Journal of Ethnopharmacology 179 (2016) 128136

Contents lists available at ScienceDirect

Journal of Ethnopharmacology
journal homepage: www.elsevier.com/locate/jep

Anti-inuenza virus effects of crude phenylethanoid glycosides

isolated from ligustrum purpurascens via inducing endogenous
interferon-
Xiao-peng Hu a,b,1, Min-ming Shao c,1, Xun Song a,b,d,1, Xu-li Wu a,b, Ling Qi c, Kai Zheng a,b,
Long Fan a,b, Cheng-hui Liao a,b, Chen-yang Li a,b, Jiang He a,b, Ying-jie Hu c, Hai-qiang Wu a,b,
Shi-he Li a,b, Jian Zhang a,b,n, Feng-xue Zhang c,nn, Zhen-dan He a,b,n
a
Department of Pharmacology, School of Medicine, Shenzhen University, Shenzhen 518060, China
b
Institute of Biotherapy, Shenzhen University, Shenzhen 518060, China
c
Institute of Tropical Medicine, Guangzhou University of Chinese Medicine, Guangzhou 510405, China
d
School of Chinese Medicine, Hong Kong Baptist University, Hong Kong 999077, China

art ic l e i nf o a b s t r a c t

Article history: Ethnopharmacological relevance: Ligustrum purpurascens Y.C. Yang (Oleaceae) is traditionally recorded as
Received 12 March 2015 Ku Ding Cha, a kind of functional tea in southern China for about two thousand years, which has been
Received in revised form reported with sore throat alleviating and pathogenic heat expelling effects. However, there are no sci-
18 March 2015
entic studies demonstrating its antiviral activity.
Accepted 16 July 2015
The aim of the study: This study is aimed at investigating the anti-inuenza virus effects of pheny-
Available online 17 July 2015
lethanoid glycosides isolated from L. purpurascens (LPG) as well as its corresponding mechanisms.
Keywords: Materials and methods: In vitro, hemagglutination assay was employed to detect the inuenza virus titer;
Ligustrum purpurascens In vivo, C57BL/6J mice were given oral administration of LPG (100 mg/kg, 300 mg/kg, 900 mg/kg) or
Glycosides ribavirin (100 mg/kg) once daily for 5 successive days. Meanwhile, on the second day, mice were infected
Endogenous antiviral cytokine
intranasally (i.n.) with A/FM/1/47 H1N1 virus. Mice survival rate and other clinical index were monitored
Interferon-
for 15 days. Infected mice were sacriced to measure the lung lesion and stained with hematoxylin
Inuenza virus
eosin. Flow cytometry analyses spleen lymphocytes and interferon- (IFN-) level. The IFN- knockout
mice (IFN-
/
mice, C57BL/6J) which had been veried lacking IFN- through Western Blot, were
applied in the death-protection test to identify the role of IFN- played in LPG antiviral effect.
Results: In vitro, LPG at 0.5 mg/ml inhibited Inuenza A Virus H1N1 type (H1N1) infection of MDCK cells.
In vivo, LPG at 300 and 900 mg/kg signicantly decreased the mouse lung index (po 0.05), alleviated
inuenza-induced lethality and clinical symptoms, and therefore enhanced mouse survival (po 0.05).
More detailed experiments demonstrated that antiviral cytokine IFN- was involved in the antiviral
effect of LPG. Flow cytometric analysis revealed that LPG (900 mg/kg) signicantly induced secretion of
IFN- by splenic CD4 and CD8 cells (po0.05). Moreover, LPG (900 mg/kg) protected wild-type C57BL/
6J mice from H1N1 injury, whereas LPG-mediated survival protection disappeared in IFN-
/
mice.
Conclusion: These results suggest that up-regulating endogenous IFN- by LPG may represent a novel
therapeutic approach for H1N1 infection.
& 2016 Published by Elsevier Ireland Ltd.

Abbreviations: LPG, Ligustrum purpurascens phenylethanoid glycosides; IFN, Interferon; H1N1, Inuenza A Virus H1N1 type; MTT, Methyl Thiazolyl Tetrazolium; MEM,
Modied Eagle Medium; MDCK cell, Madin-Darby canine kidney cell; WT mice, wild-type mice; IFN-
/
mice, Interferon- knockout mice; IACUC, Institutional Animal Care
and Use Committee; WBC, white blood cell; RBC, red blood cell; PLT, platelet; Hb, hemoglobin; LD50, 50% lethal dosage; TC50, 50% toxic concentration; ANOVA, Analysis of
Variance; EDTA, Ethylene Diamine Tetraacetic Acid
n
Corresponding authors at: Department of Pharmacology, School of Medicine, Shenzhen University, Shenzhen 518060, China. Fax: 86 755 86671906.
nn
Corresponding author at: Institute of Tropical Medicine, Guangzhou University of Chinese Medicine, Guangzhou 510405, China. Fax: 86 2039358390.
E-mail addresses: jzhanghappy@szu.edu.cn (J. Zhang), zhangfengxue@gzucm.edu.cn (F.-x. Zhang), hezhendan@126.com (Z.-d. He).
1
Contributed equally to this work.

http://dx.doi.org/10.1016/j.jep.2015.07.019
0378-8741/& 2016 Published by Elsevier Ireland Ltd.
 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136
1. Introduction
Inuenza A Virus H1N1 type (H1N1) belongs to the orthomyx-
oviridae family, which exclusively attacks lung alveoli and causes
acute respiratory infectious diseases (Dai et al., 2014). Approxi-
mately 35 million people are infected and 250,000500,000
deaths yearly worldwide due to the inuenza epidemics (Shirey
et al., 2013). Antiviral medicine such as Ribavirin and vaccine has
been widely used clinically. However, the drug resistance and side
effect tend to limit the availability of that medicine (Mungall et al.,
2004; Nguyen et al., 2012). Thus, there is a critical need for a safe
and effective therapeutic strategy adjunct and alternative to in-
uenza vaccines and conventional antiviral agents.
Classical antiviral medicine, which targets on viral protein such
as hemagglutintin (HA) for inuenza virus, may face the obstacles
like the variation of subtypes (Ge et al., 2014). The innate immune
system availably enables us to ght infection. Several cytokines
acting on DNA or RNA viruses, such as interferons (IFNs) (Platanias,
2005; Rathinam and Fitzgerald, 2010), interleukins (Fakruddin
et al., 2007; Rathinam and Fitzgerald, 2010), and tumor necrosis
factors (Matikainen et al., 2006), have been reported with the
function of anti-virus. Thus, immunotherapy, such as inducing
endogenous antiviral cytokines, may become a new approach that
stimulates immune cells to enlarge the anti-virus activity. By tar-
geting these antiviral cytokines, myriad natural small molecules
have been screened. For instances, quercetin, baicalein and san-
guinarine, strengthen antiviral ability, which followed by cyto-
kines regulation (Johari et al., 2012; Kim et al., 2013; Nair et al.,
2002). However, the coherent relationships between antiviral
ability and up-regulating cytokines havenot been identied. Eri-
toran, a Toll Like Receptor 4 antagonist, protected PR8-infected
mice via inducing endogenous IFN- and IFN- (Shirey et al.,
2013). It gives us the inspiration that natural small molecules from
antiviral medicinal plants could be screened base on inducing
antiviral cytokines.
Ku Ding Cha, which has been used as a kind of functional tea in
southern China for about two thousand years, was rst docu-
mented in the Shen Nong's Herbal Classic. Its original materials
contain 30 species in 12 families, which belong to the genera Li-
gustrum (Oleaceae), Cratoxylum (Hypericaceae), Ehretia (Ehretia-
ceae), Ilex (Aquifoliaceae), and Photinia (Rosaceae) (He et al.,
1992). The leaves of Ligustrum purpurascens Y.C. Yang (Oleaceae)
was often used as a kind of famous original plant of Ku Ding Cha.
The research proved that the tea could enhance physical tness to
resist the diseases, and is claimed to be good for inammation (He
et al., 2010; Lau et al., 2002). According to historical record, its
leaves were used as common part of herbal tea in treating com-
mon cold (including u) and rhinitis, bacterial infection, paralleled
with the immune enhancement effects (Li et al., 2013). Besides, In
the ancient encyclopedia Compendium of Materia Medica writ-
ten by Li Shizhen in Ming Dynastic (Gao et al., 2006), Ku Ding Cha
has also been described as a functional tea, which could alleviate
pathogenic windheat and promote uid production to quench
thirst (He et al., 2003; Li et al., 2013). Apart from Ilex latifolia
Thunb, L. purpurascens Y. C. Yang is also another main original
plant of Ku Ding Cha (Li et al., 2013). Previous studies revealed that
it comprises phenylethanoid glycosides (Wong et al., 2001), which
have been considered as anti-u agents in some traditional herbal
medicines, such as Strobilanthes cusia (Tanaka et al., 2004) and
Forsythia suspense (Li et al., 2014; Zhou et al., 2014).
In previous study, we found that phenylethanoid glycosides
from L. purpurascens (LPG) had high solubility in water as well as
immunomodulatory effects (Song et al., 2012), which illustrate
potential antiviral ability.
In this study, we demonstrated the anti-inuenza virus effect of
LPG both in vitro and in vivo. LPG inhibited H1N1-induced
129
hemagglutination in vitro and alleviated inuenza-induced leth-
ality and clinical symptoms in vivo. Moreover, the role of IFN- in
the antiviral process of LPG was conrmed in IFN- knockout
mice. These endings indicate that modulating the antiviral func-
tion of cytokines, especially IFN-, is promising approaches in the
development of antiviral therapies to inuenza virus infection.
2. Materials and methods
2.1. Materials
2.1.1. Plant and drugs
L. purpurascens Y.C. Yang was purchased from Suijiang county,
Yunnan province, China, and was authenticated by Peng Hua (The
specimen voucher number: NPCR-Lipu-1, deposited with the her-
barium of Kunming Institute of Botany, the Chinese Academy of
Sciences). In addition, L. purpurascens Y.C. Yang (record kew-
354137) is the accepted name of a species in the genus Ligustrum
(Oleaceae) according to the latest revision in The Plant List
www.theplantlist.org. The extraction of LPG from the leaves was
previously described (Song et al., 2012). Powder was suspended in
ultra-pure-grade water. Ribavirin injection was purchased from
Hangzhou Minsheng Pharmaceutical Co., Ltd., Zhejiang, China
(Standard: 100 mg/injection, product batch number: 1406304).
2.1.2. Reagents and antibodies
ELISA kits were purchased from eBioscience (California, US).
The following antibodies were used: for surface staining, anti-CD3
(553067; BD), anti-CD4 (553046; BD), anti-CD8 (553035; BD); for
intracellular staining, anti-IFN- (554412; BD); for western blot
analysis, anti--actin (60008-1; Proteintech); anti-IFN- (513205;
Biolegend) and secondary horseradish peroxidase-conjugated an-
tibody (E030120-01; EarthOx). The lymphocyte separation liquid
was purchased from Dakewe (Shenzhen, China).
2.1.3. Animals and cells
Specic pathogen-free C57BL/6J mice of either sex (1620 g),
were purchased from the Animal Supply Center of Guangdong
Academy of Medical Science; IFN-
/
mice on a C57BL/6 back-
ground, were purchased from the Model Animal Research Center
of Nanjing University (Model No. J002287). All animals were kept
in an environmentally controlled breeding room (temperature:
257 1 C, humidity 55 75%, and 12/12 h light/dark cycle) for at
least 1-week acclimatization before experiments. All animal ex-
periments were conducted with institutional IACUC approval from
Shenzhen University School of Medicine. MDCK cells were kindly
donated by the Institute of Tropical Medicine, Guangzhou Uni-
versity of Chinese Medicine.
2.1.4. Virus
Mouse-adapted inuenza virus (A/FM/1/47 H1N1, FM1) was
kindly donated by the Institute of Tropical Medicine, Guangzhou
University of Chinese Medicine. The virus was amplied in allan-
toic cavity of embryonated eggs for 48 h at 36 C, and then stored
at
80 C. All tests were performed in class II biosafety safety
cabinets (Dai et al., 2014).
2.2. Methods
2.2.1. Toxicity measurements
Toxicity of LPG in MDCK cells was measured by a standard
method. Briey, MDCK cells were grown to conuence in 96-well
tissue culture plates at 2  105 cells/well and treated with LPG at
serial dilution from 20 to 0.1 mg/ml. Four days after the nal ad-
dition of compound, culture media was removed and LPG was
130 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136
added for an additional 4 days. MTT solution (10 ml/100 ml med-
ium) was added to all wells, and incubation continued at 37 C for
4 h, followed by the addition of acid-isopropanol (100 ml of 0.04 N
HCl in isopropanol) to dissolve the dark blue crystals developed
from MTT.
2.2.2. Virus challenge and treatments in vitro
An amount of 100 l MEM with approximately 1  105 MDCK
cells was added to each well under 37 C, 5% CO2. After cells had
grown in monolayers, serially diluted LPG was added to MDCK
cells, 100 l per well, for 24 h. Then the suspension of each well
was thrown away, adding serially diluted virus solution for 2 h.
The medium were then removed and the MEM containing serially
diluted LPG was added again. For pre-treatment assay, cells were
pre-treated with LPG for 24 h; for after-treatment assay, cells were
after-treated with LPG for 24 h; for incubated-treatment assay,
H1N1 incubated with LPG rst on ice for 1 h, then the mixture was
added to cells for 2 h. Four days later, chicken erythrocyte agglu-
tination assay was used to evaluate LPG antiviral function. The
assay was performed by using chicken red blood cells with stan-
dard procedures (Ge et al., 2014).
2.2.3. Virus challenge and treatments in vivo
Mice were anesthetized with diethylether. WT (wild-type) or
IFN-
/
mice were infected with mouse-adapted H1N1 virus
(FM1; 5LD50 intranasal, 25 ml/nare). Infected mice were divided
into 4 groups for treatment (n 18: 10 mice for survival study and
8 mice for the remaining tests): LPG (100, 300 and 900 mg/kg) and
positive drug (ribavirin, 100 mg/kg). The mice were orally or in-
traperitoneally administrated respectively. The remaining 2 groups
(control and infected mice) received equivalent amounts of sterile
water. Some mice were sacriced at the indicated time of post-
infection and the blood and spleen were collected separately for
the blood routine test and IFN- detection.
2.2.4. Histopathology
Lungs were inated and perfused and xed with 4% paraf-
ormaldehyde. Fixed sections (8 mm) of parafn-embedded lungs
were stained with hematoxylin and eosin. Slides were rando-
mized, read blindly, and examined for tissue damage necrosis,
apoptosis, and inammatory cellular inltration (Dai et al., 2014).
2.2.5. Blood routine examination
After mice were sacriced, blood was collected by removal of
the eyeball. The percentage of white blood cells (WBC), red blood
cells (RBC), and platelets (PLT) and hemoglobin (Hb) level was
measured. Blood cell analyzer (BC-3000 Plus, Mindray) was used
for tests (Gadad et al., 2010).
2.2.6. Lung index
After administrating orally or intraperitoneally, the lung index
was used as an indicator of pulmonary edema, according to the
following equation: lung index wet weight (g)/body weight (g) 
100% (Dai et al., 2014).
2.2.7. Preparation of lymphocytes
After being removed from sacriced mice, lungs and spleens
were ground and disrupted mechanically in 5 ml 1X Mouse Lym-
phocyte Separation Medium (DKW33-R0100, EZ-SepTM, Dakewe,
Shenzhen, China) (Tang et al., 2014). Finally, T lymphocytes were
diluted to per milliliter-containing 5  106 cells and resuspended
in RPMI 1640 medium for the following experiments.
2.2.8. Western blot analysis
Whole-cell extracts were prepared by lysing 5  106 cells as
reported previously (blood-2010). Total protein concentration in
the supernatant was determined by bicinchoninic acid assay (Be-
yotime biotechnology, China). The next steps were previously
described (Chen and Chang, 2013).
2.2.9. Flow cytometry and cytokine detection
Cells were analyzed by ow cytometry after culture from 7 to
8 days. For surface staining, cells were stained with the respective
antibodies for 20 min in phosphate buffered saline containing 0.5%
fetal calf serum and 2 mM EDTA before analysis. For intracellular
staining, cells were stimulated for 45 h with phosphomolybdic
acid (50 ng/ml) and ionomycin (250 ng/ml) in the presence of
GolgiPlug (BD Biosciences), and xed and permeabilized (Fix/
Perm; BD) according to the manufacturer's instructions. Also,
monensin (8 g/ml) was employed to block cytokines translating
outside the cell, and then cells were stained with respective an-
tibodies for intracellular cytokine detection for 3045 min (Klei-
newietfeld et al., 2013). All data were acquired by use of BD FACS
Calibur and analyzed with FlowJo software (TreeStar).
2.2.10. Statistical analyses
Data are expressed as mean 7S.E.M. Statistical differences be-
tween 2 groups were determined by Student t test. For multiple
groups, one-way ANOVA analysis was used to compare means.
Analysis involved use of SAS 9.1 (SAS Institute Inc., Cary, NC).
po 0.05 was considered statistically signicant. For survival stu-
dies, a log-rank (MantelCox) test involved use of GraphPad Prism
(GraphPad 5.0 Software).
3. Results
3.1. LPG inhibited H1N1-induced hemagglutination in vitro
Chicken erythrocyte agglutination test was used to evaluate the
antiviral effect of LPG (TC50 0.89 mg/ml). As shown in Fig. 1A,
pre-treatment of LPG (0.5 mg/ml) signicantly decreased the virus
titer, while no signicant changes were observed in after-treated
with LPG, and H1N1 incubated with LPG (Fig. 1B and C).
3.2. Acute toxicity study in vivo
The mice were administrated orally with LPG which maximal
concentration up to 34 g/kg showed no changes in behaviour or
body weight. Besides, there were no signicant pathological
changes on anatomizing and viscerating organs such as lung, liver,
kidney and heart. Moreover, there were no deaths during 14 days
of observation. Thus, LPG was safe at any level on and below 34 g/
kg.
3.3. LPG alleviates clinical features and reduces mortality in H1N1
infected mice
Both the control and ribavirin-treated mice showed no mor-
tality (Fig. 2). Infected mice showed typical symptoms of inuenza
such as being inactive, rufed fur, respiratory distress and lack of
appetite. Furthermore, death occurred at day 6 post-infection (10/
10 mice), so the animal model of H1N1 infection was successfully
established. The mortality with LPG or ribavirin treatment was
reduced (p o0.05), thus prolonging the life span during the 15
days before mice were sacriced. Meanwhile, infected mice were
relieved from the symptoms of inuenza throughout the LPG
treatment period as compared with infection alone.
3.4. Effect of LPG on arterial blood analysis
Infected mice showed decreased WBC count as compared with
 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136 131

Fig.2. The percent survival of ribavirin and LPG compared with those treated with
sterile water. n 10 for each group. A log-rank (MantelCox) test was used for the
statistical analysis. ##p o 0.05, compared with the control group; **p o 0.05, com-
pared with the infected group.

Table 1
Effect of LPG on arterial blood routine analysis.

Group WBC RBC Hb PLT

(  109/L) (  1012/L) (g/L) (  109/L)

Control 6.58 7 0.73 4.677 0.33 153.337 2.52 0.26 7 0.02

Infection 4.90 7 0.53## 4.87 7 0.25 155.337 8.08 0.26 7 0.02
Ribavirin (10 mg/ 9.447 0.41** 4.75 7 0.75 158.337 5.13 0.29 7 0.10
kg)
LPG (900 mg/kg) 9.107 0.95** 4.45 7 0.12 161.337 6.66** 0.317 0.02
LPG (300 mg/kg) 6.63 7 0.09** 4.05 7 0.10 164.337 7.23** 0.29 7 0.06
LPG (100 mg/kg) 6.337 0.32** 4.357 0.11 163.337 8.96** 0.34 7 0.01

Data are expressed as mean 7 S.E.M. with n 3 mice in each group. One-way AN-
OVA was used for the statistical analysis.
##
p o 0.05 compared with the control group. **p o0.05 compared with the infected
group.

Fig. 1. LPG's hemagglutination inhibition activities. (A) Pre-treated with LPG;

(B) after-treated with LPG; (C) H1N1 incubated with LPG. Bars represent 50% tissue
culture infective dose (TCID50). Viral infectivity was titrated by the chicken ery-
throcyte agglutination assay. Data are presented as mean 7S.E.M. Three in-
dependent experiments were performed in triplicate. One-way ANOVA was used
for the statistical analysis. **p o 0.05, compared with the infected group.

controls (p o0.05; Table 1). The ascent of WBC count was observed
in ribavirin and LPG group (p o0.05), whereas not signicantly
changes occurred in RBC and PLT count.
Fig.3. Lung wet weight/body weight (%). Data are presented as mean 7 S.E.M.
(n 10). One-way ANOVA was used for the statistical analysis. ##p o 0.05, compared
with the control group; **p o0.05, compared with the infected group.
132 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136

Fig.4. The images of histopathological change of lungs from a representative animal in treatment group are shown. Control group lung (A1A3); Infected group lung (B1
B3); Ribavirin group (100 mg/kg) lung (C1C3); LPG group (900 mg/kg) lung (D1D3); the images are represented with magnication of 100  and 400  , respectively.

3.5. Effect of LPG on lung edema
We quantied the magnitude of pulmonary edema by the lung
index. The infected group ranked top, which suggested the worst
lung injury, and both ribavirin and LPG orally treatment sharply
down-regulated the lung index (p o0.05; Fig. 3). However, in-
traperitoneally treatment didnot work (Data not shown).
3.6. Effect of LPG on histopathology of lungs
Control mice showed no lesions in lungs (Fig. 4A1). Infected
mice lungs showed pulmonary congestion and edema, severe
petechial or cruentate lesions. Moreover, large-scale lung
consolidation had formed (Fig. 4B1). Ribavirin treatment pre-
sented slight edema and slight petechiae (Fig. 4C1). Pathological
changes in LPG-treated mice lungs were minor as compared with
infection alone (Fig. 4D1). To be more specic, control mice
showed visible bronchus with ciliated columnar epithelium, as
well as clear alveolar sac structure and small arteries. No ber
proliferation and bubble wall thickness were observed (Fig. 4A2).
In mice with infection alone, the artery surrounding the structure
became loose, and the wall collagen ber proliferated and thick-
ened. Inammatory cells inltrated the bronchus (Fig. 4B2). Ri-
bavirin and control mice did not differ in histological patterns
(Fig. 4C2). Pathological changes were not obvious with LPG treat-
ment, except some local bubble wall thickening and some
 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136 133

Fig.5. (A) Western blot were utilized to detect the expression of IFN- in peripheral blood mononuclear cells, splenocytes and lung lymphocytes from C57BL/6 mice, which
were then quantied by densitometry. -actin was used as an internal control. All mice were not infected with H1N1. Data are mean7 S.E.M. (n 3). ###p o0.01 by Student t
test. (B) ELISA assayed levels of IFN- in spleen during H1N1 infection. Data are mean7 S.E.M. (n 10). ##p o 0.05, compared with control; **p o0.05, compared with infected
group by one-way ANOVA.

congestion in broken blood vessels (Fig. 4D2). LPG may ameliorate
lung lesions in mice with acute respiratory distress syndrome.
3.7. LPG markedly induced spleen lymphocyte expression of IFN-
The protein levels of IFN- were measured in mouse serum,
lung and spleen without viral infection. IFN- was mainly ex-
pressed in the spleen and was induced signicantly with LPG
(p o0.05; Fig. 5A). After challenging with virus, infected mice
showed suppression of IFN-, while ribavirin (100 mg/kg) or LPG
(900 mg/kg) up-regulated the expression, compared with control
mice (p o0.05; Fig. 5B).
3.8. Flow cytometry analysis of T lymphocyte subsets
To further study the increased IFN- level in treated groups, we
investigated the ratios of T-lymphocyte subsets and which subset
was activated and secreted IFN- after infection. Both CD4 and
CD8 T lymphocytes, the two main subsets of CD3 T lympho-
cytes, were activated either by ribavirin (100 mg/kg) or LPG
(900 mg/kg) treatment to produce IFN- (po0.05; Fig. 6A1A3).
Besides, the ratio of CD4 /CD8 which was disturbed by H1N1
restored to normal level after LPG treatment (Fig. 6B). Meanwhile,
the amounts of CD3 T cells were signicantly increased (Fig. 6B).
3.9. The role of IFN- in LPG treatment protects mice against lethal
inuenza challenge
LPG induced the production of IFN- from splenic lymphocytes
in WT mice, whereas IFN-
/
mice were resistant to being
induced by LPG (Fig. 7A).
We examined 50% lethal dosage (LD50) of H1N1 from 10
1 to
10
6 in WT and IFN-
/
mice, respectively. The survival rate of
mice was calculated during 15 consecutive days after infection.
The LD50 for WT mice was 10
4.34 and 10
4 for IFN-
/
mice.
In the following survival protection test, WT and IFN-
/
mice
were infected with 5 LD50. LPG signicantly protects WT mice
against H1N1-induced lethality with 32.4% (survival proportion) as
compared with infection alone, which was paralleled by weight
loss recovery (Fig. 7C). However, H1N1-infected IFN-
/
mice
were not protected by LPG and showed 100% mortality by day 9
(Fig. 7B). Besides, the weight loss of WT mice receiving LPG re-
covered faster than mice with infection alone (Fig. 7C). IFN-
/

mice lost weight constantly regardless of LPG treatment (Fig. 7C).
Thus, IFN- may be essential for the antiviral protection which
ascribed to LPG treatment.
4. Discussion
L. purpurascens (Oleaceae) is the original plant for Ku Ding Cha
tea, with extensive biological activities, including antioxidant,
anti-inammatory, hepato-protectant and cell apoptosis regula-
tion activities. The main active compound is glycosides (Song et al.,
2012). In previous experiments, mice with different doses of LPG
showed increased hemagglutination titers as compared with
controls, especially at 440 mg/kg and 1.32 g/kg in terms of anti-
body production of spleen cells, macrophage phagocytosis of
chicken RBCs and natural killer cell activity (Song et al., 2012). In
this study, we showed that LPG can inhibit viral infection by
134 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136

Fig.6. (A1A3) LPG treatment induces T-cell secretion of IFN-. C57BL/6 mice were orally administrated with LPG or ribavirin for 5 consecutive days, and infected with H1N1
on day 2. Flow cytometry quantication of IFN-, CD3 , CD4 and CD8 cells in splenic lymphocytes. Data are mean 7 S.E.M. (n 3). One representative FACS plot is
presented. ##po 0.05, compared with control;**po 0.05, compared with infected group by one-way ANOVA. (B) The T lymphocyte subsets in the spleen were analyzed by
ow cytometry. Data are presented as mean 7S.E.M. (n 10). ##p o 0.05, compared with the control group; **p o0.05, compared with the infected group.

enhancing the body's immunity, such as inducing the endogenous
antiviral cytokines. Firstly, only pre-treatment with LPG can re-
duce viral titers in vitro (Fig. 1), suggesting that LPG might inhibit
viral attachment or entry rather than viral replication, indicating
some LPG-inducing antiviral cytokines carry out. Secondly, LPG
increased the survival rate of H1N1-infected mice by alleviating
the severe pneumonia caused by H1N1 in vivo. Furthermore, the
role of IFN- induced by LPG in protecting mice against H1N1
infection was identied (Fig. 7B). Therefore, we conrmed that
LPG which inhibited H1N1 replication via inducing endogenous
IFN- represents a novel therapeutic approach for H1N1 infection.
All H1N1-infected mice died during the survival test (Fig. 2).
Respiratory diseases are often considered as the main cause of
mortality in H1N1 infection (Dai et al., 2014). Thus, the lung index
has been widely used to evaluate the degree of pulmonary edema
and inammatory exudation. The lung index sharply declined
when mice were administrated orally with LPG (300 mg/kg,
900 mg/kg) (p o0.05; Fig. 3). However, intraperitoneal adminis-
tration didnot work (Data not shown). Infected lungs showed
pulmonary congestion and edema and severe petechial or cruen-
tate lesions (Fig. 4). On hematoxylin and eosin staining, lesions
accompanied inammatory cell inltration occurred in the in-
fected group. LPG treatment reduced the lung edema and in-
ammation in infected mice (Fig. 4). Therefore, LPG alleviates the
development of lung edema and inammation. But the mechan-
isms still need further study.
Leukopenia commonly occurred during inuenza infection
(Wang et al., 2014). Blood testing revealed that the WBC count of
H1N1-infected mice greatly decreased (p o0.05), with no change
in RBC or PLT count or Hg level, as compared with controls
(Table 1). It is ascribed to viral infection may inhibit the leukocyte
activity, thus escaping from immunological surveillance. Com-
pared with infected group, the low WBC count caused by H1N1
infection would be restored higher level after being administrated
with LPG (p o0.05; Table 1).
In addition, signicant changes in CD3 count and ratio of
CD4 /CD8 change in spleen were observed during viral infection
(Liu et al., 2014; Tang et al., 2014; Yang et al., 2013). We found that
H1N1 inhibit the generation of CD3 T cells and disturb the dif-
ferentiation of T-lymphocyte subsets (Fig. 6B); But after treatment
with LPG, the account of CD3 T cells signicantly increased ac-
companying with the down-regulating the rate of CD4 /CD8 .
Actually, both CD4 and CD8 activation are essential to immune
network in the immune response to the virus by secreting antiviral
cytokines (Nayak et al., 2010). IFN- was secreted by natural-killer
cell and T cells and B cells, bridges the innate and adaptive im-
munity soon after the recognition of pathogen-associated mole-
cular patterns by the infected host (Bonjardim et al., 2009; Deng
et al., 2014). IFN- can directly induce the targeted cells to express
antiviral protein, such as Mx and OAS (Bonjardim et al., 2009;
Cheng et al., 2014; Wolfersttter et al., 2014), and enhances the
cytotoxic effect of natural killer cells and cytotoxic T lympocyte,
thus enlarging the anti-viral capability (Wu et al., 2008). We found
higher proportion of both IFN- CD4 and IFN- CD8 T cells
in spleen after LPG treatment as compared with infection alone
(Fig. 6A1A3). In addition, ribavirin, a widely used antiviral med-
icine, had a similar inducing effect. Its antiviral mechanism needs
further study. Also, CD8 T cells are taken as vital innate immune
cells in spleen, and were signicantly activated by LPG to secreted
IFN- rather than CD4 T cells (Fig. 6A3).
 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136 135

Fig.7. (A) Western analysis was performed to detect protein levels of IFN- in WT and IFN-
/
mice and then quantied by densitometry. -actin was used as an internal
control. Mice were not challenged with H1N1. Data are mean7 S.E.M. (n 3). ###p o 0.01 by Student t test. (B) Percentage survival of WT or IFN-
/
mice (n 8) was
observed after being challenged with H1N1. ##p o 0.05, compared with control; **p o 0.05, compared with infected group by log-rank (MantelCox) test. (C) Weight loss of
WT or IFN-
/
mice with H1N1 infection (n 8). Mice were weighted on day 2 post-infection for 5 consecutive days and weighed daily. Data are mean 7S.E.M. (n 8).

Besides, it is undeniable that IFN- accumulating in some or-
ganic would cause several side effects, such as lung inammation
and thymic atrophy (Hillesheim et al., 2014; Liu et al., 2014).
Moreover, Some studies also showed that hypercytokinemia which
occurred in the lungs, thymus and serum caused the mortal injury
during viral infection, but neglecting a moderate antiviral cyto-
kines modulation in spleen may also contribute to resist viral in-
fection (Wu et al., 2013). In our study, after treating with LPG in
infected mice, IFN- was mainly expressed in the spleen without
causing deadly lung inammatory injury, indicating the IFN-
expressed in different organs may present different functions
(Fig. 5A). Thus we take spleen as the central part of humoral and
cellular immunity against virus. Apart from spleen, whether IFN-
induced by LPG in other organs is involved in the antiviral me-
chanism should be further evaluated.
IFN-
/
mice lack antimicrobial ability because of IFN- de-
ciency inhibiting the murine immune system (Fig. 7A) (Dalton
et al., 1993). All IFN-
/
mice administrated with LPG died on day
9 (Fig. 7B), whereas WT mice administrated with LPG hold
approximately 30% of survival rate (Fig. 7B). Hence, IFN- is es-
sential for the antiviral protection mediated by LPG.
5. Conclusions
To sum up, LPG extracted from L. purpurascens signicantly
protect against H1N1 infection in vivo and in vitro, and oral ad-
ministration is more effective than intraperitoneal administration
in vivo. Moreover, LPG enhances the innate and adaptive immunity
via up-regulating endogenous IFN-, which is essential for LPG-
mediated protection during viral infection. Therefore, LPG up-
regulating endogenous IFN- may represent a novel therapeutic
approach for anti-H1N1 infection.
Funding
This study was funded by Shenzhen strategic emerging in-
dustry development project funding (Project no.
136 X.-p. Hu et al. / Journal of Ethnopharmacology 179 (2016) 128136
JCYJ20120829101455848), Shenzhen and Hong Kong collaborative
innovation of science and technology plan (Project no.
SGLH20120926161415784), Natural micro-molecule drug innova-
tion engineering laboratory funding (Project no. Shenfagai (2013)
180), Guangdong natural science program (Project no.
S2012030006598), Natural Science Foundation of Shenzhen Uni-
versity (Grant no. 201410), National natural science foundation of
China (Project no. 31500285), Natural science foundation of
Guangdong Province (Project no. 2015A030310529), and Post-
doctoral science foundation of China (Project no. 2015M582420,
2015M570726).
Acknowledgments
We appreciate Ni Liu, Fang Zhao, Xiang-Yang Li, Ling Feng (In-
stitute of Tropical Medicine, Guangzhou University of Chinese
Medicine) for technical support.
References
Bonjardim, C.A., Ferreira, P.C., Kroon, E.G., 2009. Interferons: signaling, antiviral and
viral evasion. Immunol. Lett. 122, 111.
Chen, Y.J., Chang, L.S., 2013. Hydroquinone-induced miR-122 down-regulation eli-
cits ADAM17 up-regulation, leading to increased soluble TNF-alpha production
in human leukemia cells with expressed Bcr/Abl. Biochem. Pharmacol. 86,
620631.
Cheng, C.H., Chou, C.M., Chu, C.Y., Chen, G.D., Lien, H.W., Hwang, P.P., Chang, M.S.,
Huang, C.J., 2014. Differential regulation of Tetraodon nigroviridis Mx gene
promoter activity by constitutively-active forms of STAT1, STAT2, and IRF9. Fish
Shellsh Immunol. 38, 230243.
Dai, W.P., Li, G., Li, X., Hu, Q.P., Liu, J.X., Zhang, F.X., Su, Z.R., Lai, X.P., 2014. The roots
of Ilex asprella extract lessens acute respiratory distress syndrome in mice
induced by inuenza virus. J. Ethnopharmacol. 155, 15751582.
Dalton, D.K., Pitts-Meek, S., Keshav, S., Figari, I.S., Bradley, A., Stewart, T.A., 1993.
Multiple defects of immune cell function in mice with disrupted interferon-
gamma genes. Science 259, 17391742.
Deng, Y., Chu, J., Ren, Y., Fan, Z., Ji, X., Mundy-Bosse, B., Yuan, S., Hughes, T., Zhang, J.,
Cheema, B., Camardo, A.T., Xia, Y., Wu, L.C., Wang, L.S., He, X., Kinghorn, A.D., Li,
X., Caligiuri, M.A., Yu, J., 2014. The natural product phyllanthusmin C enhances
IFN- production by human NK cells through upregulation of TLR-mediated
NF-kB signaling. J. Immunol. 193, 29943002.
Fakruddin, J.M., Lempicki, R.A., Gorelick, R.J., Yang, J., Adelsberger, J.W., Garcia-
Pineres, A.J., Pinto, L.A., Lane, H.C., Imamichi, T., 2007. Noninfectious papilloma
virus-like particles inhibit HIV-1 replication: implications for immune control
of HIV-1 infection by IL-27. Blood 109, 18411849.
Gadad, B.S., Daher, J.P., Hutchinson, E.K., Brayton, C.F., Dawson, T.M., Pletnikov, M.V.,
Watson, J., 2010. Effect of fenbendazole on three behavioral tests in male
C57BL/6N Mice. J. Am. Assoc. Lab. Anim. Sci. 49, 821825.
Gao, X.M., Zhang, D.Q., Zhang, J.J., 2006. Applied Illustrated Compendium of Materia
Medica vol. 1. Foreign Languages Press, Beijing, China, p. 111.
Ge, H., Liu, G., Xiang, Y.F., Wang, Y., Guo, C.W., Chen, N.H., Zhang, Y.J., Wang, Y.F.,
Kitazato, K., Xu, J., 2014. The mechanism of poly-galloyl-glucoses preventing
Inuenza A virus entry into host cells. PLoS One 9, e94392.
He, Z.D., Lau, K.M., But, P.P.H., Jiang, R.W., Dong, H., Ma, S.C., Fung, K.P., Ye, W.C., Sun,
H.D., 2003. Antioxidative glycosides from the leaves of ligustrum robustum. J.
Nat. Prod. 66, 851854.
He, Z.D., Liu, Y.Q., Yang, C.R., 1992. Glycosides from Ligustrum purpurascens. Acta
Bot. Yunnanica 14, 328336.
He, Z.D., Peng, Y., Xiao, P.G., 2010. The textual research of Kudingcha. In: Wang, X.
(Ed.), The Research and Development of Kudingcha vol. 1. Science Press, Beijing,
China, pp. 13.
Hillesheim, A., Nordhoff, C., Boergeling, Y., Ludwig, S., Wixler, V., 2014. Beta-catenin
promotes the type I IFN synthesis and the IFN-dependent signaling response
but is suppressed by inuenza A virus-induced RIG-I/NF-kappaB signaling. Cell
Commun. Signal. 12, 29.
Johari, J., Kianmehr, A., Mustafa, M.R., Abubakar, S., Zandi, K., 2012. Antiviral activity
of baicalein and quercetin against the Japanese Encephalitis virus. Int. J. Mol.
Sci. 13, 1678516795.
Kim, D.H., Lee, J.H., Park, S., Oh, S.S., Kim, S., Kim, D.W., Park, K.H., Kim, K.D., 2013. 6-
Acetonyl-5,6-dihydrosanguinarine (ADS) from Chelidonium majus L. triggers
proinammatory cytokine production via ROS-JNK/ERK-NF-B signaling path-
way. Food Chem. Toxicol. 58, 278279.
Kleinewietfeld, M., Manzel, A., Titze, J., Kvakan, H., Yosef, N., Linker, R.A., Muller, D.
N., Haer, D.A., 2013. Sodium chloride drives autoimmune disease by the in-
duction of pathogenic TH17 cells. Nature 496, 518522.
Lau, K.M., He, Z.D., Dong, H., Fung, K.P., But, P.P., 2002. Anti-oxidative, anti-in-
ammatory and hepato-protective effects of Ligustrum robustum. J. Ethno-
pharmacol. 83, 6371.
Li, C., Dai, Y., Zhang, S.X., Duan, Y.H., Liu, M.L., Chen, L.Y., Yao, X.S., 2014. Quinoid
glycosides from Forsythia suspensa. Phytochemistry 104, 105113.
Li, L., Xu, L.J., Ma, G.Z., Dong, Y.M., Peng, Y., Xiao, P.G., 2013. The large-leaved Ku-
dingcha (Ilex latifolia Thunb and Ilex kudingcha Ilex C.J. Tseng): a traditional
Chinese tea with plentiful secondary metabolites and potential biological ac-
tivities. J. Nat. Med. 67, 425437.
Liu, B., Zhang, X., Deng, W., Liu, J., Li, H., Wen, M., Bao, L., Qu, J., Liu, Y., Li, F., An, Y.,
Qin, C., Cao, B., Wang, C., 2014. Severe inuenza A(H1N1)pdm09 infection in-
duces thymic atrophy through activating innate CD8( )CD44(hi) T cells by
upregulating IFN-gamma. Cell Death Differ. 5, e1440.
Matikainen, S., Siren, J., Tissari, J., Veckman, V., Pirhonen, J., Severa, M., Sun, Q., Lin,
R., Meri, S., Uze, G., Hiscott, J., Julkunen, I., 2006. Tumor necrosis factor alpha
enhances inuenza A virus-induced expression of antiviral cytokines by acti-
vating RIG-I gene expression. J. Virol. 80, 35153522.
Mungall, B.A., Xu, X., Klimov, A., 2004. Surveillance of inuenza isolates for sus-
ceptibility to neuraminidase inhibitors during the 20002002 inuenza sea-
sons. Virus Res. 103, 195197.
Nair, M.P., Kandaswami, C., Mahajan, S., Chadha, K.C., Chawda, R., Nair, H., Kumar,
N., Nair, R.E., Schwartz, S.A., 2002. The avonoid, quercetin, differentially reg-
ulates Th-1 (IFNgamma) and Th-2 (IL4) cytokine gene expression by normal
peripheral blood mononuclear cells. Biochim. Biophys. Acta 1593, 2936.
Nayak, J.L., Richards, K.A., Chaves, F.A., Sant, A.J., 2010. Analyses of the specicity of
CD4 T cells during the primary immune response to inuenza virus reveals
dramatic MHC-linked asymmetries in reactivity to individual viral proteins.
Viral Immunol. 23, 169180.
Nguyen, J.T., Smee, D.F., Barnard, D.L., Julander, J.G., Gross, M., de Jong, M.D., Went,
G.T., 2012. Efcacy of combined therapy with amantadine, oseltamivir, and ri-
bavirin in vivo against susceptible and amantadine-resistant inuenza A viru-
ses. PLoS One 7, e31006.
Platanias, L.C., 2005. Mechanisms of type-I- and type-II-interferon-mediated sig-
nalling. Nat. Rev. Immunol. 5, 375386.
Rathinam, V.A., Fitzgerald, K.A., 2010. Inammasomes and anti-viral immunity. J.
Clin. Immunol. 30, 632637.
Shirey, K.A., Lai, W., Scott, A.J., Lipsky, M., Mistry, P., Pletneva, L.M., Karp, C.L.,
McAlees, J., Gioannini, T.L., Weiss, J., Chen, W.H., Ernst, R.K., Rossignol, D.P.,
Gusovsky, F., Blanco, J.C., Vogel, S.N., 2013. The TLR4 antagonist Eritoran pro-
tects mice from lethal inuenza infection. Nature 497, 498502.
Song, X., Li, C.Y., Zeng, Y., Wu, H.Q., Huang, Z., Zhang, J., Hong, R.S., Chen, X.X., Wang,
L.Y., Hu, X.P., Su, W.W., Li, Y., He, Z.D., 2012. Immunomodulatory effects of crude
phenylethanoid glycosides from Ligustrum purpurascens. J. Ethnopharmacol.
144, 584591.
Tanaka, T., Ikeda, T., Kaku, M., Zhu, X.H., Okawa, M., Yokomizo, K., Uyeda, M., No-
hara, T., 2004. A new lignan glycoside and phenylethanoid glycosides from
Strobilanthes cusia BREMEK. Chem. Pharm. Bull. 52, 12421245.
Tang, W., Liu, F., Chen, Y., Song, L., Dai, W., Li, C., Weng, D., Chen, J., 2014. Reduction
of IL-17A might suppress the Th1 response and promote the Th2 response by
boosting the function of Treg cells during silica-induced inammatory response
in vitro. Mediat. Inamm. 2014, 570894.
Wang, L., Chang, L.S., Lee, I.K., Tang, K.S., Li, C.C., Eng, H.L., You, H.L., Yang, K.D., 2014.
Clinical diagnosis of pandemic A(H1N1) 2009 inuenza in children with ne-
gative rapid inuenza diagnostic test by lymphopenia and lower C-reactive
protein levels. Inuenza Other Resp. Viruses 8, 9198.
Wolfersttter, M., Schweneker, M., Spth, M., Lukassen, S., Klingenberg, M., Brink-
mann, K., Wielert, U., Lauterbach, H., Hochrein, H., Chaplin, P., Suter, M.,
Hausmann, J., 2014. Recombinant modied vaccinia virus Ankara generating
excess early double-stranded RNA transiently activates protein kinase R and
triggers enhanced innate immune responses. J. Virol. 88, 1439614411.
Wong, I.Y., He, Z.D., Huang, Y., Chen, Z.Y., 2001. Antioxidative activities of pheny-
lethanoid glycosides from Ligustrum purpurascens. J. Agric. Food Chem. 49,
31133119.
Wu, L., Adams, M., Carter, T., Chen, R., Muller, G., Stirling, D., Schafer, P., Bartlett, J.B.,
2008. lenalidomide enhances natural killer cell and monocyte-mediated anti-
body-dependent cellular cytotoxicity of rituximab-treated CD20 tumor cells.
Clin. Cancer Res. 14, 46504657.
Wu, X.L., Ju, D.H., Chen, J., Yu, B., Liu, K.L., He, J.X., Dai, C.Q., Wu, S., Chang, Z., Wang,
Y.P., Chen, X.Y., 2013. Immunologic mechanism of Patchouli alcohol anti-H1N1
inuenza virus may through regulation of the RLH signal pathway in vitro. Curr.
Microbiol. 67, 431436.
Yang, S., Niu, S., Guo, Z., Yuan, Y., Xue, K., Liu, S., Jin, H., 2013. Cross-protective
immunity against inuenza A/H1N1 virus challenge in mice immunized with
recombinant vaccine expressing HA gene of inuenza A/H5N1 virus. Virol. J. 10,
291.
Zhou, W., Tan, X.B., Shan, J.J., Liu, T., Cai, B.C., Di, L.Q., 2014. Effect of chito-oligo-
saccharide on the intestinal absorptions ofphenylethanoid glycosides in Fructus
Forsythiae extract. Phytomedicine 21, 15491558.