Get documents about "Thl and Th2 T-Helper Cells Exert Opposite Regulatory Effects on
Document Sample
Thl and Th2 T-Helper Cells Exert Opposite Regulatory Effects on
Procoagulant Activity and Tissue Factor Production by Human Monocytes
By Gianfranco Del Prete, Marco De Carli, Regina M. Lammel, Mario M. D’Elios, Kenn C. Daniel,
Betti Giusti, Rosanna Abbate, and Sergio Romagnani
of
The role of T-cell subsets in the induction tissue factor (TF) addition of fixed Th2 cells and T h l supernatants or IFN-y
production by human monocytes in vitro was investigated. inducednoticeable TF production. The addition ofeither
Mitogen stimulation enabled both unfractionated T cells and anti-IFN-y antibody or Th2 supernatants t o monocytes stim-
their CD4+ or CD8+subsets t o promote procoagulant activity ulated with activated andfixed T h l cells plus theirsuperna-
(PCA). After mitogen or antigen activation, all seven T-cell tant resulted in a dose-dependent inhibition of TF synthesis,
clones with T h l cytokine profile, but none of seven Th2 which was partially restored by neutralization of interleukin-
clones, induced TF production and PCA. T-cell blasts from 4 (IL-4) or IL-10. Addition of recombinant IL-4, IL-13, or IL-
four T h l activated clones were fixed with paraformaldehyde TF
10, but not IL-5, inhibited the Thl-induced production by
and added t o monocytes in the presence of medium alone monocytes. Data indicate that bothCD8’ and CD4+ Thl, but
or their supernatants. Addition of either fixed T h l cells or not Th2, T cells can help TF production and PCA. Both cell-
their supernatants inducedlow TF production (0.2 t o 0.6 ng/ to-cell contact with activated T cells and Thl-type cytokines,
mL), whereas addition of both resulted in much higher TF in particular IFN-y, are required for optimal TF synthesis,
synthesis (1.8 t o 3.4 ng/mL). Among Thl-type cytokines, whereas Th2-derived cytokines (IL-4,IL-13, and IL-IO) are
only interferon-y (IFN-y) induced minimal TF production (0.1 inhibitory. This may be ofpotential interestfor future thera-
t o 0.4 ng/mL). No TF synthesis was induced by activated and peutic strategies.
fixed Th2 cells and/or their supernatants, whereas combined 0 1995 by The American Society of Hematology.
F IBRIN ACCUMULATION is a prominent and consis-
tent feature of the classic cell-mediated, delayed-type
hypersensitivity responses.’ In immune-mediated reactions,
sults from collaborativeinteractionsbetweenT
monocytes.“’.” In the murine system, two main parallel and
independent models of cellular collaboration between T cells
cellsand
initiation of coagulation is due to the procoagulant activity and monocytes for T-cell induction and regulation of mono-
(PCA) expressed by cells of the macrophage/monocyte lin- cyte PCA response have been suggested.” One pathway is
eage, which synthesizeandexpress tissuefactor (TF) on mediated by a specific cytokine termed monocyteprocoagu-
their membrane in response to a variety of in vitro stimuli, lant inducing factor, which is produced by both CD4’ and
such as mitogens, bacterial lipopolysaccharide, antigedanti- CD8’ clones after antigen stimulation, whereas the second
body complexes, products of complement activation,’ pro- pathway is mediated by contact interactions between T cells
tein, and allogeneic antigen^.^.^ The expression of cell-medi- and monocytes.’’
PCA
ated been
has implicated in the activation of In the last few years, it has been shown that both murine
intravascular and extravascular coagulation in malignancy,’ and humanCD4’ T helper (Th) cells represent a functionally
endotoxaemia,” and immunopathology, such as rejection of heterogeneous population,including at least three distinct
renal allografts’ or Crohn’s disease.8 Evidence has recently Th cell subsets termed Thl, Th2, and Tho cells on basis the
been provided that increased thrombin formation and trig- Thl
of their profile of cytokine ~ e c r e t i o n . ’ ~ . ‘ ~ cells produce
geringof theacutethromboticevents inunstable angina interleukin-2 (IL-2), interferon-y (IFN-y), and tumor necro-
can result from lymphocyte-inducedactivation of circulating sis factor-p (TNF-0) and promote both macrophage activa-
monocytes, which express high procoagulant TF-like activ- tion resulting in delayed-type hypersensitivity and produc-
ity.’ tion of complement-fixing and opsonizing antibodies. Th2
Induction of monocyte TF synthesis and related PCA re- cells secrete IL-4, IL-5, IL-6, IL-10, and IL-13, provide opti-
mal help for antibody production, and promote bothmast
cell growth and eosinophil differentiation and activation. In
From the Division of Clinical Immunology and Allergology, Isti- the absence of polarizing conditions leading to stereotyped
tuto diClinica Medica 3, and Istituto di Clinica Medica 1, University Thl or Th2 patterns, CD4’ T cells secreting both Thl- and
of Florence, Florence, Italy. Th2-typecytokines(Tho cells)usually arise and mediate
Submitted September 3, 1994; accepted February 16, 1995.
intermediate effects,depending on the ratio of cytokines
Supported in part by a grant from the Consiglio Nazionale delle
Ricerche (projects “Clinical Applicationof Research in Oncology” produced and the nature of responding cells.’’,’‘ Clonal anal-
94.01198.PF39 and “FATMA” 94.00637.PF41), a grant from the ysis of murine CD4’ Th cells has revealed that the ability
Associazione Italiana per la Ricerca sul Cancro, and a grant ,from to produce monocyte procoagulantinducing factor and to
the European Community Biotechnology Network (Transplantation) drive monocyte procoagulant response is a peculiar function
Project No. PL920300. of both Thl and Tho, but not of Th2, clone^.'^
Address reprint requests to Gianfranco Del Prete, MD, Istituto In the present study, two questions have been addressed:
di Clinica Medica 3, V i d e Morgagni 85, 1-50134 Florence, Italy. ( 1 ) which of the human T-cell subsets is primarily involved
The publication costs of this article were defrayed in part by page in inducing monocyte T F activity, and (2) which of the T
charge payment. This article must therefore be hereby marked
“advertisement” in accordance with 18 U.S.C. section 1734 sole1.y to cell-derived cytokines is involved in either induction or inhi-
indicate this fact. bition of such aT cell-dependent monocyte effectorfunction.
0 1995 by The American Society of Hematology. Results show that T cells producing Thl-type cytokines are
0006-497//95/8601-0005$3.00/0 responsible for induction of TF synthesis by monocytes and
250 Blood, Vol 86,N o 1 (July I ) , 1995: pp 250-257
HUMAN Thl/Th2 CELLS AND MONOCYTE TF SYNTHESIS 25 1
that a combination of contact-mediated signal(s) and IFN-7 tometry, and their antigen-specificity was assessed by measuring 'H-
is required for induction of optimal TF production, whereas TdR uptake after 60-hour stimulation with the appropriate antigen,
IL-4, IL-13, and &l0 exert a negative regulatory effect. as rep~rted.'~
Characterization o the cytokine projle o T-cell clones. To in-
f f
duce cytokine production by T-cell clones in response to their spe-
MATERIALS AND METHODS cific antigens, lo6 T-cell blasts from each clone were cocultured in
Reagents. Purified protein derivative (PPD) of Mycobacterium 1 mL complete medium with 5 X 10' irradiated autologous non-T
tuberculosis was obtained from Istituto Sieroterapico e Vaccinogeno cells (as antigen-presenting cells) andthe specific antigen (2 pgl
Sclavo (Siena, Italy), and 70-kD heat shock protein (70-HSP) was mL).I4 After 48 hours, culture supernatants were collected, filtered,
purchased from StressGen Biotechnologies CO (Victoria, BC, Can- and stored in aliquots at -70°C until use. To induce cytokine produc-
ada). Purified extract of the house dust mite Dermutophagoides pter- tion in response to nonspecific activation, T-cell blasts were resus-
onyssinus (DP) and the excretory/secretory antigen of Toxocara pended at 106/mLcomplete medium and cultured in the presence of
canis (TES) were provided by Dr P. Falagiani (Lofarma Allergeni, PHA. After 36 hours, culture supernatants were collected, filtered,
Milan, Italy). Phytohemagglutinin (PHA) was purchased from Gibco and stored in aliquots at -70°C until use. Supernatants were assayed
Laboratories (Grand Island, NY). OKT3 (anti-CD3), OKT4 (anti- for IL-2, IFN-y, TNF-P, IL-4, and IL-5 content. The quantitative
CD4), OKT8 (anti-CDg), and O M 1 4 (anti-CD14) monoclonal anti- determinations of IFN-y, TNF-P, and IL-4 were performed by com-
bodies (MoAbs) were purchased from Ortho Pharmaceutical Corp mercial assays (Centocor Inc, Malvern, PA [IFN-y]; and Quantikine,
(Raritan, NJ). Goat antimouse IgG-coated beads (Dynabeads) were R&D Systems, Minneapolis, MN [TNF-P, IL-41. For the measure-
purchased from Dynal AS (Oslo, Norway). Recombinant human IL- ment of IL-2 and IL-5, the murine CTLL-2 and LyH7.BI3 cell
2 (rIL-2) was a gift from Eurocetus (Milan, Italy). Human rIL-4 was lines were used as source of indicator cells, respectively, as detailed
a gift from Dr J. Banchereau (Dardilly, France), and human rIL-l0 e1~ewhere.I~ According to their cytokine secretion profile, seven T-
was a gift from Dr M. Howard (DNAX, Palo Alto, CA). Recombi- cell clones usedin this study expressed the Thl phenotype, and
nant IL-5 was purchased from Amgen Biologicals (Thousand Oaks, seven were Th2. Thl clones included three PPD-specific (P.2, P.8,
CA); rIL-13 was purchased from PeproTech Inc (Rocky Hill, NJ). and P.16) and four 70-HSP-specific (HS.18, HS.34, HS.36, and
Human rIFN-y, TNF-P, anti-IFN-y MoAb, and anti-TNF-P and HS.43) clones, whereas the Th2 clones included three DP-specific
anti-IL-2 polyclonal rabbit antibodies were purchased from Gen- (DP.6, DP.14, and DP.60) and four TES-specific (TC.4, TC.19,
zyme CO (Boston, MA). The anti-TNF-a MoAb (B154.2.1) was a TC.39, and TC.41) clones.
gift from Dr G.Trinchieri (The Wistar Institute, Philadelphia, PA), Culture conditions for generation of monocyte PCA and TF pro-
and the anti-TNF-aR MoAbs Utr-l and Htr-9, reactive with type duction. Unfractionated, CM-enriched and CD8-enriched T cells
A and type B TNF-a receptors, respectively, were provided by Dr ( 106/mL) were stimulated for 48 hours with PHA (1% vol/vol) in
M. Brockhaus (Hoffman-La Roche AG, Basel, Switzerland). Neu- the presence of irradiated autologous non-T cells (104/mL). After
tralizing anti-IL-4, anti-IL-5, and anti-IL-l0 polyclonal goat anti- three washings, 8 X IO5 activated T cells were cocultured for 8
bodies were purchased from R&D Systems (Minneapolis, MN). All hours with 4 X 10' autologous or allogeneic monocytes in 1 mL
reagents were negative for endotoxin contamination at the level of complete medium. Likewise, T-cell blasts of each clone were washed
0.005 nglmL (limulus amoebocyte lysate assay; E-Toxate, Sigma three times, counted, resuspended in complete medium (106/mL),
Chemical CO, St Louis, MO). and stimulated for 24 hours with PHA in the presence of irradiated
Cell preparations. Peripheral blood mononuclear cells, obtained autologous non-T cells (104/mL) as feeder cells. PHA-stimulated
from the citrated peripheral blood of healthy volunteers, were pre- cloned T cells were fixed with 0.1% paraformaldehyde (Sigma
pared by density gradient centrifugation (Ficoll-Hypaque), exten- Chemical CO; 1 mL for 3 X lo6 T-cell blasts for 90 seconds),
sively washed, counted, and resuspended inRPM1 1640 medium extensively washed, and cultured for an additional 24 hours in me-
supplemented with 10% heat-inactivated fetal calf serum (Hyclone dium alone, as reported elsewhere." After washings, 8 X 10' fixed
Laboratories Inc, Logan, UT) (complete medium). After incubation clonal T cells were cocultured for 8 hours with 4 X 1 6 autologous
on plastic for 1 hour at 3 7 T , nonadherent cells were removed by or allogeneic monocytes in 1 mL complete medium. In some experi-
repeated washings, and adherent cells were recovered by scraping. ments, T-cell clones (8 X 105/mL) were stimulated for 16 hours
Nonadherent cells were then fractionated into T and non-T cells with the specific antigen (5 &mL) in the presence of autologous
by rosetting with neuraminidase-treated sheep red blood cells, as monocytes (4 X l@/mL) as antigen-presenting cells and effector
detailed.'* The T cell-enriched suspensions were washed, counted, cells at the same time. In four of these experiments, both unstimu-
and resuspended in complete medium. In some experiments, T-cell lated and antigen-activated T-cell blasts were recovered by incuba-
suspensions were further fractionated by negative selection into tion with carbonyl iron (37°C for 30 minutes) followed by magne-
CD4' and CD8' enriched T cells, as reported.'* The adherent and tism, washed three times, and cocultured for an additional 16 hours
the T-cell-enriched preparations were routinely analyzed by flow at a 2-to-l ratio with freshly prepared autologous or allogeneic
cytometry with anti-CD14 and anti-CD3 MoAbs, respectively. When monocytes.
their purity was lower than 95%, cell preparations were discarded. Detection of monocyte PCA and TF production. At the end of
For simplicity, adherent and T-cell-enriched preparations will be culture period, cell suspensions consisting of monocytes alone,
referred to as monocytes and T cells, respectively. In some experi- monocytes plus clonal supernatants, or monocytes plus activated T
ments, monocytes were recovered from mononuclear cell samples cells or activated and fixed clonal T cells were disrupted by repeated
by either adherence to plastic or centrifugation on density gradients freezing and thawing followed by sonication. Total cellular content
(at the interface between 40% and 50% of Percoll solutions). of PCA was determined in a one-stage clotting assay and expressed
Generation o antigen-specz$c T-cell clones. T-cell clones see-
f in arbitrary units (U/105 monocytes) assigned by comparison with
cific for PPD, 70-HSP, mite allergen DP, or excretoryhecretory a standard curve derived from rabbit brain thromboplastin standard
antigen of TES were generated from peripheral blood mononuclear (Manchester Comparative Reagents, Manchester, UK), as detailed
cells of three donors showing in vitro lymphocyte proliferation to elsewhere? Our log-log plot was linear up to 200 seconds clotting
PPD, 70-HSP, DP, and TES antigens, as previously described in time. Values less than 10 U110' monocytes corresponded to clotting
detail.14The phenotype of T-cell clones was examined by flow cy- times ranging from 170 to 80 seconds. One thousand units corres-
252 DEL PRETE ET AL
Table 1. Monocyte PCA Induced by PHA-Activated DP and four Th2 clones specific for the excretorykecretory
Human 1-Lymphocyte Subsets antigen of TES. To assess the ability of both Thl and Th2
PCA lU/105 monocytes) clones to induce monocyte TF production and PCA. T-cell
Preactivation blasts of each clone were cultured for 24hours in the absence
Cells in Culture With PHA Exp 1 E Ep p
xx
ExpZ 3 4
or presence of PHA and then added at a 2: 1 ratio to autolo-
Monocytes
alone c10 ~ <l0 <l0 <l0
Monocytes + gous or allogeneic monocytes. Neither T h l nor Th2unstimu-
Unfractionated T cells ~< l 0< l 0< l 0 <l0 lated clones were able to induce monocyte TF production/
+ 116 472 3
10
7129 213 6 211
PCA over background levels detected in cultures of mono-
CD4' T cells <l0 <l0 <l0
~ <l0
+ 7728 9628 3322 4925 cytesalone(Fig 1). However,24-hour preactivation with
CD8' T cells <l0 <l0 <l0
~ <l0 PHA enabled all seven Thl clones, but none of the seven
+ 2024 4524 2522 36+2
Th2, to induce remarkable PCA (Fig 1A) and T F production
Monocytes l4 X lo5) were cultured for 8 hours in the presence of medium
alone or with different T-cell subsets (8 x lo5)that had been precultured for 48
(Fig 1B) in both autologous and allogeneic (datanot shown)
hours in the absence or presence of PHA (1% vollvoll and then washed. Results monocytes. As adherence may lead to a general activation
represent mean values l+SD) of duplicate determinations of duplicate samples. of genes involved in the early monocyte response."." experi-
Paired t-test value between CD4' and CD8' T cells, 2.35 ( P = .08).
Abbreviation: Exp. experiment.
ments wereperformedinwhich monocytes to be assayed
for Th 1-induced TF production werepurified from the mono-
nuclearcellpreparations by either adherence to plastic o r
ponded to approximately22 seconds clotting time. PCA was charac-
terized as factor VII-dependent procoagulant activity by evaluating
its sensitivitytophospholipase C (Calbiochem,SanDiego, CA),
concanavalinA,andcysteineproteaseinhibitor (HgCI?), andby
TF protein
using factor VII- and factor X-deficient plasma samples.2o
was quantitated by a specific enzyme-linked immunosorbent assay
(ELISA) kit (American Diagnostica Inc, Greenwich, CT) purchased
from Ortho Diagnostic Systems (Milan, Italy). Briefly, after solubili-
zation of membrane proteins with Triton X-IO0 and ultracentrifuga-
tion (50,OOOg for l hour), supernatants were processed according to
the manufacturer's instructions.Values of diluted samples paralleled
the standard curve of TF for the ELISA. Values of immunoreactive
TF in the ELISA system were positively correlated with factor VII-
dependent PCA ( r = .93).
RESULTS
Both CD4+ and CD8+activated human T cells can induce
monocyte PCA. T cell-enriched suspensions obtained from
nonadherent peripheral blood mononuclearcells of four
healthy donorswereseparated by negativeselection into
either CD4+ or CD8+ cell populations. Unfractionated CD4'
and CD8' T cells were then cultured for 48 hours in the
absence or presence PHA and then assayed for their ability
of
to induce PCA in autologous or allogeneic monocytes. As
shown in Table 1, unstimulated T cells did not induce PCA
over background values detected in cultures of monocytes
alone (less than l0 U/105 monocytes), but stimulation with
PHA enabled both unfractionated T cells and their CD4' or
CD8+ subsets to induce PCA in both autologous (Table 1)
and allogeneic (data not shown) monocyte cultures. Levels
of PCA expressed by monocytes cocultured with autologous
activated CD4' T cells were consistently higher than those
found in parallel cocultures with the corresponding CD8' T Medium PHA Medium PHA
cells.
Fig 1. Induction of monocyte PCA and TF synthesis by activated
Human Thl, but not Th2, T-cell clones can induce PCA CD4+ Thl and Th2 human T-cell clones. Clonal T blasts from seven
and TF production by monocytes. To identify which of the Thl and seven Th2 clones were prestimulatedfor 24 hours with PHA
two main human CD4+ Th subsets was able to induce PCA (1% vol/voll, washed, and cocultured for 8 hours at a 2 1 ratio with
in monocytes,we raised a series of human CD4' T-cell autologous monocytes(4 x 106/mL).At the end of the culture period,
cells were disruptedand total cellular content of PCA(A) or TF protein
clones with defined Thl or Th2 cytokineprofile and known
(B) was quantitated as reported in Materials and Methods. Results
antigen specificity. They included three T h l clones specific represent mean values ofduplicatedeterminationsof duplicate sam-
for PPD of M tuberculosis and four T h l clones specific for ples. Dotted lines indicatethe highest levels of PCAand TF measured
70-HSP, as well as three Th2 clones specific for mite allergen in cultures of unstimulated monocytes alone.
SYNTHESIS HUMANMONOCYTE
TF T h l n h 2 CELLS AND
Table 2. TF Production Induced by Activated Thl Clones in Table 4. Monocyte TF Production Inducedby Fixed, Activated T
Monocvtes isolated bv Adherence or Densitv Gradient Blasts From Thl or Th2 Clonesandlor Their Supernatants
Procedure for Monocyte Monocyte TF Production
Purification (ng/rnL)
Adherence Percoll Gradient Culture Conditions Exp 1 Exp 2 Exp3
Culture Conditions (nglmL) (ng1mL)
Medium alone <O.l O
< . l <O.l
Monocytes alone <0.1 <0.1 Mock supernatant <O.l < .
O l <0.1
Monocytes + T h l cells 0.4 0.6 0.3
Unstimulated P.8 clone (Thl) <0.1<0.1 T h l supernatant 0.6 0.4 0.4
Preactivated P.8 clone (Thl) 1.7 2.0 T h l cells + T h l supernatant 2.5 2.9 3.1
Unstimulated P.16 clone (Thl) <0.1<0.1 T h l cells + Th2 supernatant <O.l <0.1 <O.l
Preactivated P.16 clone (Thl) 1.5 1.6 +
T h l supernatant Th2 supernatant 10.1 <O.l <o. 1
Unstimulated HS.36 clone (Thl) <O.l <O.l Th2 cells <O.l <O.l <o. 1
Preactivated HS.36 clone (Thl) 1.6 1.8 Th2 supernatant <0.1 <O.l <0.1
Unstimulated TC.41 clone (Th2) <O.l <O.l Th2 cells + Th2 supernatant <0.1 <0.1 <0.1
Preactivated TC.41 clone (Th2) <0.1 <O.l +
Th2 cells T h l supernatant 1.5 1.7 1.9
Monocytes were enriched from peripheral blood mononuclear cells Th2 cells + IFN-y (500 U/mL) 0.6 0.6 0.9
of the same donors by either adherence to plastic or centrifugation Monocytes (4 x lo5)were cultured for 8 hours in the presence of
on Percoll gradients. Monocytes prepared by the two procedures medium alone, PHA (mock supernatant), or 8 x lo5 paraformalde-
were then cultured in the absence or presence of unstimulated or hyde-fixed T-cell blasts from PHA-preactivatedT h l or Th2 clones with
PHA-preactivated T-cell blasts of Thl and Th2 clones, as detailed in or without addition of their 4
supernatants (1 to final dilution).Results
the legend to Fig 1. Results represent mean values obtained in one represent mean values of duplicate determinations of duplicate sam-
representative experiment of four similar experiments. ples. SD values were consistently lower than 17% of the mean.
stimulated Th2 clones did not upregulate TF production in
centrifugation on Percoll gradients. As shown in Table 2,
the same monocyte preparations. In subsequent experiments,
coculture with PHA-preactivated T-cell blasts of Thl clones
T-cell blasts of three Thl and two Th2 clones were stimu-
induced monocytes to produce comparable amounts of TF,
lated for 48 hours with medium alone or the specific antigen
irrespective of the procedure used for their purification. We
in the presence of autologous adherent cells. Unstimulated
then asked whether the inducer function for monocyte TF
and antigen-stimulated T-cell blasts were then washed, incu-
synthesis could be triggered by stimulating Thl clones with
bated with carboy1 iron followed by magnetism to remove
the specific antigen. T-cell blasts of three T h l and three
phagocytic cells, counted, and cocultured for an additional
Th2 clones were cocultured for 1 hours with autologous
6
16 hours at 2:l ratio with autologous or allogeneic mono-
monocytes (as antigen-presenting cells) in the absence or
cytes. As expected, antigen-stimulated T-cell blasts of Thl
presence of optimal concentrations of their specific antigens.
clones induced both autologous and allogeneic monocytes
As shown in Table 3, addition in culture of the specific
to TF synthesis (ranging from 0.6 to 1.8 and from 0.4 to 2.1
antigen enabled Thl clones to induce TF production by au-
ng/mL, respectively), whereas their unstimulated counter-
tologous, but not allogeneic (less than 0.1 ng/mL), antigen-
parts, as well as unstimulated or antigen-stimulated Th2
presenting monocytes. Under similar conditions, antigen-
clones, did not (<0.1 ng/mL).
Cell-to-cell contact with and cytokines secreted by acti-
vated Thl clones are both required for induction o optimal
f
Table 3. Monocyte TF Production Inducedby Autologous Thl and TF production by monocytes. We then asked whether acti-
Th2 Clones Stimulated With Specific Antigen vated Thl clones induced monocyte TF production through
TF Production (ng/mL) a membrane signal and/or by release of cytokines. To investi-
Cells in Culture
gate this point, T-cell blasts of four Thl clones were preacti-
With Medium With Specific Antigen
vated for 24 hours with PHA, fixed with paraformaldehyde,
Monocytes alone <0.1 10.1 and then added to monocytes in the presence of medium
Monocytes + P.16 (Thl) <0.1 0.5 t 0.3
alone or their supernatants recovered after PHA stimulation.
+
Monocytes HS.18 (Thl) <O.l 0.9 2 0.08
Addition to monocytes of either activated andfixed Thl
Monocytes + HS.36 (Th1) <O.l *
1.2 0.11
cells alone or their supernatants alone induced poor TF pro-
Monocytes + TC.39 (Th2) <O.l <O.l
Monocytes + DP.6 (Th2) <0.1 <O.l duction, ranging from 0.2 and 0.6ng/mL. In contrast, com-
Monocytes + DP.60 (Th2) <O.l <0.1 bined addition of both fixed, activated Thl cells and Thl
Monocytes (4 x lo5) were cocultured for 16 hours with T blasts (8
supernatants resulted in high TF production (ranging from
x 105) of autologous T h l or Th2 clones in the absence or in the to 3.4 ng/mL)' Data from three representative experi-
presence (5 pg/mL) of specific antigen (PPD for P.16,70-HSPfor HS.18 ments are in 4. In the same experiments*fixed
and HS.36, TES for TC.39. and DP for DP.6 and DP.60 clones). Results activated Th2 cells and/or their supernatants did not induce
represent mean values (+SD) of duplicate determinations of duplicate detectable TF production, whereas addition in culture of both
samples. fixed, activated Th2 cells and Thl supernatants resulted in
254 DEL PRETE ET AL
remarkable production of TF. In contrast, combination of
either fixed Thl or Th2 unstimulated cells and Thl superna-
tants induced the same low TF production obtained with Thl
supernatants alone (data not shown). Interestingly, either the A 1'"1 - SN
i 501
LL
combination of fixed, activated Thl cells and Th2 superna-
tants or mixtures of Thl and Th2 supernatants abrogated the
low TF production induced byfixed, activated Thl cells
alone or Thl supernatants alone, respectively (Table 4 . ) \cl Th2 SN (TC.41)
When Th2 supernatants were added to the combination of
fixed, activated Thl cells and their supernatants, inhibition
of TF production was observed (Table 4). Collectively, these
data suggest that at least two signals are required for induc-
tion of optimal TF synthesis by monocytes: one signal is
delivered by membrane interactions between monocytes and B
molecule(s) expressed by T cells only after recent stimula-
tion, and the other is provided by Thl-derived cytokines.
Data also indicate that activated Th2 clones secreted in their
supernatants some factor(s) that inhibits monocyte TF syn-
thesis.
IFN- y is involved in the Thl -mediated induction of TF
synthesis. To investigate which of the known cytokines
selectively produced by human Thl clones was involved
in the induction of TF synthesis, different concentrations
of recombinant IL-2, IFN- y , TNF-a, or TNF-@were added
(alone or in combination) to monocyte cultures. Minimal
TF production (ranging from 0.1 to 0.4 ng/mL) was detect- 0 2 20 200
able only in the presence of high amounts of IFN-y (500 Cytdche coIK(wI1Talkn
to 1,000 U/mL), whereas IL-2, TNF-a, and TNF-P were
Fig 2. Dose-dependent inhibition by Th2 supernatants, recombi-
completely ineffective. In the same experiments, addition nant human 11-4, IL-13, and IL-10 of the monocyte TF synthesis in-
in culture of graded concentrations of INF-y (but not of duced by activated Thl clones and their supernatants. (A) Clonal T-
IL-2, TNF-a, or TNF-P) resulted in dose-dependent TF cell blasts from three Thl clones were prestimulated for 24 hours
production (ranging from 0.6 to 1.1 ng/mL) by monocytes with PHA (1% vol/vol), fixed with 0.1% paraformaldehyde, washed
three times, andcoculturedfor 8 hours at a 2 1 ratio with autologous
cocultured with Th2 cells fixed after activation (Table 4).
monocytes (4 x 1O6/mL)in the presence of their PHA supernatants
In subsequent experiments, neutralizing amounts of anti- l : final dilution) and graded dilutions of supernatants from three
1I
IL-2, anti-IFN-y, anti-TNF-a, or anti-TNF-@ antibodies Th2 clones (DP.6, TC.19, and TC.41) previously stimulated with PHA
were added to monocytes stimulated with activated and for 24 hours. Mock supernatant waa obtained by reconstftutingwith
fixed Thl cells and their supernatants. Addition in culture of
PHA the pooled 24-hour supernatants unstimulated cultures of the
same Th2 clones. (B) T-cell Masts from the same Thl clones were
of anti-IFN- y antibody resulted in a partial but consistent prestimulated for 24 hours with PHA (1%vol/vol), fixed, and cocul-
(from 36%to 59%) inhibition of TF synthesis, whereas tured for 8 hours at a 2 1 ratio with autologous monocytes (4 x lo6/
none of the other anticytokine antibodies was inhibitory. mL) in the presence of their PHA supernatants (1:4 final dilution),
Rather, after addition of either anti-TNF-a or a mixture f
with or without addition o graded concentrations of human recombi-
nant IL-4,11-5,11-10 (U/mL), or human recombinant IL-13 InglmL).
of two anti-TNF-aR antibodies or both reagents, a slight the
At the end of culture period, c l membrane proteins were solubi-
el
but consistent increase (from 16%to 37%) in monocyte TF lized and TF synthesis wasquantitated by a specific EUSA assay, as
production induced by activated and fixed Thl cells and reported in Materials and Methods. Results repre-nt mean values
their supernatants was observed. IkSD) of percent variations in respect to baseline TF values obtained
IL-4, IL-IO, and IL-13 are powerfulinhibitors o TF f in cultures performed without addition of Th2 supernatants (A) or
indicated cytokines (B).
production induced by Thl clones and their supernatants.
To identify which cytokines were responsible for the inhibi-
tory effect exerted by Th2 supernatants on TF production containing Thl supernatants (data not shown). In subse-
induced by activated and fixed Thl cells and their cytokines quent experiments, the role for IL-4 and IL-l0 in the inhibi-
(Fig 2A), the effect of graded concentrations of recombi- tion exerted by Th2 supernatants was further investigated
nant IL-4, IL-5, IL-10, and IL-13 was examined. As shown by using anti-IL-4 and anti-IL-l0 neutralizing antibodies.
in Fig 2B, IL-4 and IL-13, and even more effectively, IL- As shown in Fig 3, pretreatment with neutralizing anti-IL-
10 exerted a dose-dependent inhibitory effect on the Thl- 10 antibody of Th2 supernatants to be added in culture
induced monocyte TF production, whereas IL-5 was com- substantially reduced their inhibitory effect on TF produc-
pletely ineffective. Likewise, IL-4, IL-13, and IL-10, but tion by monocytes cocultured with Thl cells and their se-
not IL-5, abrogated the low TF production induced in creted cytokines, whereas neutralization of IL-4 was only
monocytes alone by high amounts of IFN-y or IFN-y - partially effective.
,,..
I,..,
I-IUMHIY
...L."-L*
I n I/ I n r CELLS AND MONOCYTE TF SYNTHESIS 255
vations that unstimulated peripheral blood lymphocytes from
healthy donors did not induce PCA, whereas circulating lym-
phocytes from patients with unstable angina, which include
a number of in vivo activated T cells, induced detectable
PCA in both normal and patient monocytes.'
Th2 supemrtrnt Prior data on human T-cell clones demonstrated that col-
laborative signals to monocytes may be delivered by direct
Th2 supematant surface contact or by soluble cytokines and that, depending
+d I L - 4 on the experimental conditions, either mechanism can be
Th2 supematant effe~tive.'~ Similarly, highly competent induction of mono-
+ anulL-5 cyte procoagulant response by either mechanisms was docu-
Data
mented in the murine s y ~ t e m . ' ~ . ' ~ obtained in our exper-
Th2 Sqlemtani iments with fixed T-cell clones suggest that neither
+dIL-l0
membrane contact signal(s) delivered by activated Thl cells
nor Thl-derived cytokines alone can induce optimal TF pro-
duction in monocytes, while the combination of both signals
is far more potent. However, differences in experimental
Fig 3. Effectofaddeion of anti-IL-4or anti-IL-l0 neutralizing conditions, mechanism of clonal stimulation, and sensitivity
antibodies to Th2 supernatants able to inhibe the Thl-induced TF
from
production by monocytes. Clonal T-cell blasts three Thl clones
of responding monocytes may account for the partial discrep-
were prestimulated for 24 hours with PHA (1% vol/vol), fixed with ancy. More interestingly, blocking of the ongoing cytokine
0.1% pareformaldehyde, washed three times, and cocukured for 8 secretion by activated Thl and Th2 clones demonstrated that
hours at a 2:l ratio with autologous monocytes (4x 106/mL) inthe not only Thl, but also Th2, cells could deliver the membrane
presence of their PHA supernatants (Thl supernatant. 1:4 final dilu- activation signal(s) required for monocyte TF synthesis. In-
tion) and pooled supernatants from three prestimulated Th2 clones
(Th2 supernatant). Before its addition in culture (1:2 final dilution), deed, the combined addition to monocyte cultures of super-
pooled Th2 supernatant wes incubated for 2 hoursat CC with me- natants containing Thl-type cytokines and fixed, activated
dium alone or neutralizing anti-IL-4, anti-IL-5, or anti-IL-l0 poly- Th2 cells resulted in remarkable TF production. A possible
clonal antibodies(10 pg/mL). At the end of cutture period,cell mem- explanation for whyfixed Thl, butnot Th2, cells alone
brane proteins were solubilized and TF synthesis was quantitated,
as reported in Materials and Methods. Results represent mean values
induced minimal but consistent TF synthesis is that, besides
( S D I of duplicate determinations TF production in triplicate Sam-
of activation molecules, such as CD28 and CD40 ligand shared
ples assessed in one representative experiment of four consecutive by both Th-cell s ~ b s e t s ,Thl * ~ ~ also express mem-
~ ~ might
experiments. brane-bound Thl-type cytokines, as a result of the fixation
procedure just at the time of the cytokine secretion process.
An alternative explanation might be that our fixation protocol
DISCUSSION was partially ineffective, allowing Thl clones to deliver both
In this report, we provide evidence that both CD4+ and membrane and cytokine signals to monocytes. This latter
CD8+ human T cells can induce PCA in monocytes. How- explanation, however, seems to be unlikely, as neither entire
ever, as both T-cell subsets were enriched by negative selec- nor 10-fold concentrated 24-hour supernatants collected
tion, the possibility that also a few CD4-CD8- contaminat- from activated Thl clones after fixation had detectable cy-
ing cells may contribute to the induction of PCA in tokine content (ie, less than 0.5 U/mL IFN-y) or enabled
monocytes cannot be excluded. Levels of PCA induced in fixed Th2 clones to induce TF synthesis by monocytes (data
monocytes by autologous CD4+ T cells were higher than not shown). When added alone to monocyte cultures, even
those induced in the same monocyte cultures by the corre- PHA superpatants containing Thl-type cytokines induced
sponding CD8+ T cells. This observation is consistent with minimal but consistent TF production, whereas recombinant
data obtained in the murine system, where CD4+ T-cell K-2, TNF-P, TNF-a, or IFN-y, at any concentration or
clones produced higher amounts of monocyte procoagulant combination used, did not. The simplest explanation for this
inducing factor than CD8+ alloreactive clones.12 Also in discrepancy is that residual PHA was responsible for activa-
agreement with the murine modelI7 is the demonstration that tion of the few contaminating T cells present in cultures,
human CD4+ Thl, but not Th2, clones were able to express which in turn enabled monocytes to respond to the second
inducer function for monocyte TF production, provided that signal delivered by Thl-type cytokines. In this regard, IL-
adequate activation was achieved. The results reported in 2, TNF-P, and TNF-a seem to play a minimal role, if any,
this study also suggest that the complete set of molecules as neither their addition nor their neutralization in Thl super-
responsible for the T-cell signalling to monocytes is ex- natants hadany effect. As in the murine a major
pressed only by activated Thl cells. In most experiments, role for the induction of TF expression by monocytes can
activation was achieved with PHA, but stimulation with the be assigned to IFN-y. This cytokine showed costimulatory
specific antigens under major histocompatibility complex effect in combination with fixedTh2 cells, whereas the addi-
(MHC)-restricted conditions was also effective in inducing tion of anti-IFN-y antibody to Thl-derived supernatants
human Th 1, but not Th2, clones to trigger TF production by was inhibitory. However, as neutralization of IFN-y in Thl
monocytes. These findings may account for previous obser- supernatants did not result in complete abrogation of mono-
256 DEL PRETE ET AL
cyte TF synthesis, the possibility that supernatants from hu- F, Giusti B, Dabizzi P, Poggesi L, Modesti PA, Trotta F, Rostagno
man T h l clones also contain a distinct monocyte procoagu- C, Boddi M, Gensini GF: Transient intermittent lymphocyte activa-
lant inducing factor-like cytokine, as shown in the murine tion is responsible for instability of angina. Circulation 86:790, 1992
system,I2 cannot be excluded. IO. Schwartz BS: Antigen-induced monocyte procoagulant activ-
ity. J Clin Invest 76970, 1985
It has been shown in both mice and humans that, during
11. Gregory SA, Kornbluth RS, Helin H, Remold HG, Edgington
some strong immune responses, Thl and Th2 effectorpath-
TS: Monocyte procoagulant inducing factor: A lymphokine involved
ways appear to be exclusive, self-stimulatory, and mutually in the T cell-instructed monocyte procoagulant response t o antigen.
a
inhibit~ry.~”~’ Suchrule seems to be confirmed by data J Immunol 137:3231, 1986
showing theinhibitory effect exerted by Th2 supernatants on 12. Fan ST, Edgington TS: Clonal analysis of mechanisms of
the Thl-mediated induction of TF synthesis by monocytes. murine T helper cell collaboration with effector cells of macrophage
Among Th2 cytokines, IL-4, IL- 13, and even more effec- lineage. J Immunol I41 :I8 18, 1988
tively, IL-IO seem to play a major role in this phenomenon. 13. Mossman TR, Cherwinski H, Bond MW, Giedlin MA, Coff-
This is in agrement with the observations that IL-4 and IL- man RL: Two types of murine helper T-cell clone. 1. Definition
10 can inhibit both TF mRNA expression and production
TF according to profiles of lymphokine activities and secreted proteins.
by monocytes in responseto endotoxin.30,3’As expected, J Immunol 136:2348, 1986
14. Del Prete GF, De Carli M, Mastromauro C, Biagiotti R, Mac-
humanIL-13,anIL-4-likecytokine acting onmonocytes
chia D, Falagiani P, Ricci M, Romagnani S: Purified protein deriva-
and B cells but not on T cells,” consistently inhibited the
tive of Mycobacterium tuberculosis and excretory-secretory anti-
ThI-induced TF production. Of note is the stronginhibitory gen(s) of Toxocara canis expand in vitro human T cells with stable
activity of IL-IO, a cytokine producedby different cell types, and opposite (type 1 T helper or type 2 T helper) profile of cytokine
such as B cells, monocytes, most human Th2 clones, and a production. J Clin Invest 88:346, 1991
number of Thl clones,” that downregulates different func- 15. Mosmann TR, Coffman RL: Heterogeneity of cytokine secre-
tions of antigen-presenting cells, including antigen presenta- tion patterns and functions of helper T cells. Adv lmmunol 46: 1 1 I ,
tion, delivery of costimulatory signals, and cytokine produc- 1989
tion.34.15 Thus, it is reasonableto that
suggest L10 16. Sher A, Coffman RL: Regulation of immunity to parasites by
representsan important intrinsicregulatory mechanism of T cells and T cell-derived cytokines. AnnRev Immunol 10:385,
TF synthesis that is activated in monocytes during endotox- 1992
17. Fan ST, Glaserbrook AL, Edgington TS: Clonal analysis of
inemia‘ or other pathologic conditions, such as kidney graft
CD4’ T helper cell subsets that induce the monocyte procoagulant
rejection or Crohn’s disease, where selective infiltration of
a
response. Cell Immunol 128:52, 1990
target tissues by Thl-type cells has recently been demon- 18. Piccinni MP, Macchia D, Parronchi P, Giudizi MG, Bani D,
.~~
~ t r a t e dWhether the inhibitory activity of IL-4, IL-13, and Alterini R, Grossi A, Ricci M, Maggi E, Romagnani S : Human bone
IL-l0 on TF synthesis by monocytes may be exploited as a marrow non-B, non-T cells produce IL-4 in response to cross-linkage
therapeutic tool in these conditions should be matter of thor- of Fct and Fcy receptors. Proc Natl Acad Sci U S A 88:8656, 1991
ough investigation. 19. Parronchi P. Tiri A, Macchia D, De Carli M, Biswas P, Simo-
nelli C, Maggi E, Del Prete G, Ricci M, Romagnani S : Noncognate
REFERENCES contact-dependent B cell activation can promote IL-4-dependent in
vitro human IgE synthesis. J Immunol 144:2102, 1990
1. Colvin RB, Johnson RA, Mihm MC, Dvorak HF: Role of the
20. Helin H, Edgington TS: Allogenic induction of the human T
clotting system in cell-mediated hypersensitivity. I. Fibrin deposition
cell-instructed monocyte procoagulant response is rapid and is elic-
in delayed skin reactions in man. J Exp Med 138:686, 1973
ited by HLA-DR. J Exp Med 158:962, 1983
2. Edgington TS, Mackman N, Brand K, RufW: The structural
21, van Ginkel CJW, van Aken WG, Oh JIH, Vreeken J: Stimula-
biology of expression and function of tissue factor. Thromb Haemost
tionof monocyte procoagulant activity by adherence to different
66:67, I99 1
surfaces. Br J Hematol 37:35, 1977
3 . Edwards RL, Rickles FR: Delayed hypersensitivity in man:
22. Sporn SA, Eierman DF, Johnson CE, Morris J, Martin G,
Effects of systemic anticoagulation. Science 200541, 1978
Ladner M, Haskill S : Monocyte adherence results in selective induc-
4. Rothberger H, Zimmerman TS, Vaugham JH: Increased pro-
duction and expression of tissue thromboplastin-like procoagulant tion of novel genes sharing homology with mediators of inflamma-
activity in vitro by allogeneically stimulated human leukocytes. J tion and tissue repair. J lmmunol 144:4434, 1990
Clin Invest 62:649, 1978 23. Gregory SA, Edgington TS: Tissue factor induction in human
5. Auger MJ, Mackie MJ: Monocyte procoagulant activity in monocytes: Two distinct mechanisms displayed by different alloanti-
breast cancer. Thromb Res 47:77, 1987 gen responsive T cell clones. J Clin Invest 76:2440, 1985
6. Taylor FB, Chang A, RufW, Monissey JH, Hinshaw LB. 24. De Franco AL: Between B cells and T cells. Nature 351:603,
Catlett R, Blick K, Edgington TS: Lethal E. coli septic shock is 1991
prevented by blocking tissue factor with monoclonal antibody. Circ 25. Noelle RJ, Ledbetter JA, Aruffo A: CD40 and its ligand, an
Shock 33:127, 1991 essential ligand-receptor pair for thymus-dependent B-cell activa-
7. Rothberger H, Baninger M, Meredith J: Increased tissue factor tion. Immunol Today 13:431, 1992
activity of monocyteshacrophages isolated from canine renal allo- 26. Moon DK, Geczy CL: Recombinant IFN-y synergizes with
grafts. Blood 63:623, 1984 lipopolysaccharide to induce macrophage membrane procoagulants.
8. Edwards R L , Levine JB, Green R,Duffy M, Mathews E, J Immunol 141:1536, 1988
Brande W, Rickles FR: Activation of blood coagulation in Crohn’s 27. Gajewski TF, Fitch F W : Anti-proliferative effect of IFN-y in
disease. Gastroenterology 92:329, 1987 immune regulation, I. IFN-y inhibits the proliferation of TH2 but
9. Neri Serneri GG, Abbate R, Gori AM, Attanasio M, Martini not THI murine HTL clones. J Immunol 140:4245, 1988
HUMAN Thl/ThZ CELLS AND MONOCYTETFSYNTHESIS 251
28. Mosmann TR, Moore KW: The role of IL-10 in cross-regula- 33. Del Prete G, De Carli M, Almerigogna F, Giudizi MG, Biagi-
tion of TH1 and TH2 responses. Immunoparasitol Today 12:49,1991 otti R, Romagnani S : Human IL-l0 is produced by both type 1
29. Romagnani S: Induction of TH1 and TH2 response: A key role helper (Thl) and type 2 helper (Th2) T cell clones and inhibits their
for the ‘natural’ immune response? Immunol Today 13:379, 1992 antigen-specific proliferation and cytokine production. J Immunol
30. Ramani M, Ollivier V, Ternisien C, Vu T, Elbim C, Hakim 150353, 1993
J, De Prost D: Interleukin-4 prevents the induction of tissue factor 34. Ding L, Shevach EM: L - l 0 inhibits mitogen-induced T cell
mRNA in human monocytes in response to LPS or PMA stimulation. proliferation by selectively inhibiting macrophage costimulatory
Br J Haematol 85462, 1993
functions. J Immunol 148:3133, 1992
31. Pradier 0, Gerard, Delvaux A, Lybin M, Abramowicz D,
Capel P, Velu T, Goldman M: Interleukin-l0 inhibits the induction 35. de Waal Malefyt R, Abrams J, Bennett B, Figdor CG, de
of monocyte procoagulant activity by bacterial lipopolysaccharide. Vries JE: Interleukin-l0 (IL-10) inhibits cytokine synthesis by hu-
Eur J Immunol23:2700, 1993 man monocytes: An autoregulatory role of IL-l0 produced by mono-
32. Zurawski G, de Vries J E Interleukin-13, an interleukin-4- cytes. J Exp Med 174:1209, 1991
like cytokine that acts on monocytes and B cells, but not on T cells. 36. Romagnani S: Lymphokine production by human T cells in
Immunol Today 15:19, 1994 disease states. Annu Rev Immunol 12:227, 1994
