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 Claesson
1,2
, William J. Griffiths
3
,A
˚
sa Brunnstro
¨
m
2
, Frida Schain
4
, Erik Andersson
2,4
,
Stina Feltenmark
1,2
,He
´
le
`
ne A. Johnson
1
, Anna Porwit
5
, Jan Sjo
¨

(Received 14 April 2008, revised 18 June
2008, accepted 23 June 2008)
doi:10.1111/j.1742-4658.2008.06570.x
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 C
4
, eoxin D
4
and eoxin E
4
. 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
activity in L1236 cells demonstrates that this cell line comprises a

turally related enzymes, catalyzing the oxygenation of
arachidonic acid [6]. 5-Lipoxygenase (LO) catalyses the
conversion of arachidonic acid to leukotriene (LT)A
4
which can be further converted to LTB
4
or LTC
4
. The
latter metabolite can be further metabolized to LTD
4
and LTE
4
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].
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 EXA
4
, which in turn can be conjugated
with glutathione, leading to the formation of EXC
4
[11]. This metabolite can be further metabolized to

cells, including the maturation of keratinocytes and the
eye lens [17]. 15-Lipoxygenase is predominantly
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 large majority of H-RS cells are derived from
germinal centre B lymphocytes [1]. Because human B
lymphocytes express 5-LO [25], it was of interest to
determine whether HL cell lines also expressed this
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-
matographed with synthetic 15-HETE (Fig. 1). In
addition, a minor peak (UV absorbance maximum at

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
cells were incubated with NaCl ⁄ P
i
with Ca
2+
⁄ Mg
2+
;
NaCl ⁄ P
i
with Ca
2+
⁄ Mg
2+
plus calcium ionophore
A23187 (final concentration 5 lm); or calcium-free
NaCl ⁄ P
i
. Subsequently, the cells were sonicated
followed by subcellular fractionation. The separate
fractions were analyzed by SDS ⁄ PAGE followed by
western blotting (Fig. 2, lower panel). In the presence
of Ca
2+
⁄ Mg
2+

Ca
2+
⁄ Mg
2+
(0.9 and 0.5 mM, respectively); (B) NaCl ⁄ P
i
with
Ca
2+
⁄ Mg
2+
plus calcium ionophore A23187 (final concentration
5 l
M); and (C) NaCl ⁄ P
i
without Ca
2+
⁄ Mg
2+
. After 10 min of incuba-
tion at 37 °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 °C and the supernatants were transferred to new
tubes for ultracentrifugation (100 000 g at 4 °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

medium-sized mononuclear cells, large blasts and
multinucleated giant cells. Immunocytochemical
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.
Formation of eoxins by L1236 cells
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
acid is shown in Fig. 4. Four peaks (1–4) were
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 EXA
4
;
also named 14,15-LTA
4
) [12,19]. The pattern formed
by these metabolites was almost identical to that
reported for human airway epithelial cells and eosin-
ophils incubated with arachidonic acid [11,19]. The

(peaks I and III). Metabolite III was analyzed by
positive ion ESI-MS and MS ⁄ MS and the spectrum of
A
B
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) raised against recombinant human
15-LO-1 (red colour). (B) Pre-immune serum (diluted 1 : 1000) was
used as a negative control. Original magnification ·46.
H E. Claesson et al. Hodgkin lymphoma and 15-lipoxygenase
FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS 4225
this metabolite was identical to the spectrum for EXD
4
(data not shown). The exact structure of this metabo-
lite has not been established but the MS ⁄ MS spectrum
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 EXD
4
[28]. The material in peak III is there-
fore probably 8-trans-EXD
4
. 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

30 min. Figure 6 shows that EXA
4
was converted to
EXC
4
(peak II) and EXD
4
(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-
bated with arachidonic acid because the material in
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 EXA
4
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
Fig. 4. RP-HPLC chromatogram of the products formed by L1236
cells after incubation with arachidonic acid (40 l
M) 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 the material in peaks II and IV which also coeluted with
synthetic standards of EXC
4

4226 FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS
were incubated with synthetic EXD
4
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 (EXE
4
). Metabolite V
also had an identical retention time as synthetic EXE
4
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 EXC
4
, EXD
4
and EXE
4
, respectively (data not
shown). Taken together, these results show that L1236
cells convert arachidonic acid to EXC

4
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 EXE
4
in this experiment
because 8(S),15(S)-DHETE co-chromatographed with
EXE
4
in this HPLC system. The dose–response curves
of the formation of eoxins from arachidonic acid show
that significant amounts of EXC
4
and EXD
4
were
formed already at a concentration of 1 lm arachidonic
acid (Fig. 8B). The levels of these metabolites
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
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

H E. Claesson et al. Hodgkin lymphoma and 15-lipoxygenase
FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS 4227
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
lymphoma. Thus, H-RS cells express 15-LO-1 in vivo
and it is therefore very likely that these cells can
produce eoxins and other 15-LO-1 derived metabolites
in vivo, contributing to the inflammatory features of
HL.
Discussion
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
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. ),noH⁄ 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; MALT, mucosa-associated lymphoid tissue lymphoma; MCL, mantle cell lymphoma; B-CLL, B-cell

16 Male 36 IIIB HL MC Pos Few Low )
17 Female 5 IIA HL MC Pos Many High +
18 Female 77 IA HL MC ND Medium Low +
19 Female 70 IA HL MC Neg Medium Medium +
20 Male 37 IIIB HL MC Neg High High ++
NHL
21 Female 39 IIA PTCL ND NA ND )
22 Male 80 IIA DLBCL ND NA ND )
23 Female 60 IVB DLBCL ND NA ND )
24 Female 54 IA MALT ND NA ND )
25 Female 75 IVB Immunocytoma ND NA ND )
26 Male 81 IVA MCL ND NA ND )
27 Male 63 Rai IV B-CLL ND NA ND )
28 Male 32 IIA BL ND NA ND )
29 Female 72 IVA FCDL ND NA ND )
30 Male 28 IVB PMBCL ND NA ND )
Hodgkin lymphoma and 15-lipoxygenase H E. Claesson et al.
4228 FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS
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 EXC
4
, EXD
4
and EXE
4

) 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 (EXC
4
), 14(R)-cysteinyl-glycyl-
15(S)-hydroxy-5,8,10,12-(Z,Z,E,E)-eicosatetraenoic acid
(EXD
4
) and 14(R)-cysteinyl-15(S)-hydroxy-5,8,10,12-
(Z,Z,E,E)-eicosatetraenoic acid (EXE
4
), respectively.
Taken together, the UV profile and mass spectra show
that L1236 cells can metabolize arachidonic acid to
EXA
4
which in turn can be readily converted to
EXC
4
, EXD
4
and EXE
4
(Fig. 10). Eoxin C
4
was more
rapidly converted to EXD
4

respectively.
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.
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 formation of the giant H-RS cells in HL is still
a mystery although many attempts have been made to
clarify the formation of these cells. The enzyme
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 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.
Experimental procedures
Materials
Synthetic 14,15-epoxy-5,8,10,12-(Z,Z,E,E)-eicosa trae noic aci d
(EXA
4

IN, USA).
Cell lines, biopsies and patients
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 °C in an atmosphere
containing 5% CO
2
. 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 degree of eosinophilia and the number of H-RS cells

15 s, 58 °C for 1 min, 72 °C for 2 min and 72 °C for
6 min. b
2
-microglobulin was run for 24 cycles and 15-LO-1
and 15-LO-2 for 34 cycles.
Translocation assay and western blot
L1236 cells were washed twice in NaCl ⁄ P
i
without
Ca
2+
⁄ Mg
2+
and resuspended in 1 mL of NaCl ⁄ P
i
without
Ca
2+
⁄ Mg
2+
. 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 resuspending the cells in respective buffers
[buffer A: NaCl ⁄ P
i
with Ca
2+
⁄ Mg
2+

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 ⁄ P
i
-Tween at room temperature for 1 h.
The membrane was then incubated overnight at 4 °C with
a 15-LO-1 rabbit peptide antiserum (batch 632) raised
against the 15-LO-1 peptide CALDKEIEIRNAKLD
MPYEY (dilution 1 : 5000) (antibodies made by Innovagen
AB, Ideon, Lund, Sweden) in NaCl ⁄ P
i
-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 ⁄ P
i
-Tween, the membrane was incu-
bated with the anti-rabbit horseradish peroxidase serum,
diluted 1 : 10000 in NaCl ⁄ P
i
-Tween + 1% milk powder for
5 h at room temperature. The membrane was then
H E. Claesson et al. Hodgkin lymphoma and 15-lipoxygenase
FEBS Journal 275 (2008) 4222–4234 ª 2008 The Authors Journal compilation ª 2008 FEBS 4231
washed three times in NaCl ⁄ P
i
-Tween before detection with
ECL-plus. Recombinant human 15-LO-1 was expressed in

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 NH
3
) 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.
LC-MS ⁄ MS
L1236 cells were incubated at 37 °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

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.
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
incubated with 15-LO-1 antibody (1 : 1000), for 1 h at
room temperature. Subsequently, the biopsies were incu-
bated with biotinylated goat anti-(rabbit IgG) and alkaline
phosphatase conjugated avidin-biotin complex. VectorÒ
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

phoma. Int J Cancer 118, 1853–1861.
2 Poppema S & van den Berg A (2000) Interaction
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.
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.
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
immune regulation by 12 ⁄ 15-lipoxygenases. Prog Lipid
Res 45, 334–356.
7 Claesson HE & Dahlen SE (1999) Asthma and leukotri-
enes: antileukotrienes as novel anti-asthmatic drugs.
J Intern Med 245, 205–227.
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
(1993) Histochemical evidence for induction of arachid-
onate 15-lipoxygenase in airway disease. Am Rev Respir

B, Coffey MJ, Kuhn H et al. (2007) Activated platelets
and monocytes generate four hydroxyphosphatidyletha-
nolamines via lipoxygenase. J Biol Chem 282, 20151–
20163.
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 &
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.
20 Gulliksson M, Brunnstrom A, Johannesson M, Back-
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.
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.
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.
23 Nassar GM, Morrow JD, Roberts LJ 2nd, Lakkis FG
& Badr KF (1994) Induction of 15-lipoxygenase by

1498–1505.
30 Kuhn H, Heydeck D, Brinckman R & Trebus F (1999)
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.
32 Wolf J, Kapp U, Bohlen H, Kornacker M, Schoch C,
Stahl B, Mucke S, von Kalle C, Fonatsch C, Schaefer
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.
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.
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,
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.