Báo cáo khoa học: Hodgkin Reed–Sternberg cells express 15-lipoxygenase-1 and are putative producers of eoxins in vivo Novel insight into the inflammatory features of classical Hodgkin lymphoma

Hodgkin Reed–Sternberg cells express 15-lipoxygenase-1 and are putative producers of eoxins in vivo

Novel insight into the inflammatory features of classical Hodgkin lymphoma Hans-Erik Claesson1,2, William J. Griffiths3, A˚ sa Brunnstro¨ m2, Frida Schain4, Erik Andersson2,4, Stina Feltenmark1,2, He´ le` ne A. Johnson1, Anna Porwit5, Jan Sjo¨ berg4 and Magnus Bjo¨ rkholm4

1 Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden 2 Orexo AB, Uppsala, Sweden 3 Institute of Mass Spectrometry, School of Medicine, Swansea University, UK 4 Division of Hematology, Department of Medicine, Karolinska University Hospital and Institutet, Stockholm, Sweden 5 Department of Pathology Karolinska University Hospital and Institutet, Stockholm, Sweden

Keywords arachidonic acid; Hodgkin lymphoma; inflammation; L1236; lipoxygenase

Correspondence H.-E. Claesson, Department of Medical Biochemistry and Biophysics, Division of Physiological Chemistry II, Karolinska Institutet, SE-171 77 Stockholm, Sweden Fax: +46 8 324 264 Tel: +46 8 524 87627 E-mail: hans-erik.claesson@ki.se

(Received 14 April 2008, revised 18 June 2008, accepted 23 June 2008)

doi:10.1111/j.1742-4658.2008.06570.x

this cell that system to study the chemical and biological Classical Hodgkin lymphoma has unique clinical and pathological features and tumour tissue is characterized by a minority of malignant Hodgkin Reed–Sternberg cells surrounded by inflammatory cells. In the present study, we report that the Hodgkin lymphoma-derived cell line L1236 has high expression of 15-lipoxygenase-1 and that these cells readily convert arachidonic acid to eoxin C4, eoxin D4 and eoxin E4. These mediators were only recently discovered in human eosinophils and mast cells and found to be potent proinflammatory mediators. Western blot and immunocytochem- istry analyses of L1236 cells demonstrated that 15-lipoxygenase-1 was pres- ent mainly in the cytosol and that the enzyme translocated to the membrane upon calcium challenge. By immunohistochemistry of Hodgkin lymphoma tumour tissue, 15-lipoxygenase-1 was found to be expressed in primary Hodgkin Reed–Sternberg cells in 17 of 20 (85%) investigated biop- sies. The enzyme 15-lipoxygenase-1, however, was not expressed in any of 10 biopsies representing nine different subtypes of non-Hodgkin lym- phoma. In essence, the expression of 15-lipoxygenase-1 and the putative formation of eoxins by Hodgkin Reed–Sternberg cells in vivo are likely to contribute to the inflammatory features of Hodgkin lymphoma. These find- ings may have important diagnostic and therapeutic implications in Hodg- kin lymphoma. Furthermore, the discovery of the high 15-lipoxygenase-1 line comprises a activity in L1236 cells demonstrates useful model roles of 15-lipoxygenase-1.

eosinophilia,

Abbreviations cHL, classical HL; DiHETE, dihydroxy-eicosatetraenoic acid; EBV, Epstein–Barr virus; EX, eoxin; HETE, hydroxy-eicosatetraenoic acid; HL, Hodgkin lymphoma; IL, interleukin; LO, lipoxygenase; LT, leukotriene; MC, mixed cellularity; NHL, non-Hodgkin lymphoma; NS, nodular sclerosis.

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4222

Hodgkin lymphoma (HL), although a neoplastic dis- ease, has many features of an infection ⁄ inflammatory condition with symptoms and signs such as fever, night lymphadenopathy and splenomegaly. sweats, itching, The laboratory findings in this disease include neutro- lymphocytopenia and altered philia, serum phase reactants. It is now established that the large majority of classical HL (cHL) cases are B-cell

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

to be a potent chemotactic agent for human eosino- phils [14]. Thus, mediators formed via the 15-LO-1 pathway can induce inflammatory reactions and influence the immune system in man. There are also indications, however, that 15-LO-1 may have an anti- inflammatory role because this enzyme can be involved in the formation of lipoxins [15].

15-Lipoxygenase [17]. lens

lymphomas, which are characterized by the presence of Hodgkin Reed–Sternberg (H-RS) cells. These trans- formed cells appear to originate from pre-apoptotic germinal centre B cells that have lost their capacity to express a high-affinity B-cell receptor [1]. Importantly, H-RS cells constitute only a small minority (approxi- mately 1%) of the cell population in HL-affected tissue. The inflammatory cellular infiltrate in HL tumour tissue is rather heterogeneous, consisting of lymphocytes, mac- rophages, eosinophils, mast cells, plasma cells, stromal cells and fibroblasts. There is strong evidence that these infiltrating cells are involved in an inflammatory ⁄ reac- tive process creating an environment that allows, and probably promotes, the survival of H-RS cells [2]. Cytokines and chemokines operating in a complex inter- action have been suggested to be involved in the patho- genesis of HL [3]. A number of studies indicate that the release of cytokines and other biological active media- tors from H-RS cells plays an important role in the pathophysiology of HL. Among cytokines, interleukin (IL)-13 has been proposed to act as an autocrine growth factor for H-RS cells [4,5].

The enzyme 15-LO-1 not only metabolizes free fatty acids, but also can oxygenate phospholipids located in the cell membrane, and these oxidated phospholipids might contribute to 15-LO-1 signaling in inflammation [6,16]. The enzyme has been proposed to play a role in reticulocyte maturation through breakdown of mito- chondria membranes [6,12]. The enzyme may, how- ever, play a more general role in the differentiation of cells, including the maturation of keratinocytes and the is predominantly eye expressed in human eosinophils, activated monocytes, airway epithelial cells, reticulocytes and mast cells [6,18–20]. The Th2 cytokines IL-4 and IL-13 induce the expression of 15-LO-1 in monocytes, airway epithe- lial cells and mast cells [20–23]. Demethylation of the 15-LO-1 promoter is a prerequisite for gene activation [24].

In light of the characteristic inflammatory features of cHL, it was of interest to investigate the expression of lipoxygenases in H-RS cells and the formation of arachidonic acid metabolites by these cells.

Results

Expression and localization of 15-LO-1 in the HL cell lines

The mammalian lipoxygenases are a family of struc- turally related enzymes, catalyzing the oxygenation of arachidonic acid [6]. 5-Lipoxygenase (LO) catalyses the conversion of arachidonic acid to leukotriene (LT)A4 which can be further converted to LTB4 or LTC4. The latter metabolite can be further metabolized to LTD4 and LTE4 and these mediators are potent proinflam- matory mediators and bronchoconstrictors [7]. There are two forms of 15-LO, named type 1 and 2 [6,8]. It has been shown that patients with asthma and airway inflammation have increased expression of 15-LO-1 protein and increased activity of 15-LO-1 in the lung compared to healthy subjects [9,10].

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4223

Only recently, we reported on the formation of eox- ins (EX) in human eosinophils, cord blood derived mast cells and surgically removed nasal polyps. The enzyme 15-LO-1 catalyses the conversion of arachi- donic acid to EXA4, which in turn can be conjugated with glutathione, leading to the formation of EXC4 [11]. This metabolite can be further metabolized to EXD4 and EXE4. Eoxins induce increased permeability of the endothelial cell monolayer in vitro, indicating that they can modulate and enhance vascular perme- ability, a hallmark of inflammation [11]. It has been known for many years that 15-LO-1 can also catalyze the formation of 15(S)-hydroxy-eicosatetraenoic acid (HETE), 8(S,R),15(S)-dihydroxy-eicosatetraenoic acid (DiHETE), 5(S),15(S)-DiHETE and 14(R),15(S)-DiH- ETE [6,12]. This latter metabolite has been reported to inhibit natural killer cell activity [13]. Another 15-LO-1 derived mediator, 5-oxo-15-hydroxy-ETE, was found The large majority of H-RS cells are derived from germinal centre B lymphocytes [1]. Because human B it was of interest to lymphocytes express 5-LO [25], lines also expressed this determine whether HL cell enzyme. For that purpose, the metabolism of arachi- donic acid was examined in the HL cell lines L1236, L428, KMH2 and L570. These cell lines produced no or very low amounts of 5-HETE or LTs after incuba- tion with arachidonic acid in the presence or absence of calcium ionophore (data not shown). Incubation of L1236 cells with arachidonic acid for 5 min, however, led to the formation of a major product that cochro- In matographed with synthetic 15-HETE (Fig. 1). addition, a minor peak (UV absorbance maximum at 236 nm) that coeluted with synthetic 12-HETE was observed. The UV spectra of the materials in these peaks are in agreement with the reported spectra for 15-HETE and 12-HETE, respectively. Chiral chroma- formed tography analysis demonstrated that the

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

respectively);

Fig. 1. RP-HPLC chromatograms of the products formed by L1236 cells after incubation with arachidonic acid (40 lM) for 5 min. UV monitoring was carried out at 236 nm. The retention times for syn- thetic 15(S)-HETE, 12(S)-HETE and 5(S)-HETE are indicated. Inset: UV spectra of the materials that coeluted with synthetic 15(S)-HETE and 12(S)-HETE, respectively. Solid line, cells incubated with arachi- donic acid; broken line, cells incubated without arachidonic acid.

of these these products. Incubation

Fig. 2. Western blot and assay of 15-LO-1 activity after subcellular fractionation of L1236 cells. The cells were washed twice in NaCl ⁄ Pi without Ca2+ ⁄ Mg2+ and resuspended in 1 mL of NaCl ⁄ Pi without Ca2+ ⁄ Mg2+. Five million L1236 cells were added to three Eppendorf tubes and the buffer was changed to: (A) NaCl ⁄ Pi with Ca2+ ⁄ Mg2+ (0.9 and 0.5 mM, (B) NaCl ⁄ Pi with Ca2+ ⁄ Mg2+ plus calcium ionophore A23187 (final concentration 5 lM); and (C) NaCl ⁄ Pi without Ca2+ ⁄ Mg2+. After 10 min of incuba- tion at 37 (cid:2)C, the samples were homogenized by sonication three times for 10 s on ice, using a Sonics vibracell VC750 with 30% amplitude. The cell suspensions were centrifuged for 10 min at 1500 g at 4 (cid:2)C and the supernatants were transferred to new tubes for ultracentrifugation (100 000 g at 4 (cid:2)C) for 1 h. The super- natants were collected and the pellets were resuspended by soni- cation in the same buffer as used during the incubation. Lower panel: western blot analysis. An aliquot from each fraction equal to 40 000 cells was loaded on the NuPAGE 4–12% Bis-Tris gradient gel (1 mm) with running buffer followed by western blotting using antibodies raised against of 15-LO-1 (see Experimental procedures). Upper panel: an aliquot of each fraction was incubated with 40 lM arachidonic acid for 10 min at room temperature. The reaction was terminated by the addition of three volumes of methanol and the amounts of 15(S)-HETE and 12(S)-HETE were analysed by RP-HPLC. The results are the mean ± SD of three separate exp- eriments. P, 100 000 g pellet; S, 100 000 g supernatant.

cells

cells were

products were exclusively 15(S)-HETE and 12(S)- HETE (data not shown). The ratio of approximately 9 : 1 between 15(S)-HETE versus 12(S)-HETE was also in agreement with a 15-LO-1 catalyzed formation of cells (40 · 106 cellsÆmL)1) with arachidonic acid (final con- centration 40 lm) for 10 min led to the formation of 294 ± 172 pmol 15-HETE per 106 (n = 9; mean ± SD). No formation of 12- or 15-HETE was observed in the other examined HL cell lines after incubation with arachidonic acid with or without the calcium ionophore A23187. Linoleic acid is an excel- lent substrate for 15-LO-1 and L1236 cells also efficiently metabolized linoleic acid to 13-hydroxy-octa- decadienoic acid (data not shown), which also is in agreement with the expression of 15-LO-1 in these cells.

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4224

cells were incubated with NaCl ⁄ Pi with Ca2+ ⁄ Mg2+; NaCl ⁄ Pi with Ca2+ ⁄ Mg2+ plus calcium ionophore A23187 (final concentration 5 lm); or calcium-free NaCl ⁄ Pi. Subsequently, sonicated the followed by subcellular fractionation. The separate fractions were analyzed by SDS ⁄ PAGE followed by western blotting (Fig. 2, lower panel). In the presence of Ca2+ ⁄ Mg2+, with or without calcium ionophore, the majority of 15-LO-1 was found in the membrane fraction, although significant amounts were also detected in the supernatant fraction. In the absence of RT-PCR analysis revealed mRNA expression of 15-LO-1 but not of 15-LO-2 in L1236 cells (data not shown). To demonstrate the expression of the 15-LO-1 protein in L1236 cells and the cellular localization of the enzyme in the presence and absence of calcium, the

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

A

Ca2+ ⁄ Mg2+, almost all 15-LO-1 protein was detected in the supernatant fraction. The translocation of 15-LO-1 to the membrane fraction, upon calcium challenge, is in accordance with the findings reported in eosinophils [26]. The activity of 15-LO-1 in the separate fractions, measured as conversion of arachi- donic acid to 15- and 12-HETE (Fig. 2, upper panel), did not correlate well with the approximated amounts of 15-LO-1 detected by immunoblotting (Fig. 2, lower panel). This might be due to the presence of more free (cytosolic) enzyme in samples incubated without cal- cium and the cytosolic enzyme converting exogenously added arachidonic acid to 15-HETE and 12-HETE more efficiently than the membrane bound enzyme.

B

small cell population consisting of

giant

Immunostaining of L1236 cells revealed a hetero- to geneous large blasts and medium-sized mononuclear cells, multinucleated Immunocytochemical cells. 15-LO-1 staining of non-activated L1236 cells was in agreement with the western blot results and showed diffuse strong cytoplasmic staining of 15-LO-1 in both small cells and in multinuclear large cells (Fig. 3). Taken together, the results demonstrated that L1236 cells contain abundant amounts of active 15-LO-1.

raised against

Formation of eoxins by L1236 cells

Fig. 3. Immunocytochemical analysis of 15-LO-1 expression in L1236 cells. Cytocentrifuged, paraformaldehyde-fixed L1236 cells were analyzed for 15-LO-1 expression by the avidin-biotin complex alkaline phosphatase method. (A) L1236 cells were stained with an antiserum (diluted 1 : 1000) recombinant human 15-LO-1 (red colour). (B) Pre-immune serum (diluted 1 : 1000) was used as a negative control. Original magnification ·46.

shown in Fig. 4. Four peaks

identical

To determine whether L1236 cells could produce eoxins, an acetonitrile based mobile phase was used to improve separation of peaks with retention times of approximately 5–8 min (Fig. 1). A typical reverse phase HPLC chromatogram of products formed by L1236 cells after 5 min of incubation with arachidonic (1–4) were acid is observed in a cluster, all containing a conjugated triene spectrum and a UV absorbance maximum at 268 nm. These peaks had elution times corresponding to syn- thetic standards of the four 8(R,S),15(S)-DiHETE (the two double oxygenation metabolites and the two derived from non-enzymatic degradation of EXA4; also named 14,15-LTA4) [12,19]. The pattern formed by these metabolites was almost to that reported for human airway epithelial cells and eosin- ophils incubated with arachidonic acid [11,19]. The material in peak 5 (Fig. 4) had the same retention time as 14(R),15(S)-DiHETE. The cells were also found to produce 5(S),15(S)-DiHETE (data not shown), which can be converted to the chemotactic metabolite 5-oxo- 15-ETE [14,27].

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4225

due to the high 15-LO-1 activity in these cells and the chemical structures of eoxins [11]. The material in peaks II and IV also coeluted with synthetic EXC4 and EXD4, respectively. Furthermore, the UV spectra of the materials in peaks II and IV were in agreement with the UV spectra for eoxins (Fig. 4, inset) [11]. To further analyze the identity of the materials in peak II and IV, the cells were incubated with arachidonic acid for 2 or 10 min followed by analysis with positive ion LC-MS ⁄ MS. The analysis showed that the material in peak II and IV had identical MS ⁄ MS spectra to syn- thetic EXC4 and EXD4 (Fig. 5) [11]. In addition, two other minor peaks with conjugated triene spectra and a UV absorbance maximum at 278 nm were observed (peaks I and III). Metabolite III was analyzed by positive ion ESI-MS and MS ⁄ MS and the spectrum of The two major peaks in the chromatogram also possessed a conjugated triene spectrum but had a UV absorbance maximum at 282 nm (peaks II and IV). This indicated that these cells also produced eoxins

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

Fig. 6. RP-HPLC chromatogram of the products formed by L1236 cells after incubation with EXA4 (2 lM) for 30 min. The numbers 1, 3 and 5 correspond to retention times of synthetic standards for 14(R),15(S)-DiHETE, 8(S),15(S)-DiHETE and 8(R),15(S)-DiHETE, respectively. Peaks I–IV had identical retention times to the peaks shown in Fig. 4. Inset: UV spectrum of the material in peak V.

the material

Fig. 4. RP-HPLC chromatogram of the products formed by L1236 cells after incubation with arachidonic acid (40 lM) for 5 min. The numbers 1–4 correspond to the retention times of synthetic stan- dards for the four isomers of 8(S,R),15(S)-DiHETE and number 5 corresponds to 14(R),15(S)-DiHETE, respectively. The material in peaks I–IV was further analysed by LC-MS ⁄ MS. Inset: UV spectra for in peaks II and IV which also coeluted with synthetic standards of EXC4 and EXD4, respectively.

Fig. 5. Positive ion LC-MS ⁄ MS spectra of metabolites produced by (A) MS ⁄ MS L1236 cells after incubation with arachidonic acid. spectrum of the material corresponding to peak II in Fig. 4, pro- duced by L1236 cells after incubation with arachidonic acid (40 lM) for 2 min, compared to the spectrum of synthetic EXC4 ([M+H]+ 626). (B) MS ⁄ MS spectra of the material corresponding to peak IV in Fig. 4, produced by L1236 cells after incubation with arachidonic acid (40 lM) for 10 min, compared to the spectrum of synthetic EXD4 ([M+H]+ 497).

and UV profile of this metabolite are consistent with an all-trans triene structure because the spectrum is shifted 4 nm hypsochromically compared to the spec- trum of EXD4 [28]. The material in peak III is there- fore probably 8-trans-EXD4. The material in peak I also had a conjugated triene spectrum and a UV absorbance maximum at 278 nm but the amount of the material was insufficient to obtain an interpretable ESI-MS ⁄ MS spectrum. In line with the formation of 8-trans-EXD4, it is likely, however, that this metabolite is 8-trans-EXC4. Furthermore, incubation of the cells with synthetic EXC4 and EXD4 led to the formation in agreement of metabolites I and III, respectively, with the postulated all-trans trienes, formed from the parent EXC4 and EXD4, respectively (data not shown).

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4226

To determine the mechanism of formation of EXC4 and EXD4, as well as the putative formation of EXE4, the cells were incubated with synthetic 14,15-epoxy- 5,8,10,12-(Z,Z,E,E)-eicosatetraenoic acid (EXA4) for 30 min. Figure 6 shows that EXA4 was converted to EXC4 (peak II) and EXD4 (peak IV). In addition, the material in peak V also contained a conjugated triene and had a UV absorbance maximum at 282 nm. This metabolite was not detected when the cells were incu- in bated with arachidonic acid because the material this peak V coeluted on HPLC with the double dioxy- genation product 8(S),15(S)-DiHETE (Fig. 4). This product is not formed from EXA4 and metabolite V was now visible on the HPLC chromatogram and no longer hidden behind the double dioxygenation product. Metabolite V was also formed when the cells this metabolite was identical to the spectrum for EXD4 (data not shown). The exact structure of this metabo- lite has not been established but the MS ⁄ MS spectrum

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

Fig. 7. Positive ion LC-MS ⁄ MS spectrum of a metabolite produced by L1236 cells after incubation with arachidonic acid for 30 min. Positive ion LC-MS ⁄ MS spectrum of the material corresponding to peak V in Fig. 6, formed by L1236 cells after 30 min of incubation with arachidonic acid (40 lM). The lower panel shows the MS ⁄ MS spectrum of authentic EXE4 ([M+H]+ 440).

the Fig. 8. Time course (A) and dose–response (B) curves of formation of EXC4 and EXD4 by L1236 cells after incubation with arachidonic acid. (A) The concentration of arachidonic acid was 40 lM. (B) Five minutes of incubation with the indicated concentra- tion of arachidonic acid. One typical experiment out of three is shown. d, EXC4; s, EXD4.

were incubated with synthetic EXD4 for 30 min (data not shown). Therefore, L1236 cells were incubated with arachidonic acid for 30 min followed by analysis with LC-MS ⁄ MS. Figure 7 demonstrates that the MS ⁄ MS spectrum corresponding to peak V contains the charac- teristic 205 m ⁄ z fragment ion of eoxins and is identical to synthetic 14(R)-cysteinyl-15(S)-hydroxy-5,8,10,12- (Z,Z,E,E)-eicosatetraenoic acid (EXE4). Metabolite V also had an identical retention time as synthetic EXE4 on HPLC. In addition, the materials in peaks II, IV and V were separately collected and analyzed by positive ion ESI-MS ⁄ MS with a triple quadrupole mass spectrome- ter, as well as by negative ion ESI-MS ⁄ MS using a hybrid magnetic sector ⁄ TOF instrument. These analyses also demonstrated that the materials in peaks II, IV and V were EXC4, EXD4 and EXE4, respectively (data not shown). Taken together, these results show that L1236 cells convert arachidonic acid to EXC4, which is readily converted to EXD4 and EXE4.

increased with the concentration of arachidonic acid and a plateau was reached at a concentration of 40 lm arachidonic acid.

Expression of 15-LO-1 in tumour biopsies from HL and non-Hodgkin (NHL) patients

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4227

The time courses of the formation of EXC4 and EXD4 after incubation of L1236 cells with arachidonic acid are shown in Fig. 8A. A substantial amount of EXC4 (37 pmol per 106 cells) was produced already by 10 · 106 cells after 30 s of incubation with arachidonic acid (40 lm). The maximal level of EXC4 was observed after 5 min of incubation and, subsequently, the level of EXC4 declined with time. Significant amounts of EXD4 were observed after 2 min of incubation with arachidonic acid and the levels increased with time, reaching a maximal level after 30 min. It was not pos- sible to measure the level of EXE4 in this experiment because 8(S),15(S)-DHETE co-chromatographed with EXE4 in this HPLC system. The dose–response curves of the formation of eoxins from arachidonic acid show that significant amounts of EXC4 and EXD4 were formed already at a concentration of 1 lm arachidonic these metabolites acid (Fig. 8B). The levels of To determine whether H-RS cells also expressed 15-LO-1 in vivo, diagnostic biopsies from HL lymph nodes were stained with an antibody raised against 15-LO-1. In most HL tumours, there was a distinct cytoplasmic positivity for 15-LO-1 in tissue macro- phages and a strong staining in eosinophils. In 17 of 20 tumours, 15-LO-1 expression could also be detected in H-RS cells (Table 1). Figure 9A shows a typical H-RS cell, strongly stained by an antibody raised against human recombinant 15-LO-1. No staining was

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

lymphoma; MALT, mucosa-associated lymphoid tissue lymphoma; MCL, mantle cell

Table 1. Patient and tumor characteristics. The degree of eosinophilia and the number of H-RS cells in HL tumors were determined by ran- dom selection of 10 consecutive HPFs. In each HPF, the number of eosinophils and H-RS cells was determined and the sum of ten HPF was calculated. The biopsies were then classified as low eosinophilia (< 50 eosinophils per 10 HPF), medium eosinophilia (50–120 eosinoph- ils per 10 HPF) or high eosinophilia (> 120 eosinophils per 10 HPF). The number of H-RS cells were classified as few (< 5 H-RS cells per 10 HPFs), medium (5–10 H-RS cells per 10 HPFs) or many (> 10 H-RS cells per 10 HPFs). Macrophages and eosinophils were used as internal positive controls for 15-LO-1 staining. ), no H ⁄ RS cells expressed 15-LO-1; +, < 20% of the H ⁄ RS cells expressed 15-LO-1; ++, > 20% of the H ⁄ RS cells expressed 15-LO-1; +++, the majority of the H ⁄ RS cells strongly expressed 15-LO-1. PTCL, peripheral T-cell lymphoma; DLBCL, diffuse large B-cell lymphoma; B-CLL, B-cell chronic lymphocytic leukemia; BL, Burkitt lymphoma; FCDL, follicular centre derived lymphoma; PMBCL, primary mediastinal B-cell lymphoma; NA, not applicable; ND, not determined.

Patient number

Sex

Age (years)

Clinical stage

EBV status

H-RS cells

Eosinophilia

15-LO-1 expression

Tumor

++ ++ )

++ ++ ++ )

++ ++ ++ + + +++ ++ + )

Male Female Male Female Female Female Male Female Male Female Male Male Male Male Male Male Female Female Female Male

66 45 27 25 61 14 13 28 19 31 89 46 24 23 8 36 5 77 70 37

HL NSI HL NSI HL NSI HL NSI HL NSI HL NSI HL NSI HL NSII HL NSII HL NSII HL MC HL MC HL MC HL MC HL MC HL MC HL MC HL MC HL MC HL MC

ND ND Pos Neg Neg Neg Neg ND ND Neg Pos ND ND ND Pos Pos Pos ND Neg Neg

IIIA IVA IVB IIA IIIA IIIA IIB IIB IIB IIIB IA IA IVB IIB IIA IIIB IIA IA IA IIIB

Few Medium Few Medium High Medium Medium Many Many Many Few Many Many Many Medium Few Many Medium Medium High

Low Medium Low Low Low High High High High High Low High High Medium High Low High Low Medium High

+ + + ++

) ) ) ) ) ) ) ) ) )

HL 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 NHL 21 22 23 24 25 26 27 28 29 30

Female Male Female Female Female Male Male Male Female Male

39 80 60 54 75 81 63 32 72 28

PTCL DLBCL DLBCL MALT Immunocytoma MCL B-CLL BL FCDL PMBCL

ND ND ND ND ND ND ND ND ND ND

IIA IIA IVB IA IVB IVA Rai IV IIA IVA IVB

NA NA NA NA NA NA NA NA NA NA

ND ND ND ND ND ND ND ND ND ND

is therefore very likely that

lymphoma. Thus, H-RS cells express 15-LO-1 in vivo and it these cells can produce eoxins and other 15-LO-1 derived metabolites in vivo, contributing to the inflammatory features of HL.

Discussion

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4228

The present study shows that the cHL-derived cell line L1236 possesses high 15-LO-1 activity and readily converts arachidonic acid to eoxins, a recently observed with the pre-immune sera (Fig. 9B). The strongest staining was noted in biopsies from three cases of nodular sclerosis (NS) subtype II HL and in one case of mixed cellularity (MC) HL, all with marked high eosinophilia. However, there were also cases with high eosinophilia and weak staining of 15-LO-1 in H-RS cells (Table 1). By contrast, no stain- ing of 15-LO-1 was observed in tumour biopsies from ten patients with nine different subtypes of NHL (Table 1). We have not yet, however, examined the expression of 15-LO-1 in all different entities of human

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

A

B

Fig. 10. Overview of the metabolic pathway for the formation of eoxins in L1236 cells.

Fig. 9. Immunohistochemical analysis of 15-LO-1 expression by pri- mary H-RS cells. Paraffin sections from formalin-fixed HL tumours were stained with 15-LO-1 antibody (diluted 1 : 1000), raised against recombinant 15-LO-1, using the avidin-biotin complex alka- line phosphatase method. (A) HL biopsy of NS subtype with a typi- in the middle (red colour), cal 15-LO-1 positive H-RS cell surrounded of by several macrophages positive for 15-LO-1 (red colour). (B) Pre-immune serum (diluted 1 : 1000) was used as a negative control. Original magnification ·46.

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4229

NMR in the present study. However, because these metabolites were also formed from synthetic 14,15-epoxy-5,8,10,12 (Z,Z,E,E)-eicosatetraenoic acid (EXA4) and had identical retention times on HPLC as the corresponding synthetic metabolite (Figs 4 and 6), the structures of these metabolites are likely to be 14(R)-glutathionyl-15(S)-hydroxy-5,8,10,12-(Z,Z,E, E)-eicosatetraenoic acid (EXC4), 14(R)-cysteinyl-glycyl- 15(S)-hydroxy-5,8,10,12-(Z,Z,E,E)-eicosatetraenoic acid (EXD4) and 14(R)-cysteinyl-15(S)-hydroxy-5,8,10,12- (Z,Z,E,E)-eicosatetraenoic acid (EXE4), respectively. Taken together, the UV profile and mass spectra show that L1236 cells can metabolize arachidonic acid to EXA4 which in turn can be readily converted to EXC4, EXD4 and EXE4 (Fig. 10). Eoxin C4 was more rapidly converted to EXD4 and EXE4 in L1236 cells than in human eosinophils [11], indicating that L1236 identified group of proinflammatory cysteinyl-contain- ing arachidonic acid metabolites produced by human eosinophils and mast cells [11]. The structures of the metabolites produced by L1236 cells were determined by positive ion LC-MS ⁄ MS. The interpretation of the spectra showed that the structure of these metabolites corresponded to EXC4, EXD4 and EXE4 (Figs 5 and 7). These metabolites also co-chromatographed with synthetic EXC4, EXD4 and EXE4 on RP-HPLC, respectively. In addition, two minor products were formed which were postulated to be 8-trans-EXC4 and 8-trans-EXD4. Furthermore, the cells produced a series of 8(R,S),15(S)-DiHETEs and 14(R),15(S)-DiHETE (Fig. 4). The configurations of the double bonds in EXC4, EXD4 and EXE4 were not determined with

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

dipeptidase, which probably catalyse

cells express higher amounts of c-glutamyltransferase and the conversion of EXC4 to EXD4 and EXD4 to EXE4, respectively.

induce inflammatory features of HL. The eoxins increased vascular permeability and 5-oxo-15-hydroxy- ETE is chemotactic for eosinophils, which infiltrate many HL tumours [2]. In addition, H-RS cells can produce 15-HETE and this metabolite can exert either proinflammatory or anti-inflammatory effects in vari- ous model systems [6,12]. Thus, H-RS cells in vivo have the putative capacity to produce biological active ara- chidonic acid metabolites that might play an important role in the pathophysiology of HL.

The formation of the giant H-RS cells in HL is still a mystery although many attempts have been made to these cells. The enzyme clarify the formation of 15-LO-1 has been proposed to induce breakdown of mitochondria membranes during the differentiation of reticulocytes to erythrocytes and to degrade intracellu- lar membranes during the differentiation of keratino- cytes and the eye lens [6,17]. It is therefore tempting to speculate that 15-LO-1 might be involved in the forma- tion of the giant H-RS cells through remodelling of intracellular membranes. A cell

Western blot analysis demonstrated that 15-LO-1 was mainly cytosolic under calcium-free conditions and in non-activated cells (Figs 2 and 3). The enzyme was, however, mainly located to the cell membrane in the presence of calcium (Fig. 2). There was higher 15-LO-1 activity in the cytosolic fraction, although western blot analysis demonstrated a higher amount of 15-LO-1 protein in the membrane fraction. This was an unexpected finding because membrane bound 15-LO-1 has been found to possesses higher activity than the cytosolic enzyme [26,29,30] This might be due to the orientation of the enzyme in the membrane of L1236 cells and the exogenous addition of arachidonic acid reaching the active site of the cytosolic enzyme to a greater extent than that of the membrane enzyme in L1236 cells [31]. Alternatively, the membrane associ- ated enzyme in L1236 cells undergoes suicidal inactiva- tion during the oxygenation of membrane lipids. In addition, 15-LO-1 might be inactivated by 15-HPETE, which is probably rapidly inactivated in the cytosol, whereas 15-HPETE generated by membrane bound 15-LO may persist and inactivate the membrane-bound enzyme. line with high 15-LO-1 activity has been sought subsequent to the discovery of this enzyme more than 30 years ago. The identification of 15-LO-1 in L1236 cells, which possess high 15-LO-1 activity without addition of exogenous IL-4 or IL-13, opens up great possibilities for all researchers studying the func- tion of 15-LO-1.

In summary, the present study demonstrates that the HL cell line L1236 produces eoxins and that primary H-RS cells in vivo express 15-LO-1 and also have the putative capacity to produce eoxins that might contrib- ute to the inflammatory features of HL. These findings may have important diagnostic and therapeutic impli- cations.

Immunohistochemistry analysis of diagnostic HL lymph node biopsies demonstrated the expression of 15-LO-1 in H-RS cells in 17 of the 20 (85%) examined biopsies (Table 1). By contrast, 15-LO-1 was not expressed in biopsies derived from ten patients repre- senting nine different entities of NHL (Table 1). Thus, the expression of 15-LO-1 might be a useful biomarker to distinguish HL from NHL. Extensive studies on the expression of 15-LO-1 in biopsies derived from various subtypes of NHL are ongoing.

Experimental procedures

Materials

Synthetic 14,15-epoxy-5,8,10,12-(Z,Z,E,E)-eicosatraenoic acid 14(R)-glutathionyl-15(S)-hydroxy-5,8,10,12-(Z,Z, (EXA4), E,E)-eicosatetraenoic acid (EXC4), 14(R)-cysteinyl-glycyl-15 (S)-hydroxy-5,8,10,12-(Z,Z,E,E)-eicosatetraenoic acid (EXD4), 14(R)-cysteinyl-15(S)-hydroxy-5,8,10,12-(Z,Z,E,E)-eicosatet- raenoic acid (EXE4), 8(R),15(S)-dihydroxy-5,9,11,13-(Z,E, E,E)-eicosatetraenoic acid [8(R),15(S)-DiHETE], 8(S),15(S)- dihydroxy-5,9,11,13-(Z,E,E,E)-eicosatetraenoic acid [8(S), 15(S)-DiHETE], 8(R),15(S)-dihydroxy-5,9,11,13-(Z,E,Z,E)- eicosatetraenoic acid [8(R),15(S)-DiHETE], 8(S),15(S)-di- hydroxy-5,9,11,13-(Z,E,Z,E)-eicosatetraenoic acid [8(S),15 (S)-DiHETE], 14(R),15(S)-dihydroxy-5,8,10,12-(Z,Z,E,E)- eicosatetraenoic acid [14(R),15(S)-DiHETE], 15(S)-hydroxy-5,

Of all the HL cell lines investigated, the L1236 cell line was the only one that expressed 15-LO-1 like primary H-RS cells in vivo. Among all the established HL cell lines, however, only the L1236 cell line is clon- ally related to original tumour tissue [32,33]. The tran- scription of 15-LO-1 is dependent on the transcription factor STAT6 [6,23]. Through autocrine IL-13 stimula- tion, several HL cell lines, such as L1236 and L428, possess constitutively activated STAT6 [34,35]. Studies are now ongoing to clarify whether the epigenetic status of the 15-LO-1 promoter region in L1236 cells, which express 15-LO-1, and L428 cells, which do not express 15-LO-1, is different because only L1236 express 15-LO-1.

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4230

The expression of 15-LO-1 and putative formation of eoxins by H-RS cells in vivo might contribute to the

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

acid eicosatetraenoic

[15(S)-HETE], 8,11,13-(Z,Z,Z,E)-eicosatetraenoic 12(S)-hydroxy-5,8,10,14-(Z,Z,E,Z) acid [12(S)-HETE] and 13(S)-hydroxy-9,11-(Z,E)-octadecadie- noic acid [13(S)-HODE] were from Biomol Inc. (Plymouth Meeting, PA, USA). Cell culture medium RPMI 1640, fetal bovine serum, penicillin, streptomycin and glutamine were obtained from Gibco BRL (Gaithersburg, MD, USA). HPLC solvents were obtained from Rathburn Chemicals (Walkerburn, UK). Complete mini EDTA-free protease inhibitor cocktail was purchased from Roche (Indianapolis, IN, USA).

amplification forward 5¢-GAA GTT ATC AGT CGA CAT GGG TCT CTA CCG CA-3¢ and reverse 5¢-ATG GTC TAG AAA GCT TTT AGA TGG CCA CAC TGT TTT CCA CCA C-3¢ primers were used. For partial 15-LO-2 amplification, forward 5¢-AAT CTC GGC AAGGAG TTC ACT-3¢ and reverse 5¢-AGT CAA ACT GCC CTG CAC T-3¢ primers were used. b2-microglobulin was included as an internal control for RT efficiency and RNA integrity. AmpliTaq Gold(cid:3)DNA Polymerase (Applied Biosystems, Foster City, CA, USA) was used for DNA amplification. The PCR conditions were: 95 (cid:2)C for 10 min, 95 (cid:2)C for 15 s, 58 (cid:2)C for 1 min, 72 (cid:2)C for 2 min and 72 (cid:2)C for 6 min. b2-microglobulin was run for 24 cycles and 15-LO-1 and 15-LO-2 for 34 cycles.

Cell lines, biopsies and patients

in

the

respective

resuspending

The human HL cell lines L1236, HDLM2, KMH2 and L428 (kind gifts from V. Diehl, Department of Internal Medicine, University Hospital of Cologne, Germany) were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, l-glutamine (2 mm), penicillin (100 UÆmL)1) and streptomycin (100 lgÆmL)1) (Gibco, Paisley, Scotland, UK) at 37 (cid:2)C in an atmosphere containing 5% CO2. The examined cell lines were of B-cell phenotype (L1236, L428 and KMH2) and T-cell phenotype (HDLM2), and derived from cHL of MC (L1236 and KMH2) and NS (L428 and HDLM2) subtypes, respec- tively. All cell lines were negative for Epstein–Barr virus (EBV). Diagnostic HL-involved lymph node biopsies were collected from 1994 to 2004 at the Department of Pathol- ogy and Cytology, Karolinska University Hospital Solna, Stockholm, Sweden. Routine morphological and immuno- histochemical stainings of tumour biopsises were performed on paraffin sections. EBV expression was investigated by immunohistochemistry (latent membrane protein 1) and in situ hybridization (EBV encoded RNA) as described previously [36].

the

The degree of eosinophilia and the number of H-RS cells were determined by random selection of ten consecutive high power fields (HPFs) in hematoxylin and eosin stained paraffin sections. In each HPF, the number of eosinophils or H-RS cells was determined and the sum of ten HPF was calculated. The biopsies were then classified as low eosino- philia (< 50 eosinophils per ten HPF), medium eosino- philia (50–120 eosinophils per 10 HPF) or high eosinophilia (> 120 eosinophils per ten HPF). The number of H-RS cells were classified as few (< 5 H-RS cells per ten HPF), medium (5–10 H-RS cells per ten HPF) or many (> 10 H-RS cells per ten HPF). This study was approved by the local ethic committee of Karolinska University Hospital.

Translocation assay and western blot

in NaCl ⁄ Pi without L1236 cells were washed twice Ca2+ ⁄ Mg2+and resuspended in 1 mL of NaCl ⁄ Pi without Ca2+ ⁄ Mg2+. Five million L1236 cells were added to three Eppendorf tubes and the buffer was changed by spinning down the cells for 5 min at 600 g, removing the supernatant and buffers cells [buffer A: NaCl ⁄ Pi with Ca2+ ⁄ Mg2+ (0.9 and 0.5 mm, respectively); buffer B: NaCl ⁄ Pi with Ca2+ ⁄ Mg2+ plus cal- cium ionophore A23187 (final concentration 5 lm); and buffer C: NaCl ⁄ Pi without Ca2+ ⁄ Mg2+]. One tablet com- plete mini protease inhibitor without EDTA was added to 10 mL of the respective buffers. After 10 min of incubation at 37 (cid:2)C, the samples were homogenized by sonication, using a Sonics vibracell VC750 (Chemical Instruments AB, Lidingo¨ , Sweden) with 30% amplitude, three times for 10 s on ice. The homogenate was centrifuged for 10 min (1500 g at 4 (cid:2)C) and the supernatant was transferred to new tubes for ultracentrifugation, at 100 000 g at 4 (cid:2)C for 1 h. The supernatant was collected and the pellet was resuspended by sonication in the same buffer as used during the incubation. Western blot analysis was performed according to the NuPAGE reduced sample protocol (Invitrogen, Carlsbad, CA, USA). An aliquot from each fraction equal to 40 000 cells was loaded on the NuPAGE 4–12% Bis-Tris gradient gel (1 mm) with running buffer. After transfer to a polyv- inylidene difluoride membrane, it was blocked in 5% milk powder in NaCl ⁄ Pi-Tween at room temperature for 1 h. The membrane was then incubated overnight at 4 (cid:2)C with a 15-LO-1 rabbit peptide antiserum (batch 632) raised against 15-LO-1 peptide CALDKEIEIRNAKLD MPYEY (dilution 1 : 5000) (antibodies made by Innovagen AB, Ideon, Lund, Sweden) in NaCl ⁄ Pi-Tween + 1% milk powder. These antibodies did not detect human 5-LO, platelet 12-LO or 15-LO-2 (data not shown). After washing three times in NaCl ⁄ Pi-Tween, the membrane was incu- bated with the anti-rabbit horseradish peroxidase serum, diluted 1 : 10000 in NaCl ⁄ Pi-Tween + 1% milk powder for then 5 h at

room temperature. The membrane was

Total RNA was extracted from L1236 cells using RNeasy Mini kit (Qiagen GmbH, Hilden, Germany). One micro- gram of RNA was reverse-transcribed with reverse trans- length 15-LO-1 criptase and random hexamers. For full

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4231

RT-PCR

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

washed three times in NaCl ⁄ Pi-Tween before detection with ECL-plus. Recombinant human 15-LO-1 was expressed in Sf 9 cells and used as standard [29].

mobile phase B consisted of 80% acetonitrile in water and 0.1% acetic acid. All water used was of MilliQ grade (Milli- pore, Billerica, MA, USA). Starting isocratically for 1.5 min at 90% A was followed by a 6.5 min linear gradient reaching 100% of B. The system was washed at 100% B for 4 min and subsequently equilibrated at 90% A for 4 min.

The mass spectrometer was operated using an electro- spray atmospheric pressure ionization source in positive mode. The spray voltage was set to 4500 V, the capillary temperature was 375 (cid:2)C and sheath and auxiliary gas was optimal at 60 and 5 arbitrary units, respectively. Skimmer offset was at 10 V and tube lens was 99 V. MS ⁄ MS prod- uct ions of 626, 497 and 440 m ⁄ z, corresponding to the [M+H]+ of EXC4, EXD4 and EXE4, were recorded. Collision energy was set to 21 for MS ⁄ MS at 626 m ⁄ z and to 18 for MS ⁄ MS at both 497 and 440 m ⁄ z. The metabolite spectra were compared with those of synthetic standards.

Experimental conditions and HPLC analysis

The cell suspensions were centrifuged at 200 g for 5 min, suspended in NaCl ⁄ Pi and washed twice with NaCl ⁄ Pi (37 (cid:2)C) and subsequently solubilized in NaCl ⁄ Pi at a concentration of 10–20 · 106 cellsÆmL)1. The sample was pre-warmed for 2 min prior to addition of arachidonic acid, EXA4 or other indicated metabolites and subsequently incubated for the period indicated. The incubations were terminated by addition of three volumes of ice-cold metha- nol. Precipitated proteins and cell fragments were removed by centrifugation and excess water and methanol were evaporated under reduced pressure. The residues were solubilized in methanol and transferred to a test tube and evaporated to dryness under a stream of nitrogen. Subse- quently, the residues were reconstituted in the appropriate chromatographic solvent described below. The samples were applied to an octadecyl reverse-phase column (Nova- Pak C18 4 lm, 150 mm, Waters AB, Sollentuna, Sweden or Chromabond, C18, 100 mm, Macherey-Nagel, Du¨ ren, Germany for monohydroxy acids and cysteinyl-containing arachidonic acid metabolites, respectively), and eluted isocratically at a flow rate of 0.4 mLÆmin)1. The mobile phase was methanol ⁄ water ⁄ triflouroacetic acid (69 : 31 : 0.07, v ⁄ v ⁄ v) or acetonitrile ⁄ methanol ⁄ water ⁄ acetic acid (28 : 18 : 54:1, v ⁄ v ⁄ v, pH adjusted to 5.6 with NH3) for analysis of monohydroxy acids or cysteinyl-containing arachidonic acid metabolites and dihydroxy acids, respec- tively [11]. Eluted metabolites were detected and quantified utilizing a programmable Waters 991 diode array spectro- photometer connected to the HPLC system.

Immunostaining

Formalin-fixed and paraffin embedded lymph node biopsies were obtained for diagnostic purposes and used to study 15-LO-1 expression by the avidin-biotin complex alkaline phosphatase method [37]. The tumours studied were cHL NS (n = 10), cHL MC (n = 10) and ten NHL including nine different entities (Table 1) according to the WHO clas- sification [38]. A polyclonal 15-LO-1 antibody was raised in rabbit against recombinant human 15-LO-1 (the enzyme was expressed in Sf9 cells in house and purified prior to immunization, > 95% pure enzyme preparation was used). This antiserum (made by Innovagen AB) did not detect human 5-LO, platelet 12-LO or 15-LO-2 (data not shown). Briefly, antigen retrieval was performed by boiling deparaff- inized and rehydrated tissue sections in citrate buffer (pH 6.0) in a microwave oven for 20 min. The sections were for 1 h at incubated with 15-LO-1 antibody (1 : 1000), room temperature. Subsequently, the biopsies were incu- bated with biotinylated goat anti-(rabbit IgG) and alkaline phosphatase conjugated avidin-biotin complex. Vector(cid:3) Red Alkaline Phosphatase Substrate (Vector Laboratories, Burlingame, CA, USA) was used for 15-LO-1 visualization. Endogenous alkaline phosphatase activity was inhibited by levamisole (Vector Laboratories). Macrophages and eosin- ophils were used as internal positive controls for 15-LO-1 staining.

For immunocytochemistry, L1236 cells were resuspended in NaCl ⁄ Pi with 10% fetal bovine serum at a final concen- tration of 1 · 106 cellsÆmL)1. The cells were cytocentrifuged, fixed in 4% paraformaldehyde for 10 min and washed in NaCl ⁄ Pi. The 15-LO-1 staining was performed as described above, with the exceptions of deparaffinization, rehydration and antigen retrieval. Pre-immune serum (1 : 1000) was used as a negative control. For analysis, a microscope (BX60, Olympus, Tokyo, Japan) equipped with a digital camera (DKC-5000, Sony, Tokyo, Japan) was used.

L1236 cells were incubated at 37 (cid:2)C with exogenous arachi- donic acid (40 lm) for the indicated time. The incubations were terminated by adding one volume of methanol. Samples were centrifuged (1400 g for 6 min) and the super- natants were diluted with water to a maximum of 25% methanol and then transferred to a washed and equilibrated extraction cartridge (Oasis HLB 1 mL, 10 mg; Waters AB). The columns were washed with water and eluted with 200 lL of methanol to retrieve the metabolites. The samples were analyzed by LC-MS ⁄ MS on a Surveyor MS coupled to a TSQ Quantum Ultra triple quadrupole mass spectrometer (Thermo Finnigan AB, Stockholm, Sweden). Half a volume of water was added to samples prior to injec- tion. RP-LC was performed using a Sunfire 3.5 lm C18 column 2.1 · 30 mm (Waters AB) with the flow rate constantly held at 400 lLÆmin)1. Mobile phase A consisted of 2% acetonitrile in water and 0.1% acetic acid, and

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4232

LC-MS ⁄ MS

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

Acknowledgements

ase-1 pathway in human eosinophils and mast cells. Proc Natl Acad Sci USA 105, 680–685.

12 Kuhn H, Walther M & Kuban RJ (2002) Mammalian arachidonate 15-lipoxygenases structure, function, and biological implications. Prostaglandins Other Lipid Med- iat 68–69, 263–290.

13 Ramstedt U, Serhan CN, Lundberg U, Wigzell H &

chromatography

We wish to thank Dr Mats Hamberg for help with the chiral analysis. This work was supported by grants from the Swedish Cancer Society, Karolinska Institutet, Stockholm County Council, Orexo AB and European Commission Sixth Frame- work Programme Grant LSHM-CT-2004-005033.

References

Samuelsson B (1984) Inhibition of human natural killer cell activity by (14R,15S)-14,15-dihydroxy- 5Z,8Z,10E,12E- icosatetraenoic acid. Proc Natl Acad Sci USA 81, 6914–6918.

14 Schwenk U, Morita E, Engel R & Schroder JM (1992)

1 Brauninger A, Schmitz R, Bechtel D, Renne C, Hans- mann ML & Kuppers R (2006) Molecular biology of Hodgkin’s and Reed ⁄ Sternberg cells in Hodgkin’s lym- phoma. Int J Cancer 118, 1853–1861.

2 Poppema S & van den Berg A (2000) Interaction

Identification of 5-oxo-15-hydroxy-6,8,11,13-eicosatetra- enoic acid as a novel and potent human eosinophil che- motactic eicosanoid. J Biol Chem 267, 12482–12488. 15 Serhan CN (1994) Lipoxin biosynthesis and its impact in inflammatory and vascular events. Biochim Biophys Acta 1212, 1–25.

between host T cells and Reed-Sternberg cells in Hodg- kin lymphomas. Semin Cancer Biol 10, 345–350.

3 Re D, Kuppers R & Diehl V (2005) Molecular patho-

genesis of Hodgkin’s lymphoma. J Clin Oncol 23, 6379– 6386.

16 Maskrey BH, Bermudez-Fajardo A, Morgan AH, Stew- art-Jones E, Dioszeghy V, Taylor GW, Baker PR, Coles B, Coffey MJ, Kuhn H et al. (2007) Activated platelets and monocytes generate four hydroxyphosphatidyletha- nolamines via lipoxygenase. J Biol Chem 282, 20151– 20163.

4 Skinnider BF, Elia AJ, Gascoyne RD, Trumper LH, von Bonin F, Kapp U, Patterson B, Snow BE & Mak TW (2001) Interleukin 13 and interleukin 13 receptor are frequently expressed by Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma. Blood 97, 250–255.

17 van Leyen K, Duvoisin RM, Engelhardt H & Wied- mann M (1998) A function for lipoxygenase in pro- grammed organelle degradation. Nature 395, 392–395. 18 Turk J, Maas RL, Brash AR, Roberts LJ 2nd & Oates JA (1982) Arachidonic acid 15-lipoxygenase products from human eosinophils. J Biol Chem 257, 7068–7076.

19 Hunter JA, Finkbeiner WE, Nadel JA, Goetzl EJ &

5 Trieu Y, Wen XY, Skinnider BF, Bray MR, Li Z, Clau- dio JO, Masih-Khan E, Zhu YX, Trudel S, McCart JA et al. (2004) Soluble interleukin-13Ralpha2 decoy recep- tor inhibits Hodgkin’s lymphoma growth in vitro and in vivo. Cancer Res 64, 3271–3275.

6 Kuhn H & O’Donnell VB (2006) Inflammation and

Holtzman MJ (1985) Predominant generation of 15-lip- oxygenase metabolites of arachidonic acid by epithelial cells from human trachea. Proc Natl Acad Sci USA 82, 4633–4637.

immune regulation by 12 ⁄ 15-lipoxygenases. Prog Lipid Res 45, 334–356.

20 Gulliksson M, Brunnstrom A, Johannesson M, Back-

7 Claesson HE & Dahlen SE (1999) Asthma and leukotri- enes: antileukotrienes as novel anti-asthmatic drugs. J Intern Med 245, 205–227.

man L, Nilsson G, Harvima I, Dahlen B, Kumlin M & Claesson HE (2007) Expression of 15-lipoxygenase type-1 in human mast cells. Biochim Biophys Acta 1771, 1156–1165.

8 Brash AR, Boeglin WE & Chang MS (1997) Discovery of a second 15S-lipoxygenase in humans. Proc Natl Acad Sci USA 94, 6148–6152.

9 Shannon VR, Chanez P, Bousquet J & Holtzman MJ

21 Brinckmann R, Topp MS, Zalan I, Heydeck D, Ludwig P, Kuhn H, Berdel WE & Habenicht JR (1996) Regula- tion of 15-lipoxygenase expression in lung epithelial cells by interleukin-4. Biochem J 318, 305–312.

(1993) Histochemical evidence for induction of arachid- onate 15-lipoxygenase in airway disease. Am Rev Respir Dis 147, 1024–1028.

10 Chu HW, Balzar S, Westcott JY, Trudeau JB, Sun Y,

22 Conrad DJ, Kuhn H, Mulkins M, Highland E & Sigal E (1992) Specific inflammatory cytokines regulate the expression of human monocyte 15-lipoxygenase. Proc Natl Acad Sci USA 89, 217–221.

Conrad DJ & Wenzel SE (2002) Expression and activa- tion of 15-lipoxygenase pathway in severe asthma: rela- tionship to eosinophilic phenotype and collagen deposition. Clin Exp Allergy 32, 1558–1565.

11 Feltenmark S, Gautam N, Brunnstrom A, Griffiths W,

23 Nassar GM, Morrow JD, Roberts LJ 2nd, Lakkis FG & Badr KF (1994) Induction of 15-lipoxygenase by interleukin-13 in human blood monocytes. J Biol Chem 269, 27631–27634.

24 Liu C, Xu D, Sjoberg J, Forsell P, Bjorkholm M & Claesson HE (2004) Transcriptional regulation of

Backman L, Edenius C, Lindbom L, Bjorkholm M & Claesson HE (2008) Eoxins are proinflammatory arachi- donic acid metabolites produced via the 15-lipoxygen-

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4233

H.-E. Claesson et al.

Hodgkin lymphoma and 15-lipoxygenase

15-lipoxygenase expression by promoter methylation. Exp Cell Res 297, 61–67.

HE et al. (1996) Peripheral blood mononuclear cells of a patient with advanced Hodgkin’s lymphoma give rise to permanently growing Hodgkin-Reed Sternberg cells. Blood 87, 3418–3428.

25 Runarsson G, Liu A, Mahshid Y, Feltenmark S, Pet- tersson A, Klein E, Bjorkholm M & Claesson HE (2005) Leukotriene B4 plays a pivotal role in CD40- dependent activation of chronic B lymphocytic leukemia cells. Blood 105, 1274–1279.

33 Kanzler H, Hansmann ML, Kapp U, Wolf J, Diehl V, Rajewsky K & Kuppers R (1996) Molecular single cell analysis demonstrates the derivation of a peripheral blood-derived cell line (L1236) from the Hodg- kin ⁄ Reed-Sternberg cells of a Hodgkin’s lymphoma patient. Blood 87, 3429–3436.

26 Brinckmann R, Schnurr K, Heydeck D, Rosenbach T, Kolde G & Kuhn H (1998) Membrane translocation of 15-lipoxygenase in hematopoietic cells is calcium-depen- dent and activates the oxygenase activity of the enzyme. Blood 91, 64–74.

34 Skinnider BF, Kapp U & Mak TW (2001) Interleukin 13: a growth factor in hodgkin lymphoma. Int Arch Allergy Immunol 126, 267–276.

35 Skinnider BF, Elia AJ, Gascoyne RD, Patterson B,

27 Powell WS, Gravelle F & Gravel S (1992) Metabolism of 5(S)-hydroxy-6,8,11,14-eicosatetraenoic acid and other 5(S)-hydroxyeicosanoids by a specific dehydroge- nase in human polymorphonuclear leukocytes. J Biol Chem 267, 19233–19241.

28 Ingram CD & Brash AR (1988) Characterization of

Trumper L, Kapp U & Mak TW (2002) Signal trans- ducer and activator of transcription 6 is frequently acti- vated in Hodgkin and Reed-Sternberg cells of Hodgkin lymphoma. Blood 99, 618–626.

HETEs and related conjugated dienes by UV spectros- copy. Lipids 23, 340–344.

29 Andersson E, Schain F, Svedling M, Claesson HE &

36 Axdorph U, Porwit-MacDonald A, Sjoberg J, Grimfors G, Ekman M, Wang W, Biberfeld P & Bjorkholm M (1999) Epstein-Barr virus expression in Hodgkin’s dis- ease in relation to patient characteristics, serum factors and blood lymphocyte function. Br J Cancer 81, 1182– 1187.

Forsell PK (2006) Interaction of human 15-lipoxygen- ase-1 with phosphatidylinositol bisphosphates results in increased enzyme activity. Biochim Biophys Acta 1761, 1498–1505.

30 Kuhn H, Heydeck D, Brinckman R & Trebus F (1999)

37 Oertel J & Huhn D (2000) Immunocytochemical meth- ods in haematology and oncology. J Cancer Res Clin Oncol 126, 425–440.

Regulation of cellular 15-lipoxygenase activity on pre- translational, translational, and posttranslational levels. Lipids 34(Suppl.), S273–S279.

31 Walther M, Wiesner R & Kuhn H (2004) Investigations into calcium-dependent membrane association of 15-lip- oxygenase-1. Mechanistic roles of surface-exposed hydrophobic amino acids and calcium. J Biol Chem 279, 3717–3725.

38 Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller- Hermelink HK, Vardiman J, Lister TA & Bloomfield CD (2000) The World Health Organization classifica- tion of neoplastic diseases of the haematopoietic and lymphoid tissues: Report of the Clinical Advisory Com- mittee Meeting, Airlie House, Virginia, November 1997. Histopathology 36, 69–86.

32 Wolf J, Kapp U, Bohlen H, Kornacker M, Schoch C, Stahl B, Mucke S, von Kalle C, Fonatsch C, Schaefer

FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS

4234