1200 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212DOI: 10.1002/eji.201343629

Boswellic acids reduce Th17 differentiation viablockade of IL-1β-mediated IRAK1 signaling

Klarissa Hanja Sturner1, Nina Verse1, Sara Yousef1, Roland Martin∗1,2

and Mireia Sospedra∗1,2

1 Institute for Neuroimmunology and Clinical Multiple Sclerosis Research (INIMS) and Clinicfor Neurology, Center for Molecular Neurobiology, Universitatsklinikum HamburgEppendorf, Hamburg, Germany

2 Neuroimmunology and MS Research, Department of Neurology, University Hospital Zurich,Zurich, Switzerland

Interferon-gamma producing CD4+ T (Th1) cells and IL-17-producing CD4+ T (Th17) cellsare involved in the pathogenesis of several autoimmune diseases including multiple scle-rosis. Therefore, the development of treatment strategies controlling the generation andexpansion of these effector cells is of high interest. Frankincense, the resin from trees ofthe genus Boswellia, and particularly its prominent bioactive compound acetyl-11-keto-β-boswellic acid (AKBA), have potent anti-inflammatory properties. Here, we demonstratethat AKBA is able to reduce the differentiation of human CD4+ T cells to Th17 cells, whileslightly increasing Th2- and Treg-cell differentiation. Furthermore, AKBA reduces theIL-1β-triggered IL-17A release of memory Th17 cells. AKBA may affect IL-1β signaling bypreventing IL-1 receptor-associated kinase 1 phosphorylation and subsequently decreas-ing STAT3 phosphorylation at Ser727, which is required for Th17-cell differentiation. Theeffects of AKBA on Th17 differentiation and IL-17A release make the compound a goodcandidate for potential treatment of Th17-driven diseases.

Keywords: AKBA � boswellic acids � Th17 cells

� Additional supporting information may be found in the online version of this article at thepublisher’s web-site

Introduction

IFN-γ producing CD4+ T (Th1) cells and IL-17-producing CD4+ T(Th17) cells mediate tissue damage and inflammation in severalanimal models of autoimmune diseases such as EAE, collagen-induced arthritis and experimental colitis [1–3]. In addition,involvement of these cells in human autoimmune diseases includ-ing multiple sclerosis (MS) [4–6], rheumatoid arthritis [7], andCrohn’s disease [8, 9] is supported by the accumulation and upreg-ulation of Th1- and Th17 signature cytokines (IFN-γ and IL-17)

Correspondence: Dr. Mireia Sospedrae-mail: Mireia.SospedraRamos@usz.ch

in the target tissues. Treatment strategies controlling the gener-ation and expansion of these effector cells could be very usefultherapeutically.

Th17 cells are generated under different polarizing conditionsthan Th1 cells and, although polarization of human Th17 cells hasbeen extensively investigated, the differentiation requirements arestill incompletely defined. It has been reported that the coordi-nated activation of Smad 3 and STAT3 induce the transcriptionfactor RORγt necessary for Th17-cell differentiation [10]. Full acti-vation of STAT3 requires phosphorylation of Tyrosine 705 (Y705)

∗These authors contributed equally to this work.

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1201

and Serine 727 (S727) sites and these phosphorylations are con-trolled by different kinases. Janus Kinase 2 (JAK2) controls phos-phorylation of Y705 [11], while phosphorylation of S727 is con-trolled by IL-1 receptor-associated kinase 1 (IRAK1) [12]. IL-6and IL-1β are two cytokines reported to be essential for humanTh17-cell differentiation [13, 14]. Th17-cell activation requiresfull activation of STAT3 by phosphorylation, and IL-6 and IL-1β

both play roles in this process: IL-6 phosphorylates Y705 via JAK2and IL-1β phosphorylates S727 via IRAK1 [15].

Frankincense, the resin from trees of the genus Boswellia, hasbeen used as an oral anti-inflammatory agent in traditional East-ern and Asian medicine for thousands of years [16]. Boswellicacids (BAs) are the active pharmacological ingredients of frank-incense. The clinical efficacy, safety, and tolerability of BAs hasbeen shown in several pilot randomized clinical trials of differentdiseases including asthma, rheumatoid arthritis, Crohn’s disease,osteoarthritis, and collagenous colitis [17]. Among all describedBAs, acetyl-11-keto-β-boswellic acid (AKBA) has been identified asthe prominent bioactive compound [18]. AKBA possesses potentanti-inflammatory properties in vitro by interfering with signal-ing pathways such as the nuclear factor-κB route [19] and Ca2+

signaling [20] as well as by inhibiting several enzymes including5-lipoxygenase [21], platelet-type 12 lipoxygenase [22], humanleukocyte elastase [23], cathepsin G [24], cyclooxygenase-1 [25],and microsomal prostaglandin E2 synthase-1 (mPGES1) [26]. Inaddition, it has been reported that AKBA is able to suppress IL-6-induced STAT3 activation by inhibition of JAK2 phosphorylationand phosphorylation of STAT3 at Y705 in multiple myeloma cells[27]. In this study, we examine the possibility that AKBA interfereswith Th17 differentiation due to its effect on STAT3 activation.Our data indicate that AKBA is able to decrease Th17 differen-tiation by inhibition of IL-1β signaling via reduction of IRAK1phosphorylation, and AKBA also slightly reduces IL-17A releaseby memory Th17 cells. We did not observe an effect of AKBA onIL-6-induced STAT3 activation in CD4+ T cells.

Results

AKBA inhibits the polarization of Th17 CD4+ T cellsin vitro

In order to study the effect of AKBA on Th17-cell differentiation,highly purified naıve CD4+ T cells from healthy donors (HDs)were stimulated in vitro with beads in the presence of a Th17-polarizing cocktail (see Materials and methods) and in the pres-ence or absence of different concentrations of AKBA. After 12days in culture, intracellular production of IL-17A and IFN-γ wereexamined by flow cytometry. We found that AKBA reduced orcompletely abrogated Th17-cell differentiation in a concentration-dependent manner (Fig. 1A and B). 27.79 ± 2.8 (mean ± SD)%of naıve CD4+ T cells stimulated with beads in the presence ofthe Th17-polarizing cocktail produced exclusively IL-17A and 10± 2.2% produced both IL-17A and IFN-γ. The presence of AKBAat 1 and 5 μM during polarization decreased the differentiation

of both IL-17A- and IL-17A/IFN-γ-producing cells (Fig. 1A and B).The ability of AKBA to reduce or abolish Th17 polarization wasalso observed when purified naıve CD4+ T cells were preincubatedwith AKBA for 15 h, washed and then stimulated with beads inpresence of the Th17 polarization cocktail (Fig. 1B). The inhibitoryeffect of AKBA on Th17 differentiation was confirmed using a flowcytometry-based 13-plex ELISA system (Fig. 1C) and by quan-tifying the transcription factor RORγt (Fig. 1D). The presenceof AKBA during polarization and the preincubation with AKBAbefore starting Th17 polarization both significantly reduced thesecretion of IL-17A as well as the expression of RORγt mRNA.Although high doses of AKBA (>20 μM in vitro) were toxic forhuman leukocytes, the effects on Th17 differentiation induced byAKBA at the concentrations used in this study were not due toan increased death rate of AKBA-treated CD4+ T cells nor due toinhibition of T cell proliferation (Supporting Information Fig. 1Aand 1B).

As classical Th17 cells, Th17-polarized T cells in our culture sys-tem expressed the surface markers CD161, CCR6, and IL-23 recep-tor (IL-23R) (Fig. 2A). As expected, the presence of AKBA duringor before polarization significantly reduced the frequency of CD4+

CD161+ CCR6+ T cells as well as CD4+ IL-23R+ T cells (Fig. 2A).Further characterization of Th17-polarized T cells treated or pre-treated with AKBA demonstrated that AKBA does not affect acti-vation neither apoptosis since the frequency of CD4+ CD69+

(Fig. 2B) and annexin-V+ (Fig. 2C) T cells as well as the expres-sion of the antiapoptotic Bcl-2 and proapoptotic CD95 markers(Fig. 2D) were unchanged.

Next, we examined the effect of AKBA on Th17 differentiationusing an identical approach but with highly purified naıve CD4+

T cells from relapsing-remitting MS (RRMS) patients. The resultsare summarized in Fig. 3. We found that incubation or preincuba-tion with AKBA reduced or completely abrogated Th17 differentia-tion in naıve CD4+ T cells from RRMS patients. 25.86 ± 7.2 (mean± SD)% of naıve CD4+ T cells stimulated with beads in the pres-ence of the Th17-polarizing cocktail produced exclusively IL-17Aand 12.42 ± 4.3% produced both IL-17A and IFN-γ. The pres-ence of AKBA during polarization or during a 15 h preincubationdecreased the differentiation of both IL-17A- and IL-17A/IFN-γproducing cells (Fig. 3A). Accordingly, incubation or preincuba-tion with AKBA also significantly reduced the frequency of CD161+

CCR6+ as well as IL-23R+ CD4+ T cells (Fig. 3B).

Effect of AKBA on in vitro differentiation of Th1, Th2,and Treg cells

Preincubation or incubation with AKBA of naıve CD4+ T cellsstimulated with beads under Th17-polarizing conditions does notseem to affect Th1 differentiation, since no effect of AKBA onthe frequency of CD4+ IFN-γ-producing T cells, release of IFN-γnor expression of t-bet mRNA has been observed (Fig. 4A).The fact that AKBA does not affect IFN-γ release in super-natants is unexpected since AKBA significantly abrogated differ-entiation of IL-17A+IFN-γ+CD4+ T cells under Th17-polarizing

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1202 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

Figure 1. AKBA affects the Th17 differen-tiation of naıve human CD4+ T cells fromHDs stimulated under Th17-polarizing con-ditions. (A) Representative flow cytometricanalysis of intracellular IL-17A and IFN-γproduction by naive human CD4+ T cellsfrom a HD stimulated in vitro for 12 daysin Th17-polarizing conditions and in theabsence (left) or presence (right) of AKBA.Numbers in each compartment representthe percentage of positive cells. (B) The per-centage of IL-17A+ (top) and IL-17A+ IFN-γ+

(bottom) CD4+ T cells pretreated with AKBAor not, and stimulated in vitro for 12 days inTh17-polarizing conditions in the presenceor absence of AKBA, are shown. Each dotrepresents a healthy donor. Mean ± SEMand statistical significance are shown.**p < 0.01, ***p < 0.001, ****p < 0.0001,one-way ANOVA with Bonferroni’s correc-tion for multiple comparisons. (C) IL-17Acytokine levels measured in culture super-natants using a cytofluorimetry-based 13-plex ELISA. Data are shown as mean + SEMof eight HDs. (D) Expression of transcriptionfactor RORγt mRNA was analyzed by RT-PCR and shown as mean + SEM of sevenHDs. Values are relative expression com-pared to untreated Th17-polarized CD4+

T cells (calibrator = 1). ****p < 0.0001, ***p <

0.001, RM one-way ANOVA with Dunnett’scorrection for multiple comparisons with asingle control.

conditions. One explanation for these findings might be that singleIFN-γ−producing cells produce more IFN-γ than double IL-17A-IFN-γ−producing cells. IFN-γ expression measured as median flu-orescence intensity (MedFI) in naıve CD4+ T cells stimulated invitro with beads in the presence of a Th17-polarizing cocktail wasslightly higher in single IFN-γ−producing cells than in cells pro-ducing both cytokines, i.e. IL-17A-IFN-γ-double producing cellsand even higher in single IFN-γ−producing cells treated or pre-treated with AKBA during polarization, i.e. IFN-γ−producing cellsthat had been expanded and stimulated in absence of IL-17A (Sup-porting Information Fig. 2). AKBA might also affect the differentia-tion of naıve CD4+ T cells to Th2 and Treg cells in Th17-polarizingconditions. Incubation and to a lesser extent preincubation with

AKBA both tended to increase the frequency of IL-4-producing cellsas well as the release of IL-4 and the expression of GATA-3 mRNA,but these changes did not reach statistical significance (Fig. 4B).Regarding Treg cells differentiation, incubation and to a lesserextent preincubation with AKBA also tended to increase the fre-quency of IL-10-producing cells, and incubation with AKBA signif-icantly increased the release of IL-10 and the expression of Foxp3mRNA (Fig. 4C). Supporting these results, we found that preincu-bation or incubation with AKBA of naıve CD4+ T cells stimulatedwith beads in the presence of a Th17-polarizing cocktail tended toreduce the production of IL-22, to have no effect on TNF-α, andto slightly increase IL-5 and IL-13, although these differences didnot reach significance (Supporting Information Fig. 3).

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1203

Figure 2. Characterization of polarized Th17 cells. Donor cells were either pretreated or not with AKBA and stimulated in vitro for 12 days inTh17-polarizing conditions in the presence or absence of AKBA. The frequency (%) of (A) CD4+ CD161+ CCR6+ (left), CD4+ IL-23R+ (right), (B) CD4+

CD69+, and (C) CD4+ Annexin V+ cells was then measured by flow cytometry. Each dot represents single HD and bars represent mean ± SEM(n = 8). (D) Bcl-2 and CD95 expression (MedFI) was measured in naıve CD4+ T cells pretreated or not with AKBA and stimulated in vitro for 12 daysin Th17-polarizing conditions in the presence or absence of AKBA, and shown as mean + SEM of eight HDs. ****p < 0.0001, **p < 0.01, RM one-wayANOVA with Dunnett’s correction for multiple comparisons with a single control.

Stimulation of naıve CD4+ T cells under Th17-polarizingconditions was not optimal for Th1, Th2, or Treg differentiation,so we decided to confirm the lack of effect of AKBA on Th1 differ-entiation and the slight induction of Th2 and Treg differentiation.Therefore, we in vitro stimulated AKBA-pretreated or untreatedhighly purified naıve CD4+ T cells with beads in the presence of aTh1-, Th2-, or Treg-polarizing cocktail (see Methods) and in thepresence or absence of AKBA. We then quantified the frequencyof CD4+ IFN-γ-, IL-4-, and IL-10-producing cells, the release ofIFN-γ, IL-4, and IL-10, and the expression of t-bet, GATA-3, andFoxP3 mRNAs. Expression of additional markers for each cell typewas also analyzed, including IL-12R for Th1, CRTH2 and CCR4 forTh2, and CD127 and Foxp3 for Treg. 56.2 ± 2.5% of naıve CD4+

T cells stimulated with beads in presence of the Th1-polarizingcocktail produced IFN-γ, 14 ± 4.8% stimulated in presence of the

Th2-polarizing cocktail produced IL-4, and 22.6 ± 6.7% stimu-lated in presence of the Treg-polarizing cocktail produced IL-10(Fig. 5). Regarding the effect of AKBA on Th1 differentiation,neither the presence of AKBA during polarization nor a 15 hpreincubation period significantly affected the frequency of CD4+

IFN-γ-producing cells or the quantity of released IFN-γ (Fig. 5A).Interestingly, the presence of AKBA during Th1 polarizationsignificantly reduced the expression of t-bet and IL-12R mRNA(Fig. 5A). After Th1 polarization, no IL-17-producing cells anda very low percentage of IL-4 and IL-10-producing cells werefound and any significant effect of AKBA was observed in thefrequency of these cells and neither on RORγt, GATA-3, nor FoxP3mRNA expression (Supporting Information Fig. 4). AKBA appearsto influence Th2-cell and Treg-cell differentiation, since thepresence of AKBA during polarization and a 15 h preincubation

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1204 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

Figure 3. AKBA affects Th17 differentia-tion of naıve human CD4+ T cells from MSpatients stimulated under Th17-polarizingconditions. The frequency (%) of (A) IL-17A+

(left), IL-17A+ IFN-γ+ (right), (B) CD161+

CCR6+ (left), and IL-23R+ (right) CD4+ T cellspretreated or not with AKBA and stimulatedin vitro for 12 days in Th17-polarizing con-ditions in the presence or absence of AKBAwas measured by flow cytometry. Each dotrepresents a MS patient (n = 5) and bars rep-resent meant + SEM. ****p < 0.0001, RM one-way ANOVA with Dunnett’s correction formultiple comparisons with a single control.

period significantly affected the frequency of IL-4-producing,IL-10-producing, CRTh2+ CCR4+ and CD127lowFoxp3+ CD4+

T cells, as well as the quantity of released IL-5 and IL-10 and theexpression of GATA-3 and Foxp3 mRNA (Fig. 5B and 5C). SinceIL-4 was a component of the Th2 polarizing cocktail, IL-5 wasmeasured instead of IL-4. The Th2 polarization protocol usedrequired that the polarization cocktail was added on day 0, 3, 5, 7,and 10, which diluted putative cytokine release by the cells duringpartial removal of the culture medium. This dilution most likelyunderlies the low levels of IL-5 measured in the supernatants.The differentiation of other Th phenotypes under Th2 and Tregpolarizing conditions was very low (data not shown).

AKBA inhibits IL-1β-induced IRAK1 phosphorylationand IL-1β-induced STAT3 phosphorylation at Ser727

As mentioned above, Th17 differentiation induced by IL-6 andIL-1β most likely requires full activation of STAT3 with phospho-rylation at Y705 and S727. Since it has been reported that AKBAsuppresses IL-6-induced STAT3 activation in multiple myelomacells by inhibition of JAK2 phosphorylation and, consequently,of STAT3 phosphorylation at Y705 (27), we examined whetherAKBA may inhibit Th17 differentiation of CD4+ T cells bythe same mechanism. CD4+ T cells were stimulated with IL-6in presence or absence of AKBA (1 and 5 μM) and STAT3phosphorylation at Y705 was measured. Unexpectedly, AKBA didnot reduce phosphorylation of STAT3 at Y705 after stimulationwith IL-6 in CD4+ T cells (Fig. 6A). However, it is important tonote that the concentration of AKBA that was able to suppressIL-6-induced STAT3 phosphorylation at Y705 in the prior studieswas 25 μM, i.e. 5 times higher than the highest concentration weused. As mentioned above, concentrations of AKBA higher than20 μM are toxic for CD4+ T cells (Supporting Information Fig. 1Aand 1B).

Next, we examined whether AKBA may inhibit Th17 differ-entiation by interfering in STAT3 activation mediated by IL-1β

signaling. CD4+ T cells were stimulated with IL-1β in presence orabsence of AKBA (1 and 5 μM), and STAT3 phosphorylation atS727 was measured. We observed that IL-1β-induced phosphoryla-tion of STAT3 at S727 was significantly reduced by treatment with5 μM AKBA (Fig. 6A). We also analyzed whether AKBA reducedphosphorylation of IRAK1, the kinase responsible for phosphory-lation of STAT3 at S727, in CD4+ T cells stimulated with IL-1β.Initial Western blot analyses suggested lower phosphorylation ofIRAK1 pS376 in the presence of AKBA, although results werenot completely convincing (Supporting Information Fig. 5). Amuch clearer effect of AKBA on reducing IRAK1 phosphorylationin CD4+ T cells stimulated with IL-1β was observed using flowcytometric analysis. AKBA treatment significantly reduced IL-1β-induced phosphorylation of IRAK1 at pS376 and showed a reduc-tion of IRAK1 phosphorylation at pT387 without reaching statisti-cal significance (Fig. 6B). Finally, the effect of treatment and pre-treatment with AKBA on JAK2 phosphorylation at pY1007/Y1008,STAT3 at pY705 and pS727, and IRAK1 at pS376 was measured innaıve CD4+ T cells stimulated under Th17-polarizing conditionsfor 12 days (Fig. 6C). AKBA significantly reduced STAT3 phospho-rylation at pS727, and, although statistical significance was notreached, also IRAK1 at pS376. JAK2 and STAT3 phosphorylationat pY705 were not reduced by AKBA treatment or pretreatment,and unexpectedly the treatment with 5 μM AKBA slightly inducedJAK2 phosphorylation at pY1007/Y1008.

Effect of AKBA on differentiated Th17 CD4+ T cells

Next, we examined whether AKBA had an effect on IL-17A pro-duction by differentiated Th17 cells. Highly purified memoryCD45RO+ CD4+ T cells (purity > 95% measured by flow cytom-etry) from HDs and RRMS patients were stimulated in vitro

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1205

Figure 4. AKBA affects differentiation of naıve human CD4+ T cells stimulated under Th17-polarizing conditions. Naıve CD4+ T cells from HDswere pretreated or not with AKBA and stimulated in vitro for 12 days in presence or absence of AKBA and under Th17-polarizing conditions. Thepercentage of (A) IFN-γ+, (B) IL-4+, and (C) IL-10+ producing CD4+ T cells are shown (left). Each dot represents a single HD (n = 8) and bars representmean + SEM. The cytokine levels of (A) IFN-γ, (B) IL-4, and (C) IL-10 in culture supernatants were measured using a cytofluorimetry-based 13-plexELISA (middle). Data are shown as mean ± SEM of eight HDs. RT-PCR analysis was used to measure expression of transcription factors (A) t-bet,(B) GATA3, and (C) Foxp3 (right). Data are expressed as the relative expression compared to untreated Th17-polarized CD4+ T cells (calibrator = 1)and are shown as mean ± SEM of 7, HDs. *p < 0.05, ****p < 0.0001, RM one-way ANOVA with Dunnett’s correction for multiple comparisons with asingle control.

using anti-CD3, anti-CD2, anti-CD28-coated beads and either IL-1β alone or in combination with IL-23. Cells were then culturedfor 72 h in presence or absence of AKBA and IL-17A release wasmeasured in the supernatants. The presence of 5 μM AKBA duringstimulation with beads and IL1β significantly reduced the releaseof IL-17A by memory T cells (Fig. 7A). The release of IL-17A underthis stimulatory condition was higher in RRMS patients comparedwith HDs but differences did not reach statistical significance. Theaddition of IL-23 significantly increased the release of IL-17A inRRMS and only slightly in HDs. Under these stimulatory condi-tions, RRMS patients released significantly higher levels of IL-17Athan HDs. Interestingly, IL-23 abrogated AKBA’s effect on IL-1β-mediated IL-17A release in memory Th17 cells from HDs but itsignificantly enhanced AKBA’s effect in cells from RRMS patients(Fig. 7A).

Finally, we assessed whether AKBA had an effect on the expan-sion of differentiated Th17 cells. Fully differentiated memory Th17cell lines were generated from the cerebrospinal fluid (CSF) ofan untreated RRMS patient following several rounds of stimula-tion with PHA in the presence of IL-2 and IL-23. These differen-tiated Th17 cells were restimulated in vitro with PHA, and IL-2,IL-23, and AKBA were added at day 0 and every 2/3 days for12 days. After these 12 days intracellular IL-17A and IFN-γ pro-

duction were measured by flow cytometry. We did not observeany effect of AKBA on the expansion of these differentiated Th17cells (Fig. 7B). Since pure differentiated Th17 cells were not avail-able and the culture included Th1 cells capable of releasing Th1cytokines, IL-23 was added to the cultures to sustain Th17 differ-entiation. The addition of this cytokine might be the reason whywe did not see an effect of AKBA on the expansion of IL-17A+

CD4+ T cells.

Discussion

Th17 cells are considered relevant for several human autoimmunediseases including MS [4–6], rheumatoid arthritis [7], and Crohn’sdisease [8, 9]. Therefore, treatment strategies controlling Th17cell generation and expansion are of great interest. Here, we reportthat AKBA, a prominent bioactive compound of BAs, exerts impor-tant anti-inflammatory effects by inhibiting the in vitro differenti-ation of naıve human CD4+ T cells from HDs and RRMS patientsinto Th17 cells. Additionally, AKBA promotes the differentiationof Th2 and Treg cells without a clear effect on Th1 differentiation.AKBA reduced the differentiation of both IL-17A- and IL-17A/IFN-γ-producing cells, which is especially relevant in the context

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1206 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

Figure 5. AKBA affects Th1-, Th2-, and Treg-cell differentiation. Naıve CD4+ T cells from HDs were either pretreated or not with AKBA andstimulated in vitro for 12 days in presence or absence of AKBA and under (A) Th1-, (B) Th2-, and (C) Treg-polarizing conditions. The percentageof (A) IFN-γ+, (B) IL-4+, and (C) IL-10+ producing CD4+ T cells is shown (left). Each dot represents a single HD bars represent mean ± SEM. Thecytokine levels of (A) IFN-γ, (B) IL-4, and (C) IL-10 in culture supernatants were measured by ELISA (middle left). Data are shown as mean ± SEMof five HDs. RT-PCR analysis was used to measure expression of transcription factors (A) t-bet, (B) GATA3, and (C) Foxp3 (middle right) and alsoIL-12R mRNA (a right graph). Data are expressed as the relative expression compared to untreated polarized CD4+ T cells (calibrator = 1) andare shown as mean ± SEM of five HDs. The percentage of CRTh2+ CCR4+ (B) and CD127low Foxp3+ (C) CD4+ T cells (right) was analyzed by flowcytometry. Each dot represents a single HD and bars represent mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, RM one-way ANOVAwith Dunnett’s correction for multiple comparisons with a single control.

of MS since IL17A/IFN-γ-producing cells seem to be relevant in MSpathogenesis [28]. AKBA decreased Th17 differentiation not onlyduring the polarization process but also, and more surprisingly,when purified naıve CD4+ T cells were preincubated with AKBAfor 15 h, washed and then stimulated with beads in presence ofthe Th17 polarization cocktail. The fact that Th17 differentiationwas significantly reduced not only by the presence of AKBA dur-ing the differentiation process but also by a preincubation withAKBA is interesting from a therapeutic point of view. AKBA wasalso able to slightly reduce the IL-1β-mediated release of IL-17Aby memory CD4+ T cells from both HDs and RRMS patients. IL-1β-mediated IL-17A release by memory Th17 cells was higher in cellsfrom RRMS compared with HDs, and these differences were signif-icantly enhanced by IL-23. Interestingly, IL-23 abrogated AKBA’seffect on IL-1β-mediated IL-17A release in memory Th17 cells fromHDs but significantly enhanced the AKBA effect in cells from RRMSpatients. These data suggest that memory Th17 cells from RRMSmight be more sensitive to IL-23 produced by activated dendriticcells and macrophages, and in this context, the AKBA-mediatedreduction in IL-17A release might have also important additionaltherapeutic advantages. IL-17A is involved in the development ofautoimmunity and inflammation, but also plays an important role

in the host defense against bacterial and fungal infections. The factthat AKBA reduces IL-17A release by memory Th17 cells mightcompromise the protective functions of these cells during immuneresponses against pathogens but also of putative-differentiatedautoreactive Th17 cells.

As mentioned above, AKBA did not show a clear effect on Th1-cell differentiation. However, as shown in the results section, sin-gle IFN-γ−producing cells treated or pretreated with AKBA duringTh17 polarization express higher levels of IFN-γ compared withdouble IL-17A- IFN-γ- or even single IFN-γ-producing untreatedcells. These results suggest a negative regulatory role of IL-17Aon IFN-γ production as has been reported in some studies in thetumor context [29]. Since the frequency of Th17 cells in MS infil-trates is not very high, we do not expect that AKBA treatmentwill result in a significant increase in IFN-γ production by Th1cells. However, since Th1 cells are involved in the pathogenesisof MS, changes in IFN-γ level should be closely monitored in anytherapeutic approach involving AKBA.

Unexpectedly, the ability of AKBA to significantly reduceIL-17A production by Th17 T cells did not extend to other Th17cytokines such as IL-22. Although AKBA decreased IL-22 release byTh17 cells, this reduction did not reach statistical significance. It

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1207

Figure 6. AKBA affects STAT3, IRAK1, and JAK2 phosphorylation. (A) Schematic representation of IL-6-JAK2-STAT3 (top) and IL-1β-IRAK1-STAT3(bottom) axis. The Normalized Median Fluorescence Intensity (Normalized MedFI) for IL-6 induced STAT3 pY705 (middle) and IL-1β-induced STAT3pS727 (right) phosphorylation were analyzed by flow cytometry in purified CD4+ T cells stimulated with IL-6 (middle) or IL-1β in the presence orabsence of AKBA (right). (B) Phosphorylation of IRAK1 at pS376 and pT387 was analyzed by flow cytometry in purified CD4+ T cells stimulatedwith IL-1β in the presence or absence of AKBA. A representative experiment is shown (left) and normalized MedFI are shown (right). Each dotrepresents a single HD and bars represent mean ± SEM. *p < 0.05, one-way ANOVA with Bonferroni’s correction for multiple comparisons.(C) Dot plots represent normalized MedFI for JAK2 pY1007/1008, STAT3 pY705, IRAK1 pS376, and STAT3 pS727 analyzed by flow cytometry inpurified naıve CD4+ T cells stimulated with beads for 12 days in the presence or absence of AKBA and in Th17-polarizing conditions. Each dotrepresents a different HD bars represent mean ± SEM. *p < 0.05, RM one-way ANOVA with Dunnett’s correction for multiple comparisons with asingle control.

has been reported that the production of IL-22 and IL-17 by Th17cells is differentially regulated [30]. While both IL-6 and IL-1β

seem to be required for IL-17A release [13, 14], IL-22 productioncan be induced only with IL-6 [30]. The inability of AKBA to blockSTAT3 phosphorylation induced by IL-6 most likely underlies itsreduced effect on IL-22 production.

Although human Th17 polarizing conditions are still incom-pletely defined, it is well accepted that IL-6 and IL-1β are twoessential cytokines in human Th17 differentiation [13, 14], andmost likely the ability to differentiate Th17 cells is based on theircapacity to fully activate STAT3. During STAT3 activation IL-6 isresponsible for phosphorylation of STAT3 Y705 via JAK2 [11] andIL-1β for phosphorylating STAT3 S727 via IRAK1 [12]. Accordingto our data, phosphorylation of STAT3 at Y705 in IL-6-stimulatedCD4+ T cells was not affected by AKBA treatment, suggesting that

the inhibition of Th17 differentiation by AKBA does not involveIL-6 signaling. This observation is in contrast with a previous studyin which the ability of AKBA to suppress IL-6-induced STAT3 acti-vation by inhibition of JAK2 phosphorylation and, therefore, JAK2phosphorylation of STAT3 at Y705 was demonstrated in multi-ple myeloma cells [27]. As we mentioned above, in this studysuppression of IL-6-induced STAT3 phosphorylation at Y705 wasonly achieved at AKBA concentrations that we found to be toxicfor CD4+ T cells. Contrary to the results obtained for IL-6, wecould demonstrate that AKBA tended to reduce phosphorylationof STAT3 at S727 in CD4+ T cells stimulated with IL-1β, and AKBAsignificantly reduced the phosphorylation of IRAK1, the kinaseresponsible for phosphorylation of STAT3 at S727 after IL-1β

stimulation. Consistent with our results, IRAK1 knockout micehave decreased IL-17 expression and an attenuated inflammatory

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1208 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

Figure 7. AKBA affects differentiated Th17 cells. (A) Purified memory CD4+ T cells from HDs (white histograms) and MS patients (black histograms)were stimulated in vitro for 72 h with beads and IL-1β alone (left) or in combination with IL-23 (right) and in the presence or absence of AKBA.IL-17A release was then measured in culture supernatants by ELISA. Data are shown as mean + SEM of five HDs and five MS patients. HDs andMS patients stimulated only with IL-1β or in combination with IL-23 were independently analyzed (four independent analyses) using RM one-wayANOVA with Dunnett’s correction for multiple comparisons with a single control. Statistical significance are shown, *p < 0.05 and **p < 0.01. Therelease of IL-17A after stimulation with IL-1-β alone or in combination with IL-23 were compared between HDs and MS patients using unpairedt-tests with two-tailed p-values. Statistical significance is shown *p = 0.01. (B) Representative flow cytometric analysis of intracellular IL-17A andIFN-γ production by in vitro PHA expanded memory CD4+ Th17 cells in the presence or absence of AKBA. Numbers in each compartment representthe percentage of positive cells. Data shown are representative of three analyses performed.

responses [15]. As mentioned above, our data also indicated thatAKBA promotes the differentiation of naıve CD4+ T cells intoIL-10-producing Foxp3+ T cells. AKBA may also promote the dif-ferentiation of these cells by inhibiting IL-1β signaling. In supportof this, it has been recently reported that IL-1β inhibits IL-10 pro-duction in differentiating Th17 cells whereas blockade of IL-1β

increases IL-10 production by Th17 cells [14].A number of anti-inflammatory effects of AKBA in vitro have

been described including the inhibition of several enzymes such asmicrosomal prostaglandin E2 synthase-1 (mPGES1) and as a con-sequence suppression of PGE2 formation in whole human blood[26]. Interestingly, it has been reported that PGE2 plays an impor-tant role in promoting Th17 differentiation and expansion [31–33], suggesting that AKBA may also inhibit Th17-cell differentia-tion and expansion by reducing PGE2 formation. Although we didnot address this point here, the fact that PGE2 is added to the naıveCD4+ T-cell cultures as a component of the Th17 polarizationcocktail and the fact that monocytes, the main blood cells able toproduce PGE2, are absent in these polarizing cultures suggest thatthis mechanism most likely does not play a role in our in vitro sys-tem. However, the AKBA effect on PGE2 production might be rele-vant in vivo, and future studies should examine whether this mech-anism of action of AKBA affects Th17-cell differentiation in vivo.

Our observations suggest that BAs act by a novel and interestingmechanism to block IL-1β IRAK1 phosphorylation, making BAs apotential therapeutic option for several autoimmune- and autoin-

flammatory diseases [34]. IL-1β has just recently been shown toinduce a general proinflammatory T-cell phenotype on its own andin this context to be able to promote IL-17A-independent of RORγt[35]. Blocking IL-1β-mediated signaling has been discussed as anew treatment option in central nervous system diseases and inparticular in MS [36]. BAs have been reported to inhibit IL-1β

[37] in vitro and in vivo, and this may be the mechanism by whichthey exert their beneficial effects in clinical trials of osteoarthritisand rheumatoid arthritis [38, 39]. We are currently conductinga bicentric phase IIa proof-of-concept study examining the safety,tolerability, and efficacy of BAs in RRMS patients (SABA Trial).BAs are orally available and have up to now shown very good toler-ability and safety in several clinical trials of different autoimmunediseases [17]. They might therefore be a new treatment option,especially for patients with early manifestations of MS, the patientgroup in which we are currently testing BAs, but also could be ofinterest for several other autoimmune diseases.

Materials and methods

PBMCs and isolation of CD4+ T cells

Peripheral blood was obtained from HDs (mean age 28.4 ± 7.9)recruited at the Blood Bank and RRMS patients (mean age

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1209

34.8 ± 10.5, median EDSS 1.9) at the inims outpatient clinicand day hospital both at the University Medical Center Hamburg-Eppendorf. This study was approved by the local ethics committee(Ethik-Kommission der Arztekammer Hamburg, No. 2758), andinformed consent was obtained from all subjects. Patients whohad not received steroids for at least 4 weeks prior to enrolmentor any immunomodulatory or immunosuppressive agent duringthe last 3 months were considered untreated and included in thestudy. PBMCs were separated by density gradient with Histopaque(PAA, Pasching, Austria). Immediately after isolation, naıve, andmemory or total CD4+ T cells were purified by negative selectionusing BD Imag kits (BD Biosciences, NJ, USA) according to themanufacturer’s instructions. Purity was checked by flow cytome-try using the following antibodies: anti-CD3 (PE, Biolegend, SanDiego, USA), anti-CD4 (PB, DaKo Biozol, Germany or PB, DaKoBiozol, Germany), anti-CD45RA (APC, BD Biosciences) and anti-CD45RO (FITC, ebiosciences, San Diego, USA). Only preparationswith a purity of > 95% were used for further experiments. Sam-ple acquisition was conducted using a LSRII (BD Biosciences)flow cytometer and data were analyzed using the FlowJo soft-ware (Tree Star, Inc.). CD4+ CD127+ CD45RA+ CD25− cells weresorted from CD4+ enriched PBMCs using a FACSAriaTM III (BDBiosciences) and the following antibodies: anti-CD127 (PerCP-Cy5.5, Dako), anti-CD45RA (allophycocyanin, BD Biosciences),and anti-CD25 (FITC, Biolegend).

Expansion of differentiated Th17 memory T cells

Differentiated Th17 cells were expanded by seeding memory CSF-infiltrating T cells from a MS patient (2 × 103/well) in a 96-wellU-bottom microplate with 2 × 105 irradiated (45 Gy) allogeneicPBMCs in RPMI complete medium containing 50 ng/mL PGE2, 20ng/mL IL-23 (R&D), 20 IU/mL IL-2 (Tecin, Roche Diagnostics)and 1 μg/mL PHA-P (Sigma). IL-2 and IL-23 were added every3–4 days.

In vitro culture and treatment withAcetyl-11-keto-β-boswellic acid

Th17, Th1, and Th2 polarization: Purified naive CD4+ T cells wereseeded at 1 million cells/well in 96-well U-bottom microplatesand incubated overnight with or without different concen-trations of AKBA (Calbiochem R© Merck, Darmstadt, Germany)in X-VIVOTM culture medium (Lonza Walkersville, Inc., USA).Overnight untreated or AKBA pretreated naive CD4+ T cellswere seeded in 96-well U-bottom microplates at 50.000 cells/welland stimulated with 25.000 anti-CD3, anti-CD2, anti-CD28-coatedbeads (Miltenyi Biotec, GmbH, Germany) in: (i) Th17-polarizingconditions (10 ng/mL IL-1β (PeproTech, Rocky Hill, NJ, USA),50 ng/mL IL-6 (PeproTech), 50 ng/mL Prostaglandin E2 (PGE2,Sigma-Aldrich, Munich, Germany), 10 mg/mL neutralizing anti-IFN-γ antibody (PeproTech), 10 mg/mL neutralizing anti-IL-4antibody (PeproTech), and 20 ng/ml IL-23 (R&D Systems,

Wiesbaden-Nordenstadt, Germany)); (ii) Th1 polarizing condi-tions (10 ng/mL IL-12 (PeproTech)); and (iii) Th2 polarizingconditions (12.5 ng/mL IL-4 (Biolegend), 10 mg/mL neutraliz-ing anti-IFN-γ antibody (Biolegend) and 2 μg/mL neutralizinganti-IL12 antibody (Biolegend)). A 20 U/mL IL-2 (Tecin, RocheDiagnostics) and AKBA at different concentrations were addedon day 0, 3, 5, 7, and 10. For Th2 polarization the polarizationcocktail was also added on day 0, 3, 5, 7, and 10.

Treg polarization: CD4+ CD127+ CD45RA+ CD25− sortedcells were seeded at 1 million cells/well in 96-well U-bottommicroplates and incubated overnight with or without differentconcentrations of AKBA (Calbiochem R© Merck) in X-VIVOTM cul-ture medium (Lonza). Overnight untreated or AKBA pretreatedCD4+ CD127+ CD45RA+ CD25− T cells were seeded in 96-well U-bottom microplates at 50,000 cells/well and stimulatedwith 5000 anti-CD3, anti-CD2, anti-CD28-coated beads (MiltenyiBiotec) in Treg-polarizing conditions (5 ng/mL TGF-beta and 100nM retinoic acid (Sigma-Aldrich)). A 80 U/mL IL-2 (Tecin) andAKBA at different concentrations were added on day 0, 3, and 5.

Purified memory CD4+ T cells were seeded in 96-well U-bottommicroplates at 150 000 cells/well and stimulated with 5000 anti-CD3, anti-CD2, anti-CD28-coated beads (Miltenyi Biotec, GmbH,Germany) in the presence of 0, 1, or 5 μM AKBA for 72 h. Whenindicated 10 ng/mL IL-1β and 20 ng/mL IL-23 were also added.

Cytokine production

For intracellular cytokine staining, naıve CD4+ T cells were ana-lyzed 12 days after polarization was started and PHA expanded dif-ferentiated Th17 cells were analyzed 14 days after their last stim-ulation. Cells were then stimulated with PMA (50 ng/mL, Sigma)and ionomycin (1 μg/mL, Sigma) in the presence of BrefeldinA (10 μg/mL, eBioscience) for 5 h. Next, cells were stainedwith LIVE/DEAD Fixable Dead Cell Stain Kit (AmCyan, MolecularProbes, Invitrogen), fixed and permeabilized with the correspond-ing buffers (eBioscience) and stained with anti-CD3 (PE, Dako-Cytomation, Denmark), -CD8 (PB, DakoCytomation, Denmark),-IFN-γ (FITC, BDPharmingen), -IL-4 (PE-Cy7, eBioscience), -IL-10(PE, Biolegend), and -IL-17A (Alexa Fluor (-647, eBioscience) anti-bodies at room temperature. Sample acquisition was conductedusing a LSRII (BD Biosciences) flow cytometer and data were ana-lyzed using the FlowJo software (Tree Star, Inc.). Cytokine levelsin supernatants were determined using a cytofluorimetry-based13-plex ELISA system (Flowcytomix, Bender Medsystem GmbH,Austria) and data analysis was performed using the FlowCytomixTM Pro Software (Version 2.4, Bender Medsystem). Alternatively,cytokines were also measured using ELISA kits (Human IFN-γ, IL-17A, IL-4, IL-5, IL-10 ELISA MAXTM Deluxe, Biolegend) accordingto manufacturer’s instructions.

Transcription factors quantified by qRT-PCR

CD4+ T cells, harvested 12 days after polarization wasstarted, were stimulated with PMA (50 ng/mL, Sigma)

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1210 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

and ionomycin (1 mg/mL, Sigma) for 6 h and RNA wasextracted. For mRNA gene expression assays, the primer andprobe sets Hs00203436 m1, Hs01076112 m1, Hs01085834 m1,Hs00231122 m1, and Hs00538167 m1 for T-bet, RORγt, Foxp3,GATA-3, and IL12RB, respectively, were purchased from AppliedBiosystems (Foster City, CA) and used according to manufactur-ers instructions. A 18S rRNA was used as endogenous control andthe relative gene expression was calculated by the �� Ct methodusing untreated TH17-polarized CD4+ T cells as calibrator.

Analysis of proliferative responses

In order to asses an effect of AKBA on the proliferative response ofT cells, peripheral blood mononuclear cells were stimulated with1 μg anti-CD3 (clone OKT3, BioXCell, West Lebanon, NH, USA) for72 h in the presence or absence of different doses of AKBA and thencells were pulsed with (methyl-3H)thymidine (Hartmann Analyt-ics, Braunschweig, Germany) for 15 h before harvesting. Incor-poration of radioactivity was measured in a scintillation counter(Wallac 1450, PerkinElmer, Rodgau-Ju rgesheim, Germany).

Analysis of surface markers and apoptosis by flowcytometry

TH17- and Th2-polarized CD4+ T cells on day 12 or Treg-polarizedCD4+ T cells on day 7 were washed in PBS, Fc-fragments wereblocked with human IgG (Jackson ImmunoResearch) and cellswere stained for 30 min in the dark at room temperature usingthe following antibodies: Th17 (anti-CCR6 (PE, Biolegend), anti-CD161 (PE-Cy7, Biolegend), anti-IL23R (FITC, R&D Systems),anti-CD95 (PE-Cy7, Biolegend), and anti-CD69 (FITC, Biole-gend)), Th2 (anti-CRTH2 (FITC, Biolegend), anti-CCR4 (allophy-cocyanin, Biolegend)), and Treg (CD127 (PerCP-Cy5.5, Dako)).In order to measure apoptosis, cells were stained for AnnexinV (PE, Biolegend) according to manufacturer’s instructions. Formeasurement of Bcl-2 expression in Th17 polarized and Foxp3expression in Treg-polarized CD4+ T cells, cells were fixed andpermeabilized with the corresponding buffers (eBioscience) andstained with anti-Bcl-2 (PE, Biolegend) or Foxp3 (PE, Biolegend)for 35 min at room temperature in the dark. Sample acquisitionwas conducted using a LSRII flow cytometer (BD Biosciences) anddata were analyzed using the FlowJo software (Tree Star, Inc.).

Phosphorylation of IRAK1 (pT387 and pS376), JAK2pY1007/Y1008, and STAT3 (pY705 and pS727)

Purified CD4+ T cells treated in vitro with 50 ng/mL IL-6 orwith 50 ng/mL IL-1β and TH17-polarized CD4+ T cells on day12 were analyzed using BD PhosflowTM (BD Biosciences) buffersand the following antibodies: anti-STAT3 (PE, BD Biosciences),anti-STAT3 pY705 (PerCP-CyTM 5.5, BD Biosciences), anti-STAT3pS727 (Alexa Fluor R© 488), anti-JAK2 (sc-278, Santa Cruz), anti-

JAK2 pTyr1007/Tyr1008 (sc-21870, Santa Cruz), anti-IRAK1(FITC, Biorbyt), anti-IRAK1 pThr387 (PE, Biorbyt, Cambridge,United Kingdom), anti-IRAK1 pSer376 (sc-130197) with a sec-ondary antibody anti-rabbit Alexa Fluor R© 488. Sample acquisitionwas conducted using an LSRII flow cytometer (BD Biosciences)and data analysis was performed with the FlowJo (Tree Star,Inc.) software. Expression of phosphorylated proteins was nor-malized to whole protein expression by using the following for-mula: (MedFI of phosphorylated protein/MedFI of correspondingprotein) × 100.

Western blot

Purified CD4+ T cells were treated in vitro with 10 ng/mL IL-1β

for 15 min in X-VIVOTM in the presence or absence of AKBA at1 or 5 μM. Lysates were Western blotted with anti-pIRAK1, anti-IRAK1 (both from Santa Cruz Biotechnology Inc., Santa Cruz, CA)and anti-Actin (Sigma) in PBS with 0.5% BSA. Quantification wasconducted using ImageJ as described previously [40].

Statistical analysis

Statistical analyses were performed with Prism 5.02 (GraphPadSoftware Inc., San Diego, CA). Descriptive statistics are reportedas mean ± SEM. Comparisons of three groups and more wereassessed by repeated measures one-way ANOVA with Dunnett’scorrection for multiple comparisons with a single control or byone-way ANOVA with Bonferroni’s correction for multiple com-parisons as indicated. Parametric tests were applied for two-groupcomparisons using unpaired t-tests with two-tailed p-values.

Acknowledgments: We thank M. Kolster and C. Haueis for tech-nical assistance, S. Fleischer for helpful organizational support andB. Reinhart for carefully reading the manuscript. The Institute forNeuroimmunology and Clinical Multiple Sclerosis Research wassupported by the Gemeinnutzige Hertie Stiftung.

Conflict of interest: R. Martin has previously received an unre-stricted grant by Biogen Idec as well as compensation for presen-tations and advisory function by Biogen Idec, Merck & Serono,Novartis, and Teva.

References

1 Bouma, G. and Strober, W., The immunological and genetic basis of

inflammatory bowel disease. Nat. Rev. Immunol 2003. 3: 521–533.

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

Eur. J. Immunol. 2014. 44: 1200–1212 Clinical immunology 1211

2 Korn, T., Bettelli, E., Oukka, M. and Kuchroo, V. K., IL-17 and Th17 cells.

Annu. Rev. Immunol. 2009. 27: 485–517.

3 Steinman, L., A rush to judgment on Th17. J. Exp. Med. 2008. 205:

1517–1522.

4 Kebir, H., Kreymborg, K., Ifergan, I., Dodelet-Devillers, A., Cayrol, R.,

Bernard, M., Giuliani, F. et al., Human Th17 lymphocytes promote blood-

brain barrier disruption and central nervous system inflammation. Nat.

Med. 2007. 13: 1173–1175.

5 Tzartos, J. S., Friese, M. A., Craner, M. J., Palace, J., Newcombe, J., Esiri, M.

M. and Fugger, L., Interleukin-17 production in central nervous system-

infiltrating T cells and glial cells is associated with active disease in mul-

tiple sclerosis. Am. J. Pathol. 2008. 172: 146–155.

6 Lock, C., Hermans, G., Pedotti, R., Brendolan, A., Schadt, E., Garren, H.,

Langer-Gould, A. et al., Gene-microarray analysis of multiple sclerosis

lesions yields new targets validated in autoimmune encephalomyelitis.

Nat. Med. 2002. 8: 500–508.

7 Ziolkowska, M., Koc, A., Luszczykiewicz, G., Ksiezopolska-Pietrzak, K.,

Klimczak, E., Chwalinska-Sadowska, H. and Maslinski, W., High levels

of IL-17 in rheumatoid arthritis patients: IL-15 triggers in vitro IL-17 pro-

duction via cyclosporin A-sensitive mechanism. J. Immunol. 2000. 164:

2832–2838.

8 Fujino, S., Andoh, A., Bamba, S., Ogawa, A., Hata, K., Araki, Y., Bamba, T.

and Fujiyama, Y., Increased expression of interleukin 17 in inflammatory

bowel disease. Gut 2003. 52: 65–70.

9 Annunziato, F., Cosmi, L., Santarlasci, V., Maggi, L., Liotta, F., Mazzinghi,

B., Parente, E. et al., Phenotypic and functional features of human Th17

cells. J. Exp. Med. 2007. 204: 1849–1861.

10 Veldhoen, M., Hocking, R. J., Atkins, C. J., Locksley, R. M. and Stockinger,

B., TGFbeta in the context of an inflammatory cytokine milieu supports

de novo differentiation of IL-17-producing T cells. Immunity 2006. 24:

179–189.

11 Wen, Z., Zhong, Z. and Darnell, J. E., Jr., Maximal activation of transcrip-

tion by Stat1 and Stat3 requires both tyrosine and serine phosphoryla-

tion. Cell 1995. 82: 241–250.

12 Huang, Y., Li, T., Sane, D. C. and Li, L., IRAK1 serves as a novel regulator

essential for lipopolysaccharide-induced interleukin-10 gene expression.

J. Biol. Chem. 2004. 279: 51697–51703.

13 Acosta-Rodriguez, E. V., Napolitani, G., Lanzavecchia, A. and Sallusto,

F., Interleukins 1beta and 6 but not transforming growth factor-beta are

essential for the differentiation of interleukin 17-producing human T

helper cells. Nat. Immunol. 2007. 8: 942–949.

14 Zielinski, C. E., Mele, F., Aschenbrenner, D., Jarrossay, D., Ronchi, F.,

Gattorno, M., Monticelli, S. et al., Pathogen-induced human Th17 cells

produce IFN-gamma or IL-10 and are regulated by IL-1beta. Nature 2012.

484: 514–518.

15 Maitra, U., Davis, S., Reilly, C. M. and Li, L., Differential regulation of

Foxp3 and IL-17 expression in CD4 T helper cells by IRAK-1. J. Immunol.

2009. 182: 5763–5769.

16 Moussaieff, A. and Mechoulam, R., Boswellia resin: from religious cere-

monies to medical uses; a review of in-vitro, in-vivo and clinical trials. J.

Pharm. Pharmacol. 2009. 61: 1281–1293.

17 Ernst, E., Frankincense: systematic review. BMJ 2008. 337: a2813.

18 Shen, T. and Lou, H. X., Bioactive constituents of myrrh and frankin-

cense, two simultaneously prescribed gum resins in chinese traditional

medicine. Chem. Biodivers. 2008. 5: 540–553.

19 Syrovets, T., Gschwend, J. E., Buchele, B., Laumonnier, Y., Zugmaier,

W., Genze, F. and Simmet, T., Inhibition of IkappaB kinase activity by

acetyl-boswellic acids promotes apoptosis in androgen-independent PC-

3 prostate cancer cells in vitro and in vivo. J. Biol. Chem. 2005. 280:

6170–6180.

20 Poeckel, D., Tausch, L., George, S., Jauch, J. and Werz, O., 3-O-acetyl-

11-keto-boswellic acid decreases basal intracellular Ca2+ levels and

inhibits agonist-induced Ca2+ mobilization and mitogen-activated pro-

tein kinase activation in human monocytic cells. J. Pharmacol. Exp. Ther.

2006. 316: 224–232.

21 Safayhi, H., Mack, T., Sabieraj, J., Anazodo, M. I., Subramanian, L. R.

and Ammon, H. P., Boswellic acids: novel, specific, nonredox inhibitors

of 5-lipoxygenase. J. Pharmacol. Exp. Ther. 1992. 261: 1143–1146.

22 Poeckel, D., Niedermeyer, T. H., Pham, H. T., Mikolasch, A., Mundt, S.,

Lindequist, U., Lalk, M. and Werz, O., Inhibition of human 5-lipoxygenase

and anti-neoplastic effects by 2-amino-1,4-benzoquinones. Med. Chem.

2006. 2: 591–595.

23 Safayhi, H., Rall, B., Sailer, E. R. and Ammon, H. P., Inhibition by boswellic

acids of human leukocyte elastase. J. Pharmacol. Exp. Ther. 1997. 281:

460–463.

24 Tausch, L., Henkel, A., Siemoneit, U., Poeckel, D., Kather, N., Franke, L.,

Hofmann, B. et al., Identification of human cathepsin G as a functional

target of boswellic acids from the anti-inflammatory remedy frankin-

cense. J. Immunol. 2009. 183: 3433–3442.

25 Siemoneit, U., Hofmann, B., Kather, N., Lamkemeyer, T., Madlung, J.,

Franke, L., Schneider, G. et al., Identification and functional analysis

of cyclooxygenase-1 as a molecular target of boswellic acids. Biochem.

Pharmacol. 2008. 75: 503–513.

26 Siemoneit, U., Koeberle, A., Rossi, A., Dehm, F., Verhoff, M., Reckel, S.,

Maier, T. J. et al., Inhibition of microsomal prostaglandin E2 synthase-1

as a molecular basis for the anti-inflammatory actions of boswellic acids

from frankincense. Br. J. Pharmacol. 2011. 162: 147–162.

27 Kunnumakkara, A. B., Nair, A. S., Sung, B., Pandey, M. K. and Aggarwal,

B. B., Boswellic acid blocks signal transducers and activators of tran-

scription 3 signaling, proliferation, and survival of multiple myeloma

via the protein tyrosine phosphatase SHP-1. Mol. Cancer. Res. 2009. 7:

118–128.

28 Kebir, H., Ifergan, I., Alvarez, J. I., Bernard, M., Poirier, J., Arbour,

N., Duquette, P. and Prat, A., Preferential recruitment of interferon-

gamma-expressing Th17 cells in multiple sclerosis. Ann. Neurol. 2009. 66:

390–402.

29 Reppert, S., Boross, I., Koslowski, M., Tureci, O., Koch, S., Lehr, H. A. and

Finotto, S., A role for T-bet-mediated tumour immune surveillance in

anti-IL-17A treatment of lung cancer. Nat. Comm. 2011. 2: 600.

30 Zheng, Y., Danilenko, D. M., Valdez, P., Kasman, I., Eastham-Anderson,

J., Wu, J. and Ouyang, W., Interleukin-22: a T(H)17 cytokine, mediates

IL-23-induced dermal inflammation and acanthosis. Nature 2007. 445:

648–651.

31 Chizzolini, C., Chicheportiche, R., Alvarez, M., de Rham, C., Roux-

Lombard, P., Ferrari-Lacraz, S. and Dayer, J. M., Prostaglandin E2 syner-

gistically with interleukin-23 favors human Th17 expansion. Blood 2008.

112: 3696–3703.

32 Boniface, K., Bak-Jensen, K. S., Li, Y., Blumenschein, W. M., McGeachy,

M. J., McClanahan, T. K., McKenzie, B. S. et al., Prostaglandin E2 regulates

Th17 cell differentiation and function through cyclic AMP and EP2/EP4

receptor signaling. J. Exp. Med. 2009. 206: 535–548.

33 Yao, C., Sakata, D., Esaki, Y., Li, Y., Matsuoka, T., Kuroiwa, K., Sugimoto,

Y. and Narumiya, S., Prostaglandin E2–EP4 signaling promotes immune

inflammation through Th1 cell differentiation and Th17 cell expansion.

Nat. Med. 2009. 15: 633–640.

34 Dinarello, C. A., A clinical perspective of IL-1beta as the gatekeeper of

inflammation. Eur. J. Immunol. 2011. 41: 1203–1217.

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu

1212 Klarissa Hanja Sturner et al. Eur. J. Immunol. 2014. 44: 1200–1212

35 Hebel, K., Rudolph, M., Kosak, B., Chang, H. D., Butzmann, J. and

Brunner-Weinzierl, M. C., IL-1beta and TGF-beta act antagonistically in

induction and differentially in propagation of human proinflammatory

precursor CD4+ T cells. J. Immunol. 2011. 187: 5627–5635.

36 Sha, Y. and Markovic-Plese, S., A role of IL-1R1 signaling in the differen-

tiation of Th17 cells and the development of autoimmune diseases. Self

Nonself. 2011. 2: 35–42.

37 Akhtar, N., Miller, M. J. and Haqqi, T. M., Effect of a herbal-leucine mix

on the IL-1beta-induced cartilage degradation and inflammatory gene

expression in human chondrocytes. BMC Complement Altern. Med. 2011.

11: 66.

38 Vishal, A. A., Mishra, A. and Raychaudhuri, S. P., A double blind, ran-

domized, placebo controlled clinical study evaluates the early efficacy

of aflapin in subjects with osteoarthritis of knee. Int. J. Med. Sci. 2011. 8:

615–622.

39 Sengupta, K., Krishnaraju, A. V., Vishal, A. A., Mishra, A., Trimurtulu, G.,

Sarma, K. V., Raychaudhuri, S. K. and Raychaudhuri, S. P., Comparative

efficacy and tolerability of 5-Loxin and Aflapin against osteoarthritis of

the knee: a double blind, randomized, placebo controlled clinical study.

Int. J. Med. Sci. 2010. 7: 366–377.

40 Gassmann, M., Grenacher, B., Rohde, B. and Vogel, J., Quantifying West-

ern blots: pitfalls of densitometry. Electrophoresis 2009. 30: 1845–1855.

Abbreviations: AKBA: acetyl-11-keto-β-boswellic acid · BA: boswellic

acid · CSF: cerebrospinal fluid · HD: healthy donor · IRAK1: IL-1 receptor-

associated kinase 1 · MS: multiple sclerosis · RRMS: relapsing-remitting

MS

Full correspondence: Dr. Mireia Sospedra, Neuroimmunology and MSResearch, Department of Neurology, University Hospital Zurich,Frauenklinikstrasse 26, 8091 Zurich, SwitzerlandFax: +41-442558864e-mail: Mireia.SospedraRamos@usz.ch

Received: 18/4/2013Revised: 19/12/2013Accepted: 20/1/2014Accepted article online: 27/1/2014

C© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.eji-journal.eu