| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*
Division of Rheumatology, Department of Medicine, Winthrop University Hospital, Mineola, NY 11501, and State University of New York, Stony Brook, NY 11794; and
Division of Rheumatology, Department of Medicine, Nassau University Medical Center, East Meadow, NY 11554
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, IL-1, TNF) over
Th2 (IL-4) responses, whereas lymphoid DCs may facilitate the
activation of Th2 responses and promote negative selection in the
thymus (4, 8, 9, 10). Importantly, DC functions are also
related to stages of maturity. Fully matured DCs are equipped to
deliver antigenic signals to T cells in a MHC-restricted manner, but do
not efficiently handle Ag. Conversely, immature DCs present Ag poorly,
but avidly internalize and process Ag (2, 4). Recent
evidence also indicates that immature DCs may directly activate
regulatory T cells, which suppress the development of effector T cells
(2, 11, 12). Arguably, DC function is not fixed, but is
highly influenced by extracellular signals. While immature DCs and DC
precursors/progenitors may be highly susceptible to external
instruction, mature DCs appear to be more resistant
(9).
Within the myeloid DC lineage, two distinct DC pathways have been
demonstrated (5, 6, 7). One pathway yields the CD14-derived
(monocyte-derived) DC subtype. Intermediate developmental stages of
this pathway include: myelodendritic progenitor cells exhibiting
monocyte-, granulocyte-, and DC-differentiating potential,
CD14dim precursors with either DC- or
monocyte-differentiating potential, and
CD14dimCD1a+ precursors
that are DC committed and that develop from
CD14dim precursors (5, 6, 7, 13). The
second myeloid DC pathway produces the CD14-independent DC subtype
(Langerhans cell) and includes
CD14negativeCD1a+
precursors (6, 7, 13). TNF/GM-CSF +/- TGF
appear to be
primary growth factors for the CD14-independent pathway (14, 15), while the CD14-derived DC pathway develops with GM-CSF +
IL-4/IL-13 (16, 17, 18). A third DC subtype, the
lymphoid-related DC, may stem from common NK-T cell precursors under
the influence of IL-3/CD40 ligand, but not GM-CSF (1, 2, 4, 19, 20, 21).
DCs are thought to play an important role in driving immunopathogenic
responses that lead to the establishment of chronic proliferative
synovitis and joint destruction in rheumatoid arthritis (RA)
(22, 23, 24, 25, 26, 27, 28, 29, 30, 31). Recent studies concerning DC activity in RA have
revealed that different stages of DCs and distinct DC subtypes may be
represented in various synovial microenvironments (30). In
inflamed RA synovial tissue, most APCs are fully differentiated DCs
expressing high levels of class I and II MHC and T cell costimulatory
molecules (30, 31). Because these DCs are clustered around
activated T cells, it has been proposed that they are directly involved
in the generation of destructive autoimmune responses. Unlike RA
synovial tissue, RA synovial fluid (SF) is enriched in DCs expressing a
less mature phenotype, similar to that exhibited by
CD14dim DC precursors (30). When
isolated and cultured in vitro with DC growth factors (GM-CSF/IL-4),
these SF precursors do not proliferate, but mature into DCs expressing
high levels of MHC and T cell costimulatory molecules. It has been
speculated that immature DCs present in RA SF DCs are a source for DCs
involved in the DC-lymphocyte interactions of synovial tissue and that
suppressive factors such as TGF modulate DC maturity/function, while
DCs are contained in the SF space (29, 30).
Despite recent data indicating that myeloid DCs may be overrepresented in RA, little is known about the mechanisms promoting differentiation along specific DC pathways within distinct joint microenvironments or the association of specific DC subtypes with lymphocyte abnormalities, including abnormal Th1 responses. Moreover, even though mature DCs are likely major contributors to inflammatory responses in the RA joint, the potential contribution of early self-renewing or proliferating DC progenitors to joint pathology has not been described. Conceivably, the existence of progenitors with proliferative and DC-differentiating potential in the joint, along with factors that mediate DC growth and function, would greatly amplify local DC-driven events.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Twenty-four patients with RA diagnosed according to the 1987 revised criteria of the American College of Rheumatology were studied. All patients were receiving treatment at the time of sample collection. Most (>80%) of the patients were receiving disease-modifying agents such as gold, penicillamine, and methotrexate. A few were being treated with steroids and nonsteroidal antiinflammatory agents. None were receiving TNF antagonist therapy. Eight patients were male; mean age was 67 (r = 42–85). SF and peripheral blood (PB) samples were obtained as part of routine clinical care. Cell-free SF and serum were cleared of any precipitate by centrifugation (x 15 g for 15 min at 4°C). Six patients with osteoarthritis (OA) and ten normals (six females) were also included. The study was conducted according to Winthrop University Hospital institutional guidelines.
Enrichment and culture of precursors/progenitors
Because early self-renewing CD34negative progenitors for specific DC lineages are still not completely well characterized, we studied cell fractions known to be lacking in mature DCs, but enriched in DC precursors/progenitors (32). The strategy we used limited manipulation of the cells, which also diminished the problem of DC maturation initiated by excessive handling (33). Due to strong effects of endotoxins on myeloid cell/DC development and function, all phases of cell handling and culture were performed under low endotoxin conditions.
SF or PB were collected into sterile heparinized containers, diluted in
RPMI media, and layered on density gradients (lymphoprep endotoxin
poor; Nyegaard, Oslo, Norway) for the enrichment of mononuclear cells
(MNCs). For further separation of nonadherent progenitors, MNCs were
placed on nylon wool (NW) columns (32). The NW nonadherent
(NWX1) cells representing 
30% of MNCs are devoid of monocytes, B
cells, polymorphonuclear (PMN) cells, and mature DCs, and contain
precursor/progenitor cells and T and NK cells (55 and 15%,
respectively). The NWX1 cells were negative for nonspecific esterase
(<1% positive) and >95% viable, as determined by trypan blue dye
exclusion.
NWX1 cells were adjusted to 0.5–1 x 106
cells/ml, placed in RPMI 1640 medium (Life Technologies, Grand Island,
NY) with 2 mM L-glutamine, 10 mM HEPES, 50 IU/ml
penicillin, 50 µg/ml streptomycin, and 5% pooled normal human serum
(NHS/RPMI), and then incubated in either Teflon vials (Scientific
Specialties Service, Randallstown, MD) or 24-well plates at 37°C in a
5% CO2 humidified incubator. Cultures were
supplemented with various combinations of cytokines, including
cytokines known to sustain myeloid cell/DC development and to be
prevalent in the RA joint (GM-CSF/TNF/stem cell factor (SCF)/IL-13)
(34, 35, 36, 37). Although conflicting reports exist related to
IL-13 levels in RA (35, 38), we confirmed by ELISA
(Immunotech/Coulter, Hialeah, FL) that IL-13 is abundant in RA SF and
serum, compared with NHS, OA SF, and serum. After dose analysis, human
rSCF (Genzyme, Boston, MA) was utilized at 50 ng/ml, human rTNF-
(Knoll Pharmaceuticals, Whippany, NJ) at 500 U/ml, human rGM-CSF
(Genzyme) at 100 U/ml, human rIL-13 (PeproTech, Rocky Hill, NJ) at 10
ng/ml, human IL-1 (Genzyme) at 20 U/ml, human rM-CSF (Genzyme) at 50
ng/ml, and human rIL-4 (Genzyme) at 200 U/ml. To assess nonspecific
effects of T/NK cell activation, NWX1 cells were also cultured in
NHS/RPMI media containing 300 U/ml human rIL-2 (Hoffman-LaRoche,
Nutley, NJ).
Enrichment of neonatal cord blood progenitor cells
Cord blood was collected from healthy full-term infants into sterile heparinized containers during repeat caesarean sections, according to institutional guidelines. MNCs were prepared by density centrifugation on Lymphoprep gradients and placed on NW columns for the isolation of nonadherent cells (32). Further separation of CD34+ progenitor used positive immunoselection using immunomagnetic beads (Dynabeads; Dynal, Great Neck, NY), as previously detailed (32, 39). Myeloid DC hemopoiesis (both the CD14-derived and CD14-independent pathways) from these progenitors was instituted by adding GM-CSF (100 U/ml)/TNF (500 U/ml)/SCF (50 ng/ml) (GTS) (32). RA SF, RA serum, or OA SF were included at a final concentration of 10%, as indicated. The cytokine combinations used do not support the development of lymphocytes or erythrocytes (32, 39).
Myelodendritic progenitor cells
Myelodendritic cells, representing intermediate (CD34negativeCD33+ DR+CD115+) progenitors of the CD14-derived DC pathway in the cord blood model (13), were used to test the differentiating effects of RA SF in serum-free and cytokine-free RPMI media. These cells differentiate into either monocytes, granulocytes, or DCs when cultured with M-CSF, G-CSF, or CD14-DC growth factors (GM-CSF/IL-4 or GM-CSF/IL-13 ± SCF, ± TNF), respectively, and were obtained by treating GTS-instituted cord blood cultures on day 3 with 15 µg/ml rabbit polyclonal anti-TNF Ab (Genzyme), as previously described (13). After anti-TNF Ab treatment, cultures were supplemented with fresh NHS/RPMI media without exogenous cytokines on a weekly basis so as to maintain myelodendritic cells in a progenitor state (13). After 10–21 days, myelodendritic cells were removed from culture, centrifuged, and adjusted to 0.4 x 105 cells/ml in fresh serum-free RPMI 1640 medium containing 2 mM L-glutamine, 10 mM HEPES, 50 IU/ml penicillin, and 50 µg/ml streptomycin. Myeloid DC cytokines, 10% NHS, 10% RA SF, 10% RA serum, or 10% OA SF were included as indicated.
Proliferation
Proliferation was measured by the uptake of [3H]thymidine and by manual hemacytometer-assisted cell counts (Improved Neubauer). For thymidine uptake, 0.5 µCi [3H]thymidine (sp. act., 25 Ci/mmol; Amersham, Arlington, IL) was added to 100-µl aliquots taken from Teflon cultures and placed into 96-well microtiter plates. After 5 h, cells were harvested using an automated sample harvester and counted in a liquid scintillation counter. Results are expressed as the mean of triplicate samples; the SE was <25% in all experiments.
Immunofluorescence (IF) analysis
Abs to DR, CD13, CD33, and CD34 were obtained from Becton Dickinson (San Jose, CA); anti-CD1a was obtained from Biosource (Camarillo, CA); anti-CD14 was obtained from Sigma (St. Louis, MO); anti-CD86 was obtained from PharMingen (San Diego, CA); and anti-CD115 (M-CSF receptor) was obtained from Serotec (Oxford, U.K.). Dual label cytometric analysis was performed. In indirect assays, reactivity was detected with either FITC or PE-linked anti-mouse Igs or anti-rabbit Ig F(ab')2 (Boehringer Manneheim, Indianapolis, IN; Cappel, West Chester, PA; Jackson ImmunoResearch Laboratories, West Grove, PA). Matched nonimmune mouse or rabbit Ig were used as negative controls (Becton Dickinson; Coulter). Cells were analyzed on a FACScan flow cytometer (Becton Dickinson) calibrated with Calbrite beads (Becton Dickinson) for FITC and PE. The distribution of debris, dead cells, and any contaminating RBCs was assessed on the basis of forward (FSC) and right angle (side) scatter (SSC) before proceeding with the analysis. A total of 5,000–10,000 events was examined using a 488-nm wavelength excitation. Acquired events were analyzed using CellQuest Software (Becton Dickinson). Results are expressed as percentage of positive cells after subtracting negative control values.
Mixed leukocyte reaction (MLR)
Stimulator cells were removed from Teflon cultures, centrifuged in RPMI, adjusted to equal concentrations in 5% NHS/RPMI, and irradiated (60Co source, 2000 rad total). Varying numbers of stimulator cells were added to 96-well microtiter plates containing 5 x 104 responder cells/well (NW-enriched T cells obtained from normal PB). Proliferation was measured on days 6–7 by adding 0.5 µCi [3H]thymidine to each well and harvesting the cells, as described above. Control cultures containing irradiated stimulator or responder cells alone yielded <200 cpm.
T cell activation and measurement of Th1/Th2 responses
Putative DCs were irradiated as above, combined with allogeneic
T cells (NW nonadherent cells) at various DC:T cell ratios in 5%
NHS/RPMI, and incubated for 6 days in a humidified 5%
CO2 incubator. The cells were then resuspended in
fresh 5% NHS/RPMI media at 2 x 106
cells/ml. For T cell restimulation and the intracellular retention of
cytokines, cells were incubated in 25 ng/ml PMA (Sigma), 1 µg/ml
ionomycin (Sigma), and 5 µg/ml brefeldin A (Sigma) for 4 h in a
5% CO2 incubator (10, 40). Cells
were then incubated with anti-CD3 PerCP (Becton Dickinson),
permeabilized (FACS lysing and permeabilizing solutions; Becton
Dickinson), and stained with FITC anti-IFN- (Th1-restricted) and
PE anti-IL-4 (Th2-restricted) Abs (Becton Dickinson). Isotype
control samples received nonimmune mouse IgG1 labeled with PE
and FITC instead of anti-cytokine Abs. After staining, cells were
fixed in 10% buffered Formalin in PBS and stored in the dark until
analyzed. Approximately 15–30,000 cells were collected on the FACScan,
as indicated above. Calbrite beads (Becton Dickinson) were used to
calibrate the FACScan for FITC, PE, and PerCP. Region analysis was set
according to CD3 reactivity; positive values were determined according
to isotype controls.
TNFR ELISA
Soluble (s) TNFR I (p55) content was measured using a commercial sandwich ELISA with a sensitivity of 1 pg/ml (Genzyme; Predicta, ELISA human TNFRI). Experiments were performed exactly as recommended. In some instances, repeat determinations were performed with diluted samples. Results were detected using a microplate ELISA reader (Titerek) at 450 nm and are expressed as pg/ml sTNFR.
Blocking studies
Neutralization of sTNFR activity in RA SF was performed with TNF (500 U/ml) and mAb anti-human TNFR (p55) (2 µg/ml; Genzyme). Before being added to progenitor cells, RA SF was incubated for 45 min at room temperature with TNF, anti-sTNFR Ab, or control Abs nonimmune mouse IgG1).
Statistics
Student’s t test or the Mann-Whitney rank sum test was used to analyze data using Sigma Stat software (Jandel Scientific, San Rafael, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A higher proportion of RA SF MNCs than RA PB MNCs expressed
CD14dim, DR, CD86, and CD33 (Table I
, all regions), confirming that
relatively mature DCs are increased in RA SF vs RA PB (30, 31). Two subregions of RA SF and RA PB MNCs, one exhibiting
high SSC, the other exhibiting high FSC, were identified (Fig. 1A). The high SSC area was
comprised mostly of
CD33+CD13+cells, some
DR+CD86+ cells, and few
CD14dimCD1a+ cells (Table I). The lack of monocyte and DC markers in the high SSC area, together
with the known high SSC profile of PMN cells, indicated that the
majority of the high SSC area cells were
CD33+CD13+ PMN cells.
Wright stain analysis confirmed the presence (20%) of PMN cells in
the RA SF MNC subset. While most of cells in the high FSC subregion
were also CD33+CD13+, a
much larger proportion (>60%) were DCs, as indicated by the
expression of CD14 (dim), DR, and CD86.
|
|
) and in the removal of mature
cells, including PMN cells and DCs, residing in the high SSC area (Fig. 1A, right plots). The NW nonadherent SF FSC area
(Table I) was particularly enriched in a myelodendritic cell-like
progenitor compared with the SF MNC FSC area, as revealed by increases
in CD115 (3-fold, p = 0.01), decreases in CD14 and CD86
(p < 0.01), and DR, CD13, and CD33 reactivity
(13, 41). Neither the RA SF MNCs nor the RA SF nonadherent
cells expressed CD34, which is consistent with the lack of CD34 on
hemopoietic cells in the RA joint (30). As previously
shown by us (26), higher levels of
CD34+CD33+ myeloid
progenitors were detected in the RA PB nonadherent cells vs normal PB
nonadherent cells (data not shown). Thus, a very immature
(CD34+) myeloid progenitor is present in RA PB,
but not RA SF. As reported (32), passage of normal PB MNCs over NW produced mostly T and NK cells and decreases in CD14-, DR-, CD86-, and CD33-positive cells (p < 0.025, n = 3, data not shown). The high SSC region (containing CD33+CD13+ PMN cells) present in RA SF and RA PB MNCs was not present in normal PB MNCs (data not shown). Unlike RA SF nonadherent cells, normal PB nonadherent cells were not enriched in myelodendritic progenitor cells (<3% were DR, CD13, CD33, or CD115 positive).
Fig. 1B demonstrates that the APC potential of RA SF and PB
cell fractions was hierarchical, correlating with DC content. The
strongest MLR-stimulatory capacity was displayed by SF MNCs containing
the highest levels of DCs. Compared with RA SF MNCs, the MLR generated
by SF nonadherent progenitors was weak (p =
0.04), confirming that a large proportion of mature DCs had been
removed from the MNCs. Both RA PB MNCs and RA PB nonadherent cells
exhibited poor MLR-stimulatory potential. Nonetheless, unfractionated
PB MNCs exhibited higher stimulatory capacity than nonadherent (NWX1)
cells (Fig. 1B), further demonstrating the efficiency of the
separation strategy in removing mature APCs. As expected from the low
mature APC content in the samples (<10% and<1%
DR+CD86+ cells in MNC and
nonadherent cells, respectively), freshly prepared normal PB MNC cells
produced weak MLRs (data not shown) .
Lineage-specific growth of SF progenitors in response to cytokines
We evaluated the growth response of RA SF cell subsets to various
combinations of hemopoietically active myeloid cytokines. In support of
the existence of myeloid progenitor cells in RA SF, nonadherent cells
treated with myeloid DC cytokines produced marked proliferation. On day
7 (Fig. 2A), the level of
proliferation was higher in the nonadherent progenitor cell-enriched
fractions (NWX1) of RA SF than in total MNC fractions with either GTS
or GTS/IL-4 treatment (GTS MNC vs NWX1 = 1303 ± 621 vs
2980 ± 2278 cpm, respectively, and GTS/IL-4 MNC vs NWX1 =
3007 ± 1724 vs 4942 ± 2807 cpm, respectively,
n = 4–7). Within the progenitor cell group, six of
seven of the samples displayed higher levels of proliferation with
GTS/IL-4 vs GTS (n = 7, p = 0.023),
suggesting a preferential growth response to IL-4. In the remaining
sample, no change in growth occurred between treatment groups until
later time points. Temporal analysis of cell growth (Fig. 2B) demonstrated that proliferation persisted on days 14 and
18 in GTS/IL-4 cultures, but not in GTS cultures
(p = 0.04 for GTS vs GTS/IL-4 on day 18).
Hemacytometer-assisted cell counts of progenitor cell cultures revealed
that the total number of cells was higher in the GTS/IL-4 cultures than
in the GTS cultures, in further support of sustained growth with
GTS/IL-4 (data not shown). Because the growth of particular DC
progenitors/precursors (CD14-derived DCs) may be TNF independent at
certain developmental phases (13), cultures were also
established in the absence of TNF. Progenitor cells cultured with
GM-CSF/SCF/IL-4 displayed increased proliferation vs GTS
(p = 0.04, n = 3), indicating
that a growth response may be achieved independently of exogenous TNF
(data not shown).
|
An equivalent progenitor cell subset could not be demonstrated in OA SF. As reported (42), we noted that fewer MNCs were present in OA SF than in RA SF. Most of the OA SF MNCs were adherent macrophages, and few nonadherent cells (lymphocytes and precursor/progenitor cells) were present (data not shown). Because of the lack of nonadherent cells in OA SF, cultures could not be established from OA samples for comparison with RA SF.
Phenotypic and functional features of cultured SF progenitor cells
On day 0, nonadherent RA SF cells contained few
CD14dim cells (3% ± 1),
CD1a+ cells (4% ± 1) (Table I), and
CD14+CD1a+ cells (1% ±
0.5), which are transient intermediates of the CD14-derived DC pathway
developing from CD14dim cells. Thus, fresh RA SF
nonadherent cells were not enriched in late CD14-derived DC precursors.
After 8–12 days in IL-4, the proportion of total
CD1a+ cells increased (12% ± 3), and transient
CD14+CD1a+ cells were noted
in some cultures (Fig. 3A,
middle plot). Without the addition of CD14-derived
DC-restricted cytokines (GTS alone), fewer (6% ± 1)
CD1a+ cells were observed (Fig. 3A,
right plot). Fig. 3B depicts typical DC clusters
developing from RA SF progenitors cultured in myeloid DC cytokines
(GTS ± IL-4 or IL-13).
|
C).
Treatment of progenitors with GTS/IL-13 (or GTS/IL-4) resulted in the
expansion of myeloid DCs residing in the high FSC area (Fig. 3C, right plot). These cultures contained more
(>2x, p < 0.02) cells falling in the high FSC DC
area than freshly isolated RA SF nonadherent cells (Fig. 3C,
middle plot), denoting the extensive potential of the RA SF
progenitors to yield DC progeny.
We assessed the MLR- and Th1/Th2-stimulatory capacity of mature DCs
developing from the RA SF progenitor cells (Fig. 4). Culture with GTS +/- IL-4 (Fig. 4A) produced progeny with potent T cell-stimulatory capacity
in the MLR. In 67% of paired comparisons (six of nine), the T cell
response was higher (1.3-fold, p = 0.04) when cells
were cultured with GTS/IL-4 vs GTS. In the remaining 33% of the
samples, no difference in the MLR occurred between treatment groups.
IL-13, which is much more abundant than IL-4 in the RA joint
(35), produced results similar to IL-4 when combined with
GTS (75% of the samples exhibited increased MLRs vs GTS, Fig. 4B). Culture of RA SF progenitors with yet other myeloid
cytokines prevalent in the RA joint (IL-1) also produced progeny with
MLR-stimulatory potential (Fig. 4C) (GM-CSF/SCF/IL-1 vs GTS,
p > 0.05).
|
A). Because IL-2
would further activate and expand contaminating T cells present in the
progenitor cell-enriched fractions, the lack of stimulatory capacity of
the IL-2-treated cultures argues against RA SF T cells acting as APCs
in our system.
In Fig. 4D, we compared the ability of fresh and cultured SF
progenitor cells to elicit Th1/Th2/Th0 responses during the stimulation
of naive T cells in the allogeneic MLR. Fresh RA SF progenitor cells
(Fig. 4, middle plot) produced only weak Th1 (IFN-), Th2
(IL-4), and Th0 (IFN-/IL-4) responses. This was in contrast to fresh
RA SF MNCs (Fig. 4, left plot), which preferentially
produced strong Th1-type responses (NWX1 vs MNC, p =
0.004). Growth of progenitors with GTS/IL-13 (Fig. 4, right
plot) yielded increases in the number of Th1-type cells over day 0
(p = 0.04) and a similar Th1 profile as that
promoted by RA SF MNCs on day 0 (Fig. 4D). In all samples
studied (n = 5), GTS/IL-13-treated cells lacked
Th2-activating potential. Cells cultured in GTS and GTS/IL-4 also
produced strong Th1 responses (data not shown). Thus, DC progenitors
present in RA SF mature, after in vitro culture with DC cytokines
prevalent in the inflamed RA joint (GTS/IL-13), into potent APCs that
preferentially activate inflammatory-type Th1 responses over Th2 and
Th0 responses.
RA SF skews the development of CD34+ progenitors toward the CD14-derived DC pathway
In the remaining sections, we describe the impact of the RA
microenvironment on the in vitro growth of DC progenitors. GTS-driven
cord blood CD34+ progenitors were utilized as a
model for DC growth. DC hemopoiesis was instituted from these
progenitors with GTS and 10% NHS (39), or with GTS and
either 10% RA SF, 10% RA serum, or 10% OA SF. A subgroup of the RA
SF samples tested (five of eight) produced higher levels of
proliferation around day 7, compared with NHS
(p = 0.013, Fig. 5A). With three of the eight
RA SF samples, proliferation was not increased over NHS
(p > 0.05). With RA serum, trends toward
increased proliferation were noted, but differences were not
significant.
|
B,
p < 0.04). Increases in
DR+CD86+ cells occurred
with four of five RA SF and five of six RA serum samples tested vs NHS
(Fig. 6, p < 0.04).
Thus, the higher MLR potential displayed by the GTS/RA SF and GTS/RA
serum cultures was related to increases in mature DCs expressing class
II MHC and T cell costimulatory molecules. The addition of RA SF or RA
serum to GTS cultures also produced increases in transient
CD14+CD1a+ cells vs NHS
(n = 11, p < 0.03, Fig. 6) and
increases in CD80+CD11b+
(data not shown), further indicating the expansion of the CD14-derived
DC pathway (1). Culture with GTS/OA SF did not
yield increases in APCs over GTS/NHS and yielded a predominance of
monocytes/macrophages with lower MLR-stimulatory capacity (Fig. 5C). IF analysis revealed few (<1%)
CD14+CD1a+ cells and mature
DCs in the GTS/OA SF cultures.
|
The expansion of the CD14-derived DC pathway promoted by RA SF was
analogous to that previously described with an anti-TNF Ab
(13). Naturally occurring TNF antagonists are presumed to
down-regulate normal TNF-mediated responses. In chronic inflammation,
however, it has been suggested that increases in TNF antagonists such
as sTNFRs contribute to the disregulation of inflammatory responses via
an unknown mechanism (43). We associated high levels of
sTNFRs (p55) with the ability of RA SF to enhance the development of
the CD14-derived DC pathway from cord blood progenitors. An ELISA (Fig. 7) showed that p55 sTNFRs were
significantly (p = 0.0027) increased in the RA
SF samples (n = 10) reported in this work compared with
NHS samples (n = 5), which is consistent with previous
reports (44, 45). Pretreatment of RA SF with either Abs to
sTNFRs (p55) or excess TNF ligand inhibited the development of
CD14-derived DCs (Fig. 7). Thus, while untreated RA SF increased the
number of CD14+CD1a+
intermediates of the CD14 pathway, this ability was lost when the
binding capacity of sTNFRs in RA SF was blocked. Pretreatment with
isotypic Ab controls did not produce these effects, further supporting
that sTNFRs were specifically neutralizing TNF bioactivity.
|
In the experiments depicted in
Figs. 5–7, RA SF augmented the
development of the CD14-derived DC pathway in cultures that were
initially instituted from CD34+ progenitors with
DC cytokines (GTS). Thus, SF appears to enhance the effects of the DC
cytokines used. Hemopoiesis from CD34+
progenitors did not proceed with RA SF alone (except in one instance),
suggesting the absence of essential differentiation factors, decreased
distribution of preferred target cells, and/or the presence of
inhibitory factors. We tested these possibilities by studying the
differentiating effects of RA SF alone (i.e., no exogenous cytokines)
on intermediate (CD34negative) myelodendritic
progenitor cells of the CD14 DC pathway (13). Phase
microscopy revealed that myelodendritic cells placed in RA SF/RPMI
media developed typical branched and veiled DC morphology (Fig. 8B). In contrast, cells placed
in NHS/RPMI media (Fig. 8A) remained mostly undifferentiated
(as expected) (13), and cells placed in OA SF/RPMI yielded
scant progeny (Fig. 8C). Of the few viable cells that had
developed in OA SF, most resembled monocytes/macrophages and a minority
(<1%) were DCs (Fig. 8C). Dual label IF analysis showed
that myelodendritic cells placed in RA SF alone contained an increased
number of CD14+CD1a+
intermediates and DR+CD86+
cells, compared with cultures containing NHS alone
(p 
0.05, Fig. 9A), which is consistent with
the development of CD14-derived DCs (6, 7, 13). DCs
generated from myelodendritic cells in the presence of RA SF or RA
serum alone stimulated potent MLRs (Fig. 9B,
p < 0.02 vs NHS), whereas cells obtained from
myelodendritic cell cultures maintained in OA SF alone produced weak
MLRs. Thus, the RA microenvironment is capable of sustaining the in
vitro growth of early myeloid progenitor cells committed to DC
development.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We used a comparative strategy using light scatter subregion
analysis coupled with cell surface IF analysis to assess the
distribution of DCs and DC progenitors/precursors in RA SF cell
subsets (Fig. 1A, Table I). A high FSC region in RA SF
MNCs contained a large proportion of
CD33+CD13+DR+CD86+CD14+CD1a+
cells, which is consistent with a myeloid DC (immature and mature)
phenotype (30, 41). Passage of SF MNCs over NW produced a
nonadherent cell subset that lacked mature myeloid DCs
(DR+CD86+ cells) and
myeloid DC precursors, and that was enriched in myeloid progenitors
(CD13+CD33+DR+CD86negativeCD115+CD14+/-CD1anegative
cells).
The degree of MLR activity generated by fresh RA SF and PB cell subsets
correlated with mature DC content (Fig. 1, Table I). RA SF MNCs
contained higher levels of mature DCs than RA PB MNCs and were potent
stimulators of an MLR, which is in agreement with prior studies
(30, 31). In addition to stimulating a potent MLR, RA SF
MNCs stimulated Th1-inflammatory responses over Th2/Th0 responses (Fig. 4D). In contrast, RA PB MNCs were weak stimulators of either
Th1/Th2 or Th0 responses (data not shown). The polarization toward Th1
responses induced by freshly isolated RA SF DCs, but not by RA PB
cells, provides new evidence that DCs are instructed within the joint
to acquire functions associated with the selective activation of
inflammatory T cells.
The fresh RA SF progenitor cells displayed poor MLR- and
Th1/Th2-stimulatory potential. A proliferative response, potent MLR-
and Th1-stimulatory capacity, and large increases in the number of
cells residing in the high FSC DC areas were produced only after in
vitro culture of these cells with myeloid DC growth factors (
Figs. 2–4). The ability of RA SF progenitor cell-derived DCs obtained with
DC growth factors to selectively promote Th1-type responses resembled
that of mature DCs present in fresh RA SF MNCs (Fig. 4D).
Because inflammatory-type Th1 responses have been associated with
immunopathology in RA (48, 49, 50), it is conceivable that
mature myeloid DCs arising from RA SF progenitors in situ would
activate pathogenic Th1 responses (Fig. 10). Besides driving Th1-type
responses, myeloid DCs might mediate other immune responses within the
RA joint, such as Ig (rheumatoid factor) synthesis by B cells
and the production of IFN- by CD8+ T cells
(6, 7, 51).
|
A reservoir of myeloid DC progenitors in RA SF could provide a pool of
mature DCs exhibiting T cell-activating potential, and could also be a
plentiful source of immature DCs in the joint. In the context of the
RA-inflammatory environment, the ability of immature DCs to acquire Ag
from autologous apoptotic and/or necrotic cells for subsequent
cross-priming of T cells might be favored (Fig. 10)
(52, 53, 54). Such events have been linked to the autoimmune
process in lupus erythematosus and autoimmune diabetes
(52, 53, 54) and could perpetuate autoimmune responses, even
in the absence of the original instigating Ag. In theory, the process
of acquiring Ags from apoptotic/necrotic cells in the RA SF space
(55) by immature DCs would be facilitated by CD36,
v3, and
v5 (53, 56), and would prompt maturation. Further maturation would be
favored by an excess of myeloid DC growth factors and IL-1 found in RA
SF (35, 36, 56). Since immature DCs are incapable of Ag
presentation and might directly activate regulatory T cells that
suppress self-reactivity (11, 12), we propose that the
intraarticular regulation of DC maturation is a critical control point
in the RA disease process.
RA SF facilitated the maturation of DCs from myeloid progenitors,
providing direct evidence that the inflamed RA environment instructs DC
differentiation (
Figs. 5–9). While the addition of RA SF to GTS-driven
CD34+ progenitors enhanced the effects of the
cytokines used (
Figs. 5–7), RA SF alone (in the absence of exogenous
growth factors) did not generally induce DC growth from
CD34+ progenitors. In contrast, RA SF directly
promoted the differentiation of myeloid DCs from
CD34negativeCD33+
progenitors (intermediate myelodendritic cells,
Figs. 8–9), indicating
that the factors present in SF were targeting specific
progenitors/precursors that differentiate immediately upon entry into
the inflamed joint (Fig. 10). Because GM-CSF, TNF, IL-13, and SCF are
abundant in the RA joint (34, 35, 36, 37), they most likely
facilitate the local growth of myeloid DCs, especially CD14-derived
DCs. IL-4, which has been consistently shown to be present at low
levels in the RA joint (36), is unlikely to be a major
participant in this process. Mature DCs derived from progenitor cells
in vitro with myeloid DC growth factors and IFN- exhibit a greater
capacity to activate IFN--producing T cells (Th1-type cells) than
cells cultured with myeloid DC growth factors alone (57).
Thus, IFN- existing in RA SF (58, 59) may also skew DC
development and function toward a Th1 response. Other factors known to
promote CD14-DC growth and that are prevalent in RA SF, such as
hyaluronan and fibronectin (1, 2, 60, 61), may contribute
to the formation of a Th1 response as well.
The role of TNF and TNF antagonists in the regulation of myeloid DC
hemopoiesis is complex and varies with stages of DC development
(5). While TNF is required to ensure DC hemopoiesis from
CD34+ progenitors, the addition of polyclonal
anti-TNF Ab to GTS-treated progenitor cells after 3 days of culture
produces amplification of the CD14-derived pathway (13).
In the present study, we noted that the effects of RA serum/SF on the
myeloid DC pathway were analogous to those achieved with polyclonal
anti-TNF Ab. That is, RA serum/SF altered DC developmental
responses to yield increases in
CD14+CD1a+ intermediates
and mature CD14-derived DCs with strong MLR potential. The capacity of
RA SF to induce the development of CD14-derived DCs correlated with
high levels of sTNFRs (p55) (Fig. 7), which share with anti-TNF Abs
the ability to bind TNF and inhibit TNF activity. Thus, within the RA
joint, it is possible that sTNFRs (p55) also regulate the growth of
CD14-derived DCs. This suggests that TNF blockade, including
therapeutic TNF inhibition, may have important suppressive and
stimulatory effects on DC activity in RA.
Notwithstanding the positive effects of the RA extracellular
environment on DC development, factors present in RA SF such as TGF
may negatively regulate DC-mediated functions, both at the level of
immature DCs (late DC precursors) and during DC-T cell interactions
(29, 30). We noted that the substitution of TGF for
IL-13 or IL-4 (GM-CSF/SCF/TGF) in RA SF progenitor cell cultures
appeared to interfere with DC maturation and yielded cells with a
diminished potential to induce Th1 responses (40% decreases vs
IL-13 and IL-4; data not shown). Because TGF is the preferred growth
factor for CD14-independent DCs (Langerhans cell type), we cannot
discount the possibility that the reduced Th1 response was related to a
lack of TGF-responsive progenitor/precursor cells.
The inability to obtain nonadherent progenitor cells from OA SF is consistent with differences in the distribution of MNC types in OA SF compared with RA SF (42). Reduced levels of sTNFRs (p55) and proinflammatory cytokines (TNF, IL-1, IL-13) in OA SF vs RA SF (44, 45) further imply that OA SF lacks factors that would instruct the maturation of myeloid (CD14) DCs. Our findings do not exclude the possibility that other DC subtypes with distinct effector functions are amplified in RA.
In summary, we provide new insight into how selection of specific DC
progenitor cells by the RA synovial microenvironment may play a central
role in the RA disease process (Fig. 10). The final outcome of
DC-driven events in the various joint microenvironments may ultimately
be determined by the proportion of factors promoting and inhibiting the
maturation and function of particular members of the DC lineage system,
and by the nature of factors regulating the trafficking of DCs into the
synovial space. By understanding the precise mechanisms regulating
these events, it may ultimately be possible to provide some basis for
the heterogeneity of the immune response in RA and to devise highly
directed therapeutic strategies aimed at either blocking and/or skewing
the differentiation and/or functions of particular DC subtypes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Frances Santiago-Schwarz, Director Cellular Immunology Research, Division of Rheumatology, Winthrop University Hospital, 222 Station Plaza North, Suite 430, Mineola, NY 11501. E-mail address: fschwarz{at}winthrop.org
3 Abbreviations used in this paper: DC, dendritic cell; FSC, forward angle light scatter; GTS, GM-CSF/TNF/SCF; IF, immunofluorescence; MNC, mononuclear cell; NHS, normal human serum; NW, nylon wool; OA, osteoarthritis; PB, peripheral blood; PMN, polymorphonuclear; RA, rheumatoid arthritis; SCF, stem cell factor; SF, synovial fluid; SSC, right angle (side) light scatter; sTNFR, soluble TNFR.
Received for publication September 14, 2000. Accepted for publication May 18, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
inhibits dendritic cell-T lymphocyte interactions in patients with chronic arthritis. Arthritis Rheum. 42:507.[Medline]
is highly expressed in synovium of rheumatoid arthritis compared with seronegative spondyloarthropathies. Ann. Rheum. Dis. 59:263.-cell apoptosis: a trigger for autoimmune diabetes?. Diabetes 49:1.[Abstract]
2-glycoprotein I antibodies. Arthritis Rheum. 42:1412.[Medline]
This article has been cited by other articles:
|
|
 |
|
  C. A. Martin, D. L. Kurkowski, A. M. Valentino, and F. Santiago-Schwarz Increased Intracellular, Cell Surface, and Secreted Inducible Heat Shock Protein 70 Responses Are Triggered during the Monocyte to Dendritic Cell (DC) Transition by Cytokines Independently of Heat Stress and Infection and May Positively Regulate DC Growth J. Immunol., July 1, 2009; 183(1): 388 - 399. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Lakey, T. G. Morgan, A. D. Rowan, J. D. Isaacs, T. E. Cawston, and C. M. U. Hilkens A novel paradigm for dendritic cells as effectors of cartilage destruction Rheumatology, May 1, 2009; 48(5): 502 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L Salazar, O Aravena, P Abello, A Escobar, J Contreras-Levicoy, N Rojas-Colonelli, D Catalan, A Aguirre, R Zuniga, B Pesce, et al. Modulation of established murine collagen-induced arthritis by a single inoculation of short-term lipopolysaccharide-stimulated dendritic cells Ann Rheum Dis, September 1, 2008; 67(9): 1235 - 1241. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Chang, S. A. Mihalcik, and D. F. Jelinek B Lymphocyte Stimulator Regulates Adaptive Immune Responses by Directly Promoting Dendritic Cell Maturation J. Immunol., June 1, 2008; 180(11): 7394 - 7403. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E Smolewska, J Stanczyk, H Brozik, M Biernacka-Zielinska, B Cebula, T Robak, and P Smolewski Distribution and clinical significance of blood dendritic cells in children with juvenile idiopathic arthritis Ann Rheum Dis, June 1, 2008; 67(6): 762 - 768. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. L. Zhang, P. Colmenero, U. Purath, C. Teixeira de Matos, W. Hueber, L. Klareskog, I. H. Tarner, E. G. Engleman, and K. Soderstrom Natural killer cells trigger differentiation of monocytes into dendritic cells Blood, October 1, 2007; 110(7): 2484 - 2493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Zhu, F. Ji, Y. Wang, Y. Zhang, Q. Liu, J. Z. Zhang, K. Matsushima, Q. Cao, and Y. Zhang Synovial Autoreactive T Cells in Rheumatoid Arthritis Resist IDO-Mediated Inhibition J. Immunol., December 1, 2006; 177(11): 8226 - 8233. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. B. Joosten, M. G. Netea, S.-H. Kim, D.-Y. Yoon, B. Oppers-Walgreen, T. R. D. Radstake, P. Barrera, F. A. J. van de Loo, C. A. Dinarello, and W. B. van den Berg IL-32, a proinflammatory cytokine in rheumatoid arthritis PNAS, February 28, 2006; 103(9): 3298 - 3303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T R D J Radstake, A W T van Lieshout, P L C M van Riel, W B van den Berg, and G J Adema Dendritic cells, Fc{gamma} receptors, and Toll-like receptors: potential allies in the battle against rheumatoid arthritis Ann Rheum Dis, November 1, 2005; 64(11): 1532 - 1538. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Moghaddami, G. Mayrhofer, and L. G. Cleland MHC class II compartment, endocytosis and phagocytic activity of macrophages and putative dendritic cells isolated from normal tissues rich in synovium Int. Immunol., August 1, 2005; 17(8): 1117 - 1130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Moghaddami, L. G. Cleland, and G. Mayrhofer MHC II+ CD45+ cells from synovium-rich tissues of normal rats: phenotype, comparison with macrophage and dendritic cell lineages and differentiation into mature dendritic cells in vitro Int. Immunol., August 1, 2005; 17(8): 1103 - 1115. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T R D J Radstake, R van der Voort, M ten Brummelhuis, M de Waal Malefijt, M Looman, C G Figdor, W B van den Berg, P Barrera, and G J Adema Increased expression of CCL18, CCL19, and CCL17 by dendritic cells from patients with rheumatoid arthritis, and regulation by Fc gamma receptors Ann Rheum Dis, March 1, 2005; 64(3): 359 - 367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rivollier, M. Mazzorana, J. Tebib, M. Piperno, T. Aitsiselmi, C. Rabourdin-Combe, P. Jurdic, and C. Servet-Delprat Immature dendritic cell transdifferentiation into osteoclasts: a novel pathway sustained by the rheumatoid arthritis microenvironment Blood, December 15, 2004; 104(13): 4029 - 4037. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Cloosen, M. Thio, A. Vanclee, E. B. M. van Leeuwen, B. L. M. G. Senden-Gijsbers, E. B. H. Oving, W. T. V. Germeraad, and G. M. J. Bos Mucin-1 is expressed on dendritic cells, both in vitro and in vivo Int. Immunol., November 1, 2004; 16(11): 1561 - 1571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Lande, E. Giacomini, B. Serafini, B. Rosicarelli, G. D. Sebastiani, G. Minisola, U. Tarantino, V. Riccieri, G. Valesini, and E. M. Coccia Characterization and Recruitment of Plasmacytoid Dendritic Cells in Synovial Fluid and Tissue of Patients with Chronic Inflammatory Arthritis J. Immunol., August 15, 2004; 173(4): 2815 - 2824. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Martin, S. E. Carsons, R. Kowalewski, D. Bernstein, M. Valentino, and F. Santiago-Schwarz Aberrant Extracellular and Dendritic Cell (DC) Surface Expression of Heat Shock Protein (hsp)70 in the Rheumatoid Joint: Possible Mechanisms of hsp/DC-Mediated Cross-Priming J. Immunol., December 1, 2003; 171(11): 5736 - 5742. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
