In the present study, anti-BI MAb was applied to the immunostaining technique for the visual detection of BI on the membrane. First, the specificity of anti-BI MAb was tested in a dot blot assay using BI and other glycosides such as flavonoids and terpenoids. From the staining data as shown in Figure 2, anti-BI MAb specifically recognized BI when 10 μg was present on the membrane. This result agrees with cross-reactivity data of anti-BI MAb against various glycosides in competitive ELISA [15]. In the methodology of immunostaining of BI, BI molecule was separated into two functional parts, the epitope part (mainly aglycone) and the sugar part. The sugar part of BI is oxidized to give aldehyde groups which react with amino groups of lysine or arginine or both on the protein that can bind strongly to the adsorbent membrane. The aglycone part of BI was bound by anti-BI MAb as shown diagrammatically in Figure 3. From this, it becomes evident that anti-BI MAb interacts with the A ring including the hydroxyl groups at C-5 and C-6 positions of BI. Figure 4 showed the sensitivity of anti-BI MAb and detection of BI under dot blot assay conditions. As little as 0.5 μg of BI was still detected by dot blot assay. These characteristics of anti-BI MAb might be an important advantage for the specific analysis of BI which is a major pharmacologically active component of S. radix using on-membrane assisted immunostaining in a plant extract. Furthermore, the results of dot blot assay suggest that the PES membrane is suitable for immunodetection of BI.

Direct detection of small molecular compounds on a TLC plate by immunostaining is impossible, because they are easily washed out by buffer solution without fixing. We designed the system that BI was blotted onto a PES membrane from the developed TLC plate for immunostaining. After BI was transferred to the PES membrane, it was treated with NaIO₄ solution, followed by the addition of BSA. This reaction enhanced the fixation of BI via BI-BSA conjugate preparation on the PES membrane. When the blotted PES membrane was incubated in the absence of BSA, it was essentially free of immunostaining (data not shown). A comparison of the detection limit of BI was made between the eastern blotting technique and the TLC plate stained with FeCl₃ (Figure 5). It became apparent that the eastern blotting technique was almost same sensitivity compared to FeCl₃ staining. The eastern blotting technique could detect 1 μg of BI on the PES membrane. In regard to blotting, two points should be taken into consideration. First, BI on the TLC plate should be easily dissolvable in the blotting solution. Second, diffusion should be avoided so that the spots of BI on the TLC plate are not destroyed during the transferring manipulation. It is difficult to estimate the blotting efficiency, because the percentage of BI that can be blotted onto the PES membrane is unknown.

Various S. radix samples and KMs were qualitatively analyzed by eastern blotting technique (Figure 6), and the amount of BI was quantitatively determined by competitive ELISA using anti-BI MAb (Table 1). Lane 6 is shakuyakukanzoto extract which does not contain S. radix and no band was indicated by immunostaining. On the other hand, the stained band in shosaikoto and saikokaryukotsuboreito extracts (Figure 6, lanes 7 and 8), which contain S. radix, and various extracts of S. radix samples clearly appeared (Figure 6, lanes 2–5). Furthermore, the concentration of BI was quantitatively analyzed by competitive ELISA using anti-BI MAb (Table 1). Collectively, this established eastern blotting technique is more useful than the TLC analysis for the specific detection of BI in samples containing various crude drugs such as KMs with various colored components (e.g., flavonoids, anthraquinones and alkaloids) and impurities.

As another application of eastern blotting technique, the immunostaining of BI was investigated to identify the distribution of BI in fresh root of S. baicalensis. A piece of PES membrane was placed onto a glass microfiber filter sheet. A sliced fresh root of S. baicalensis was treated with methanol to inactivate endogenous peroxidase. Then the sliced root was placed onto the PES membrane, and pressed evenly for 10 min. The blotted PES membrane was immunostained using a procedure similar to that described in dot blot assay. Figure 7 illustrates the immunohistochemical localization of BI in fresh S. baicalensis root. The cortex (part B) contained a higher concentration of BI than the xylem (part A). In order to confirm this result, these tissues were analyzed individually by competitive ELISA. The concentrations of BI, as determined by ELISA, were 33.7 ± 3.3 and 72.2 ± 6.8 μg/mg dry weight for xylem and cortex, respectively.

BI, BE and other flavonoids were purchased from Wako Pure Chemicals (Osaka, Japan). Bovine serum albumin (BSA) and peroxidase-labeled anti-mouse IgG (whole molecule) were purchased from Sigma (Steinheim, Germany). Mustang^TM E positive-charged PES membrane was purchased from Pall Corporation (East Hills, NY, USA). Glass microfiber filter sheets (GF/A) were purchased from Whatman International Ltd. (Maidstone, England). All other chemicals were standard commercial products of analytical-reagent grade.

Samples of various S. radixes were purchased from Tochimototenkaido Corporation (Osaka, Japan) and collected from the botanical garden of Nagasaki International University. The commercial KMs were procured from Tsumura & Co. (Tokyo, Japan).

Dried powders of various S. radixs (100 mg) and KMs (100 mg) were extracted five times with 1.2 mL methanol containing 0.1% NH₄OH with sonication and combined into the tube. Then, the solvent of the extract was evaporated using N₂ gas evaporation system and freeze-dryer. After freeze-drying, the residue was dissolved in 1.0 mL methanol containing 0.1% NH₄OH, filtered using a Cosmonice filter W (0.45 μm filter unit, Nacalai Tesque, Kyoto, Japan) and used for the experiments of eastern blotting and ELISA.

Immunization, hybridization, purification and characterization of MAb against BI and its development of ELISA system for determination of BI have been established and described previously [15].

BI and various glycosides were dissolved in methanol (1 mg/mL) and aliquots (10 μg) were dotted onto a positive-charged PES membrane. All following steps were done at room temperature. After air drying, the PES membrane was dipped into water containing NaIO₄ (10 mg/mL) and kept for 30 min. After washing with water, 50 mM carbonate buffer solution (pH 9.6) containing BSA (1%) was added, and shaken for 3 h. After washing the PES membrane with phosphate buffer saline (PBS), the membrane was treated with PBS containing 5% skim-milk for 2 h to reduce nonspecific adsorption. The PES membrane was immersed in anti-BI MAb and shaken for 3 h. After washing the PES membrane twice with PBS containing 0.05% Tween 20 (T-PBS), a peroxidase-labeled goat anti-mouse IgG (whole molecule) in T-PBS was added and shaken for 1 h. The PES membrane was washed twice with T-PBS and once with PBS and exposed to freshly prepared 1 mg/mL 4-chloro-1-naphthol-0.03% H₂O₂ in PBS for 15 min. The reaction was stopped by washing with PBS, and the immunostained PES membrane was allowed to dry.

BI (10 μg) and appropriate other sample extracts were applied to a TLC plate (Silica gel 60, Merck Ltd., Darmstadt, Germany) and developed with 1-butanol-water-acetic acid (4:2:1, by volume). After drying the developed TLC plate, it was sprayed by FeCl₃ solution for the detection of BI.

BI (10 μg) and appropriate other sample extracts were applied to a TLC plate and developed with 1-buranol-water-acetic acid (4:2:1, by volume). The developed TLC plate was dried and then sprayed with a blotting solution mixture of isopropanol-methanol-water (1:4:20, by volume). It was placed on a stainless steel plate and then covered with a PES membranes sheet. After covering with a glass microfiber filter sheet, the whole assembly was pressed evenly for 70 s with a 120 °C hot plate. The PES membrane was separated from the TLC plate and dried. The blotted PES membrane was treated for the immunostaining of BI as described in dot blot assay.

Fresh root of S. baicalensis was collected from the botanical garden of Nagasaki International University at the summer season. To identify the distribution of BI in the root, a sliced fresh root of S. baicalensis (thickness 2–3 mm) was placed on the PES membrane, and they were pressed together evenly 10 min for transfer all of the contents from the section to the PES membrane. Then, the blotted PES membrane was stained using the same procedure described in the dot blot assay.