ORIGINAL RESEARCH

The Antimalarial Drug Artesunate Inhibits Primary Human CulturedAirway Smooth Muscle Cell ProliferationSheryl S. L. Tan1, Benjamin Ong1, Chang Cheng2, Wanxing Eugene Ho4, John K. C. Tam3, Alastair G. Stewart5,Trudi Harris5, W. S. Fred Wong2, and Thai Tran1

1Department of Physiology, 2Department of Pharmacology, and 3Department of Surgery, Yong Loo Lin School of Medicine, NationalUniversity of Singapore, Singapore; 4Saw Swee Hock School of Public Health, National University Health System, Singapore;and 5Department of Pharmacology, University of Melbourne, Australia

Abstract

Airway smooth muscle (ASM) cell hyperplasia contributes to airwaywall remodeling (AWR) in asthma. Glucocorticoids, which are usedas first-line therapy for the treatment of inflammation in asthma,have limited impact on AWR, and protracted usage of high doses ofglucocorticoids is associated with an increased risk of side effects.Moreover, patientswith severe asthmaoften show reduced sensitivityto glucocorticoids.Artesunate, a semisynthetic artemisinin derivativeused to treat malaria with minimal toxicity, attenuates allergicairway inflammation in mice, but its impact on AWR is not known.We examined the effects of artesunate on ASM proliferation in vitroand in vivo. Primary human ASM cells derived from nonasthmaticdonors were treated with artesunate before mitogen stimulation.Artesunate reducedmitogen-stimulated increases in cell number andcyclin D1 protein abundance but had no significant effect on ERK1/2phosphorylation. Artesunate, but not dexamethasone, inhibitedphospho-Akt and phospho-p70S6K protein abundance. Artesunate,but not dexamethasone, inhibited mitogen-stimulated increases incell number, cyclin D1, and phospho-Akt protein abundance onASM cells derived from asthmatic donors. In a murine model ofallergic asthma, artesunate reduced the area of a-smooth muscleactin–positive cells and decreased cyclin D1 protein abundance. Ourstudy provides a basis for the future development of artesunate asa novel anti-AWR agent that targets ASM hyperplasia via the

PI3K/Akt/p70S6K pathway and suggests that artesunate may be usedas combination therapy with glucocorticoids.

Keywords: asthma; airway wall remodeling; PI3K; cyclin D1;p70S6K

Clinical Relevance

Our study describes, for the first time, a novel application forthe antimalarial drug artesunate, which shows anti-airway wallremodeling properties that target airway smooth muscleproliferation via the PI3K/Akt/p70S6K pathway, a pathwaythat is not targeted by glucocorticoids. Our study may offer anadditional therapeutic strategy in combination therapy for useby patients who are insensitive to glucocorticoids due to anelevated PI3K/Akt signaling pathway. Results from our studywill increase our understanding of the mechanisms bywhich airway wall remodeling occurs, and the findings haveimplications for other lung-related diseases, such as chronicobstructive pulmonary disease and lymphangioleiomyomatosis,where hyperproliferation of abnormal airway smoothmuscle cells in the lungs of women of child-bearing age aretargets.

An essential component of airway wallremodeling (AWR) in asthma is thethickening of the airway smooth muscle(ASM) layer, which is attributed to ASMhypertrophy and/or hyperplasia (1).Hyperplasia of ASM cells further amplifiescontraction, leading to obstruction of theairways (2). ASM has been identified as an

attractive target for the development ofantiasthma agents (3, 4). However, thedegree to which treatment can reverseremodeling of ASM is unknown.

Glucocorticoids are antiinflammatoryagents that reduce AHR (3, 5) but show lowefficacy regarding features of AWR (6).Glucocorticoid therapy is accompanied by

many side effects (7), such as infection thataccompanies attained immunodeficiencydue to chronic steroid use (8). Therefore,there is a need for alternative andsupplementary anti-AWR drugs with highefficacy and minimal side effects.

Artesunate is a partially syntheticartemisinin derivative extracted from the

(Received in original form June 18, 2013; accepted in final form September 10, 2013 )

Correspondence and requests for reprints should be addressed to Thai Tran, Ph.D., Department of Physiology, MD9, 2 Medical Drive, National University ofSingapore, Singapore 117597. E-mail: tran_thai@nuhs.edu.sg

Am J Respir Cell Mol Biol Vol 50, Iss 2, pp 451–458, Feb 2014

Copyright © 2014 by the American Thoracic Society

Originally Published in Press as DOI: 10.1165/rcmb.2013-0273OC on September 25, 2013

Internet address: www.atsjournals.org

Tan, Ong, Cheng, et al.: Artesunate Inhibits ASM Cell Proliferation 451

Chinese herb sweet wormwood (Artemisiaannua) (9). With minimal toxicity andthe absence of apparent side effects,artesunate has been approved for thetreatment of malaria (10). Artesunateexerts antiproliferative effects in humanosteosarcoma cells and myeloma cells viaunknown mechanisms (11, 12). In asthma,artesunate attenuates allergic airwayinflammation in mice (13). However, nostudies to date have assessed the role ofartesunate against features of AWR. Wehypothesize that treatment of humanASM cells with artesunate may inhibithyperplasia to influence AWR.

We demonstrate for the first time thatartesunate exerts inhibitory effects on thePI3K/Akt/p70S6K pathway to regulate cyclin

D1, resulting in antiproliferative effects inprimary human cultured ASM cells derivedfrom nonasthmatic and asthmatic donorsand in a murine model of allergic asthma.

Materials and Methods

Cell CulturePrimary human ASM cell cultures wereobtained from macroscopically healthysegments of the second to fourth generationmain bronchus from patients with asthma(n = 4; 50% male; age, 35 6 17 yr [mean 6SD; range, 14–54 yr]) or without asthma(n = 4; 50% male; age, 58 6 15 yr [range,38–70 yr]) who underwent lung resectionsurgery (14). All procedures were approved

by the Institutional Review Board of theNational University of Singapore and theUniversity of Melbourne.

Experimental SystemHuman ASM cells (60–70% monolayerconfluence) were serum deprived (24 h)in DMEM with 1% ITS (insulin 5 mg/ml;transferrin 5 mg/ml; selenium 5 mg/ml) and1% penicillin streptomycin.

Various mitogens were used to assesswhether the effects of artesunate weremitogen dependent. ASM cells werepretreated (1 h) with dexamethasone(100 nM) (15) (Sigma-Aldrich, St. Louis,MO) or artesunate (0.3–30 mM) (Sigma-Aldrich).

Cell Number EnumerationAfter incubation with mitogens for48 hours, cell counting by trypan blueexclusion method was performed (14).

Immunoblot AnalysisAfter incubation with mitogens, the cellswere lysed, and proteins (6–12 mg) wereseparated on a 12% SDS-PAGE andtransferred onto nitrocellulose membranes(Bio-Rad, Hercules, CA) as previouslydescribed (16). Antibodies used were cyclinD1, phospho-ERK1/2, total ERK, phospho-Akt (S473), total Akt, phospho-p70S6K

(T389), total p70S6K (all 1:1,000 dilution)(Cell Signaling, Danvers, MA), or b-actin(1:15,000 dilution) (Sigma-Aldrich).

AnimalsAll animal experiments were performedaccording to the Institutional Guidelinesfor Animal Care and Use Committee(National University of Singapore).Female BALB/c mice (6–8 wk old) wereused in these experiments (Centre forAnimal Resource, Australia). Briefly,mice were sensitized by intraperitonealinjections of 20 mg OVA and 4 mg Al(OH)3 suspended in 0.1 ml saline onDays 0 and 14. On Days 22, 23, and 24,mice were challenged with 1% OVAaerosol for 30 minutes (17). Artesunate(30 mg/kg) or vehicle (5% DMSO) in0.1 ml saline was given by intraperitonealinjection 1 hour before each OVA aerosolchallenge. Saline aerosol was used asa negative control. Mice were killedon Day 25, and lung tissues werecollected for histological analysis andimmunoblotting (20 mg).

Figure 1. Effect of increasing concentrations of artesunate on primary human airway smooth muscle(ASM) cell number. (A) Quiescent cells were stimulated for 48 hours with FBS (5%) and countedusing trypan blue exclusion to identify nonviable cells. Dexamethasone (Dex, 100 nM) andartesunate (ART, 0.3–30 mM) were added 1 hour before stimulation. (B) Quiescent cells werestimulated for 48 hours with thrombin (Thr) (0.3 U/ml), basic fibroblast growth factor (bFGF) (0.3 nM),and epidermal growth factor (EGF) (10 ng/ml) and counted using trypan blue exclusion to identifynonviable cells. ART (30 mM) was added 1 hour before stimulation. Data are expressed as foldincrement over basal (unstimulated cells). *P , 0.05 compared with basal. †P , 0.05 comparedwith FBS, Thr, bFGF, or EGF (n = 4).

ORIGINAL RESEARCH

452 American Journal of Respiratory Cell and Molecular Biology Volume 50 Number 2 | February 2014

Harris Hematoxylin andEosin StainingLung sections (5 mm) were deparaffinizedwith xylene, rehydrated with graded alcohol,and subjected to Harris hematoxylinstaining, differentiation in 0.5% acid alcohol,and counterstaining with eosin.

ImmunohistochemistryImmunohistochemical staining of a-smoothmuscle actin (a-SMA) (1:100) was performed(16). The area of a-SMA–positive cells(measured by a single observer in a blindedfashion) was corrected for airway basementmembrane length (18). CellSens dimensionsoftware (Olympus, Hamburg, Germany) wasused for analysis. Digital photographs of lungsections were analyzed at a magnification of 203.

Statistical AnalysisDifferences were determined using one-wayANOVA with repeated measures

(GraphPad Prism 6; GraphPad Software,La Jolla, CA) followed by Bonferroni’s posthoc test (for cell number and immunoblotanalysis) or uncorrected Fisher’s leastsignificant difference test (for a-SMAquantification). A P value , 0.05 wasdeemed significant.

Results

Artesunate Inhibits Primary HumanASM Cell Number Independent ofMitogen UsedIncubation of quiescent primary humanASM cells with FBS for 48 hourssignificantly increased ASM cell numberby approximately 1.8-fold when comparedwith basal (unstimulated ASM cells) (Figure1A). Pretreatment of ASM cells withdexamethasone or artesunate significantlyinhibited the effects of FBS on cell number

(Figure 1A). There was no difference in thepercentage of trypan blue positive cellsacross all treatment groups, with an averagepercentage of trypan blue–positive cellsof 15 6 4% (n = 4). Moreover, ourpreliminary studies show that artesunatealone had no significant effect on basalhuman ASM cell number (basal, 1.0 6 0.0;100 nM dexamethasone alone, 1.0 6 0.2;30 mM artesunate alone, 1.0 6 0.1),suggesting that the effects of artesunate onASM cells were not due to drug-inducedcytotoxicity. To examine whether thisphenomenon was a general inhibitory effectof artesunate, we tested other well-characterized mitogens of ASM cells. Weshowed that artesunate was able to inhibitASM cell proliferation induced by thrombin(Thr) (Sigma-Aldrich), basic fibroblastgrowth factor (Promega, Madison, WI),and epidermal growth factor (EGF) (Sigma-Aldrich) (Figure 1B), whereas previousstudies indicate that EGF is relativelyresistant to modulation by dexamethasone(19). Taken together, these resultsdemonstrate that artesunate inhibits primaryhuman ASM cell number and that theeffects appear to be non–mitogen dependent.

Artesunate Decreases CyclinD1 Protein Abundancebut Not ERK1/2 PhosphorylationWe next investigated the effect ofartesunate on a number of biomarkers ofcell proliferation using immunoblot analysis(i.e., namely cyclin D1, ERK1/2, Akt, andp70S6K). The incubation period of ASMcells with the respective mitogens for themeasurement of ERK1/2, Akt, p70S6K

phosphorylation, and cyclin D1 are chosenon the basis of published work by ourgroup and others (14, 20–23). FBS causeda significant increase in the protein levelsof cyclin D1 when compared with basal(Figure 2A). FBS-induced increase in cyclinD1 protein abundance was significantlyinhibited when cells were pretreated withdexamethasone and artesunate (Figure 2A).A similar pattern was observed when themitogen was switched to thrombin, withartesunate exerting similar inhibitory effectson cyclin D1 protein abundance (Figure 2B).

Because artesunate influences cyclinD1 protein levels, we investigatedwhether artesunate regulates ERK1/2phosphorylation. FBS increased thephosphorylation of ERK1/2 (Table 1).However, the effects of FBS on ERK1/2phosphorylation were not significantly

Figure 2. Immunoblot analysis showing the effect of ART on cyclin D1 protein expression in primaryhuman ASM cells. Cells were stimulated for 20 hours with FBS (5%) (A) or Thr (0.3 U/ml) (B).Dex (100 nM) and ART (3, 10, and 30 mM) were added 1 hour before stimulation. Data areexpressed as fold increment over basal (unstimulated cells) relative to b-actin protein abundance.*P , 0.05 compared with basal. †P , 0.05 compared with FBS or Thr (n = 4–5).

ORIGINAL RESEARCH

Tan, Ong, Cheng, et al.: Artesunate Inhibits ASM Cell Proliferation 453

inhibited by pretreatment withdexamethasone or artesunate. Given thateach primary human ASM culture wasderived from an individual patient, weattribute the large variation in ERKphosphorylation responses to interpatientvariability. For example, the magnitude of

ERK phosphorylation to FBS stimulationranged from 2- to 20-fold. However, the useof a paired test analysis enabled interdonorvariation to be removed from the analysis,and on this basis it is clear that neitherdexamethasone nor artesunate influencedERK phosphorylation (normalized to FBS

100%, dexamethasone 151.4%, andartesunate 96.8%). Similar results wereobtained in experiments using thrombinas the mitogen (normalized to thrombin100%, dexamethasone 115.3%, andartesunate 93.8%) (Table 1).

Artesunate, but Not Dexamethasone,Decreases Akt andp70S6K PhosphorylationGiven that the inhibition of cell number andcyclin D1 by artesunate was nonmitogendependent, we focused our subsequentstudies on FBS as a stimulus to study theeffects of artesunate on downstream signalingpathways that regulate human ASMproliferation. The PI3K/Akt pathway worksin parallel to that of the ERK1/2 pathwayto mediate ASM cell proliferation (24–26).Downstream of Akt, p70S6K has been shownto mediate DNA synthesis in human (26)and bovine (27) ASM cells. Inhibition ofp70S6K activity inhibits cyclin D1 proteinexpression (28). We showed that FBSsignificantly increased the levels of phospho-Akt and phospho-p70S6K relative to basal(Figure 3). Pretreatment with artesunate,but not dexamethasone, inhibited the FBS-stimulated increase in phospho-Akt andphospho-p70S6K (Figure 3).

Effects of Artesunate on PrimaryHuman ASM Cells Derived fromDonors with AsthmaAs a more pathophysiologically relevantin vitro model, we examined whether theantiproliferative effects of artesunate wereobserved in ASM cells taken from subjectswith asthma. Similar to that of nonasthmaticdonors, incubation with FBS for 48 hourssignificantly increased ASM cell numberby approximately 2-fold when comparedwith basal (Figure 4A). FBS also causeda significant increase in the protein levelsof cyclin D1 and phospho-Akt whencompared with basal (Figures 4B and 4C).Pretreatment with artesunate, but notdexamethasone, inhibited the effects of FBSon cell number, cyclin D1, and phospho-Aktprotein abundance (Figures 4A–4C). Theseresults demonstrate that artesunate exertsantiproliferative effects on primary humanASM cells derived from nonasthmaticdonors and from subjects with asthma.

Effects of Artesunate in a MurineModel of Allergic AsthmaTo verify the antiproliferative effects ofartesunate in vivo, we used a murine model

Table 1: Effect of Artesunate on Phospho-Erk1/2 Protein Abundance in PrimaryHuman Airway Smooth Muscle Cells*

FBS (5%) Thrombin (0.3 U/ml)

Basal 1.0 6 0.0 1.0 6 0.01FBS/thrombin 9.6 6 8.0† 4.0 6 2.2†

1Dex (100 nM) 13.7 6 10.2 4.1 6 1.11ART (30 mM) 8.0 6 5.8 3.3 6 1.7

Definition of abbreviations: ART, artesunate; Dex, dexamethasone.*Quiescent cells were stimulated for 6 h with FBS or thrombin. Dex and ART were added 1 h beforestimulation. Data are expressed as fold increment over basal relative to total ERK1/2 proteinabundance.†P , 0.05 compared with basal (n = 4).

Figure 3. Immunoblot analysis showing the effect of ART on phospho-Akt (Ser473) (A) and phospho-p70S6K (Thr389) (B) protein expression in primary human ASM cells. Human ASM cells werestimulated for 20 hours with FBS (5%). Dex (100 nM) and ART (30 mM) were added 1 hour beforestimulation. Data are expressed as fold increment over basal (unstimulated cells) relative to totalprotein abundance. *P , 0.05 compared with basal. †P , 0.05 compared with FBS (n = 5–8).

ORIGINAL RESEARCH

454 American Journal of Respiratory Cell and Molecular Biology Volume 50 Number 2 | February 2014

of allergic asthma as previously described(13, 17). When compared with saline-challenged mice, ovalbumin (OVA)-sensitized and -challenged mice showed anincrease in the area of a-SMA–positivecells (Figures 5A and 5B). This increasewas significantly inhibited with artesunatepretreatment (Figures 5A and 5B).Moreover, compared with saline-challengedmice, mouse lung lysates from the OVA-challenge group showed increased cyclin D1

protein abundance, and this increase wassignificantly inhibited with artesunatepretreatment (Figure 5C).

Discussion

Using in vitro and in vivo approaches, weshowed for the first time that artesunateinhibited primary human cultured ASMcell number and decreased cyclin D1

protein levels in a non–mitogen-dependentmanner. This was evident in primaryhuman ASM cells derived fromnonasthmatic and asthmatic donors.Furthermore, artesunate, but notdexamethasone, decreased Akt and p70S6K

phosphorylation, suggesting that artesunateexerts antiproliferative effects in humanASM cells via actions on this pathway. Weshowed, in a murine model of allergicasthma, that artesunate treatment decreasedthe area of a-SMA–positive cells and cyclinD1 protein levels in the lungs of OVA-sensitized and -challenged mice.

The PI3K/Akt pathway and the ERKpathway work in parallel to mediateproliferation of ASM cells. We observed thatdexamethasone did not decrease the levelof Akt phosphorylation. In certain cell types,such as auditory hair cells, dexamethasoneactivated the PI3K/Akt pathway (29). Unlikedexamethasone, artesunate significantlyinhibited FBS-stimulated Akt and p70S6K

phosphorylation. This is supported byour previous studies, which showed thatartesunate blocked EGF-inducedphosphorylation of Akt and p70S6K innormal human bronchial epithelial cells (13).PI3K/Akt regulates p70S6K activity, whichin turn regulates cyclin D1 protein levels topromote ASM cell proliferation (26–28).Stewart and colleagues (22) have shown thatdexamethasone inhibits ASM proliferationby regulating cyclin D1 at the mRNA level.Taken together with current findingsindicating a lack of impact of dexamethasoneon PI3K, these observations are consistentwith the notion that dexamethasone worksvia mechanisms that are downstream ofPI3K/Akt. Thus, dexamethasone andartesunate appear to work through differentmechanisms to inhibit cell proliferation. Themechanism by which artesunate inhibitsthe PI3K/Akt/p70S6K pathway remains to befully elucidated. Cheng’s recent findings inmast cells suggest that Syk and phospholipaseC g are potential targets of artesunatethat are upstream of PI3K (30). Furtherinvestigation is warranted to determinewhether these two proteins upstream ofPI3K are involved in human ASM cellproliferation. Nonetheless, our study is thefirst to show that artesunate exerts inhibitoryeffects on the PI3K/Akt/p70S6K/cyclin D1pathway in human ASM cells in vitro andin vivo. This is favorable in asthma treatmentbecause it suggests that artesunate targets oneof the dominant pathways by which ASMcells proliferate, which is not targeted by

Figure 4. Effect of ART on primary human asthmatic ASM cell number and on cyclin D1 andphospho-Akt (Ser473) expression. (A) Quiescent cells were stimulated for 48 hours with FBS (5%) andcounted using trypan blue exclusion to identify nonviable cells. Dex (100 nM) and ART (3, 10, and30 mM) were added 1 hour before stimulation. Data are expressed as fold increment over basal(number of unstimulated cells). (B and C ) Cells were stimulated for 20 hours with FBS (5%). Data areexpressed as fold increment over basal (unstimulated cells) relative to b-actin protein abundance.*P , 0.05 compared with basal. †P , 0.05 compared with FBS (n = 3).

ORIGINAL RESEARCH

Tan, Ong, Cheng, et al.: Artesunate Inhibits ASM Cell Proliferation 455

Figure 5. Effect of ART in a murine model of allergic asthma. Female BALB/c mice were sensitized with ovalbumin (OVA) and challenged with salineaerosol (O/S), OVA aerosol (O/O) with or without DMSO, or ART (30 mg/kg). Lung tissues were collected 24 hours after last challenge. (A) Hematoxylin andeosin and immunohistochemistry staining of a-smooth muscle acrtin (a-SMA) were performed on the lung tissue sections. (B) Quantification of a-SMA–positive staining (1:100) in mouse lung tissues (O/S sections, n = 6; O/O sections, n = 5; DMSO and artesunate sections, n = 4). Data are expressed as

ORIGINAL RESEARCH

456 American Journal of Respiratory Cell and Molecular Biology Volume 50 Number 2 | February 2014

dexamethasone. Whether the effect ofa combined treatment of artesunate withdexamethasone would result in greaterantiproliferative effects on human ASM cellsremains to be investigated.

Although human ASM cell numberwas reduced with the pretreatment ofartesunate at concentrations below 30 mM,only 30 mM of artesunate was sufficientto reduce cyclin D1 protein abundance.We reason that, at low concentrationsof artesunate, other biomarkers of cellproliferation besides cyclin D1 may beaffected. In support of this hypothesis,Chen and colleagues showed thatderivatives of artemisinin, such asdihydroartemisinin, can down-regulatecyclin E and CDK2 and up-regulate theCDK inhibitor p27Kip1 (31). In normalhuman bronchial epithelial cells, artesunatealso inhibits cyclin D3 (unpublishedresults).

Dexamethasone did not inhibit FBS-and thrombin-stimulated ERK1/2phosphorylation. This is consistent with thework done by Fernandes and colleagues, whoshowed that glucocorticoids do not directlyaffect the phosphorylation of ERK1/2 (15).Similar to dexamethasone treatment,artesunate did not inhibit the effects of thesemitogens on ERK1/2 phosphorylation.Previous studies have found that artesunatealso had no effect on ERK phosphorylationin TNF-a– or hypoxia-stimulated humanrheumatoid arthritis fibroblast-likesynoviocytes (32). Although we cannotexclude the role of ERK1/2 in mediatingASM proliferation, the fact that inhibitionof cyclin D1 is associated with a significant

decrease in cell number in vitro and witha marked decrease in the total area ofa-SMA in vivo suggests a minor role forphospho-ERK1/2 in the regulation of ASMproliferation in this experimental model.

We previously demonstrated thatartesunate attenuated allergic airwayinflammation and reduced airwayhyperresponsiveness in a murine modelof allergic inflammation (13). Using thesame lung samples taken from this earlierstudy and as a proof-of-concept for theantiproliferative effects of artesunate in vivo,we now show that pretreatment of OVA-sensitized and -challenged mice withartesunate decreased the area of a-SMA–positive cells and cyclin D1 protein levels.It is possible that the reduction in area ofa-SMA and cyclin D1 protein in ASMcells could account for the reduction inairway reactivity that we previouslyreported (13). In further support, thelevels of phospho-Akt and phospho-p70S6K from OVA-sensitized and -challengedmice that were treated with artesunatewas also reduced relative to untreatedmice (13).

We have shown in our previousstudy that treatment of OVA-sensitizedand -challenged mice with artesunatesignificantly reduced eosinophil, neutrophil,and lymphocyte cell recruitment into theairways (13). Artesunate also reduced thelevels of proinflammatory cytokines andchemokines in the bronchoalveolar lavagefluid of these mice. Thus, we cannotexclude the possibility that artesunate maybe indirectly affecting ASM proliferationvia inhibiting inflammation. Future

time-course experiments to determinewhether changes in ASM mass areconsequential to reduction in inflammatorymarkers or vice versa are required.

Because therapies for airway diseasessuch as asthma are based on symptom relief,further understanding of how artesunatemodulates ASM function holds promise forthe design of more effective therapies thattarget the underlying cellular andmolecular mechanism(s) that governAWR. The finding that artesunate can haveantiproliferative properties via regulatingthe PI3K/Akt/p70S6K/cyclin D1 pathwayin vitro and in vivo may offer an additionaltherapeutic strategy in combination therapyfor use by patients who are insensitive toglucocorticoids due to an elevated PI3K/Akt signaling pathway (33). These groupsof patients who are glucocorticoidinsensitive (often seen in severe asthma)make up more than half the total healthcarecosts associated with asthma (34).Moreover, the findings from our study maybe applicable to other disease states whereASM proliferation may play a role, suchas in lymphangioleiomyomatosis, wherehyperproliferation of abnormal ASM cellsin the lungs of women of child-bearing ageare targets. Finally, given the establishedsafety and patient tolerability profile forartesunate in the treatment of malaria (35)and its low production cost, furtherresearch into its use for the treatment ofother diseases such as lung disorders iswarranted. n

Author disclosures are available with the textof this article at www.atsjournals.org.

References

1. Ebina M, Takahashi T, Chiba T, Motomiya M. Cellular hypertrophy andhyperplasia of airway smooth muscles underlying bronchial asthma:a 3-D morphometric study. Am Rev Respir Dis 1993;148:720–726.

2. Lambert RK, Wiggs BR, Kuwano K, Hogg JC, Pare PD. Functionalsignificance of increased airway smooth muscle in asthma and COPD.J Appl Physiol 1993;74:2771–2781.

3. Hirst SJ, Lee TH. Airway smooth muscle as a target of glucocorticoidaction in the treatment of asthma. Am J Respir Crit Care Med 1998;158:S201–S206.

4. Stewart AG, Tomlinson PR, Wilson JW. Regulation of airway wallremodeling: prospects for the development of novel antiasthmadrugs. Adv Pharmacol 1995;33:209–253.

5. Sotomayor H, Badier M, Vervloet D, Orehek J. Seasonal increase ofcarbachol airway responsiveness in patients allergic to grasspollen: reversal by corticosteroids. Am Rev Respir Dis 1984;130:56–58.

6. Jeffery PK, Godfrey RW, Adelroth E, Nelson F, Rogers A, JohanssonSA. Effects of treatment on airway inflammation and thickening ofbasement membrane reticular collagen in asthma: a quantitative lightand electron microscopic study. Am Rev Respir Dis 1992;145:890–899.

7. Nash JJ, Nash AG, Leach ME, Poetker DM. Medical malpractice andcorticosteroid use. Otolaryngol Head Neck Surg 2011;144:10–15.

8. Poetker DM, Reh DD. A comprehensive review of the adverse effects ofsystemic corticosteroids. Otolaryngol Clin North Am 2010;43:753–768.

Figure 5. (Continued). fold increment over O/S samples. BM = basement membrane. *P , 0.05 compared with O/S. †P , 0.05 compared with O/O.Digital photographs of lung sections were analyzed at a magnification of 203. (C ) Immunoblot analysis showing the effect of ART on cyclin D1 proteinexpression in mouse lung tissue. Data are expressed as fold increment over O/S samples relative to b-actin protein abundance. *P , 0.05 comparedwith O/S. †P , 0.05 compared with vehicle control DMSO (n = 3).

ORIGINAL RESEARCH

Tan, Ong, Cheng, et al.: Artesunate Inhibits ASM Cell Proliferation 457

9. Krishna S, Bustamante L, Haynes RK, Staines HM. Artemisinins: theirgrowing importance in medicine. Trends Pharmacol Sci 2008;29:520–527.

10. Efferth T, Kaina B. Toxicity of the antimalarial artemisinin and itsdervatives. Crit Rev Toxicol 2010;40:405–421.

11. Xu Q, Li ZX, Peng HQ, Sun ZW, Cheng RL, Ye ZM, Li WX. Artesunateinhibits growth and induces apoptosis in human osteosarcomaHOS cell line in vitro and in vivo. J Zhejiang Univ Sci B 2011;12:247–255.

12. Li S, Xue F, Cheng Z, Yang X, Wang S, Geng F, Pan L. Effect ofartesunate on inhibiting proliferation and inducing apoptosis of SP2/0 myeloma cells through affecting NFkappaB p65. Int J Hematol2009;90:513–521.

13. Cheng C, Ho WE, Goh FY, Guan SP, Kong LR, Lai WQ, Leung BP,Wong WS. Anti-malarial drug artesunate attenuates experimentalallergic asthma via inhibition of the phosphoinositide 3-kinase/Aktpathway. PLoS ONE 2011;6:e20932.

14. Tran T, Stewart AG. Protease-activated receptor (PAR)-independentgrowth and pro-inflammatory actions of thrombin on human culturedairway smooth muscle. Br J Pharmacol 2003;138:865–875.

15. Fernandes D, Guida E, Koutsoubos V, Harris T, Vadiveloo P, WilsonJW, Stewart AG. Glucocorticoids inhibit proliferation, cyclin D1expression, and retinoblastoma protein phosphorylation, but notactivity of the extracellular-regulated kinases in human culturedairway smooth muscle. Am J Respir Cell Mol Biol 1999;21:77–88.

16. Tran T, Teoh CM, Tam JK, Qiao Y, Chin CY, Chong OK, Stewart AG,Harris T, Wong WS, Guan SP, et al. Laminin drives survival signals topromote a contractile smooth muscle phenotype and airwayhyperreactivity. FASEB J 2013;27:3991–4003.

17. Bao Z, Guan S, Cheng C, Wu S, Wong SH, Kemeny DM, Leung BP,Wong WS. A novel antiinflammatory role for andrographolide inasthma via inhibition of the nuclear factor-kappaB pathway. Am JRespir Crit Care Med 2009;179:657–665.

18. Robinson PJ, Hegele RG, Schellenberg RR. Increased airway reactivityin human RSV bronchiolitis in the guinea pig is not due to increasedwall thickness. Pediatr Pulmonol 1996;22:248–254.

19. Vlahos R, Lee KS, Guida E, Fernandes DJ, Wilson JW, Stewart AG.Differential inhibition of thrombin- and EGF-stimulated humancultured airway smooth muscle proliferation by glucocorticoids.Pulm Pharmacol Ther 2003;16:171–180.

20. Ravenhall C, Guida E, Harris T, Koutsoubos V, Stewart A. Theimportance of ERK activity in the regulation of cyclin D1 levels andDNA synthesis in human cultured airway smooth muscle. Br JPharmacol 2000;131:17–28.

21. Maddika S, Ande SR, Wiechec E, Hansen LL, Wesselborg S, Los M.Akt-mediated phosphorylation of CDK2 regulates its dual role incell cycle progression and apoptosis. J Cell Sci 2008;121:979–988.

22. Stewart AG, Harris T, Fernandes DJ, Schachte LC, Koutsoubos V,Guida E, Ravenhall CE, Vadiveloo P, Wilson JW. Beta2-adrenergic receptor agonists and cAMP arrest human culturedairway smooth muscle cells in the G(1) phase of the cell cycle:role of proteasome degradation of cyclin D1. Mol Pharmacol1999;56:1079–1086.

23. Orsini MJ, Krymskaya VP, Eszterhas AJ, Benovic JL, Panettieri RA Jr,Penn RB. MAPK superfamily activation in human airway smoothmuscle: mitogenesis requires prolonged p42/p44 activation. Am JPhysiol 1999;277:L479–L488.

24. Burgess JK, Lee JH, Ge Q, Ramsay EE, Poniris MH, Parmentier J, RothM, Johnson PR, Hunt NH, Black JL, et al. Dual ERK andphosphatidylinositol 3-kinase pathways control airway smooth muscleproliferation: differences in asthma. J Cell Physiol 2008;216:673–679.

25. Page K, Li J, Wang Y, Kartha S, Pestell RG, Hershenson MB.Regulation of cyclin D(1) expression and DNA synthesis byphosphatidylinositol 3-kinase in airway smooth muscle cells. Am JRespir Cell Mol Biol 2000;23:436–443.

26. Krymskaya VP, Penn RB, Orsini MJ, Scott PH, Plevin RJ, Walker TR,Eszterhas AJ, Amrani Y, Chilvers ER, Panettieri RA Jr.Phosphatidylinositol 3-kinase mediates mitogen-induced human airwaysmooth muscle cell proliferation. Am J Physiol 1999;277:L65–L78.

27. Scott PH, Belham CM, al-Hafidh J, Chilvers ER, Peacock AJ, GouldGW, Plevin R. A regulatory role for cAMP in phosphatidylinositol3-kinase/p70 ribosomal S6 kinase-mediated DNA synthesis inplatelet-derived-growth-factor-stimulated bovine airway smooth-muscle cells. Biochem J 1996;318:965–971.

28. Hashemolhosseini S, Nagamine Y, Morley SJ, Desrivieres S, Mercep L,Ferrari S. Rapamycin inhibition of the G1 to S transition is mediatedby effects on cyclin D1 mRNA and protein stability. J Biol Chem1998;273:14424–14429.

29. Haake SM, Dinh CT, Chen S, Eshraghi AA, Van De Water TR.Dexamethasone protects auditory hair cells against TNFalpha-initiated apoptosis via activation of PI3K/Akt and NFkappaBsignaling. Hear Res 2009;255:22–32.

30. Cheng C, Ng DS, Chan TK, Guan SP, Ho WE, Koh AH, Bian JS, LauHY, Wong WS. Anti-allergic action of anti-malarial drug artesunate inexperimental mast cell-mediated anaphylactic models. Allergy 2013;68:195–203.

31. Chen H, Sun B, Wang S, Pan S, Gao Y, Bai X, Xue D. Growth inhibitoryeffects of dihydroartemisinin on pancreatic cancer cells: involvementof cell cycle arrest and inactivation of nuclear factor-kappaB.J Cancer Res Clin Oncol 2010;136:897–903.

32. He Y, Fan J, Lin H, Yang X, Ye Y, Liang L, Zhan Z, Dong X, Sun L, Xu H.The anti-malaria agent artesunate inhibits expression of vascularendothelial growth factor and hypoxia-inducible factor-1a in humanrheumatoid arthritis fibroblast-like synoviocyte. Rheumatol Int 2011;31:53–60.

33. Marwick JA, Adcock IM, Chung KF. Overcoming reduced glucocorticoidsensitivity in airway disease: molecular mechanisms and therapeuticapproaches. Drugs 2010;70:929–948.

34. Chung KF, Godard P, Adelroth E, Ayres J, Barnes N, Barnes P, Bel E,Burney P, Chanez P, Connett G, et al.; European RespiratorySociety. Difficult/therapy-resistant asthma: the need for anintegrated approach to define clinical phenotypes, evaluate riskfactors, understand pathophysiology and find novel therapies. ERSTask Force on Difficult/Therapy-Resistant Asthma. Eur Respir J1999;13:1198–1208.

35. Gordi T, Lepist EI. Artemisinin derivatives: toxic for laboratory animals,safe for humans? Toxicol Lett 2004;147:99–107.

ORIGINAL RESEARCH

458 American Journal of Respiratory Cell and Molecular Biology Volume 50 Number 2 | February 2014