PRI/DynCorp,. NCI-FCRDC, for cytokine assays. We appreciate. Joyce. Vincent for editorial assistance and. Drs. Marco. Cippitell,. Pantosh. Ghosh, and Antonio.
Cellular and molecular mechanisms of IFN-y production by IL-2 and lL-12 in a human NK cell line Jianping
Ye, John
A. Ortaldo,
Laboratory
ofExperimental
Immunology,
Kevin Biological
Conlon, Response
Abstract: Interferon-y munoregulatory protein cells and large granular to different extracellular
(IFN-y) is an important improduced predominantly by T lymphocytes (LGL) in response signals. In particular, two interleukins (ILs), IL-2 and IL-12, have been shown to be potent inducers of IFN-y gene expression in both T cells and LGL. Although it has been reported that there are some T cell lines that produce IFN-y in response to IL-2 and !L-12 stimulation, there has as yet been no report of a natural killer (NK) cell line that responds in a similar manner. In this report we present evidence that the cell line NK3.3 derived from human NK cells, responds to both IL-2 and IL-12, as measured by increases in IFN-y and granulocyte-macrophage colony-stimulating factor (GM-CSF) cytoplasmic mRNA and protein expression. In addition, when used together IL-2 and IL-12 synergized in the induction of IFN-’y and GM-CSF and this synergy was attributed to an increased accumulation and stability of the IFN-y and GM-CSF mRNAs. To investigate the signaling pathways involved in the gene induction, five inhibitors, cyclosporin A (CsA), transforming growth factor-, cycloheximide, genistein, and staurosporine A, were used in analyzing the effects of IL-2 and IL.12 on NK3.3 cells. The results suggest that activation of protein kinase C, but not new protein synthesis, is required for IL-2 induction of IFN-y and GM-CSF cytoplasmic mRNA. In contrast, IL-12 induction of IFN-’y cytoplasmic mRNA appears to only partially depend on activation of protein kinase C. Furthermore, both transforming growth factorand genistein, a tyrosine kinase inhibitor, could suppress IL-2 and IL-12 signaling but CsA was generally inactive. It also was observed that suppression of cytokine gene expression by these agents was independent of the inhibition ofproliferation. In addition, IL-2 but not IL-12 induced nuclear factors NF-icB and APi, and regulation of the nuclear levels of these two DNA binding protein complexes is correlated with IFN-y and GM-CSF gene expression. These data indicate that IL-2 and IL-12 may have distinct signaling pathways leading to the induction of IFN-y and GM-CSF gene expression, and that the NK3.3 cell line may serve as a novel model for dissecting the biochemical and molecular events involved in these pathways.J. Leukoc. Biol. 58: 225-233; 1995. Key
Words:
IFN-y
.
IL-2
#{149} IL-12
#{149} NK3.3
Robin
Winkler-Pickett,
ModIers
Program,
induced
and Howard
NCI-FCRDC,
A. Young
Frederick
Maryland
natural killer (NK) and T cells [3, 4], including the induction of IFN-’y production from NK and T cells [5]. Recently, IL-12 has been proposed as a prospective cytokine for the treatment of cancer [6, 7] and viral [6] and parasitic infections [8-1 1] because IL-12 appears to be an initiation cytokine for cell-mediated immunity [12, 13]. As NK cells are a primary target ofboth IL-12 and IL-2, these cells serve as an important target for understanding the mechanism of IL-12/IL-2 induction of gene expression. Unfortunately, NK cells comprise a very small subpopulation of peripheral blood mononuclear cells and the difficulty in obtaining sufficient numbers of highly purified cells is a major limitation for elucidating the biochemical and molecular mechanisms of IL-12/IL-2 signal transduction. It is obvious that identification of an IL-12-responsive NK cell line that is also responsive to IL-2 would be important in elucidating the signaling events resulting in gene induction by these lymphokines. In this report we have analyzed the effects of IL-2 and IL-12 on the human cell line, NK3.3. NK3.3 is a human NK cell line developed from a primary mixed lymphocyte culture by using human peripheral blood lymphocytes. This cell line has been characterized morphologically, phenotypically, and functionally and was found to resemble peripheral blood NK cells [14]. The growth of NK3.3 is strictly dependent on IL-2, and, on IL-2 stimulation, this cell line expresses an increased cytolytic function; accumulates mRNA for c-myb, c-myc, and IL-2 receptor [15]; and produces IFN-’y CD16, and CD56 [16]. Recently, Bacon et al. [17] have used this cell line to demonstrate that different JAK kinases are activated in response to IL-2 and IL-12. In this report we present data that NKS.3 cells produce both IFN-y and granulocyte-macrophage colony-stimulating factor (GM-CSF) in response to both IL-2 and IL-12. In addition, multiple aspects of the effects of these two ILs on this cell line have been explored, including cytokme production, cell surface marker expression, cell proliferation, signal transduction, and activation of nuclear proteins. Evidence is presented that IL-2 and IL-12 have
cell Abbreviations:
IFN-y,
cyte-macrophage
INTRODUCTION Interleukin-12 (IL-12) (natural killer cell stimulatory factor) is a 70-kDa heterodimeric cytokine composed of two covalently linked subunits (p40 and p35) [1] that are produced primarily by peripheral blood adherent and nonadherent lymphocytes [2]. IL-12 has multiple effects on
factor4l, acetate; cyanate;
interferon-’
colony-stimulating
IL,
factor;
interieukin;
TGF-fI,
PMSF, phenylmethylsulfonyl fluoride; PMA, PE, phycoerythrin; CsA, cyclosporin A; FITC, Act D, actinomycin D; EMSA, electrophoretic
NK, natural
killer
cell; PKC,
protein
kinase
granulo-
GM-CSF,
transforming
C; NF-icli,
phorboi fluorescein mobility
nuclear
growth myristate isothioshift assay;
factor
Current address: Kevin Conlon, Center for Biologics Evaluation Research, FDA, Rockville, MD 20852. Reprint requests: Howard Young, NCI-FCRDC, Building 560, 31-93, Frederick, MD 21702-1201.
Journal
of
Leukocyte
Biology
Volume
58, August
1995
icB. and Room
225
different biological effects on this cell line and these effects are mediated by distinct signaling pathways. In addition, IL-2 and IL-12 demonstrate a strong synergy in enhancing IFN-y and GM-CSF gene expression. We propose that the NK3.3 cell line may serve as a unique model for analyzing the molecular mechanisms of the signaling pathways involved in lymphokine gene activation induced by IL-2 or IL-12.
MATERIALS
AND
METHODS
determined
at each time point to normalize RNA loading variability (actin mRNA was stable over the time points in this experiment). The zero time point value was considered as 100% and used to determine the mRNA values at different time points.
Cell proliferation Cells
were
assay
adjusted
containing
to a density
IL-2 or IL-12,
text, ume
then transferred of 200 Isl/well.
cells
were
of 5 x 1 06/mi
in addition
into After
a 96-well, incubating
harvested
for
the
added
to each
flat-bottom at 37C
proliferation
in 5%
assay.
CO2
One
48
h the
of
Nuclear
protein
Recombinant human IL-2 was obtained from (Nutley, NJ). The recombinant human 1L-12 Institute under Material Transfer Agreement
Eiectrophoretic mobility shift assays (EMSAS) were used to detect nuclear DNA binding proteins, as described previously [19]. An APi binding sequence from the human IL-2 promoter
Staurosporine A, genistein, from Calbiochem-Novabiochem heximide was purchased
(PE)-conjugated fluorescein anti-CD16
and from
mouse isothiocyanate antibodies
actinomycin International Sigma (St.
anti-human
was purIN). CyAustria).
D (Act D) were purchased (San Diego, CA). CycloLouis, MO). Phycoerythrin
IL-2Ra,
anti-CD56,
(FITC)-conjugated were obtained from
mouse Becton
anti-CD69,
anti-human Dickinson
and
IL-2Rl, (San Jose,
CA). Cell culture NK3.3 cells were maintained in RPMI 1640 medium (BioWhitaker, Walkersviile, MD), suppiemented with 10% fetal calf serum (Flow Laboratories, Rockville, MD), 4 mM L-giutamine, 100 U/mI penicillin, 100 sWml streptomycin (Biofluids Inc., Rockville, MD), and 25-50 U/mI recombinant human IL-2. The cells were cultured at a density of 2-3 x
lO’/ml
in an incubator
with
105/ml. without
In most experiments IL-2 at a density
10%
CO2 and
the NK3.3 of 1 x 106/ml
used
at a density
cells were rested for 24 h before
in the use.
of 6-8
x
cells
indicated cell-free protein
were
in the figure
with
different
legends)
cytokines
for
at a cell density
various
of 1 x 106/ml
times
(as
then
the
supernatant analysis Resources,
gram
EUSA
was harvested for cytokine assays. IFN-’y or GM-CSF was performed by Clinical Immunology Services (ProInc., NCI-FCRDC, Frederick, MD) using commercial (for IFN-’y, INCSTAR Corporation, Stiliwater, MN; for GMSystems, Minneapolis, MN).
kits
CSF.
R&D
Flow cytometry analysis The 1 x
106 NK3.3
cells
were
treated
for
washed once with phosphate-buffered conjugated antibody, and analyzed (Becton contour
able
Dickinson) with 10,000 cells plots were generated by using the FACScan.
on
mRNA
24 h with
being standard
indicated
agents,
with PE- or FITCFlow Cytometer
analyzed. analysis
Isotogram programs
and avail-
analysis
Total cellular phenol/chloroform Friendswood,
RNA
was isolated from 1 x iO extraction procedure Ten micrograms of total
TX).
fractionated
on a formaldehyde-denaturing
to Magnabond (MSI, Westboro, MA). hybridized in Fasthyb (Digene, Silver probes prepared by using a random
Jolla,
CA).
All cDNA
probes
cpmfltg and all hybridizations Blots were exposed to Kodak
mANA
half-life
cells by using a singie-step (RNAsoI; Cinna Biotecx, cytoplasmic RNA was size
0.8%
ferred were cDNA
scanning
gel and
trans.
in the extract.
EMSA assay to detect APi The protein-DNA binding
exposed
Atlanta, to an
(Richmond,
of Leukocyte
CA),
Biology
and
whereby NK3.3 cells had of Act D 1 x i0 cells then the same number different time points, as of mRNA half-life, intenusing a Bio-Rad GS-670 the IFN-’yJactin ratio was
Volume
58, August
1995
and NF-icB protein in the nudear reaction was conducted in a 20-jsl
x-ray
GA). film
After overnight
electrophoresis
the gel was dried
and
at -70-C.
RESULTS
Induction
of IFN-y by IL-2 stimulation
To characterize IFN-y production by NK3.3 cells, the effects of IL-2 dose and stimulation time were investigated. As shown in Figure l#{192}, IFN-y production was stimulated by IL-2 in a dose-dependent pattern (IL-2 dosages from 1 to 500 U/mi). To examine the effect of stimulation time the NKS.3 cells were treated with 100 U/mi of IL-2 and the cell-free supernatant was collected at several time points and assayed for IFN-’y. The results (Fig. 1B) showed that IL-2-induced IFN-y was detected as early as 3 h and peaked at 48 h after addition of IL-2. In addition to IFN-y induction, IL-2 also induced GMCSF production (Fig. 1D) and expression of specific cell surface markers in the NK3.3 cells. Although the rested NKS.3 cells still express IL-2Ra (CD25), CD16, CD56, and CD69 on the cell surface, an increase in surface density of IL-2Ra, CD56, and CD69 was observed in the IL-2treated cells (Fig. 1C).
Induction
analysis
densitometer
Journal
agarose
After UV cross-linking blots Spring, MD) to ‘2P-labeied priming kit (Stratagene, La had a specific activity of at least 2-8 x 108 were performed with 1 x 106 cpm/ml. X-OMAT X-ray film for 6-72 h at -70C.
Act D (10 sgJmi) was added to the ccii culture, been treated with cytokine for 3 h. After addition were harvested immediately as a zero time point, of cells were harvested for RNA extraction at indicated in the figure legend. For calculation sities of IFN-y and -actin were quantitated by
226
the
saline, stained on an FACScan
analysis
(5’GAAATFCCAAAGAGTCATCAGA 3’) and a nuclear factor icB (NFKB) binding sequence from the IL-2 receptor a-chain promoter (5’AGGGGAATCTCCC 3’) [20] were synthesized by the phosphoramidite method on a DNA/RNA synthesizer (model 392; Applied Biosystems, Foster City, CA). The double-stranded DNA was labeled with “P-dCTP (Amersham, Arlington Heights, IL) by using Klenow fragment (BRL, Gaithersburg, MD). The NK3.3 cells were treated with cytokine for 3 h then the nuclear protein was prepared as follows: 1 x 108 cells were treated with 500 p1 lysis buffer (50 mM KC1, 0.5% NP-40, 25 mM HEPES, pH 7.8, 1 mM phenylmethylsulfonyl fluoride [PMSFJ, 10 gsg/ml leupeptin, 20 tWml aprotinin, and 100 tM DTF) and the nudei were washed once with the same buffer without NP-40. The nudear extract was prepared from the nudei with 300 p1 of extraction buffer (500 mM KC1, 10% glycerol with the same concentrations of HEPES, PMSF, ieupeptin, aprotinin, and WIT as the lysis buffer) dialyzed against the same buffer with 50 mM KC1 then stored at -70C. ‘2P-labeled APi or NF-icB binding oligonudeotides were used as probes
Diagnostics, treated
of culture.
reaction mixture with 0.5 izg poly(dll.dC) (Sigma), 2 jsg nudear protein, 5 x io cpm of probe, and 10 gxl of 2X GS buffer (40 mM Tris, pH 7.4, 120 mM KC1, 8% Ficoli, 4 mM EDTA, 1 mM DTF), and the DNA-protein complexes were resolved in a 5% nondenature acrylamide gel (National
medium
Cytokine assays NK3.3
end
for
microcurie
Reagents
recombinant human transforming growth factor-D (TGF-3) chased from Boehringer Mannheim Corporation (Indianapolis, ciosporin A (CsA) was obtained from Sandoz Ltd. (Vienna,
the
medium
was
Roche, Inc. from Genetics 1517 92. The
18 h before
complete
as described in the culture plate at a vol-
[‘HJTdR
Hoffmann-La was a gift CA-MTA#2
well
with
to the inhibitors
of IFN-’y by IL-12 stimulation
In peripheral blood NK cells IL-12 is a powerful inducer of IFN-y [3]. This prompted us to investigate the effects of IL-12 on the NKS.3 cell line. In contrast to strong effects on the peripheral blood cells, IL-12 stimulation resulted in approximately a threefold induction of IFN-y in this cell line (Fig. 2A). The induced IFN-y was initially
detected with an IL-2 dose of 0. 1 U/ml and it reached a maximal level with 10 U/ml of IL-12. The results from a time course study showed that a prolonged stimulation could induce more IFN-y between 24 and 96 h (Fig. 2B). GM-CSF also was induced in NK3.3 cells by IL-12 (Fig. 2D). However, in the IL-12-treated cells no change was detected in the cell surface markers (Fig. 2C) that were induced by IL-2. These results demonstrate that although both IL-2 and IL-12 induce IFN-y and GM-CSF expression, they have a different spectrum of effects on the NK3.3 cells.
Kinetics of IFN-yand GM-CSF induction by IL-2 or IL-i 2
cytoplasmic
mRNA
Total cytoplasmic RNA was prepared from NK3.3 cells treated with IL-2 or IL-12 for different times and analyzed for IFN-’y in a Northern blot assay (Fig. 3). The results showed that an increased accumulation of IFN-y cytoplasmic mRNA could be detected at 1 h and its level peaked at 3 h after stimulation with either cytokine. In addition, at 6 h of stimulation the IFN-’y cytoplasmic mRNA was reduced to the background level in the IL-12 but not
IL-2-treated cells. A similar pattern was observed when GM-CSF cytoplasmic mRNA was analyzed (data not shown). These results suggest that IL-12-induced cytokine cytoplasmic mRNA may have a shorter half-life than that induced by IL-2.
Synergistic
effect
of IL-i 2 and lL-2 on IFN-y
induction
Because it has been reported that IL-2 and IL-12 can act synergistically in the induction of IFN-y gene expression [5, 21], we next investigated whether a similar synergistic effect could be observed in the NK3.3 cells. To address this question, both IL-2 (100 U/ml) and IL-12 (10 U/ml) were added to NK3.3 cells and the IFN-y level in the cell supernatant was assayed at different time points (Fig. 4A). The results demonstrated that IL-2 and IL-12 had a strong synergistic activity in IFN-y induction that could be detected as early as 6 h after stimulation. To test whether the synergistic effect was specific for IFN-’y production, expression of cell surface markers and production of GM-CSF were monitored. The results from FACScan analysis indicated that compared with IL-2treated cells there is no significant change in IL-2Ra,
A
B 800
.!.500 400
z
300
184
200 100
0
1
100
10
IL-2
0
500
3
( U/mi)
6
9
Hour
C
12
after
IL-2
15
24
48
72
stimulation
D CoQtrot
-
1L-2
‘I. ‘SI 481
81
121
8) 41
318 04
‘I
61 .
1.
184 Cl)
E
248
4)
128 S ‘I
38
Coatrol
I
Fig. 1. Effect were harvested
ofIL-2 for
on NK.3.3 cells. (A) NKS.3 IFN-y ELISA. Each point
cells were represents
treated with varyingdosages the mean value of results
ofIL-2, as indicated from three different
IL-2
in the x-axis experiments.
for 24 h, and the cell (B) The cells were
supernatants treated with
U/mi for different times, as indicated in the x-axis, then the cell supernatants were harvested for IFN-y ELISA. Each point represents the of results from three different experiments. (C) The cells were treated with IL-2 at 100 U/mi for 24 h then harvested for flow cytometry analysis as described in Materials and Methods. Mean intensity of fluorescence is used to express marker density. Each bar represents the mean value of results from two different experiments. (D) GM-CSF levels were measured in culture supernatants after treatment with IL-2 at 100 U/mI for 24 h. Each bar represents the mean value of results from two separate experiments. IL-2
at 100
mean value
Ye et al. Characterization
of IFN-y
production
by NKS.3
cells
227
A
B
18
2ll
5. ill
‘.
41
a
121
31
z184
z
21
184 41
I I 1.11
1.1
IL-12
1
dose
II
24
4$
Stimulation
time
72
18
(Hour)
(Unit/ml)
C
D
-
Co.troI II
12
II.
IL-2
treated III
#{149}
(I)
1
48
Dl
__ Dl
4
3
184
1..
Cl) 24 121
11 12
IL-12
Coatrol
IL-2Ra
Fig. 2. Effects supernatants
of were
IL-12 on harvested
CDSI
CDI,
the x-axis for 24 h, and the cell experiments. (B) The cells were treated with IL-12 at 10 U/mI for different times, as indicated in the x-axis, then the cell supernatants were harvested for IFN-’y ELISA. Each point represents the mean value of results from three different experiments. (C) The cells were treated with IL-12 at 10 U/mi for 24 h then harvested for flow cytometry analysis as described in Materials and Methods. Mean intensity of fluorescence is used to express marker density. Each bar represents the mean value of results from two different experiments. (D) GM-CSF levels were measured in culture supernatants after treatment with IL-12 at 10 U/mI for 24 h. Each bar represents the mean value of results from two separate experiments.
CD56, and CD69 However, synergy GM-CSF induction
NK3.3 cells. (A) The cells were treated for IFN-y ELISA. Each point represents
in the IL-2/IL-12-treated between IL-2 and IL-12 (Fig. 4C).
cells (Fig. was observed
with the
4B). in
varying dosages of IL-12, mean value of results from
Sti
as indicated in three different
m uIationHour-”
Mechanism
of synergistic
IL-2
IL-12
I
0
1
1
3
6
I
F
1
3
6
effect of IL-2 and IL-12
To determine
whether the increased expression of IFN-y by the combination of IL-2 and IL-12 was observed at the RNA level, we compared cytoplasmic mRNA levels and mRNA half-life in cells treated by IL-2, IL-12, or a combination of both ILs. As seen in Figure 5A, the combined stimulation resulted in an additive mRNA level at 3 h after treatment. Furthermore, analysis of IFN-y mRNA half-life (Fig. 5B) showed that the combined treatment significantly increased the IFN-’y cytoplasmic mRNA half-life. This is consistent with results seen in T cells [5, 22]. Thus, the increased IFN-’y protein levels induced by the combined IL-12/IL-2 stimulation is due, at least in large part, to an increased accumulation and stability of IFN-y mRNA.
IFN-y--
induced
228
Journal
of Leukocyte
Biology
Volume
58, August
1995
fl-Actin 1234567 Fig. 3. Induction of IFN-y were treated with IL-2 (100 as indicated at the top of analysis of IFN-y mRNA as
cytoplasmic
mRNA
by IL-2 or IL-12.
U/mI) or IL-12 (10 U/mI) for different each lane, then harvested for Northern described in Materials and Methods.
The
cells times, blot
A
accumulation of IFN-y cytoplasmic mRNA induced by phorbol myristate acetate (PMA) plus ionomycin (Fig. 7, lanes 7 and 8). This result suggests that the calcium-regulated phosphatase calcineurin [30] may not be directly involved in the IL-2 signaling pathway. Cycloheximide did not affect IFN-y cytoplasmic mRNA accumulation induced by IL-2 (data not shown), indicating that new protein synthesis is not required for IL-2-induced IFN-y cytoplasmic mRNA accumulation. The effects of inhibitors on IL-2-induced IFN-’y protein were confirmed at the cytoplasmic mRNA level (Fig. 7, lanes 1-6). Staurosporine completely blocked IFN-’y mRNA accumulation, whereas genistein and TGFpartially inhibited mRNA accumulation. In contrast to the results observed when IL-2 was used as the inducing agent, staurosporine A only partially inhibited IL-12-induced IFN-y cytoplasmic mRNA (Fig. 7, lane 12) and genistein had no obvious inhibitory effect (Fig. 7, lane 1 1) on IL-12-induced IFN-’y cytoplasmic mRNA levels. However, similar to that observed with IL-2, TGFpartially inhibited and CsA did not inhibit IL-12 induction of IFN-’y cytoplasmic mRNA (Fig. 7, lanes 10 and 13). These data strongly indicate that IL-2 and IL-12
“I
1200
.960
E
-
L-2
--
IL-2+L-12
-.-
IL-12
720
0
480
.-
z 184 -
240
0
3
6
Hours
9
after
12
24
15
stimulation
B 1311
1141 61 -
c1
61
IL.2
781
61
-
IL.2+1L12
521
.; 61
C’b11 C\l
211
I
I
+
C’1
-
I
A IFN-y
Fig.
4. Synergistic
1L.25a
effects
with IL-2 (100 U/mi) in the x-axis, Each point
experiments.
and the represents
CD*
and
-CSE
IL-12.
(10 U/mI)
(A)
cells were
value
treated
the supernatants cells were analyzed
of results
from
NKS.3
for different
cell supernatants were the mean value of
(B) The
the mean
of IL-2
andIL-12
(10 U/ml) for 24 h then GM-CSF ELISAS and the
represents
CL)S6
cells
were
times,
treated
as indicated
harvested for IFN-y ELISA. results from three separate
with IL-2 (100
U/mi)
and IL-12
were harvested for by flow cytometry.
two separate
IFN-y Each
and bar
3-Actin---’
experiments.
B Effects of inhibitors
on lL-2 and IL-i 2 induction
of IFN-y
To further compare the effects ofIL-2 and IL-12 on NK3.3 cells, we used a number of signaling inhibitors, including CsA, TGF[23], genistein [24, 25], staurosporine A [25, 26], and cycloheximide to analyze signaling pathways of IL-2 and IL-12. As shown in Figure 6, IL-2 induction of IFN-’y IL-2Ra, and CD56 could be strongly inhibited by 2 x 10-8 M staurosporine A (Fig. 6A-C), a commonly used protein kinase C (PKC) inhibitor. However, under the same conditions expression of CD69 was not affected (Fig. 6D), indicating that the suppression of gene expression is not due to a nonspecific cytotoxicity of staurosporine. When the cells were treated with genistein, induction of all the tested proteins was suppressed by this tyrosine kinase inhibitor (Fig. 6A-D), confirming that IL-2 signal transduction is tyrosine kinase dependent. This is consistent with the recent reports of association of specific tyrosine kinase molecules with the IL-2 receptor complex [27-29]. TGFwas able to partially inhibit IL-2-induced expression of IFN-’y and IL-2Rct but did not suppress GM-CSF, CD56, and CD69. It is interesting that CsA did not exhibit a significant suppressive effect on the IL-2-induced expression of any the proteins but it did block
4) Dl
80 4) U
-e--
L-2+IL-12
60 4) -..4)
40
....
IL-2
IL-12
4.)
.
20
........
4) .
. #{149}#{149}..
U
0 0
0.5
1
1.5
Time
2
2.5
3
3.5
(Hours)
Fig. 5. IFN-ycytoplasmic mRNA analysis after induction by IL-2, 11,12, or 11,2 and IL-12. (A) The cells were stimulated with II2 (100 U/mi), IL-12 (10 U/ml), or IL-2 (100 U/mi) and IL-12 (10 U/mi), as indicated at the top of each bar, for 3 h then harvested for IFN-y cytoplasmic mRNA
analysis.
(B) The
cells were
treated
under
the same
conditions
as stated
in A for 3 h then treated with 5 tg/ml of Act D for different times as indicated in thex-axis. IFN-’ymRNA was quantitated by usinga densitometer at each time point and corrected for fractin mRNA levels that did not change during the time course of this experiment.
Ye et al. Characterization
of IFN-’y production
by NK3.3
cells
229
have
distinct,
yet
overlapping
signal
transduction
path-
ways for the induction of IFN-y. Similar effects of these inhibitors were observed when GM-CSF mRNA was analyzed (data not shown), indicating that induction of IFN-y and GM-CSF by IL-2 or IL-12 may share the same signal transduction pathway.
Relationship production In
between
cell
proliferation
and IFN-y
addition
to the induction of IFN-y and GM-CSF cytomRNA and protein, IL-2 also induces NK3.3 cells to proliferate [15]. Because it has been reported that in the T cells IL-2-induced proliferation does not depend on PKC [31], we tested the effects of PKC inhibitors on the proliferation of NK3.3 cells. We observed that both IL-2
plasmic
and IL-12 are able to induce cell proliferation. However, in the presence of staurosporine IL-2-induced proliferation was suppressed by approximately 40% (Fig. 8). These data suggest that IL-2-induced NK cell proliferation may be at least partially PKC dependent. Because our data indicated that IL-2-induced IFN-y production was also inhibited by staurosporine, it was possible that the inhibition of IFN-y production might be due to a suppression of cell proliferation. To test the hypothesis we examined the effect of TGFon cell proliferation. Our results demonstrated that TGF-, which inhibited IFN-y production by 50%, had no suppressive effect on IL-2-induced cell proliferation (Fig. 8). These results indicate that suppression of IFN-’1’ gene expression is probably independent of any effects on cell proliferation.
A
D
ill
II.
4)
I.
41 SI
-
U
41
4)
a-
II
.41 U
21 21
Co.troI
c.A
TGF-I
Ge.
S
Sta.r
C..tr.l
II.
Gsa
Slur
ill
-
41
4)
II
‘41
TGV-S
E
B ..
C.t
55
hi
61 U) ‘9
,;
4.
.
2l
25
41 .41 41
41 61
41 c..tr.a
c.A
TGF-1
G..
Sisar
C..ir.I
TGP-S
G
Slur
Ge.
Slur
F
C ..
C.A
15S
ISI 4)
I)
41
S.
-
0 U
II 4)
II
S II
41
45
a-
4)
a-
‘61 4) U 41
21 41 61
25
41
41 C.A
C.aIr.I
Fig. 6. Effects ofinhibitors A (20 nM) for 1 h then
TGF.I
Ge.
Culul
St..r
Journal
of
Leukocyte
Biology
Volume
58,
August
1995
TGF-3(3 In F the
ng/mi), pretreated
TGV-S
tM), or staurosponne treated with IL-2 (100 U/mi) and IL-12 (10 U/ml) for 24 h in the presence of the inhibitors. At the end of the treatment the cell supernatants were harvested for IFN-y or GM.CSF ELISAs and the cells were harvested for flow cytometry analysis. In each panel the value from a control group that received IL-2 and/or IL-12 but no inhibitor represented the 100% vaiue and a ratio ofthe data from the cells treated with inhibitors versus control was used to express the percentage of inhibition. Each bar represents the mean value of resuits from three individual experiments.
230
on IL-2 gene induction. The cells were pretreated with CsA(I0 ng/mi), treated with IL-2 (100 U/mi) for 24 h in the presence of the inhibitors.
CiA
genistein(30 cells were
present
PMA/ Ionom
IL-2 I
IL-12 U
1
I
a)
0) C
C a-. .-
0 .
Cl)
Cl)
-
a-.
0
0 0.
a)
o_o(
ECl)
4
-.
p
ELl
LI
I
(-‘Cl,,-
U)I-
lFN--
fl-.Actin .
1 Fig. 7. Effect were pretreated
the
same
3
of inhibitors
4 on
with inhibitors,
conditions
with IL-2 (100 U/mI), (10 U/mI). At the mined by Northern
Activation IL-i 2
2
as stated
5
IFN-y
6 cytoplasmic
as indicated in Figure
PMA(10
7
ng/mi)and
8
910111213
mRNA
levels.
The
cells
at the top of each lane under
6, then
followed
ionomycin
by S-h
(1 jig/mI),
treatment
or IL-12
end of the treatment IFN-y mRNA levels were deterblot analysis as described in Materials and Methods.
of nuclear
transcription
here strong evidence that NK3.3 is an IL-12-reNK cell line. Although IL-12 also induced IFN-y and GM-CSF at both cytoplasmic mRNA and protein 1evels, it did not induce cell surface expression of IL-2Ra, CD56, and CD69. This is in contrast to the reported effects ofIL-12 on peripheral blood NK cells (after a 6-day stimulation) [33], suggesting that the NK3.3 cell line may have a more limited capacity to respond to IL-12 or may differ from peripheral blood NK cells in their IL-12 receptor numbers or density. Note that IL-2 and IL-12 synergized in the production of both IFN-y and GM-CSF and this synergy was due, at least in part, to both increased accumulation and enhanced stabilization of cytoplastic mRNA, paralleling what was observed in fresh lymphocytes [5]. This stabilization of cytokine mRNA may be due to some type of interference with or inhibition of nudeases that interact with the AUAA regions present in the 3’-untranslated region of both the GM-CSF and IFN-y mRNAs. The fact that synergy was not observed in expression of the cell surface markers suggests that the mRNAs for these genes may not be stabilized by the treatment. These hypotheses require further investigation. To analyze the signal transduction pathway involved in the induction of gene expression by IL-2 and IL-12, five different inhibitors were used in this report. Of interest was our observation that CsA, a potent inhibitor of cytokme gene expression in lymphoid cells, did not significandy block induction of IFN-y by IL-2, IL-12, or IL-2 and IL-12. In addition, CsA did not block IL-2-enhanced expression of IL-2Ra, CD56, and CD69 expression. As calcineurin is a target of CsA [30], these data suggest that calcineurin may not play an important role in the IL-2 or IL-12 signaling pathway in this cell line. In contrast to the results with GsA, TGF-3, another inhibitor of lymphokine gene expression, partially inhibited IL-2-induced expression of IFN-’y (50%) and IL-2Rcz (40%) but did not affect the induction of GM-CSF or expression of CD56 and CD69 (Fig. 6). It also partially inhibited IL-12-induced IFN-y mRNA accumulation (Fig. 7). This inhibitory effect on NK3.3 cells is comparable to that reported for TGFon T cells in which TGFdecreases anti-CD3 and PMA-induced IFN-y production by approximately 70% [23]. Taken together, these results suggest that both IL-2 and IL-12 signaling pathways for the induction of IFN-’y are sensitive to TGF-f inhibitory activity, and there may be TGFresponse elements in the IFN-y
sponsive
factors by IL-2 and
IL-2 and IL-12 both are able to induce IFN-’y transcription but it is not clear which nuclear DNA binding proteins are activated in NK cells after treatment with either IL-2 or IL-12. EMSA assays were conducted to investigate whether IL-2 or IL-12 can activate specific DNA binding proteins. Analysis was focused on NF-icB and APi because these two nuclear factors are involved in the transcriptional activation of many cytokine genes [32]. EMSA resuits showed that IL-2 is able to induce both NF-icB and APi (Fig. 9, lane 2) in the nucleus of NK3.3 cells but IL-12 did not induce these two factors (Fig. 9, lane 3). In addition, in the presence of IL-2 these two factors were not further induced by IL-12 (Fig. 9, lane 4). Consistent with the supernatant results, staurosporine inhibited but CsA had no effect on the IL-2 induction of these DNA binding protein complexes (Fig. 9, lanes 5 and 6). These results suggest that IL-2- and IL-12-induced transcriptional activation of the IFN-y gene may involve the activation of different panels of nuclear transcription factors.
‘5
52
DISCUSSION NKS.3 is an IL-2-dependent human NK cell line that retains many of the markers and functional properties of peripheral blood NK cells. This cell line is able to produce IFN-’y in response to IL-2 stimulation but this feature was not well studied [16]. With this cell line we have performed detailed analyses of IL-2-induced IFN-y production at both cellular and molecular levels and observed that IL-2 also resulted in the production of GM-CSF (Fig. 1D) and expression of cell surface markers including IL2Ra, CD56, and CD69 (Fig. 1C). These data indicate that the NK3.3 cell line is a useful NK cell model for studying IL-2 gene induction at the molecular level. In addition, we
3,
I-
21
‘3
0 Control
CIA
TGR-5
Ge.
Staur
Fig. 8. Suppression of celi proliferation. The cells were pretreated with inhibitors under the same conditions, as stated in Figure 6, then treated with IL-2 (100 U/mi) or IL-12 (10/mi) for 48 h and harvested for the proliferation assay as described in Materials and Methods. Each bar represents the mean value of cpm from two individual experiments.
Ye et al. Characterization
of IFN-y
production
by NK3.3
cells
231
a) 0 0. U)
0 (\1 ,,
U) -j
+
Cs.’ -
Cq1 I
0 + (‘1
Cq1
1-
I
Cl) +
I
I
I
._J
NF-xB
AP1-’.
2 Fig. 9. Effects were treated harvested for
ofIL-2 under EMSA
and IL-12 the same as described
3
4
5
6
on nuclear factors NF-icB and APi. Cells conditions, as stated in Figure 7, then in Materials and Methods.
gene that can suppress IL-2- and IL-12-induced transcriptional activation. Previous reports have clearly demonstrated that [24] the IL-2 receptor signaling pathway is tyrosine kinase dependent (Fig. 6A) [34-36]. To test the role of tyrosine kinase activity in IL-2 induction of IFN-y gene expression, we used genistein, a tyrosine kinase inhibitor. We observed that IL-2-induced IFN-y cytoplasmic mRNA was at least partially inhibited by genistein. The effects on IL-12 are less clear because little decrease in IL-12-induced IFN‘1 cytoplasmic mRNA was observed. However, the recent report by Bacon et al. that IL-12 activated specific tyrosine kinases in NK3.3 cells suggest that additional tyrosine kinase inhibitors may in fact block the IL-12 signal transduction pathway. The use of staurosporine, a commonly used but not completely specific PKC inhibitor, also demonstrated differences between the IL-2 and IL-12 signaling pathways. IL-2 induction of IFN-y mRNA and protein was completely blocked by staurosporine at 2 x 10 M, a concentration of staurosporine thought to be more specific for PKC mactivation. However, because staurosporine was not able to suppress IL-2-induced CD69 expression (Fig. 6), there may be at least two signaling pathways mediating biological activities ofIL-2: one being PKC dependent and leading to the induction of IFN-’y, IL-2Ra, and CD56 and the other PKC independent, leading to induction of CD69. This
232
Journal
of Leukocyte
Biology
Volume
58,
August
1995
hypothesis is supported by the report that in T cells phytohemagglutinin/PMA-induced expression of IL-2Ra is also PKC dependent [37]. In contrast, IL-12-induced IFN-y was only partially blocked by staurosporine (Fig. 7, lane 12), suggesting that IL-12 induction of IFN-y might also occur through at least two signal pathways, one PKC dependent and the other PKC independent. However, because staurosporine is not uniquely specific for only PKC, this effect may be due to suppression of other signal mediators that are sensitive to staurosporine. It is possible that kinases other than PKC also are inhibited by this drug, and a combined inhibition of these kinases and PKC accounts for the suppressive activity of this drug. In fact, in preliminary experiments using another reported PKC inhibitor, bisindolylmaleimide [38], only a small suppressive effect on IL-2-induced IFN-ywas observed (data not shown). In addition to the differences observed by using metabolic inhibitors, IL-2 and IL-12 also showed differences in the activation of nuclear transcriptional factors. NF-icB and APi are two important nuclear factors involved in the transcription control of many cytokines [30, 32] and their receptors [20]. In T cells it has been reported that IL-2 signal transduction involves the activation of NF-icB [39] and APi [30] but similar information has not been available in NK cells. Here we have found that in NKS.3 cells NF-icB and APi both are induced by IL-2 but not by IL-12 (Fig. 9). This induction can be attenuated by staurosporine, suggesting that activation of these two nuclear factors is associated with activation of PKC, as previously reported [40]. As NF-icB is able to enhance IL-2Ra gene promoter activity [41], activation of this nuclear factor may be responsible for IL-2-induced expression of IL-2Ra. In contrast, IL-12 did not induce any of these two nuclear factors (Fig. 9). As IL-12 did not induce NF-cB, this may explain why IL-12 did not induce IL-2Ra. Because IL-2 and IL-12 both induce IFN-y, we propose that in the regulation of IFN-y gene transcription there might be two groups of response elements: one that enhances transcription in response to IL-2 and the other in response to IL-12. Because induction and inhibition of IFN,r is correlated with the activation and suppression of NF-KB and APi in the IL-2-stimulated cells, these data indicate that NF-icB and APi may be involved in the transcriptional activation of the IFN-y gene by IL-2. This hypothesis is supported by the fact that we have previously reported the presence of a c-Rel binding site in the human IFN-y first intron [42] and that computer analysis of the human IFN-y genomic DNA has revealed the presence of an NF-icB site and multiple APi sites in the promoter (unpublished observations). Experiments to conclusively prove the involvement of these DNA binding proteins in the control of human IFN-y gene expression are currently ongoing in our laboratory.
ACKNOWLEDGMENTS We thank Dr. Jacki Kornbluth, University of Arkansas School of Medicine, for the NKS.3 cell line; Dr. George Ricca, Rorer Biotechnology, for the human IFN-y cDNA; Dr. Stephen Clark, Genetics Institute, for human GM-CSF cDNA; Dr. Stan Wolf, Genetics Institute, for human IL12; and Clinical Immunology Services, PRI/DynCorp, NCI-FCRDC, for cytokine assays. We appreciate Joyce Vincent for editorial assistance and Drs. Marco Cippitell, Pantosh Ghosh, and Antonio Sica for helpful discussions.
REFERENCES 1.
Kobayashi,
M.,
S., Loudon,
Fitz,
L,
Ryan,
R., Sherman,
Identification and (NKSF), a cytokine
M.,
Hewick,
F., Perussia,
R.M.,
Clark,
S.C.,
B., Trinchieri,
purification ofnatural with multiple biologic 170, 827-845.
22.
Chan,
G. (1989)
killer cell stimulatory effects on human
factor lympho-
23.
2.
cytes.J. Exp. Med. D’Andrea, A., Rengaraju,
3.
Aste, M., Chan, S.H., Kobayashi, M., Young, D., Nickbarg, E., Chizzonite, R., Wolf, S.F., Trinchieri, G. (1992) Production of natural killer cell stimulatory factor (interleukin 12) by peripheral blood mononuclear cells.J. Exp. Med. 176, 1387-1398. Chan, S.H., Perussia, B., Gupta, J.W., Kobayashi, M., Pospisil, M.,
Young,
4.
5.
6. 7.
8.
9. 10.
1 1.
H.A., Wolf,
13. 14.
15.
Hall,
S.S. (1994)
AIDS
hope.
17.
19.
20.
Kubin,
S.C., Trinchieri,
M.,
IL-12
Science 263,
holds
promise
against
cancer,
25.
sion tein
G. (1991)
glimmer
26.
27.
28.
of 29.
Afonso,
L.C.C.,
30.
Trinchieri,
G.,
Scharton, Scott,
TM.,
P. (1994) Leishmania
in a vaccine against Locksiey, R.M. (1993) Interleukin bial pathogens. Proc. Nati. Acad.
Vieira,
The adjuvant major. Science
L.Q.,
Wysocka,
effect
M.,
of interleukin-12 235-237.
263,
12 in host defense
against
micro-
USA 90, 5879-5880. Gazzineili, R.T., Hieny, S., Wynn, T.A., Wolf, S., Sher, A. (1993) Interieukin 12 is required for the T-lymphocyte-independent induction ofinterferon yby an intracellular parasite and induces resistance in T-cell-deficient hosts. Proc. Nati. Acad Sci. USA 90, 6115-6119. Tripp, CS., Wolf, S.F., Unanue, E.R. (1993) Interleukin 12 and tumor necrosis factor a are costimulators ofinterferon yproduction
killer
cells in severe
with
IFN-3
and
combined
immunodeficiency
IL-2-induced augmentation Immunol. 141, 3234-3240.
C.M., J.A.
McVicar,
D.W.,
(1995)
Interleukin
Ortaldo,J.R., 12 (IL-12)
Rees, induces
R.C.,
M., D’Andrea, A., Kubin, M., Trinchieri, G. (1994) of natural killer cell stimulatory factor/interieukin-l2 stimuli in the induction of cytokines and cytotoxic
Russell,
SM.,
Johnston,
J.A.,
Noguchi,
M., Kawamura,
M., Bacon,
Friedman, M., Berg, M., McVicar, D.W., Witthuhn, B.A., Siivennoinen, 0., Goldman, A.S., Schmalstieg, F.C., Ihie, J.N., O’Shea,J.J., Leonard, W.J. (1994)Interaction ofIL-2Rf3 andy chains withJakl andJak3: implications for XSCID and XCID. Science 266, 10421045. Miyazaki, T., Kawahara, A., Fujii, H., Nakagawa, Y., Minami, Y., Liu, z.-J., Oishi, I., Siivennoinen, 0., Witthuhn, B.A., Ihle, J.N., Taniguchi, T. (1994) Functional activation of Jail and JakS by selective association with IL-2 receptor subunits. Science 266, 1045-1047. Rao, A. (1994) NF-ATp: a transcription factor required for the coordinate induction of several cytokine genes. Immunol. Today 15, 274-281. Mills, G.B., Girard, P., Grinstein, S., Gelfand, E.W. (1988) Interleukin-2 induces proliferation of T lymphocyte mutants lacking protein kinase C. Cell 55, 91-100. Grilli, M., Chiu,J.J.-S., Lenardo, M.J. (1993) NF-icB and Rei: participants in a multiform transcriptional regulatory system. ml. Rev. CytoL Rabinowich,
H.,
Herberman,
of IL-12 and
R.B.,
34.
36.
37.
38.
39.
40.
41.
molecules on 481-498. Johnston,J.A., Kawamura, Shibuya, K., Ortaldo,J.R.,
purified
human
(1993)
function
natural
killer
Differen-
of cellular cells.
CelL
and
M., Kirken, McVicar,
activation
R.A., Chen, Y.-Q., Blake, T.B., D.W., O’Shea,J.J. (1994) Phosof the Jak-3 Janus kinase in response to
interleukin-2. Nature 370, 151-153. Witthuhn, B.A., Silvennoinen, 0., Miura, 0., Lal, KS., Cwik, C., Liu, E.T., Ihle, J.N. (1994) Involvement of the Jak-3 Janus kinase in signalling by interleukins 2 and 4 in lymphoid and myeloid cells. Nature 370, 153-157. Taniguchi, T., Minami, Y. (1993) The IL-2/IL-2 receptor system: a current overview. Cell 73, 5-8.
loannides, O’Brian, production inhibitor Toullec, Ajakanes, Duhamel,
C.G.,
Freedman,
R.S.,
Liskamp,
R.M.,
Ward,
N.E.,
C.A. (1990) Inhibition ofIL-2 receptor induction and IL-2 in the human leukemic cell linejurkat by a novel peptide of protein kinase C. CelL Immunol. 131, 242-252. D., Pianetti, P., Coste, H., Believergue, P., Grand-Perret, T., M., Baudet, V., Boissin, P., Boursier, E., Loriolle, F., L, Charon, D., Kirilovsky, J. (1991) The bisindolylmale-
imide
GF 109203X
kinase Arima, duced
C.]. Biol. Chem. N., Kuziel, WA, signal transduction
is a potent
and
selective
inhibitor
of protein
266,
15771-15781. Grdina, TA., Greene, W.C. (1992) involves the activation of nuclear expression.J. ImmunoL 149, 83-91. Shirakawa, F., Mizei, S.B. (1989) In vitro activation and translocation of NF-icB catalyzed by cyclic AMP-dependent kinase and protein kinase C. MoL CelL BioL 9, 2424-2430. B#{246}hnlein, E., Lowenthal,J.W., Siekevitz, M., Ballard, D.W., BR., Greene, W.C. (1988) The same inducible nuclear
regulates 42.
T.L.
and
152,
phorylation 35.
Whiteside,
IL-2 on expression
adhesion
ImmunoL
phos-
of multiple cytokine gene expression and IRF-1 mRNA by flavone acetic acid in a murine macrophage cell line. CelL Immunol. 157, 21 1-222. Ye,J., Ghosh, P., Cippitelli, M., Subieski,J., Hardy, K.J., Ortaldo,J.R, Young, HA. (1994) Characterization ofa silence regulatory element in the human interferon-y promoter. J. Biol. Chem. 269, 25728-25732. Tan, T.-H., Huang, G.P., Sica, A., Ghosh, P., Young, H.A., Longo, D.L., Rice, N.R. (1992) icB site-dependent activation of the interleukin-2 receptor a-chain gene promoter by human c-rd. Mol. Cell.
1039-1043. Tamaoki, T. (1991) Use and specificity of staurosporine, UCN-O1, and calphostin C as protein kinase inhibitors. Methods EnzymoL 201, 340-347. Boussiotis, V.A., Barber, DL, Nakai-ai, T., Freeman, G.J., Gribben, J.G., Bernstein, G.M., D’Andrea, A.D., Ritz,J., Nadler, LM. (1994) Prevention ofT ccii anergy by signaling through the y chain of the IL-2 receptor. Science 266, 1039-1042.
tial effects
O’Shea,J.J.,
tyrosine
involvement of proEur. j ImmunoL 23,
143, 1-62. 33.
of human
phorylation of JAK2 and TYK2: differential use of Janus family tyrosine kinases by IL-2 and IL-12.J. Exp. Med. 181, 399-404. Eader, L.A., Gusella, L, Dorman, L., Young, HA. (1994) Induction
Aste-amerzaga, Cooperation with other
32.
mice
natural killer cell function.]. Hayakawa, K., Salmerson, M.A., Kornbluth,J., Bucana, C., Itoh, K. (1991)The role ofIL-4 in proliferation and differentiation of human natural killer cells: study ofan IL-4-dependent versus an IL-2-dependent natural killer cell clone.j ImmunoL 146, 2453-2460.
Bacon,
31.
Sri.
with listeriosis, and interleukin 10 is a physiologic antagonist. Proc. Nati. Acad. Sri. USA 90, 3725-3729. Scott, P. (1993) IL-12: Initiation cytokine for cell-mediated immunity. Science 260, 496-497. Trinchieri, G. (1993) Interleukin-12 and its role in the generation of THI cells. Immunol. Today 14, 335-338. Kornbluth,J., Flomenberg, N., Dupont, B. (1982) Cell surface phenotype ofa cloned line ofhuman natural killer cells.]. Immunol. 129, 2831-2837. Kornbluth, J., Hoover, R.G. (1988) Changes in gene expression
on human natural killer cells: differential kinase C and protein tyrosine kinases.
CM.,
1685-1686.
Biol. 12, 4067-4075. 21.
24.
CL, Edington, H.D., McKinney, T.G., Tahara, H., Nalesnik, M.A., Brunda, M.J., Gately, M.K., Wolf, S.F., Schreiber, R.D., Storkus, W.J., Lotze, MT. (1994) Recombinant IL-12 administration induces tumor regression in association with IFN-’y production. J. Immunol. 153, 1697-1706.
Johnston,
18.
D., Clark,
Chehimi,J.,
Nastala,
associated 16.
S.F., Young,
N.M.,
Induction ofinterferon ‘yproduction by natural killer cell stimulatory factor: characterization ofthe responder cells and synergy with other inducers.J. Exp. Med. 173, 869-879. Schmitt, E., Hoehn, P., Hueis, C., Goedert, S., Palm, N., Rude, E., Germann, T. (1994) T helper type 1 development of naive CD4 T cells requires the coordinate action ofinterleukin-12 and interferon-r and is inhibited by transforming growth factor-(. Eur.J. Immunol. 24, 793-798. Chan, S.H., Kobayashi, M., Santoli, D., Perussia, B., Trinchien, G. (1992) Mechanisms ofIFN-y induction by natural killer cell stimulatory factor (NKSF/IL-12): role of transcription and mRNA stability in the synergistic interaction between NKSF and IL-2.J. Immunol. 148, 92-93.
by natural
12.
M., Vaiiante,
cell-associated moiecules in human T and NK cells. Cell. Immunol. 156, 480-492. Nagy, E., Buhlmann, J.E., Henics, T., Waugh, M., Rigby, W.F.C. (1994) Selective modulation of IFN-y mRNA stability by IL12/NKSF. Cell. Immunol. 159, 140-151. Hoiter, W., Kalthoff, F.S., Pickl, W.F., Ebner, C., Majdic. 0., Kraft, D., Knapp, W. (1994)Transforminggrowth factor-a inhibitslL-4 and IFN-y production by stimulated human T cells. Ira. ImmunoL 6, 469-475. Akiyama, T., Ishida, J., Nakagawa, S., Ogawara, H., Watanabe, S., Itoh, N., Shibuya, M., Fukami, Y. (1987) Genistein, a specific inhibitor oftyrosine-specific protein kinases.j Biol. Chem. 262, 5592-5595. Borrego, F., Pena, J., Solana, R. (1993) Regulation of CD69 expres-
mitogen
activation
ofboth
the interleukin-2
IL-2-inNF-icli nudear protein Franza, proteins
receptor-alpha
1 HIV. Cell 53, 827-836. T.-H., Rice, N., Kretzschmar, M., Ghosh, P., Young, HA. (1992) The c-rd protooncogene product c-Rd but not NFKB binds to the intronic region of the human interferon-y gene at a site related to an interferon-stimulable response element. Proc. NaIL Acad. Sci. USA 89, 1740-1744. gene Sica,
and type A., Tan,
Ye et al. Characterization
of IFN-y production
by NK3.3
cells
233
