www.e l sev i e r. com/ loca te / i n t imp

International Immunopharmacology (2008) 8, 1196–1206

The herb medicine formula “Yang Wei Kang Liu”improves the survival of late stage gastric cancerpatients and induces the apoptosis of human gastriccancer cell line through Fas/Fas ligand andBax/Bcl-2 pathways☆

Jie Li ⁎, Gui-Zhi Sun, Hong-Sheng Lin, Ying-Xia Pei, Xin Qi, Cheng An,Jun Yu, Bao-Jin Hua

Department of Oncology, Guang An Men Hospital, China Academy of Chinese Medical Sciences, Beijing, 100053, PR China

Received 19 February 2008; received in revised form 8 April 2008; accepted 14 April 2008

Abbreviations: YWKLF, formula “Yaargyrophilic nucleolar organizer region☆ This project was supported by gran⁎ Corresponding author. Tel./fax: +86E-mail addresses: drjieli2007@gma

1567-5769/$ - see front matter © 200doi:10.1016/j.intimp.2008.04.007

Abstract

The herb medicine formula “Yang Wei Kang Liu” (YWKLF) has been used to inhibit the metastasisof human gastric cancer to prolong patient survival. In this study, we evaluated the effect ofcombination of chemotherapy with YWKLF on the survival of stage IV gastric cancer patients andthe potential mechanisms of YWKLF by focusing on its capacity to activate apoptotic pathways inhuman gastric cancer cell line MGC-803. We found that combination of chemotherapy with oraladministration of YWKLF significantly increased the survival of stage IV gastric cancer patients. Inan approach of “serum pharmacology” in which sera were collected from rabbits orallyadministered with YWKLF and examined for their anti-tumor cell activity in vitro, we observedthat sera from rabbits administered with YWKLF induced the apoptosis of MGC-803 cells bycausing the loss of mitochondrial membrane potential, increasing the expression of Fas proteinand Bax mRNA, as well as down-regulating Fas-L mRNA. Our results suggest that activation ofmajor pro-apoptotic pathways may account for the anti-gastric cancer activity of YWKLF, whichmay provide a basis for isolation and identification of more highly effective anti-cancercomponents.© 2008 Elsevier B.V. All rights reserved.

KEYWORDSHerb medicine;Serum pharmacology;Apoptosis;Gastric carcinoma;Bax/Bcl-2;Fas/Fas ligand

ng Wei Kang Liu”; Sub-F1, Sub-Fs; AO, acridine orange; Δ ψm, mitts No. 30600839 and No. 306727610 88001340.

il.com, drjieli@yahoo.com (J. Li).

8 Elsevier B.V. All rights reserved.

ormula 1; Sub-F2, Sub-Formula 2; Sub-F3, Sub-Formula 3; AgNORs,ochondrial membrane potential.4 from National Natural Science Foundation of China.

1197YWKLF prolonged the overall survival time of stage late gastric cancer patients

1. Introduction

Gastric cancer is one of the most common malignant tumorsin China [1]. Facing the problems of lower rate of earlydiagnosis, higher possibility of relapse and metastasis aftersurgery, and poor sensitivity to radiotherapy or/and che-motherapy, herb medicine plays a role in preventing gastriccancer from relapse and metastasis [2]. A herb medicineformula “Yang Wei Kang Liu” (YWKLF), consisting of sixherbs, has been used for a number of years in traditionalmedicine to promote circulation and human body resistanceto pathogens and cancer. Preliminary clinical studies haveshown that YWKLF modulates immune function, improvesthe quality of life, and prolongs survival time of cancerpatients [3]. However, the molecular mechanisms underlyingthe anti-cancer effect of YWKLF are poorly understood.

Apoptosis is an important component of various biologicalprocesses, such as development, maintenance of tissuehomeostasis and elimination of transformed cells. Theaberrant regulation of apoptosis contributes to many typesof human diseases including cancer. Apoptosis uses two mainsignaling mechanisms: the “extrinsic” and the “intrinsic”pathways, to trigger programmed death of cells. The Fas–Fasligand (FasL) pathway is the most studied in the cells throughextrinsic pathway [4–6]. The Bcl-2 family of proteins

2. Materials and methods

constitutes a critical intrinsic checkpoint coupled to cellapoptosis. The ratio of anti- to pro-apoptotic molecules,such as Bax/Bcl-2, sets the threshold of cell susceptibility toapoptosis of the Bax/Bcl-2 pathway, which utilizes organellessuch as mitochondria to amplify death signal [7]. Ourprevious study found YWKLF regulated Fas/FasL expressionon T lymphocytes of gastric cancer patients [8]. Thus, it ispossible that YWKLF may inhibit human cancer by inducingapoptosis.

Due to the difficulties in standardizing the studies offormulated herb medicine, “serum pharmacology” has beenproposed by Japanese scholars in the 1980s [9], in which asingle herb or herb mixture is administered orally to animals,followed by collecting animal sera for testing on thebiological behavior of tumor cells. The methodology hasbeen believed to better reflect the maintenance of herbcomponents and metabolism through a host, therefore maybe more informative than drug extracts when tested in vitro[10].

In the present study, we report the effect of combinationof YWKLF with chemotherapy on prolonging survival of stageIV gastric cancer patients. In addition, we report that serafrom rabbits administered with YWKLF promote the apopto-sis of human MGC-803 gastric cancer cells through Fas/Fasligand and Bax/Bcl-2 pathways.

2.1. Patients and the treatment

A total of 123 patients with stage IV gastric cancer were enrolled during the period of 1990 to 1997. The study was permitted by InstitutionalReview Board (IRB), and informed consents (IC) were obtained from the patients before they were recruited into this study. All patients had

Karnofsky performance status N60 and showed adequate hematological, liver and renal functions. Patients were recruited to YMKLF+chemotherapy and chemotherapy alone cohorts by random number tables with informed consent. Patients assigned to chemotherapy groupreceived as MFP chemotherapy (cisplatin 60 mg/m2 on day 1, 5-FU 500 mg/m2/day, on day 1–5 and MMC 6 mg/m2/day, on day 1, 8) every4 weeks, for 4 cycles. In chemotherapy+YWKLF group: patients received MFP chemotherapy in combination with oral administration oflyophilized YWKLF granules, 12 g twice daily. Changes in tumor size and Karnofsky performance status were analysed analyzed every2 months.

2.2. Cells and reagents

Human gastric cancer cell line MGC-803 was purchased from Cell Culture Centre, Institute of Basic Medical Sciences, Chinese Academy ofMedical Sciences and cultured in RPMI 1640 (GIBCO/BRL) containing 10% fetal bovine serum (FBS). Rabbit anti-human Fas and FasL polyclonalantibodies and FITC-labeled goat anti-rabbit IgG antibody were obtained from Santa Cruz Corporation (Beijing, China). Annexin V-FITCApoptosis Detection Kit was from BD Corporation (Shenzhen, China). TRIzol Reagent was purchased from Invitrogen/GIBCO (Beijing, China).Reverse Transcriptase, Taq DNA polymerase and other reagents used in reverse transcription PCR (RT-PCR) were obtained from PromegaCorporation (Beijing, China).

2.3. Animals

Japanese flap-eared rabbits were purchased from the Institute of Laboratory Animal Science, Chinese Academy of Medical Sciences (CAMS)(Beijing, China) and maintained in the animal facility approved by institutional policies.

2.4. Formula preparation

YWKLF is composed of Astragalus mongholicus Bge (450 g), Panax notoginseng (Burk.) F. H. Chen (40 g), Panax ginseng C.A. Mey (100 g), Parispolyphylla Smith var. chinensis (Franch.) Hare (150 g), Caesalpinia sappan L (60 g) and Oldenlandia diffusa (Willd.) Roxb (180 g). Based on theprinciples of herb action, YWKLF was sub-formulated into Sub-Formula1 (Sub-F1): Astragalus mongholicus Bge (450 g), Panax ginseng C.A. Mey(100 g), that improve the general body conditions [11]; Sub-Formula 2 (Sub-F2): Panax notoginseng (Burk.) F. H. Chen (40 g) and Caesalpiniasappan L (60 g), that improve circulation [12–14]; and Sub-Formula 3 (Sub-F3): Paris polyphylla Smith var. chinensis (Franch.) Hare (150 g) andOldenlandia diffusa (Willd.) Roxb (180 g), that contain anti-oxidants [15]. All medicinal plants used to prepare formulae were provided byGuang An Pharmacy (Beijing, China), the species, plant parts, and origin used in the formula as Table 1. The plant were homogenized with aWaring blender, then soaked in 12 Liter double distilled water (DDW) for 24 h. The mixture was heated to 100 °C for 1 h, and the decoction wasfiltrated. The filtrates obtained from 3 cycles of the procedures were mixed, concentrated by heating and granulated by lyophilization. Total

Table 1 The component of YWKLF

Species (family) Plant part Origin

Sub-Formula 1Astragalus mongholicus Bge Root Neimenggu, ChinaPanax ginseng C. A. Mey Root JiLin, China

Sub-Formula 2Panax notoginseng (Burk.) F. H. Chen Root Yunnan, ChinaCaesalpinia sappan L Heartwood Yunnan, China

Sub-Formula 3Oldenlandia diffusa (Willd.) Roxb Leaves Henan, ChinaParis polyphylla Smith var. chinensis (Franch.) Hare Rhizome Yunnan, China

1198 J. Li et al.

yield of the YWKLF extract is 125 g lyophilized powder from water extract of 1 kg raw mixed herb. Aqueous solution was prepared by dissolvingthe granulated formulae in sterile water for animal administration at 1.63 g raw mixed herb/ml for YWKLF, 0.92 g raw mixed herb/ml for Sub-F1, 0.17 g raw mixed herb/ml for Sub-F2 and 0.55 g raw mixed herb/ml for Sub-F3. The quality control on YWKLF preparation, includingdefinition of the correct plants, origin of production, implantation, harvesting, and processing. Good manufacturing practice has been used toproduce lyophilized particles. Contamination of hazardous agents such as pesticide residues, arsenic and heavy metals, microorganismsincluding aflatoxins, and sulfur dioxide residue were excluded. Consequently, YWKLF and its preparations were standardized, regulated andquality controlled according to the guidelines defined by Chinese State Food and Drug Administration (SFDA).

2.5. Method of “serum pharmacology”

“Serum pharmacology” has been used to study the in vitro pharmacological activity of herb medicine. The preparation of “serumpharmacology” was based on the described procedures [16]. Briefly, 18 rabbits were randomly divided into six groups and administered bygavage of YWKLF, Sub-F1, Sub-F2 and Sub-F3. The rabbits of saline and chemotherapy groups were fed with 10 ml saline twice daily for 3 daysper rabbit. YWKLF and its sub-formula groups were treated with 10 ml YWKLF or a sub-formula twice daily for 3 days per rabbit. The formuladose was estimated as 10 times of doses given to patients in the clinic. Two hours after the fifth treatment on the third day, blood was obtainedfrom aorta of the rabbits under sterile conditions, and was allowed to coagulate at 25 °C for 4 h. The sera were separated by centrifugation at2500 rpm for 20 min. After filtration through 0.45 μm cellulose acetate membrane twice, the sera were treated in 56 °C water bath for 30 minthen were stored at −20 °C until use.

2.6. Cell viability assay by MTT

Cell viability was assessed using 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) dye reduction assay (Sigma, USA). MGC-803 cells were seeded at a density of 1×104 cells per well into 96-well plates in RPMI 1640 containing 10% FBS. After 24 h, the cells were washedand further cultured in the following media: the saline groups: RPMI 1640 containing 10%, 20% and 30% sera from rabbits treated with saline;the YWKLF groups: RPMI 1640 containing 10%, 20% and 30% sera from rabbits treated with herb formulae; the chemotherapy groups: 10%, 20%and 30% sera from saline saline-treated rabbits in RPMI 1640 plus cisplatin at a final concentration of 2 μg/ml. The cells were incubated in 5%CO2 at 37 °C for 48 h, then with 5 mg/ml MTTat 10 μl/well for 4 h. The culture media were discarded and substituted with DMSO at 100 μl/well.The plates were mounted on a micro-mixer for 5 min to thoroughly dissolve the blue MTT/DNA granules. The plates were then placed in amicroplate reader. The OD was measured at 570 nm. The inhibition rate was calculated according to the following formula:

Inhibition rate kð Þ ¼ 100k� 1� mean absorbance of cells in sera f rom drug treated rabbits=mean absorbance for cells in sera f rom rabbits treated with salineð Þf g:

2.7. Morphometrical study of argyrophilic protein associated with the nucleolar organizer regions (AgNORs) inMGC-803 cells

After 24 h treatment with rabbit sera, 200 μl cell suspensions was loaded into a cytospin chamber and spun for 10 min at 300 rpm. Slides werefixed by a mixture of methanol and glacial acetic acid at 3:1 vol/vol. The silver colloid solution for AgNORs staining was prepared by dissolvinggelatin in 1% aqueous formic acid to a concentration of 2%, and this solution was mixed at 1:2 vol/vol with 50% aqueous silver nitrate in whichslides were immersed for 45 min at 20 °C in dark [17,18]. After rinsing in H2O, the slides were immersed in 5% sodium thiosulfate for 45 min atroom temperature. Counterstaining was carried out using methyl green.

The parameters of nucleolin and AgNOR were assessed by an image analysis system (KL-2, Beijing). The stained dots of only one focal planeper nucleus, scanned from light microscopy image using a ×100 oil-immersion lens, were converted via a TV camera to appear as dots on a high-resolution monitor. The mean number and area of stained dots were measured using an image analyzer. Measurements were made for 150 cellsin random areas of each specimen. The morphometric parameters were means of areas and numbers of AgNORs per nucleus and the ratio ofAgNORs area to nucleolin/nucleus area per cell.

2.8. Detection of cell apoptosis by flow cytometry

MGC-803 cells seeded in 6-well plates at 106/well were cultured in the presence of sera from rabbits received saline or formulae for 24 h. Thecells were then washed and fixed in 70% ethanol at 4 °C overnight. The cells were then re-suspended in 10 μl propidium iodide (PI) solution

1199YWKLF prolonged the overall survival time of stage late gastric cancer patients

(2 mg/ml) with 50 μl RNAase (10 mg/ml) and incubated in the dark at 37 °C for 30 min, followed by DNA content analysis using a FACSort flowcytometer (Becton Dickinson). The CellQuest program was used to analyze the results. Different phases of cell cycle were assessed bycollecting the signal at channel FL2-A. The percentage of the cell population at a particular phase was estimated by ModFit LT.

Annexin V-FITC/PI staining was performed after the cells were treated with rabbit sera for 24 h in vitro. Rabbit sera treated cells (2×106/group) were rinsed with ice-cold PBS and then re-suspended in 200 μl of reaction buffer. Annexin V stock solution (10 μl) was added to the cellsand incubated for 30 min at 4 °C. The cells were further mixed with 5 μl PI and immediately analyzed on a FACSort equipped with CellQuestsoftware.

2.9. Assessment of morphological changes by acridine orange (AO) staining

Cell slides were fixed in 95% ethanol for 10 min on ice and air dried for 10 min. The slides were stained with AO (6 μg/ml; sigma), thenexamined by laser scanning confocal microscopy (Radiance2000, BIO-RAD) using a 488 nm argon laser excitation beam, a 515 nm beam filterfor DNA, and a 630 nm beam filter for RNA. Apoptotic cells were characterized by morphological alterations such as condensed nuclei andcell shrinkage. AO is an intercalating, nucleic acid specific fluorochrome, which emits green fluorescence when bound to DNA. Underfluorescence microscopy, viable cells show bright green nuclei with intact structure while apoptotic cells exhibit bright green nuclei withcondensed chromatin in dense green areas. Three to five randomly chosen fields were viewed in each slide with a minimum of 300 cells foreach group.

2.10. Analysis of mitochondrial membrane potential

Mitochondrial membrane potential (Δ ψm) was assessed by rhodamine 123, a mitochondrial potential sensor (Molecular Probes). Cells wereseeded in six-well plates for 6 h. The medium of each well was discarded and treated with 1 ml of medium containing rhodamine 123 at25 μM for 30 min at 37 °C and 5% CO2 in the dark. The cells were washed three times in PBS, then were analysed analyzed by flowcytometry.

2.11. Fas and FasL detection by flow cytometry

Sera-treated cells washed twice in PBS buffer were incubated with a rabbit anti-human Fas antibody or a rabbit anti-human FasL antibody at4 °C for 30 min. FITC-labeled goat anti-rabbit IgG antibody was then added to cells for 30 min at 4 °C. Flow cytometric analysis was performedusing a Becton Dickenson FACSort and CellQuest software. A total of 10,000 events were acquired for each analysis.

2.12. Detection of FasL, Bax, Bcl-2 mRNA in MGC-803 cells by RT-PCR

After rabbit sera treatment for 24 h, MGC-803 cell RNA was extracted with TRIzol Reagent, and used for reverse transcription RT-PCR. For PCRamplification, specific primers used were as follows: FasL (299 bp): 5′-GGAGCTGGCAGAACTCCGAGAG-3′ and 5′GGCTCAGGGGCAGGTTGTTG-3′; Bax (312 bp): 5′-GCGTCCACCAAGAAGCTGA-3′ and 5′-ACCACCCTGGTCTTGGATCC-3′; Bcl-2 (231 bp): 5′-CAGCTGCACCTGACGCCCTT-3′ and5′-GCCTCCGTTATCCTGGATCC-3′; β-actin primers: 5′-CGTCTGGACCTGGCTGGCCGGGACC-3′ and 5′-CTAGA AGCATTTGCGGTGGACGATG-3′(600 bp). The PCR conditions for FasL were: denaturation at 94 °C for 50 s, annealing at 63 °C for 50 s, and extension at 72 °C for 1 min,40 cycles. For Bcl-2 and Bax: denaturation at 94 °C for 50 s, annealing at 62 °C for 50 s, and extension at 72 °C for 1 min, 35 cycles. The PCRproducts were separated by electrophoresis on 2% agarose gels which were stained with ethidium bromide.

2.13. Statistical analysis

All in vitro experiments were performed for at least 3 times and the mean±SD of representative results are presented. Significant differencesamong the groups were determined using two way analysis of variance. Comparison of the rate of the patient survival was analyzed by Kaplan–Meier survival analysis method. Pb0.05 was considered as an indication of statistical significance.

Table 2 Clinical characteristics of patients with stage IV gastric cancer

Chemotherapy Chemotherapy+YWKLF

Total no. of patients 46 77Sex: male/female 35/11 59/18Histological typeAdenocarcinoma 32 53Mucinous adenocarcinoma 12 20Adenosquamous carcinoma 2 4

Disease statusRelapse after surgery 8 20Bypass surgery 14 21Laparotomy 12 17No surgery 12 19

Figure 2 Inhibition ofMGC-803 cell proliferation in vitro by serafrom formula-treated rabbits. MGC-803 cells were treated withsera from rabbits administered with YWKLF, Sub-F1, Sub-F2, andSub-F3 for 48 h. Cell growth was determined by MTT. A. The effectof rabbit sera at different concentrations in cell culture. B.Inhibition of cell growth in the presence of sera (20%), 1–2 h afterthe last formula administration in rabbits. ⁎⁎Pb0.01, comparedwith the effect of serum from rabbits given saline (0% inhibition).

1200 J. Li et al.

3. Results

3.1. The effect of combination therapy of YWKLF withchemotherapy on the survival of stage IV gastric cancerpatients

77 patients were assigned to treatment with chemotherapy plusYWKLF and 46 patients were assigned to chemotherapy alone.The background factors of patients as Table 2. There is nostatistically significant difference of the background in thepatients between chemotherapy alone and chemotherapy withYWKLF cohorts in terms of age, sex and tumor histology. The rateof response (RR) as reduction by at least 50% in the size ofprimary tumor in 46 patients receiving chemotherapy alone was13.04% (6/46), and 33.77% (26/77) of patients receivingchemotherapy and YWKLF showed responses. The mediansurvival time was 14 months for chemotherapy alone and21.5 months for YWKLF+chemotherapy patients (Pb0.001)(Fig. 1). These results indicate the advantage of combinationof chemotherapy with YWKLF in prolonging the survival of stageIV gastric cancer patients over chemotherapy alone.

3.2. The effect of YWKLF rabbit serum on MGC-803 cellviability

We next examined the effect of YWKLF on the biologicalbehavior of human gastric cancer cells. Pervious studies ofserum pharmacology suggested an eight to eighteen times ofdrug doses should be needed to administer into rabbits ascompared to humans. Based on our clinical data and preliminaryexperiments, we chose a rabbit administration dose thatequaled to 10 times of doses given to patients. Administrationwith herbal mixtures did not cause the loss of body weight,decreased activity, and change in fur quality of the rabbits (datanot shown). By using “serum pharmacology”, we harvest serafrom rabbits administered with YWKLF and sub-formulae and

Figure 1 Survival of stage IV gastric cancer patients withchemotherapy alone or in combination with YWKLF. Kaplan–Meieranalysis of survival of patients treated with chemotherapy orchemotherapy+YWKLF. Themedian survival timewas 14months forchemotherapy group and 21.5 months for chemotherapy+YWKLFgroup, respectively. Pb0.001, significantly increased survival ascompared with patients receiving chemotherapy alone.

tested their effects on the growth of the MGC-803 human gastriccells in vitro. As shown in Fig. 2A, after treatment for 48 h, withserum from rabbits received YWKLF, MGC-803 cells showedsignificantly reduced rate of growth. The maximal cell growthinhibition was observed when serum from YWKLF rabbits wasused at 20% concentration in the cell culture with slightlydecreased effect at 30% concentration reflecting “plateaued”biological responses seen in many drug experiments. The resultsalso suggest that serum harvested 2 h after last gavageadministration of YWKLF was optimal in inhibiting cell growth.When sera from rabbits administered with sub-formulae weretested, Sub-F1 and Sub-F3 sera showed significant inhibition ofcell growth, although the efficacy was lower than the intactformula serum. In contrast, Sub-F2 serum showed littleinhibitory activity on cell growth. Thus, components in bothSub-F1 and F3 contributed to the overall activity of YWKLF in“serum pharmacology” test. It should be noted that YWKLF andits sub-formulae did not affect the viability of normal gastriccells in primary culture or endothelial cells (data not shown).

3.3. YWKLFserum inhibitedAgNORsactivity inMGC-803cells

The levels of argyrophilic nucleolar organizer regions (AgNORs)represent cell cycle kinetics and the shorter the cell cycle, thegreater rate of the synthesis of rRNA, therefore, the increasedquantity of Ag NORs present in the nucleoli. Thus, AgNOR valuewas used to measure the rate of cell proliferation. Previous

1201YWKLF prolonged the overall survival time of stage late gastric cancer patients

studies showed a linear increase in both the area and the numberof AgNOR stainings in each tumor cell with increasing degree ofmalignancy [18]. We found AgNORs as distinct black–brown dotson a yellow pale background (Fig. 3A), and the black–brown dotsin the YWKLF serum treatment cell group were smaller than cellstreated with serum from saline administered rabbits (Fig. 3B).The ratio of AgNORs area to nucleolin/nucleus area per cell (I/S)is significantly decreased in cells treated by YWKLF serum andsaline serum containing cisplatin (⁎Pb0.05). Sub-F2 and Sub-F3sera did not affect I/S in MGC-803 cells, while Sub-F1 serumdecreased I/S in MGC-803 cells, but with the efficacy lower thanYWKLF serum (Fig. 3C). Thus, YWKLF rabbit serum inhibitedMGC-803 cell proliferation as shown by reduced AgNORs activityin the nuclei.

3.4. YWKLF rabbit serum induced MGC-803 cell apoptosis

We evaluated sub-diploid DNA content (sub-G1) in MGC-803 cellsusing flow cytometry. As shown in Fig. 4A, YWKLF rabbit serumtreatment resulted in an increase in the sub-G1 phase of MGC-803cells after 24 h (19.58±2.1% versus 2.66±1.2% cells with salinerabbit serum). We did not observe significant effect on sub-G1peak by Sub-F2 and Sub-F3 sera (Fig. 4A). However the sub-G1peak in cells treated by Sub-F1 rabbit serum was 12.36±2.3%,lower than YWKLF serum, but the effect was statistically

Figure 3 Argyrophilic nucleolar organizer regions (AgNORs) in MGCrabbits given saline. B. Cells cultured in the presence of 20% serumdefines I/S%. C. The ratio of AgNORs area to nucleolin/nucleus areamicroscope (×400). ⁎Pb0.05 compared with cells cultured with seru

significant. The number of cells in the G2/M phase wassignificantly decreased after treatment with YWKLF rabbitserum for 24 h with an increase in S-phase cell number. Theseresults indicate MGC-803 cell cycle arrest at S phase aftertreatment with YWKLF rabbit serum (Fig. 4B). The chemotherapyagent cisplatin has been reported to increased the fraction of S-phase cells in prostate cancer cells [19] and induced S-phase arrestof human ovarian carcinoma cells [20]. In this study, Fig. 4A showscisplatin (2 μg/ml) in saline rabbit serum induced S-phase arrest ofMGC-803 cells after 24 h treatment, with an increase in cells inapoptotic sub-G1 phase (25.86±3.1%). These results that indicateYWKLF rabbit serum was able to prevent gastric cancer celldivision.

We then evaluated MGC-803 cell apoptosis by using annexinV-FITC and PI staining. Fig. 4C showed that MGC-803 cellstreated with saline rabbit serum remained Annexin V−/PI−.YWKLF rabbit serum treated cells increased Annexin V+/PI−

staining. In contrast, cispaltin in saline rabbit serum inducedAnnexin V+/PI+ cells. Cisplatin caused both the early and latestage apoptosis in gastric cancer cells, while YWKLF rabbit serumpreferentially induced early stage cancer cell apoptosis. Sub-F2and Sub-F3 rabbit sera did not increase Annexin V+/PI− andAnnexin V+/PI+ cells (Fig. 4C).

We further determined morphological changes in MGC-803cells by confocal scanning laser microscopy. Fig. 5A-1 shows the

-803 cells. A. Cells cultured in the presence of 20% serum fromfrom rabbits given YWKLF. X axis indicates cell numbers, Y axisper cell. A total of 150 random cells were counted under a lightm from rabbits given saline.

1202 J. Li et al.

change in gastric cancer cell morphology after exposure to saline-treated rabbit serum, which failed to induce apoptosis. Mean-while, MGC-803 cells showed features of apoptosis, including cellshrinkage, nuclear condensation, DNA fragmentation, break downof the cells and formation of apoptotic bodies, after 24 h exposureto YWKLF rabbit serum and Sub-F1 rabbit serum (Fig. 5B). AfterYWKLF rabbit serum treatment, the nuclear structure of MGC-803cells became incomplete (Fig. 5C), supporting the notion thatYWKLF rabbit serum induced cell apoptosis.

3.5. YWKLF rabbit serum inducedmitochondrial dysfunctionin tumor cells

Mitochondria play a critical role in apoptosis induced by cancertherapeutic agents [21,22], which cause mitochondrial mem-

Figure 4 Induction of MGC-803 cell apoptosis by sera from rabdistribution of cultured MGC-803 cells with PI staining following 24 h ecells with sub-G1 phase DNA content, suggestive of apoptotic cell deaThe values shown are the mean±SEM of 3 experiments. ⁎⁎Pb0.01 cDetection of apoptosis with Annexin V-FITC/PI binding to MGC-803 capoptotic cells (Annexin V+/PI−) in the lower right, post apoptotic sprimary necrotic cells (Annexin V−/PI+) in the upper left quadrants,

brane potential (Δ ψm) loss and cytochrome c release into thecytosol, inducing caspase-9-dependent activation of caspase-3and cleavage of DNA reparatory protein PARP. We examined theeffect of YWKLF rabbit serum on Δ ψm using rhodamine 123, amitochondrial potential sensor. A marked loss of Δ ψm wasobserved after exposure of MGC-803 cells to YWKLF rabbit serumfor 4 to 24 h. The Sub-F2 and Sub-F3 rabbit sera did not affectcell Δ ψm, whereas, Sub-F1 rabbit serum caused Δ ψm loss incancer cells at the levels similar to YWKLF rabbit serum (Fig. 6).

3.6. Fas/FasL expression in MGC-803 cells

The Fas–FasL interaction is a key physiological regulator ofprogrammed cell death [23]. We investigated the expression ofFas/FasL in MGC-803 cells by flow cytometry and found that

bits given formulae. A. Flow cytometry analysis of cell cyclexposure of cells to rabbit sera. Arrows indicate the population ofth. B. Distribution of MGC-803 cell cycle indicated by PI staining.ompared with the effect of serum from rabbits given saline. C.ells. Viable cells (Annexin V−/PI−) are located in the lower left,econdary necrotic cells (Annexin V+/PI+) in the upper right andrespectively. Numbers in each quadrant are percentage of cells.

Figure 5 Changes in nuclear shape in MGC-803 cells. A. The change of gastric cancer cell morphology after exposure to sera in thefluorescent microscope: 1, saline serum treated cells; 2, YWKLF serum treated cells; B. YWKLF serum treated cells: 1, green fluorescenceindicating DNA; 2, red fluorescence indicating RNA; 3, superimposed DNA and RNA fluorescence. Sub-F1 serum treated cells: 4, DNA ingreen fluorescence; 5, RNA in red fluorescence; 6, superimposed DNA and RNA fluorescence. C. Changes in the fluorescence intensity ofnuclei: saline serum treated cells: 1, cell with normal morphology; 2, topographic map of nuclear DNA fluorescence distribution. YWKLFserum treated cells: 3, DNA fragmentation and formation of apoptotic bodies. 4, incomplete nuclear structure.

1203YWKLF prolonged the overall survival time of stage late gastric cancer patients

Figure 6 Evaluation of mitochondrial membrane potential(Δ ψm) with rhodamine 123. MGC-803 cells incubated with rabbitsera were treated with rhodamine 123 and analyzed by flowcytometry at the indicated time points. Fluorescence intensity (FI)indicates membrane potential of mitochondria in the cells.

Figure 8 Changes in the expression of FasL, Bax and Bcl-2mRNAin MGC-803 cells. The cells were cultured for 24 h with sera fromformula formula-treated rabbits. Relative levels of Fas-L, Bax andBcl-2 mRNA were determined using RT-PCR. Lane 1, saline rabbitserum, Lane 2, Sub-F1 rabbit serum; Lane3, Sub-F2 rabbit serum;Lane 4, Sub-F3 rabbit serum; Lane 5, YWKLF rabbit serum.

1204 J. Li et al.

treatment of the cells by YWKLF rabbit serum and Sub-F1 rabbitserum increased Fas expression by two fold (Fig. 7). The FasLexpression on MGC-803 was low and was not significantlychanged by any treatment (Fig. 7).

Since recent studies have suggested that resistance toapoptosis is associated with the loss of Fas function in severalmalignancies [24–27], up-regulating Fas expression on MGC-803cells by YWKLF rabbit serum may increase cell sensitivity to FasLkilling.

3.7. FasL and Bax/Bcl-2 mRNA in MGC-803 cells

RT-PCR showed that the level of FasL mRNA in MGC-803 declinedsignificantly after treatment by YWKLF rabbit serum for 24 h.However, the level of Bax mRNAwas markedly increased (Fig. 8).There was no quantitative difference in Bcl-2 mRNA levels in

Figure 7 FACS analysis of Fas and Fas-L, expression on MGC-803cells. Tumor cells treated with sera from rabbits given formulaewere stained with a rabbit anti-human Fas antibody or a rabbitanti-human FasL antibody followed by staining with a FITC-conjugated secondary goat anti-rabbit IgG antibody. ⁎Pb0.05compared with cells treated with serum from rabbits given saline.

cells after rabbit sera treatment (Fig. 8). Sub-F1 rabbit serumalso markedly increased the level of Bax mRNA in MGC-803 cells.

4. Discussion

Alternative medicine has been considered as a complemen-tary therapy for late stage gastric cancer for a number ofyears in Southeast Asia because the lack of effective meansto prolong the life span of patients with this deadly tumortype [28–31]. In this study, we provided evidence thatcombination of chemotherapy with herbal medicine formulaYWKLF prolonged the overall survival time of stage IV gastriccancer patients. Although the therapeutic regime stillrequires multi-centered clinical evaluation with lagerpatient cohorts, YWKLF may represent a basis on whichspecific herb components possessing anti-cancer effectcould be identified by further screening.

Medicinal herbs constitute challenges to clinicians andbasic researchers for their complexity and the potential ofactive components generated after metabolism in vivo. It isnow realized that adding herb extracts to cancer cell culturemay not be able to better evaluate the pharmacologicalactivity of the herbs. Therefore, “serum pharmacology” hasrecently been proposed and utilized in Southeast Asian, inwhich sera collected from small animals administered withherb extracts would retain the major components of theextracts and also potential metabolites of the components[10]. Based on the “serum pharmacology ” principle, in thisstudy, we used sera from rabbits administered YWKLF and itssub-formulae to show that YWKLF rabbit serum significantlyinhibited the growth of human gastric cancer cells in vitro bytriggering the apoptosis signaling pathways. These findingswere supportedby the induction of tumor cell cycle arrest, lossof mitochondrial membrane potential and increased expres-sion of Fas and Bax in tumor cells. Previous studies indicatethat gastric carcinomas express Fas ligand but low level Fas toescape killing from host immune system. Clinical data alsoshowed that apoptosis in gastric carcinoma may be anindependent indicator of better prognosis [32]. This apoptosis

Table 3 Major components contained in YWKLF with known biological activities

Herb Known components Biological activity Reference

Astragalusmongholicus Bge

Polysaccharides Stimulation of cell-mediated immunity;improvement of cardiovascular and neuroendocrinesystems; anti-oxidation; inhibition of tumorgrowth; anti-inflammation

[11,37,41]Stragaloside I–IVIsoflavonoids 1–5Saponinsγ-aminobutyric acid

Panax ginsengC. A. Mey

More than forty kinds of saponin(ginsenosides Rg3, Rh2, Rb1, Rb2, etc)

Anti-allergy; anti-tumor; improvement of learningand memory anti-aging

[11,33,38]

Panax notoginseng(Burk.) F. H.Chen

Ginsenosides Rb1, Rg1, Rd; NotoginsenosideR1,Rb1, Re, Rg1 and Rh1; Dencichine,flavonoid and polysaccharide

Anti-thrombosis; anti-atherosclerosis; anti-inflammation; liver and kidney protection;immunological adjuvant activity andimmunostimulatory action; anti-tumor; chemo-preventive effect

[12,13,39]

Caesalpinia sappanL

Triterpenoids Anti-oxidation; anticonvulsant; anti-complement;anti-inflammation

[14]FlavonoidsPhenolicsSteroidsBrazilin

Oldenlandiadiffusa (Willd.)Roxb.

Ursolic acid Anti-cancer; improvement of immune function;anti-oxidation

[15,36,40]Oleanolic acidIridoid glucosides flavone

Paris polyphyllaSmith var.chinensis(French) Hare

Steroidal saponins Anti-cancer; analgesic; support immune function [34,35]DiosgeninPennogenin

1205YWKLF prolonged the overall survival time of stage late gastric cancer patients

is associated with tumor Fas expression. Thus up-regulatingFas expression on MGC-803 cells by YWKLF rabbit serum mayincrease gastric cancer cell sensitivity to apoptotic killing byFasL positive cells (such as T lymphocytes).

YWKLF is composed of six herbs that include 3 medicinalprinciples, and on this basis, we divided the formula into 3 sub-formulae, and tested the effect of each sub-formula ongeneration of rabbit serum that may affect the biologicalbehaviour of tumor cells. We found Sub-F1 composed of herbsthat strengthen the host defense [11], also inhibited MGC-803cell proliferation and induced apoptosis, albeit with lowerefficacy as compared with the intact YWKLF. Therefore,combination of therapeutic principles represented bymultipleherbs may yield better results in cancer treatment.

In-depth studies are needed to clarity the componentscontained in rabbit sera that induce the apoptosis of gastriccancer cells. The “serum pharmacology” presumes that someherb components will be retained in sera in their original ormetabolized form [9]. In fact, some components of theYWKLF and metabolites have been showed to be capable ofinducing apoptosis of cancer cells [33–36] (Table 3). Forexample, Astragalus saponins induce growth inhibition andapoptosis in human colon cancer cells and tumor xenograft[37]. An intestinal bacterial metabolite of Ginseng proto-panaxadiol saponins induces apoptosis of human hepatocel-lular carcinoma SMMC7721 cells [38]. Ginsenoside Rd from P.notoginseng is cytotoxic to HeLa cells and induces theirapoptosis [39]. The aqueous extract of O. diffusa was foundto be directly toxic to the HL-60 myeloid leukemia cells,inducing cell apoptosis [40].

Despite abundance of evidence for the capacity of herbalcomponents to inhibit tumor cell growth, it remains to beelucidated whether in “serum pharmacology” the herbal

formula might induce host factors such as TNF a or othercytostatic molecules that could be “indirectly” suppress cancercells [41]. Also in human, it is unknown whether herbal formulaacts directly on tumor cells through absorption by gastro-intestinal tract or indirectly throughmobilizing the host defensesystem or both. Our present study calls for more thoroughinvestigation into the mechanisms of herbal medicine and itspotential in the treatment of cancer and other human diseases.

Acknowledgments

This project was supported by grants No. 30600839 and No.30672764 from National Natural Science Foundation ofChina. The authors thank Dr. Ji Ming Wang of NCI-Frederick,NIH, USA, for critical reviewing the manuscript.

References

[1] Yang L. Incidence and mortality of gastric cancer in China.World J Gastroenterol 2006;12:17–20.

[2] Xu YQ, Xue HN, Zhu XQ, Chen XD. Clinical observation ontreatment for postoperative metastasis of gastric cancer byJianpi Huoxue Jiedu prescription. Zhong Xi Yi Jie He Xue Bao2003;1:192–4.

[3] Lu WP, Sun GZ, Piao BK. Clinical and experimental research ingastric cancer treatment by “Yangwei kang liu” granules. J TradChin Med 1996;37:350–2.

[4] Letai AG. Diagnosing and exploiting cancer's addiction to blocksin apoptosis. Nat Rev Cancer 2008;8:121–32.

[5] Fulda S, Debatin KM. Extrinsic versus intrinsic apoptosis pathwaysin anticancer chemotherapy. Oncogene 2006;25:4798–811.

[6] Curtin JF, Cotter TG. Live and let die: regulatory mechanisms inFas-mediated apoptosis. Cell Signal 2003;15:983–92.

1206 J. Li et al.

[7] Danial NN, Korsmeyer SJ. Cell death: critical control points.Cell 2004;116:205–19.

[8] Li J, Sun GZ, Qi X. Effect of “YangWeiKangLiu” granules onperipheral blood T-lymphocyte rDNA transcription activity andFas expression in gastric cancer metastasis. Chin J Surg Integr2002;8:253–5.

[9] Tashino S. “Serum pharmacology” and “serum pharmaceuticalchemistry”: from pharmacology of Chinese traditional medi-cines to start a new measurement of drug concentration inblood. Ther Drug Monit Res 1988;5:54–64.

[10] Bochu W, Liancai Z, Qi C. Primary study on the application ofSerum Pharmacology in Chinese traditional medicine. ColloidsSurf B Biointerfaces 2005;43:194–7.

[11] BlockKI, MeadMN. Immune systemeffects of echinacea, ginseng,and Astragalus: a review. Integr Cancer Ther 2003;2:247–67.

[12] Li WL, Zheng HC, Bukuru J, De Kimpe N. Natural medicines usedin the traditional Chinese medical system for therapy ofdiabetes mellitus. J Ethnopharmacol 2004;92:1-21.

[13] Ng TB. Pharmacological activity of sanchi ginseng (Panaxnotoginseng). J Pharm Pharmacol 2006;58:1007–19.

[14] Ye M, Xie WD, Lei F, Meng Z, Zhao YN, Su H, et al. Brazilein, animportant immunosuppressive component from Caesalpiniasappan L. Int Immunopharmacol 2006;6:426–32.

[15] Kim DH, Lee HJ, Oh YJ, Kim MJ, Kim SH, Jeong TS, et al. Iridoidglycosides isolated fromOldenlandia diffusa inhibit LDL-oxidation.Arch Pharm Res 2005;28:1156–60.

[16] Li YK. Medicine serum pharmacology test methods a number ofissues. Traditional Chinese Medicine Research & ClinicalPharmacology, vol. 10; 1999. p. 95–7.

[17] Sirri V, Roussel P, Hernandez-Verdun D. The AgNOR proteins:qualitative and quantitative changes during the cell cycle.Micron 2000;31:121–6.

[18] K. Vacharadze, G. Burkadze, G. Turashvili, N. Kiria (2005)Argyrophilic nucleolar organizer regions in benign and malignantmesothelial lesions. Georgian Med News, 91–3.

[19] Qiao L, Pizzolo G, Melamed MR. Effects of selected chemother-apeutic agents on PCNA expression in prostate carcinoma celllines. Urol Res 1994;22:171–6.

[20] Nguyen HN, Sevin BU, Averette HE, Perras J, Ramos R, DonatoD, et al. Cell cycle perturbations of platinum derivatives on twoovarian cancer cell lines. Cancer Investig 1993;11:264–75.

[21] Hoye AT, Davoren JE, Wipf P, Fink MP, Kagan VE. Targetingmitochondria. Acc Chem Res 2008;41:87–97.

[22] Lalier L, Cartron PF, Juin P, Nedelkina S, Manon S, Bechinger B,et al. Bax activation and mitochondrial insertion duringapoptosis. Apoptosis 2007;12:887–96.

[23] Konopleva M, Zhao S, Xie Z, Segall H, Younes A, Claxton DF, et al.Apoptosis. Molecules and mechanisms. Adv Exp Med Biol1999;457:217–36.

[24] Wohlfart S, Sebinger D, Gruber P, Buch J, Polgar D, Krupitza G,et al. FAS (CD95) mutations are rare in gastric MALT lymphomabut occur more frequently in primary gastric diffuse large B-celllymphoma. Am J Pathol 2004;164:1081–9.

[25] Radfar S, Davrinche C, Hollande E. Serial in vivo loss and in vitrogain of Fas expression and function in human cancerouspancreatic duct cells. Int J Cancer 2005;115:214–23.

[26] von Reyher U, Strater J, Kittstein W, Gschwendt M, KrammerPH, Moller P. Colon carcinoma cells use different mechanisms toescape CD95-mediated apoptosis. Cancer Res 1998;58:526–34.

[27] Sikora J, Dworacki G, Zeromski J. Expression of Fas and Fas ligandandapoptosis in tumor-associated lymphocytes and in tumor cellsfrom malignant pleural effusions. Nat Immun 1998;16:244–55.

[28] Glimelius B, Ekstrom K, Hoffman K, Graf W, Sjoden PO, HaglundU, et al. Randomized comparison between chemotherapy plusbest supportive care with best supportive care in advancedgastric cancer. Ann Oncol 1997;8:163–8.

[29] Kodera Y, Yamamura Y, Torii A, Uesaka K, Hirai T, Yasui K, et al.Postoperative staging of gastric carcinoma. A comparisonbetween the UICC stage classification and the 12th edition ofthe Japanese General Rules for Gastric Cancer Study. ScandJ Gastroenterol 1996;31:476–80.

[30] Hundahl SA, Menck HR, Mansour EG, Winchester DP. TheNational Cancer Data Base report on gastric carcinoma. Cancer1997;80:2333–41.

[31] Kim SG, Park EC, Park JH, Hahm MI, Lim JH, Choi KS. Initiationand discontinuation of complementary therapy among cancerpatients. J Clin Oncol 2007;25:5267–74.

[32] Ohno S, Tachibana M, Shibakita M, Dhar DK, Yoshimura H,Kinugasa S, et al. Prognostic significance of Fas and Fas ligandsystem-associated apoptosis in gastric cancer. Ann Surg Oncol2000;7:750–7.

[33] Hofseth LJ, Wargovich MJ. Inflammation, cancer, and targets ofginseng. J Nutr 2007;137:183S–5S.

[34] Sun J, Liu BR, Hu WJ, Yu LX, Qian XP. In vitro anticancer activityof aqueous extracts and ethanol extracts of fifteen traditionalChinese medicines on human digestive tumor cell lines.Phytother Res 2007;21:1102–4.

[35] Zhao Y, Kang LP, Liu YX, Zhao Y, Xiong CQ, Ma BP, et al. Threenew steroidal saponins from the rhizome of Paris polyphylla.Magn Reson Chem 2007;45:739–44.

[36] Liang Z, He M, Fong W, Jiang Z, Zhao Z. A comparable, chemicaland pharmacological analysis of the traditional Chinesemedicinal herbs Oldenlandia diffusa and O. corymbosa and anew valuation of their biological potential. Phytomedicine2008;15:259–67.

[37] Tin MM, Cho CH, Chan K, James AE, Ko JK. Astragalus saponinsinduce growth inhibition and apoptosis in human colon cancercells and tumor xenograft. Carcinogenesis 2007;28:1347–55.

[38] Ming YL, Song G, Chen LH, Zheng ZZ, Chen ZY, Ouyang GL, et al.Anti-proliferation and apoptosis induced by a novel intestinalmetabolite of ginseng saponin in human hepatocellularcarcinoma cells. Cell Biol Int 2007;31:1265–73.

[39] Yang ZG, Sun HX, Ye YP. Ginsenoside Rd from Panax notoginsengis cytotoxic towards HeLa cancer cells and induces apoptosis.Chem Biodivers 2006;3:187–97.

[40] Willimott S, Barker J, Jones LA, Opara EI. Apoptotic effect ofOldenlandia diffusa on the leukaemic cell line HL60 andhuman lymphocytes. J Ethnopharmacol 2007;114:290–9.

[41] Spelman K, Burns J, Nichols D, Winters N, Ottersberg S,Tenborg M. Modulation of cytokine expression by traditionalmedicines: a review of herbal immunomodulators. Altern MedRev 2006;11:128–50.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具