 mL of heparinized ICR mice whole blood. The samples were incubated at 37uC on a rotator. Aliquots were removed and plated on BHY agar to determine the number of CFU after 0, 30 and 60 min. Data were expressed as bacterial CFU counts at each times.

Analysis of serum cytokines during infection
We determined the levels of serum cytokines, such as IL-12, IFN-c and TNF-a during the course of infection in S. pyogenesinfected ICR mice treated with Hainosankyuto. Blood from both infected and treated/infected groups was collected under anaesthesia at 1 and 3 days after infection. Each blood sample was centrifuged, and the resultant serum was stored at 280uC until the day of assay. IL-12, IFN-c and TNF-a levels were determined by sandwich ELISA (Quantikine, R&D Systems Inc., MN, USA). For the assay, combinations of capture and biotinylated mAb were used as recommended by the manufacturer. Cytokine levels were calculated using standard murine recombinant cytokine curves run on the same immunoplate.

Macrophage cells
In the Hainosankyuto-treated group, mice were force-fed Hainosankyuto (0.1 mL/10 g body weight (bw)) on 1 day before peritoneal lavage. Mice in the control group were given an equal volume of PBS. Mouse peritoneal macrophage cells were harvested by lavage with 6 ml of cold Hank's balanced salt solution (Wako pure chemical industry, Osaka, Japan) from the peritoneal cavities. The cells were washed twice and suspended in culture medium to give 10 6 cells/mL. Five hundred mL of the cell suspension was dispensed into 1.5 mL tube and incubated for 2 hour at 37uC in 5% CO 2 . The culture medium was removed gently by aspiration and changed to fresh culture medium. For cell culture, RPMI-1640 (Wako pure chemical industry) supplemented with 5% inactivated fetal bovine serum was used. All cells were suspended in the culture medium, plated onto 24-well culture plate at dose of 6610 5 cells per well (final volume, 0.6 mL) and cultured in a humidified chamber at 37uC with 5% CO 2 and 95% air. The 5 hour culture was washed three times with Hanks' balanced salt solution to remove nonadherent cells and the adherent cells only were used as the macrophage culture.

Macrophage phagocyte assay
Mouse peritoneal macrophage were mixed with S. pyogenes and incubated for 1 hour with shaking at 37uC. Briefly, macrophage phagocyte assays were performed using polypropylene tubes containing 100 ml mouse macrophage (2610 5 cells) and 100 mL S. pyogenes to yield a final concentration of 10 5 CFU/mL. S. pyogenes was opsonized with mouse serum for 30 min before mixing. The samples were incubated at 37uC on a rotator, and aliquots were removed for quantitative and plated on BHY agar to determine the number of CFU after 0, 30 and 60 min. Data were expressed as bacterial CFU counts at each times.

Quantitative real time RT-PCR
Five mg of RNA was reverse transcribed using SuperScript firststrand synthesis system (Invitrogen Co., Carlsbad, CA, USA). The cDNA samples were then subjected to PCR analysis. Primers used were described as above (Northen blotting analysis section). To quantitate mRNA, a real-time RT-PCR was performed by using the GeneAmp 7900 sequence detection system with SYBR green reagent (Life technologies Corp., Carlsbad, CA). Samples were subjected to 40 cycles of amplification consisting of 95uC for 15 s followed by 60uC for 1 min according to the manufacture's instructions. Each assay was normalized to HPRT mRNA.

Statistical analysis
Comparison between groups regarding blood bactericidal and cytokine assays were made using the t-test. Survival data were assessed by Kaplan-Meier survival analysis and tested for significance using the log-rank test. Lesion size data were also tested for significance using the unpaired two-tailed t-test. A p,0.05 were considered significant.