Background
Cancer is one of the leading causes of deaths worldwide [
1]. Chemotherapy is a standard treatment of cancers along with surgical operation and radiotherapy. Cytotoxic chemotherapy generally targets uncontrolled growth of cancer cells, and therefore, fast growing normal bystander cells like bone marrow cells are also affected by these therapies. As a member of taxane anticancer agents, docetaxel, brand name Taxotere, is widely used to treat a wide range of solid tumors including breast, ovary, prostate and non-small-cell lung cancers [
2]. It suppresses tumor cell growth by inhibiting mitotic spindle assembly leading to G2/M cell cycle arrest and eventually apoptotic cell death [
2,
3]. However, the limitation of docetaxel in clinical use is its side effects, such as hair loss, numbness, vomiting and especially myelosuppression. Serious hematopoietic injuries like decreased bone marrow cells lead to a delay in the drug treatment schedule, dose reduction, or early termination of the chemotherapy. Therefore, suitable treatment for myelosuppression is very important for the overall outcome of the chemotherapy as well as for the quality of life of cancer patients.
Traditional medicines have emerged as alternative treatments for chemotherapy-induced myelosuppression. Recent studies involving animal models or human trials have shown that traditional medicine, especially medicinal herb extracts, have benefits in the treatment of chemotherapy-induced myelosuppression. For example,
Fufang E’jiao Jiang, a famous formulation used in traditional Chinese medicine (TCM), efficiently relieved radiotherapy/chemotherapy-induced myelotoxicity in animal models [
4].
Panax ginseng with the active ingredient ginsenoside was able to recover experimental animals from 5-fluorouracil-induced myelotoxicity [
5]. Prescription of TCM decoctions,
Jian Pi and
Yi Qi Yang Xue Sheng Sui, could lower the risk of leukopenia and neutropenia and febrile neutropenia in breast cancer patients receiving anthracyclines combined with paclitaxel or docetaxel [
6]. Thus, traditional herbal medicines are considered good starting natural sources for the development of novel pharmaceuticals that would manage chemotherapy-induced myelosuppression.
Phragmites communis Trinius is a perennial plant that is globally distributed in the temperate and boreal regions of the world [
7]. The dried root of
P. communis,
Phragmitis rhizoma, has been prescribed in traditional Korean medicine (TKM) to relieve fever and vomiting and to nourish the body fluid (Korean Traditional Knowledge Portal) [
8]. It is registered in the Korean Herbal Pharmacopoeia as well as in the Chinese Pharmacopeia. In our in vitro activity screening test, the aqueous extract of
Phragmitis rhizoma (EPR) exhibited a protective effect against myelotoxicity induced by diverse chemotherapeutic agents with different mechanisms of action. In this study we show that EPR can ameliorate chemotherapy, especially, docetaxel-induced myelotoxicity in both in vitro hematopoietic colony forming unit (CFU) assay as well as in vivo animal models with primary endpoints focusing on changes in blood cell counts and microenvironmental structures of immune organs. To the best of our knowledge, this is the first report showing the pharmaceutical value of
Phragmitis rhizoma to treat chemotherapy-induced myelosuppression in cancer patients.
Methods
Chemicals and reagents
p-hydroxycinnamic acid, a reference standard for chromatographic analysis, was purchased from ChemFaces (Wuhan, Hubei, China). The purity of the p-hydroxycinnamic acid was ≥99.1% for high-performance liquid chromatography (HPLC). HPLC grad acetonitrile and analytical grad formic acid were purchased from JT Baker chemicals (Center Valley, PA, USA) and Wako chemicals (Osaka, Japan), respectively. Deionized distilled ultra-pure water was produced by Millipore RiQs & Milli-Q-Gradient water purification system (Millipore, Bedford, MA, USA). Docetaxel was purchased from LC laboratories (D1000, Woburn, MA, USA). CPT11 (camptothecin, C9911) and doxorubicin (D1515) were obtained from Sigma (St. Louis, MO, USA).
Plant materials and extract preparation
Phragmitis rhizoma was purchased from Kwangmyungdang Medicinal Herbs Co. (Ulsan, Republic of Korea) and identified by Dr. Goya Choi, K-herb Research Center, Korea Institute of Oriental Medicine, Republic of Korea. A voucher specimen (KIOM-CRC-#172) was deposited at the KM Convergence Research Division, Korea Institute of Oriental Medicine. The dried Phragmitis rhizoma (500.0 g) was finely ground using a mechanical pulverizer and then subjected to repeated extraction in boiling distilled water (100 °C, 2 h, 10 L × 2 times,) using a reflux extraction system from KOC Biotech (Daejeon, Republic of Korea). The extract solution was filtered through cotton wool, concentrated in the EYELA rotary evaporator system (Tokyo Rikakikai, Tokyo, Japan) and freeze-dried (FD8518, Ilshin Biobase, Dongducheon, Republic of Korea) to produce a final water extract (EPR, 60.21 g, 12.0%). The extract was stored at 4 °C in a dark plastic bottle with a silica gel desiccant until use.
Phytochemical analysis of the EPR was performed with the 1290 Infinity UHPLC system (Agilent Technologies, Waldbronn, Germany). The EPR was dissolved in 50% (
v/v) methanol at 10 mg/ml. A reference standard stock solution was prepared by dissolving
p-hydroxycinnamic acid in 50% methanol at 1 mg/ml. The EPR and standard stock solution were cleared with a 0.22 μm syringe filter. The standard stock solution was further 10-fold diluted with 50% methanol right before the UHPLC analysis. The EPR or diluted standard solution was loaded on the auto-sampler which was maintained at 4 °C. Chromatographic analysis was performed on the ZORBAX Eclipse Plus C
18 column (2.1 × 50 mm, 1.8 μm pore size, Agilent Technologies, Santa Clara, CA, USA). The linear gradient mobile phase was 0.1% formic acid in water (A) and 100% acetonitrile (B). The detailed conditions for the UHPLC analysis are summarized in Table
1. Chromatographic data were further processed by the OpenLAB CDS (ChemStation Edition) software (Agilent).
Table 1
UHPLC analytical conditions
Column temperature | 40 °C |
Detection wave length | UV at 290 nm |
Injection volume | 2.0 μl |
Flow rate | 0.3 ml/min |
Run time | 60 min |
Post time | 10 min |
Mobile phase | Time (min) | A (0.1% formic acid) | B (acetonitrile) |
0 | 100 | 0 |
50 | 60 | 40 |
51 | 0 | 100 |
60 | 0 | 100 |
In vitro toxicity assays
Normal human lung fibroblast IMR-90 (CCL-186) and neonatal foreskin fibroblast (HFFn, PC501A-HFF) cells were obtained from the American Type Culture Collection (Rockville, MD, USA) and Systems Biosciences (Mountain View, CA, USA), respectively. They were maintained in DMEM basal media which included 10% (v/v) heat inactivated fetal bovine serum (FBS), 100 U/ml penicillin and 100 μg/ml streptomycin antibiotics. DMEM, FBS and antibiotics were obtained from Thermo Fisher Scientific (Waltham, MA, USA). Human umbilical vein endothelial cells (HUVEC) were purchased from Lonza (Walkersvill, MD, USA) and maintained in EGM-2 basal medium containing endothelial growth medium supplements (Lonza). To determine the toxicity of the EPR itself, cells were inoculated at 50,000 cells/well in a 96-well plate and incubated for 24 h at 37°C in a humidified atmosphere of 5% CO2. Then, the cells were exposed to various concentrations of EPR for 48 h. Cell viabilities were determined using the Ez-Cytox cell viability assay kit (Daeil Lab Service, Seoul, Republic of Korea) according to the manufacturer’s instructions.
The hepatotoxicity of the EPR was assayed in vitro with the human hepatic cell line HepaRG. Frozen cells, William’s E basal medium, supplements for general maintenance or the toxicity assay were purchased from Thermo Fisher Scientific. Preparation of the media for general maintenance or the toxicity assay and differentiated HepaRG were done according to the manufacturer’s instructions. Cells that were differentiated in a collagen type I pre-coated 96-well plate (Corning, Kennebunk, ME, USA) were exposed to various concentrations of EPR for 24 h and then, hepatotoxicity parameters including cell viability (WST-1 reagent, Roche, Mannheim, Germany), intracellular glutathione (GSH) contents (GSH-Glo glutathione assay system, Promega, Madison, WI, USA), caspase 3/7 activity (Caspase-Glo 3/7 assay system, Promega) and lactate dehydrogenase (LDH) leakage (Cytotoxicity Detection kit, Roche) were determined as described by the manufacturer’s instructions.
CFU assay of hematopoietic progenitor cells
The in vitro CFU assay has been used to evaluate drug-related myelotoxicity in previous studies [
9,
10] and the CFU-GM assay was endorsed by the members of European Centre for the Validation of Alternative Methods (ECVAM) Scientific Advisory Committee (ESAC) as a substitute to experimental animals to predict myelotoxicity in humans [
11]. Mouse bone marrow nucleated cells (BMNCs) were isolated from the femoral bones of C57BL/6 mice as described previously with slight modifications [
12]. In brief, femoral bones were removed from a CO
2 euthanized mouse under sterile conditions and immersed in ice-cold Hank’s balanced salt solution (HBSS, Thermo Fisher Scientific) containing 100 U/ml penicillin and 100 μg/ml streptomycin. Both epiphyses of the femurs were removed with sterile scissors, and bone marrow cells were collected by strongly flushing the diaphysis with ice-cold HBSS using a 1 ml syringe. Red blood cells (RBC) were lysed using a 0.9% (
w/
v) NH
4Cl solution (StemCell Technologies, Cambridge, MA, USA). After washing with HBSS, BMNCs were resuspended in DMEM containing 1% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. Viable cells were counted using the ADAM-MC auto cell counter (NanoEnTek, Seoul, Republic of Korea). Viable 1×10
4 BMNCs were homogeneously dispersed in the Methocult GF M3434 medium (StemCell Technologies) containing a combination of EPR and anticancer drugs, plated in a 24-well culture plate (StemCell Technologies) and incubated at 37°C in a humidified atmosphere of 5% CO
2. After 7 days, colonies consisting of more than 30 cells were manually counted under an inverted microscope (IX71, Olympus, Center Valley, PA, USA).
Experimental animals
Six-week-old C57BL/6 male mice were supplied by OrientBio (Seongnam, Republic of Korea) and housed in a specific pathogen free laboratory animal care facility. Microbiological status of animals was evaluated and documented by the supplier. All animals were housed (5 mice/polycarbonate cage) and acclimated for 1 week before experiments in an environmentally controlled vivarium (temperature, 22 ± 2 °C; humidity, 45 ± 10; and dark and light cycle, 12 h/12 h). The animals accessed to water and commercial standard chow ad libitum. All procedures for the animal studies were in accordance with the relevant national raw and reviewed by the Institutional Animal Care and Use Committee of Korea Institute of Oriental Medicine (Protocol #16–075).
Animal studies
Docetaxel was dissolved at 60 mg/ml in a 1:1 mixture of ethanol and polysorbate 80 (Sigma) to make a stock solution. They were always freshly prepared when needed. To minimize crystallization of the docetaxel, the stock solution was further diluted to 2.4 mg/ml with injectable normal saline (Samyang Anipham, Seoul, Republic of Korea) right before injection. Experimental animals were randomly allocated in to 5 groups. Animals (Group 2 to Group 5) received docetaxel intraperitoneally at 30 mg/kg/day or vehicle (Group 1) for 3 consecutive days. Then, mice orally received EPR (Group 3, 30 mg/kg; Group 4, 100 mg/kg; Group 5, 300 mg/kg; n = 10 mice/group) or 0.5% (w/v) carboxymethylcellulose (CMC, Group 2, n = 10 mice, Sigma, St. Louis, MO, USA) as a vehicle for 4 days. At day 5, the animals were anaesthetized by intraperitoneal injection of 50 mg/kg pentobarbital solution (Entobar, Hanlim Pharmaceuticals Inc., Seoul, Republic of Korea). Whole blood was collected from the heart using a 1 ml syringe with 25 gauge pre-coated with 10% (w/v) ethylenediaminetetraacetic acid (EDTA) dipotassium salt (Sigma) for the blood cell counts. Half of the spleens were rapidly frozen in liquid nitrogen and store at −70 °C until total RNA preparation. The other half of the spleen and the whole thymus and femurs isolated from the mice were soaked and fixed in a neutral buffered formalin (pH 6.8–7.2, BBC Biochemical, Mount Vernon, WA, USA) for 72 h at 4 °C with gentle shaking.
Complete blood count (CBC) analysis
CBC analysis of EDTA-treated whole blood was determined using a HEMAVET 950 automatic analyzer (Drew Scientific, Miami Lakes, FL, USA).
Isolation and culture of splenocytes and thymocytes
All procedures except for animal euthanasia were performed in an aseptic area. Thymus and spleen were isolated from a normal 6-week-old male C57BL/6 mouse that was euthanized by CO2 inhalation. The tissues were transferred to a 70 μm pore-sized meshed cell strainer (Thermo Fisher Scientific) that was submerged in ice-cold RPMI medium containing 100 U/ml penicillin and 100 μg/ml streptomycin. The tissues were gently mashed with a 1 ml syringe rubber plunger, and then, the cells were transferred to a 50 ml conical tube. The cell were washed once using RPMI with antibiotics and collected at 1600 rpm for 5 min at 4 °C. Then, the cells were subjected to osmotic shock by adding 450 μl of sterile distilled water for 5 s to lyse the RBC and neutralized by adding 50 μl of 10× phosphate buffered-saline (PBS, pH 7.4). The cells were centrifuged and resuspended in a 1 ml pre-warmed complete medium which was RPMI supplemented with 10% FBS and antibiotics. Viable cells at 5 × 105 in 95 μl of complete medium were seeded in a 96-well culture plate (Nunc, Naperville, IL, USA) pre-loaded with 5 μl of 20× concentrated drugs and incubated at 37 °C in a humidified 5% CO2 incubator. After 48 h, cell growth was determined using a WST-1 cell proliferation reagent (Sigma) according to the manufacturer’s instruction.
Enzyme-linked immunosorbent assay
Quantification immunomodulatory cytokines released from splenocytes was determined using commercially available enzyme-linked immunosorbent assay (ELISA). The ELISA kits detecting murine interleukin-3 (IL3, ab113345) and IL6 (ab100712) were obtained from Abcam (Cambridge, MA, USA). The kits for murine interferon-gamma (IFNγ, MIF00) and tumor necrosis factor-alpha (TNFα, MTA00B) were obtained from R&D systems (Minneapolis, MN, USA). The culture medium was clarified by centrifugation at 12,000 rpm for 1 min, and the supernatant was subjected to ELISA assay following the manufacturer’s instruction. Horseradish peroxidase-mediated
3,3',5,5'-tetramethylbenzidine color development was quantified at 450 nm using the Emax microplate reader (Molecular Devices, Sunnyvale, CA, USA). The concentrations of the cytokines released in the medium were quantified by comparison with serially diluted standards that were included in the kits.
Determination of cytokine expression in splenic tissues and BMNCs
Total RNAs were isolated from frozen spleens using the Easy-Spin™ total RNA extraction kit (iNtRON Biotechnology, Seoul, Republic of Korea) according to the manufacturer’s instructions. BMNCs were prepared as described above. After determining the cell viability, viable cells at 4-5 × 10
6 in 490 μl of RPMI supplemented with 10% FBS and antibiotics were plated in a 12-well culture plate (Nunc) pre-loaded with 10 μl of 50× concentrated drugs. After 24 h, the cells were washed with ice-cold PBS and total RNAs were isolated using the Easy-Spin™ total RNA extraction kit. Single-stranded cDNA was synthesized from 1 μg of total RNA using the iScript™ cDNA synthesis kit (Bio-Rad, Hercules, CA, USA). Quantitative real-time polymerase chain reaction (qPCR) was performed using the SsoAdvanced Universal SYBR Green Super mix (Bio-Rad), and fluorescence intensities were detected by the CFX96 real time system (Bio-Rad). All primers except those for IL3 (MQP029461, GeneCopoeia, Rockville, MD, USA) were synthesized and provided by Genotech (Daejeon, Republic of Korea). The information for the primer sets and enzyme reactions are summarized in Table
2.
Table 2
Primers and quantitative PCR conditions to determine expression of murine immunomodulatory cytokines
Target | Forward (5′ → 3′) | Reverse (5′ → 3′) | GenBank Accession # |
GAPDH | TTGATGGCAACAATCTCCAC | CGTCCCGTAGACAAAATGGT | NM_001001303 |
GM-CSF | CCGTAGACCCTGCTCGAATA | TGCCTGTCACATTGAATGAA | NM_009969.4 |
IFNγ | ATGAACGCTACACACTGCATC | CCATCCTTTTGCCAGTTCCTC | NM_008337 |
IL1β | GCAACTGTTCCTGAACTCAACT | ATCTTTTGGGGTCCGTCAACT | NM_008361 |
IL3 | | NM_010556.4 |
IL6 | TAGTCCTTCCTACCCCAATTTCC | TTGGTCCTTAGCCACTCCTTC | NM_031168 |
IL7 | TTCCTCCACTGATCCTTGTTCT | AGCAGCTTCCTTTGTATCATCAC | X07962.1 |
IL12β | TGGTTTGCCATCGTTTTGCTG | ACAGGTGAGGTTCACTGTTTCT | NM_008352 |
SCF | ATGTTCCCCGCTCTCTTTGG | GTGTGGCATAAGGGCTCACT | NM_013598.2 |
TNFα | CCCTCACACTCAGATCATCTTCT | GCTACGACGTGGGCTACAG | NM_013693 |
Step | Temperature (°C) | Time (s) | # of cycle |
Activation | 95 | 30 | 1 |
Amplification | 95 | 15 | 40 |
60 | 30 |
Melting-curve | 65–95 | 2 s/step | 1 |
Histological analysis
Formalin fixed femurs were treated with Decalcifying solution-Lite (Sigma) according to the manufacturer’s instructions. Formalin-fixed spleen, thymus and decalcified femurs were embedded in paraffin. Microscopic slides with a 4 μm thickness were prepared from the paraffin-embedded tissues and then subjected to hematoxylin-eosin (H&E) staining. Histological examination and capture of digital images were performed under a light microscope (BX41, Olympus).
Statistics
Statistical analyses were done with SigmaPlot (v11.0, Systat Software, San Jose, CA, USA). Differences among the means were compared with one-way analysis of variance followed by the Tukey post-hoc test. The differences with a p < 0.05 were considered significant.
Discussion
Docetaxel is widely prescribed as an adjuvant chemotherapy to treat various types of solid tumors [
2]. However, the adverse side effects of docetaxel, such as myelosuppression, limit its clinical use [
2]. Modification of the docetaxel formulation using nanotechnologies is under investigation to minimize the docetaxel myelotoxicity [
2]. However, hematopoietic growth factors (HGFs) like recombinant granulocyte-colony stimulating factor (G-CSF) and erythropoietin are now first-line choices for the treatment of myelosuppression in cancer patients. The drawback of HGFs, however, are their high costs and their own adverse effects like osteomuscular pain, arthralgia and allergic reactions [
4]. Furthermore, a recent study raised a critical safety issue of G-CSF in cancer patients, that is, its potential to promote tumor growth by enhancing neovascularization in a tumor [
14]. Therefore, the development of efficient and safer therapeutics or preventives are still needed to manage cancer patients.
A variety of traditional herbal medicines have been studied to evaluate their preventive or therapeutic efficacies against chemo−/radiotherapy-induced myelosuppression, and they have shown some promising pharmaceutical values. They have consisted of extracts of a single medicinal herb [
15‐
17], an herbal decoction or extract combination [
4‐
6,
18‐
21], or active fractions or single components isolated from medicinal herbs [
1,
22‐
25]. Their protective effects have been investigated using experimental animal models [
1,
4,
5,
15‐
18,
20‐
25] or clinical human studies enrolling cancer patients [
6,
19]. However, to the best of our knowledge, this is the first report on using a traditional folk medicine to reverse docetaxel-induced myelotoxicity and on the myelo-protective effect of
Phragmitis rhizoma.
In this study, we showed that the aqueous extract of
Phragmitis rhizoma (EPR) ameliorated docetaxel-induced myelosuppression in both in vitro and in vivo experimental models. In summary, EPR efficiently reduced myelotoxicity induced by anticancer agents that have different mechanisms of action (docetaxel, CPT11 and doxorubicin; Fig.
2 and Additional file
1: Figure S1). In our mouse model, repeated injection of docetaxel remarkably decreased the WBC (38%), neutrophil (38%) and lymphocyte (46%) counts in the blood. RBC (15%) were more resistant to docetaxel toxicity than that of the WBC, which may be due to their longer lifespan compared to WBC in the circulating blood [
25]. Unexpectedly, the platelet count was increased by docetaxel which was consistent with the previous clinical study reporting that co-administration of docetaxel could spare platelet counts in cancer patients who received carboplatin by elevating circulation TPO concentration in blood [
26]. EPR could slightly attenuate this docetaxel-induced platelet increase. Thus, the EPR treatment had the following benefits: blood cell counts rebounded (Table
3), microenvironmental niches in the bone marrow improved, and the structural integrity of the thymus was preserved from docetaxel toxicity (Fig.
3).
In our animal model, IL3 expression was significantly suppressed in splenic tissues following exposure to docetaxel. IL6 expression did not change but SCF and GM-CSF were slightly increased by the docetaxel treatment. Elevated expression of immune-stimulating cytokines has also been observed in other different myelosuppressed animal models. For example Liu et al. reported that IL1β, IL3 and IL6 expression was increased in spleens, and IL6, SCF and GM-CSF expression was increased in bone marrow stromal cells in combined radiotherapy/cyclophosphamide/chloramphenicol-induced myelosuppressed mice [
4]. In our docetaxel model, it was shown that EPR relieved the suppressed IL3 expression and promoted the expression of immune-stimulating cytokines like IL6, SCF and GM-CSF in splenic tissues (Fig.
4). Furthermore, IL3, IL6, IFNγ and TNFα were induced by EPR in normal splenocytes (Fig.
6) whereas IL1β and IFNγ were induced in bone marrow cells (Table
4). The different patterns of cytokine expression response to EPR may be partially due to different cell populations residing in two immune organs.
IL3 is known to potentially support the proliferation and expansion of early myeloid progenitors leading to bone marrow reconstitution, aplastic anemia and hematopoietic recovery following chemotherapy [
27]. GM-CSF stimulates hematopoietic stem cells to produce granulocytes and macrophage myeloid cells [
5]. Except for GM-CSF and IL3, other cytokines and SCF are known to have little or no effect on the proliferation of hematopoietic progenitor cells when acting alone but enhance the response to other HGFs [
28]. IL1β, an isoform of IL1, is known to induce a variety of HGFs including GM-CSF. IL6 is known to stimulate hematopoietic stem cells, granulocyte-macrophage progenitor cells and erythroid progenitor cells, and can accelerate hematopoietic recovery [
29]. A synergistic myelostimulating effect was observed between IL1β and IL6 in cyclophosphamide-induced myelosuppressed mice which was not recovered by single regime [
28]. Administration of IL1 with IL6 reconstitutes hematopoietic functions in chemotherapy/radiation-induced myelosuppressed mice [
30]. Enhancement of IFNγ expression may have a positive role in immune suppressed cancer patients by providing them protective immunity against infections [
5]. TNFα is known to suppress the growth of erythroid and myeloid progenitors resulting in the inhibition of erythropoiesis. However, TNFα can mediate the maturation of a granulocyte-macrophage lineage and may protect myelosuppressed cancer patients from bacterial infections by increasing bacteriocidal phagocytic cytotoxicity [
5,
31]. Therefore, EPR may exert its myelo-protective effect by inducing the expression of diverse hematopoietic cytokines and by their combinatorial effects. In addition, EPR and its major component
p-hydroxycinnamic acid could ameliorated docetaxel-induced myelosuppression in vitro (Fig.
2) and promote the proliferation of splenocytes and thymocytes (Fig.
5). However,
p-hydroxycinnamic acid did not remarkably modulate immune-stimulating cytokines in splenocytes as much as EPR (data not shown). Taken together,
p-hydroxycinnamic acid partially contributes to the protective activity of EPR against docetaxel myelotoxicity. Myelo-protective effect of
p-hydroxycinnamic acid should be confirmed in in vivo animal model in our future study. In the present study we cannot rule out the fact that active substances other than
p-hydroxycinnamic acid or combination among them may also contribute to myelo-protective potential of EPR. The identity of other component(s) should be further elucidated to fully understand the mechanism of the EPR.
Prediction of potential toxicity of a candidate at the early drug developmental stage is critical to minimize the risk of failure. Especially, drug-induced liver injury is the most frequent cause of post-marketing warnings and withdrawals [
32]. In this study, we also estimated the toxicity of the EPR in 2 normal human fibroblasts (IMR-90 and HFFn) and 1 normal endothelial (HUVEC) cells. In all the tested cells, the EPR did not show any remarkable toxicities within 0–200 μg/ml ranges of the EPR. Furthermore, we predicted the hepatotoxicity of the EPR using multi-parametric hepatotoxicity tests including cell viability, apoptotic caspase 3/7 activation, intracellular GSH contents and LDH leakage in HepaRG. Although the HepaRG hepatic cell line was constructed from a human hepatocellular carcinoma [
33], they still contain a variety of enzyme activities seen in primary human hepatocytes, such as cytochrome P450s, P-gp membrane transporters and phase II metabolizing enzymes that are responsible for normal liver function [
32,
34]. No remarkable hepatotoxicity in the HepaRG was observed even when the cells were exposed to up to 500 μg/ml EPR. Overall, EPR is a safe and potential candidate for novel drug development to treat chemotherapy-induced myelosuppression in cancer patients based on the present efficacy and toxicity studies.