key: cord-0002076-2s6hpakk
authors: Kwok, Hoi-Hin; Poon, Po-Ying; Fok, Siu-Ping; Ying-Kit Yue, Patrick; Mak, Nai-Ki; Chan, Michael Chi-Wai; Peiris, Joseph Sriyal Malik; Wong, Ricky Ngok-Shun
title: Anti-inflammatory effects of indirubin derivatives on influenza A virus-infected human pulmonary microvascular endothelial cells
date: 2016-01-06
journal: Sci Rep
DOI: 10.1038/srep18941
sha: dc24e1b58aa068d810743e726f98ef82af89f349
doc_id: 2076
cord_uid: 2s6hpakk

Influenza A virus (IAV) poses global threats to human health. Acute respiratory distress syndrome and multi-organ dysfunction are major complications in patients with severe influenza infection. This may be explained by the recent studies which highlighted the role of the pulmonary endothelium as the center of innate immune cells recruitment and excessive pro-inflammatory cytokines production. In this report, we examined the potential immunomodulatory effects of two indirubin derivatives, indirubin-3′-(2,3-dihydroxypropyl)-oximether (E804) and indirubin-3′-oxime (E231), on IAV (H9N2) infected-human pulmonary microvascular endothelial cells (HPMECs). Infection of H9N2 on HPMECs induced a high level of chemokines and cytokines production including IP-10, RANTES, IL-6, IFN-β and IFN-γ1. Post-treatment of E804 or E231 could significantly suppress the production of these cytokines. H9N2 infection rapidly triggered the activation of innate immunity through phosphorylation of signaling molecules including mitogen-activated protein kinases (MAPKs) and signal transducer and activator of transcription (STAT) proteins. Using specific inhibitors or small-interfering RNA, we confirmed that indirubin derivatives can suppress H9N2-induced cytokines production through MAPKs and STAT3 signaling pathways. These results underscore the immunomodulatory effects of indirubin derivatives on pulmonary endothelium and its therapeutic potential on IAV-infection.

Scientific RepoRts | 6:18941 | DOI: 10.1038/srep18941 manufacturer's instruction. Complementary DNA was synthesized from DNase-treated total RNA using Superscript II first-strand synthesis system (Invitrogen). The relative expression of target gene was quantified by real-time RT-PCR using KAPA SYBR Fast ABI prism qPCR Kit (KAPA Biosystems, Woburn, MA, USA) and detected by a StepOnePlus real-time PCR system (Applied Biosystems, Foster City, CA, USA). The relative expression of target gene was normalized by the level of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and then calculated by the comparative Ct method.

Plaque assay. The virus titers were determined by standard plaque assay on Madin-Darby canine kidney (MDCK) cells. In brief, MDCK cells were grown in MEM and seeded onto 6-well plates. Diluted cell culture medium from influenza virus-infected HPMECs were added to the confluent MDCK cells monolayers for 1h. Then, the inoculum was removed, and a mixture of agarose (2%, w/v) containing L-(tosylamido-2-phenyl) ethyl chloromethyl ketone (TPCK) (1 μ g/ml) was added onto the MDCK cells monolayers. After 72 h of incubation, the plate was fixed by formaldehyde (4%, v/v) overnight and then the agarose was discarded. The plaques were counted after staining with crystal violet (0.2%, w/v).

were prepared by NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Scientific, Rockford, IL, USA) according to the manufacturer's protocol. For extraction of whole-cell lysate, cells were lysed by CytoBuster TM Protein Extraction Reagent (Novagen, Madison, WI, USA) containing protease (0.5%, v/v) and phosphatase inhibitor cocktails (0.5%, v/v) (Calbiochem, San Diego, CA, USA). The cell lysate was collected after centrifugation. Protein concentration of the sample was determined by the detergent-compatible protein assay (Bio-Rad, Hercules, CA, USA). Equal amounts of protein were loaded and separated by 10% SDS-PAGE followed by electroblotting onto nitrocellulose membrane. The membrane was soaked in blocking buffer (1% non-fat milk in TBS-T, w/v) and then incubated with specific primary antibodies overnight at 4 °C and secondary antibody for 1 h at room temperature. Immunoreactive bands were visualized using SuperSignal West Pico Kit (Thermo Scientific).

In vitro mitogen-activated protein kinases assay. To detect the activity of individual MAPKs after treatment with IAV and indirubin derivatives, the non-radioactive In vitro protein kinase assay kit from Cell Signaling Technology was used. In brief, the Sepharose bead-immobilized antibody was used to immunoprecipitate active MAPKs from an equal amount of total cell lysate (200 μ g) overnight. The immunoprecipitate was washed twice with cell lysis buffer and kinase reaction buffer. The immunoprecipitate were then incubated with indirubin derivatives E804 or E231 (1 μ M) for 3 min before addition of ATP. Subsequently, kinase reactions using corresponding protein substrate were performed at 37 °C for 30 min. The kinase reaction was stopped with SDS loading buffer. Phosphorylation of protein substrate was detected by immunoblotting with specific antibody. Immunofluorescence microscopy. HPMECs at a density 1 × 10 4 were seeded onto a glass coverslip in a 24-well plate. After treatment for the indicated time, cells were fixed with 4% paraformaldehyde for 15 min at room temperature. Cells were incubated with primary antibody (1: 200 dilution rabbit anti-phospho-STAT3 (Tyr 705 ) antibody) overnight at 4 °C. The coverslip was washed and then incubated with FITC-conjugated goat anti-rabbit secondary antibody (1: 250 dilution) (Invitrogen) for 2 h at room temperature. Nuclei were visualized by staining with DAPI (0.5 μ g/ml). The coverslip was washed and mounted on a slide using DAKO fluorescence mounting medium (Carpinteria, CA, USA). Fluorescence image was captured by the Olympus Fluoview FV1000 confocal laser-scanning microscope (Tokyo, Japan).

Small interfering RNA (siRNA) transfection. Transfection of siRNA was performed using Lipofectamine RNAiMAX (Invitrogen). Non-targeting-siRNA (50 nM) was used in parallel with STAT3-specific siRNA (50 nM) (Ambion, Austin, TX, USA). Cells plated at 80% confluence were transfected in Opti-MEM medium (Gibco BRL, Grand Island, NY, USA) for 24 h. After transfection, cells were rinsed with Opti-MEM prior to further treatment.

All results were expressed as mean ± standard derivation (S.D.) of at least 3 independent experiments. Statistical significance between groups was determined by one-way ANOVA with Tukey's post hoc test. p < 0.05 was considered to be statistically significant.

Influenza A virus H9N2 is a potent inducer of cytokines production in pulmonary endothelial cells. Recent studies suggested that lung endothelium is the central regulator of cytokine amplification during influenza A virus infection, while dysregulation of cytokines production may result in systemic inflammation 22 . In this study, we found that the infection of influenza A virus subtype A/Quali/Hong Kong/G1/97 (H9N2) on HPMECs induced excessive production of various pro-inflammatory cytokines and chemokines, including IP-10 ( 

To examine the immunomodulatory effects of indirubin derivatives, HPMECs were infected with H9N2 for 1 h followed by incubation with indirubin derivatives E804 or E231 for another 24 h. We have tested the cytotoxicities of indirubin derivatives in HPMECs prior to the ELISA. As shown in Fig. 2 was observed at or below 10 μ M in HPMCEs. Next, indirubin derivatives were found to suppress H9N2-induced IP-10 ( Fig. 3A) , RANTES (Fig. 3B ) and IL-6 ( Fig. 3C ) expression in a concentration-related manner. E804 significantly inhibited cytokines expression at 1 μ M, a similar inhibitory effect was observed when a higher concentration of E231 (10 μ M) was used. Both indirubin derivatives slightly induced and inhibited the basal level of RANTES and IL-6 respectively. It may be due to the regulatory effects of indirubins on innate immunity 45 . For 46 . Similar to the results of direct virus infection, viral RNA stimulated cytokines expressions, and the inhibitory effects of MAPKs inhibitors were very similar to the experiments in direct infection. This confirmed that the suppression effects of MAPKs inhibitors on cytokines expression were not due to their potential effects on viral load. As a result, the anti-inflammatory effects of indirubin derivatives are mainly due to its inhibitory effects on different kinases.

In contrast, STAT3-specific siRNA had no effects on IP-10, RANTES and IL-6 production induced by H9N2 infection in HPMECs (Fig. 5D-F) . Time-course experiments showed that H9N2 rapidly induced p38 and JNK phosphorylation in 15 min after addition of the virus, then a second wave of p38 and JNK phosphorylation were induced at 24 h.p.i. and 6 h.p.i., respectively (Fig. 6A,B) . No activation of ERK was found after H9N2 infection. Similar to the early phosphorylation of stress-related MAPKs, STAT3 was phosphorylated early in 30 min after addition of the virus. However, the second wave of STAT3 phosphorylation was found at 2 h.p.i. Taken together, these results suggested that H9N2 infection on HPMECs could activate p38, JNK and STAT3 signaling pathways rapidly, and the expression of cytokines including IP-10, RANTES and IL-6 were mainly due to the activation of p38 and JNK.

Indirubin derivatives suppress H9N2-induced cytokines expression through direct inhibition of p38 and JNK activity. To study the underlying mechanism of the anti-inflammatory effects of indirubin derivatives, HPMECs were treated with indirubin derivatives E804 or E231 after H9N2 infection. As shown in Fig. 7 , E804 can significantly reduce H9N2-induced phosphorylation of p38 (Fig. 7A) and JNK (Fig. 7B ) at 24 and 6 h.p.i., respectively, and E804 demonstrated a more potent effect than E231. It has been suggested that indirubin and its derivatives are potent inhibitors of various kinases, including MAPKs. In vitro kinase assay on p38 and JNK showed that E804 but not E231 inhibited p38 and JNK kinases activity. This action was reflected by the reduced phosphorylation of their direct downstream substrates ATF2 (Fig. 7C ) and c-Jun (Fig. 7D) respectively.

Indirubin derivatives prevent H9N2-induced IFN-β expression through inhibition of STAT3 phosphorylation and nuclear translocation. Though we found no relationship between STAT3 and H9N2-induced IP-10, RANTES and IL-6 expressions (Fig. 5D-F) , STAT signaling pathway is indispensable for the induction of interferons. We showed that knockdown of STAT3 strongly inhibited H9N2-induced IFN-β (Fig. 8B ) mRNA expression. To elucidate the inhibitory effects of E804 on STAT3, Western blot analysis was performed. Treatment with E804 or E231 inhibited H9N2-induced STAT3 tyrosine phosphorylation (Fig. 8C ). Upon activation, STAT3 forms homo-or heterodimers that translocate to the nucleus. HPMECs fractionation of nucleus and cytoplasm was obtained by means of subcellular fractionation followed by Western blot analysis. The results showed that H9N2 infection increased phosphorylated STAT3 in the nuclear fraction, while treatment with E804 significantly reduces the nuclear translocation (Fig. 8D) . Similar to the result of Western blot analysis, the confocal image also showed that increased fluorescence signal was found in the nucleus after H9N2 infection in HPMECs (Fig. 8E) . Treatment with E804 reduced STAT3 fluorescence signal in the nucleus.

The emergence of IAV poses a serious global threat to human health. Besides regular epidemic outbreaks, severe pandemics like the 1918 Spanish flu and the more recent 2009 swine flu had caused enormous social and economic burden. Current treatment of IAV infection by M2-ion channel inhibitors or NA inhibitors emerged high frequency of resistance, and the efficacy and effectiveness of these antiviral drugs are limited by disappointing success rate 1 , so alternative or complementary therapies that modulate the signaling pathways utilized by IAV came into focus.

In this report, we demonstrated that indirubin derivatives, particularly E804 is a potent immunomodulatory compound for IAV-infection in vitro by inhibiting intracellular signaling pathways in pulmonary endothelial cells.

During the early stage of IAV infection, innate immune cells are recruited to the site of infection and are associated with an overwhelming production of pro-inflammatory cytokines and chemokines. Endothelial cells in the pulmonary vasculature form a barrier between the blood and interstitium. This strategic position suggests that pulmonary endothelial cells are prone to be affected by the cytokines and viral particles released from the IAV-infected epithelial cells. A recent study by Teijaro et al. identified endothelial cells as the central orchestrator which contribute to the aberrant pro-inflammatory cytokine and chemokine production during early IAV infections 22 . Concomitant with our in vitro data, we showed that H9N2 virus can efficiently infect HPMECs (Fig. 4A,B) and induce a significant amount of IP-10, RANTEs, IL-6, IFN-β and IFN-γ 1. IP-10 and RANTES are the chemoattractants for leukocytes including T cells, NK cells, and granulocytes, while production of IL-6 by endothelial cells initiates infiltration of neutrophils in the early phase of infection 47 . These cytokines have been found histopathologically in the lungs (including epithelial and endothelial cells) of H5N1 infected patients, who showed acute respiratory distress syndrome (ARDS) 48 , and ARDS can be characterized by progressive pulmonary endothelial damage. It has been suggested that treatment with antibodies against IP-10 in H1N1 infected mice can improve the survival rate and reduce acute lung injury 49 . Furthermore, suppression of early innate cytokine and chemokine production in the pulmonary endothelium can significantly improve survival of mice infected with lethal H1N1 Swine IAV 22 . These studies suggested that inhibition of cytokines production of the pulmonary endothelium is an attractive therapeutic strategy against IAV-induced cytokine storm.

Since severe infection of the influenza virus triggers the activation of the innate immune response and sometimes results in the induction to a cytokine storm. In this study, we demonstrated the immunomodulatory effects of indirubin derivatives, particularly E804 on IAV-infected pulmonary endothelial cells. Over the past two decades, many studies have identified that indirubin derivatives are potent inhibitors of various kinases, including MAPKs 35,36 , Src kinase 37 , glycogen synthase kinase-3β (GSK-3β ) 50 and cyclin-dependent kinases (CDKs) 51 . Based on these findings, potential functions of indirubin derivatives have been proposed, including anti-inflammation 35, [40] [41] [42] , anti-leukemia 36 , antiviral 38, 39 and angiosuppression 52, 53 . Crystal structure analysis revealed that indirubin can form three hydrogen bonds with the ATP-binding pocket of CDKs, thereby competitively inhibiting ATP binding in the catalytic domain of CDKs 36 . The results from our in vitro kinase assay also demonstrated that E804 is a potent inhibitor of p38 and JNK (Fig. 7C,D) . The cytokine ELISA data also suggest that H9N2-induced IP-10 expression of HPMECs was dependent on p38 and JNK activation, while RANTES and IL-6 were controlled by JNK and p38, respectively ( Fig. 5A-C) . However, the Western blot analysis showed that E804 could also inhibit phosphorylation of p38 (Fig. 7A) and JNK (Fig. 7B) , which mean upstream kinases of p38 and JNK may also be inhibited by E804. MAPKs are important mediators of influenza-induced cytokine expression. In fact, p38 has been shown to regulate the stability of IL-6 mRNA 54 . Meanwhile, another study also indicated the critical function of p38 on IP-10 during viral infection 55 . Furthermore, inhibition of p38 by specific inhibitor SB202190 in vivo can greatly diminish H5N1 lethal infection 28 . However, the role of JNK in IAV infection has not been fully examined. Nacken et al. elucidated that influenza viral RNA induces JNK phosphorylation in an RIG-I dependent manner, but the NS1 of IAV has also an intrinsic JNK activating property 56 . Taken together, the potent inhibitory effect of p38 and JNK signaling pathways by E804 strongly correlates with its anti-inflammatory function.

IFNs are pro-inflammatory cytokines crucial for antiviral responses to IAV infection. STAT1 and STAT2 are predominant and essential transcription factors of type I and II interferons signaling pathway, but the role of STAT3 activation after IFN binding remains controversial. Undeniably, STAT3 is indispensable for downstream signaling pathway of many other cytokines like IL-6, VEGF or EGF 57 . Our data showed that H9N2 infection on HPMECs can significantly induce IFN-β and IFN-γ 1 expression. Interestingly, STAT3-specific siRNA has no effect on H9N2-induced IL-6 ( Fig. 5F ) and IFN-γ 1 (Fig. 8B ), but strongly inhibit IFN-β expression level, indicating the involvement of STAT3 in IFN-β induction. The Western blot data showed that IAV-infection could activate STAT3 in early stage (15 min after addition of virus) followed by another activation at 2 h.p.i. (Fig. 6) . H9N2 was found to upregulate TLR-8 and MyD88, which is critical to the induction of IFN-β 58 . In line with this finding, the early activation of STAT3 may function together with the downstream signaling molecules of TLR-8, possibly the interferon regulatory factors (IRFs), to induce rapid expression of IFN-β mRNA at 2 h.p.i. (Fig. 1H) . And then the autocrine effect of IFN-β induces a further amplification of IFN-β expression and result in a cytokine storm. A previous study suggested that E804 could inhibit STAT3 dimerization and subsequent DNA binding 37 . Our translocation experiments clearly indicated that E804 could inhibit STAT3 phosphorylation and nuclear translocation. Since the induction of many pro-inflammatory cytokines requires type I interferon signaling, inhibition of IFN-β production by E804 may blunt the early induction of these cytokines. Though IFN-β is a well-known antiviral cytokine, it is also involved in the pathogenesis of influenza infection. In vivo studies focusing on the S1P 1 receptor and p38 pathway also suggested that, even the IAV-induced IFNs were suppressed by an S1P 1 receptor agonist 22 or p38 inhibitor 28 , the survival rate of the infected mice could still be significantly improved if those strongly induced cytokines were suppressed. In many studies, increased viral load were concomitant of reduced interferons expressions. However, in the present study, viral titer did not further increased in indirubin derivatives treated cells (Fig. 4B ) even IFNβ and IFN-γ 1 were suppressed, the results indicated that suppression of cytokines produced by the infected pulmonary endothelium could reduce IAV pathogenicity independent of the viral clearance. In fact, many kinases including CDK 39 and MAPKs, which are suggested being involved in the influenza replication process are also the target of indirubin derivatives, this might explain the partial antiviral effect of indirubin derivatives in this model. Meanwhile the phosphorylation and nuclear translocation of STAT3 leaded to induction of IFN-β . Indirubin derivatives particularly E804 is a potent inhibitor of p38 and JNK signaling pathways. E804 could also reduce the phosphorylation and nuclear translocation of STAT3. By inhibition of these signaling pathways, E804 could significantly suppress H9N2-induced cytokine burst in HPMECs.

Scientific RepoRts | 6:18941 | DOI: 10.1038/srep18941

Combination therapies coupling with antiviral and immunomodulatory drugs have been investigated intensively 29 . The encouraging results from in vitro and pre-clinical studies have led to an increased interest on this topic. However, to achieve the best clinical outcome, antiviral and immunomodulatory drugs should be administrated at the appropriate time during an infection. Further understanding of the immune dynamic could allow us to design an optimum therapy strategy. In this report, we demonstrated for the first time, the potent immunomodulatory effects of indirubin derivatives on pulmonary endothelial cells and their therapeutic potential for IAV-infection (Fig. 9) . As a result, the combinational effects of indirubin derivatives and antiviral drug in animal model warrant further investigation.

Perspectives on influenza evolution and the role of research

Structure and assembly of the influenza A virus ribonucleoprotein complex

Pathogenesis of influenza virus infections: the good, the bad and the ugly

Innate immunity to influenza virus infection

Detrimental contribution of the Toll-like receptor (TLR)3 to influenza A virus-induced acute pneumonia

Efficient influenza A virus replication in the respiratory tract requires signals from TLR7 and RIG-I

Characterization of conserved viral leader RNA sequences that stimulate innate immunity through TLRs

Toll-like receptor 10 is involved in induction of innate immune responses to influenza virus infection

Differential role of TLR-and RLR-signaling in the immune responses to influenza A virus infection and vaccination

Innate immune responses to influenza A H5N1: friend or foe?

Mitogen-activated protein kinases in innate immunity

N-terminal kinases as potential therapeutic targets

Induction of proinflammatory cytokines in primary human macrophages by influenza A virus (H5N1) is selectively regulated by IFN regulatory factor 3 and p38 MAPK

Targeting cell signalling pathways to fight the flu: towards a paradigm change in anti-influenza therapy

RNA viruses and the mitogenic Raf/MEK/ERK signal transduction cascade

Influenza virus non-structural protein 1 (NS1) disrupts interferon signaling

Mammalian innate resistance to highly pathogenic avian influenza H5N1 virus infection is mediated through reduced proinflammation and infectious virus release

Human and avian influenza viruses target different cell types in cultures of human airway epithelium

Transmission of influenza A in human beings

Influenza A viruses target type II pneumocytes in the human lung

Influenza H5N1 virus infection of polarized human alveolar epithelial cells and lung microvascular endothelial cells

Endothelial cells are central orchestrators of cytokine amplification during influenza virus infection

H5N1 virus activates signaling pathways in human endothelial cells resulting in a specific imbalanced inflammatory response

Human pulmonary microvascular endothelial cells support productive replication of highly pathogenic avian influenza viruses: possible involvement in the pathogenesis of human H5N1 virus infection

Into the eye of the cytokine storm

Pathogenesis of influenza-induced acute respiratory distress syndrome

Dissecting influenza virus pathogenesis uncovers a novel chemical approach to combat the infection

Inhibition of p38 mitogen-activated protein kinase impairs influenza virus-induced primary and secondary host gene responses and protects mice from lethal H5N1 infection

Newer influenza antivirals, biotherapeutics and combinations

New strategies to overcome the drawbacks of currently available flu vaccines

The epidemiology and spread of drug resistant human influenza viruses

Targeting the "cytokine storm" for therapeutic benefit

Influenza neuraminidase: a druggable target for natural products

Dual functions of ginsenosides in protecting human endothelial cells against influenza H9N2-induced inflammation and apoptosis

Indirubin-3-monoxime exhibits anti-inflammatory properties by down-regulating NF-kappaB and JNK signaling pathways in lipopolysaccharide-treated RAW264.7 cells

Indirubin, the active constituent of a Chinese antileukaemia medicine, inhibits cyclin-dependent kinases

Indirubin derivatives inhibit Stat3 signaling and induce apoptosis in human cancer cells

Anti-inflammatory and antiviral effects of indirubin derivatives in influenza A (H5N1) virus infected primary human peripheral blood-derived macrophages and alveolar epithelial cells

Indirubin-3′ -monoxime, a derivative of a Chinese antileukemia medicine, inhibits P-TEFb function and HIV-1 replication

5′ -nitro-indirubinoxime inhibits inflammatory response in TNF-alpha stimulated human umbilical vein endothelial cells

Indirubin inhibits inflammatory reactions in delayed-type hypersensitivity

Inhibition of RANTES expression by indirubin in influenza virus-infected human bronchial epithelial cells

Amino acid residues 253 and 591 of the PB2 protein of avian influenza virus A H9N2 contribute to mammalian pathogenesis

The ginsenoside protopanaxatriol protects endothelial cells from hydrogen peroxide-induced cell injury and cell death by modulating intracellular redox status

Indirubin-3′ -(2,3 dihydroxypropyl)-oximether (E804) is a potent modulator of LPSstimulated macrophage functions

The Rac1 inhibitor NSC23766 exerts anti-influenza virus properties by affecting the viral polymerase complex activity

Understanding the roles of cytokines and neutrophil activity and neutrophil apoptosis in the protective versus deleterious inflammatory response in pneumonia

Pathological study of archival lung tissues from five fatal cases of avian H5N1 influenza in Vietnam

Monoclonal antibody against CXCL-10/IP-10 ameliorates influenza A (H1N1) virus induced acute lung injury

Soluble 3′ ,6-Substituted Indirubins with Enhanced Selectivity toward Glycogen Synthase Kinase -3 Alter Circadian Period

Indirubins Inhibit Glycogen Synthase Kinase-3β and CDK5/P25, Two Protein Kinases Involved in Abnormal Tau Phosphorylation in Alzheimer's Disease: A peoperty common to most cyclin-dependent kinase inhibitors

An indirubin derivative, E804, exhibits potent angiosuppressive activity

Automated, Quantitative Screening Assay for Antiangiogenic Compounds Using Transgenic Zebrafish

The p38 MAP kinase pathway signals for cytokine-induced mRNA stabilization via MAP kinase-activated protein kinase 2 and an AU-rich region-targeted mechanism

Innate immune response to H3N2 and H1N1 influenza virus infection in a human lung organ culture model

Activation of c-jun N-Terminal Kinase upon Influenza A Virus (IAV) Infection Is Independent of Pathogen-Related Receptors but Dependent on Amino Acid Sequence Variations of IAV NS1

STATs in cancer inflammation and immunity: a leading role for STAT3

Human intestinal epithelial cells are susceptible to influenza virus subtype H9N2

This work was supported by the Area of Excellence Scheme of the University Grants Committee, Hong Kong SAR Government (AoE/M-12/06) and Dr. Gilbert Hung Ginseng Laboratory Fund.