Background
Cancer has long been the leading cause of death in the Republic of Korea. The National Cancer Information Center reported that the cancer incidence in the Republic of Korea was 285.7 per 100,000 in 2013, which is higher than the average (270.3) of the OECD countries [
1]. More than 76,000 people died from cancer and colorectal cancer (CRC) was ranked the 4
th accounting for 16.5 deaths per 100,000 people in 2014 [
2]. CRC is occasionally considered as a disease of “developed countries” because the incidence of CRC is relatively higher among the countries with high-income economy than the other counterparts [
3]. Since the western life style has prevailed among Koreans, the incidence of CRC has also increased continuously, and now, it has become one of the critical socio- economic burdens.
Oxaliplatin, a trade name for Eloxatin, is a third-generation platinum-based chemotherapeutic consisting of platinum coordinated with diaminocyclohexane and oxalate [
4]. Oxaliplatin was the first platinum-based chemotherapeutic approved for advanced CRC and has been commonly used to treat CRC in combination with folinic acid and 5-fluorouracil known as FOLFOX [
4,
5]. The cytotoxic activity of oxaliplatin is attributed to the formation of DNA adducts resulting from the intra- and inter-strand DNA crosslinks which inhibit DNA replication and transcription, followed by the generation of reactive oxygen species (ROS) and the ensuing DNA lesions and reduced glutathione depletion [
6,
7]. Along with its pharmaceutical values in oncology, however, oxaliplatin is also known to induce adverse side effects which are related to its hematopoietic, gastrointestinal, and neuronal toxicities [
4,
8,
9]. Oxaliplatin-induced peripheral neuropathy (OXIPN) is a member of a disease category called chemotherapy-induced peripheral neuropathy (CIPN), and is the most notorious, dose-limiting adverse side effect related with oxaliplatin [
10]. Clinical symptoms of OXIPN are onset of pain, numbness, tingling, cramping, and hypersensitivity to cold but not to heat in the hands or feet [
11‐
13]. It was reported that almost 50 % of cancer patients receiving a cumulative dose of oxaliplatin of 1,000 mg/m
2 or higher, which is the standard treatment protocol, experience OXIPN-related symptoms [
10], 40 % of patients continue to experience neuropathy for six months or longer, and 13 % of patients discontinue chemotherapy due to neuropathy [
14]. Unfortunately, however, the molecular mechanisms underlying development of CIPN including OXIPN have not been fully elucidated and no preventive or therapeutic agents for CIPN have been approved by the US FDA [
15]. Therefore, only symptomatic managements are provided to CIPN patients in clinics [
16]. As diagnostic technologies of cancer are getting more advanced and the structure of the world population moves fast towards aged societies, it is evident that the number of cancer patients receiving chemotherapy will increase and accordingly, the number of patients suffering from CIPN as well. Therefore, the development of novel efficient preventive or therapeutic agents targeting CIPN still remains a major medical unmet need in oncology.
In our preliminary study, we tried to find effective materials to relieve oxaliplatin-mediated neurotoxicity using a library of medicinal herb extracts that have been traditionally used. We screened extracts with biological activities to relieve oxaliplatin-mediated neurotoxicity using in vitro cell-based assays which employed the nerve growth factor (NGF)-stimulated neurite growth from rat pheochromocytoma PC12 cells, and found that the aqueous extract of Lithospermi radix (WLR) could effectively rescue the cells from the neurotoxicity of oxaliplatin. In the present study we further evaluated the in vivo efficacies of WLR using a neuropathic animal model, which suggested the possible mechanism underlying its neuroprotective potential. To our knowledge this is the first study reporting that WLR can efficiently attenuate the oxaliplatin-induced neurotoxicity in both in vitro and in vivo models without affecting the cytotoxicity profiles of oxaliplatin.
Methods
Plant materials and WLR preparation
The dried Lithospermi radix was purchased from Kwangmyungdang Medicinal Herbs Co. (Ulsan, Republic of Korea) which had been collected from the area of Cheongsong-gun, Gyeongsangbuk-do, Republic of Korea in March 2015. The information of collection time and site was written on the batch label. It was identified by Dr. Go Ya Choi, at the K-herb Research Center, Korea Institute of Oriental Medicine, Republic of Korea. A voucher specimen (KIOM-CRC#187) was deposited at the KM Convergence Research Division, Korea Institute of Oriental Medicine, Republic of Korea. The powder of the plant material (100.0 g) was extracted twice in 2 L distilled water using a reflux extraction system at 100 °C for 2 h. The extract was filtered through Whatman filter paper No. 2 (Whatman International, Maidstone, UK), concentrated at 40 °C using a rotary evaporation system (N-1200A, EYELA, Tokyo Rikakikai, Tokyo, Japan), and freeze-dried using a freeze dryer (FD8518, IlshinBioBase, Dongduchun, Republic of Korea) to produce a soluble-aqueous extract (WLR, 35.2 % yield). The freeze-dried WLR was homogenized, stored in a tight-sealed bottle containing silica gel pouches, and kept in a refrigerator with light protection. For the in vitro studies WLR was dissolved in phosphate-buffered saline (PBS), sterilized using a 0.22 μm syringe filter (Satorius, Goettingen, Germany), aliquoted in a small volume, and then stored at −80 °C until used. For the animal studies, WLR was suspended in 0.5 % (w/v) carboxymethylcellulose (CMC, Sigma, St. Louis, MO, USA) dissolved in distilled water right before the experiments.
In vitro cell culture
Rat pheochromocytoma PC12 cells (CCL-1721), human colorectal carcinoma HCT116 cells (CCL-247), and human lung carcinoma A549 cells (CCL-185) were obtained from the American Type Culture Collection (Rockville, MD, USA). Human neonatal foreskin fibroblast (HFFn, PC501A-HFF) cells were obtained from Systems Biosciences (Mountain View, CA, USA). PC12 cells were maintained on a 100 mm culture dish pre-coated with collagen type I (Corning, Kennebunk, ME, USA) containing DMEM supplemented with 10 % heat-inactivated horse serum, and 5 % non-heat-inactivated fetal bovine serum (FBS). A549, HCT116, and HFFn cells were maintained with RPMI1640, McCoy’s 5A, and DMEM, respectively, supplemented with 10 % heat-inactivated FBS. All culture media were supplemented with 100 U/mL penicillin and 100 μg/mL streptomycin to minimize microbial contamination. Basal media, serum, and antibiotics were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Cells were maintained at 37 °C in a humidified incubator aerated with 5 % CO2.
Cell viability
Cells were seeded at a density of 1,000 (HFFn) or 5,000 (HCT116 and A549) cells/well into a 96-well tissue culture plate and cultured overnight. The cells were exposed to drugs for 48 h and then, cell viabilities were determined with the EZ-Cytox cell viability assay kit (Daeil Lab Service, Seoul, Republic of Korea) measuring viable cells with normal mitochondrial function as described in the manufacturer’s instructions. The development of color was monitored and quantified at 450 nm using the Emax microplate reader (Molecular Devices, Sunnyvale, CA, USA). The relative viabilities were determined by comparing the color intensity of the drug treatment with those of the vehicle treatment. The half maximal inhibitory concentration (IC50) was calculated with the SoftMax Pro 4.0 software (Molecular Devices).
Neurite outgrowth assay
PC12 cells (1 × 104 cells/well) were plated on a 24-well tissue culture plate pre-coated with collagen type IV (Corning) containing a complete culture medium. After 24 h, the medium was replaced with serum free DMEM containing 100 ng/mL recombinant rat beta-NGF (R&D systems, Cat # 556-NG, Minneapolis, MN, USA), N2 supplement (Thermo Fisher Scientific, Cat # 17502048) and 0.5 % FBS to induce neurite growth in the presence or absence of oxaliplatin (200 nM, LC Laboratories, Woburn, MA, USA). WLR (25 and 100 μg/mL) or amifostine (0.5 mM, Santa Cruz Biotechnology, Dallas, TX, USA) were co-treated with oxaliplatin. Three days after the drug treatment, a morphometric analysis was done on digitalized images of PC12 cells taken with an inverted microscope (IX71, Olympus, Center Valley, PA, USA). Images of fields with more than 20 cells were captured and used for the image analysis. The differentiation of cells induced by NGF was examined, and cells that had at least one neurite with a length equal to the cell body diameter were counted. The number of neurite bearing cells was expressed as a percentage of the total cells. Total neurite length was determined by manually tracing the length of the neurite per cell using the MetaMorph image software (Molecular Devices).
Experimental animals
Eight-week-old male C57BL/6 mice were obtained from OrientBio Inc (Seongnam, Republic of Korea) and maintained in a specific-pathogen-free laboratory animal care facility. Animals were housed in a temperature (22 ± 2 °C)- and humidity (45 ± 10 %)-controlled vivarium (12/12-h dark/light cycle with lights on at 8 a.m.) with free access to food and water. Animals were acclimated for 1 week before the experiments. All procedures for animal care and experiments were reviewed and approved by the Korea Institute of Oriental Medicine Institutional Animal Care and Use Committee.
Induction of peripheral neuropathy in animals
Oxaliplatin was dissolved in a physiological saline solution (Samyang Anipharm, Seoul, Republic of Korea) and administered intraperitoneally (i.p.) at dose of 5 mg/kg. A week later, another dose of oxaliplatin (5 mg/kg, i.p.) was given to induce peripheral neuropathy (total 10 mg/kg). Control mice received the same volume of saline as oxaliplatin treatment.
Drug administration
One week after the final oxaliplatin injection, animals were divided into 3 groups, and each animal group received WLR (250 mg/kg in 0.5 % (w/v) CMC, oral, 6 days per week,
n = 8), vehicle (0.5 % CMC, oral, 6 days per week,
n = 8), or amifostine (100 mg/kg in 0.9 % saline, intraperitoneal, once a week,
n = 10) between 9–10 a.m. for 4 weeks. For comparison, saline control mice (i.e., no oxaliplatin treatment,
n = 8) received 0.9 % saline (vehicle for oxaliplatin) and 0.5 % CMC (vehicle for WLR). In pharmacopoeia of Korean traditional medicine, the
Lithospermi radix has been usually prescribed for nourishment of blood, enhancement of blood stream, detoxification, anti-inflammation, skin damage, etc. In those cases, decoction of 3–10 g of
Lithospermi radix is recommended as a daily dosage. However, there was no record of
Lithospermi radix for relieving or ameliorating the neuronal pain. Therefore, we used the recommended doses of
Lithospermi radix for other cases to evaluate its protective effect against OXIPN. In our lab, the yield of
Lithospermi radix extract in hot water (WLR) was almost 40 % and we could expect to get 1.2-4 g of WLR from 3–10 g of raw plant materials. On the assumption that the average body weight of healthy adults is 60 kg, the daily doses of WLR is 20–67 mg/kg. So, we chose 250 mg/kg for initial dose for our animal studies using mice, which is equivalent to 20 mg/kg in adult humans [
17].
Behavioral tests
The sensitivity of experimental animals to external mechanical stimuli were determined by an operator blinded to the status of the drug treatment in a quiet, temperature (22 ± 2 °C)-controlled behavior test room between 1–5 PM. Mice were individually placed in a black plastic chamber (5 × 5 × 8 cm) set on the perforated metal mesh-like open grid of square holes (5 × 5 mm). Animals were acclimated for at least 30 min before the behavior test. Mechanical sensitivity was determined using a calibrated von Frey filament with 0.4 g bending force (Aesthesio, BanMic Global, LLC, San Jose., CA, USA). The filament was applied to the plantar surface of the right hind paw for 1 s with an interval of 5 s between stimulations. The test was repeated 5 times with an interval of at least 5 min between each test cycle. The frequencies of paw withdrawal were calculated as a percentage of response (rapid and sudden lifting, shaking, or licking) occurring at least 3 times from a total of 5 trials.
Immunohistochemistry
Four weeks after the last oxaliplatin injection, the mice were deeply anesthetized by intraperitoneal injection of sodium pentobarbital (50 mg/kg, Hanlim Pharmaceuticals Inc, Seoul, Republic of Korea) and transcardially perfused with 4 % (w/v) paraformaldehyde prepared in 0.1 M PBS, pH 7.4). After perfusion, the L4-L5 spinal cords and hind footpads were removed and post-fixed in the same fixative solution for immunohistochemical and morphometric analyses. The L4-L5 spinal cords were verified by identification of the lumbar enlargement and nerve roots. Immunohistochemical analyses were done with specific antibodies against glial fibrillary acidic protein (GFAP) and ionized calcium binding adaptor molecule 1 (IBA1) as markers of activated astrocytes and microglias, respectively, or with a specific antibody against tumor necrosis factor-alpha (TNF-α) as a marker of inflammation using the fixed spinal cords. Immunohistochemical analysis was also done with a specific antibody against protein gene product 9.5 (PGP9.5) as markers of intraepidermal nerve fibers (IENF) using the fixed hind footpads. The paraffin-embedded spinal cords and footpads were sectioned at a 4 μm thickness and then mounted on silane- coated slides (Muto Pure Chemicals, Tokyo, Japan). Sections were deparaffinized and immersed in unmasking solution (Vector laboratory, Burlingame, CA, USA) and microwaved for 10 min at 98 °C. After cooling, sections were washed in Tris-buffered saline with 0.5 % tween 20 (TBS-T) and then, incubated with primary antibody solutions (mouse monoclonal anti-GFAP, 1:200; rabbit polyclonal anti-PGP9.5, 1:200, Millipore, Temecula, CA, USA; mouse monoclonal anti-TNF-α, 1:200, Abcam, Carlsbad, CA, USA; rabbit polyclonal anti-IBA1, 1:200, WAKO, Wako, Osaka, Japan) at 4 °C overnight. After 3 washes with TBS-T, sections were incubated in a secondary antibody labeled with Alexa Fluor 488 and 548 (1:2,000, Abcam) for 1 h at room temperature. After washing 3 times with TBS-T, sections were covered with an aqueous mounting medium (Vector shield, Vector laboratory), examined, and photographed under a fluorescence microscope (BX41, Olympus, 200X for GFAP and IBAI, 400X for TNF-α and PGP9.5). The images were optimized using an image analyzing software (Cellsense, Olympus). The densities of the immunoreactive GFAP and IBA1 antibody in the superficial dorsal horn laminae were determined with Image J (v1.43, NIH, USA) and expressed as a percentage of the immunostained area (% area) compared to the total image area. Quantification of IENFs in the foot pad was conducted as follows; 5 randomly chosen slices per animal were analyzed using a fluorescence microscope. Nerve fibers that perpendicularly penetrated into the epidermis from dermal/epidermal junction were counted in 3 fields of view from each slice. The length of the epidermis within each field of view was measured using the Cellsense image analyzing software. The IENF density was determined as the total number of fibers per field length of epidermis that branched after crossing the basement membrane (IENF/mm) as previously described [
18].
Morphometric analysis on the dorsal root ganglia
Dorsal root ganglions (DRGs) were obtained from L4-L5 and fixed as described above. Post-fixed DRGs were embedded in paraffin, sliced at a 4 μm thickness, mounted on charged slides, and subjected to Azan-Mallory staining (Azan Trichrome kit, Gennova Scientific, Sevilla, Spain) as described in the manufacturer’s instructions. The numbers of DRG neurons with multiple and/or eccentric nucleoli were counted at a high magnification (400X). The nucleolus was classified eccentric when located in the outer half of the radius of the nucleus, and the cell was classified multinucleolated when it showed more than 1 nucleolus in the cell body. At least 5 sections for each animal were analyzed, and the values were expressed as a percentage of DRG neurons with multiple and/or eccentric nucleoli compared to the total DRG neurons.
Statistical analysis
All statistical analyses were performed with SigmaPlot (v11.0, Systat Software, San Jose, CA, USA). The results are presented as the mean ± standard deviation (SD) of at least triplicate experiments. Differences were compared using one-way analysis of variance (ANOVA, parametric) or Kruskal-Wallis one-way ANOVA on ranks (nonparametric) according to Shapiro-Wilk normality test followed by the Tukey post-hoc test. The values with p < 0.05 were considered significant.
Discussion
As a member of the platinum-based chemotherapeutic anticancer drug family, oxaliplatin has widely been used as a key drug to treat patients with advanced and metastatic colorectal cancers as well as other cancer types, such as lung, breast, and ovarian cancers [
24]. It is one of the key anticancer drugs because it was chosen to be on a complementary list of the 19
th WHO model list of essential medicine (April 2015), category anti-neoplastics and immunosuppressives, subcategory cytotoxic and adjuvant medicine, representing essential medicines for priority diseases [
25]. However, oxaliplatin is also known to cause severe and disabling sensory peripheral neuropathy which is a mildly disrupting tingling sensation to extremely painful paresthesia [
26]. DRG neurons are a primary target of oxaliplatin-mediated peripheral neurotoxicity because the administered oxaliplatin accumulates and exerts neurotoxicity in the DRG where sensory neurons are located [
20]. Acute neuropathy that appears soon after drug administration includes acral paresthesia and dysesthesia which is especially enhanced by exposure to cold. However, chronic neuropathy that appears after repeated administration of the drug leads to dysfunction of sensory and motor neurons [
27]. The incidence and severity of chronic neuropathy induced by oxaliplatin in cancer patients can be predicted by those of acute neurological symptoms by oxaliplatin [
28]. Oxaliplatin is metabolized to oxalate and platinum metabolite, dichloro (1,2-diaminocyclohexane) platinum [
29]. It was known that chronic neuropathy involves morphological damages of the nerve caused by the platinum metabolite of oxaliplatin; however, acute neuropathy caused by oxalate is due to alterations in diverse signaling channels including voltage-gated Ca
2+, K
+, and Nav1.9 sodium channels, transient receptor potential-melastatin 8, and -ankyrin 1 channels [
13,
27]. According to the severity of the chronic neuropathy the chemotherapeutic dosing paradigm should be modified or delayed, or sometimes, the use of oxaliplatin should be discontinued. Therefore, sustained and chronic neuropathic pain by oxaliplatin is a dose-limiting adverse side effect and may greatly reduce the chemotherapy efficacy as well as quality of life of cancer patients. Unfortunately, however, efficient standard interventions against CIPN including OXPN are not yet clinically available.
In the present study, we demonstrated for the first time that the water extract of
Lithospermi radix, which has been used as a traditional medicine to treat diverse diseases including acute hepatitis, cancer, and wound healing [
30], can reduce the neurotoxicity of oxaliplatin in an NGF-PC12 in vitro model, and it can attenuate the oxaliplatin-induced nociceptive hypersensitivity in an animal model. Our study revealed that twice administration of oxaliplatin at a dose 5 mg/kg (total 10 mg/kg) in C57BL/6 mice that mimics the chronic OXIPN in patients induced enhanced sensitivity to nociceptive mechanical stimuli, and oral administration of WLR can relieve the oxaliplatin-induced mechanical hypersensitivity. We also showed that such mechanical hypersensitivity induced by oxaliplatin is accompanied by activation of both microglia and astrocytes in the dorsal horn of the mouse spinal cords. Astrocytes are related with regulation of CNS inflammation by exerting its diverse cytokines and perivascular microglia supplied from vessels are increased during CNS inflammation and after peripheral nerve damage [
31]. Involvement of the activation of spinal glial cells during the development of OXIPN has been previously reported with contradictory results [
23,
24,
32‐
34]. For example, Yoon [
32], Robinson [
33], and Zheng [
34] reported that only spinal astrocytes but not microglias were activated by oxaliplatin treatment. However, Di Cesare Mannelli [
23] and Ahn [
24] reported that the activation of both spinal microglias and astrocytes was observed following oxaliplatin treatment which is consistent with our results. These contradictory results may be due to the fact that they used different methods to make the oxaliplatin-induced neuropathic animal models, such as the dosing paradigm administered, time of analysis, and choice of experimental animals.
In addition to glial cell activation in the central nerve system, we also observed a significant loss of IENFs in the skin of oxaliplatin-treated C57BL/6 mice. However, the loss of IENF in the skin of foot pad by oxaliplatin was significantly recovered by oral administration of WLR. An IENF is a bare nerve ending comprised of unmyelinated axons from small-diameter sensory neurons [
35], and the loss of IENFs in the hands and feet is known to be a consistent pathological finding in patients receiving chemotherapy and experiencing numbness and loss of vibratory senses. Furthermore, loss of IENFs can also be observed in experimental animals by administration of neuropathy-inducing anti-tumor agents including oxaliplatin [
14,
36‐
38]. Xiao et al. reported results similar to ours in a rat OXIPN model showing that repeated administration of low doses of oxaliplatin in rats mimics the clinical development of chronic OXIPN in cancer patients including the induced hypersensitivities in mechanical/cold but not heat stimuli as well as loss of IENFs in the hind paw skin. They also reported the occurrence of atypical swollen and vacuolated mitochondria in peripheral nerve axons by electron microscopy analysis and slowing down of sensory neurons by a nerve conduction study [
13]. However, it has not been clearly determined whether the loss of IENFs in the skin is the cause or effect of the changes in neuronal sensitivity following chemotherapy treatment [
36].
The precise molecular mechanisms underlying the development of OXIPN remains unclear and should be fully elucidated to prevent or treat established OXIPN. Some research groups suggested that the development of OXIPN involves the activation of the immune system and subsequent induction of inflammation in the satellite cells of the DRG [
39] or in the dorsal horn of the spinal cord [
36]. Bioenergetic deficits by reactive nitrogen species-mediated inhibition of mitochondrial respiratory complex I and II in peripheral nerve sensory axons [
40] and mitochondrial dysfunction by enhanced production of ROS [
36] were also suggested as putative mechanisms of OXIPN. Remodeling of ion channel expression in nociceptors such as TREK1 [
41], up-regulation of N-methyl-D-aspartate (NMDA) receptor containing subtype 2B (NR2B) and its downstream target, neuronal nitric oxide synthase (nNOS) [
27], and enhanced phosphorylation of calcium/calmodulin-dependent protein kinases II [
42] were also reported be related with development of OXIPN. It is unambiguous that the development of efficient preventives or therapeutics targeting OXIPN is largely dependent on the understandings of the mechanisms of disease development.
In the present study, we showed that WLR can suppress the spinal activation of microglias and astrocytes and attenuate the mechanical hypersensitivity following oxaliplatin treatment. This finding suggests that WLR may exert an anti-inflammatory activity in neuronal immune cells to attenuate OXIPN. Administration of agents that can ameliorate the neurotoxicity of chemotherapeutics may allow for a stronger chemotherapy regimen like dose escalation. Unfortunately, however, conventional symptomatic management of neuropathic pain such as nortriptyline, amitriptyline, gabapentin, and lamotrigine failed to improve neuropathic patients in placebo-controlled, double-blinded clinical trials [
16]. Ca/Mg infusion and venlafaxine, a serotonin-norepinephrine reuptake inhibitor, are known to be effective in preventing OXIPN; however, they are not routinely used because of concerns related with decreased chemotherapy efficacies [
43]. Based on the mechanisms underlying the development of OXIPN, several preclinical studies reported some promising agents that exhibited neuroprotection against OXIPN. They include mitoprotective agents like acetyl-L-carnitine and olesosime [
13], peroxynitrite scavenger Mn(III) 5, 10, 15, 20-tetrakis (N-n-hexylpyridinium-2-yl) porphyrin [
23], immunomodulatory agents minocycline (microglial inhibitor) and fluorocitrate (astrocyte inhibitor) [
14,
23], NMDA receptor antagonists KM-801 and memantine or NR2B selective antagonist Ro25-6981 and ifenprodil [
27], non-selective NOS inhibitor L-N
G-nitroarginine methyl ester or nNOS selective inhibitor 7-nitroindazole [
27], and glucagon-like peptide-1 agonist exenatide [
42]. Such hopeful preclinical findings should be confirmed by prospective randomized, double-blinded, placebo-controlled clinical trials.
When developing supportive drugs to be co-administered with anticancer agents, the interaction between two drugs should be evaluated because a co-administered supportive drug can affect the antitumor activity of anticancer drugs in patients. In the present study, we demonstrated that co-administered WLR did not affect the antitumor activity profiles of oxaliplatin in both human colorectal (HCT116) and non-small cell lung carcinoma (A549) cells. In addition, WLR was revealed not to exert any cytotoxicity in normal cells. Based on this in vitro toxicity study, we expect that WLR can attenuate OXIPN without affecting the anti-tumor effects of oxaliplatin itself and it could be safe for use in preclinical and clinical studies. However, this herb-drug interaction in terms of the anti-tumor effect of oxaliplatin should be confirmed using a suitable xenograft animal model bearing established human cancer cells or patient-derived tumor tissues because we cannot exclude the possibility that co-administered WLR may affect the pharmacokinetics of oxaliplatin in animals.
Acknowledgments
The authors sincerely thank to Dr. Go Ya Choi, K-herb Research Center, Korea Institute of Oriental Medicine, for identifying Lithospermi radix. We also thank to Korea Institute of Oriental Medicine for the grant (K16294) supporting this study.