| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Biology (L.R., P.C.-O., Y.C., T.S., P.O.) and Division of Bone and Mineral Research (P.C.-O., Y.C., L.C., L.A., P.O.), Washington University, St. Louis, Missouri 63130; and the Department of Rheumatology (A.T., S.G.), Harvard Institutes of Medicine, Boston, Massachusetts 02115
Address all correspondence and requests for reprints to: Dr. Philip Osdoby, Department of Biology, Box 1229, Washington University, St. Louis, Missouri 63130. E-mail: osdoby{at}biodec.wustl.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
.
At a cellular level, in situ hybridization analysis
revealed that IL-8 mRNA was expressed in resorbing hOC of rheumatoid
arthritic pannus and was substantially greater than that expressed in
hOC of noninflammatory giant cell tumor of bone tissue. Therefore, the
potential inflammation-mediated induction of IL-8 was directly assessed
using cultured hOCL. IL-8 release was stimulated by proinflammatory
signals (IL-1, tumor necrosis factor-, lipopolysaccharide, or
phorbol 12-myristate 13-acetate), unaffected by various other
osteotropic modulators (transforming growth factor-ß1 and -ß3,
IL-6, 17ß-estradiol, or calcitonin) and was decreased by
interferon-
, vitamin D3, and the antiinflammatory
glucocorticoid dexamethasone. Changes in IL-8 secretion were paralleled
by corresponding changes in IL-8 mRNA steady state levels. We conclude
that hOC and hOCL synthesize and secrete high constitutive and
inflammation-stimulated levels of the chemokine IL-8. Consequently,
hOC-derived IL-8 could act as an important regulatory signal for bone,
vascular, and immune cell recruitment and activation during normal and
pathological bone remodeling. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(TNF), or granulocyte-macrophage
colony-stimulating factor] or other immune stimuli [bacterial
lipopolysaccharide (LPS), lectins, and protein kinase C activation].
The precursor protein (99 amino acids) undergoes cell type-dependent
sequential N-terminal proteolytic processing to yield products (69–79
amino acids) of varying bioactivities (6). In addition to serving as a
potent chemoattractant activator of neutrophils, IL-8 exerts other
diverse proinflammatory or physiological effects on target cells,
including the stimulation of hematopoiesis, angiogenesis, mitogenesis,
bioactive lipid formation, protease activation, the release of
lysosomal enzymes, reactive oxygen species, cytokines, and nitric
oxide, and rapid changes in the expression or conformation of cell
surface adhesion proteins (e.g. ß-integrins and Mac-1),
F-actin, and cell shape (2, 3, 4, 5, 6, 7, 8, 9, 10, 11). Such biological actions are typically
mediated via elevated intracellular Ca2+ levels (12) that
occur following IL-8 ligand binding to specific cell surface receptors
that belong to the superfamily of seven-transmembrane spanning G
protein-coupled receptors (2, 3, 4, 5, 6, 7, 8, 9, 13). Consistent with its
proinflammatory actions in vitro, IL-8 levels are markedly
elevated in various inflammatory disorders, such as rheumatoid
arthritis, osteoarthritis, periodontal disease, psoriatic lesions,
pulmonary diseases, endotoxemia and sepsis, gastrointestinal
inflammation, and immune complex glomerulonephritis (2, 7, 8, 9). Whereas
ip or sc injections of IL-8 cause neutrophil migration, infiltration,
and activation at such sites (2, 7, 8, 9), neutralizing antibodies to IL-8
conversely suppress acute inflammatory reactions in various
inflammatory conditions (8, 14). These and many other studies have
shown IL-8 to be a major mediator involved in biological inflammatory
responses.
Bone loss is a typical hallmark of many inflammatory conditions and is
thought to be a consequence of both reduced osteoblast bone formation
and increased osteoclast (OC) bone resorption. Recently, we showed that
IL-8 (0.1–1.0 nM) directly inhibits osteoblast
characteristics important for their bone formative function (15, 16).
Therefore, a local rise in IL-8 levels could presumably contribute to
regional bone loss through suppressive effects on osteoblasts. At
higher concentrations (
10 nM), IL-8 also partially
inhibits the overall bone-resorptive activity of isolated rat and chick
OC; however, this appears to reflect IL-8-evoked increases in OC
motility to new sites of resorption rather than actual declines in the
resorptive capability of individual OC per se (15, 17, 18).
Formerly, human osteoblast-like cells (hOBL) were shown to also produce
this chemokine, the levels of which were elevated by inflammatory
cytokine stimulation and conversely suppressed by the antiinflammatory
glucocorticoid dexamethasone (19). However, aside from these initial
studies, little is known about the role of IL-8 in bone physiology,
specifically in relation to either osteoblast or OC development or
function. During the course of studies employing human OC-like cells
(hOCL), which exhibit most of the attributes of bone-resorptive OC
(20), we discovered that high levels of IL-8 were released into the
culture medium. This prompted us to ask whether human OC (hOC) might
also produce this potent intercellular signal molecule, a question that
had not been addressed before the current work. Over the past few
years, OC have been increasingly recognized as capable of producing a
variety of autocrine/paracrine-acting bone-active cytokines, free
radicals, matrix proteins, and other signal molecules. Thus, we
explored whether bone-resorbing hOC and related hOCL expressed
messenger RNA (mRNA) for IL-8, released measurable levels of this
chemokine, and exhibited altered IL-8 production in response to
inflammatory stimuli. Moreover, IL-8 mRNA expression levels in
resorbing hOC were assessed by in situ hybridization in
inflammatory and noninflammatory human bone tissues to learn whether
IL-8 levels were correlated with conditions of localized
osteopenia.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
107) were isolated
from femoral heads discarded during hip replacement surgery on
osteoporotic patients or from segments of bone removed from sites of
implant loosening during resection (21). Briefly, the bone was cut into
small pieces and trypsin/collagenase treated with shaking, and the
released cells were enriched (to >40% and >80% on a per cell and a
per nuclear basis, respectively) for hOC by Percoll fractionation or
serum sedimentation. Highly purified (
90% on a per cell basis,
>98% on a per nuclear basis) hOC preparations were obtained by
immunomagnetic sorting using the specific anti-OC monoclonal antibody
(Mab) 121F (22). hOC preparations were cultured in OC medium [phenol
red-free medium 199 Earle’s salts with 8.3 mM
NaHCO3, 100 mM HEPES (pH 6.8), 5% FBS (Life
Technologies, Gaithersburg, MD), and 2.5% antibiotic/antimycotic] at
37 C in a 95% air-5% CO2 moist environment. The cells
expressed characteristic OC features: high tartrate-resistant acid
phosphatase activity, vitronectin receptor
(Vß3 integrin), 121F Mab-reactive antigen,
mRNAs for calcitonin receptor, cathepsin K/O, and carbonic anhydrase
II, and resorption pit formation during culture on ivory or devitalized
bovine cortical bone slices (21). In vitro formed hOCL were
generated using bone marrow from discarded surgical specimens of human
rib bone, the mononuclear cells were Ficoll-Hypaque (Pharmacia,
Piscataway, NJ) separated, and the cells were resuspended in MEM
(Life Technologies) with 10% FBS and 2% antibiotic/antimycotic for
culture at 40 x 106 cells/100-mm dish (21, 22). On
day 6, nonadherent cells were replated in medium with 10 nM
1,25-dihydroxyvitamin D3 and thereafter refed every 3 days
with medium containing 25% conditioned medium (CM) from UMR 106-01 OBL
to further OC-like differentiation (22, 23). After 2–3 weeks, stromal
cells were selectively removed by brief trypsinization, and hOCL were
detached to replate into 24-well dishes at 0.8 x 105
cells/well (21). After 48 h, hOCL were switched to phenol red-free
medium, modulators were given in fresh medium 1–2 days later, and CM
was collected, spun, and stored at -80 C. These multinucleated hOCL
express the key OC phenotypic features listed above for hOC, but form
few, if any, resorption pits in culture on ivory or bone slices (20, 21).
Human bone marrow stromal (hBMS) and hOBL cells
hBMS were obtained as the first adherent cell population of the
initial plating of human marrow mononuclear cells (above) based on the
report of Cheng et al. (24). Briefly, after removal of the
day 6 nonadherent cells, the adherent cells were cultured to
confluence, trypsin detached, and replated at 2 x 105
cells/100-mm dish (21). hBMS were refed before 48-h CM collection from
near-confluent cultures. For hOBL differentiation, 10-7
M dexamethasone was added to the proliferating hBMS (day 2)
and withdrawn (day 7) before collection of 48-h CM (24).
Modulator treatments
The following reagents were obtained from Sigma Chemical Co.
(St. Louis, MO): Escherichia coli LPS serotype 0111:B4
stored at 4 C as a 5 mg/ml stock in Hanks’ Balanced Salt Solution,
phorbol 12-myristate 13-acetate (PMA) stored at -80 C as a 8.1 x
10-4 M stock in ethanol, dexamethasone freshly
prepared as a 10-3 M stock in ethanol,
17ß-estradiol stored at -20 C as a 3.7 x 10-4
M stock in ethanol, and BSA fraction V (<0.1 ng/ml
endotoxin). 1,25-(OH)2D3, a gift from
Hoffmann-La Roche (Nutley, NJ), was stored dry at -80 C and
reconstituted at 10-3 M in ethanol before use.
Salmon calcitonin, also a gift from Hoffmann-La Roche, was stored as
10-3-M aliquots (4411 IU/mg) in acetate
solution at -80 C. Human recombinant cytokines [interferon-
(IFN), IL-1, IL-6, TNF, transforming growth factor-ß1
(TGFß1), and TGFß3] were purchased from R&D Systems (Minneapolis,
MN), reconstituted in PBS with 0.1% low endotoxin BSA (and 4
mM HCl for TGFß stocks), and stored as aliquots at -80
C. Modulators were diluted in medium just before their administration
to cells, vehicle controls were tested in parallel, and CM was
collected after 24 h, spun, and stored at -80 C.
Quantification of cytokine release
Cytokine levels in CM were measured using specific enzyme-linked
immunoassay kits (R&D Systems) as recommended. Standard curves were run
with every assay. Each modulator was tested in three or more separate
culture wells per trial, and two to six independent trials were
performed per condition. Corresponding control medium for each trial
exhibited insignificant reactivity with each cytokine enzyme-linked
immunosorbent assay. Minimum detectable cytokine levels were 3.0 pg/ml
IL-8, 0.064 pg/ml IL-6, 0.1 pg/ml IL-1ß, and 0.11 pg/ml TNF.
Results are expressed as nanograms of cytokine released per ml culture
medium or as nanograms of cytokine released per 106 cells.
Normalization to cell protein was performed using harvested cells in
the bicinchoninic acid protein assay (Pierce, Rockford, IL) with BSA as
a standard.
RNA isolation and RT-PCR
Total RNA was extracted using RNA STAT-60 (Tel-Test,
Friendswood, TX), first strand complementary DNA (cDNA) synthesized
using a cDNA cycle kit (Invitrogen, San Diego, CA), and PCR reactions
were performed using specific primer pairs from these locations: 1)
192–212 and 338–317 of GenBank locus HSMDNCF for the 147-bp IL-8
amplicon, 2) 340–359 and 494–475 from locus HUMACP5 for the 155-bp
tartrate-resistant acid phosphatase amplicon, 3) 409–426 and 661–642
from locus HUMCALREC for the 252-bp calcitonin receptor
amplicon, 4) 3325–3344 and 3764–3747 from locus HUMGAPDHG for the
221-bp glyceraldehyde-3-phosphate dehydrogenase amplicon, and 5)
749–770 and 1093–1071 for the first round, and 800–822 and
1093–1071 for the second round, from locus HUMERMCF for the 249-bp
estrogen receptor (ER) amplicon as previously described (25). PCR
products were separated in a 1.5% agarose gel and ethidium bromide
stained. All were of the expected size for the primers used, and
sequencing of select amplicon products, including IL-8 and estrogen
receptor, was performed to confirm their identity.
Northern blot analysis
Total RNA (20–25 µg) was separated on denaturing 1%
agarose/formaldehyde gels (26), ethidium bromide stained to assess RNA
integrity/loading, transferred to nylon membranes (Schleicher and
Schuell), fixed by UV cross-linking (UV Stratalinker 2400, Stratagene,
La Jolla, CA), and prehybridized, and the blots were hybridized
overnight at 42–45 C with a labeled 406-bp human IL-8 cDNA probe
(provided by Dr. S. Kunkel, Pathology, University of Michigan Medical
School, Ann Arbor, MI) (27) generated by random primer extension using
[32P]deoxy-CTP (New England Nuclear, Boston, MA) or by
digoxygenin-11-deoxy-UTP labeling (Genius, Boehringer Mannheim,
Indianapolis, IN). After hybridization with 106 cpm/ml
32P-labeled probe (see Fig. 7A
) or 10-10
M digoxygenin-11-deoxy-UTP-labeled probe (see Fig. 7B),
specific signals were detected by autoradiography or chemiluminescence
on XAR-5 film (Eastman Kodak, Rochester, NY). Stripped blots were
rehybridized with a labeled cDNA probe for 18S RNA (provided by Dr. J.
Lian (26)) for normalization of IL-8 signals. Blots were quantified
(sum of the gray area) using a densitometer (Hewlett-Packard Scanjet,
Palo Alto, CA) linked to a Leica Quantimet image analysis system
(Leica, Deerfield, IL), and the results were expressed in arbitrary
densitometric units as relative ratios for IL-8/18S.
|
5 weeks) in 14% EDTA,
washed in graded ethanols, transferred to xylenes, and embedded in
paraffin, and thin sections (5 µm) were cut and dried onto
poly-L-lysine-coated slides (28, 29). Slides were
deparaffinized in xylenes, rehydrated in graded ethanols, rinsed (5 min
each in 0.85% NaCl and PBS), refixed (4% PF, 20 min), PBS washed,
proteinase K treated (20 µg/ml, 20 min), dipped in PBS (5 min) and
then in 4% PF (5 min), acetylated in 0.25% acetic anhydride/100
mM triethanolamine, washed (5 min each in PBS and 0.85%
NaCl), dehydrated in graded ethanols, and air-dried. Sense and
antisense riboprobes were prepared from linearized IL-8 cDNA vector
(above) by incubation with T7 or SP6 RNA polymerase and
[35S]UTP (New England Nuclear) in the presence of
unlabeled nucleotides, 10 mM dithiothreitol, and
ribonuclease (RNase) inhibitor (Promega, Madison, WI), and the probes
were purified using NucTrap push columns (Stratagene, La Jolla, CA).
Probes (5 x 104 cpm/µl) were diluted in
hybridization buffer [4 x SSC (standard saline citrate), 50%
formamide, 10 mM Tris-HCl (pH 7.5), 1 mM EDTA
(pH 8.0), 10% dextran sulfate, 1 x Denhardt’s solution, 0.5
mg/ml total yeast RNA, 50 µg/ml yeast transfer RNA, 100 µg/ml
denatured salmon sperm DNA, 100 mM dithiothreitol, 0.1%
SDS, and 0.1% sodium thiosulfate], 50 µl were added per slide, a
glass coverslip was placed on top, and the slides were incubated (55 C,
16 h), washed (four times, 5 min each time, 25 C, 2 x SSC-10
mM 2-mercaptoethanol-1 mM EDTA), and RNase A
treated (20 µg/ml, 30 min; Pharmacia Biotech, Piscataway, NJ). Washed
slides (three times, 5 min each time, 60 C, 0.1 x SSC) were
dehydrated in graded ammonium acetate-ethanol solutions, air-dried,
dipped in Kodak NTB-2 emulsion, drained, air-dried (1 h), and placed in
a light-proof container with dessicant at 4 C for 4–8 weeks. Slides
were developed in Kodak D-19 developer, fixed, and hematoxylin/eosin
counterstained. Controls included sense probes and RNase treatment of
tissue sections before hybridization.
Statistical analysis
Data are presented as the mean ± SEM of at
least three culture wells in each of two to six independent trials.
Differences between treatments were analyzed using one-factor ANOVA.
For simultaneous comparisons between multiple treatments, significant
differences were determined using the post-ANOVA Bonferroni test.
Differences were considered significant for P <
0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). By
comparison, the same CM samples contained much lower basal release
levels of the three other key inflammatory cytokines, IL-6 (150-fold),
TNF (1,000-fold), and IL-1ß (17,000-fold), on either a per cell or
a per ml basis in relation to IL-8 (Table 1). Whether authentic
bone-resorptive hOC also released significant basal amounts of IL-8 was
analyzed using hOC that were isolated and partially purified in
quantity from human femoral head samples, and subsequently cultured for
24 h in vitro (Table 1). Like hOCL, hOC exhibited high
levels of IL-8 release (197 ng/106 cells or 48 ng/ml) into
the CM, and hOC secretion of IL-8 exceeded the release of TNF (by
750-fold), IL-6 (by 25- to 40-fold), and IL-1ß (by 200-fold).
Somewhat unexpected was the finding that IL-1ß release was
substantially greater (by 150-fold) in authentic hOC cultures than
in hOCL cultures (Table 1); possibly this relates to having isolated
hOC from osteoporotic bone, whereas marrow cells were obtained from
normal human bone for hOCL formation. Otherwise, hOC and hOCL released
comparable amounts within the same order of magnitude of IL-8, IL-6,
and TNF. In contrast to hOC and hOCL, basal IL-8 release by human
bone cell populations unrelated to the hematopoietic lineage, such as
hBMS and stromal-derived hOBL cells, was notably lower on both a
nanograms per ml and per cell basis (Table 2). Although IL-8 release from
differentiated hOBL was 10-fold greater than that from hBMS
osteoprogenitor cells, such levels were still 10-fold below the high
constitutive levels of IL-8 released by hOC and hOCL. Consistent with
detecting IL-8 protein in the CM, IL-8 mRNA was present in both hOC and
hOCL based on RT-PCR analysis (Fig. 1).
Therefore, both hOC and hOCL express IL-8 mRNA and release high levels
of IL-8 into the CM during their culture in vitro.
|
|
|
) was seen for hOC in sections from a
noninflammatory sample such as human giant cell tumor of bone (Fig. 2, B and C), a more pronounced and specific hybridization signal was
localized to hOC of rheumatoid arthritic bone that had been similarly
processed (Fig. 2, E and G). Negligible hybridization signals were
obtained using control sense probes on parallel tissue sections from
either the rheumatoid arthritic (Fig. 2, D and F) or giant cell tumor
of bone samples (Fig. 2A), or if the sections were pretreated with
RNase before hybridization with the IL-8 antisense probe (not shown).
Therefore, the IL-8 antisense signal localization to individual
resorbing hOC of human bone tissue indicates that hOC synthesize mRNA
for IL-8 and, consequently, may contribute to or account for the
release of this potent proinflammatory factor in isolated hOC cultures.
Moreover, hOC IL-8 mRNA levels in vivo may be significantly
elevated in inflammatory bone disorders associated with osteopenia,
such as rheumatoid arthritis.
|
. Because hOCL express
multiple OC characteristics and release IL-8 at levels comparable to
hOC, and larger numbers of hOCL are more readily available and of
greater purity than isolated hOC, hOCL were employed in further studies
to investigate the potential regulation of IL-8 production in response
to inflammatory signals. The proinflammatory cytokines IL-1 and
TNF each significantly and dose dependently raised the levels of
IL-8 released into the CM beyond the already high basal level exhibited
by hOCL (Fig. 3, A and B), and such rises
were initially detectable as early as 6 h after stimulation (not
shown). Bacterial LPS generated a striking rise in the IL-8 release by
these cells (Fig. 3C), as did the phorbol ester PMA, which directly
activates protein kinase C, an intracellular signal commonly triggered
by inflammatory stimuli (Fig. 3D).
|
or TNF, other
bone-active cytokines such as IL-6
(10-8–10-11 M), TGFß1
(10-8–10-12 M), or TGFß3
(10-8–10-12 M) did not stimulate
IL-8 release from hOCL (not shown), whereas IFN (50, 100, or 200
U/ml) reduced constitutive IL-8 release from hOCL (by 60 ± 2%,
62 ± 3%, and 63 ± 4%, respectively; all P
< 0.005). Neither of the antiresorptive osteotropic hormones,
17ß-estradiol (10-6–10-10 M)
or calcitonin (10-6 to 10-10 M),
altered the basal level of IL-8 secretion from hOCL (not shown),
although mRNA for receptors of both of these hormones were present in
cultured hOCL (Fig. 4). Small, but
significant, reductions in basal IL-8 release by hOCL occurred after
treatment with 10-6–10-8 M
1,25-(OH)2D3 (Fig. 5A). However, a more marked and
dose-dependent suppression of basal IL-8 release by hOCL was achieved
in response to 10-9 or 10-8 M of
the antiinflammatory glucocorticoid dexamethasone (Table 3), which was not further reduced at a
concentration of 10-7 M of this osteotropic
hormone (20 ± 4% of control; P < 0.001).
|
|
|
and TNF on stimulating
IL-8 release from hOCL, IL-1 significantly and dose dependently
increased the hOCL release of both TNF and IL-6 (Table
4), and TNF dose dependently elevated
both IL-1ß and IL-6 release (Table 3
). However, just as IL-6
treatment of hOCL did not modify IL-8 release, neither did IL-6 alter
IL-1ß or TNF release from hOCL (Table 3). The antiinflammatory
glucocorticoid dexamethasone conversely reduced basal IL-8 release
concurrent with that of IL-1ß, IL-6, and TNF (Table 3). Calcitonin
(10-10–10-6 M) and
17ß-estradiol (10-10–10-6 M),
which had no effects on basal IL-8 production by hOCL, also did not
alter IL-1ß or IL-6 release by hOCL (not shown). Thus, complex
interdependent regulatory pathways govern the production of these
inflammatory-associated bone-active cytokines from hOCL, which might
generate autocrine feedback and amplification regulatory circuits.
Cytokine stimulated levels of IL-8 release from hOCL are inhibited
by 1,25-(OH)2D3 or by dexamethasone
As 1,25-(OH)2D3 (Fig. 5A) and dexamethasone (Table 4) each reduced basal IL-8 release from
hOCL, their effects on stimulated IL-8 release were investigated.
IL-1 stimulated IL-8 release was decreased by cotreatment of hOCL
with 1,25-(OH)2D3, although
such IL-8 levels remained significantly raised over those of
unstimulated control cultures in the presence of as much as
10-7 M of this hormone (Fig. 5B). The very
high levels of IL-8 release stimulated by either LPS (0.1–10 µg/ml)
or PMA (10-8–10-6 M) were not
reduced by cotreatment of hOCL with 10-8 M
1,25-(OH)2D3 (not shown).
Dexamethasone was more potent than
1,25-(OH)2D3 in diminishing
both basal (Fig. 6A) and
IL-1-stimulated (Fig. 6B) IL-8 release. In addition, dexamethasone
potently and dose dependently decreased hOCL release of IL-8 stimulated
by either TNF (Fig. 6C) or LPS (Fig. 6D). In each case,
10-7 M dexamethasone was sufficient to return
stimulated IL-8 release to levels indistinguishable from control
values. In contrast, IL-1 (10-10
M)-stimulated IL-8 release by hOCL was not altered by
cotreatment with calcitonin (10-8 M),
17ß-estradiol (10-8 M), TGFß1
(10-10–10-8 M), or TGFß3
(10-10–10-8 M; not shown).
|
). Similarly, steady state IL-8
mRNA levels were elevated by IL-1 treatment of hOCL (Fig. 7B).
Conversely, cotreatment with dexamethasone dramatically reduced, by 4-
to 6-fold, the elevated IL-8 steady state mRNA levels elicited by
either LPS or IL-1 (Fig. 7B). Consistent with its reduction of basal
IL-8 release from hOCL (Table 4), dexamethasone also decreased (by
2-fold) the basal IL-8 mRNA steady state levels in hOCL (not shown).
These findings were confirmed and extended by RNase protection assays,
which also demonstrated that IL-8 mRNA levels were partially elevated
by 2 h, peaked within 4–8 h, and thereafter declined by 24 h
in response to IL-1 stimulation (not shown). Thus, the regulation of
IL-8 cytokine release by hOCL in response to LPS, PMA, IL-1, and
dexamethasone is reflected in corresponding parallel changes in hOCL
steady state expression levels for IL-8 mRNA. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and IL-1ß in vivo (30); that rabbit OC
contain mRNA encoding IL-1ß, IL-6, IL-8, and the chemokine epithelial
neutrophil activating peptide-78 (31); and that mouse OC express
mRNA for the chemokine macrophage inflammatory protein-1 (MIP-1)
in vivo (32). This and other recent evidence suggest that OC
may function as both effector and target cells in local
autocrine/paracrine regulatory circuits that govern normal and
pathological bone remodeling. Here, we report on our detailed studies
of the expression, production, and regulation of the proinflammatory
chemokine IL-8, in relation to IL-1ß, TNF, and IL-6, by isolated
authentic bone-resorptive hOC and closely related in vitro
formed bone marrow-derived hOCL. The findings have demonstrated for the
first time that hOC and hOCL express mRNA for IL-8 and secrete high
basal levels of this potent chemokine, which can be further increased
by inflammatory conditions or signals. Overall, the similar basal
levels (within 1.5- to 4-fold) of IL-8, TNF, and IL-6 secreted by
cultured hOC and hOCL provide further evidence of the OC-like nature
that develops during the in vitro formation of hOCL (20, 22, 33). More importantly, compared with the relatively low basal levels of
IL-1ß, TNF, and IL-6 produced by cultured hOC and hOCL, IL-8
release was orders of magnitude higher (on a per cell or a per ml
basis). In fact, the levels of IL-8 constitutively released by cultured
hOC (48 ng/ml; 6 nM) and hOCL (22 ng/ml; 2.75
nM) or released by cultured hOCL in response to
inflammatory stimulation by IL-1 or TNF (elevated 2- to 4-fold to
5–11 nM) were similar to those reported for endothelial
cells (34, 35, 36) and activated neutrophils or macrophages (37, 38, 39), cells
that are considered primary effector cells for mediating physiological
and pathophysiological immune-related processes. Thus, IL-8 release by
hOC and hOCL was at concentrations known to be biologically effective
for physiological cell responses via either of the two known high
affinity cell surface IL-8 receptors (CXCR-1 and CXCR-2), each of which
exhibits Kd values ranging from 0.1–10 nM in
various cell types (40). Although it is possible that other cells
present in the hOC preparations may have contributed to the levels of
IL-8 release measured, we confirmed that IL-8 mRNA was expressed in
highly purified 121F Mab immunomagnetically isolated hOC preparations
as well as in highly purified hOCL cultures and showed by in
situ hybridization analysis that IL-8 mRNA expression was
localized to individual resorbing hOC of human bone tissue. Therefore,
hOC potentially represent an additional source of this potent
chemoattractant and activating factor in bone, and its regulated
release by hOC could exert important paracrine or autocrine modulatory
influences on cells within the bone microenvironment, affecting cell
recruitment, attachment, differentiation, and physiological
function.
Virtually all IL-8-producing cell types respond to inflammatory
mediators, especially IL-1 and TNF, with increased IL-8 mRNA
expression and IL-8 production (2, 3, 4, 19, 27). Several transcriptional
regulatory elements in the promoter region of the human IL-8 gene,
including activating protein-1 (AP-1), AP-2, nuclear factor-
B,
nuclear factor-IL-6-C/EBP, hepatocyte nuclear factor-1 (HNF-1),
glucocorticoid response element, and IFN regulatory factor-1 sites,
mediate these responses (3). In hOCL, as in other cells, both IL-8 mRNA
levels and IL-8 release were raised by the inflammatory stimuli PMA,
LPS, IL-1, and TNF, and conversely decreased by the
antiinflammatory glucocorticoid dexamethasone. In contrast, IL-6 did
not promote hOCL release of IL-8, IL-1ß, or TNF, consistent with
its general inability to induce such cytokines in a broad range of cell
types (41), although in one report IL-6 induced the chemokine monocyte
chemoattractant protein-1 in human and rat osteoblasts (42). Although
TGFß isoforms have variably had no effect on (2), augmented (43), or
inhibited (44) cytokine-stimulated IL-8 production in other cells, and
inhibited MIP-1 production by bone marrow macrophages (45), neither
TFGß1 nor TGFß3 altered either basal or IL-1-induced IL-8
secretion by hOCL. Whereas 1,25-(OH)2D3
decreased both basal and IL-1-stimulated IL-8 release by hOCL,
similar to its suppression of IL-8 production by monocytes,
keratinocytes, or fibroblasts (3), such inhibition in hOCL was only
weakly dose dependent and did not surpass a 30% decline in stimulated
IL-8 levels even at hormonal concentrations as high as
10-7 M, in contrast to the potent inhibitory
effects of dexamethasone. Finally, neither of the bone resorption
inhibitory hormones, calcitonin or 17ß-estradiol, directly influenced
either basal or IL-1-stimulated IL-8 release from hOCL, although
hOCL expressed mRNA for each receptor and have responded in other ways
to calcitonin (46) and 17ß-estradiol (47). Estrogen also does not
affect basal or stimulated IL-8 release by hBMS, hOB, or human MG-63
osteosarcoma cells (19, 48, 49), although estrogen can inhibit
cytokine-induced MCP-1 production in fibroblasts (50). Thus, hOCL
production of IL-8 is predominantly regulated by inflammatory mediators
and not by other osteotropic factors or hormones. Recent work has also
suggested more complex temporal modes of regulating IL-8 production in
hOCL, such as via estrogen modulation of IL-1 receptor expression in
hOCL, which subsequently modifies IL-1-mediated induction of IL-8 mRNA
(47).
Evidence for the in vivo expression of IL-8 by resorbing hOC and its regulation by inflammatory stimuli was obtained from in situ hybridization studies of human bone tissues. Thus, IL-8 mRNA levels were higher in the resorbing hOC of rheumatoid arthritic pannus than in hOC of giant cell tumors of bone. Although both of these conditions represent pathological states in which OC numbers are dramatically increased over the low number usually observed in human bone, inflammation is not considered a hallmark of giant cell tumors of bone, whereas rheumatoid arthritis is an autoimmune disease characterized by chronic joint inflammation that leads to a progressive destruction of cartilage and bone tissue in the afflicted joint (30, 51, 52, 53). Preliminary in situ hybridization studies on human osteonecrotic bone samples have similarly evidenced only low IL-8 mRNA levels in the resorbing hOC of this noninflammatory tissue (15). Therefore, hOC recruited into the developing pannus region in rheumatoid arthritis may respond to local stimulatory signals elaborated in this microenvironment by increasing their synthesis of IL-8, which could then contribute to the maintenance and amplification of the inflammatory response in this disorder. Chemokines, including IL-8, play key roles in both the initiation and maintenance phases of the evolutionary pathogenesis of rheumatoid arthritis by virtue of their capacity to recruit and activate leukocyte populations (monocytes and neutrophils) infiltrating the synovial space and joint tissue, to stimulate synovial cell proliferation and pannus formation at the juncture of the synovium lining the joint capsule with the cartilage and bone, and to activate endothelial cells and promote neovascularization, which further favors the delivery of immune cells from the vascular compartment into the inflamed joint (51, 52, 53, 54). Pre-OC are also likely to be recruited, whereupon their exposure to locally high concentrations of inflammatory mediators may promote their development and bone-resorptive activity. Whereas limiting chemokine levels appear optimal for the recruitment and passage of blood cells across the vascular endothelium into the inflamed tissue site, adhesion and activation of cells once in this location typically require high saturating concentrations of chemokines (55), such as those detected (up to 220 ng/ml IL-8) in the synovial fluid and serum of rheumatoid arthritic individuals (38) and shown here to be released by cultured hOC and hOCL. Consequently, hOC may significantly contribute to such IL-8 increases and thereby exert paracrine inflammatory or immunoregulatory influences on other cells infiltrating or present within the arthritic lesion.
Both stimulatory and suppressive factors are commonly coproduced during
inflammation, and their relative balance is thought to provide a
mechanism for the sensitive and rapid modulation of the inflammatory
response (38). Although IL-8 appears to inhibit overall bone resorption
by isolated avian (17) or rat (18) OC, this has been attributed at
least in the rat system to a decrease in the proportion of OC actively
resorbing bone due to an IL-8-stimulated increase in their motility,
rather than to a suppression of the actual bone-resorptive capacity of
individual OC. Therefore, it is also possible that hOC-derived IL-8
exerts important autocrine actions, similar to those seen in
neutrophils (30), which might enhance OC degradation of bone at
inflammatory sites. This is consistent with the elevated numbers of OC
and augmented bone resorption activity exhibited in rheumatoid
arthritis seen here and in other studies. One potential
resorption-stimulating mechanism might be via IL-8 triggering of OC
release of superoxides (56), reactive oxygen species involved in OC
bone resorption (57, 58). In addition, hOC-derived IL-8 might elicit
recruitment and activation of additional OC, because select chemokines,
including IL-8 (17, 18) and MIP-1 (59, 60), promote migration or
chemotaxis of avian, rat, or human OC (42, 47, 48, 59, 61, 62). In
conclusion, we have shown that isolated bone-resorptive hOC and
in vitro formed hOCL synthesize and secrete high levels of
the potent proinflammatory chemokine IL-8, that mRNA and cytokine
levels for IL-8 are further raised in hOCL in response to various
inflammatory stimuli, and that IL-8 mRNA expression is localized
in situ to resorbing hOC of human bone tissues, wherein it
is particularly evident in inflammatory rheumatoid arthritic pannus.
Therefore, hOC-derived IL-8 may serve as an important
autocrine/paracrine mediator of bone cell physiology and
immunoregulation involved in normal or pathological bone
remodeling.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received January 27, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-8 stimulate the motility but suppress
the resorption of isolated rat osteoclasts. J Immunol 154:6065–6072[Abstract]
-induced interleukin-8 production in human bone
marrow stromal and osteoblast-like cells. Calcif Tissue Int 55:16–20[CrossRef][Medline]
and ß mRNA during differentiation of human osteoblast
SV-HFO cells. Endocrinology 138:5067–5070. J Immunol 156:1566–1571[Abstract]
, and 1,25-D3
induce distinct temporal patterns of cytokine expression by
osteoblasts. J Bone Miner Res [Suppl 1] 11:S166
receptor are chemotactic for human osteoclast precursors. J Bone
Miner Res [Suppl 1] 11:S285
This article has been cited by other articles:
 |
|
 |
|
  D. Onan, E. H. Allan, J. M. W. Quinn, J. H. Gooi, S. Pompolo, N. A. Sims, M. T. Gillespie, and T. J. Martin The Chemokine Cxcl1 Is a Novel Target Gene of Parathyroid Hormone (PTH)/PTH-Related Protein in Committed Osteoblasts Endocrinology, May 1, 2009; 150(5): 2244 - 2253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lorenzo, M. Horowitz, and Y. Choi Osteoimmunology: Interactions of the Bone and Immune System Endocr. Rev., June 1, 2008; 29(4): 403 - 440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Yoneda, N. Ishimaru, R. Arakaki, M. Kobayashi, T. Izawa, K. Moriyama, and Y. Hayashi Estrogen Deficiency Accelerates Murine Autoimmune Arthritis Associated with Receptor Activator of Nuclear Factor-{kappa}B Ligand-Mediated Osteoclastogenesis Endocrinology, May 1, 2004; 145(5): 2384 - 2391. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A S Taha, V Grant, and R W Kelly Urinalysis for interleukin-8 in the non-invasive diagnosis of acute and chronic inflammatory diseases Postgrad. Med. J., March 1, 2003; 79(929): 159 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. H. Cottler-Fox, T. Lapidot, I. Petit, O. Kollet, J. F. DiPersio, D. Link, and S. Devine Stem Cell Mobilization Hematology, January 1, 2003; 2003(1): 419 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Gurlek, M. R. Pittelkow, and R. Kumar Modulation of Growth Factor/Cytokine Synthesis and Signaling by 1{alpha},25-Dihydroxyvitamin D3: Implications in Cell Growth and Differentiation Endocr. Rev., December 1, 2002; 23(6): 763 - 786. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Bendre, D. Gaddy-Kurten, T. Mon-Foote, N. S. Akel, R. A. Skinner, R. W. Nicholas, and L. J. Suva Expression of Interleukin 8 and not Parathyroid Hormone-related Protein by Human Breast Cancer Cells Correlates with Bone Metastasis in Vivo Cancer Res., October 1, 2002; 62(19): 5571 - 5579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. N. Das Nitric Oxide as the Mediator of the Antiosteoporotic Actions of Estrogen, Statins, and Essential Fatty Acids Experimental Biology and Medicine, February 1, 2002; 227(2): 88 - 93. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Masood, S. Nagpal, T. Zheng, J. Cai, A. Tulpule, D. L. Smith, and P. S. Gill Kaposi sarcoma is a therapeutic target for vitamin D3 receptor agonist Blood, November 1, 2000; 96(9): 3188 - 3194. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Schierano, F. Bassi, G. Gassino, K. Mareschi, G. Bellone, and G. Preti Cytokine Production and Bone Remodeling in Patients wearing Overdentures on Oral Implants Journal of Dental Research, September 1, 2000; 79(9): 1675 - 1682. [Abstract] [PDF]
|
|
|
|
 |
|
 P. Collin-Osdoby, L. Rothe, F. Anderson, M. Nelson, W. Maloney, and P. Osdoby Receptor Activator of NF-kappa B and Osteoprotegerin Expression by Human Microvascular Endothelial Cells, Regulation by Inflammatory Cytokines, and Role in Human Osteoclastogenesis J. Biol. Chem., June 1, 2001; 276(23): 20659 - 20672. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
