12/15-Lipoxygenase Regulates the
Inflammatory Response to Bacterial Products
In Vivo
This information is current as
of July 31, 2017.
Vincent Dioszeghy, Marcela Rosas, Benjamin H. Maskrey,
Chantal Colmont, Nicholas Topley, Pavlos Chaitidis,
Hartmut Kühn, Simon A. Jones, Philip R. Taylor and Valerie
B. O'Donnell
References
Subscription
Permissions
Email Alerts
This article cites 33 articles, 16 of which you can access for free at:
http://www.jimmunol.org/content/181/9/6514.full#ref-list-1
Information about subscribing to The Journal of Immunology is online at:
http://jimmunol.org/subscription
Submit copyright permission requests at:
http://www.aai.org/About/Publications/JI/copyright.html
Receive free email-alerts when new articles cite this article. Sign up at:
http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by
The American Association of Immunologists, Inc.,
1451 Rockville Pike, Suite 650, Rockville, MD 20852
Copyright © 2008 by The American Association of
Immunologists All rights reserved.
Print ISSN: 0022-1767 Online ISSN: 1550-6606.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
J Immunol 2008; 181:6514-6524; ;
doi: 10.4049/jimmunol.181.9.6514
http://www.jimmunol.org/content/181/9/6514
The Journal of Immunology
12/15-Lipoxygenase Regulates the Inflammatory Response to
Bacterial Products In Vivo1
Vincent Dioszeghy,* Marcela Rosas,* Benjamin H. Maskrey,* Chantal Colmont,*
Nicholas Topley,* Pavlos Chaitidis,† Hartmut Kühn,† Simon A. Jones,* Philip R. Taylor,*
and Valerie B. O’Donnell2*
T
he 12/15-Lipoxygenase (LOX)3 belongs to a family of
lipid-peroxidizing enzymes that catalyze the oxygenation of polyunsaturated fatty acids to their corresponding hydroperoxy derivatives (1). In mice, the enzyme generates
12(S)-hydroperoxyeicosatetraenoic acid (HpETE) from arachidonate, which is subsequently converted into secondary
products including 12-hydroxyeicosatetraenoic acid (12-HETE),
while in humans 15-HpETE is the predominant isomer. In mice,
the predominant population expressing 12/15-LOX is resident
peritoneal macrophage (M␾), with mRNA levels ⬃1000 times
higher than any other tissue (2). Resident M␾ play a key role in
peritoneal inflammation so the physiological role of 12/15-LOX in
mice could involve the regulation of peritoneal M␾ functions and
*Department of Medical Biochemistry and Immunology, School of Medicine, Cardiff
University, Cardiff, United Kingdom; and †Institute of Biochemistry, Humboldt University, Berlin, Germany
Received for publication May 21, 2008. Accepted for publication August 18, 2008.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1
Funding from the Wellcome Trust was provided to V.B.O., V.D., B.M., C.C., S.A.J.,
and N.T. Financial support to H.K. was provided by Deutsche Forschungsgemeinschaft (Ku 961/8-2) and by the European Commission (FP6, LSHM-CT-20040050333). P.R.T. is an Medical Research Council Senior Fellow (G0601617).
inflammatory responses in that tissue (3). In support, deficiency of
12/15-LOX in thioglycolate-elicited peritoneal M␾ alters their in
vitro phenotype, resulting in decreased IL-4 induction of the scavenger receptor CD36, decreased stimulation of IL-12 synthesis by
LPS, and attenuated phagocytic removal of apoptotic cells (4 – 6).
High levels of exogenous 12/15-LOX products (e.g., lipoxins) and
plant 15-LOX administered to the peritoneal cavity during inflammation appear to suppress neutrophil influx (7, 8); however, the
activity and function of the endogenous murine enzyme and its
products in this organ during inflammation have not been
examined.
In this study, we characterized the phenotype of 12/15-LOX
expressing cells in the mouse peritoneal cavity, both basally and
during acute sterile inflammation. We also sought to determine
whether 12/15-LOX expressed by the cells was required for their
in vivo function in this tissue.
Materials and Methods
Animals
All animal experiments were performed in accordance to the United Kingdom Home Office Animals (Scientific Procedures) Act of 1986. 12/15LOX knockout mice generated as described previously and wild-type (WT)
male C57BL/6 mice (25–30 g) from Charles River Laboratories were kept
in constant temperature cages (20 –22°C) and given free access to water
and standard chow (9).
2
Address correspondence and reprint requests to Dr. Valerie B. O’Donnell, Department
of Medical Biochemistry and Immunology, School of Medicine, Cardiff University,
Cardiff CF14 4XN, United Kingdom. E-mail address: [email protected]
HETE quantitation using liquid chromatography/mass
spectrometry/mass spectrometry
3
Abbreviations used in this paper: LOX, lipoxygenase; DC, dendritic cell; HETE,
hydroxyeicosatetraenoic acid; LXA4, lipoxin A4; MMR, mannose receptor; Mo,
monocyte; M␾, macrophage; SES, Staphylococcus epidermidis cell-free supernatant;
WT, wild type.
Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00
www.jimmunol.org
12-HETE-d8 (10 ng) was added to each lavage before extraction, as internal standard. Hydroperoxides were then reduced to their corresponding
stable alcohols using 1 mM SnCl2. Lipids were extracted by adding a
solvent mixture (1 M acetic acid/2-isopropanol/hexane (2/20/30, v/v/v)) to
the sample at a ratio of 2.5 ml of solvent mixture to 1 ml of sample,
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
The peritoneal macrophage (M␾) is the site of greatest 12/15-lipoxygenase (12/15-LOX) expression in the mouse; however,
its immunoregulatory role in this tissue has not been explored. Herein, we show that 12/15-LOX is expressed by 95% of
resident peritoneal CD11bhigh cells, with the remaining 5% being 12/15-LOXⴚ. 12/15-LOXⴙ cells are phenotypically defined
by high F4/80, SR-A, and Siglec1 expression, and enhanced IL-10 and G-CSF generation. In contrast, 12/15-LOXⴚ cells are
a dendritic cell population. Resident peritoneal M␾ numbers were significantly increased in 12/15-LOXⴚ/ⴚ mice, suggesting
alterations in migratory trafficking or cell differentiation in vivo. In vitro, M␾ from 12/15-LOXⴚ/ⴚ mice exhibit multiple
abnormalities in the regulation of cytokine/growth factor production both basally and after stimulation with Staphylococcus
epidermidis cell-free supernatant. Resident adherent cells from 12/15-LOXⴚ/ⴚ mice generate more IL-1, IL-3, GM-CSF, and
IL-17, but less CCL5/RANTES than do cells from wild-type mice, while Staphylococcus epidermidis cell-free supernatantelicited 12/15-LOXⴚ/ⴚ adherent cells release less IL-12p40, IL-12p70, and RANTES, but more GM-CSF. This indicates a
selective effect of 12/15-LOX on peritoneal cell cytokine production. In acute sterile peritonitis, 12/15-LOXⴙ cells and LOX
products were cleared, then reappeared during the resolution phase. The peritoneal lavage of 12/15-LOXⴚ/ⴚ mice showed
elevated TGF-␤1, along with increased immigration of monocytes/M␾, but decreases in several cytokines including RANTES/CCL5, MCP-1/CCL2, G-CSF, IL-12-p40, IL-17, and TNF-␣. No changes in neutrophil or lymphocyte numbers were
seen. In summary, endogenous 12/15-LOX defines the resident M⌽ population and regulates both the recruitment of monocytes/M␾ and cytokine response to bacterial products in vivo. The Journal of Immunology, 2008, 181: 6514 – 6524.
The Journal of Immunology
6515
vortexing, and then adding 2.5 ml hexane. Following vortex and centrifugation, lipids were recovered in the upper hexane layer. The samples
were then reextracted by addition of an equal volume of hexane, followed
by vortex and centrifugation. The combined hexane layers were dried under N2 flow, resuspended in 100 ␮l methanol, and stored at ⫺80°C until
analysis. Samples were separated on a C18 ODS2, 5-␮m, 150 ⫻ 4.6-mm
column (Waters) using a gradient of 50% to 90% B over 20 min (A,
water/acetonitrile/acetic acid, 75/25/0.1, B, methanol/acetonitrile/acetic
acid, 60/40/0.1) at 1 ml/min. Products were quantitated by monitoring specific multiple reaction monitoring transitions (negative mode, [M ⫺
H]⫺) on an Applied Biosystems 4000 Q-Trap, using m/z 319 3 219
(15-HETE), 319 3 179 (12-HETE), 319 3 115 (5-HETE), 319 3 155
(8-HETE), 319 3 167 (11-HETE), 351 3 115 (lipoxin A4, LXA4), and
327 3 184 (12-HETE-d8) with collision energies of ⫺20 to ⫺28 V.
Standard curves with purified HETEs and LXA4 vs 12-HETE-d8 were
constructed for metabolite quantitation, with sensitivities down to 0.1–1
pg on-column, allowing for detection of down to ⬃1–10 pg per lavage
of each eicosanoid.
Staphylococcus epidermidis cell-free supernatant (SES)-induced
peritoneal inflammation
Peritoneal inflammation was established in mice through i.p. administration of a defined 500 ␮l dose of SES, prepared from a clinical isolate of S.
epidermidis (10). At defined intervals following SES administration, animals were sacrificed and the peritoneal cavity was lavaged with 2 ml of
ice-cold PBS. Composition of the leukocyte infiltrate was assessed using a
Coulter Z2 counter (Beckman Coulter), differential cell staining of cytospins with Accustain (Sigma-Aldrich), and flow cytometric analysis (see
below). Lavage fluids were rendered cell-free by centrifugation for analysis
of inflammatory mediators. In some experiments, 10 ng 12-HETE was
administered at the same time as SES.
Immunochemistry
Cytospin preparations of peritoneal cells were methanol-fixed on glass
slides, permeabilized using 0.1% (w/v) Triton X-100/PBS, and blocked
using 1% (w/v) BSA/PBS. 12/15-LOX, F4/80, or mannose receptor
(MMR) expression were detected using guinea pig anti-12/15-LOX (generated in our laboratory), in combination with either rat anti-F4/80 or rat
anti-MMR (both Serotec), with goat anti-rat IgG-Alexa 568 and goat antiguinea pig IgG-Alexa 488 as secondaries (Molecular Probes). Negative
controls used equivalent concentrations of isotype-matched rat or guinea
pig IgG Ab. Nuclei were stained using DRAQ5 (Biostatus). Imaging was
performed on an Axiovert 100 inverted microscope connected to a Bio-Rad
MRC 1024ES laser scanning system (Bio-Rad Microscience) using standard analysis software (Lasersharp 2000, Bio-Rad Microscience). Images
were acquired using a ⫻40 oil lens, with excitation at 488 nm and emission
522/35 nm, and excitation at 568 nm and emission 595/35 nm, at room
temperature. For each slide, three separate regions were imaged and cells
counted to calculate percentages.
Flow cytometric analysis
Leukocytes were incubated with mouse Fc block (BD Pharmingen) before immunolabeling for 30 min at 4°C. Cells were incubated 1 h with
primary fluorochrome-conjugated or nonconjugated Abs, as follows:
anti-7/4 Ag (7/4, Serotec), anti-B220 (RA3-6B2, BD Pharmingen), antidectin-1 receptor (2A11, Serotec), anti-CD3 (17A2, BD Pharmingen),
anti-CD4 (GK1.5, BD Pharmingen), anti-CD8a (53-6.7, BD Pharmingen), anti-CD11b (M1/70, BD Pharmingen), anti-CD11c (HL3, BD
Pharmingen), anti-CD14 (Sa2-8, e-Bioscience), anti-CD40 (3/23, BD
Pharmingen), anti-CD54 (3E2, BD Pharmingen), anti-CD62L (MEL14, BD Pharmingen), anti-CD80 (16-10A1, BD Pharmingen), antiCD83 (Michel-17, e-Bioscience), anti-CD86 (PO3, BD Pharmingen),
anti-CD115 (604B5 2E11, Serotec), anti-CD138 (281-2, BD Pharmingen), anti-CD206 (MR5D3, Serotec), anti-CCR2 (E68, Abcam), antiCCR5 (C34-3448, BD Pharmingen), anti-CX3CR1 (Torrey Pines Biolabs), anti-CXCR4 (2B11/CXCR4, BD Pharmingen), anti-cectin-2
(D2.11E4) (11), anti-F4/80 Ag (CI:A3-1, Serotec), anti-Gr-1 (RB68C5, BD Pharmingen), anti-I-A/I-E (2G9, BD Pharmingen), antiMARCO (macrophage receptor with collagenous structure) (ED31, Serotec), anti-TLR2 (6C2, e-Bioscience), streptavidin-RPE (Serotec),
streptavidin-RPE-Cy5 (Serotec), and RPE anti-rabbit IgG (Serotec).
GM-CSF receptor was stained using a chimeric protein fusing the
mouse GM-CSF to a mutated Fc region of human IgG1 (12). Where
necessary, cells were incubated for a further 30 min at 4°C with appropriate fluorochrome-conjugated secondary Abs or streptavidin. Cells
were washed and fixed with CellFix (BD Biosciences) and then analyzed by using a FACSCalibur flow cytometer and CellQuest software
(BD Biosciences). Data were acquired from 50,000 events, and staining
was compared with fluorochrome-conjugated isotype control Abs.
For cell cycle analysis, in brief, cells were fixed in 70% ethanol (2 h
at ⫺20°C) and then resuspended in 45 mM Na2HPO4, 2.5 mM citric
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 1. Phenotypic characterization of 12/15-LOX⫹ cells and 12/
15-LOX⫺ cells in a WT mouse. A,
Left panels, 12/15-LOX (green) colocalizes with CD11b⫹ cells (red), but a
small proportion (5%) are 12/15LOX⫺. Resident peritoneal lavage
cells were cytospun onto glass slides
and immunostained for 12/15-LOX or
CD11b as described in Materials and
Methods. Right panel, FACS analysis
shows two populations of CD11b⫹
cells, characterized as F4/80high (R1)
or F4/80med (R2). Resident peritoneal
lavage was analyzed for expression of
CD11b or F4/80 as described in Materials and Methods. B, R1 and R2
cells are defined by 12/15-LOX or
mannose receptor expression, respectively. Total lavage (left), FACSsorted R1 cells (middle), and R2 cells
(right) were immunostained for 12/
15-LOX (green) or mannose receptor
(red), as described in Materials and
Methods. C, 12/15-LOX is not expressed by T or B cells. Total resident
lavage cells were immunostained for
CD4, CD8, or CD45R (red) and 12/
15-LOX (green) as described in Material and Methods.
6516
12/15-LIPOXYGENASE IN INFLAMMATION
acid, and 0.1% Triton X-100 (30 min at 37°C). Cells were stained by
addition of 3.75 vol of 10 mM PIPES (pH 6.8), 100 mM NaCl, 2 mM
MgCl2, 0.1% Triton X-100, 40 ␮g/ml RNase A, and 20 ␮g/ml propidium iodide (30 min at room temperature in the dark) and analyzed by
flow cytometry on a FACSCalibur flow cytometer and analyzed with
CellQuest software.
FIGURE 3. Comparison of cytokine sythesis by R1 and R2 cells in
response to SES stimulation. FACS-sorted cells (200,000) were cultured
with and without SES for 24 h and culture fluids were assayed for cytokine
generation by ELISA or Bio-Plex assay (n ⫽ 4, mean ⫾ SEM; ⴱ, p ⬍ 0.05
for effect of cell type, two-way ANOVA; see Table I for full details and
explanation of statistical analysis).
Cell population determination
Peritoneal cell populations were identified as follows: M␾/monocytes (Mo)
are CD11b⫹/F4/80⫹, dendritic cells (DC) are CD11bhigh/F4/80med/
CD11c⫹, lymphocytes are CD11b⫺/low/F4/80⫺ and show characteristic
forward light scatter-side light scaller pattern, and neutrophils are
CD11bhigh/F4/80⫺ cells. Blood Mo were identified as CD11b⫹/F4/80⫹
cells in whole-blood samples.
Ag presentation assays
Cultures were established in 96-well U-bottom tissue culture plates
(Costar). CFSE-labeled CD4⫹ T cells (30,000 cells per well) from spleens
of DO11.10 mice (TCR with specificity for an OVA peptide) were cultured
with increasing concentrations of APCs (0, 2,500, 5,000, 10,000, and
20,000 cells) in RPMI 1640 supplemented with 10% low-endotoxin FBS
and 50 ␮M ␤-mercaptoethanol. OVA (Sigma-Aldrich) was used at final
concentration of 0.5 mg/ml per well. Bone marrow-derived DC and the
synthetic OVA peptide (323–339) were used as controls. After 72 h of
incubation at 37oC, supernatants were collected and kept at ⫺20°C for
subsequent assay of IL-2 production by ELISA. The cell pellets were
washed with PBS, resuspended in blocking buffer (PBS supplemented with
5% heat-inactivated rabbit serum, 0.5% BSA, 5 mM EDTA, and 2 mM
NaN3), and incubated for 1 h at 4°C. Cells were incubated with 10 ␮g/ml
of anti-CD3-biotin (clone KT3, a gift from Dr. S. Cobbold) for 1 h at 4°C
and washed twice in washing buffer (PBS supplemented with 0.5% BSA,
5 mM EDTA, and 2 mM NaN3). Subsequently, streptavidin-allophycocyanin (BD Pharmingen) was added (1/100) and incubated for 1 h at 4°C.
After three washes, cells were resuspended in 1% formaldehyde (in PBS)
and analyzed on a FACSCalibur (Becton Dickinson). FACS analysis was
performed using FlowJo software (Tree Star). CD3⫹ T cells were gated and
the CFSE fluorescence as a measure of T cell proliferation was determined.
T cell proliferation data were expressed as a division index (average number of divisions that dividing cells underwent) as determined using FlowJo
software.
Cell isolation and culture
Cell subpopulations were isolated from peritoneal lavages by MoFlo
sorting. In brief, peritoneal cells from 15 to 20 mice were pooled and
then labeled with F4/80-FITC and CD11b-allophycocyanin Abs. The
cells were then washed three times with PBS, 0.5% BSA before flowsorting based on their level of expression of CD11b and F4/80. Sorted
cells were then cultured in RPMI 1640 medium supplemented with 10%
FCS, 2 mM L-glutamine, 10 U/ml penicillin, and 10 ␮g/ml streptomycin
for 3 h. Nonadherent cells were removed and the remaining cells were
cultured in fresh medium with or without SES (1/50). After 24 h of
culture, cell culture supernatants were collected and stored at ⫺20°C
for future cytokine determination. Sorted cells were also used as APC
for Ag presentation assays.
Cytokine determination
IL-1␣, IL-1␤, TNF-␣, IL-10, G-CSF, GM-CSF, IL-13, MIP-1␣/CCL3,
IL-12p40, and IL-12p70 concentrations in culture supernatant fluids and
in peritoneal fluids were measured by using a commercial Bio-Plex
mouse cytokine assay kit (Bio-Rad). IL-6 and MCP-1/CCL2 concentrations in culture supernatant fluids and in peritoneal fluids were measured by using BD OptEIA mouse-specific ELISA set according to the
manufacturer’s protocol (BD Biosciences). RANTES/CCL5 and KC/
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 2. Comparison of surface expression of Ags between R1 and
R2 cells. FACS-sorted R1 or R2 cells were analyzed for surface expression
of Ags as described in Materials and Methods.
The Journal of Immunology
6517
Table I. Cytokine production of resident peritoneal M␾ populationsa
Mean (⫾SEM), pg/ml
Two-way ANOVA ( p-value)
Cytokine/Growth
Factor
R1
R1 ⫹ SES
R2
R2 ⫹ SES
n
Interaction
SES
Cell type
IL-1␣
IL-1␤
IL-6
IL-10
IL-12p40
IL-12p70
IL-13
RANTES
MCP-1
MIP-1␣
KC
TNF
G-CSF
GM-CSF
52.58 (21.78)
23.11 (6.39)
5,093 (2,248)
57.51 (15.58)
14.96 (6.31)
15.05 (7.09)
92.25 (77.02)
1,412 (449)
3,781 (2,884)
1,598 (456)
4,568 (2,095)
86.14 (68.95)
707 (330)
36.63 (21.47)
391.20 (215.26)
62.35 (21.50)
10,356 (2,640)
139.60 (49.95)
38.69 (17.03)
48.30 (23.26)
194.39 (139.26)
2,289 (839)
6,883 (3,747)
7,122 (4,037)
11,117 (2,933)
347.14 (159.82)
7,835 (2,569)
174.67 (98.77)
7.13 (5.00)
11.67 (5.86)
759 (169)
6.17 (2.11)
31.27 (18.13)
3.19 (1.31)
13.85 (12.27)
595 (204)
366 (342)
3,942 (3,504)
1,902 (1,557)
47.61 (47.61)
46 (27)
16.10 (10.50)
39.96 (22.04)
22.57 (10.26)
4,104 (1,378)
12.43 (4.49)
96.08 (40.28)
9.77 (3.43)
56.78 (48.45)
1,756 (818)
630 (571)
9,243 (8,430)
6,170 (2,201)
97.94 (72.55)
337 (217)
41.98 (21.13)
4
4
4
4
4
4
4
4
3
4
4
4
4
4
NS
NS
NS
0.0806
NS
NS
NS
NS
0.0512
NS
NS
NS
0.0249
NS
NS
0.0193
0.0137
0.0498
0.0306
0.0507
0.0914
0.0591
0.0309
NS
0.0111
0.0367
0.0180
0.0824
NS
NS
0.0633
0.0333
NS
NS
NS
NS
NS
NS
NS
NS
0.0293
NS
CXCL1 concentrations in culture supernatant fluids and in peritoneal
fluids were measured by using mouse specific DuoSet ELISA kit according to the manufacturer’s protocol (R&D Systems). For TGF-␤1
concentrations measurement, latent TGF-␤1 was activated to immunoreactive TGF-␤1 by adding 0.1 ml of 1 N HCl to 0.5 ml of sample for 10 min.
Samples were then neutralized with 0.1 ml of 1.2 N NaOH/0.5 M HEPES.
Total TGF-␤1 was then measured using a mouse-specific DuoSet ELISA kit.
lymphocytes that comprise most of the remaining lavage cells
express 12/15-LOX, immunostaining was performed for CD4,
CD8, or CD45R, along with 12/15-LOX. As shown, 12/15-LOX
was not associated with any of these lymphocyte Ags, indicating that it is not expressed by these cells (Fig. 1C).
Real-time PCR analysis of gene expression
Real-time PCR was conducted on a DNA Engine Opticon 2 (MJ Research/Biozym) using the QuantiTect SYBR Green PCR kit from Qiagen according to the recommendations of the vendor, and the following
PCR protocol was applied: 15-min hot start at 95°C, followed by variable numbers of amplification cycles consisting of denaturation (30 s at
94°C), annealing (30 s at 60°C), and synthesis (30 s at 72°C) phases in
a total volume of 20 ␮l. Amplicons were used as external standards for
each gene product including GAPDH. For melting curve analyses, the
temperature was elevated slowly from 60°C to 95°C. Data were acquired and analyzed with the Opticon Monitor software (version 2). The
amplification kinetics were recorded as sigmoid progress curves, for
which the fluorescence was plotted against the number of amplification
cycles. Homogeneity of the amplified PCR products was tested by melting-curve analysis and the following primer pairs were used: GAPDH
(forward), 5⬘-CCA TCA CCA TCT TCC AGG AGC GA-3⬘, GAPDH
(reverse), 5⬘-GGA TGA CCT TGC CCA CAG CCT TG-3⬘; COX1 (forward), 5⬘-CTG CGG CTC TTT AAG GAT GGG A-3⬘, COX1 (reverse),
5⬘-GCG AGA GAA GGC ATC CAC CAG-3⬘; COX2 (forward), 5⬘GTC CCT GAG CAT CTA CGG TTT G-3⬘, COX2 (reverse), 5⬘-CTC
TGG TCA ATG GAA GCC TGT G-3⬘.
Statistical analysis
Data were analyzed using Student’s t test or two-way ANOVA, as indicated in the text. p values of ⬍0.05 were considered significant.
Results
12/15-LOX expression in peritoneal cells
Basally, ⬃95% of resident peritoneal M␾ of WT C57BL/6
mice, identified by F4/80 and CD11b expression, express 12/
15-LOX while 5% are negative (Fig. 1A, left and middle panels). Phenotypic characterization by FACS showed two distinct
populations of CD11bhigh cells: one characterized by high levels of F4/80 (R1) and a second that had only medium levels of
F4/80 (R2) (Fig. 1A, right panel). FACS sorting of the two M␾
populations followed by immunostaining defined the R1 population as the 95% of cells expressing 12/15-LOX, whereas R2
cells are largely negative (Fig. 1B). In contrast, the 12/15LOX⫺ cells were positive for MMR (CD206), an Ag that is
absent in 12/15-LOX⫹ cells (Fig. 1B). To determine whether
FIGURE 4. R2 cells present Ag and stimulate naive lymphocytes to
proliferate and generate IL-2. A, Stimulation of T cell division by R1 or
R2 cells. FACS-sorted R1 and R2 cells were used in Ag presentation
assays with CFSE-labeled BALB/c.DO.11.10 transgenic OVA-specific
CD4⫹ T cells in the presence or absence of OVA. Histograms show the
proliferation-induced reduction of CFSE labeling of the CD4⫹ T cells
(gated on CD3) caused by the R2 cells, but not the R1 cells, in the
presence of OVA. Histograms represent experiments with 10,000 APC
and 30,000 CD4⫹ T cells in the presence or absence of 0.5 mg/ml OVA.
B and C, Stimulation of T cell proliferation or IL-2 synthesis by R1 or
R2 cells. Shown are summaries of T cell proliferation (B) (division
index is the average number of divisions that the dividing cells underwent) and IL-2 generation (C) in the presence of OVA and the indicated
numbers of APC (mean ⫾ SEM of two independent experiments, each
performed in triplicate).
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
a
12/15-LOX⫹ (R1) and 12/15-LOX⫺ (R2) peritoneal populations were MoFlo sorted and cultured with or without SES for 24 h. Experiments were repeated four times (n)
on separate occasions, except for MCP-1 (n ⫽ 3), and data were analyzed using a two-way repeated-measures ANOVA calculated in GraphPad Prism. Boldface type indicates
significant p-values, and underlined values are those that show a trend toward significance. Nonsignificant values are not shown. Interaction significant implies that R1 and R2
respond differently to SES. p-values for SES or cell type ⬍ 0.05 indicate that there is a significant effect of either on the response.
6518
12/15-LIPOXYGENASE IN INFLAMMATION
FIGURE 5. 12/15-LOX-deficient mice have more resident M␾ but similar
rates of proliferation and apoptosis to WT mice. A, 12/15-LOX⫺/⫺ mice have
more resident M␾ but similar numbers of DC and lymphocytes to WT mice.
Cells were quantified in peritoneal lavage of control mice using FACS as
described in Materials and Methods (n ⫽ 10 for both strains). Right panel
shows no difference in blood Mo, determined using FACS. B, 12/15-LOX
deficiency does not alter cell cycle or apoptosis rates of peritoneal cells. Cells
were fixed and stained using propidium iodide and analyzed using FACS as
described in Materials and Methods. C, Both strains have similar spleen
weights. Ten- to 12-wk-old mice were sacrificed and spleens weighed.
Functional and phenotypic characterization of 12/15-LOX⫹ and
12/15-LOX⫺ cells
A full phenotypic characterization of the cells was undertaken by
FACS analysis (Fig. 2). 12/15-LOX⫺ cells expressed higher levels
of MMR (CD206), dectin-1, M-CSF-R, GM-CSF-R, CCR5, MHC
class II, CD11c, Syndecan1, and MARCO than did the 12/15LOX⫹ cells. In contrast, 12/15-LOX⫹ cells express more SR-A
(CD204) and Siglec1 (CD169) than do 12/15-LOX⫺ cells. Both
populations lacked lymphoid lineage markers, such as CD4, CD3,
and CD8 (T cells), CD45R/B220 (B cells), polymorphonuclear cell
markers including 7/4 Ag and Ly6C/G (Gr-1), and also dectin-2,
CCR6, CCR2, CD49b, and CD83 (Fig. 2 and data not shown).
Both cell types expressed comparable levels of CD14, CD40,
TLR2, ICAM-1, CD62L, and the costimulatory molecules CD80
and CD86 (Fig. 2 and data not shown). These differences indicate
that the expression of 12/15-LOX is associated with distinct monocytic cell populations and is not expressed by other peritoneal leukocyte populations.
Functional differences between these M␾ subpopulations were
examined in vitro by cytokine profiling the responses of MoFlosorted cells to activation by SES, which acts predominantly via
TLR2 (C. Colmont, N. Topley, S. Jones, M. Labeta unpublished).
12/15-LOX⫹ M␾ produced greater quantities of IL-10 and G-CSF
than did 12/15-LOX⫺ cells (Fig. 3, A and B, and Table I). There
was a trend toward increased generation of IL-1␣, IL-1␤, IL-6,
TNF-␣, IL-12p70, IL-13, MCP-1/CCL2, and GM-CSF (Fig. 3C–J
and Table I). In contrast, LOX⫺ cells appeared to generate more
IL-12p40 (Fig. 3K and Table I). Both cell populations secreted
comparable levels of RANTES/CCL5, KC/CXCL1, and MIP-1␣/
CCL3 levels (Fig. 3L–N and Table I). There was, however, little or
no production of IL-2, IL-3, IL-4, IL-5, IL-17A, and IFN-␥ detected from either population (data not shown).
To assign an immunological function to these distinct monocytic
populations, studies examined their capacity to present Ag to naive T
cells. FACS-sorted 12/15-LOX⫺ cells stimulated DO11.10 CD4⫹ T
cell proliferation in presence of OVA, but 12/15-LOX⫹ cells did not
(Figs. 4, A and B). Furthermore, only 12/15-LOX⫺ cells stimulated
secretion of IL-2 by lymphocytes (Fig. 4C). Collectively, these studies
show that 12/15-LOX expression defines two phenotypically and
functionally distinct populations of resident M␾-like cells in the murine peritoneal cavity, and they suggest that 12/15-LOX⫺ cells
represent a novel population of DC.
Comparison of M␾ numbers and phenotype between WT and
12/15-LOX⫺/⫺ mice
There were 1.7-fold more CD11bhigh/F4/80high M␾ in the peritoneal cavity of 12/15-LOX⫺/⫺ mice compared with WT animals ( p ⫽ 0.00082). However, DC, lymphocytes, and blood Mo
numbers were identical (Fig. 5A). Propidium iodide staining of
total lavage cells or resident M␾ between WT and 12/15LOX⫺/⫺ mice was identical, indicating that the proportion of
cells undergoing proliferation or apoptosis was the same for the
two strains (Fig. 5B). This suggests that 12/15-LOX may exert
selective negative effects on Mo/M␾ migration into the peritoneal cavity. Furthermore, in the light of recent reports regarding
a role for 12/15-LOX in myeloid cell proliferation, we detected
no alteration in spleen size for mice aged 10 –12 wk between
strains (Fig. 5C) (13).
CD11bhigh/F4/80high resident cells isolated by adhesion from
12/15-LOX⫺/⫺ mice were phenotypically identical to those from
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 6. Resident peritoneal M␾ from WT and 12/15-LOX⫺/⫺ mice
show identical expression of surface Ags. Cells were analyzed for surface
expression of Ags using FACS. Resident M␾ were identified based on
forward light scatter/side light scatter and being CD11b⫹/F4/80high.
The Journal of Immunology
6519
WT animals for expression of Ags including CCR5, MHC class II,
dectin-1, dectin-2, MMR, CD11c, CD62L, ICAM-1, M-CSF-R,
MARCO, and CD40 (Fig. 6). However, some differences were
apparent in terms of cytokine generation in response to stimulation
with or without bacterial products (SES) in vitro (Fig. 7). Specifically, 12/15-LOX⫺/⫺ cells isolated by adhesion generated more
IL-1␣, IL-3, GM-CSF, and IL-17, but less RANTES/CCL5, than
did WT mice (Fig. 7 and Table II). SES-elicited cells behaved in
a somewhat overlapping manner, with 12/15-LOX⫺/⫺ cells showing reduced RANTES and increased GM-CSF production, but also
significantly reduced generation of IL-12p40 and IL-12p70 as
compared with WT (Fig. 7 and Table III). Overall, elicited cells
from either strain generated far greater quantities of cytokines in
response to stimulation than did resident cells. All other cytokine
levels remained unchanged between strains. Detailed statistical
analyses of all cytokines analyzed are in Tables II and III. A caveat
is that these cells isolated by adhesion were not a pure M␾ population, with resident cells comprising predominantly M␾ (82%)
and B lymphocytes (13.5%) with small numbers of T lymphocytes
and NK cells. SES-elicited cells comprised M␾ (75.5%), neutrophils (18.5%), B lymphocytes (5%), and NK cells (1%). This indicates that cytokines may not have solely originated from M␾ in
these experiments.
Changes in M␾ populations during peritoneal inflammation
To characterize how 12/15-LOX⫹ cells responded to acute inflammation, mice were administered (i.p.) SES and the temporal
changes in leukocyte numbers were recorded for 7 days. Following initiation of inflammation, peritoneal M␾ numbers rapidly declined during the first 3 h. However, these were replaced
by CD11bmed/F4/80med Mo after 6 –12 h (Fig. 8, A and B). This
population expressed 7/4 Ag, dectin-1, CD40, CCR5, and
CX3CR1 (data not shown). Concurrent analysis of 12/15-LOX
expression using immunohistochemistry showed a decline in
12/15-LOX expressing cells consistent with a decrease in the number of resident peritoneal M␾ (Fig. 8, C and D). The number of infiltrating Mo/M␾ increased during 12–72 h post-SES injection (Fig. 8,
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 7. Deficiency of 12/15-LOX causes selective alterations in response to SES stimulation in vitro. Left panels, Altered cytokine generation by
resident M␾. Resident peritoneal M␾ were isolated by adhesion, then cultured for 24 h with or without SES, and culture fluids were assayed for cytokine
generation by ELISA or Bio-Plex assay (n ⫽ 4, mean ⫾ SEM, ⴱ, p ⬍ 0.05 for effect of genotype using two-way ANOVA). Right panels, Altered cytokine
generation by elicited M␾. Mice were administered SES for 24 h before lavage. Elicited M␾ were isolated by adhesion, then cultured for 24 h with or
without SES, and culture fluids were assayed for cytokine generation by ELISA or Bio-Plex assay (n ⫽ 4, mean ⫾ SEM, ⴱ p ⬍ 0.05 for effect of genotype
using two-way ANOVA). Full statistical analyses of all cytokines analyzed in these populations are in Tables II and III.
6520
12/15-LIPOXYGENASE IN INFLAMMATION
Table II. Cytokine production by resident adherent peritoneal cellsa
Mean (⫾SEM), pg/ml
Cytokine/Growth
Factor
IL-1␣
IL-1␤
IL-3
IL-9
IL-10
IL-12p40
IL-12p70
IL-13
IL-17
RANTES
MCP-1
MIP-1␣
MIP-1␤
KC
TNF
G-CSF
GM-CSF
Two-way ANOVA ( p-value)
WT
WT ⫹ SES
KO
KO ⫹ SES
n
Interaction
SES
Genotype
48.52 (6.17)
24.35 (1.43)
2.74 (0.67)
33.24 (2.54)
11.81 (0.80)
58.81 (13.48)
10.98 (0.76)
41.95 (3.12)
43.87 (18.34)
250.02 (26.46)
223.96 (12.57)
363.84 (26.95)
223.49 (23.51)
214.05 (5.64)
52.51 (4.90)
6.75 (1.02)
65.12 (1.36)
95.97 (5.21)
92.75 (9.88)
24.51 (0.97)
84.12 (1.49)
170.78 (52.23)
559.18 (85.29)
1,510.15 (52.67)
1,274.91 (119.07)
2,126.59 (358.09)
80.92 (12.70)
44.33 (15.79)
5.47 (0.92)
32.88 (2.37)
11.47 (1.80)
48.66 (20.03)
11.37 (1.30)
42.65 (3.02)
160.33 (41.77)
60.45 (11.03)
253.62 (45.86)
416.53 (148.85)
473.18 (278.19)
324.45 (328.88)
98.76 (31.69)
10.88 (1.91)
61.55 (2.51)
82.42 (13.04)
104.07 (14.74)
26.70 (2.60)
85.27 (4.63)
397.98 (101.66)
186.11 (64.26)
1,303.48 (166.06)
1,113.48 (253.51)
3,535.53 (2,443.42)
0.0372
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
⬍0.0001
0.0346
0.0014
⬍0.0001
⬍0.0001
0.0088
⬍0.0001
⬍0.0001
0.0092
0.0012
⬍0.0001
0.0001
0.0621
0.0008
0.0826
0.0127
NS
NS
NS
NS
NS
0.0130
0.0001
NS
NS
NS
26.55 (4.28)
702.71 (208.87)
38.64 (7.71)
134.78 (22.50)
40,712.19 (3,426.16)
1,065.82 (353.42)
13.09 (4.19)
867.20 (171.84)
156.29 (33.18)
263.83 (89.88)
35,103.92 (6560.98)
2,776.11 (281.71)
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
5
NS
NS
0.0029
0.0014
⬍0.0001
⬍0.0001
NS
NS
0.0010
A and B). After 7 days, an F4/80high/CD11bhigh population was restored; however, a large population of F4/80med/CD11bmed monocytic cells remained within the peritoneal cavity (Fig. 8B). Parallel
analysis of 12/15-LOX expression within this emerging monocytic
infiltrate demonstrated a significant increase in the number of both
F4/80⫹/12/15-LOX⫹ and F4/80⫹/12/15-LOX⫺ cell types (Fig. 8, B
and C).
By day 7, greater numbers of both 12/15-LOX⫹ and 12/15LOX⫺ M␾ were found than at day 0 (Fig. 8C). It is possible that
some of the infiltrating Mo already express 12/15-LOX, while
other cells induce it during differentiation to a “resident-like”
phenotype. Significantly, blood Mo do not express 12/15-LOX,
suggesting that its expression in infiltrating Mo/M␾ is induced
early following their trafficking to the peritoneal cavity.
12/15-LOX activity profile generation during peritonitis
Mass spectrometry profiling of HETEs in lavage showed that 12HETE is the predominant positional isomer, confirming its enzymatic
production by the 12/15-LOX (Fig. 8E). Also, 12-HETE was virtually
absent in lavage from 12/15-LOX⫺/⫺ mice (Fig. 8E, right bars). Basally, low levels of 15-HETE, also formed by this isoform, were detected, but other isomers were largely absent. During SES peritonitis,
the concentration of 12-HETE decreased from 7.4 ng to ⬍1 ng per
mouse after 24 h of inflammation (Fig. 8E). This decrease followed
the disappearance of 12/15-LOX⫹ cells in the cavity and recovered
partially by day 7. There was a small increase in 5-HETE levels
during inflammation, but LXA4 was not detected (Fig. 8E and data
not shown).
Table III. Cytokine production by SES-elicited peritoneal adherent cellsa
Mean (⫾SEM), pg/ml
Cytokine/Growth
Factor
IL-1␣
IL-1␤
IL-3
IL-6
IL-9
IL-10
IL-12p40
IL-12p70
IL-13
IL-17
RANTES
MCP-1
MIP-1␣
MIP-1␤
KC
TNF
G-CSF
GM-CSF
Two-way ANOVA ( p-value)
WT
WT ⫹ SES
KO
KO ⫹ SES
n
Interaction
SES
Genotype
61.91 (6.01)
188.10 (39.94)
324.11 (56.47)
32,570 (8,259)
92.46 (8.34)
7.60 (0.80)
38,376 (18,269)
500.60 (87.14)
190.10 (77.69)
145.16 (41.84)
80,665 (2,836)
1,143.00 (127.70)
1,145.43 (118.15)
1,050.64 (46.45)
94,740 (9,729)
207.06 (7.20)
29.94 (0.86)
53,382 (15,495)
1,461.34 (148.34)
762.33 (54.54)
615.81 (53.24)
83,370 (2,300)
354.27 (117.94)
414.81 (138.52)
551.25 (132.00)
57,280 (26,440)
102.27 (4.88)
10.64 (1.13)
11,545 (8,217)
246.46 (57.88)
199.28 (38.54)
497.08 (281.30)
70,441 (3,109)
1,421.00 (323.84)
1,056.26 (144.64)
1,035.18 (119.25)
87,730 (820)
211.36 (7.53)
32.25 (2.58)
5,561 (730)
733.12 (152.49)
732.08 (22.02)
677.29 (244.89)
7,6210 (1,264)
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
NS
NS
NS
NS
NS
NS
NS
0.0686
NS
NS
NS
⬍0.0001
⬍0.0001
⬍0.0001
0.0102
⬍0.0001
⬍0.0001
NS
⬍0.0001
⬍0.0001
NS
NS
NS
NS
NS
NS
NS
0.0923
0.0122
0.0014
NS
NS
0.0043
53,658 (24898)
213.79 (64.07)
135.39 (9.29)
55.42 (7.06)
122,654 (3,308)
4,700.27 (2,061.84)
3,099.63 (248.11)
2,150.68 (141.19)
31,094 (16,476)
243.92 (33.61)
662.89 (186.37)
288.26 (109.34)
10,7811 (3,849)
3,262.93 (2,441.72)
4,542.16 (116.62)
3,105.91 (384.21)
NS
0.0004
NS
NS
NS
0.0369
⬍0.0001
⬍0.0001
NS
NS
NS
0.0160
a
SES-elicited peritoneal cells isolated by adhesion from individual C57BL/6 mice were cultured with and without SES. Data were analyzed using a two-way ANOVA in
GraphPad Prism (n ⫽ 4 separate mice). Boldface indicates significance ( p ⬍ 0.05) and underlined numbers are those that show a trend toward significance. Nonsignificant values
are not shown. Interaction significant implies that WT and knockout (KO) respond differently to SES. p-values for SES or genotype of ⬍0.05 indicate that there is a significant
effect of either on the response.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
a
Resident cells isolated by adhesion from individual C57BL/6 mice were cultured with and without SES. Data were analysed using a two-way ANOVA calculated in
GraphPad Prism (n ⫽ 5 separate mice). Boldface indicates significance ( p ⬍ 0.05) and underlined numbers are those that show trend toward significance. Nonsignificant values
are not shown. Interaction significant implies that WT and knockout (KO) respond differently to SES. p-values for SES or genotype of ⬍0.05 indicate that there is a significant
effect of either on the response.
The Journal of Immunology
6521
12/15-LOX suppresses Mo/M␾ numbers in the peritoneal cavity
during inflammation
Analysis of Mo/M␾ numbers (CD11b⫹/F4/80⫹) during SES-induced
peritonitis in WT mice showed a decrease in numbers during the first
3 h followed by influx of Mo/M␾ (CD11b⫹/F4/80⫹), which peaked
at 72 h (Figs. 8B and 9A). 12/15-LOX⫺/⫺ mice displayed the same
kinetics of Mo/M␾ migration into the peritoneal cavity, but with a
2-fold increase in the number of recruited cells after 24 h persisting to
7 days ( p ⫽ 0.0477 and p ⫽ 0.0435 at 24 and 7 days, respectively)
(Fig. 9A). Similarly, migration of DCs had significantly increased at
24 h, although numbers of these cells were still relatively low (Fig.
9B). In contrast, neutrophil and lymphocyte numbers in the peritoneal
cavity were unaffected by 12/15-LOX deletion (Fig. 9, C and D).
peak production of these inflammatory mediators in WT animals (Fig.
10A–F). Significantly higher levels of TGF-␤1 were found in 12/15LOX⫺/⫺ mice 24 h after the induction of peritonitis, correlating with
elevated numbers of inflammatory Mo/M␾ present at this time (see
above) (Fig. 10G). No substantive differences in IL-1␣, IL-9, IL-1,3 and
MIP-1␤/CCL4 were noted, while IL-2, IL-3, IL-4, IL-5, MIP-1␣/CCL3,
or IFN-␥ were low or below the limit of detection (data not shown).
During inflammation, lavage cell expression of the eicosanoid-generating
enzymes COX-1 and COX-2 were regulated in an inverse manner, with
COX-2 being transiently induced and COX-1 being suppressed but recovering by 24–72 h, similar to 12/15-LOX (Fig. 10, H and I). In
12/15-LOX⫺/⫺ mice, COX-1 and COX-2 expression followed a similar pattern during peritonitis, but, overall, expression of the two enzymes was lower than in WT mice (Fig. 10, H and I).
12/15-LOX regulates cytokines and eicosanoid-generating
enzymes during inflammation
Significantly lower levels of MCP-1/CCL2, RANTES/CCL5, IL12p40, IL-17, G-CSF, and TNF-␣ were generated in 12/15-LOX⫺/⫺
mice, particularly around 6 h after SES injection, which is the time of
Discussion
The role of 12/15-LOX expressed by peritoneal M␾ is unknown.
Some reports suggest a proinflammatory action of the enzyme in
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 8. Resident 12/15-LOX-expressing M␾ are rapidly cleared during inflammation, then reappear by day 7. A, FACS analysis shows disappearance
of R1 cells by 12 h after SES, with reappearance by day 7. Total peritoneal lavage was analyzed for F4/80 and CD11b as described in Materials and Methods.
Resident M␾ (res-M␾) are defined as CD11bhigh/F4/80high (left panel). Infiltrating Mo/M␾ (inf-Mo/M␾) are CD11bmed/F4/80med (center panel). B, Time course
of disappearance and reappearance of resident and infiltrating Mo/M␾. Cell numbers for res-M␾ and inf-Mo/M␾ were obtained by FACS analysis of lavage
obtained during the SES time course, as described in Materials and Methods (n ⫽ 10, mean ⫾ SEM). C, 12/15-LOX and F4/80high cells disappear during peritonitis
and reappear by day 7. Cytospins of peritoneal lavage were immunostained and analyzed by confocal microscopy as described in Materials and Methods (n ⫽
5, mean ⫾ SEM). D, Representative slides from C. Disappearance and reappearance of 12/15-LOX/F4/80⫹ cells are seen by day 7. E, 12/15-LOX; green, F4/80;
red. Peritoneal lavage contains 12-HETE that disappears during inflammation and reappears by day 7. HETE isomers were determined in lavage using liquid
chromatography/mass spectrometry/mass spectrometry as described in Materials and Methods (n ⫽ 8 or 5 for WT or 12/15-LOX⫺/⫺, respectively).
6522
12/15-LIPOXYGENASE IN INFLAMMATION
FIGURE 9. 12/15-LOX regulates
infiltration of Mo/M␾ and DC during
inflammation. 12/15-LOX⫺/⫺ mice
have significantly higher influx of
M␾/Mo and DC, but no differences in
neutrophil or lymphocytes, during
peritonitis. Cells were monitored using FACS, as described in Materials
and Methods. M␾/Mo (A), DC (B),
lymphocytes (C), and neutrophils (D)
(n ⫽ 10 for both strains, mean ⫾
SEM, ⴱ, p ⬍ 0.05 by Student’s t test).
crease in proliferation can mediate dramatic increases in peritoneal
M␾ numbers and detecting a small change may be beyond the
limit of detection for our method (26). Recently, it was proposed
that 12/15-LOX is a suppressor of myeloproliferative disease (13).
However, we found no increase in peripheral blood Mo nor splenomegaly in our mice (Fig. 5, A and C). Thus, the higher numbers of
peritoneal M␾ could result from greater rates of inward migration
and local differentiation or from slower rates of egress from the
peritoneal cavity. The significantly lower generation of MCP-1/
CCL2 and RANTES/CCL5 during acute inflammation in 12/15LOX⫺/⫺ mice indicates that the elevated Mo/M␾ levels are not
due to elevated levels of classical chemokines (Fig. 10).
During peritonitis, the rapid clearance of 12/15-LOX-expressing
M␾ as part of the “M␾ disappearance reaction” probably occurred
via migration to draining lymph nodes and increased adherence to
peritoneal tissues (Fig. 8, A and B) (27, 28). Resoration of local
expression may involve up-regulation of the enzyme in the recruited cells, as peripheral blood Mo do not express 12/15-LOX
(data not shown). Since substantial recovery in the levels of 12/
15-LOX products was not achieved until as long as 7 days poststimulation, it is possible that the enzyme is not fully active until
later time points (Fig. 8E). By this time, cells with the phenotype
of resident peritoneal M␾ have begun to repopulate the cavity. It
is not clear whether these cells have been recruited and differentiated from peripheral blood Mo or if they are from an alternate
source; however, their antigenic phenotype coupled with the repletion of the cavity with 12/15-LOX products provides compelling evidence that the inflammatory reaction is largely resolved.
To examine the impact of 12/15-LOX deficiency on the inflammatory response to SES in the context of the M␾ heterogeneity that
exists in vivo, we examined the response of both resident- and SESelicited adherent cells that consist primarily of M␾ to SES stimulation. This would model both the effects of initial in vivo challenge of
the resident population and the impact of 12/15-LOX on the differentially programmed M␾ recruited from the periphery during the response. Our analysis highlighted some overlapping and distinct aspects of the effect of 12/15-LOX expression on these two cell types
(Fig. 7 and Tables II and III). For example, both resident and SESelicited cells exhibited similar increases in GM-CSF and reductions in
RANTES in the absence of 12/15-LOX (Fig. 7, F and G). SES-elicited cells produced very high levels of IL-12p40 and IL-12p70, which
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
atherosclerosis (2, 14, 15). In contrast, studies on the pharmacological actions of exogenously administered lipoxins and resolvins
suggest an antiinflammatory role (7, 8, 16). Since the peritoneal
cavity is by far the major site of 12/15-LOX expression in vivo, we
sought to characterize its function in regulating inflammation in
this organ. High concentrations of exogenous 12/15-LOX products
or plant LOX are known to inhibit neutrophil migration into the
peritoneal cavity (7, 8). However, unlike the murine enzyme that
generates 12S-HETE, the plant LOX generates the 15S-positional
isomer, and the role of the endogenous 12/15-LOX has not been
explored in vivo. Herein, a model of sterile peritonitis was used
where a cell-free supernatant from S. epidermidis, one of the principal causative agents of human peritonitis in peritoneal dialysis
patients, is administered i.p. to create a well-characterized model
of mild acute and resolving inflammation (17).
Basally, 12/15-LOX⫹ cells accounted for 95% of the recovered
CD11bhigh cells with their phenotype consistent with previous
studies describing the expression pattern of resident peritoneal M␾
(11, 18 –21) (Fig. 2). In contrast, 12/15-LOX⫺ cells appeared to be
DC (Figs. 1, 2, and 4) (22). Although it is known that a small
number of human and rat peritoneal cells are DC, their phenotype
has not been described before (23, 24). Unlike the 12/15-LOX⫹
cells, the peritoneal DC expressed high levels of mannose receptor,
previously only documented on specific murine DC located in T
cell areas of lymphoid organs draining from the periphery (10, 12,
25). The comparison of SES response of purified resident peritoneal M␾ and DC populations in vitro demonstrates clear differences between the cell types. The greater generation of IL-10 and
G-CSF, with a trend toward increased production of several other
inflammatory mediators, suggests that the peritoneal M␾ is likely
to be the more important of these two populations as a source of
these cytokines during the early stages of the inflammatory response (Fig. 3 and Table I). The lack of differences in Ag expression between WT and 12/15-LOX⫺/⫺ resident M␾ indicates that
the enzyme is not required for maintaining the major phenotypic
characteristics of these cells (Fig. 6). However, the nearly 2-fold
elevation in peritoneal M␾ number in 12/15-LOX⫺/⫺ mice compared with WT mice, both before and during inflammation, shows
that it exerts selective negative effects on the number of M␾ (Figs.
5A and 9A). The reason for this did not appear to be either increased proliferative capacity or decreased apoptosis of these cells
within the peritoneal cavity (Fig. 5B), although only a slight in-
The Journal of Immunology
6523
in the case of IL-12p70 was further increased by in vitro SES stimulation (Fig. 7, C and D). Both IL-12p40 and IL-12p70 were significantly reduced in the 12/15-LOX⫺/⫺ cells, consistent with previous
reports of impaired IL-12 production in response to LPS of thioglycolate-elicited peritoneal M␾ (6). However, resident peritoneal cells
produced minimal IL-12 even after in vitro SES stimulation, indicating that these cells are a different population in terms of response to
bacterial products. Separately, the in vivo impact of 12/15-LOX deficiency was assessed by measuring the production of cytokines and
growth factors in peritoneal lavage of SES-challenged mice (Fig. 10).
Consistent with our results using SES-elicited cells, IL-12p40 and
RANTES were deficient in the absence of 12/15-LOX, but several
additional cytokines were also decreased, including MCP-1/CCL2,
IL-17, G-CSF, TNF-␣, COX-1, and COX-2 (Fig. 10). These results
suggest that infiltrating Mo may be the source of RANTES and IL12p40 in vivo, but that others (MCP-1, IL-17, G-CSF, TNF-␣) may
originate from other cell types. The increase in TGF-␤ production in
12/15-LOX⫺/⫺ mice at 24 h correlated with the marked increase in
Mo/M␾ numbers present in the peritoneal cavity, suggesting they
may be the source of this cytokine. Thus, the impact of 12/15-LOX on
the production of inflammatory mediators is influenced by cellular
heterogeneity and likely affects multiple cellular subsets in vivo.
Previous studies administering exogenous 12/15-LOX products
and plant 15-LOX proposed a role for the enzyme in regulating
neutrophil influx in peritoneal inflammation, through synthesis of
LXA4 and related mediators from omega3 fatty acids. Our results
show key differences: 1) there was no deficit in neutrophil influx,
with instead a selective suppression of Mo/M␾ migration; and 2)
LXA4 was not detected in our model even though the detection
limit of our mass spectrometry assay is well below the levels previously determined using ELISAs in other studies (29 –32). A caveat is that S. epidermidis-induced inflammation is bacterial rather
than fungal, and in terms of neutrophil recruitment is less severe
than the zymosan model (32, 33). The peak of LXA4 reported in
zymosan peritonitis occurs at 2– 4 h and decreases during resolution (32). In contrast, we found the opposite pattern, with 12/15LOX and 12-HETE disappearing during inflammation (Fig. 8).
Pharmacological effects of lipoxin analogs on leukocyte recruitment utilize far higher doses (300 ng/mouse, i.p.) than reported to
be generated in vivo (0.4 ng/mouse) (8, 32). In summary, a role for
LXA4 derived from endogenous 12/15-LOX in regulating peritoneal responses to bacterial products appears unlikely.
Receptor-dependent stimuli for 12-HETE production by 12/15LOX in M␾ are unknown. The high levels of 12-HETE in the
lavage of naive mice indicate that the enzyme is generating product in vivo without inflammatory stimulation (Fig. 8E). Treatment
of peritoneal lavage cells with SES in vivo or in vitro did not
activate 12/15-LOX (data not shown). Several additional agents
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
FIGURE 10. 12/15-LOX regulates production of cytokines and COXs during inflammation. A–G, Generation of several cytokines is altered in 12/15-LOX⫺/⫺
mice during peritonitis. Cytokine levels in peritoneal
lavage were determined using Bio-Plex mouse cytokine
assay kit, PCR (TNF-␣), or ELISA (TGF-␤1) as described in Materials and Methods (n ⫽ 8 –10 for both
strains, mean ⫾ SEM, ⴱ, p ⬍ 0.05 by Student’s t test).
H and I, Expression of COX-1 (H) and COX-2 (I) was
determined using PCR, as described in Materials and
Methods (n ⫽ 4 for both strains, mean ⫾ SEM, ⴱ, p ⬍
0.05 using unpaired Student’s t test).
6524
Disclosures
The authors have no financial conflicts of interest.
References
1. Kuhn, H., and B. J. Thiele. 1999. The diversity of the lipoxygenase family: many
sequence data but little information on biological significance. FEBS Lett. 449: 7–11.
2. Huo, Y., L. Zhao, M. C. Hyman, P. Shashkin, B. L. Harry, T. Burcin,
S. B. Forlow, M. A. Stark, D. F. Smith, S. Clarke, et al. 2004. Critical role of
macrophage 12/15-lipoxygenase for atherosclerosis in apolipoprotein E-deficient
mice. Circulation 110: 2024 –2031.
3. Cailhier, J. F., M. Partolina, S. Vuthoori, S. Wu, K. Ko, S. Watson, J. Savill,
J. Hughes, and R. A. Lang. 2005. Conditional macrophage ablation demonstrates
that resident macrophages initiate acute peritoneal inflammation. J. Immunol.
174: 2336 –2342.
4. Huang, J. T., J. S. Welch, M. Ricote, C. J. Binder, T. M. Willson, C. Kelly,
J. L. Witztum, C. D. Funk, D. Conrad, and C. K. Glass. 1999. Interleukin-4dependent production of PPAR-␥ ligands in macrophages by 12/15-lipoxygenase.
Nature 400: 378 –382.
5. Miller, Y. I., M. K. Chang, C. D. Funk, J. R. Feramisco, and J. L. Witztum. 2001.
12/15-Lipoxygenase translocation enhances site-specific actin polymerization in
macrophages phagocytosing apoptotic cells. J. Biol. Chem. 276: 19431–19439.
6. Zhao, L., C. A. Cuff, E. Moss, U. Wille, T. Cyrus, E. A. Klein, D. Pratico,
D. J. Rader, C. A. Hunter, E. Pure, and C. D. Funk. 2002. Selective interleukin-12
synthesis defect in 12/15-lipoxygenase-deficient macrophages associated with reduced atherosclerosis in a mouse model of familial hypercholesterolemia. J. Biol.
Chem. 277: 35350 –35356.
7. Bannenberg, G. L., J. Aliberti, S. Hong, A. Sher, and C. Serhan. 2004. Exogenous
pathogen and plant 15-lipoxygenase initiate endogenous lipoxin A4 biosynthesis.
J. Exp. Med. 199: 515–523.
8. Bannenberg, G. L., N. Chiang, A. Ariel, M. Arita, E. Tjonahen, K. H. Gotlinger,
S. Hong, and C. N. Serhan. 2005. Molecular circuits of resolution: formation and
actions of resolvins and protectins. J. Immunol. 174: 4345– 4355.
9. Cyrus, T., J. L. Witztum, D. J. Rader, R. Tangirala, S. Fazio, M. F. Linton, and
C. D. Funk. 1999. Disruption of the 12/15-lipoxygenase gene diminishes atherosclerosis in apo E-deficient mice. J. Clin. Invest. 103: 1597–1604.
10. Mackenzie, R. K., G. A. Coles, and J. D. Williams. 1991. The response of human
peritoneal macrophages to stimulation with bacteria isolated from episodes of
continuous ambulatory peritoneal dialysis-related peritonitis. J. Infect. Dis. 163:
837– 842.
11. Taylor, P. R., D. M. Reid, S. E. Heinsbroek, G. D. Brown, S. Gordon, and
S. Y. Wong. 2005. Dectin-2 is predominantly myeloid restricted and exhibits
unique activation-dependent expression on maturing inflammatory monocytes
elicited in vivo. Eur. J. Immunol. 35: 2163–2174.
12. Rosas, M., S. Gordon, and P. R. Taylor. 2007. Characterisation of the expression
and function of the granylocyte-macrophage colony-stimulating factor (GMCSF) reeptor ␣-chain in mice. Eur. J. Immunol. 37: 2518 –2528.
13. Middleton, M. K., A. M. Zukas, T. Rubinstein, M. Jacob, P. Zhu, L. Zhao,
I. Blair, and E. Pure. 2006. Identification of 12/15-lipoxygenase as a suppressor
of myeloproliferative disease. J. Exp. Med. 203: 2529 –2540.
14. Harats, D., A. Shaish, J. George, M. Mulkins, H. Kurihara, H. Levkovitz, and
E. Sigal. 2000. Overexpression of 15-lipoxygenase in vascular endothelium accelerates early atherosclerosis in LDL receptor-deficient mice. Arterioscler.
Thromb. Vasc. Biol. 20: 2100 –2105.
15. Reilly, K. B., S. Srinivasan, M. E. Hatley, M. K. Patricia, J. Lannigan, D. T. Bolick,
G. Vandenhoff, H. Pei, R. Natarajan, J. L. Nadler, and C. C. Hedrick. 2004. 12/15Lipoxygenase activity mediates inflammatory monocyte/endothelial interactions and
atherosclerosis in vivo. J. Biol. Chem. 279: 9440 –9450.
16. Serhan, C. N., A. Jain, S. Marleau, C. Clish, A. Kantarci, B. Behbehani,
S. P. Colgan, G. L. Stahl, A. Merched, N. A. Petasis, et al. 2003. Reduced
inflammation and tissue damage in transgenic rabbits overexpressing 15-lipoxygenase and endogenous anti-inflammatory lipid mediators. J. Immunol. 171:
6856 – 6865.
17. Hurst, S. M., T. S. Wilkinson, R. M. McLoughlin, S. Jones, S. Horiuchi,
N. Yamamoto, S. Rose-John, G. M. Fuller, N. Topley, and S. A. Jones. 2001. Il-6
and its soluble receptor orchestrate a temporal switch in the pattern of leukocyte
recruitment seen during acute inflammation. Immunity 14: 705–714.
18. Hughes, D. A., I. P. Fraser, and S. Gordon. 1995. Murine macrophage scavenger
receptor: in vivo expression and function as receptor for macrophage adhesion in
lymphoid and non-lymphoid organs. Eur. J. Immunol. 25: 466 – 473.
19. Kraal, G., L. J. van der Laan, O. Elomaa, and K. Tryggvason. 2000. The macrophage receptor MARCO. Microbes Infect. 2: 313–316.
20. Taylor, P. R., G. D. Brown, D. M. Reid, J. A. Willment, L. Martinez-Pomares,
S. Gordon, and S. Y. Wong. 2002. The ␤-glucan receptor, dectin-1, is predominantly expressed on the surface of cells of the monocyte/macrophage and neutrophil lineages. J. Immunol. 169: 3876 –3882.
21. Wu, Q., Y. Feng, Y. Yang, Jingliu, W. Zhou, P. He, R. Zhou, X. Li, and J. Zou.
2004. Kinetics of the phenotype and function of murine peritoneal macrophages
following acute inflammation. Cell. Mol. Immunol. 1: 57– 62.
22. Trinchieri, G. 1998. Interleukin-12: a cytokine at the interface of inflammation
and immunity. Adv. Immunol. 70: 83–243.
23. Betjes, M. G., C. W. Tuk, D. G. Struijk, R. T. Krediet, L. Arisz, and R. H. Beelen.
1993. Antigen-presenting capacity of macrophages and dendritic cells in the peritoneal cavity of patients treated with peritoneal dialysis. Clin. Exp. Immunol. 94:
377–384.
24. van Vugt, E., J. M. Arkema, M. A. Verdaasdonk, R. H. Beelen, and
E. W. Kamperdijk. 1991. Morphological and functional characteristics of rat
steady state peritoneal dendritic cells. Immunobiology 184: 14 –24.
25. McKenzie, E. J., P. R. Taylor, R. J. Stillion, A. D. Lucas, J. Harris, S. Gordon,
and L. Martinez-Pomares. 2007. Mannose receptor expression and function define a new population of murine dendritic cells. J. Immunol. 178: 4975– 4983.
26. Metcalf, D., M. J. Elliott, and N. A. Nicola. 1992. The excess numbers of peritoneal macrophages in granulocyte-macrophage colony-stimulating factor transgenic mice are generated by local proliferation. J. Exp. Med. 175: 877– 884.
27. Soesatyo, M., T. Thepen, M. Ghufron, J. Biewenga, and T. Sminia. 1993. Peritoneal cell labelling: a study on the migration of macrophages and dendritic cells
towards the gut. Adv. Exp. Med. Biol. 329: 321–326.
28. Taylor, P. R., G. D. Brown, A. B. Geldhof, L. Martinez-Pomares, and S. Gordon.
2003. Pattern recognition receptors and differentiation antigens define murine
myeloid cell heterogeneity ex vivo. Eur. J. Immunol. 33: 2090 –2097.
29. Aliberti, J., C. Serhan, and A. Sher. 2002. Parasite-induced lipoxin A4 is an
endogenous regulator of IL-12 production and immunopathology in Toxoplasma
gondii infection. J. Exp. Med. 196: 1253–1262.
30. Gronert, K., N. Maheshwari, N. Khan, I. R. Hassan, M. Dunn, and
M. Laniado Schwartzman. 2005. A role for the mouse 12/15-lipoxygenase pathway in promoting epithelial wound healing and host defense. J. Biol. Chem. 280:
15267–15278.
31. Levy, B. D., C. B. Clish, B. Schmidt, K. Gronert, and C. N. Serhan. 2001. Lipid
mediator class switching during acute inflammation: signals in resolution. Nat.
Immunol. 2: 612– 619.
32. Schwab, J. M., N. Chiang, M. Arita, and C. N. Serhan. 2007. Resolvin E1 and
protectin D1 activate inflammation-resolution programmes. Nature 447: 869 – 874.
33. Arita, M., S. F. Oh, T. Chonan, S. Hong, S. Elangovan, Y. P. Sun, J. Uddin,
N. A. Petasis, and C. N. Serhan. 2006. Metabolic inactivation of resolvin E1
and stabilization of its anti-inflammatory actions. J. Biol. Chem. 281:
22847–22854.
Downloaded from http://www.jimmunol.org/ by guest on July 31, 2017
from organisms, including yeast and bacteria, also failed to stimulate significant 12/15-LOX activity in isolated peritoneal cells
(including LPS, zymosan, ␤-glucan, flagelin, Pam3CSK4) (data
not shown). Overall, the pattern of 12-HETE generation in peritoneal cells is not consistent with a classic proinflammatory pathway. Studies have suggested that formation of 12-HETE or any
other known free acid product made by 12/15-LOX is not required
for its bioactivity. Both IL-12p40 generation in response to LPS
and phagocytosis of apoptotic thymocytes by elicited peritoneal
M␾ require 12/15-LOX, but both phenomena occur in WT M␾
without significant 12-HETE generation (5, 6). Our data extend
this idea, since while 12/15-LOX regulated the cellular and cytokine response to bacterial products in vivo, the enzyme was not
acutely activated. Also, we could not restore the WT phenotype to
12/15-LOX⫺/⫺ mice by i.p. administration of 12-HETE, live WT
lavage cells, or WT lavage cell lipid extract (data not shown). In all
of these experiments, lipids were added immediately before inflammatory activation, and thus an acute effect of exogenous lipids
does not seem to account for the effect of 12/15-LOX. It is possible
that the long-term absence of 12/15-LOX causes a fundamental
change to the biology of the peritoneal cavity so that it responds
differently to bacterial products. Alternatively, deletion of 12/15LOX during development may play a role in macrophage phenotype. Thus, further aspects of the mechanisms by which 12/15LOX regulates immune responses in vivo remain to be elucidated.
In summary, we show that endogenous 12/15-LOX is a marker
for resident peritoneal M␾ that regulate Mo/M␾ homeostasis and
cytokine production during acute inflammation. We propose that
the enzyme is a central player in regulating the M␾-dependent
response to an inflammatory stimulus in vivo.
12/15-LIPOXYGENASE IN INFLAMMATION