Chinese Journal of Natural Medicines  2018, Vol. 16Issue (3): 175-183  DOI: 10.1016/S1875-5364(18)30045-1
0

Cite this article as: 

ZHAO Yu-Nan, SHAO Xiao, OUYANG Liu-Feng, CHEN Lin, GU Ling. Qualitative detection of ginsenosides in brain tissues after oral administration of high-purity ginseng total saponins by using polyclonal antibody against ginsenosides[J]. Chinese Journal of Natural Medicines, 2018, 16(3): 175-183.
[Copy]

Research funding

The work was supported by the National Natural Science Foundation of China (Nos. 81303246, 81703732) and Projects of Priority-Discipline for Colleges and Universities of Jiangsu Province (integration of traditional Chinese and western medicine)

Corresponding author

ZHAO Yu-Nan, Tel/Fax: 86-24-85811922, E-mail: zhaoyunan-js@163.com

Article history

Received on: 29-Sep.-2017
Available online: 20 Mar., 2018
Qualitative detection of ginsenosides in brain tissues after oral administration of high-purity ginseng total saponins by using polyclonal antibody against ginsenosides
ZHAO Yu-Nan1 , SHAO Xiao1 , OUYANG Liu-Feng2 , CHEN Lin1 , GU Ling3     
1 Laboratory of Pathological Sciences, College of Medicine and Life, Nanjing University of Chinese Medicine, Nanjing 210046, China;
2 Laboratory of Pathological Sciences, College of Medicine, Yan'an University, Yan'an 716000, China;
3 Research Center, Basic Medical College, Nanjing University of Chinese Medicine, Nanjing 210046, China
[Abstract]: Given the limited studies and conflicting findings, the transport character of ginsenosides crossing the blood–brain barrier (BBB) remains unclear. The present study was designed to qualitatively determine the distribution of ginsenosides in brain tissues after oral administration of ginseng total saponins, using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) combined with immunohistochemistry. In brain tissue homogenates, ginsenoside Rg1 was detectable and no other ginsenosides or their metabolites were found. No ginsenosides were detected in cerebrospinal fluid. Immunohistochemistry staining of brain tissue sections by using anti-ginsenoside polyclonal antibodies revealed the localization of ginsenosides in brain tissues. Furthermore, immunofluorescence double staining revealed that ginsenosides widely existed in vascular endotheliocytes and astrocytes, and in few neurons. These results indicated that Rg1 was the main component that entered the brain after oral administration of ginseng total saponins and that ginsenosides could cross the BBB, although the transport capability of ginsenosides through the BBB may be poor.
[Key words]: Ginseng     Ginsenoside     Immunostaining     Mass spectrometry     Blood-brain barrier    
Introduction

In Asia, ginseng is the most valuable traditional medicinal herb that has been used for over 2000 years. Ginsenosides, the triterpenoid saponins extracted from ginseng, are the major bioactive components of ginseng [1]. Recent studies have demonstrated that ginsenosides exert protective effects against neurodegenerative disorders through attenuation of excitotoxicity and neuroinflammation, regulation of ion channels and receptors, maintenance of neurotransmitter balance, anti-apoptotic effects, and mitochondrial stabilization [2-5].

Studies have investigated the in vivo distribution of ginsenosides in brain tissues by using high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) to determine whether the pharmacological effects of ginsenosides in the central nervous system (CNS) are caused by components that target the brain. Ten such studies are found in PubMed. For example, by using a fast HPLC-MS/MS analytical procedure, Liu et al. have determined the distribution behavior of ginsenosides Rg1, Re, Rb1, Rc, and Rd in brain tissues after intravenous administration of ginseng extracts [6]. Shi et al. and Xue et al. have found that subcutaneously administered Rg1 or Re is rapidly distributed in the cerebrospinal fluid (CSF) and dialysates in the hippocampus and medial prefrontal cortex [7-8]. These findings indicate that ginsenosides can cross the blood-brain barrier (BBB) and directly target the neurons, possibly causing pharmacological effects in the CNS.

However, after carefully reading these literatures, we found several limitations inherent in their experimental designs. First, the authors did not perfuse the brain with saline solution before preparing brain tissue homogenate [6, 9-12]. Failure to perform such procedure possibly produced false positive results because of blood contamination in the tissue homogenate. Second, markers for BBB integrity were not used to monitor changes at the interfaces between blood, brain, and CSF while implanting dialysis probes into brain tissues or while inserting the needle of microsyringe into the cerebral ventricle [7-8, 13]. Positive detection of ginsenosides in the CSF or dialysates is likely a consequence of BBB disruption. In Wang et al.'s study, the ginsenoside Rg1 showed a very poor BBB permeability and failed to be detected in brain tissues [14]. Third, ginseng extracts or ginsenosides used in these studies were administered intravenously or subcutaneously instead of orally, and such approach does not represent the administration route of ginseng in the clinic. These limitations and conflicting findings render the findings on the transport and disposition of ginsenosides in CNS uncertain and thus must be clarified by using advanced analytical and detection approaches.

In addition to HPLC-MS/MS, an immunological approach using monoclonal or polyclonal antibodies can be used for quantitative or qualitative analysis of bioactive small-molecule compounds [15-16]. Antibodies against small-mole-cule compounds have been successfully applied to quantify small-molecule compounds in various extracts or body fluids by using Eastern blotting assay [17] or enzyme-linked immunosorbent assay [18], and bioactive constituents can be purified through a one-step immunochromatographic separation [19]. Additionally, antibodies can be used in immunohistochemistry staining, allowing visualization of small molecule compounds in a tissue section. Furthermore, immunofluorescence double staining, a specific example of immunohistochemistry, allows visualization of small-molecule compounds in certain cells by using specific biomarkers within cells. The present study aimed to elucidate the disposition of ginsenosides in CNS through immunohistochemistry by using anti-ginsenoside polyclonal antibodies.

This work included three experiments involving male rats and mice. In Experiment 1, after oral administration of ginseng total saponins, ginsenosides were qualitatively detected in brain tissue homogenates of mice using HPLC-MS/MS. In Experiment 2, we detected the presence of ginsenosides in the CSF of rats by using HPLC-MS/MS and determined whether ginsenosides can cross the BBB. In Experiment 3, we qualitatively detected ginsenosides in brain tissue sections of mice through immunohistochemistry and determined the disposition of ginsenosides in brain tissues through immunofluorescence double staining to confirm the results from Experiments 1 and 2.

Materials and Methods Chemicals

The ginsenosides Rg1, Re, Rh1, Rb1, F1, Rd, Rh2, protopanaxatriol, protopanaxadiol, and compound K standards (purity > 98%) were purchased from the National Institute for Food and Drug Control (Beijing, China). Deionized water was prepared using a Milli-Q water purification system (Millipore, Billerica, MA, USA). Acetonitrile, methanol, and HPLC-grade formic acid were supplied by Fisher Chemicals (Loughborough, UK). All other solvents used in the present study were of analytical grade and obtained from Nanjing Chemical Reagents Co. (Nanjing, China).

Animals

Male Sprague-Dawley rats (weighing 180-210 g) and C57BL/6N mice (weighing 18-22 g) were obtained from the Laboratory Animal Center of Nanjing Medical University. The animals were housed in groups of five in home cages with sawdust on the floor, and food and water were freely available. They were maintained under a standard dark-light cycle (lights on between 7: 00 and 19: 00) at room temperature (22 ± 2 ℃) and adapted to daily handling during the week after delivery. All animal treatments were approved by the IACUC (Institutional Animal Care and Use Committee) of Nanjing University of Traditional Chinese Medicine and carried out in accordance with the Guidelines of Accommodation and Care for Animals, which was formulated by the Chinese Convention for the protection of vertebrate animals used for experimental and other scientific purposes.

Preparation of high-purity ginseng total saponins (HPGTS)

HPGTS were prepared as previously reported [20]. In brief, total saponins were efficiently purified from a water decoction of Chinese ginseng root through dynamic anion-cation exchange following the removal of hydrophilic impurities by macroporous resin. Rg1, Re, Rf, F3, Rh1, Rg2, Rb1, Rc, Rb2, Rb3, Rd, F2, and Rg3 were detected in HPGTS. The total saponin content of HPGTS reached up to approximately 106.8% ± 4.5% and 74.97% ± 2.63% as determined using a colorimetric method and the determination of total analogs by single marker approach based on charged aerosol detection with post-column inverse gradient, respectively [21].

Preparation and characterization of polyclonal antibody against ginsenosides

First, ginsenoside-bovine serum albumin (BSA) conjugate was synthesized as previously described [22]. In brief, 80% ethanol solution (0.7 mL) of HPGTS (10 mg) was added dropwise into a solution (0.5 mL) containing NaIO4 (4 mg); the solution was stirred at room temperature for 1 h. Carbonate buffer (100 mmol·L-1, pH 9.0, 2 mL) containing BSA (10 mg) was added to the mixture, which was then stirred at room temperature for 5 h. The reaction mixture was dialyzed against water at 4 ℃ for 2 days and then lyophilized to obtain ginsenoside-BSA conjugate. Second, the levels of antibodies were raised by subcutaneously injecting ginsenoside-BSA conjugate emulsified with ImjectTM Freund's complete adjuvant (Thermo Scientific Pierce) into a New Zealand rabbit. The antiserum was collected after three injections given at a 10-day interval. Third, the immunoglobulin fractions IgG was purified through precipitation in 50% saturated (NH4)2SO4 and by using immobilized Pierce® Protein A/G Agarose column according to the manufacturer's instruction. Finally, the antibody was characterized through biolayer interferometry-based assay and immunocytochemistry.

Procedures Experiment 1: Qualitative detection of ginsenosides in brain tissues after oral administration of HPGTS

The mice were randomly assigned into two groups (n = 5/group): the control and HPGTS groups. The HPGTS group was administered daily with HPGTS (100 mg·kg-1, dissolved in physiological saline) via gastric gavages, whereas the control group was given the vehicle. After administration for 1 week, the mice were deeply anesthetized with an overdose of sodium pentobarbital and perfused with saline solution through their left ventricle. Their skulls were subsequently opened and their brain tissues were removed.

The brain tissues were homogenized in saline solution and centrifuged at 8 000 r·min-1 for 10 min. The supernatants were pretreated through solid-phase extraction (SPE). The supernatants (1 mL) were loaded onto a C18 SPE cartridge (1.5 mL) preconditioned with 3 mL of acetonitrile and 3 mL of water. After loading the supernatants, the SPE cartridge was washed with 6 mL of water to elute hydrophilic impurities and 6 mL acetonitrile to elute the analytes. The eluate was evaporated until dryness under a stream of nitrogen. The residue was dissolved in methanol (100 μL) and centrifuged at 12 000 r·min-1 for 10 min, prior to ultra-HPLC-MS/MS analysis.

The SPE recoveries of ginsenosides from brain tissues were measured by comparing the peak areas of brain tissue extracts with those of the same amounts of HPGTS solutions in the mobile phase. The recoveries of ginsenosides at three concentrations (low, medium, and high) were acceptable and reached up to approximately 70%.

Experiment 2: Qualitative detection of ginsenosides in the CSF after oral administration of HPGTS

The rats were randomly assigned into two groups (n = 10/group): the control and HPGTS groups. The HPGTS group was administered daily with HPGTS (80 mg·kg-1) via gastric gavages, whereas the control group was given the vehicle. After administration for 1 week, the rats were anesthetized with sodium pentobarbital and positioned in a stereotaxic apparatus (RLJD68506, RWD Biotechnology Co. Ltd., Shenzhen, China). The needle of a microsyringe was inserted into the cisterna magna through the foramen magnum in the occipital crest. CSF (100 μL) was then collected by gently moving the syringe plunger. The merged CSF (1 mL, 10 rats) was dried under a stream of nitrogen; the residue was dissolved in methanol (100 μL) and centrifuged at 12 000 r·min-1 for 10 min prior to ultra-HPLC-MS/MS analysis.

Experiment 3: Qualitative detection of ginsenoside in brain tissue sections in mice after oral administration of HPGTS

The mice were randomly assigned into two groups (n = 5/group): the control and HPGTS groups. The HPGTS group was administered daily with HPGTS (100 mg·kg-1) via gastric gavages, whereas the control group was given the vehicle. After administration for 1 week, the mice were euthanized by an overdose of anesthesia and perfused with 0.9% ice cold saline followed by 4% paraformaldehyde in phosphate buffer. The brain tissues were subsequently removed and serially sectioned (20 μm) on a cryostat (CM1950, Leica, Germany). The free-floating sections in saline solution in the wells were collected for the subsequent immunohistochemistry and immunofluorescence double staining.

Ultra-HPLC-MS/MS analysis

A PerkinElmer Flexar SQ 300 MS system (Santa Clara, CA, USA) equipped with an electrospray ionization (ESI) source was used to separate and qualitatively detect ginsenosides in brain tissue or CSF samples. Chromatographic separation was performed on a Dionex Acclaim RSLC120 C18 reversed-phase column (150 mm × 2.1 mm, 2.2 μm) at 30 ℃. The mobile phase consisted of 0.2% formic acid (V/V) (solvent A) and acetonitrile (solvent B) at a flow rate of 0.4 mL·min-1. The injection volume was 10 μL. The linear gradient was as follows: 0-25 min, 18% B; 25-30 min, 18%-23% B; 30-65 min, 23%-38% B; 65-75 min, 38%-58% B; 75-83 min, 58%-78% B; and 83-100 min, 78% B.

ESI in MS was performed using the following parameters: spray voltage, 4.0 kV; sheath gas pressure, 35 psi; auxiliary gas pressure, 15 psi; capillary temperature, 350 ℃; and collision pressure, 1.5 psi. Full scan of ions ranging from m/z 100 to m/z 1500 was conducted in the negative-ion mode.

Biolayer interferometry

The binding affinity of anti-ginsenoside polyclonal antibody and each ginsenoside was determined using biolayer interferometry [23] performed on an Octet Red96 system (ForteBio, Menlo Park, CA, USA), which is a label-free, real-time detection system [24]. In brief, assays were performed in 96-well black microplates at 25 ℃, and all incubation volumes were 200 μL. Biotinylated polyclonal antibody was immobilized onto Super Streptavidin (SSA) Biosensor tips (ForteBio) in phosphate-buffered saline (PBS). Reference SSA Biosensor tips were blocked with biocytin in PBS buffer. The coated tips were subsequently incubated with each ginsenoside (1-100 μmol·L-1 in PBS buffer containing 0.05% Tween-20 and 0.5% DMSO) to generate an association curve. The dissociation curve was generated by incubating these tips with the above PBS buffer containing Tween-20 and DMSO. The ForteBio Data Analysis Octet software was applied to process the curve by subtracting a sample curve from the reference curve. The association and dissociation curves were fit using a single-exponential fitting model, and the affinity constant KD was calculated by curve fitting.

Immunocytochemistry

Human hepatoma SMMC-7721 cells [25] were cultured in RPMI-1640 medium containing 10% fetal bovine serum. After the cells stabilized in a six-well plate, each well was treated with 2 mL of HPGTS solution (1 μg·min-1) and then the plate was incubated at 37 ℃ in 5% CO2 incubator for 2 h. The cells were subsequently washed three times with cold PBS and fixed with 4% paraformaldehyde for 30 min. After being washed twice with PBST (PBS containing 0.05% Triton X-100), the cells were preincubated in 5% goat serum in PBS for 30 min and then incubated with a polyclonal antibody against ginsenosides (5 μg·mL-1) at 4 ℃ overnight. After being washed thrice with PBST, the cells were incubated in a polyperoxidase-anti-rabbit IgG (PV9000, ZSGB-BIO) for 1 h, washed again, developed for 5 min in diaminobenzidine (DAB), and then thoroughly washed. The stained cells were observed under an inverted microscope (Olympus Ⅸ 51, Japan).

Immunohistochemistry

Selected sections were washed thrice with PBST and blocked with 5% goat serum for 1 h. The sections were incubated with the polyclonal antibody against ginsenosides (10 μg·mL-1) overnight at 4 ℃. The sections were subsequently rinsed several times with PBS, incubated in a polyperoxidase-anti-rabbit IgG for 1 h, rinsed again, developed for 5 min in DAB, and then thoroughly rinsed. The sections were mounted onto gelatin-coated slides and dried, dehydrated through stepped alcohol washes, cleared in xylene, and coverslipped. The stained sections were observed under an upright microscope (Olympus BX 51, Japan).

Immunofluorescence staining

The primary antibodies included rabbit polyclonal anti-ginsenoside (10 μg·min-1), mouse monoclonal anti-glial fibrillary acidic protein (anti-GFAP, 1 : 250, Abcam), mouse monoclonal anti-Neuronal Neclei (anti-NeuN, 1 : 200, Millipore), and mouse monoclonal anti-von Willebrand factor (anti-vWF, 1 : 250, Santa Cruz). The secondary antibodies included Alexa Fluor 594-labelled goat anti-rabbit IgG (1 : 200, Invitrogen) and Alexa Fluor 488-labelled goat anti-mouse IgG (1 : 200, Invitrogen).

The selected sections were washed thrice with PBST, blocked with 5% goat serum for 1 h, and incubated with primary antibody at 4 ℃ overnight. After being washed thrice with PBST, the sections were incubated with appropriate secondary antibody for 1 h in the dark, followed by washing three times with PBST. After being stained with Hoechst 33342 (Sigma), the sections were treated with 1 mol·L-1 CuSO4 in ammonium acetate buffer for 20 min and then placed back into PBS. The sections were finally mounted using FluoromountTM aqueous mounting medium (Sigma) and viewed under a Zeiss Axio Imager M2 fluorescent microscope.

Results Rg1 is detectable in brain tissue samples but not in CSF samples after oral administration of HPGTS

Figs. 1A and 1B show the total ion chromatograms (TIC) of ginseng components in brain tissue homogenates of the control and HPGTS groups. Compared with that in the control, the peak with a retention time of 28.76 min was observed in the HPGTS group; this peak was extracted to produce a main ion of m/z 845 (Fig. 1C), which can be ascribed to [800 + COOH]. In known ginseng saponins, the m/z value of ginseno-sides Rg1 and Rf is 800. As the retention time is in accordance with that of Rg1 standard, the peak was identified-as Rg1.

Figure 1 Representative TIC of the ginseng components in brain tissue homogenates obtained from the control (A) and HPGTS (B) groups. The ion chromatogram of the peak indicated by "↓" was extracted, and the main ion at m/z 845 was obtained (C)

Selective ion monitoring (SIM) was used for the other peaks in the TIC, and deprotonated molecules [M + COOH] were detected at m/z 991 for Re or Rd, m/z 815 for F3, m/z 683 for Rh1 or F1, m/z 829 for Rg2, F2, or Rg3, m/z 1153 for Rb1, m/z 1123 for Rc, Rb2, or Rb3, m/z 647 for Compound K or Rh2, m/z 521 for protopanaxatriol, and m/z 505 for protopanaxadiol. No other ginsenosides or their metabolites [26-27] were found in the TIC of the HPGTS group. Structures and m/z of these ginsenosides were based on our previous reports [21].

Figs. 2A and 2B show the TIC of the ginseng components in the CSF samples of the control and HPGTS groups. As mentioned above, the SIM was used to detect ginsenosides or their metabolites. With the lower detection limit of ginsenosides at approximately 2.5 ng·mL-1 using HPLC/MS assay, no ginsenosides were detected in the CSF samples. As the CSF samples were concentrated by up to 10 times prior to ultra-HPLC-MS analysis, the concentration of each ginsenoside in the CSF samples was zero or less than 0.25 ng·mL-1.

Figure 2 Representative TIC of the ginseng components in CSF samples obtained from the control (A) and HPGTS (B) groups
Positive staining with polyclonal antibody against ginsenosides is observed in brain tissue sections after oral administration of HPGTS Characterization of anti-ginsenoside polyclonal antibody

Biolayer interferometry is a label-free method that enables real-time kinetic characterization of biomolecular interactions and commonly used to detect the binding affinity in protein-protein or small molecule-protein interactions. This technique showed that the affinity constants KD for the interaction of anti-ginsenoside polyclonal antibody with HPGTS, Rg1, Re, Rh1, Rh2, F1, Compound K, protopanaxatriol, and protopanaxadiol were 3.99 × 10-7 mol·L-1, 2.59 × 10-9 mol·L-1, 4.69 × 10-9 mol·L-1, 2.40 × 10-6 mol·L-1, 1.62 × 10-4 mol·L-1, 1.66 × 10-10 mol·L-1, 1.08 × 10-7 mol·L-1, 1.32 × 10-6 mol·L-1, and 2.64 × 10-7 mol·L-1, respectively. Polyclonal antibody against ginsenosides showed a high affinity not only with ginsenosides in HPGTS (i.e., Rg1, Re, and Rh1) but also with their metabolites (i.e., F1, Rh2, Compound K, and protopanaxatriol) in vivo. Additionally, this polyclonal antibody partially cross-reacted with BSA. The cross-reactivity may make it difficult to precisely quantify ginsenosides in plasma, but would not interfere with the qualitative detection of ginsenosides in brain tissues in situ, because the brain tissues were perfused with saline in advance.

The reactive specificity of anti-ginsenoside polyclonal antibody with ginsenosides was verified using immunocytochemistry. First, no specific immunoreactivity was observed in cells not treated with HPGTS (Fig. 3A). Second, after incubation with HPGTS for 2 h, SMMC-7721 cells were fixed, washed, and reacted with the antibody. The localized ginsenosides within the cells appeared as brown structures (Fig. 3B), suggesting that ginsenosides crossed the cell membrane and thus were detected clearly in the cytoplasm. Third, HPGTS-treated SMMC-7721 cells were fixed, washed, and incubated with the antibody spiked with HPGTS to partially block the specific immunoreactivity of the HPGTS-treated cells (Fig. 3C). The result demonstrated that polyclonal antibody against ginsenosides can be applied to specifically detect ginsenosides in cells.

Figure 3 Positive staining of ginsenosides in HPGTS in SMMC-7721 cells by using immunocytochemistry. After HPGTS was treated to SMMC-7721 cells for 2 h, the cells were washed, fixed and incubated with polyclonal antibody against ginsenosides. Compared with non-HPGTS treated cells (A), the brown color indicated the localization of ginsenosides in the cytoplasm of HPGTS treated cells (B). Additionally, co-incubation of polyclonal antibody with HPGTS inhibited positive staining of ginsenosides in HPGTS treated cells (C)
Ginsenosides are detectable in vascular endotheliocytes, astrocytes, and neurons in the CNS

Figs. 4A and 4B show the immunohistochemistry staining of brain tissue sections obtained from the control and HPGTS groups. Compared with the control, the cells in brain tissues showed brown coloration, indicating the localization of ginsenosides in brain tissues of mice after oral administration of HPGTS. To determine which cells in brain tissues can ginsenosides enter into, we chose vWF, GFAP, and NeuN as immunolabelling markers of vascular endotheliocytes, astrocytes, and neurons; additionally, we performed double staining. Positive staining was widely observed in vascular endotheliocytes (Fig. 5) and astrocytes (Fig. 6). Moreover, positive staining was found in a few neurons, indicating that ginsenosides can pass through the BBB (Fig. 7).

Figure 4 Immunohistochemistry staining of brain tissue sections by using polyclonal antibody against ginsenosides. Compared with that in the control (A), an obvious specific immunoreactivity in brain tissue sections was observed in the HPGTS group (B). The HPGTS group was administered daily with HPGTS for 1 week via gastric gavages, whereas the control group was given with vehicle
Figure 5 Ginsenosides were detected in vascular endotheliocytes through immunofluorescence double staining. The brain tissue section was treated with Hoechst 33342 (A) and immunostained with a monoclonal antibody against vWF (B, green) and a polyclonal antibody against ginsenosides (C, red). D is the merged image of A, B, and C. The yellow arrow indicates positive staining in vascular endotheliocytes. Scale bar is 100 μm
Figure 6 Ginsenosides were detected in astrocytes through immunofluorescence double staining. The brain tissue section was treated with Hoechst 33342 (A) and immunostained with a monoclonal antibody against GFAP (B, green) and a polyclonal antibody against ginsenosides (C, red). D is the merged image of A, B, and C. The yellow arrow indicates positive staining in astrocytes. Scale bar is 100 μm
Figure 7 Ginsenosides can pass through the BBB and enter few neurons. The brain tissue section was treated with Hoechst 33342 (A) and immunostained with a monoclonal antibody against NeuN (B, green) and a polyclonal antibody against ginsenosides (C, red). D is the merged image of A, B, and C. The yellow and red arrows indicate positive and negative staining in neurons, respectively. Scale bar is 100 μm
Discussion

An increasing number of studies have shown that, after subcutaneous or intravenous administration of ginseng extracts or ginsenosides, ginsenosides can be detected in brain tissue homogenates through HPLC-MS/MS [6, 9-12]. However, the authors did not clearly describe in their methods the brain perfusion before preparing tissue homogenates; false positive results were possibly obtained because of blood contamination in tissue homogenates. In the present study, the mice were deeply anesthetized and perfused thoroughly with saline solution through the left ventricle to remove blood in brain tissues. After adopting these treatments, we still found ginsenoside Rg1 in brain tissues through ultra-HPLC-MS/MS analysis, indicating that ginsenosides administrated either via intravenous injection or gastric gavages target the brain.

Molecule and ion transportation between the blood and the brain is tightly regulated to maintain normal brain function. Such tight regulation is maintained by the BBB, which constitutes microvascular endothelial cells with tight junctions, pericytes, and astrocytic end-feet [28-29]. Tight junctions between adjacent cells block diffusion of polar molecules through the intercellular cleft. However, endogenous polar compounds and drugs may still cross the BBB through the influx transporter systems of BBB. For example, solute carrier transporters present in the membrane of the capillary endothelium and astrocytes mediate the uptake of various endogenous compounds and drugs [30]. Moreover, an efflux transporter system exists in BBB. For example, ATP-binding cassette transporters localized at the luminal side of brain capillaries are ATP-driven drug efflux pumps [31]. Tight junctions and efflux transporter systems collectively impede neuron uptake of a variety of lipophilic molecules, xenobiotics, and toxic metabolites.

Consequently, the detection of Rg1 in brain tissue homogenates does not necessarily mean that Rg1 acts on neurons, given that BBB is present in the CNS. Ginsenoside Rg1 may simply enter into microvascular endothelial cells or astrocytes through the influx transporter systems. Indeed, in immunofluorescence double staining, positive staining of ginsenosides was widely observed in microvascular endothelial cells and astrocytes. These findings indicated that ginsenosides could at least reach the BBB. Astrocytes play very important roles in cerebral functions, and astrocyte dysfunction substantially contributes to neurodegenerative processes [32]. Thus, ginsenosides possibly exert pharmacological effects in the CNS by targeting astrocytes, even if they cannot cross the BBB to act on neurons.

Detection of drugs in CSF or brain tissue dialysates has been used as evidence to demonstrate the ability of drugs to cross the BBB. Wang et al. have reported that the capacity of ginsenoside Rg1 administrated intravenously to cross the BBB is quite limited, as Rg1 is not detected in brain tissue dialysates [14]. Similarly, in the present study, no ginsenosides were detected in the CSF samples through ultra-HPLC-MS/MS analysis after oral administration of ginseng total saponins for 1 week. However, non-detection of ginsenosides in the CSF or dialysates does not mean that ginsenosides cannot cross the BBB. It is likely that the ginsenoside concentration in the CSF or dialysates was lower than the limit of detection. In the present study, the detection limit of ginsenosides in the CSF samples was approximately 0.25 ng·mL-1 in the negative-ion mode. However, positive ion ultra-HPLC-MS/MS generally showed higher sensitivities than negative ion ultra-HPLC [33]. In Ferrer et al.'s study, the sensitivity in positive ion mode was approximately 10 times higher than that in negative ion mode [34]. Consequently, non-detection of ginsenosides in the CSF might be due to low detection sensitivity caused by non-opti-mized ionization polarity and mobile phase modifier.

In three recent studies, ginsenosides were successfully detected in the CSF or brain tissue dialysates through ultra-HPLC-MS/MS analysis after intravenous administration of individual ginsenoside or Sanqi extracts [7-8, 13]. Their results indicate that ginsenosides can cross the BBB and that they display good BBB permeability. However, these studies did not use markers (i.e., Evans blue or visualizable dextrans [35]) to monitor the BBB integrity after implanting dialysis probes or inserting the needle of microsyringe, possibly producing false positive results as a consequence of BBB disruption. Given the above study limitations, adopting other approaches to further clarify the transport character of ginsenosides through the BBB is necessary.

The present study used an immunological approach to qualitatively detect ginsenosides in brain tissues. Ginseng saponins are metabolized extensively by intestinal bacteria after oral administration [26-27]. In the protopanaxadiol group, Rb1 and Rd were metabolized into compound K, whereas Rg3 was biotansformed into Rh2. In the protopanaxatriol group, Rg1 and Re were converted into Rh1, F1 and protopanaxatriol. After oral ingestion, ginsenoside metabolites are absorbed from the gut into systemic circulation [36-37]. On this basis, we designed anti-ginsenoside polyclonal antibodies, which demonstrated a high affinity with not only parent ginsenosides but also their metabolites. Immunohistochemistry staining using such polyclonal antibodies can render the parent ginsenosides or their metabolites visible in brain tissue sections, revealing the distribution of ginsenosides. We found ginsenosides in neurons, indicating that ginsenosides can cross the BBB and enter into the neurons. However, ginsenosides were found only in few neurons, suggesting that ginsenoside transport through the BBB might be poor, or that the distribution of ginsenosides in neurons might show a certain degree of sele-ctivity.

Finally, our study had one limitation that was inherent in the antibody design. Although we found ginsenosides in neurons using anti-ginsenoside polyclonal antibodies, we failed to determine which ginsenoside can cross the BBB because of the high affinity of polyclonal antibodies with each ginsenoside. In addition, we failed to assess the transport capability of ginsenosides through the BBB, because we only qualitatively detected ginsenosides in brain tissues using ultra-HPLC-MS/MS combined with immunohistochemistry.

Conclusion

Ginsenoside Rg1 was the main ginseng component that targets the brain after oral administration of ginseng total saponins. Ginsenosides could not only reach the BBB but also cross this barrier. To the best of our knowledge, the present study was the first to employ immunohistochemistry using anti-ginsenoside polyclonal antibody and detect ginsenosides in brain tissues. The ultra-HPLC-MS/MS analysis combined with immunohistochemistry is an appropriate technique to determine the in vivo distribution of active ingredients in brain tissues.

References
[1]
Shin BK, Kwon SW, Park JH. Chemical diversity of ginseng saponins from Panax ginseng[J]. J Ginseng Res, 2015, 39(4): 287-298. DOI:10.1016/j.jgr.2014.12.005
[2]
Navavi SF, Sureda A, Habtemariam S, et al. Ginsenoside Rd and ischemic stroke:a short review of literatures[J]. J Ginseng Res, 2015, 39(4): 299-303. DOI:10.1016/j.jgr.2015.02.002
[3]
Sheng C, Peng W, Xia ZA, et al. The impact of ginsenosides on cognitive deficits in experimental animal studies of Alzheimer's disease:a systematic review[J]. BMC Complement Altern Med, 2015, 15(1): 386.
[4]
Ong WY, Farooqui T, Koh HL, et al. Protective effects of ginseng on neurological disorders[J]. Front Aging Neurosci, 2015, 7: 129.
[5]
Nah SY. Ginseng ginsenoside pharmacology in the nervous system:involvement in the regulation of ion channels and receptors[J]. Front Physiol, 2014, 5(5): 98.
[6]
Liu M, Wang H, Zhao S, et al. Studies on target tissue distribution of ginsenosides and epimedium flavonoids in rats after intravenous administration of Jiweiling freeze-dried powder[J]. Biomed Chromatogr, 2011, 25(11): 1260-1272. DOI:10.1002/bmc.v25.11
[7]
Xue W, Liu Y, Qi WY, et al. Pharmacokinetics of ginsenoside Rg1 in rat medial prefrontal cortex, hippocampus, and lateral ventricle after subcutaneous administration[J]. J Asian Nat Prod Res, 2016, 18(6): 587-595. DOI:10.1080/10286020.2016.1177026
[8]
Shi J, Xue W, Zhao WJ, et al. Pharmacokinetics and dopamine/acetylcholine releasing effects of ginsenoside Re in hippocampus and mPFC of freely moving rats[J]. Acta Pharmacol Sin, 2013, 34(2): 214-220. DOI:10.1038/aps.2012.147
[9]
Hao K, Gong P, Sun SQ, et al. Mechanism-based pharmacokinetic-pharmacodynamic modeling of the estrogen-like effect of ginsenoside Rb1 on neural 5-HT in ovariectomized mice[J]. Eur J Pharm Sci, 2011, 44(1-2): 117-126. DOI:10.1016/j.ejps.2011.06.014
[10]
Feng L, Wang L, Hu C, et al. Pharmacokinetics, tissue distribution, metabolism, and excretion of ginsenoside Rg1 in rats[J]. Arch Pharm Res, 2010, 33(12): 1975-1984. DOI:10.1007/s12272-010-1213-2
[11]
Li T, Shu YJ, Cheng JY, et al. Pharmacokinetic and efficiency of brain targeting of ginsenosides Rg1 and Rb1 given as Nao-Qing microemulsion[J]. Drug Dev Ind Pharm, 2015, 41(2): 224-231. DOI:10.3109/03639045.2013.858734
[12]
Zhang Y, Lin L, Liu GY, et al. Pharmacokinetics and brain distribution of ginsenosides after administration of sailuotong[J]. China J Chin Mater Med, 2014, 39(2): 316-321.
[13]
Long W, Zhang SC, Wen L, et al. In vivo distribution and phar-macokinetics of multiple active components from Danshen and Sanqi and their combination via inner ear administration[J]. J Ethnopharmacol, 2014, 156: 199-208. DOI:10.1016/j.jep.2014.08.041
[14]
Wang R, Li YN, Wang GJ, et al. Neuroprotective effects and brain transport of ginsenoside R1[J]. Chin J Nat Med, 2009, 7(4): 315-320. DOI:10.3724/SP.J.1009.2008.00315
[15]
Kim H, Yoon S, Chung J. In vitro and in vivo application of anti-cotinine antibody and cotinine-conjugated compounds[J]. BMB Rep, 2014, 47(3): 130-134. DOI:10.5483/BMBRep.2014.47.3.006
[16]
Zhao YN. Advance of studies on application of hapten antibodies[J]. China J Chin Mater Med, 2013, 38(5): 657-660.
[17]
Morinaga O, Tanaka H, Shoyama Y. Detection and quantification of ginsenoside Re in ginseng samples by a chromatographic immunostaining method using monoclonal antibody against ginsenoside Re[J]. J Chromatogr B, 2006, 830(1): 100-104.
[18]
Zhu S, Shimokawa S, Shoyama Y, et al. A novel analytical ELISA-based methodology for pharmacologically active saikosaponins[J]. Fitoterapia, 2006, 77(2): 100-108. DOI:10.1016/j.fitote.2005.11.005
[19]
Xu J, Tanaka H, Shoyama Y. One-step immunochromatographic separation and ELISA quantification of glycyrrhizin from traditional Chinese medicines[J]. J Chromatogr B, 2007, 850(1-2): 53-88. DOI:10.1016/j.jchromb.2006.10.062
[20]
Zhao YN, Wang ZL, Dai JG, et al. Preparation and quality assessment of high-purity ginseng total saponins by ion exchange resin bombined with macroporous adsorption resin separation[J]. Chin J Nat Med, 2014, 12(5): 382-392.
[21]
Ouyang LF, Wang ZL, Dai JG, et al. Determination of total ginsenosides in ginseng extracts using charged aerosol detection with post-column compensation of the gradient[J]. Chin J Nat Med, 2014, 12(11): 857-868.
[22]
Joo EJ, Ha YW, Shin H, et al. Generation and characterization of monoclonal antibody to ginsenoside Rg3[J]. Biol Pharm Bull, 2009, 32(4): 548-552. DOI:10.1248/bpb.32.548
[23]
Wartchow CA, Podlaski F, Li S, et al. Biosensor-based small molecule fragment screening with biolayer interferometry[J]. J Comput Aided Mol Des, 2011, 25(7): 669-676. DOI:10.1007/s10822-011-9439-8
[24]
Abdiche Y, Malashock D, Pinkerton A, et al. Determining kinetics and affinities of protein interactions using a parallel real-time label-free biosensor, the Octet[J]. Anal Biochem, 2008, 377(2): 209-217. DOI:10.1016/j.ab.2008.03.035
[25]
Feng Y, Tian ZM, Wan MX, et al. Protein profile of human hepatocarcinoma cell line SMMC-7721:identification and functional analysis[J]. World J Gastroenterol, 2007, 13(18): 2608-1614.
[26]
Qi LW, Wang CZ, Du GJ, et al. Metabolism of ginseng and its interactions with drugs[J]. Curr Drug Metab, 2011, 12(9): 818-822. DOI:10.2174/138920011797470128
[27]
Peng D, Wang H, Qu C, et al. Ginsenoside Re:its chemistry, metabolism and pharmacokinetics[J]. Chin Med, 2012, 7(1): 2. DOI:10.1186/1749-8546-7-2
[28]
Obermeier B, Daneman R, Ransohoff RM. Development, maintenance and disruption of the blood-brain barrier[J]. Nat Med, 2013, 19(12): 1584-1596. DOI:10.1038/nm.3407
[29]
Alvarez JI, Katayama T, Prat A. Glial influence on the blood brain barrier[J]. Glia, 2013, 61(12): 1939-1958. DOI:10.1002/glia.22575
[30]
Serlin Y, Shelef I, Knyazer B, et al. Anatomy and physiology of the blood-brain barrier[J]. Semin Cell Dev Biol, 2015, 38: 2-6. DOI:10.1016/j.semcdb.2015.01.002
[31]
Bhowmik A, Khan R, Ghosh MK. Blood brain barrier:a challenge for effectual therapy of brain tumors[J]. Biomed Res Int, 2015. DOI:10.1155/2015/320941
[32]
Finsterwald C, Magistretti PJ, Lengacher S. Astrocytes:new targets for the treatment of neurodegenerative diseases[J]. Curr Pharm Design, 2015, 21(25): 3570-3581. DOI:10.2174/1381612821666150710144502
[33]
Park SY, Jung MY. UHPLC-ESI-MS/MS for the quantification of eight major gingerols and shogaols in ginger products:Effects of ionization polarity and mobile phase modifier on the sensitivity[J]. J Food Sci, 2016, 81(10): C2457-C2465. DOI:10.1111/1750-3841.13429
[34]
Ferrer I, Zweigenbaum JA, Thurman EM. Analytical methodologies for the detection of sucralose in water[J]. Anal Chem, 2013, 85(20): 9581-9587. DOI:10.1021/ac4016984
[35]
Saunders NR, Dziegielewska KM, Møllgård K, et al. Markers for blood-brain barrier integrity:how appropriate is Evans blue in the twenty-first century and what are the alternatives[J]. Front Neurosci, 2015. DOI:10.3389/fnins.2015.00385
[36]
Hasegawa H. Proof of the mysterious efficacy of ginseng:Basic and clinical trials:Metabolic activation of ginsenoside:Deglycosylation by intestinal bacteria and esterification with fatty acid[J]. J Pharmacol Sci, 2004, 95(2): 153-157. DOI:10.1254/jphs.FMJ04001X4
[37]
Lee J, Lee E, Kim D, et al. Studies on absorption, distribution and metabolism of ginseng in humans after oral administration[J]. J Ethnopharmacol, 2009, 122(1): 143-148. DOI:10.1016/j.jep.2008.12.012