| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Production by Human Monocytes
Department of Medicine, Division of Rheumatology, University of Massachusetts Medical School, Worcester, MA 01655.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
from human
monocytes stimulated with LPS. LPS-induced IL-1 release is followed
by IL-1-induced IL-1 release, an amplification process termed
autoinduction. We show here with peripheral blood monocytes from normal
volunteers and from patients with rheumatoid arthritis by using IL-1R
antagonist to block autoinduction and IL-1
stimulation to simulate
autoinduction that 
40% of IL-1 released from LPS-stimulated
cells is attributable to autoinduction and that GLA reduces
autoinduction of IL-1 while leaving the initial IL-1 response to
LPS intact. Experiments with cells in which transcription and protein
synthesis were blocked suggest that GLA induces a protein that reduces
pro-IL-1 mRNA stability. IL-1 is important to host defense, but
the amplification mechanism may be excessive in genetically predisposed
patients. Thus, reduction of IL-1 autoinduction may be protective in
some patients with endotoxic shock and with diseases characterized by
chronic inflammation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
from
LPS-stimulated human PBMC (11).
IL-1 is an important mediator of inflammation and joint tissue
injury, and it has become a target of therapy for patients with RA
(12). However, like most mediators, IL-1 also modulates
cell function and participates in host defense (13, 14).
Thus, appropriate regulation of IL-1 production and secretion is
necessary for prevention of excessive immune/inflammatory responses.
The final biological effects of IL-1 depend on the balance between
the different elements of the IL-1 system (pro-IL-1, converting enzyme,
receptors, and receptor antagonist), most especially, perhaps, between
IL-1 and IL-1R antagonist (IL-1Ra). Pro-IL-1 is a 35-kDa molecule
that is converted by the action of IL-1-converting enzyme (ICE, now
referred to as caspase-1) to a 17-kDa "mature" IL-1, which is
the biologically active compound secreted by monocytes
(14). Peripheral blood monocytes (PBM) from RA patients
exhibit increased production of IL-1 compared with cells from normal
subjects and patients with osteoarthritis (15). Synovium
from RA patients is characterized by an imbalance between IL-1 and
IL-1Ra production, in favor of IL-1 (16). In the
studies presented here, mechanisms whereby GLA modulates IL-1
secretion were investigated in isolated PBM stimulated with either LPS
or by IL-1 itself. LPS stimulates an initial increase in IL-1
production and secretion, which in turn induces further IL-1
production, an amplification system termed autoinduction
(17). We show here in experiments with human cells in
vitro that GLA reduces mainly autoinduction of IL-1 in
LPS-stimulated cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mononuclear cells (PBMC) were isolated from peripheral blood of
healthy volunteers and of patients with RA using Ficoll-Hypaque
separation. Monocytes (PBM) were isolated by adherence in 24-well
tissue culture plates. Adherent cells were washed, and cells were
maintained overnight at 37°C, 5% CO2, in RPMI
1640 2% autologous serum. Cells were washed again, medium was
replaced, and experiments were performed with the rested cells
(18). Cells were treated or not with GLA for 60 min and
then stimulated with either 10 ng/ml LPS or 1 ng/ml recombinant human
IL-1 for 4 h (for mRNA determination in cells) or 18 h
(for IL-1 assay in supernatants).
ELISA
Mature, 17-kDa IL-1 was measured in supernatants with an
ELISA kit from R&D Systems (Minneapolis, MN). The Ab cross-reacts
10–15% with pro-IL-1. Our previous studies (19)
indicate that only 2% of IL-1 secreted from LPS-stimulated
monocytes is pro-IL-1. The sensitivity of the assay is 2 pg/ml.
Assay for gene expression
RNA isolation. Cells (up to 1 x 107), either in 250 µl of medium or with 250 µl of buffer added, were lysed in 0.75 ml TRI Reagent LS (Molecular Research Center, Cincinnati, OH). RNA was purified according to manufacturer’s instructions. Cells were homogenized by allowing the lysed samples to incubate at room temperature for 10 min, and then 200 ml of chloroform was added and samples were extracted. The aqueous phase was removed to new tubes, 500 ml of isopropanol was added, and samples were incubated at room temperature for 15 min. Samples then were centrifuged for 15 min at full speed in an Eppendorf model 5415c centrifuge (Eppendorf Scientific, Westbury, NY) to precipitate the RNA. Supernatant fluid was withdrawn and pellets were washed with 75% ethanol and centrifuged. Supernatants were withdrawn again, and pellets were allowed to air dry for 5 min.
Probe generation. A probe template set consisting of templates for the genes of interest and housekeeping genes L32 and GAPDH (BD PharMingen, San Diego, CA) were biotin labeled. A mixture consisting of 1 µl of the template set, 2 µl of biotin labeling mix (Boehringer-Mannheim, Indianapolis, IN), 2 µl of DTT, 4 µl of 5x transcription buffer, 1 µl of RNaisin, 1 µl of T7 polymerase, and 9 ml of water (all components of mixture with the exception of the biotin labeling mix are from the BD PharMingen in vitro transcription kit) was mixed and then incubated at 37°C for 1 h. DNase (20 µl) was added and the reaction mixture was incubated at 37°C for an additional 30 min. EDTA (26 µl of 20 mM), 25 µl of Tris-saturated phenol (Sigma, St. Louis, MO), 25 µl of chloroform/isoamyl alcohol (50:1), and 2 µl of yeast tRNA were added. The mixture was vortexed to emulsion and microcentrifuged at full speed for 5 min. The aqueous phase was removed to a new tube, and 50 µl of chloroform/isoamyl alcohol (50:1) was added. Subsequent to extraction, the aqueous phase was removed to a new tube, and 50 µl of 4 M ammonium acetate was added with 250 µl of ethanol. This mixture was incubated at -80°C for a minimum of 30 min and microcentrifuged at full speed at 4°C for 15 min. The supernatant was removed, and 100 µl of cold 90% ethanol added, followed by centrifugation at 4°C for 5 min. Subsequent to removal of all fluid, the pellet was allowed to air dry for 5 min. The labeled probe was dissolved in 50 µl of hybridization buffer.
RNase protection assay
RNA was dissolved in 18 µl of hybridization buffer, and 2 µl of a 1/1 dilution of probe was added to each RNA sample. The assay was conducted with a BD PharMingen RNase protection assay kit according to the manufacturer’s instructions. Subsequent to electrophoresis, RNA was transferred to a positively charged nylon membrane with a SemiPhor semidry transfer device (Hoeffer Scientific Instruments, San Francisco, CA). The membrane was UV-cross-linked in a Stratalinker (Stratagene, Palo Alto, CA) for 5 min. The membrane then was incubated with constant motion in blocking buffer for at least 1 h. Next, the membrane was incubated with Avidx-AP (Tropix, Bedford, MA) 1:20,000 in blocking buffer with constant motion for 30 min. The membrane then was washed five times for 10 min each in washing buffer. The membrane then was blotted on absorbent paper until damp, and placed into a plastic sleeve. CDP-Star RTU (Tropix) was added to cover the membrane entirely. The membrane was incubated at room temperature for 5 min, blotted to remove excess CDP-Star, placed in a new plastic cover, and exposed to film. Densitometry was conducted by scanning an image of the membrane into the NIH Image program (National Institutes of Health, Bethesda, MD). All reported levels are normalized for GAPDH levels.
Stability of pro-IL-1 mRNA
To investigate whether steady-state levels of pro IL-1 mRNA
are altered by changing its stability, new transcription was arrested
by addition to cells of 5 µg/ml actinomycin D 4 h after
stimulation. Steady-state mRNA levels were assayed at intervals to
determine decay of mRNA. Experiments in which 50 µg/ml cyclohexamide
also was added to halt new protein synthesis were performend in a
similar manner. Cytoplasmic RNA was obtained by a modification of the
method of Borun et al. (20). Cells suspended in 1.5 ml of
cold PBS were pelleted (microcentrifuged at full speed for 15 s)
and then washed in cold PBS. Pellets were resuspended in 45 µl of
cold Tris-EDTA (10 µM Tris and 1 µM EDTA), 5 µl of 5% Nonidet
P-40 was added, and the samples were placed on ice for 5 min. An
additional 5 µl of 5% Nonidet P-40 was added and the mixture was
microcentrifuged for 2.5 min at full speed. Supernatants (50 µl) were
transferred to tubes containing a fresh mixture of 30 µl 20x SSC and
20 µl of formaldehyde. These tubes were incubated at 60°C for 15
min and were either blotted directly to a positively charged nylon
membrane with a dot blot device or were stored at -80°C for future
blotting. The amount of RNA blotted was normalized for the number of
cells harvested.
After blotting, the membrane was UV-cross-linked for 5 min in a UV
Stratalinker. The membrane then was blocked with a hybridization
buffer, which consisted of: 50% formamide, 5x SSC, 0.1% sarcosyl,
0.02% SDS, 2% blocking buffer, and 0.2% salmon sperm DNA for 1
h at 42°C. A biotin-labeled probe for IL-1 was added to 10 ml of
hybridization buffer, boiled for 10 min, and added to the membrane. The
membrane was maintained with motion overnight at 42°C. The membrane
then was subjected to a series of washes in reagents provided in the
digoxigenin wash and block buffer was set according to the
manufacturer’s instructions, with the exceptions that the reagent used
in this protocol was AvidX-AP at a dilution of 1/20,000 in 1x blocking
buffer and the luminescent reagent was CDP Star (Tropix). The treated
blot then was exposed to Kodak XAR film (Eastman-Kodak, Rochester,
NY).
Statistical analysis
Data were analyzed and compared by paired Student’s t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
When freshly isolated monocytes were cultured in polypropylene
tubes and were used as soon as they adhered to plates, variable amounts
of IL-1 secretion were observed in the absence of cell stimulation.
However, when cells were cultured in polystyrene plates overnight,
background levels of IL-1 were low and induction of IL-1 and
effects of GLA were consistent. Thus, the mean ± SD secretion of
IL-1 from freshly isolated cells was 940 ± 670 pg/ml
(n = 4), whereas the baseline level of IL-1 secreted
from cells first cultured in polystyrene plates overnight was 48.8 ±
24 pg/ml (n = 5).
Separation of LPS induced from IL-1 induced IL-1
secretion
Because LPS-induced IL-1 includes IL-1 induced by LPS and
further IL-1 induction by IL-1 (autoinduction), we separated
LPS-induced IL-1 from autoinduction by pretreating LPS stimulated
cells with IL-1Ra. A 100-fold excess of IL-1Ra to IL-1 (100 ng/ml
IL-1Ra added to cells stimulated with 1 ng/ml IL-1) completely
prevented secretion of IL-1-induced IL-1 (Fig. 1
A), whereas IL-1 secretion
from PBM exposed to 100 ng/ml IL-1Ra and stimulated with LPS was
reduced by 38.9% (Fig. 1B). Thus, in the experiment shown,
LPS stimulation accounted for 61.1% of IL-1 secretion, and the
remaining 38.9% of IL-1 secretion was attributable to
autoinduction. This ratio of LPS induced to autoinduced IL-1 varies
from donor to donor.
|
secretion
We then determined the effect of GLA on LPS-induced IL-1
secretion from cells first treated with IL-1Ra and on IL-1-induced
IL-1 secretion (simulated autoinduction). Addition of GLA in vitro
reduced secretion of IL-1 in a dose-dependent manner. However, GLA
reduced IL-1 secretion much more effectively when cells were
stimulated by IL-1 than when they were stimulated by LPS (Fig. 2). In a series of experiments, a 10
µg/ml concentration of GLA reduced LPS-induced IL-1 secretion by
37.9 ± 10.7% (mean ± SD; n = 9;
p < 0.01 vs untreated control cells). In contrast,
treatment of cells with 10 µg/ml GLA reduced IL-1-induced IL-1
secretion by 61.5 ± 19.3% (mean ± SD; n =
6; p < 0.01 vs untreated control cells).
|
gene expression
Concentrations of GLA that reduced LPS-induced IL-1 secretion
(Fig. 2
) did not reduce pro-IL-1 mRNA (Figs. 3 and 4).
However, GLA did reduce IL-1-induced pro-IL-1 mRNA markedly (Figs. 3 and 4). The observation suggests further that the GLA effect on IL-1
secretion was on autoinduction rather than on the initial LPS-induced
IL-1 production.
|
|
mRNA stability
The different effects of GLA on steady-state levels of LPS and
IL-1-induced pro-IL-1 mRNA were not attributable to an influence on
mRNA stability as determined by the change in cellular message content
after cells were treated with actinomycin D to stop transcription. GLA
did not change pro-IL-1 mRNA stability whether cells were stimulated
with LPS (Fig. 5, A and
C) or IL-1 (data not shown). However, when protein
synthesis was inhibited by cycloheximide in addition to inhibition of
transcription by actinomycin D, a difference in mRNA stability between
untreated and GLA treated cells emerged (Fig. 5, B and
D). In GLA-treated cells (Fig. 5D), stability of
the pro-IL-1 message was reduced markedly (53% remaining after
16 h; t1/2 = 17.1 h)
compared with untreated cells (94% remaining after 16 h;
t1/2 = 128.9 h). The data shown in
Fig. 5, A and D suggest that GLA treatment
induced a protein that reduced pro-IL-1 mRNA stability.
|
To determine whether the continuous presence of GLA in the
stimulated cell cultures was required for the inhibitory activity,
washout studies were done in which PBM were treated with GLA for 1
h and then washed extensively and further incubated with LPS or IL-1
alone for 18 h. The results (Table I) show that GLA had to be present
continuously to suppress IL-1 secretion.
|
In general, PBM from RA patients are more activated than PBM from
controls (15). Nonetheless, the differential effects of
GLA on LPS-induced and autoinduced IL-1 secretion also were seen in
PBM from RA patients (Fig. 6 and Table II). In these experiments,
autoinduction of IL-1 (as simulated by stimulation with IL-1) was
blocked by incubation of cells with 100 ng/ml of IL-1Ra before
stimulation with LPS. Results of the experiment with cells from a
61-year-old woman with an 8-year history of RA (Fig. 6A)
indicate that 52.0% of LPS-induced IL-1 secretion was attributable
to LPS alone (IL-1Ra plus LPS: autoinduction blocked). This portion of
IL-1 secretion was not reduced by addition to cells of GLA (IL-1Ra
plus LPS plus GLA). IL-1 induced by LPS was reduced 50.6% (LPS plus
GLA). Responses (documented in Fig. 6 and Table II) varied among the
four RA patients.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
production and secretion and
thereby modulate its biological activity. Joint tissue injury in RA
patients appears attributable in large part to activation of
mononuclear cells and synovial cells with subsequent release of IL-1
and other cytokines (21). Interference with production and
action of these polypeptide mediators of inflammation has emerged as a
major strategy for treatment of RA patients (12).
The addition of GLA to PBM in vitro reduces release of IL-1 from
LPS-stimulated cells, but GLA does not influence steady-state
pro-IL-1 mRNA levels, suggesting that GLA affects a process that
occurs after LPS induces an initial production of IL-1. One such
event, which contributes to total secretion of IL-1 from
LPS-stimulated cells, is the subsequent autoinduction of pro-IL-1
production by the IL-1 initially secreted (22).
Treatment of cells with IL-1Ra before stimulation with LPS prevents
autoinduction and separates LPS-induced from autoinduced IL-1. The
addition of GLA reduces markedly IL-1 autoinduction as simulated by
IL-1 stimulation. In contrast, GLA reduces the LPS-induced portion
of IL-1 release modestly, as illustrated by experiments using
exogenous IL-1Ra to block autoinduction. Donor cell variability is
reflected in the amount of IL-1 secreted in response to cell
stimulation, the response to GLA, and the proportion of LPS-induced
IL-1 secretion that is due to autoinduction. In general,
autoinduction accounted for 40% of total IL-1 secretion. Our
studies confirm others, which indicate that self-induction of IL-1
contributes significantly to IL-1 synthesis in stimulated human
monocytes (23).
Variability of responses appears to be even greater among patients with
RA (Fig. 6 and Table II). It is not clear whether these cellular
responses will predict clinical responses to treatment with GLA. If
they do, we would predict that patients C and D (Table II), whose cells
exhibit prominent autoinduction of IL-1, will respond well to GLA,
whereas patient B, in whom autoinduction is minimal, will not respond
well clinically to GLA.
The host defense properties of IL-1 include reduction of mortality from
bacterial and fungal infections in animals. Administration of IL-1 to
humans increases the number of bone marrow precursor cells and
circulating platelets and neutrophils. However, increasing doses cause
fever, gastrointestinal disturbances, myalgia, arthralgia, and
hypotension (14). It is possible that an initial insult to
the cell provokes a protective burst of IL-1 and that the amplification
system (autoinduction) is a backup defense mechanism. If in endotoxic
shock or in diseases characterized by chronic inflammation,
autoinduction of IL-1 is not well regulated, the excessive activity of
IL-1 would then be pathological rather than protective. Among
similar patients who suffer similar bacterial infections, some survive
while others succumb to shock and multiple organ failure
(24), perhaps due to different genetic predispositions to
exuberant inflammation. Proper regulation of IL-1 autoinduction may be
one component of an appropriate response to LPS and other stimuli.
We have shown (25) that GLA alters the cascade that
transduces signals from plasma membrane to nucleus during activation of
human T lymphocytes. The signaling pathways leading to IL-1
production after stimulation of human PBM with LPS and IL-1 have not
been characterized fully. It does appear that the pathways diverge.
Although both LPS and IL-1 initiate a broad range of signaling
cascades, several of which may influence IL-1 production
(26), IL-1 initiates a more narrow spectrum of cascades
that leads to production of IL-1 (27). For example, p38
mitogen-activated protein kinase is needed for successful induction of
IL-1 by LPS (28), but the enzyme is not necessary for
IL-1 induction of IL-1 (29). It is likely that one or
more of the redundant pathways activated by LPS are not affected by
GLA. It is possible that GLA influences signal transduction
pathways through an effect on peroxisome
proliferator-activated receptors (PPARs). PPARs,
intracellular receptors that act as transcription factors and mediate
ligand-dependent transcriptional activation and repression
(30), are cellular receptors for long chain fatty
acids, including GLA (31). It will be important to compare
in detail the varied LPS- and IL-1-induced signaling pathways in an
effort to better understand regulation of the amplification of IL-1
production observed in chronic inflammation and, even more tragically,
in septic shock.
Another potential mechanism whereby GLA might reduce IL-1 autoinduction
is suggested by the observations of pro-IL-1 mRNA stability
illustrated in Fig. 5. In these experiments, inhibition of translation
by cycloheximide in actinomycin D-treated, LPS-stimulated cells results
in stability of pro-IL-1 message that is greater than in
LPS-stimulated cells treated with actinomycin alone, in which
transcription is halted and translation proceeds (Fig. 5, B
vs A). These observations suggest that a protein synthesized
de novo upon LPS stimulation leads, directly or indirectly, to
destabilization of pro-IL-1 mRNA, or that translation itself leads
to destabilization of the message. When transcription is stopped in
GLA-treated cells and translation proceeds, pro-IL-1 mRNA is
degraded as it is in cells not treated with GLA (Fig. 5, C
vs A), and GLA does not appear to alter pro-IL-1 mRNA
stability. However, when translation is stopped by cycloheximide, the
innate stability of the pro-IL-1 mRNA is observed (Fig. 5B). In GLA-treated cells in which translation is stopped,
the instability of mRNA induced by GLA, unrelated to the translation of
this specific pro-IL-1 mRNA (translation of which has been stopped
by the cycloheximide), is revealed (Fig. 5D). The
observation suggests that GLA induces a protein that destabilizes
(accelerates decay of) pro-IL-1 mRNA. If GLA is indeed destabilizing
message induced by LPS, then it is possible that by the time the
LPS-induced IL-1 is ready to induce (autoinduce) further IL-1
production, the pro-IL-1 mRNA is too unstable to support
transcription. That GLA reduces LPS-induced pro-IL-1 mRNA 20 h
after cells are stimulated but not at 4 h after stimulation (data
not shown) also suggests that possibility.
Although the biological activity of IL-1 is mainly extracellular,
the protein lacks a secretory signal sequence (32).
IL-1 is synthesized by monocytes as a 35-kDa precursor (pro-IL-1)
that accumulates in the cytosol (33) and is
proteolytically processed to the mature 17-kDa IL-1 by the
IL-1-converting enzyme caspase-1 (34). In LPS-activated
human monocytes, IL-1 is contained in endolysosomes, exocytosis of
which leads to extracellular release of the cytokine (35).
Although it appears that the major mechanism whereby GLA
reduces secretion of IL-1 from LPS-stimulated cells is by reduction
of autoinduction, an effect on secretion itself is not excluded by our
studies. Increases in cellular concentrations of cAMP result in reduced
secretion of lysosomal products from activated human neutrophils
(36), and DGLA increases cAMP in human synovial cells
(2). Thus, the ability of GLA to alter IL-1 secretion
directly needs to be addressed.
These studies suggest that GLA, an unsaturated fatty acid with the
ability to reduce active synovitis in patients with RA, may exert its
therapeutic effects, at least in part, by modulation of IL-1
secretion from activated cells. GLA does not interfere significantly
with production of IL-1 due to LPS stimulation, but does suppress
the secondary, autoinduced (IL-1-induced) stimulation of IL-1
production. It is possible that dysregulation of the amplification
process known as autoinduction leads to unrestrained inflammation in
genetically predisposed individuals.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert B. Zurier, Department of Medicine, Division of Rheumatology, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655.
3 Abbreviations used in this paper: GLA, gammalinolenic acid; DGLA, dihomogammalinolenic acid; RA, rheumatoid arthritis; IL-1Ra, IL-1R antagonist; PBM, peripheral blood monocytes; PPAR, peroxisome proliferator-activated receptor.
Received for publication June 27, 2000. Accepted for publication April 30, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and tumor necrosis- secretion by stimulated human peripheral blood monocytes: studies in vitro and in vivo. J. Invest. Med. 47:246.[Medline]
in rheumatoid arthritis. Arthritis Rheum. 38:151.[Medline]
, interleukin-1, and interleukin-6: mediation by decreased mRNA stability. Mol. Pharmacol. 43:176.[Abstract]
production by human monocytes. Cytokine 9:1008.[Medline]
mitogen-activated protein kinase in lipopolysacharide-stimulated neutrophils. J. Clin. Invest. 103:851.[Medline]
(PPAR-) mediates the action of gammalinolenic acid in breast cancer cells. Prostaglandins Leukot. Essent. Fatty Acids 62:19.
is localized in the cytoplasmic ground substance but is largely absent from the Golgi apparatus and the plasma membrane of stimulated human monocytes. J. Exp. Med. 167:389. converting enzyme. Science 256:97. involves exocytosis of endolysosome-related vesicles. Mol. Biol. Cell. 10:1463.This article has been cited by other articles:
|
|
 |
|
  C. Song, M. S. Manku, and D. F. Horrobin Long-Chain Polyunsaturated Fatty Acids Modulate Interleukin-1{beta}-Induced Changes in Behavior, Monoaminergic Neurotransmitters, and Brain Inflammation in Rats J. Nutr., May 1, 2008; 138(5): 954 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Aragona, C. Bucolo, R. Spinella, S. Giuffrida, and G. Ferreri Systemic Omega-6 Essential Fatty Acid Treatment and PGE1 Tear Content in Sjogren's Syndrome Patients Invest. Ophthalmol. Vis. Sci., December 1, 2005; 46(12): 4474 - 4479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Xing and D. G. Remick Mechanisms of Dimethyl Sulfoxide Augmentation of IL-1{beta} Production J. Immunol., May 15, 2005; 174(10): 6195 - 6202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Song, X. Li, B. E. Leonard, and D. F. Horrobin Effects of dietary n-3 or n-6 fatty acids on interleukin-1{beta}-induced anxiety, stress, and inflammatory responses in rats J. Lipid Res., October 1, 2003; 44(10): 1984 - 1991. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
