Interferon- receptor 2 expression as the deciding factor in ... - CiteSeerX

5 downloads 0 Views 458KB Size Report
INTRODUCTION. Interferon (IFN)- is a lymphokine produced by activated T ...... Garotta, G. (1997) The interferon gamma receptor: a paradigm for the multichain ...
Interferon-␥ receptor 2 expression as the deciding factor in human T, B, and myeloid cell proliferation or death Paola Bernabei,*† Eliana M. Coccia,‡ Laura Rigamonti,*† Marita Bosticardo,*† Guido Forni,*† Sidney Pestka,§ Christopher D. Krause,§ Angela Battistini,‡ and Francesco Novelli*† *Department of Clinical and Biological Sciences, University of Turin, I-10043 Orbassano, †Centro Ricerche di Medicina Sperimentale, S. Giovanni Battista Hospital, I-10126 Turin, and ‡Immunology and Virology Laboratories, Istituto Superiore di Sanita´, I-00161 Rome, Italy; and §Department of Molecular Genetics and Microbiology, UMDNJ, Piscataway, New Jersey

Abstract: The heterodimeric interferon (IFN)-␥ receptor (IFN-␥R) is formed of two chains. Here we show that the binding chain (IFN-␥R1) was highly expressed on the membranes of T, B, and myeloid cells. Conversely, the transducing chain (IFN-␥R2) was highly expressed on the surfaces of myeloid cells, moderately expressed on B cells, and poorly expressed on the surfaces of T cells. Differential cell membrane expression of IFN-␥R2 determined the number of receptor complexes that transduced the IFN-␥ signal and resulted in a different response to IFN-␥. After IFN-␥ stimulation, high IFN-␥R2 membrane expression induced rapid activation of signal transducer and activator of transcription-1 (STAT-1) and high levels of interferon regulatory factor-1 (IRF-1), which then triggered the apoptotic program. By contrast, low cell membrane expression resulted in slow activation of STAT-1, lower levels of IRF-1, and induction of proliferation. Because the forced expression of IFN-␥R2 on T cells switched their response to IFN-␥ from proliferative to apoptotic, we concluded that the surface expression of IFN-␥R2 determines whether a cell stimulated by IFN-␥ undergoes proliferation or apoptosis. J. Leukoc. Biol. 70: 950 –960; 2001. Key Words: human 䡠 IFN-␥ 䡠 IFN-␥ receptor 䡠 signal transduction 䡠 apoptosis

INTRODUCTION Interferon (IFN)-␥ is a lymphokine produced by activated T lymphocytes and natural killer cells that plays important roles in host defense mechanisms by exerting antiviral, antineoplastic, immunoregulatory, and proinflammatory activities on a wide range of cell types [1] through its interaction with a heterodimeric receptor (IFN-␥R). The IFN-␥R complex consists of two chains, namely the IFN-␥R1 chain (also called ␣ chain) that binds IFN-␥ and the IFN-␥R2 chain (also called accessory factor-1 or ␤ chain) that transduces signals to the nucleus [2]. Interaction of IFN-␥ with its receptor results in phosphorylation of Janus kinase 1 and 2, which mediate acti950

Journal of Leukocyte Biology Volume 70, December 2001

vation via tyrosine phosphorylation of the signal transducer and activator of transcription-1 (STAT-1) [2, 3]. The phosphorylated STAT-1 homodimer translocates to the nucleus, where it binds to the cis-acting DNA response element SBE (STATbinding element) in the promoters of IFN-␥-stimulated genes to induce transcription [2– 4]. Several primary responsive genes are themselves transcription factors required for induction of secondary components of the cellular response to IFN-␥. Among them, the interferon regulatory factor-1 (IRF-1) [5] is transcriptionally regulated by both SBE and ␬B sites [6, 7] and is in turn responsible for the activation of IFN-␥-responsive genes [4]. Less clear is how the IFN-␥-induced genes modulate all the pleiotropic activities elicited by IFN-␥, including cell proliferation and differentiation [8]. In particular, the interaction between IFN-␥ and the IFN-␥ receptor (IFN-␥R) complex triggers different behaviors in target cells. It has been shown, in fact, that IFN-␥ can induce either apoptosis or proliferation in hematopoietic precursor cells [9 –12], myeloid cells [13], B cells [14, 15], and T cells [16 –18]. A correlation exists between differential expression of the IFN-␥R2 chain and the delivery by IFN-␥ of proliferative or apoptotic signals [19 –21]. In addition, it has been reported that many inhibitory effects of IFN-␥ on cell growth are mediated by the transcriptional factor IRF-1. When IRF-1 is inhibited, IFN-␥ activates proliferative signals, whereas when IRF-1 expression is increased, IFN-␥ activates apoptotic signals [22, 23]. In this respect, caspase-1 [formerly termed interleukin (IL)-1␤-converting enzyme], a molecule involved during the induction and effector phases of programmed cell death, is the target of IRF-1 [24, 25]. We investigated whether the differential expression of the two IFN-␥R chains on T, B, and myeloid cell lineages is responsible for IFN-␥’s ability to differently induce STAT-1 activation and cell proliferation or death. Here we show that low IFN-␥R2 chain expression on the T cell membrane is sufficient to elicit a functional response to IFN-␥, such as enhanced major histocompatibility complex

Correspondence: Francesco Novelli, Dipartimento di Scienze Cliniche e Biologiche, Universita´ di Torino, Ospedale S. Luigi Gonzaga, 10043 Orbassano, Italy. E-mail: [email protected] Received January 18, 2001; revised July 9, 2001; accepted August 6, 2001.

http://www.jleukbio.org

(MHC) class I expression, but not to trigger the apoptotic pathway; cell proliferation, on the other hand, is stimulated. Conversely, when T cell surface expression of IFN-␥R2 chain is up-regulated after serum starvation or overexpressed after transfection, IFN-␥ induces apoptotic signals. In conclusion, these data demonstrated that differences in IFN-␥R2 membrane expression govern the way in which target cells physiologically respond to IFN-␥.

MATERIALS AND METHODS Chemicals and reagents RPMI 1640 was from BioWhittaker (Walkersville, MD); fetal calf serum (FCS), penicillin, streptomycin, gentamycin, and trypan blue dye were from Life Technologies (Grand Island, NY); paraformaldehyde, 2-mercaptoethanol, EDTA, EGTA, Tween 20, propidium iodide (PI), HEPES, KCl, phenylmethylsulfonyl fluoride (PMSF), dithiothreitol (DTT), leupeptin, glycerol, NaCl, NaF, sodium azide, sodium dodecyl sulfate, sodium molybdate, sodium orthovanadate, sodium carbonate, Nonidet P-40, glycine, Ponceau S, urea, PIPES, proteinase K, RNase A, RNase T1, formamide, rabbit anti-actin polyclonal antibodies, and Tris were from Sigma Chemical Co. (St. Louis, MO); Ficoll type 400 was from Pharmacia (Uppsala, Sweden); 24-well plates and cell culture flasks were from Corning Costar Corp. (Cambridge, MA); isopropanol, methanol, dimethyl sulfoxide, phosphate-buffered saline (PBS), bovine serum albumin (BSA), ethanol, phenol, chloroform, and bromophenol blue were from Merck Chemicals (Milan, Italy); acrylamide, N,N⬘-methylenebisacrylamide, ammonium persulfate, and N,N,N⬘,N⬘-tetramethylethylenediamine (TEMED) were from Bio-Rad (Hercules, CA); streptavidin-phycoerythrin (PE), isotypenegative control mouse immunoglobulin (Ig) G2a, IgG1, fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse Ig, and PE-conjugated anti-human CD14, CD19, CD16/CD56, and CD3 were from Becton Dickinson (Mountain View, Ca); rabbit anti-IRF-1, anti-caspase-1 polyclonal antibodies, and horseradish peroxidase-conjugated goat anti-rabbit IgG were from Santa Cruz Biotechnology (Santa Cruz, CA); anti-MHC class I W6.32 was kindly provided by G. Trinchieri (The Wistar Institute, Philadelphia, PA); biotinylated rabbit anti-mouse Ig, PE-conjugated anti-human CD33, and CD34 were from Dako (Glostrup, Denmark).

Media The culture medium was RPMI 1640 supplemented with penicillin, streptomycin, gentamycin, 2.5 ⫻ 10⫺4 M 2-mercaptoethanol, and 10% FCS (referred to hereinafter as complete medium). All the in vitro cultures were performed at 37°C in a humidified 5% CO2 atmosphere.

IFN-␥ and monoclonal antibody (mAb) to IFN-␥R The human recombinant IFN-␥ ⌬10 and mAb to IFN-␥R were produced at Hoffmann-La Roche, Basel, Switzerland. IFN-␥ ⌬10 (108 U/mg of protein) contains the NH2 terminal MQDP and lacks the COOH terminal 10 amino acid residues. Mouse mAb ␥R99 is an IgG1 that specifically interacts with the extracellular domain of human IFN-␥R1 and inhibits the binding of IFN-␥ [19]. Mouse mAb C.11 is an IgG2a that specifically interacts with the extracellular domain of human IFN-␥R2 [19].

Peripheral blood mononuclear cells (PBMCs) and malignant cells PBMCs from heparinized venous blood from healthy donors or from a patient with non-Hodgkin lymphoma were isolated by density gradient centrifugation with Ficoll type 400. ST4 T cells (CD1⫹, CD2⫺, CD3⫺, CD4⫺, CD8⫹, and CD25⫺) display large irregular nuclei with deep indentations typical of childhood, convoluted-type T cell lymphoma; PF382 is a human T-acute lymphoblastic leukemia (CD1⫹, CD2⫺, CD3⫺, CD4⫺, CD8⫹, and CD25⫺) stabilized both in vitro and in nu/nu mice starting from biopsy material [26]. Molt-4 (CD1⫹, CD2⫺, CD3⫹, CD4⫹, CD8⫹, and CD25⫺) [American Type Culture Collection (ATCC, Rockville, MD) CRL1582] and Jurkat (CD1⫹, CD2⫹,

CD3⫹, CD4⫹, CD8⫺, CD25⫺) (ATCC CRL8161) are human T cells from acute lymphoblastic leukemia. HL-60 (ATCC CCL240) is a promyelocytic cell line; U937 (ATCC CRL1593) a promonocytic line; THP-1 (ATCC TB202) is a monocytic line. Ramos (ATCC CRL1596), Namalwa (ATCC CRL1432), Daudi (ATCC CCL213), and Raji (ATCC CCL86) cells are human B lymphocytes from patients with Burkitt’s lymphoma; RPMI 8866 is a human Epstein-Barr virus-transformed B cell line and was kindly provided by G. Trinchieri (The Wistar Institute, Philadelphia, PA).

Flow cytometry Malignant cells were recovered, washed twice in cold PBS supplemented with 0.2% BSA and 0.1% sodium azide, and stained for surface protein with ␥R99 and C.11 mAbs, followed by biotinylated rabbit anti-mouse Ig and streptavidin-PE. All labeling steps were followed by incubation for 30 min at 4°C and were separated by two washes with cold PBS supplemented with 0.2% BSA and 0.1% sodium azide. PBMCs were recovered, washed, and simultaneously stained with FITC-conjugated ␥R99 or FITC-conjugated C.11 and PE-conjugated anti-CD14, CD19, CD16/CD56, CD3, CD33, or CD34 mAbs. For simultaneous staining of IFN-␥R1 and IFN-␥R2, cells were recovered and stained with FITC-conjugated C.11 mAb and biotinylated ␥R99 mAb followed by streptavidin-PE. To evaluate IFN-␥-induced MHC class I antigen expression, cells were cultured in complete medium in the absence or presence of 100 U/mL of IFN-␥. At 72 h, cells were recovered, washed, and stained for MHC class I expression with mAb W6.32, followed by FITC-conjugated goat antimouse Ig. For intracytoplasmic detection of the two IFN-␥R chains and analysis of DNA content, 106 cells were stained with unconjugated ␥R99 or C.11 to block the membrane-bound chains, fixed and permeabilized as previously described [19, 20], and stained with FITC-conjugated ␥R99 or FITC-conjugated C.11, or with 1 mL of PBS supplemented with 2% FCS and 0.1% NaN3 containing 25 ␮g/mL of PI and 11.25 Kunitz U of RNase for at least 30 min. Membrane, cytoplasmic expression, and DNA content were analyzed with a FACScan flow cytometer (Becton Dickinson). Each plot represented the results from 10,000 events.

Cell proliferation assay Malignant cells (0.5⫻106/mL) were cultured in 24-well microtest plates in 2 mL of complete medium supplemented or not with serum and IFN-␥. Transfected IFN-␥R2 T cells (0.5⫻106/mL) were cultured in 24-well microtest plates in 2 mL of complete medium supplemented with 100 U/mL of IFN-␥. At the indicated times, a small aliquot of the cell suspension was removed and mixed with an equal volume of trypan blue-containing solution, and viable cells were counted. The results are expressed as the arithmetic mean ⫾ SD of cell numbers from triplicate cultures. The experiments were performed independently at least three times and representative results are shown below.

DNA transfection pcDNA3 is a mammalian expression vector with the promoter of cytomegalovirus (CMV) and neomycin-resistance genes. PcDNA3␥R2 was constructed by inserting human IFN-␥R2 cDNA into its KpnI and EcoRI sites. Plasmid DNAs were transfected into the cells by the lipofectamine procedure (Gibco-BRL, Gaithersburg, MD). Briefly, 2 ⫻ 106 cells were mixed with 2 ␮g of plasmid and 20 ␮L of lipofectamine and incubated for 5 h at 37°C in a humidified 5% CO2 atmosphere. Next, 4 mL of medium containing 10% FCS were added. After 72 h, the transduced cells were selected in 1 mg/mL of G418-containing medium (geneticin sulfate; Gibco-BRL) for 4 weeks. Neomycin-resistant cells were maintained in medium containing 0.5 mg/mL of G418 for more than 1 month before further experimentation.

Western blot analysis Treated cells (5⫻106) were washed twice in cold PBS and then collected by centrifugation. Total and nuclear proteins (25 or 30 ␮g of protein) were extracted as previously described [21] and separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis at 140 V on 8% miniprotein gels. Gels were electroblotted onto a polyvinylidene fluoride membrane (Bio-Rad, Hercules, CA) at 100 V for 1 h, and the equality of the amount of protein analyzed was checked by nonspecific staining with Ponceau S. The membranes were blocked with TTBS [20 mM Tris-HCl (pH 7.5), 500 mM NaCl, and 0.05%

Bernabei et al. IFN-␥R2 distribution on human hematopoietic cells

951

Tween 20] and 5% nonfat dry milk overnight and then incubated with a 1:1,000 dilution of anti-IRF-1 or anti-caspase-1 rabbit polyclonal antibodies or with a 1:200 dilution of anti-actin rabbit polyclonal antibody. After washing with TTBS, blots were reacted with 1:2,000 horseradish peroxidase-conjugated goat anti-rabbit IgG antibody. Antibody reactions were visualized by enhanced-chemiluminescence reagents according to the manufacturer’s instructions (ECL plus; Amersham International, Bucks, United Kingdom). Actin was used as a control for equal protein loading. Fold increases of treated cells relative to untreated cells were quantitated after normalization with actin. Density scanning was performed using The University of Texas Health Science Center (San Antonio, TX) ImageTools for Windows 2.0.

RNase protection assay Total cellular RNA was extracted with RNAzol solution (Cinna/Biotech, Houston, TX). Total RNA (5–10 ␮g) was hybridized for 18 h to the RNA probes (3 ⫻

105 counts per minute) at 55°C in 25 ␮L of 80% formamide, 0.4 M NaCl, 40 mM PIPES (pH 6.8), and 1 mM EDTA. Subsequently, samples were incubated with RNase A (40 ␮g/mL) and RNase T1 (1 ␮g/mL) for 1 h at 33°C and then subjected to proteinase K digestion, phenol-chloroform extraction, and ethanol precipitation. Gel electrophoresis was performed on standard 8% polyacrylamide 8 M urea sequencing gels. To obtain the pBS IRF-1 construct, the plasmid pUC IRF-1 was digested with SmaI, and the 400-bp-long fragment was cloned into the same sites of pBleuscript/KS (Stratagene, La Jolla, CA). To generate the 32P-labeled 280-bp-long antisense IRF-1 RNA probe, the plasmid pBS IRF-1 was linearized with EcoRI and transcribed with T7 polymerase.

Electrophoretic mobility shift assay (EMSA) EMSA was performed as described by Kotenko et al. [27]. Briefly, cells resuspended to a concentration of 107/mL were incubated in a water bath at 37°C in the presence or absence of 100 U/mL of IFN-␥. At the appropriate

Fig. 1. IFN-␥R1 and IFN-␥R2 chain expression in malignant T, B, and myeloid cells. (A) Membrane and cytoplasmic expression of IFN␥R1 and IFN-␥R2 on four T (ST4, PF382, Molt-4, and Jurkat), five B (Namalwa, Raji, Ramos, Daudi, and RPMI 8866), and three myeloid (HL60, U937, and THP-1) cell lines was evaluated by cytofluorimetric analysis. Values, expressed as percentages of positive cells, were calculated by subtracting the positivity of nonspecific fluorescence detected with isotypematched control Ig from that obtained with specific anti-IFN-␥R1 ␥R99 and anti-IFN-␥R2 C.11 mAbs. Results of one representative experiment of three independently performed are shown. (B) Cytofluorimetric analysis of simultaneous membrane expression of IFN-␥R1 and IFN-␥R2 on representative T (ST4), B (Raji), and myeloid (U937) cell lines. Cells were washed and simultaneously stained using specific biotinylated anti-IFN-␥R1 ␥R99 mAb followed by streptavidin-PE and FITC-conjugated anti-IFN-␥R2 C.11 mAb. (C) Cytofluorimetric analysis of constitutive cytoplasmic expression of IFN-␥R1 and IFN-␥R2 on representative T (ST4), B (Raji), and myeloid (U937) cells. Thin lines represent nonspecific fluorescence detected with isotype-matched control Ig. (D) Cytofluorimetric analysis of IFN-␥-mediated induction of MHC class I antigen on representative malignant human T (ST4), B (Raji), and myeloid (U937) cells. Cells (0.5⫻106) were treated (continuous line) or untreated (broken line) with 100 U/mL of IFN-␥. After 72 h, cells were washed and stained for MHC class I antigen expression using W6.32 mAb, followed by FITC-conjugated goat anti-mouse Ig.

952

Journal of Leukocyte Biology Volume 70, December 2001

http://www.jleukbio.org

time intervals, 100 ␮L were removed, immediately diluted into 1.0 mL of ice-cold PBS, and maintained at 4°C in an ice bath. Then each suspension was centrifuged at 1,100 g, the supernatant was removed, and the pellet was

suspended in 25 ␮L of lysis buffer [0.5% Brij-96, 10% glycerol, 0.1 mM EDTA, 50 mM Tris-HCl (pH 8.0), 3 ␮g/mL of aprotinin, 1 ␮g/mL of leupeptin, 1 ␮g/mL of pepstatin, 1 mM Na3VO4, 1 mM DTT, 0.2 mM PMSF, and 150 mM NaCl] and stored at ⫺80°C. After 30 min of incubation on ice, 2.5 ␮L of lysate were added to a mixture of 0.2 mg/mL of polydeoxyinosinic acid-polydeoxycytidylic acid, 1% Ficoll, 4 mM HEPES (pH 7.4), 30 ␮g/mL of BSA, and 0.2 ng of a double-stranded [32P]dCTP-radiolabeled oligonucleotide encoding the SBE element from the IRF-1 gene promoter (5⬘-GATCGATTTCCCCGAAATCATG-3⬘) [25, 28]. After 20 min of incubation, samples were loaded onto a 5% native polyacrylamide gel and electrophoresed for 4 h at 450 V. The vacuumdried gel was then autoradiographed either on X-ray film or with a GS-525 Molecular Analyzer phosphorimager with its own software (Bio-Rad, Hercules, CA). All data analyses were performed with SigmaPlot 4.03 (Jandel Scientific, San Rafael, CA). To lower the scatter in the nonnormalized data, we divided the intensity of the STAT-1-oligonucleotide band by the intensity of a nonspecific band possessing twice the mobility to normalize the STAT-1 signal intensity for differing amounts of lysate added to the binding buffer or to the gel [29, 30]. The zero time point was subtracted from each normalized value to secure an initial zero signal. The experiments were performed independently at least three times. The results were expressed as the arithmetic means ⫾ SD of normalized STAT-1 signal intensity.

Statistical analysis The statistical significance of IFN-␥-induced proliferation or growth inhibition between untreated cells versus treated cells was evaluated by Student’s t-test (GraphPad Prism 3; GraphPad Software, Inc., San Diego, CA). P values of ⬍ 0.05 (*) and ⬍ 0.005 (**) were considered significant.

RESULTS IFN-␥R1 and IFN-␥R2 chain expression of normal and malignant mononuclear cells IFN-␥R chain expression was evaluated on four T (ST4, PF382, Molt-4, and Jurkat), five B (Namalwa, Raji, Ramos, Daudi, and RPMI 8866), and three myeloid (HL60, U937 and THP-1) cell lines. High mRNA transcripts (data not shown) and cytoplasmic protein levels (Fig. 1A–C) of both chains were detected in all of these lines. Striking differences were observed, however, when their membrane expression was evaluated. All cell lines expressed high membrane levels of IFN␥R1. In contrast, T cell lines expressed barely detectable levels of IFN-␥R2, whereas its expression was significantly higher on B cells and even higher on myeloid cells (Fig. 1A–B). Simultaneous staining with mAbs to IFN-␥R1 and IFN-␥R2 indicated that T cells showed substantial staining with IFN␥R1 antibodies but only 2% of double-positive cells compared with 42% and 85% in B and myeloid cells, respectively (Fig. 1B). When the functional capacity of IFN-␥R was assessed, it was found that IFN-␥ increased MHC class I antigen expression equally in T, B, and myeloid cells (Fig. 1D), thus demonstrating that IFN-␥R is functional irrespective of the membrane density of IFN-␥R2. 4 Fig. 2. IFN-␥R1 and IFN-␥R2 chain expression in different subsets of normal PBMCs and hematopoietic precursor cells. IFN-␥R1 (A) and IFN-␥R2 (B) cell membrane expression on normal peripheral blood CD3⫹, CD14⫹, CD19⫹, and CD16⫹/56⫹ cells and on peripheral blood CD33⫹ and CD34⫹ cells from a patient with non-Hodgkin lymphoma (C) were evaluated by cytofluorimetric analysis. Results of one representative experiment of four independently performed are shown.

Bernabei et al. IFN-␥R2 distribution on human hematopoietic cells

953

Similar IFN-␥R chain expression and distribution were observed in the normal counterparts from PBMCs, all of which expressed high membrane levels of IFN-␥R1 (Fig. 2A), whereas IFN-␥R2 was expressed by few CD3⫹ cells (T lymphocytes) and CD16⫹CD56⫹ cells (natural killer cells), about half of CD19⫹ cells (B lymphocytes), and most CD14⫹ cells (monocytes) (Fig. 2B). Because IFN-␥ can induce either apoptosis or proliferation in hematopoietic precursor cells [9 –12], the surface expression of IFN-␥R chains on these cells was also evaluated. CD33⫹ and CD34⫹ cells expressed significant levels of IFN-␥R1. By contrast, few CD34⫹ and most CD33⫹ cells expressed the IFN-␥R2 chain (Fig. 2C).

Effect of IFN-␥ on proliferation and apoptosis of T, B, and myeloid cells We have previously shown that T lymphocytes expressing high membrane density of IFN-␥R1 respond to IFN-␥ by increasing their proliferation, whereas those expressing high membrane density of both chains are susceptible to IFN-␥-mediated

apoptosis [19, 21]. Because IFN-␥R is still functioning on T, B, and myeloid cells irrespective of membrane density of IFN␥R2, we used a broad IFN-␥ dose response to evaluate the effect of exogenous IFN-␥ on proliferation. This effect was easy to study, because none of the malignant lines secreted IFN-␥ constitutively [21]. ST4, Raji and U937 cells were cultured for 72 h in the absence or presence of IFN-␥ (from 1 to 1,000 U/mL) with or without 50 ␮g/mL of ␥R99, which hampers the binding of IFN-␥ with IFN-␥R1 [19, 26]. Addition of scalar doses of IFN-␥ clearly induced the growth of ST4 cells, whereas it inhibited that of Raji and U937 cells in a dose-dependent manner (Fig. 3A). Both effects were strongly reduced or abolished by ␥R99. Similarly, IFN-␥ increased proliferation of PF382, Molt-4, and Jurkat cells, but it inhibited that of the Ramos, Raji, and Daudi B lines and two myeloid lines (HL60 and THP-1) (data not shown). To investigate whether the IFN-␥-induced inhibition of cell growth could be related to apoptosis, the effect of IFN-␥ addition on the DNA content of the T, B, and myeloid cells was

Fig. 3. Effect of IFN-␥ on proliferation and apoptosis of T, B, and myeloid cells in the presence of serum. (A) ST4, Raji, and U937 cells (0.5⫻106/mL) were cultured in complete medium supplemented or not with 1, 10, 100, or 1,000 U/mL of IFN-␥, or with 100 or 1,000 U/mL of IFN-␥ plus 50 ␮g/mL of anti-IFN-␥R1 ␥R99 mAb. At 72 h, a small aliquot of the cell suspension was removed and mixed with an equal volume of trypan blue-containing solution, and viable cells were counted. Results are expressed as numbers of viable cells ⫻ 10⫺6/mL. The experiments were performed independently at least three times, and results of representative experiments are shown. (B) At the same time, the cells treated or not with 100 U/mL of IFN-␥ were recovered, and apoptotic/hypodiploid cells were evaluated for DNA content by PI staining. Results of one representative experiment of three independently performed are shown.

954

Journal of Leukocyte Biology Volume 70, December 2001

http://www.jleukbio.org

further evaluated. When ST4 cells were cultured for 72 h in the presence of 100 U/mL of IFN-␥, no increase in the apoptotic/ hypodiploid cell population was observed (Fig. 3B, upper panels). By contrast, when Raji or U937 cells were exposed to IFN-␥, there was a significant increase (Fig. 3B, middle and lower panels). In addition, we tested whether the IFN-␥-induced proliferation and apoptosis corresponds to variations in nuclear factor ␬B (NF-␬B) activation. EMSA of NF-␬B activation showed that this transcriptional factor was induced by IFN-␥ in T cells, but not in B and myeloid cells (data not shown). Thus these data suggested that NF-␬B activation was not concurrent with the antiproliferative response, whereas a specific NF-␬B response occurred with an IFN-␥-induced proliferative signal.

Enhanced expression of IFN-␥R2 and IFN-␥induced apoptosis on T, B, and myeloid cells We previously observed that serum or growth factor deprivation up-regulates the expression of both IFN-␥R chains on normal and malignant T cells [16, 19, 26]. To evaluate the influence of serum on IFN-␥R2 chain expression on malignant T cells, ST4 and Jurkat cells were cultured for 24 h in the presence or absence of serum, and the expression of IFN-␥R2 was detected by flow cytometry. In the presence of serum, ST4 and Jurkat T cells displayed low IFN-␥R2 expression (Fig. 4A, left panels), whereas in its

absence (Fig. 4A, right panels) there was a marked increase to levels observed in B and myeloid cells, which are sensitive to the antiproliferative effect of IFN-␥. By contrast, no significant variations in surface IFN-␥R2 expression were observed on Raji and U937 cells cultured in the presence or absence of serum (data not shown). We thus evaluated whether this increase switches the direction of the signal delivered by IFN-␥ to serum-deprived T cells from proliferative to apoptotic. ST4, Jurkat, Raji, and U937 cells were cultured in parallel under the following conditions: (1) medium with serum for 72 h, (2) medium with serum for 24 h followed by medium with serum containing 100 U/mL of IFN-␥ for 48 h, (3) medium without serum for 24 h followed by medium with serum for 48 h, and (4) medium without serum for 24 h followed by medium with serum containing 100 U/mL of IFN-␥ for 48 h (Fig. 4B). Direct cell counts showed that 48 h of exposure to IFN-␥ inhibited the growth of Raji and U937 cells cultured in the presence or absence of serum (Fig. 4B, lower panel). By contrast, IFN-␥ enhanced the growth of ST4 and Jurkat cells cultured in the presence of serum (IFN-␥R2 chain low) but inhibited that of all cells cultured in its absence (IFN-␥R2 chain high) (Fig. 4B, upper panel). Although serum deprivation induces a slow apoptosis of T cell lines [26], in the control condition in which T cells were cultured for 24 h in medium without serum followed by medium with serum for

Fig. 4. Enhanced expression of IFN-␥R2 and IFN-␥-induced apoptosis on T cells cultured with or without serum. (A) ST4 and Jurkat cells were cultured with (left panels) or without (right panels) serum. Cytofluorimetric analysis was performed 24 h later. IFN-␥R2 expression was evaluated with the anti-IFN-␥R2 C.11 mAb. Solid lines, IFN-␥R2 C.11 mAb; broken lines, background of mouse IgG2a as negative control. (B) Representative T (ST4, and Jurkat), B (Raji), and myeloid (U937) cell lines were cultured (0.5⫻106/mL) in parallel under the following conditions: (1) medium without serum (24 h) followed by medium with serum (48 h); (2): medium without serum (24 h) followed by medium with serum supplemented with IFN-␥ (48 h); (3) medium with serum (72 h); and (4) medium with serum (24 h) followed by medium supplemented with IFN-␥ (48 h). At 72 h, a small aliquot of the cell suspension was removed and mixed with an equal volume of trypan blue-containing solution, and viable cells were counted. Results are expressed as the arithmetic means ⫾ SD of numbers of viable cells from triplicate cultures and reported as cell numbers ⫻ 10⫺6/mL. The experiments were performed independently at least three times, and results of representative experiments are shown. Statistical significance of the difference in proliferation between untreated cells and treated cells as evaluated by Student’s t-test is indicated. *, P ⬍ 0.05; **, P ⬍ 0.005.

Bernabei et al. IFN-␥R2 distribution on human hematopoietic cells

955

peak of the response after 20 min was about 80% reduced compared with that of U937 cells. In contrast, when Jurkat cells were cultured in the absence of serum, STAT-1 peaked after 10 –15 min, resembling the response observed in U937 cells (Fig. 5B). Thus these data suggest that concurrent with the antiproliferative response, there was a rapid STAT-1 activation, whereas a slow STAT-1 response occurred with a proliferative signal.

Induction of IRF-1 by IFN-␥ in T, B, and myeloid cells It has been reported that IFN-␥ stimulates or inhibits proliferation of hemopoietic cells, depending on the relative expression of IRF-1, because inhibition of its expression favors their IFN-␥-induced proliferation [22, 23]. We thus tested whether the IFN-␥-induced switch of malignant T cells from proliferation to apoptosis in the absence of serum corresponds to variations in IRF-1 levels. ST4 and Jurkat cells were cultured in the presence or absence of serum. After 24 h, each culture was split and recultured for a further 8 h with or without 100 U/mL of IFN-␥. Total RNA was extracted and analyzed by RNase protection assay with a specific riboprobe for IRF-1. Figure 6A shows that in the presence of serum, IFN-␥ increased IRF-1 mRNA as expected (Fig. 6A, lanes 2, 6). There was an approximately twofold further increase in the absence of serum (Fig. 6B, lanes 4, 8).

Fig. 5. STAT-1 activation kinetics induced by IFN-␥ in cells cultured in the presence or absence of serum. (A) U937 and (B) Jurkat cells were cultured in the presence or absence of serum for 24 h and treated with 100 U/mL of IFN-␥ for the indicated intervals. Quantitative results of the EMSA of STAT-1 activation were calculated as described in Materials and Methods. Results are expressed as the arithmetic means ⫾ SD of normalized STAT-1 signal intensity values from three experiments independently performed.

48 h, neither growth arrest nor apoptosis was observed (data not shown).

Kinetics of STAT-1 activation by IFN-␥ in T and myeloid cells To determine whether the differential expression of IFN-␥R2 chain induces differential STAT-1 activation after IFN-␥ interaction, EMSA was performed on Jurkat and U937 cells treated with IFN-␥ for various times in the presence or absence of serum. Figure 5A shows that in U937 cells, STAT-1 was rapidly activated to a maximum level after 5–10 min, irrespective of the presence or absence of serum. Conversely, in Jurkat cells treated with IFN-␥, different kinetics of STAT-1 activation were observed. When cultured with serum, a slow increase of STAT-1 activation was observed (Fig. 5B). However, the 956

Journal of Leukocyte Biology Volume 70, December 2001

Fig. 6. Analysis of IRF-1 mRNA induction by IFN-␥ in T cells cultured in the presence or absence of serum. ST4 and Jurkat cells were cultured in the presence or absence of serum. After 24 h, each culture was split and recultured for a further 8 h with or without 100 U/mL of IFN-␥. (A) IRF-1 mRNA transcripts were analyzed by RNase protection assay. (B) The results were quantified by scanning autoradiograph from three separated experiments. Values are expressed as mRNA relative fold increases after normalization with GAPDH signal. A representative of three independently performed experiments is shown.

http://www.jleukbio.org

Increased IRF-1 expression and caspase-1 induction in T and myeloid cells We next evaluated the expression of caspase-1, an IFN-␥induced gene product involved in autonomous cell death [13, 24, 31, 32]. Cells were cultured in the presence or absence of serum. After 24 h, the cultures were split and recultured for a further 24 h with or without 100 U/mL of IFN-␥, and proteins were extracted. Western blot analysis showed that IFN-␥ always induced IRF-1 in ST4 cells, much more so in the absence of serum (Fig. 7). IFN-␥ also induced a barely detectable increase of caspase-1 expression in ST4 cells cultured in the presence of serum, whereas in its absence this enhancement was substantially higher (Fig. 7). Similar results were obtained with Jurkat and PF382 cells (data not shown). In U937 cells, the IFN-␥-induced IRF-1 or caspase-1 was not influenced by the presence or absence of serum. Similar results were also obtained with Raji cells (data not shown).

Forced expression of IFN-␥R2- and IFN-␥induced apoptosis on T cells Given that serum-starved cells undergo many changes including growth arrest and increased sensitivity to apoptosis [33– 35], to provide a mechanistic explanation of the extent to which the selective overexpression of IFN-␥R2 chain on T cells renders them susceptible to IFN-␥-apoptotic signals, we transfected Jurkat T cells with a vector containing IFN-␥R2 chain. Jurkat T cells transfected with the control vector pcDNA3 displayed low IFN-␥R2 expression (Fig. 8A, left panel), whereas those transfected with the vector pcDNA3␥R2 containing IFN-␥R2 chain (Fig. 8A, right panel) expressed higher surface levels of IFN-␥R2. Addition of 100 U/mL of IFN-␥ for 48 h to Jurkat pcDNA3␥R2 induced a marked decrease of proliferation (Fig. 8B). By contrast, addition of IFN-␥ to Jurkat

Fig. 7. IRF-1 and caspase-1 induction by IFN-␥ on serum-deprived hematopoietic cells. ST4 and U937 cells were cultured under the following conditions: (1) medium with serum for 48 h; (2) medium with serum for 24 h followed by medium with 100 U/mL of IFN-␥ for 24 h; (3) medium without serum for 48 h; and (4) medium without serum for 24 h followed by medium containing 100 U/mL of IFN-␥ for 24 h. IRF-1 induction was evaluated by Western blot analysis on nuclear cell extracts and caspase-1 induction by Western blot analysis on total cell extracts. Western blot filters were subsequently challenged with an anti-actin antibody to confirm equal protein loading in each lane of the gel. Fold induction of IFN-␥-treated cells relative to untreated cells was quantitated after normalization with actin using UTHSCSA ImageTools for Windows 2.0.

pcDNA3 control cells caused an increase of their proliferation. Jurkat pcDNA3␥R2 showed a faster and higher STAT-1 activation after IFN-␥ interaction compared with Jurkat pcDNA3 (data not shown). As expected, Jurkat pcDNA3␥R2 cells displayed a higher IRF-1 expression and caspase-1 induction in response to IFN-␥ compared with Jurkat pcDNA3 (Fig. 8C). Thus these data demonstrated that high IFN-␥R2 chain expression was the deciding factor in IFN-␥-mediated proliferation or death of human hematopoietic cells.

DISCUSSION In this report we show that among T, B, and myeloid cells, the ligand-binding chain IFN-␥R1 was highly and uniformly expressed on the membrane whereas IFN-␥R2 was highly expressed on the surface of B and myeloid cells but very limited on T cells. Observation of the same pattern in the normal counterpart of PBMCs suggested that cell-type-specific internalization is a physiologic homeostatic function. Moreover, hematopoietic precursor cells displayed an analogous differential IFN-␥R2 distribution, because CD33⫹ cells expressed higher levels than CD34⫹ cells. These observations might account for the proliferative response to IFN-␥ reported in human myeloid leukemia cell lines [36]. All these data suggest that differential membrane expression of the IFN-␥R2 chain has broad ramifications and a role in modulating growth and apoptosis of hematopoietic cells in both physiologic and pathologic conditions. We thus present evidence that differential expression of the IFN-␥R2 chain could be considered the limiting factor determining the number of functional receptor complexes that transduce IFN-␥ signals. In effect, optimal IFN-␥R2 membrane expression might result in prompt IFN-␥-mediated STAT-1 activation and optimally induced IRF-1 expression. This expression pattern could lead to apoptosis of IFN-␥-sensitive cells, e.g., B and myeloid cells. Conversely, cells that express low levels of surface IFN-␥R2 chain functionally might upregulate MHC class I antigen expression and might not respond to apoptotic signals induced by IFN-␥. Because the apoptotic or proliferative response to IFN-␥ in the same cell population was not changed by different doses of IFN-␥, our data indicate that the switch of IFN-␥ response from proliferative to apoptotic was strictly dependent on the density of IFN-␥R2 expression. In human T cells, IFN-␥R2 chain expression is prevalently intracytoplasmic [see above; 16, 37]. This results from the fast and continuous recycling between surface and clathrin-coated vesicles involved in the protein-recycling pattern [37–39]. Through this recycling mechanism a few receptor molecules are continuously expressed on the surface of T lymphocytes and allow a few heterodimeric receptors to be engaged by IFN-␥, resulting in induction of a slow STAT-1 activation, and low levels of IRF-1 without triggering apoptosis (see above; 37). The intracellular traffic of IFN-␥R2 is completely IFN-␥independent, because IFN-␥R2 chain internalization was equally observed in T cells from children with inherited IFN␥R1 gene deficiency and in healthy donors [37]. The T cell lines used in our study were not in the same differentiation

Bernabei et al. IFN-␥R2 distribution on human hematopoietic cells

957

Fig. 8. Forced expression of IFN-␥R2- and IFN-␥-induced apoptosis on transfected IFN-␥R2 T cells. (A) Cytofluorimetric analysis of IFN-␥R2 expression on transfected Jurkat pcDNA3 or Jurkat pcDNA3␥R2. IFN-␥R2 expression was evaluated with the anti-IFN-␥R2 C.11 mAb. Solid lines, IFN-␥R2 C.11 mAb; broken lines, background of mouse IgG2a as negative control. (B) Transfected Jurkat pcDNA3 or Jurkat pcDNA3␥R2 was cultured (0.5⫻106/mL) in medium supplemented with IFN-␥. At 48 h, a small aliquot of the cell suspension was removed and mixed with an equal volume of trypan blue-containing solution, and viable cells were counted. Results are expressed as the arithmetic means ⫾ SD of numbers of viable cells from triplicate cultures and reported as cell number ⫻ 10⫺6/mL. The experiments were performed independently at least three times, and representative experiments are shown. Statistical significance of the difference in proliferation between untreated cells and treated cells as evaluated by Student’s t-test is indicated. *, P ⬍ 0.05; **, P ⬍ 0.005. (C) IRF-1 and caspase-1 induction by IFN-␥ on transfected Jurkat pcDNA3 or Jurkat pcDNA3␥R2 cells. Jurkat pcDNA3 or Jurkat pcDNA3␥R2 cells were cultured in the absence or presence of 100 U/mL of IFN-␥ for 24 h. IRF-1 induction was evaluated by Western blot analysis on nuclear cell extracts and caspase-1 induction by Western blot analysis on total cell extracts. Western blot filters were subsequently challenged with an anti-actin antibody to confirm equal protein loading in each lane of the gel. Fold induction of IFN-␥-treated cells relative to untreated cells was quantitated after normalization with actin using the UTHSCSA ImageTools for Windows 2.0.

stages, did not produce IFN-␥ constitutively, and displayed the IFN-␥R2 chain preferentially in the cytoplasm. This expression pattern resembles that of human Th1 and Th2 clones, in which IFN-␥ still induces IRF-1 and MHC class I antigen expression without affecting cell viability [16]. Preferential cytoplasmic expression of IFN-␥R2 in T cells may reflect the need to limit the apoptotic effect of IFN-␥ that occurs when the membrane expression of IFN-␥R is high [37]. On B and myeloid cells, a different rate of trafficking between the early endosomes and the cell surface could be responsible for higher IFN-␥R2 membrane expression. Moreover, the absence of serum or growth factors might modify this trafficking on T cells and render them susceptible to IFN-␥-induced apoptosis [16, 26]. Here we show that the absence of serum increased IFN-␥R2 expression in T cells and thus determined the signal switch from proliferation to apoptosis. Serum deprivation, like IL-2 deprivation [16], mimicked the passive apoptosis induced by growth factor deprivation that T cells encounter in vivo [40]. Up-regulation of IFN-␥R2 in serum-deprived T cells suggested that serum factors such as hormones or ions might keep IFN␥R2 expression low. It is interesting that this up-regulation was completely abolished by iron, whereas the iron chelator deferoxamine increases IFN-␥R2 [41]. Studies addressing the role 958

Journal of Leukocyte Biology Volume 70, December 2001

of ions or hormones in regulating the expression of IFN-␥R2 are currently in progress in our laboratory. Present data indicated that overexpression of IFN-␥R2 induces an optimal STAT-1 activation that results in IFN-␥mediated apoptosis in line with previous observations [42– 44]. Besides optimal activation of STAT-1, that of IRF-1, which is also involved in the IFN-␥-mediated apoptosis [22, 23], appeared to be induced to a lesser extent, implying that there is no direct relationship between STAT-1 activation and the levels of IRF-1 expression. This suggests that the apoptosis induced by IFN-␥ is strictly dependent on the extent of STAT-1 activation rather than IRF-1 levels, indicating that IFN-␥ might, through STAT-1, induce additional factors that are critical for IFN-␥-mediated apoptosis. Even a relative increase in IRF-1 expression might be critical for IFN-␥induced apoptosis, however [8], because the partial blockade of IRF-1 expression by mRNA antisense switches the antiproliferative signal delivered by IFN-␥ to a growth-promoting signal [22]. Caspase-1 is up-regulated by IFN-␥ in B, myeloid, serumdeprived, and IFN-␥R2-transfected T cells to the same extent as IRF-1, confirming that a relationship exists between IFN␥-induced IRF-1 and activation of the caspase-1 [8, 25, 28]. There is strong evidence that caspase-1 plays an obligatory role http://www.jleukbio.org

in IFN-␥-induced apoptosis [44]. In effect, caspase inhibitors reverse IFN-␥-induced apoptosis in IFN-␥R2 overexpressing cells (data not shown). However, because caspase-1 is generally regarded as an inflammatory protease rather than a key enzyme involved in apoptosis, other death effectors might well be involved in such events [31, 45]. An interesting feature of our data is the involvement of NF-␬B activation in IFN-␥-induced proliferation. This transcriptional factor, in fact, was activated by IFN-␥ only in T cells, not in B and myeloid cells. These observations are in agreement with that of Deb et al., who found that IFN-␥ induces NF-␬B activation in a STAT-1-independent manner [46]. These data suggest that proliferation of T cells in response to IFN-␥ could result from a combined action of the antiapoptotic effects of NF-␬B and slow STAT-1 activation. The fact that B cells respond to IFN-␥ by undergoing apoptosis and T cells respond by proliferating could represent a way to switch off activation of the T helper response through the down-regulation of helper functions. It has been shown, in fact, that cytolytic Th1 cells kill antigen-pulsed autologous B cells and that this cytolytic activity is increased by IFN-␥ [47]. Because the cytolytic ability of Th1 cells is observed only when the T-B ratio is high, IFN-␥ might expand the T cell population and thus limit the effect of Th1 helper cells [48]. In conclusion, the present data provide evidence for the fine regulation of the response of human T, B, and myeloid cells to IFN-␥ through an interplay between the density of IFN-␥R2 surface levels, STAT-1 activation, and the cell death signal cascade. This could be the starting point for the elaboration of the new strategies to control their growth or apoptosis.

ACKNOWLEDGMENTS This work was supported in part by grants from the Istituto Superiore di Sanita´ (special projects on AIDS) to F. N. and A. B., Fondazione Piemontese Studi e Ricerche sulle Ustioni (FPSRU) to F. N., Associazione Italiana per la Ricerca sul Cancro (AIRC) to F. N. and G. F., Ministero dell’Universita´ della Ricerca Scientifica (MURST) ex 40% to G. F., MURSTCNR Biotechnology Program to G. F., MURST Molecular Engineering L.488/92 to G. F., National Institutes of Health (CA-46465 and AI-36450) to S. P., and the New Jersey Cancer Commission (98-2003-CCR-00) to C. D. K. P. B. was supported by a fellowship from Fondazione Italiana Ricerca sul Cancro (FIRC). We thank J. Iliffe and G. Garotta for critically reading the manuscript.

REFERENCES 1. Billiau, A. (1996) Interferon ␥: biology and role in pathogenesis. Adv. Immunol. 62, 61–130. 2. Pestka, S., Kotenko, S. V., Muthukumaran, G., Izotova, L. S., Cook, J. R., Garotta, G. (1997) The interferon gamma receptor: a paradigm for the multichain cytokine receptor. Cytokine Growth Factor Rev. 8, 189 –206. 3. Darnell, J. E., Jr., Kerr, I. M., Stark, G. R. (1994) Jak-Stat pathway and transcriptional activation in response to IFN and other intracellular signaling proteins. Science 264, 1415–1421. 4. Boehm, U., Klamp, T., Groot, M., Howard, J. C. (1997) Cellular responses to interferon-␥. Annu. Rev. Immunol. 15, 749 –795.

5. Nguyen, H., Hiscott, J., Pitha, P. M. (1997) The growing family of interferon regulatory factors. Cytokine Growth Factor Rev. 8, 293–312. 6. Harada, H., Takahashi, E., Itoh, S., Harada, K., Kori, T. A., Taniguchi T. (1994) Structure and regulation of the human interferon regulatory factor 1 (IRF-1) and IRF-2 genes: implications for a gene network in the interferon systems. Mol. Cell. Biol. 14, 1500 –1509. 7. Coccia, E. M., Marziali, G., Stellacci, E., Perrotti, E., Ilari, R., Battistini, A. (1995) Cells resistant to interferon-␤ respond to interferon-␥ via STAT1-IRF-1 pathway. Virology 211, 113–122. 8. Taniguchi, T., Harada, H., Lamphier, M. (1995) Regulation of the interferon system and cell growth by the IRF transcription factors. J. Cancer Res. Clin. Oncol. 121, 516 –520. 9. Selleri, C., Maciejewski, J. P., Sato, T., Young, N. S. (1996) Interferongamma constitutively expressed in the stromal microenvironment of human marrow cultures mediates potent hematopoietic inhibition. Blood 87, 4149 – 4157. 10. Kawano, Y., Takaue, Y., Hirao, A., Abe, T., Saito, S., Matsunaga, K., Watanabe, T., Hirose, M., Ninomiya, T., Kuroda, Y., Yokobayashi, A., Asano, S. (1991) Synergistic effect of recombinant interferon-gamma and interleukin-3 on the growth of immature human hematopoietic progenitors. Blood 77, 2118 –2121. 11. Brugger, W., Mocklin, W., Heimfeld, S., Berenson, R. J., Mertelsmann, R., Kanz, L. (1993) Ex vivo expansion of enriched peripheral blood CD34⫹ progenitor cells by stem cell factor, interleukin-1beta (IL-1beta), IL-6, IL-3, interferon-gamma, and erythropoietin. Blood 81, 2579 –2584. 12. Shiohara, M., Koike, K., Nakahata, T. (1993) Synergism of interferongamma and stem cell factor on the development of murine hematopoietic progenitors in serum-free culture. Blood 81, 1435–1441. 13. Tamura, T., Ueda, S., Yoshida, M., Matsuzaky, M., Mohri, H., Okubo, T. (1996) Interferon-gamma induces Ice gene expression and enhances cellular susceptibility to apoptosis in the U937 leukemia cell line. Biochem. Biophys. Res. Commun. 229, 21–26. 14. Grawunder, U., Melchers, F., Rolink F. (1993) Interferon-␥ arrests proliferation and causes apoptosis in stromal/interleukin-7-dependent normal murine pre-B cell lines and clones in vitro, but does not induce differentiation to surface immunoglobulin-positive B cells. Eur. J. Immunol. 23, 544 –551. 15. Romagnani, S., Giudizi, M. G., Biagiotti, R., Almerigogna, F., Mingari, C., Maggi, E., Liang, C. M., Moretta, L. (1986) B cell growth factor activity of interferon-gamma. Recombinant human interferon-gamma promotes proliferation of anti-mu-activated human B lymphocytes. J. Immunol. 136, 3513–3516. 16. Novelli, F., D’Elios, M. M., Bernabei, P., Ozmen, L., Rigamonti, L., Almerigogna, F., Forni, G., Del Prete, G. (1997) Expression and role in apoptosis of the ␣- and ␤-chains of the IFN-␥ receptor on human Th1 and Th2 clones. J. Immunol. 159, 206 –213. 17. Liu, Y., Janeway, C. A. (1990) Interferon-␥ plays a critical role in induced cell death of effector T cells: a possible third mechanism of self tolerance. J. Exp. Med. 172, 1735–1739. 18. Novelli, F., Giovarelli, M., Reber-Liske, R., Virgallita, G., Garotta, G., Forni, G. (1991) Blockade of physiologically secreted IFN-␥ inhibits human T lymphocytes and natural killer cell activation. J. Immunol. 147, 1445–1452. 19. Novelli, F., Bernabei, P., Ozmen, L., Rigamonti, L., Allione, A., Pestka, S., Garotta, G., Forni, G. (1996) Switching on of the proliferation or apoptosis of activated human T lymphocytes by IFN-␥ is correlated with the differential expression of the alpha and beta chains of its receptor. J. Immunol. 157, 1935–1943. 20. Allione, A., Wells, V., Forni, G., Mallucci, L., Novelli, F. (1998) ␤-Galactoside-binding protein (␤-GBP) alters the cell cycle, up-regulates expression of the ␣- and ␤-chains of the IFN-␥ receptor and triggers IFN-␥ mediated apoptosis of activated human T lymphocytes. J. Immunol. 161, 2114 –2119. 21. Allione, A., Bernabei, P., Bosticardo, M., Ariotti, S., Forni, G., Novelli, F. (1999) Nitric oxide suppresses human T lymphocyte proliferation through IFN-␥-dependent and IFN-␥-independent induction of apoptosis. J. Immunol. 163, 4182– 4191. 22. Sato, T., Selleri, C., Young, N. S., Maciejewski, J. P. (1997) Inhibition of interferon regulatory factor-1 expression results in predominance of cell growth stimulatory effects of interferon-␥ due to phosphorylation of STAT-1 and STAT-3. Blood 90, 4749 – 4758. 23. Sato, T., Selleri, C., Young, N. S., Maciejewski, J. P. (1995) Hematopoietic inhibition by interferon-␥ is partially mediated through interferon regulatory factor-1. Blood 86, 3373–3380. 24. Villa, P., Kaufmann, S. H., Earnshaw, W. C. (1997) Caspases and caspase inhibitors. Trends Biochem. Sci. 10, 388 –393. 25. Tamura, T., Ishihara, M., Lamphier, M. S., Tanaka, N., Oishi, I., Aizawa, S., Matsuyama, T., Mak, T. W., Taki, S., Taniguchi, T. (1997) DNA

Bernabei et al. IFN-␥R2 distribution on human hematopoietic cells

959

26.

27.

28.

29. 30.

31.

32. 33.

34.

35.

36.

37.

damage-induced apoptosis and Ice gene induction in mitogenically activated T lymphocytes require IRF-1. Leukemia 11, 439 – 440. Novelli, F., Di Pierro, F., Francia di Celle, P., Bertini, S., Affaticati, P., Garotta, G., Forni, G. (1994) Environmental signals influencing expression of the IFN-␥ receptor on human T cells control whether IFN-␥ promotes proliferation or apoptosis. J. Immunol. 152, 496 –504. Kotenko, S. V., Izotova, L. S., Pollak, B. P., Muthukumaran, G., Paukku, K., Silvennoinen, O., Ihle, J. N., Pestka, S. (1996) Other kinases can substitute for Jak2 in signal transduction by interferon-␥. J. Biol. Chem. 271, 17174 –17182. Tamura, T., Ishihara, M., Lamphier, M. S., Tanaka, N., Oishi, I., Aizawa, S., Matsuyama, T., Mak, T. W., Taki, S., Taniguchi, T. (1995) An IRF-1dependent pathway of DNA damage-induced apoptosis in mitogen-activated T lymphocytes. Nature 376, 596 –599. Klug, J. (1997) Ku autoantigen is a potential major cause of nonspecific bands in electrophoretic mobility shift assay. BioTechniques 22, 212–216. Krause, C. D., Lunn, C. A., Izotova, L. S., Mirochnitchenko, O., Kotenko, S. V., Lundell, D. J., Narula, S. K., Pestka, S. (2000) Signaling by covalent heterodimers of interferon-gamma: evidence for one-sided signaling in the active tetrameric receptor complex. J. Biol. Chem. 275, 22995–23004. Dai, C., Krantz, S. B. (1999) Interferon ␥ induces upregulation and activation of caspase 1, 3, and 8 to produce apoptosis in human erythroid progenitor cells. Blood 10, 3309 –3316. Cohen, G. M. (1997) Caspases: the executioners of apoptosis. Biochem. J. 326, 311–316. Fukami-Kobayashi, J., Mitsui, Y. (1999) Overexpression of proliferating cell nuclear antigen in mammalian cells negates growth arrest by serum starvation and cell contact. Jpn. J. Cancer Res. 90, 286 –293. Hasan, N. M., Adams, G. E., Joiner, M. C. (1999) Effect of serum starvation on expression and phosphorylation of PKC-alpha and p53 in V79 cells: implications for cell death. Int. J. Cancer 80, 400 – 405. Amundson, S. A., Zhan, Q., Penn, L. Z., Fornace, A. J., Jr. (1998) Myc suppresses induction of the growth arrest genes gadd34, gadd45, and gadd153 by DNA-damaging agents. Oncogene 17, 2149 –2154. Drexler, H. G., Zaborsky, M., Quentmeier, H. (1997) Interferon-gamma induced proliferation of human myeloid leukemia cell lines. Br. J. Haematol. 98, 699 –710. Rigamonti, L., Ariotti, S., Losana, G., Gradini, R., Russo, M. A., Jouanguy, E., Casanova, J. L., Forni, G., Novelli, F. (2000) Surface expression of the

960

Journal of Leukocyte Biology Volume 70, December 2001

38. 39.

40. 41.

42. 43. 44. 45.

46.

47.

48.

IFN-␥R2 chain is regulated by intracellular trafficking in human T lymphocytes. J. Immunol. 164, 201–207. Trowbridge, I. S. (1991) Endocytosis and signals for internalization. Curr. Opin. Cell Biol. 3, 634 – 641. Chuang, E., Alegre, M. L., Duckett, C. S., Noel, P. J., Vander Heiden, M. G., Thompson, C. B. (1997) Interaction of CTLA-4 with the clathrinassociated protein AP50 results in ligand-independent endocytosis that limits cell surface expression. J. Immunol. 159, 144 –151. Lenardo, M. J. (1996) Fas and the art of lymphocyte maintenance. J. Exp. Med. 183, 721–724. Allione, A., Pescarmona, G. P., Bernabei, P., Rigamonti, L., Forni, G., Novelli, F. (1997) Iron and nitric oxide regulate the expression of the IFN-␥ receptor and the apoptosis of human neoplastic T cells. Biochem. Soc. Trans. 24S, 604 (Abstr.). Lee, C. K., Smith, E., Gimeno, R., Gertner, R., Levy, D. E. (2000) STAT1 affects lymphocyte survival and proliferation partially independent of its role downstream of IFN-gamma. J. Immunol. 164, 1286 –1292. Lee, K. Y., Anderson, E., Madani, K., Rosen, G. D. (1999) Loss of STAT1 expression confers resistance to IFN-gamma-induced apoptosis in ME180 cells. FEBS Lett. 459, 323–326. Chin, Y. E., Kitagawa, M., Kuida, K., Flavell, R. A., Fu, X. Y. (1997) Activation of the STAT signaling pathway can cause expression of caspase-1 and apoptosis. Mol. Cell. Biol. 17, 5328 –5337. Ossina, N. K., Cannas, A., Powers, V. C., Fitzpatrick, P. A., Knight, J. D., Gilbert, J. R., Shekhtman, E. M., Tomei, L. D., Umansky, S. R., Kiefer, M. C. (1997) Interferon-␥ modulates a p53-independent apoptotic pathway and apoptosis-related gene expression. J. Biol. Chem. 272, 16351–16357. Deb, A., Haque, J., Mogensen, T., Silverman, R. H., Williams, B. R. G. (2001) RNA-dependent protein kinase PKR is required for activation of NF-␬B by IFN-␥ in a STAT1-independent pathway. J. Immunol. 166, 6170 – 6180. Parronchi, P., De Carli, M., Manetti, R., Simonelli, C., Sampognaro, S., Piccinni, M. P., Macchia, D., Maggi, E., Del Prete, G., Romagnani, S. (1992) IL-4 and IFN (␣ and ␥) exert opposite regulatory effects on the development of cytolytic potential by Th1 or Th2 human T cell clones. J. Immunol. 149, 2977–2983. Del Prete, G., De Carli, M., Ricci, M., Romagnani, S. (1991) Helper activity for immunoglobulin synthesis of T helper type 1 (Th1) and Th2 human T cell clones: the help of Th1 clones is limited by their cytolytic capacity. J. Exp. Med.174, 809 – 813.

http://www.jleukbio.org