Fifteen healthy volunteers aged from 19 to 59 years were recruited and were administered HF for 15 days. Fifteen milliliters of blood were drawn from the forearm vein one hour before the first administration of HF and 15 days after the last HF administration (day 30). All volunteers provided informed consent prior to participation for this trial. This study was approved by Ethics Committee of Kanazawa Medical University.

Dried haemopoietic formula SDT and BYT in granules were chosen as representatives of HF [5] -[12] . Dang- Gui-Liu-Huang-Tang (dRHT) and Syao-Chin-Rong-Tang (sCRT) were chosen as controls. A dose of 5.25 g dried powder of each formula (supplied by Tsumura Pharmaceutical Co. Ltd., Japan) was administered two times per day for 15 days. Physiological functions were checked and possible side effects of the drug were inquired for all the subjects to ensure the safety of the trial.

The assessments including a total number of leukocytes was ordered to count with blood chemical test for the medical diagnosis of public institution (Ishikawa Preventive Medicine Association, Ishikawa, Japan). In the differential counting, 200 cells were counted on a May-Grünewald-Gimsa stained slide, and percentages of lymphocytes and granulocytes were determined.

The assessments including a total number of leukocytes was ordered to count with blood chemical test for the medical diagnosis of public institution (Ishikawa Preventive Medicine Association, Ishikawa, Japan). In the differential counting, 200 cells were counted on a May-Grunewald-Gimsa stained slide, and percentages of lymphocytes and granulocytes were determined [13] [14] .

The whole blood obtained from the subjects was washed twice in phosphate buffered saline (PBS). One hundred micro-liters of the suspensions were stained with 20 μl of fluorescent monoclonal antibodies (anti-human CD2⁺, CD4⁺, CD8⁺, CD11b⁺, CD14⁺, CD16⁺, CD19⁺ and CD56; antibodies). About 10,000 stained cells were re-sus- pended in PBS, then surface markers were detected by flow cytometry (FACS Calibur; Becton Dickinson Immnocytometry Systems, CA, USA).

To test whether HF affected the functional maturation of immunocytes within a short period of time, we examined the number of cytokine containing cells using FACS analysis. This method reveals cytokine producing cell by peering off the surface of lymphocyte, enable to assess the cells in a festival evening, compare than serum cytokine level that correspond to paper tips of post festival [15] .

Blood cell suspensions were cultured in phorbol 12-myristate 13-acetate (PMA), ionomycin and bovine serum albumin (BSA) (Sigma CO. Ltd., Mo, USA) for 4 - 5 h at 37˚C. Subsequently, the cell suspensions were stained with monoclonal antibodies (Percp-CD3, Percp-CD45, FITC-interferon (IFN)-γ, PE-interleukin (IL)-4, FITC-IL- 1β) and analyzed by flow cytometry. All antibodies used in this study were purchased from Becton Dickinson Immunocytometry System (CA, USA).

In the animal model of immuno-competency reduction, male C57BL/6J mice, aged 8 - 9 weeks, were injected with Mitomycin-C (MMC) (5 mg/kg) to inhibit the bone marrow. Then, HF extracts was administered orally at a dosage of 1 g/kg/day for five consecutive days. tRHT and sCRT were chosen as controls [16] -[24] .

The bone marrow-suppressed mice were administered herbal decoction HF 1g/kg dairy for 5 days and after 1 week later, their blood were withdrawn from their tail vain. Then, the number of leukocytes was counted in Bürker-Türk solution.

Bone marrow-suppressed mice were administered with herbal decoction of dLHT (1g/kg/day) for five days. One week later, the blood from their tail vain was withdrawn. Then the granulocyte and lymphocyte subsets were counted in Bürker-Türk solution.

Cells from peritoneal exudates were collect from the peritoneal cavity of bone marrow-suppressed mice. Phagocytes were purified using adherent technique to get cell suspensions which contained more than 95% of phagocytes. The purified cells were loaded to the upper room of Boyden chamber to test migration ability at a concentration of 1 × 10⁴ cell/ml. Human serum treated at 56˚C for 30 min was for the chemo tactic agent of mouse phagocyte [25] .

The same cells suspension was purified by adherent technique for phagocyte, which produces cells contained more than 95% of phagocytes. The purified cells were adjusted to 1 × 10⁴ cell/cm² and mixed with latex beads that are 5 μm in granule with fluorescence isochianate. After 90 min of incubation, remained granule were washed out from the glass slide. Number of phagocytic cell and their ability to catch up the latex beads were automatically measured by ACAS system, which outputs the result in a digital form (Adeherent cell activity evaluating system; Shimazu, Kyoto, Japan).

The same phagocyte suspension, which contained over 95% of phagocytes, was produced by adherent technique for phagocyte. The purified cells were adjusted to 1 × 10⁵ cell/ml to examine the macrophage activity of killing by the nitroblue tetrazolium (NBT) reduction test [15] .

The bone marrow suppressed mice were administered herbal decoction of HF (1g/kg/day) for 5 days. One week later, mice were immunized with sheep red blood cells, (2 × 10⁸/mouse) intraperitoneally. Five days later, their spleen cells were collected. Plague-forming cells (PFC) were developed, and the ability of IgM and IgG antibody production was tested by the method reported by Jerne and Nordin [19] [20] .

Whole blood obtained from the subjects was washed twice with PBS. One hundred micro-liters of the suspensions were stained with 20 μl of fluorescent monoclonal antibodies (anti-human CD2⁺, CD4⁺, CD8⁺, CD11b⁺, CD14⁺, CD16⁺, CD19⁺ and CD56⁺ antibodies). Ten thousands stained cells were re-suspended in PBS to detect surface markers by flow cytometry (FACS Calibur; Becton Dickinson Immnocytometry Systems, CA, USA).

The blood cell suspensions were cultured with PMA (phorbol 12-myristate 13-acetate), ionomycin and BSA (bovine serum albumin) for 4 - 5 hours at 37˚C. After that, the cell suspensions were stained using the monoclonal antibodies of PE-IL-4, FITC-IFN-g and FITC-IL-1b. Then they were analyzed by the FACScan (Becton Dickinson Co. Ltd. USA). The antibodies and reagents used in the test were purchased from Becton Dickinson Immunocytometry system (USA).

The peritoneal exudates cells were collect from the peritoneal cavity of bone marrow-suppressed mice. Cell suspensions were purified by adherent technique for phagocyte, getting a suspension which contained over 95% of phagocytes. The purified cells were adjusted to 1 × 10⁴ cell/ml and loaded at the upper chamber of Boyden chamber for test migration. Human serum with treated at 56˚C for 30 min was for the chemotactic agent for mouse phagocyte.

Phagocytic activity and antibody production of macrophages were analyzed using a classical test that could test the total activity of the immune system by examine chemotaxis, phagocytosis and intracellular degradation of macrophage. For identifying antibody-forming cells, plaque-forming cells were detected using heterogeneous erythrocyte; sheep erythrocyte was a target antigen. Peritoneal macrophages were collected and purified in fetal calf serum (FCS)-coated petri-dishes. The cell population was approximately 97% uniform in function and morphology. These cells were applied to the nuclepore-membrane (pore size: 5 μm; Neuro Probe Co. Ltd., Cabin John MD, USA) with a chemotaxis chamber (Neuro Probe Co. Ltd.). After 90 minutes’ incubation, the membrane was vigorously washed with saline (37˚C), fixed, and then stained with methylene blue dye. After counting under a microscope for the total field of the membrane, the average number of migrating cells was expressed as cell counts/mm².

The same cells suspension was purified by adherent technique for phagocyte, which contained over 95% of phagocytes. The purified cells were adjusted to 1 × 10⁴ cell/cm² and mixed with latex beads that were 5 um in granule with fluorescence isochianate. After 90 min of incubation, the remained granule were washed out from the glass slide and counting automatically by ACAS system, which outputs digital presentation, for evaluating phagocytes in number and in their ability to catch up the latex beads (Adeherent cell activity evaluating system, Shimazu, Kyoto, Japan). Latex beads in 5 μm with fluorescence were used to test phagocytic activity and Candida albicans was cell killing activity. A macrophage-target cell ratio of 1:10 was considered to be optimum. Ten minutes after incubating phagocytes and target cells, intracellular Candida cells were cultured on an agar dish with conventional medium 1640 until the next day to perform the colony forming assay. In this way, the phagocytic ability of the macrophages was monitored. To document intracellular killing activity, the same procedures were performed excepting that the incubation time was changed to 90 minutes.

Sheep erythrocyte (SRBC), a T-dependent antigen, was used for antibody formation cell study. Ten days after tumor transplantation, each antigen was intra-peritoneally injected. After four and six days, the antibody-form- ing cells were detected using localized hemolysis in an agar gel. Plaque-forming cells were developed by the method of Jerne and Nordin [19] [20] .

Eight week-old female C57BL/6 were purchased from Sankyo Laboratory Service Corporation (Shizuoka, Japan). All mice were kept under specific pathogen-free conditions. Mice food and distilled water were freely accessible for each mouse. Housing temperature and humidity were controlled 25˚C ± 1˚C and 60% (Figure 1).

As for the basic medium, HEPES buffer (HEPES 17mM, NaCl 120mM, Glucose 5mM, KCl 5mM, CaCl₂ 1mM, MgCl₂ 1mM) was prepared and sterilized by filtration. Phorbol 12-myristate 13-acetate (PMA, Sigma, USA) was diluted to 10⁻⁶ M by dimethyl sulfoxide (DMSO, Sigma, USA) and used as a stimulant for super oxide anion generation for murine peritoneal exudates cells. Cytochrome-c (Sigma, USA) was diluted to 1mM by HEPES buffer. Since cytochrome-c reduced by super oxide showed maximum absorbance at 550 nm, we used cytochrome-c to measure the amount of super oxide anion generation through spectro-photometrical technique. Oyster Glycogen (type, Sigma, USA) was diluted in the purified water (10% w/v, Wako, Japan) and autoclaved at 120˚C for 20 min. This solution was used for intraperitoneal injection to mice in order to induce peripheral neutrophils into the abdominal cavity.

3) Modified TCM and Other Materials

Agaricus was purchased from Ohbiken Co. Ltd. (Yamanashi, Japan), Chlorella was purchased from Taiwan Chlorella Co. Ltd. (Taipei, Taiwan) and Propolis was purchased from Epimedix Co. Ltd. (Kanazawa, Japan). The herbal medicine (TCM), Shi-Quan-Da-Bu-Tang and Bu-Zong-Ye-Qi-Tang were purchased from Tsumura Co. Ltd. (Tokyo, Japan) and fermented at Ohbiken Co. Ltd. (Yamanashi, Japan). Propolis was also degradated at

Futaba Co. Ltd. (Shizuoka, Japan). Both micronized and not micronized Propolis, i.e. native Propolis were insoluble in water, so we used 10% DMSO as a solvent.

4) The Measuring the Amount of Super Oxide Anion Generated by Murine Peritoneal Exudates Cells

Each drug was orally administered to mice (500 mg/kg) for one week. Two milliliters of 10% Oyster glycogen was injected intraperitoneally 10 hours before the assay. Sufficient murine peritoneal exudative cells were induced ten hours after the stimulation. Mice were euthanized by cervical dislocation, murine peritoneal exudates cells (PEC) suspension was centrifuged twice for 5 minutes at 1500 rpm at 4˚C. Then PEC was prepared to 1 × 10⁶ cells/ml of HEPES buffer. One hundred microliters of cytochrome-c and 10 μl of PMA were added to the cell suspension and this was incubated for 20 minutes at 37˚C. The reaction mixture was then centrifuged for 10 minutes at 1500 rpm, 4˚C. An OD of supernatant was measured at both 550 nm and 540 nm, the amount of generated super oxide anion was shown in the formula; increased absorbance at 550 nm (ΔA_550-540)/19.1 × 10³ (mmol/ml). In order to ensure if we really measured the amount of generated super oxide anion or not, we tried to add super oxide anion dismutase (SOD), an enzyme for its anti-oxidative effect, into our experimental system. The result was as expected that the reduction of cytochrome-c was inhibited after the addition of SOD (Figure 2). This showed us that our experimental system could be used properly for measuring the amount of generated super oxide anion.

Data are expressed as means ± standard deviations. The differences between HF-treated and non-treated condi- tions were compared using a one-tailed analysis of variance. A P value < 0.05 was considered to be statistically significant.

Leukocyte numbers have been counted one hour before and 15 days after the treatment of hemopoitic formula. The cell number measured one hour before the administration was set as 100%. Relative percentage of cell number on the 15th day was calculated. No significant changes were observed in G-group after the administration of BYT. However, significant change was found in L-type group (Table 1).

The volunteers were healthy subject, with no drastic change for the total number of leukocytes. However, we

tried to check the regulative effect of herbal formulae for two different constitution, G-rich type and L-rich type. Analysis that mixed both groups together showed no significant differences in total leukocyte number except that for HF; in G-type group, total number of leukocytes was down regulated by SQT. This was a results of the down regulation of major group of leukocyte, granulocyte.

As for the L-type, no significant changes were found after the treatment of both HFs. In the L-type group, SQT, on the other hand, increased the tonal leukocyte and granulocyte in number, on the contrary to the down regulation for lymphocytes. To further clarify the influence of hemopoietic formula, we divided the subjects into two groups: the G-type group, who had a granulocyte count over 60%, and the L-type group, who had a lymphocyte count over 40%. In the L-type group, lymphocyte counts tended to decrease on day 15, accompanied by an increase in granulocyte numbers by SQT but not by BYT. On the contrary, the granulocyte counts of G-type group tended to decrease on day 15. The decrease of granulocyte count was raised by BYT, but not by SQT on day 15 (Tables 1-3).

After HF treatment, cell counts of CD2⁺, CD4⁺, CD8⁺, CD11b⁺, CD16⁺, CD19⁺ and CD56⁺ were tested to evaluate variations in T cells, B cells, macrophages and NK cells. These values were measured one hour before hemopoietic formula and 15 days thereafter. Our results showed that CD2 and CD4 cells were increased by both BYT and SQT. CD11b⁺ and CD14⁺ cell counts, which are closely associated with macrophage activity, increased by SDT in the L-type subjects. In particular, there was a remarkable increase in CD11b⁺ cell number on day 15. T cell subsets that are closely associated with activity of immature T cells, (CD2⁺, CD4⁺ and CD8⁺), the CD2⁺ (P < 0.05) showed an increase with the treatment of BYT 15 days after administration. The number of

CD19⁺ cells, which is closely associated with B cell activity, was not changed by both HF throughout the trial, neither were the numbers of CD16⁺ and CD56⁺ cells (Table 2).

To test whether herbal decoction affected the functional maturation of immunocytes in a short time, we investigated the number of cytokine producing/containing cells by FACS analysis. This method reveals cytokine producing cell number by peering off the surface of lymphocyte, enable to express the number of cells in festival evening, compare than serum cytokine level that correspond to the paper tips of post festival. To determine whether HF influences functional maturation of immuno-competent cells, levels of IL-1β-, IL-4- and IFN-γ-ex- pressed T cells were further examined using fluorescence-activated cell sorter analyses. There was a significant increase in the levels of IFN-γ and IL-4 containing cells after administration of SQT. The result revealed that IFN-γ expression, which increased highly on the 15th day after treatment, was different from the expression of IL-1β and IL-4, those on the other hand, exponentially increased on day 15 after the administration of SQT. The augmentation of cytokine expression was confirmed by a classical method in the lymphoid organ, i.e. antibody-forming cells and plaque-forming cells. Both HFs down-regulated IL-1β producing cells in both G-type and L-type groups (Table 3).

The body weight and thymus weight reduced in bone marrow-suppressed mice, resulting in the reduction of peripheral blood leukocyte to around 40%. After administered each herbal decoction 1 g/kg dairy for 5 days and after 1 week later, their blood were recovered to around 90% of normal value (Figure 3).

The bone marrow-suppressed mice were administered HF 1 g/kg dairy for 5 days, and one week later, their blood was withdrawn from their tail vein. The cell count of the peripheral blood is showed in Figure 3 shows that the thymus weight decreased to half of normal control after 5 mg/kg of MMC was injected. However, all the three HFs recovered thymus weight to about 70% of the control.

CD3, CD4 and CD19 cells of MMC treated mice were recovered to almost normal values after the administration of HFs. As for the functional recovery, IFN-γ and IL-4 producing cells were also recovered by the all three decoction, including HFs and a functionally depressive agent of TCM. In cytokine producing cells, IFN-r and IL-4 producing cell were recovered with HF. In all the three HFs tested, cytokine producing cells were recovered with HF and even by the formulae of sCRT (Figure 4).

Figure 5 shows that MMC clearly suppressed the phagocytic activity of mice both in number and function. After the treatment of HF, the mice recovered their phagocytic activity to normal range. With a precise observation, the recovery activity was different between SQT and BYT. SQT was the strongest HF among the four formulae to augment in number and function of phagocytes. On the other hand, the augmentation by BYT was less than that by SQT (Figures 5-9).

The bone marrow-suppressed mice were administered herbal decoction HF for five days. One week later, mice were immunized with sheep red blood cells, (2 × 10⁸/mouse) intraperitoneally. Four and six days later, their plague-forming cells (PFC) were developed. The ability of IgM and IgG antibody production was tested by the method reported by Jerne and Nordin [19] [20] . In this mouse model, MMC did not reduce the antibody forming cells significantly but the tendency was the same as shown in the former section. In this test, B was the most effective than that of A. BYT was the strongest material to augment antibody secreting cell among the four formulae (Figure 10).

So as to detect the supportive effect and important immunological stimulation by tonic agent, Bu-Ji, we traced the augmentation pattern of each remedy. As results of this trial, the phagocytic patterns by tonic agents, Bu-Ji, were clearly different from MMC-treated mice. Moreover, augmentation of phagocytes were different between each HF. SDT was prominent in activating phagocytes quantitatively and qualitatively compared to BYT.

We showed the diversity in the recovery pattern of HFs. Famous tonic remedies in China and Japan, SDT, strongly recovered phagocytic activity in compromised hosts, but the recovery by BYT was much less than that by SDT. The recovery level of phagocytic activity by sCRT was between SDT and BYT (Figures 5-9, Table 4).

The amount of generated super oxide anion was calculated in the formula shown above. The generated super oxide anion after one week administration of Agaricus and Chlorella were 2.64 and 1.95 × 10⁻⁵ mmol/ml, respectively, whereas that was 2.85 × 10⁻⁵ mmol/ml in control group. The generated super oxide anion after one week administration of herbal medicine, SDT, BYT and sCRT were 1.24, 1.25 and 2.88 × 10⁻⁵ mmol/ml, respectively. The generated super oxide anion after one week administration of Propolis was 2.55 × 10⁻⁵ mmol/ml. All these drugs, except for sCRT, decreased super oxide anion generation after administration for one week in

mice (Figure 11).

Since the antioxidative effects of herbal medicine were demonstrated, we investigated the way to reinforce this effect. The fermentation is one of the possibilities. Since the fermentation is preceded by bacterial digestion and degradation, less efficient constituents would be lost than commonly used extraction by hot water. Therefore, we decided to ferment the herbal medicine by yeast (Saccharomyces cerevisiae), expecting the enhancement of its antioxidative effects. The generated super oxide anion after one week administration of fermented herbal medicine TCM, SDT, BYT and sCRT were 0.62, 0.84 and 1.50 × 10⁻⁵ mmol/ml, respectively. All the fermented herbal medicine decreased super oxide anion generation in compare with their corresponding unfermented ones (Figure 12, Table 5).

The antioxidative activity of Propoils has been demonstrated, however, the particle of native Propolis was seen to be gross. In order to reinforce its antioxidative activity from physical constructive view point, we tried to micrified Propolis into 0.5 μm, expecting enlarged attachment area with reaction mixture. The generated super oxide anion after one week administration of micrified Propolis was 2.52 ×10⁻⁵ mmol/ml, whereas that of non- micrified Propolis was 2.55 × 10⁻⁵mmol/ml. The antioxidative activity was slightly enhanced by micrifying the drug (Figure 11, Table 6).

^*p < 0.01 comparing to MMC control. ^**p < 0.05 comparing to non-fermented formulae.

Sheep erythrocyte (SRBC), a T-dependent antigen, was used for antibody formation cell study. Ten days after tumor transplantation, each antigen was intra-peritoneally injected. After four and six days, the antibody-form- ing cells were detected using localized hemolysis in an agar gel. Plaque-forming cells were developed by the method of Jerne and Nordin [19] [20] .

The fermentation is preceded by bacterial digestion and degradation, less of the efficient constituents would be lost than commonly used extraction by hot water. Therefore we decided to ferment the herbal medicine by yeast (Saccharomyces cerevisiae), expecting the enhancement of lymphocyte activating effects through antibody forming cells. The antibody forming cells after one week’s administration of fermented SQT and BYT were 135% and 140%, respectively. All the fermented herbal medicines from HF increased PFC (Table 6).