Experimental Hematology 2010;38:116–123

Interleukin-13 stimulation of the mediastinalB-cell lymphoma cell line Karpas-1106P induces

a phenotype resembling the Hodgkin lymphoma cell line L1236

Erik Anderssona,b, Frida Schaina, Jan Sjoberga, Magnus Bjorkholma, and Hans-Erik Claessonb,c

aDepartment of Medicine, Karolinska Institutet and Karolinska University Hospital, Solna,

Sweden; bOrexo AB, Uppsala, Sweden; cDepartment of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden

(Received 26 August 2009; revised 23 October 2009; accepted 16 November 2009)

Offprint requests to

Uppsala, Sweden; E-m

0301-472X/10 $–see

doi: 10.1016/j.exph

Objective. Primary mediastinal B-cell lymphoma (PMBCL) and Hodgkin lymphoma (HL)share many biological and clinical characteristics supporting a common pathogeneticpathway. Interleukin (IL)-13 has an important pathophysiological role in HL. In this study,we asked the question of whether IL-13 is a major contributor to the observed differencein features of inflammation between HL and PMBCL.

Materials and Methods. Expression of IL-13 and IL-4 receptors was studied by flow cytome-try, expression of a functional cysteinyl leukotriene receptor type 1 (CysLT1R) was investi-gated by calcium flux measurement, expression and activity of 15-lipoxygenase type 1 (15-LO-1) was determined by Western blot and reversed-phase high-performance liquid chroma-tography, respectively, and cytokines were quantified by Bioplex detection.

Results. Stimulation of the PMBCL cell line Karpas-1106P with IL-13 or IL-4 induceda proinflammatory phenotype similar to that of the HL cell line L1236. Upon interleukin stim-ulation of the PMBCL cell line, the cellular size increased and cells became multinucleated.Cells also expressed CysLT1R and 15-LO-1, and produced the proinflammatory eoxins.The IL-13 or IL-4 treated PMBCL cell line and the HL cell line secreted a similar set of cyto-kines such as IL-6, tumor necrosis factorLa, interferon-inducible proteinL10, interferon-g,and RANTES.

Conclusions. IL-13 or IL-4 stimulation of the PMBCL cell line Karpas-1106P induced aninflammatory phenotype that resembles that of the HL cell line. Our results suggest thatthe autocrine release of IL-13 in HL is one critical factor that can at least partly explainthe difference in phenotype between these two lymphoma entities. � 2010 ISEH - Societyfor Hematology and Stem Cells. Published by Elsevier Inc.

Hodgkin lymphoma (HL) and primary mediastinal B-celllymphoma (PMBCL) share certain clinical characteristics,but are classified as two different disease entities with sepa-rate therapeutic approaches. The World Health Organiza-tion International Classification of Diseases placesPMBCL in the group of diffuse large B-cell lymphomas,but gene expression profiling has recently shown that it ismore likely related to HL [1–3]. The existence of medias-tinal gray-zone lymphoma emphasizes that there is a spec-trum of features that can be shared by HL and PMBCL [4].

: Erik Andersson, Ph.D., Orexo AB, Box 303, 751 05

ail: erik.andersson@orexo.com

front matter. Copyright � 2010 ISEH - Society for Hemat

em.2009.11.005

One prominent feature of HL, which distinguishes itfrom other lymphomas, is the Hodgkin Reed-Sternberg(H-RS) cell. The H-RS cell is a giant multinucleated cellsurrounded by an infiltrate of inflammatory immune cellsand it constitutes only 0.5% to 5% of all cells in the tumortissue [5]. Other important features of the H-RS cells is thegerminal B-cell origin, the absence of typical B-cell genesand expression of several genes found in other types ofimmune cells involved in inflammation [6,7]. The PMBCLcells are of thymic B-cell origin with conserved B-cellmarkers that are absent in HL [8].

Leukotrienes are arachidonic acid metabolites formedvia the 5-lipoxygenase pathway. These mediators are potentproinflammatory agents [9]. Inflammation is an importantfeature of HL and we have shown that the cysteinyl

ology and Stem Cells. Published by Elsevier Inc.

117E. Andersson et al. / Experimental Hematology 2010;38:116–123

leukotriene receptor 1 (CysLT1R) is expressed in HL biop-sies as well as in HL cell lines [10]. In addition to HL,CysLT1R was found to be expressed in biopsies fromfour patients with PMBCL but not in other types of non-Hodgkin lymphoma (NHL) [11].

The enzyme 15-lipoxygenase type 1 (15-LO-1) isconstitutively expressed in a few cell types, such as airwayepithelial cells, eosinophils, alveolar macrophages, andreticulocytes [12]. Its expression can also be induced inmonocytes, dendritic cells, and mast cells after stimulationwith IL-13 or IL-4 [12–14]. Recently, we found that 15-LO-1 was expressed in the HL cell line L1236 as well asin 85% of primary HL tumors biopsies and that L1236 cellscan produce the proinflammatory eoxins through the 15-LO-1 pathway [15,16]. We also investigated the expressionof 15-LO-1 in a broad range of NHL by immunohisto-chemistry [11]. The enzyme 15-LO-1 was expressed inonly 1 of 58 examined NHL tumor biopsies. Thus, 15-LO-1 is a rather unique biomarker for HL [15]. Studieson two PMBCL cell lines demonstrated expression of 15-LO-1 messenger RNA (mRNA), but not of the 15-LO-1protein [11].

Expression of IL-13 is a common feature of H-RS cellsand a potential role of this cytokine as an autocrine growthfactor for H-RS cells has been suggested [17]. In contrast,IL-13 expression in NHL is uncommon [17]. In the presentreport, features normally associated with HL were investi-gated on a PMBCL cell line stimulated with IL-13 andIL-4.

Material and methods

MaterialsRPMI-1640, L-glutamine, fetal bovine serum, penicillin, strepto-mycin, Dulbecco’s phosphate-buffered saline (PBSþCa2þ/Mg2þ),TRIS-HCl pH 7.4, NuPAGE-BT gels, and loading buffers werepurchased from Invitrogen (Carlsbad, CA, USA). Human IL-4,IL-13, mouse IgG2b isotype control, mouse antisera against humanIL-13a1 receptor (IL-13a1R) and human IL-4 receptor (IL-4R)were purchased from R&D Systems (Minneapolis, MN, USA). b-actin antibody, dimethyl sulfoxide, NaCl, Nonidet P-40, butylhy-droxytoluene, and phenylmethylsulfonyl fluoride were purchasedfrom Sigma-Aldrich (St Louis, MO, USA). NaOH, HCl, and May-Grunewald/Giemsa staining were purchased from Merck (White-house Station, NJ, USA). Super Frost Plus glasses were boughtfrom Menzel-Glaser (Braunschweig, Germany). Complete miniethylenediamine tetraacetic acid�free protease inhibitor waspurchased from Roche (Basel, Switzerland). The Bradford proteindetection reagent and Bioplex cytokine detection kit were purchasedfrom Bio-Rad (Hercules, CA, USA). ECL-plus and secondaryhorseradish peroxidase antirabbit IgG were purchased from GEHealthcare (Chalfont St Giles, UK). Eoxin C4 (EXC4), Eoxin D4

(EXD4), 15(S)-HETE, 12(S)-HETE, 13(S)-HODE, leukotriene C4

(LTC4), leukotriene D4 (LTD4), leukotriene B4 (LTB4) werepurchased from Biomol Inc. (Plymouth Meeting, PA, USA). Arach-idonic acid and linoleic acid were purchased from Nu-Chek Prep

Inc. (Elysian, MN, USA). N-(2-Hydroxyethyl)-5Z,8Z,11Z,14Z-eicosatetraenamide was purchased from Tocris Bioscience (Bristol,UK). 15-LO-2 antisera, montelukast and zafirlukast were purchasedfrom Cayman Chemicals (Ann Arbor, MI, USA). Antiserum against15-LO-1 was described in reference [15]. The antisera against 5-lip-oxygenase and cyclooxygenases 1 and 2 were a gift from Merck-Frost (Kirkland, Quebec, Canada).

Cell culturesL1236 (kind gift from V. Diehl, Department of Internal Medicine,University Hospital of Cologne, Germany), Karpas-1106P(DSMZ, Braunschweig, Germany), and A549 cells were cultivatedin RPMI-1640 medium, supplemented with 2 mM L-glutamine,10% fetal bovine serum, penicillin (100 U/mL), and streptomycin(100 mg/mL), in a humified atmosphere with 5% CO2 at 37�C. IL-13 or IL-4 was added to the final concentration 10 ng/mL.

May-Grunewald/Giemsa stainingCells were washed twice in PBS with Ca2þ/Mg2þ containing 10%fetal bovine serum before 0.1 � 106 cells were cytospinned ontoSuper Frost Plus glasses and stained by May-Grunewald/Giemsa.Photographs were taken with a Nikon ECLIPS E600 microscopewith 600� magnification.

15-LO-1 activity assays and Western blotCells were washed twice and resuspended in 0.5 mL PBS withCa2þ/Mg2þ. Cells were preincubated for 2 minutes in a waterbath at 37�C with shaking. Arachidonic acid was added and theincubation proceeded for 10 minutes before termination by addi-tion of 0.5 mL ice-cold methanol. After centrifugation at 7,200gat 4�C for 10 minutes, the supernatant was saved at �80�C untilreversed-phase high-performance liquid chromatography (RP-HPLC) analysis. Samples were analyzed on a Waters 2795HPLC linked to the diode array detector W2996 and the mobilephase methanol:water:triflouroacetic acid (70:30:0.007) at 1.2mL/min, and the column Zorbax eclipse plus a C18 columnfrom Agilent (Berkshire, UK). In the eoxin assay, the supernatantwas centrifuged at 7,200g at 4�C for 10 minutes, then diluted 10times in water, and subsequently purified by solid-phase C18extraction (Oasis HLB 1 mL, 10 mg; Waters, MA, USA). Thematerial in the column was eluted with 300 mL methanol andmixed with 200 mL water prior to subjection on a RP-HPLCsystem with the mobile phase acetonitrile:methanol:water:aceticacid (29:19:72:0.8), pH 5.6, at 1 mL/min. The protein extractionand Western blot detection was performed as described previously[15]. Briefly, 10 mg protein per sample was loaded on Nu-PAGEBisTris gels according to Invitrogen’s protocol. Proteins wereblotted onto polyvinylidene fluoride membrane, blocked in TBS-T with milk powder and subsequently incubated with 15-LO-1antisera, secondary anti�rabbit horseradish peroxidase and b-actin-horseradish peroxidase antibodies followed by detectionwith ECL-plus [14].

Flow cytometry analysisL1236 cells and Karpas-1106P cells were analyzed by flow cytom-etry for the expression of the IL-4R and IL-13a1R. Briefly, 0.6� 106

cells were washed in PBS including 0.5% bovine serum albumin and0.1 M NaN3, and incubated on ice for 30 minutes with 0.5 mg mouseantisera against human IL-13a1R and human IL-4R. The cells werewashed twice in PBS, incubated with phycoerythrin-labeled goat-anti mouse immunoglobulins (DAKO, Glostrup, Denmark) for 30

Figure 1. Expression of interleukin (IL)-4 and IL-13a1 receptors in L1236 cells and Karpas-1106P cells. (A) Expression of IL-4R and IL-13a1R in L1236

cells. (B) Expression of IL-4R and IL-13a1R in Karpas-1106P cells; 71.5% and 11.5% of L1236 cells expressed the IL-4R and the IL-13a1R, respectively;

corresponding figures in Karpas-1106P cells were 46.7% and 20.6%, respectively. Surface expression of the indicated marker M2 (IL-4R or IL-13a1R) is

overlaid with the isotype control M1 to compensate for background staining. Histograms from one representative experiment of three are shown.

118 E. Andersson et al./ Experimental Hematology 2010;38:116–123

minutes on ice and resuspended in 1% paraformaldehyde. FACSCa-libur flow cytometer CellQuest Pro Software (BD Biosciences, SanJose, CA, USA) was used and receptor expression was measured aspercentage of gated or total cells and as the mean fluorescence inten-sity value of the antibody of interest compared to the isotype-matched control. The lung epithelial cell line A549 was used asa positive control.

Receptor screenThe FLIPR calcium 3 kit from Molecular Devices (Sunnyvale,CA, USA) was used to screen Karpas-1106P cells for receptorsthat bind any of the following ligands: LTB4, LTC4, LTD4,EXC4, EXD4, 12(S)-HETE, 15(S)-HETE), arachidonic acid, N-(2-hydroxyethyl)-5Z,8Z,11Z,14Z-eicosatetraenamide, linoleic acid,13(S)-HODE, 5(S)-HETE. L1236 and Karpas-1106P cells culti-vated with or without IL-13 or IL-4 were washed twice in PBSwith Ca2þ/Mg2þ and loaded with flourophore according to thecalcium 3 kit. Then, the compounds at final concentration of100 nM were added in PBS along with solvent and ionomycincontrols. The calcium flux was measured on a Flexstation fromMolecular Devices. In the leukotriene assay, the cells were alsoincubated with zafirlukast or montelukast for 30 minutes at37�C, prior the addition of 300 nM LTD4.

Cytokine releaseKarpas-1106P cells were cultivated for 10 days with IL-13 or IL-4(10 ng/mL). Cells were washed once and diluted to 1 millioncells/mL in RPMI-1640. L1236 and Karpas-1106P cells werecultivated with or without interleukins for 16 hours. Subsequently,cells were spun down for 10 minutes at 7,200g, and the superna-tants were saved in �80�C. BioRad’s Biolplex assay kit was usedfor cytokine detection and each sample was diluted 2 or 10 times.

Results

Karpas-1106P and L1236 cells express IL-4 and IL-13a1 receptorsFlow cytometry was used to investigate if the HL cell lineL1236 and the PMBCL cell line Karpas-1106P expressedthe IL-4R and the IL-13a1R. Antibodies were validatedby detecting respective receptor in A549 cells (data notshown) [18]. The IL-4R and the IL-13a1R were expressedin 71.5% and 11.5% of the gated L1236 cells (Fig. 1A) andin 46.8% and 20.6% of the gated Karpas-1106P cells,respectively (Fig. 1B).

Stimulation of Karpas-1106P cells with IL-13 or IL-4change the cell morphologyKarpas-1106P cells increased in size and became multi-nucleated after cultivation for 10 days in media containingIL-4 (Fig. 2A) or IL-13 (Fig. 2B). Unstimulated Karpas-1106P cells, which were also cultivated for 10 days, weresmaller and had a single nucleus (Fig. 2C). The HL cellline L1236 has the characteristic large- and multinucleatedcells (Fig. 2D). Furthermore, Karpas-1106P cells cultivatedin the presence of IL-13 or IL-4 proliferated more slowlythan untreated cells (data not shown).

Karpas-1106P cells stimulated with IL-13 or IL-4express 15-LO-1 and produce eoxinsTo investigate if Karpas-1106P cells could be induced toexpress 15-LO-1, cells were stimulated with IL-13 or

Figure 2. May-Grunewald-Giemsa staining of Karpas-1106P cells and L1236 cells. After interleukin (IL)-13 or IL-4 (10 ng/mL) stimulation of Karpas-

1106P cells for 10 days, the cells were washed twice in phosphate-buffered saline (PBS) with Ca2þ/Mg2þ containing 10% fetal bovine serum before the

cells (0.1 � 106) were cytospinned onto Super Frost Plus glasses and stained by May-Grunewald-Giemsa. Photographs were taken with Nikon ECLIPS

E600 microscope with 600� magnification. (A) IL-4 stimulated Karpas-1106P cells; (B) IL-13 stimulated Karpas-1106P; (C) nonstimulated Karpas-

1106P cells; (D) L1236 cells.

119E. Andersson et al. / Experimental Hematology 2010;38:116–123

IL-4 for 10 days. Samples were taken daily from thecultures and 10 million cells were incubated with arachi-donic acid (25 mM) for 10 minutes at 37�C. Unstimulatedcells at day 0 did not produce any 15-HETE. However,15-HETE was detected after 1 day of IL stimulation andlevels increased with time and reached a plateau at day 9(Fig. 3A). Similar results were obtained with either IL-13or IL-4. The formation of 12-HETE and 15-HETE wasalways in the ratio 1:9. Chiral HPLC analysis confirmedthat the metabolite formed was 15(S)-HETE (data notshown). In parallel, samples were taken daily from theIL-4 cultures for western blot analysis. No 15-LO-1 protein

was detected at day 0 before IL-4 stimulation (Fig. 3B).After 1 day of IL-4 stimulation, the 15-LO-1 band washardly visible, but thereafter the amount of 15-LO-1 proteinincreased with time. Stimulation with IL-13 induced the 15-LO-1 protein to the same extent as with IL-4 (data notshown). We have earlier reported that Karpas-1106P andMed-B1, another PMBCL-derived cell line, expressed 15-LO-1 mRNA but no protein [11]. In contrast to Karpas-1106P, only a modest increased expression of 15-LO-1protein was observed in Med-B1 cells after IL-treatment(data not shown). Therefore, Karpas-1106P cells wereused in this study. The amount of 15-LO-1 in IL-treated

Figure 3. Time course of the formation of 15-HETE and expression of 15-lipoxygenase type 1 (15-LO-1) protein in Karpas-1106P cells during cultivation with

interleukin (IL)-13 or IL-4 and biosynthesis of eoxins. The Karpas-1106P cells were cultivated for 10 days in the presence of IL-13 or IL-4 (10 ng/mL). Each day

an aliquot of cell culture was spun down and either incubated with arachidonic acid or saved for protein extraction and Western blot analysis of 15-LO-1. (A) After

washing twice in phosphate-buffered saline, 10� 106 cells/mL were incubated with arachidonic acid (25 mM) for 10 minutes at 37�C. The amounts of 15-HETE

were quantified by reversed-phase high-performance liquid chromatography with a photo diode array detector (RP-HPLC-PDA) set at 235 nm and results are

presented as mean 6 standard deviation from triplicate samples from one representative experiment. (B) Western blot analysis. After protein extraction, 10

mg total cellular protein was loaded per sample on BT-Nupage gels and Western blot was performed according to Invitrogen’s protocol. L1236 cells and recombi-

nant 15-LO-1 were used as positive controls. A b-actin antibody was used as loading control. (C) Formation of eoxins in IL-4�treated Karpas-1106P cells. The

cells (10� 106 cells/mL) were incubated as described here. The sample was purified by C18 solid-phase extraction before injection on the RP-HPLC-PDA system

with the mobile phase acetonitrile:methanol:water:acetic acid (29:19:72:0.8), pH 5.6, 1 mL/min. Retention time of the material in peak I and peak II corresponded

to the retention time of synthetic EXC4 and EXD4, respectively. (Inset) Ultraviolet spectra of the material in peak I and peak II.

120 E. Andersson et al./ Experimental Hematology 2010;38:116–123

Karpas cells was lower than in L1236 cells (Fig. 3B).Another band was also detected above the 15-LO-1 band.This band was found to be an artifact from the b-actin anti-body used as a control of protein loading. Western blotanalysis demonstrated that Karpas-1106P cells 6 IL-13or IL-4 did not express 5-lipoxygenase, 15-lipoxygenasetype 2, cyclooxygenase type 1 or 2 (data not shown).

The HL cell line L1236 releases the inflammatory medi-ators eoxins, upon arachidonic acid stimulation [15]. Incu-bation of IL-4�treated Karpas-1106P cells witharachidonic acid led also to the formation of eoxins. Severalpeaks with an ultraviolet absorbance maximum of 270 to280 nm were observed in the HPLC chromatogram(Fig. 3C). Retention times of the material in two peaks (Iand II) corresponded to retention times of EXC4 andEXD4, respectively. The ultraviolet spectra of the materialin these two peaks showed also the characteristic triene-spectra of eoxins with the maximum at 282 nm. Theseresults demonstrate that IL-4�treated Karpas-1106P cellscan produce eoxins. IL-13�stimulated Karpas-1106P cellsproduced similar amounts of eoxins as IL-4�treated cells(data not shown).

Karpas-1106P cells stimulated with IL-13 or IL-4express a functional CysLT1RA functional CysLT1R has been shown to be expressed inH-RS cells and L1236 cells [10]. In a previous study, wereported that the CysLT1R was expressed in biopsiesfrom patients with PMBCL. Furthermore, the cell line Kar-pas-1106P was found to have a low expression of CysLT1R[11]. Therefore, we investigated if IL-13 or IL-4 couldstimulate the expression of CysLT1R in Karpas-1106Pcells. The addition of LTC4 and LTD4 gave a robust calciumsignal in IL-13� or IL-4�stimulated Karpas-1106P cells,but not in unstimulated cells, demonstrating that IL-4 andIL-13 induced expression of CysLT1R (Fig. 4A). Inparallel, we tested the effect of 10 different eicosanoids(Fig. 4A). No significant effect was observed with theseother lipid mediators except for a weak signal after stimu-lation with EXD4 (Fig. 4A). To clarify if the calcium signalwas mediated by CysLT1R, IL-4�stimulated Karpas-1106P cells were preincubated with the CysLT1R antago-nists, zafirlukast or montelukast. These drugs completelyblocked calcium signals, indicating that the signal wasmediated through CysLT1R (Fig. 4B).

Figure 4. Effect of cysLTs on calcium mobilization in Karpas-1106P and L1236 cells. Karpas-1106P cells were cultivated with or without interleukin (IL)-

13 or IL-4 (10 ng/mL for 10 days). (A) Cells were loaded with Calcium 3 according to the Molecular Devices protocol. The indicated compounds (100 nM)

were added to the cells and calcium flux was monitored as relative fluorescence units (RFU) for 1 minute. (B) The IL-4�stimulated Karpas-1106P cells were

preincubated with montelukast, zafirlukast, or dimethyl sulfoxide for 30 minutes before the stimulation with 300 nM LTD4. The IC50 values for Montelukast

and Zafirlukast were 2.1 and 0.7 nM, respectively. Data show mean RFU 6 standard deviation of triplicate determinations of the maximum response from one

representative experiment out of two.

121E. Andersson et al. / Experimental Hematology 2010;38:116–123

Cytokine profile is similar in L1236 cells and IL-13 orIL-4�stimulated Karpas-1106P cellsThe HL cell line L1236 releases certain cytokines [19]. In thisstudy, the Bioplex assay from Bio-Rad was used to investi-gate the cytokine release from IL-13 or IL-4�stimulatedKarpas-1106P cells and L1236 cells. Both L1236 and IL-treated Karpas-1106P cells released the cytokines IL-6,tumor necrosis factor�a, interferon-inducible protein�10,interferon-g, and RANTES (Table 1). In contrast, IL-2 andIL-4 were not detected in any sample. Macrophage inflam-matory protein�1a was released by unstimulated and IL-stimulated Karpas-1106P cells as well as L1236 cells.Thus, the profile of proinflammatory cytokines spontane-ously released from the cells was similar in IL-13� or IL-4�treated Karpas-1106P cells and L1236 cells.

Surface antigens did not change on Karpas-1106P cellsafter IL-13 or IL-4 stimulationSurface antigens are used for classification of differentlymphomas. Unstimulated and stimulated Karpas-1106P

Table 1. Release of cytokines from Karpas-1106P cells and L1236 cells

L1236 Karpas

TNF-a 169 6 10 0

RANTES 140 6 53 0

IL-6 789 6 91 0

IP-10 1,766 6 427 0

MIP-1a 112 6 16 54 6 2

IL-2 0 0

IL-4 0 0

IL-13 13 6 2 0

IFN-ga 0/398 6 46 0

IP-10 5 interferon-inducible protein�10; MIP-1a 5 macrophage inflammatory

L1236 cells and Karpas-1106P cells, cultivated with or without interleukin (IL)-4

medium and incubated overnight at 37 �C in fresh RPMI medium. Subsequently,

tified by the Bio-Rad’s Bioplex assay kit. Results are presented as the mean (pg/

Each value is the mean of triplicate determinations.aIFN-g was detected in the medium of L1236 cells in two of four experiments (

cells were positive for CD20, CD30, and CD45, but nega-tive for CD10 (data not shown). In comparison, the HLcell line L1236 was positive for CD30 but negative forCD10, CD20, and CD45. Thus, although many biochemicaland cellular features of Karpas-1106P cells were changedrendering a L1236-like phenotype after treatment withIL-13 or IL-4, the expression of certain surface antigensremained different.

DiscussionHL and PMBCL share many biological and clinicalfeatures. Several lines of evidence indicate that these twoentities are pathogenically related [2,4,7,20]. However,the inflammatory component of HL is a unique featurethat distinguishes it from PMBCL.

Autocrine IL-13 signaling in HL results in constitutivesignal transducers and activators of transcription (STAT)6 activation [21,22]. In contrast, the constitutive STAT6activation in PMBCL is independent of IL-13 or IL-4

Karpas þ IL-4 Karpas þ IL-13

226 6 30 144 6 12

96 6 1 73 6 5

882 6 152 531 6 28

851 6 116 822 6 10

24 6 4 12 6 2

0 0

72 6 4 0

0 822 6 209

9 6 1 9 6 2

protein�1a.

(10 ng/mL) or IL-13 (10 ng/mL) for 10 days, were washed in RPMI-1640

cells were spun down and the cytokines in the cell-free media were quan-

mL) 6 standard deviation from one representative experiment out of two.

mean of two positive experiments).

122 E. Andersson et al./ Experimental Hematology 2010;38:116–123

signaling [23]. In this study, flow cytometry analysisdemonstrated that both the HL cell line and the PMBCLcell line under study expressed IL-4R and IL-13a1R(Fig. 1). However, there is no evidence to indicate thatIL-4 is important in the pathophysiology of HL or PMBCL.Expression of the IL-4 receptor in these cells might reflectthe common B-cell origin because IL-4 is involved in classswitching of immunoglobulins [24].

A distinctive feature of HL and L1236 is the giant H-RScell with at least two nuclear lobes or nuclei with prominentnucleoli [6]. Nonstimulated Karpas-1106P cells have onenonlobated nucleus and they are smaller than L1236 cells(Fig. 2). However, the IL-stimulated Karpas-1106P cellsincreased in size and became multinucleated, resemblingL1236 cells morphologically.

The enzyme 15-LO-1 is expressed in a restricted numberof primary cells. We have identified the HL cell line L1236as the first cell line with a constitutive high expression of15-LO-1 [15]. In addition, we detected 15-LO-1 in 85%of tumor biopsies from HL patients, but in only one biopsyfrom 58 NHL tumors, an anaplastic large-cell lymphoma[11]. Our studies show that the expression of 15-LO-1 isan almost unique biomarker for HL. Here we show thata PMBCL cell line, stimulated with IL-13 or IL-4, alsoexpressed a functional 15-LO-1, which suggests a molecularrelationship between these two diseases.

It is important to note that IL-4 does not necessarilyinduce 15-LO-1 expression in a cell with IL-4R [25,26].The amount of the 15-LO-1 protein and enzymatic activityincreased with time during 10 days of incubation of thePMBCL cell line with IL-13 or IL-4 (Fig. 3). Furthermore,the IL-4�stimulated PMBCL cell line produced the proin-flammatory eoxins, formed through the 15-LO-1 pathway,which are also produced by the HL cell line L1236.Because IL-13 or IL-4 induced a strong 15-LO-1 expres-sion in the PMBCL cell line, this cell line might be a usefulon-off model system of 15-LO-1 to study the biochemicaland biological role of this enzyme.

Figure 5. A cartoon of the phenotypic changes of the PMBCL cell line Karpa

lymphoma (HL) cell line L1236. (A) Summary of the characteristic features of

nucleoli, the autocrine interleukin (IL)-13 signaling, 15-lipoxygenase type 1 (15

(CysLT1R) expression, and release of inflammatory cytokines. (B) Changes ind

into an L1236-like phenotype with similar morphology and proinflammatory sig

CysLTs are well-known mediators of inflammation andwe have reported that both primary H-RS cells and celllines express the CysLT1R [10]. Previous studies in ourlaboratory have demonstrated that both PMBCL cell linesKarpas-1106P and Med-B1 express a functional CysLT1R,but calcium response was very low in Karpas-1106P cellsafter challenge with cysLTs [11]. In this study, the PMBCLcell line treated with IL-13 or IL-4, but not untreated cells,was found to express a functional CysLT1R that respondedwith a robust calcium signal upon LTC4 and LTD4 chal-lenge (Fig. 4). Expression of CysLT1R in tumor biopsiesfrom PMBCL patients indicates that proinflammatorycysLT signaling may also play a role in PMBCL [11].

IL-4 and IL-13 have been shown to induce the expressionof CysLT1R and 15-LO-1 at the transcriptional level throughactivation of the transcription factor STAT6 [27,28]. There-fore, the effects of IL-13 and IL-4 on expression of functionalCysLT1R, as well as 15-LO-1 protein in the PMBCL cellline, were unexpected because this cell line constitutivelyexpress mRNA for both CysLT1R and 15-LO-1 [11], prob-ably due to the constitutive STAT6 activation in PMBCL[23]. One explanation could be that these ILs also influencethe translational process or stabilization of mRNA [29].

There is limited knowledge about the release of cyto-kines and chemokines from PMBCL cells. In contrast,HL is associated with release of proinflammatory cytokinesthat attract the immune cells, which build up the tumortissue [30,31]. IL-13 is an important autocrine cytokine inHL and it was also found in culture medium of the HLcell line, but not in any samples of the PMBCL cell line.The interleukin-treated PMBCL cell line, but not untreatedcultures, released several cytokines such as tumor necrosisfactor�a, RANTES, IL-6, interferon-inducible protein�10,and interferon-g. All these cytokines are associated withinflammation and are produced in HL tumors as well asHL cell lines [19,22,32].

Taken together, these results suggest that IL-13 andprobably IL-4 are important cytokines in the induction of

s-1106P cells during treatment with IL-13 in comparison to the Hodgkin

a Hodgkin Reed-Sternberg (H-RS) cell (L1236): a giant cell with multiple

-LO-1) expression and eoxin production, cysteinyl leukotriene receptor 1

uced by IL-13 stimulation of the Karpas-1106P cells, which transformed

naling but unchanged CD antigens.

123E. Andersson et al. / Experimental Hematology 2010;38:116–123

the characteristic phenotype of H-RS cells (Fig. 5). Further-more, IL-13 and IL-4 stimulation of the PMBCL cell lineinduced the release of proinflammatory cytokines simulta-neously with expression of 15-LO-1 and CysLT1R. Thesefindings support a role of IL-13 in generating the inflamma-tory cell infiltrate around H-RS cells. Our results indicatethat IL-13 might be a critical pathogenetic factor, whichcan partly explain observed differences in phenotypebetween HL and PMBCL.

AcknowledgmentsThis work was supported by grants from the Swedish CancerSociety, Karolinska Institutet (Stockholm, Sweden), StockholmCounty Council (Stockholm, Sweden), Orexo AB (Uppsala,Sweden) and European Commission Sixth Framework ProgrammeGrant LSHM-CT-2004-005033. We would like to thank HeleneJohnson, Margareta Andersson and Yilmaz Mashid for technicalassistance, Anja Porwit for help with morphologic evaluation ofcells and Mats Hamberg for help with the chiral chromatographyanalysis.

Conflict of Interest DisclosureErik Andersson was employed by Orexo AB. Frida Schain wasemployed by Karolinska University Hospital, Solna. Jan Sjobergwas employed by Karolinska University Hospital, Solna. MagnusBjorkholm was employed by Karolinska Institutet and KarolinskaUniversity Hospital, Solna. Hans-Erik Claesson was employed byKarolinska Institutet and Orexo AB.

References1. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of

primary mediastinal B cell lymphoma identifies a clinically favorable

subgroup of diffuse large B cell lymphoma related to Hodgkin

lymphoma. J Exp Med. 2003;198:851–862.

2. Savage KJ, Monti S, Kutok JL, et al. The molecular signature of medi-

astinal large B-cell lymphoma differs from that of other diffuse large

B-cell lymphomas and shares features with classical Hodgkin

lymphoma. Blood. 2003;102:3871–3879.

3. Mani H, Jaffe ES. Hodgkin lymphoma: an update on its biology with

new insights into classification. Clin Lymphoma Myeloma. 2009;9:

206–216.

4. Traverse-Glehen A, Pittaluga S, Gaulard P, et al. Mediastinal gray

zone lymphoma: the missing link between classic Hodgkin’s

lymphoma and mediastinal large B-cell lymphoma. Am J Surg Pathol.

2005;29:1411–1421.

5. Kuppers R. The biology of Hodgkin’s lymphoma. Nat Rev Cancer.

2009;9:15–27.

6. Thomas RK, Re D, Wolf J, Diehl V. Part I: Hodgkin’s lymphomad

molecular biology of Hodgkin and Reed-Sternberg cells. Lancet

Oncol. 2004;5:11–18.

7. Eberle FC, Mani H, Jaffe ES. Histopathology of Hodgkin’s

Lymphoma. Cancer J. 2009;15:129–137.

8. Rodriguez J, Gutierrez A, Piris M. Primary mediastinal B-cell

lymphoma: Treatment and therapeutic targets. Leuk Lymphoma.

2008;49:1050–1061.

9. Claesson H-E, Dahlen S- E. Asthma and leukotrienes: antileukotrienes

as novel anti-asthmatic drugs. J Intern Med. 1999;245:205–227.

10. Schain F, Tryselius Y, Sjoberg J, et al. Evidence for a pathophysiolog-

ical role of cysteinyl leukotrienes in classical Hodgkin lymphoma. Int

J Cancer. 2008;123:2285–2293.

11. Schain F, Schain D, Mahshid Y, et al. Differential expression of cys-

teinyl leukotriene receptor 1 and 15-lipoxygenase-1 in non-Hodgkin

lymphomas. Clin Lymphoma Myeloma. 2008;8:340–347.

12. Claesson HE. On the biosynthesis and biological role of eoxins and

15-lipoxygenase-1 in airway inflammation and Hodgkin lymphoma.

Prostaglandins Other Lipid Mediat. 2009;89:120–125.

13. Kuhn H, O’Donnell VB. Inflammation and immune regulation by

12/15-lipoxygenases. Prog Lipid Res. 2006;45:334–356.

14. Andersson E, Schain F, Svedling M, Claesson HE, Forsell PK. Inter-

action of human 15-lipoxygenase-1 with phosphatidylinositol bisphos-

phates results in increased enzyme activity. Biochim Biophys Acta.

2006;1761:1498–1505.

15. Claesson HE, Griffiths WJ, Brunnstrom A, et al. Hodgkin Reed-Stern-

berg cells express 15-lipoxygenase-1 and are putative producers of

eoxins in vivo: Novel insight into the inflammatory features of clas-

sical Hodgkin lymphoma. FEBS J. 2008;275:4222–4234.

16. Feltenmark S, Gautam N, Brunnstrom A, et al. Eoxins are proinflam-

matory arachidonic acid metabolites produced via the 15-lipoxyge-

nase-1 pathway in human eosinophils and mast cells. Proc Natl

Acad Sci U S A. 2008;105:680–685.

17. Skinnider BF, Kapp U, Mak TW. The role of interleukin 13 in classical

Hodgkin lymphoma. Leuk Lymphoma. 2002;43:1203–1210.

18. LaPorte SL, Juo ZS, Vaclavikova J, et al. Molecular and structural

basis of cytokine receptor pleiotropy in the interleukin-4/13 system.

Cell. 2008;132:259–272.

19. Wolf J, Kapp U, Bohlen H, et al. Peripheral blood mononuclear cells of

a patient with advanced Hodgkin’s lymphoma give rise to permanently

growing Hodgkin-Reed Sternberg cells. Blood. 1996;87:3418–3428.

20. Poppema S, Kluiver JL, Atayar C, et al. Report: workshop on medias-

tinal grey zone lymphoma. Eur J Haematol 2005;(Suppl l):45–52.

21. Skinnider BF, Elia AJ, Gascoyne RD, et al. Signal transducer and acti-

vator of transcription 6 is frequently activated in Hodgkin and Reed-

Sternberg cells of Hodgkin lymphoma. Blood. 2002;99:618–626.

22. Skinnider BF, Elia AJ, Gascoyne RD, et al. Interleukin 13 and inter-

leukin 13 receptor are frequently expressed by Hodgkin and Reed-

Sternberg cells of Hodgkin lymphoma. Blood. 2001;97:250–255.

23. Guiter C, Dusanter-Fourt I, Copie-Bergman C, et al. Constitutive

STAT6 activation in primary mediastinal large B-cell lymphoma.

Blood. 2004;104:543–549.

24. Tangye SG, Ferguson A, Avery DT, Ma CS, Hodgkin PD. Isotype

switching by human B cells is division-associated and regulated by

cytokines. J Immunol. 2002;169:4298–4306.

25. Brinckmann R, Topp M, Zalan I, et al. Regulation of 15-lipoxygenase

expression in lung epithelial cells by interleukin-4. Biochem J. 1996;

318(Pt 1):305–312.

26. Brinckmann R, Kuhn H. Regulation of 15-lipoxygenase expression by

cytokines. Adv Exp Med Biol. 1997;400B:599–604.

27. Conrad DJ, Lu M. Regulation of human 12/15-lipoxygenase by Stat6-

dependent transcription. Am J Respir Cell Mol Biol. 2000;22:226–234.

28. Thivierge M, Stankova J, Rola-Pleszczynski M. IL-13 and IL-4 up-

regulate cysteinyl leukotriene 1 receptor expression in human mono-

cytes and macrophages. J Immunol. 2001;167:2855–2860.

29. Zhai Y, Zhong Z, Chen CY, et al. Coordinated changes in mRNA turn-

over, translation, and RNA processing bodies in bronchial epithelial cells

following inflammatory stimulation. Mol Cell Biol. 2008;28:7414–7426.

30. Skinnider BF, Mak TW. The role of cytokines in classical Hodgkin

lymphoma. Blood. 2002;99:4283–4297.

31. Maggio E, van den Berg A, Diepstra A, et al. Chemokines, cytokines

and their receptors in Hodgkin’s lymphoma cell lines and tissues. Ann

Oncol. 2002;13(Suppl 1):52–56.

32. Teruya-Feldstein J, Tosato G, Jaffe ES. The role of chemokines in

Hodgkin’s disease. Leuk Lymphoma. 2000;38:363–371.