Báo cáo Y học: Interferon-alpha inhibits Stat5 DNA-binding in IL-2 stimulated primary T-lymphocytes

Eur. J. Biochem. 269, 29–37 (2002) (cid:211) FEBS 2002

Interferon-alpha inhibits Stat5 DNA-binding in IL-2 stimulated primary T-lymphocytes

Sven Erickson1,*, Sampsa Matikainen2,*, Lena Thyrell1, Olle Sangfelt1, Ilkka Julkunen2, Stefan Einhorn1 and Dan Grande´ r1

1Department of Oncology and Pathology, Cancer Centre Karolinska (CCK), Karolinska Hospital and Institute, Stockholm, Sweden; 2Department of Virology, Mannerheimintie 166, Helsinki, Finland

induced proliferation in activated T-lymphocytes, was associated with a suppressed Jak3 protein expression as well as an inhibited prolonged Stat5 DNA binding, and a par- tially reduced expression of the Stat5 inducible gene IL-2Ra. Our results provide a possible molecular link between the prominent antiproliferative e(cid:128)ects of IFN-a on IL-2 induced T-cell proliferation and the signal transduction pathways emerging from the IL-2 receptor.

interferon-alpha, T-lymphocytes; Stat5;

inter-

Keywords: leukin-2; proliferation.

It has previously been shown that IFN-a is a potent inhibitor of IL-2 induced proliferation in primary T-lymphocytes, by selectively abrogating the downstream e(cid:128)ects of IL-2 on the core cell cycle machinery regulating the G1/S transition. Theoretically this could be mediated through cross-talk between the signalling cascades activated by these cytokines, as several signalling components are known to be shared. IL-2 activates multiple signalling pathways that are impor- tant for T-cell proliferation and di(cid:128)erentiation. In the pre- sent study, the e(cid:128)ects of IFN-a on IL-2 signal transduction was investigated. The IFN-a induced inhibition of IL-2

Interferons (IFNs) constitute a family of proteins first isolated because of their antiviral abilities [1]. Today, IFNs are known to be therapeutically effective in the treatment of a number of malignancies [2], viral diseases, as well as in immunorelated disorders such as multiple sclerosis. IFNs mechanism of action in these diseases is unclear, but its well known ability to inhibit proliferation has been suggested to be of major importance. The antiproliferative effect of IFN-a in malignant cell lines has recently been linked to potent effects on the basic cell cycle machinery [3–5]. The investigated cell lines are commonly arrested by IFN in the G1 phase of the cell cycle, an effect that is coinciding with the up-regulation of the cyclin-dependent kinase inhibitors (CKIs), such as p15, p21 and p27 [4,5]. We have recently shown that IFN-a also inhibits interleukin-2 (IL-2) induced proliferation in nontransformed human peripheral T-lym- phocytes, by inhibiting the activation of key cell cycle molecules [6]. This finding has prompted us to investigate whether this effect may relate to interactions with the signalling from the IL-2 receptor.

The lymphocyte-derived cytokine IL-2 plays a pivotal role in the regulation of immune responses, and is a major regulator of T-lymphocyte proliferation [7,8]. IL-2 signalling is mediated by a multimeric receptor (IL-2R), consisting of two obligate signalling subunits, IL-2Rb and cc and the variably expressed IL-2Ra subunit, which regulates the affinity for IL-2 [8]. The IL-2 receptor exhibits no intrinsic kinase activity but rather relies on the association of intracellular protein kinases for signalling. Ligation of IL-2 to the receptor triggers several signal transduction path- ways, including the Jak-Stat pathway as well as alternative pathways leading to the activation of molecules such as c-myc, and phosphatidylinositol-3-kinase (PtdIns3K) [7,8]. Heterodimerization of the IL-2Rb and cc chains results in the rapid activation of several protein tyrosine kinases including Jak1, Jak3 and Syk. Syk activation has been suggested to be of importance for the up-regulation of c-myc, thereby being crucial for the proliferative response to IL-2 [7,9]. Jak1, however, seems dispensable for IL-2 induced proliferation [10], whereas in contrast, Jak3 activa- tion seem to be essential for IL-2 dependent mitogenic signalling. This conclusion is supported by a number of manipulative studies. For example, fibroblasts expressing a reconstituted IL-2R complex will fail to proliferate in the presence of IL-2 unless Jak3 is coexpressed [11], and furthermore, an inactive form of Jak3 severely inhibits IL-2 mediated proliferation in BAF3 cells [12]. Receptor associ- ated Jak3 will phosphorylate specific residues in the cytoplasmatic domains of the IL-2Rb. These phosphotyro- sine motifs serve as docking sites for the SH2 domain in Stat5, a member of the Stat family of transcription factors. Upon binding to the receptor complex, Stat5 will become phosphorylated, dimerize and translocate into the nucleus inducing transcription of its target genes [8].

Stat5 exists in two different forms, Stat5a and Stat5b, encoded by separate genes. The two genes are highly

Correspondence to D. Grande´ r, Department of Oncology and Pathology, Cancer Centre Karolinska (CCK), Karolinska Hospital and Institute, S-171 76 Stockholm, Sweden. Fax: + 46 8 339031, Tel.: + 46 8 51776262, E-mail: dan.grander@cck.ki.se Abbreviations: IFN, interferon; CKI, cyclin-dependent kinase inhibi- tor; IL, interleukin; IL-2R, interleukin-2 receptor; PtdIns3K, phos- phatidylinositol-3-kinase; Jak, Janus kinase; Stat, signal transduction activator of transcription; EMSA, electrophoretic mobility shift assay; GAS, gamma activation site; MEK, mitogen-activated/extra-cellular regulated kinase; ERK, extracellular regulated kinase; Cdk, cyclin dependent kinase; SOCS, suppressor of cytokine signalling. *Note: the authors contributed equally to this manuscript. (Received 17 August 2001, revised 18 October 2001, accepted 19 October 2001)

30 S. Erickson et al.

(Eur. J. Biochem. 269)

(cid:211) FEBS 2002

Recombinant IL-2 was a generous gift form A. O¨ sterborg (Karolinska Institute, Stockholm, Sweden). The following antibodies were used for Western blotting and immunopre- cipitation in this study. The phosphospecific polyclonal Akt antibodies (Ser473 and Thr308) and the polyclonal Akt antibody were from New England Biolabs, polyclonal Jak3 from Santa Cruz and monoclonal phosphotyrosine anti- bodies (4G10) from Upstate Biotechnology (Lake Placid, NY, USA).

it

Electrophoretic mobility shift assay (EMSA)

homologous but differ in the C-terminus region [13]. Stat5 is expressed in a variety of tissues and its biological effects are still incompletely understood. Mice with homozygous inactivation of the Stat5a gene exhibit a failure in postpar- tum mammary gland differentiation and lactation, demon- strating that Stat5 is required for mammopoesis and lactation. The hematopoietic system is not severely affected in Stat5a–/–, Stat5b–/– or Stat5a–/– Stat5b–/– double knockout mice [14]. Although Stat5 is apparently not required for hematopoiesis, is however, of crucial importance for the proliferative response to IL-2, as thymocytes from Stat5 double knockout mice fail to proliferate in response to IL-2 [14,15].

We have recently demonstrated that IFN-a is a potent inhibitor of IL-2 induced proliferation in PHA stimulated human peripheral T-lymphocytes [16] where IFN-a abro- gated the activation of the basic cell cycle machinery as well as the inhibited the down-regulation of the CKI p27 [16]. The effect on IL-2 signalling was however, selective, as IFN- a did not inhibit the IL-2 dependent up-regulation of c-myc or Cdc25A. In order to obtain a better understanding whether these previously observed effects may be due to cross-talk with the more upstream IL-2 signalling, we have here investigated how IFN-a affects the activation of IL-2R associated molecules, and subsequent activation of IL-2 responsive genes.

instructions

M A T E R I A L S A N D M E T H O D S

T-Lymphocyte isolation and cell culture

Buffy coats from healthy blood donors were heparinized, mononuclear cells isolated by Lymphoprep gradient cen- trifugation (Nycomed, Oslo, Norway), and T-lymphocytes isolated using nylon wool columns, as previously described [17]. In each experiment performed, > 95% of the cells were viable, as determined by trypan blue exclusion. The enriched human T-lymphocytes were tested for purity before stimulation as previously described [17] and were found to be more then 94% pure.

To prepare nuclear extracts, 10 · 106 cells were harvested by centrifugation, 600 g, for 8 min at 4 (cid:176)C, and washed once in 1.5 mL ice-cold NaCl/Pi. Subsequently cells were resuspended in 400 lL ice-cold Buffer A (10 mM Hepes/ KOH pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dith- iothreitol and CompleteTM protease inhibitor cocktail according to the manufacturers instructions (Boehringer Mannheim, Mannheim, Germany) by flicking the tube. Cells were allowed to swell on ice for 10 min, and then vortexed for 10 s. Samples were centrifuged for 10 s, and the supernatant discarded. The pellets were resuspended in 100 lL ice-cold Buffer B (20 mM Hepes/KOH pH 7.9, 25% glycerol, 420 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, and CompleteTM protease inhibitor cocktail according the manufacturers (Boehringer Mannheim, Mannheim, Germany) and incubated on ice for high-salt extraction. Cellular debris was removed by centrifugation for 2 min at 4 (cid:176)C, the supernatant removed and saved at ) 70 (cid:176)C. Protein concentration was determined spectrophotometrically with the Bradford method, accord- ing to instructions of the manufacturer (Bio-Rad). Nuclear protein/DNA binding reactions were performed as previ- ously described [19]. The following oligonucleotides were used: IFP 53 GAS (5¢-GATCAATCACCCAGATTCT CAGAAACACTT- 3¢), IRF-1 GAS (5¢-AGCTTCAG CCTGATTTCCCCGAAATGACGGA-3¢) and IL-2Ra (5¢-TTTCTTCTAGGAAGTACCAAA GAS-c/GAS-n probes were The CATTTCTGATAATAGAA-3¢). 32P-labelled by T4 polynucleotide kinase and the binding reaction was performed at room temperature for 30 min. Samples were separated on 6% nondenaturating low-ionic strength polyacrylamide gels in 0.25 · Tris/borate/EDTA. Subsequently gels were dried and bands visualized by autoradiography. Antibodies used in supershift experiments were purchased from Santa Cruz Biotechnology. The following antibodies were used: anti-Stat1 (sc-245X), anti-Stat3 (sc-482X), anti-Stat4 (sc-486X) and anti-Stat5 (sc-835). In supershift experiments, antibodies (1 : 20 dilu- tion) were incubated with the nuclear extracts for 1 h on ice.

Western blot analysis

The lymphocytes were seeded at a density of 5 · 105 cells per ml in complete MEM (MEM supplemented with 10% heat inactivated AB+ serum, 2 mM glutamine 50 mgÆmL)1 of streptomycin and 50 mgÆmL)1 of penicillin) and kept in a humid incubator with 5% CO2. The cells were counted and reseeded every second day (5 · 105 cells per mL) when fresh medium was added. The resting T-lymphocytes were stimulated to proliferation in a two-stage process by the sequential addition of 0.8 lgÆmL)1 PHA (Sigma Chemical Co.) for 72 h and, subsequently IL-2 at a concentration of 100 UÆmL)1 [17,18]. Restimulation was performed every second day by inclusion of IL-2 (100 UÆmL)1) in the fresh medium added to the cells. In cultures receiving IFN-a (5000 UÆmL)1), it was added to the cultures alone or together with IL-2.

Each experiment was repeated with cells from two to seven different donors. All figures show the results from representative experiments.

Cytokines and antibodies

Recombinant IFN-a2b (from Schering-Plough, Kenil- worth, NJ, USA) was used in the different experiments.

Western blot analysis was performed essentially as previ- ously described [4]. Briefly, whole extracts were prepared by lysis through sonication in LSLD buffer containing protease inhibitors. The protein concentration was determined as described above. Seventy lg of protein was loaded in each well. Proteins were resolved by SDS/PAGE on 12% gels and electroblotted to poly(vinylidene difluoride) membranes (Boerhinger Mannheim GMbH, Germany) by semidry transfer. For protein detection, the filters were hybridized

(cid:211) FEBS 2002

Interferon-a inhibits Stat5 induced transcription (Eur. J. Biochem. 269) 31

for 1 h with the appropriate antibody. Antibody–antigen interaction was detected by incubation with horseradish peroxidase-conjugated anti-(rabbit IgG)Ig or anti-(mouse IgG)Ig for one hour, and subsequent detection by enhanced chemiluminescence, ECL (Amhersham).

Immunoprecipitation

from Firpo et al. [18] and subsequent data from our own group [16]. Exogenous IL-2 addition allows for S-phase entry and activation the G1 Cdk complexes [16,18]. As described elsewhere, cotreatment with IFN-a abrogated the proliferative response to IL-2 and severely inhibited the activation of the cell cycle machinery [16]. The effect of IFN- a on the proportion of apoptotic cells was also measured by flow cytometry for DNA content and AnnexinV positivity. There were no significant differences in the proportion of apoptotic cells in IL-2 treated or IFN-a plus IL-2 treated cells up to 48 h of culture as measured by sub-G1 content (data not shown) or Annexin V staining (Fig. 1).

For immunoprecipitation whole protein extracts were prepared as described above. Lysates were precleared with protein A–Sepharose beads (Pharmacia Biotech AB, Uppsala, Sweden) with gentle agitation for at least 2 h at 4 (cid:176)C. Precleared extracts representing 400 lg of whole protein extract were incubated with polyclonal Jak3 anti- bodies for 2 h at 4 (cid:176)C. Antibody associated complexes were then bound to prewashed protein A–Sepharose beads, washed four times in LSLD buffer containing protease inhibitors and denatured as previously described [5]. Precipitated proteins were resolved on an 8% SDS/PAGE, electroblotted and analysed by sequential immunoblotting with Jak3 and anti-phosphotyrosine Ig.

Northern blot analysis

Recent studies have clearly demonstrated that Jak3 is required to activate the IL-2R pathway for T-cell prolifer- ation. Jak3 is rapidly phosphorylated upon IL-2 stimula- tion, an event essential for a proper proliferative signal [8,20,21]. In order to investigate if IFN-a affects the IL-2 induced expression of Jak3, Western blotting was per- formed. IL-2 stimulation for 24 h (day 4) increased the Jak3 expression significantly from the low levels present in unstimulated cells, whereas prolonged stimulation did not lead to any further increase in Jak3 protein levels (Fig. 2A). IFN-a suppressed the IL-2 dependent Jak3 expression in cells treated for more than 24 h (Fig. 2A).

Because low levels of Jak3 are present in the cells before IL-2 stimulation we analysed whether IFN affected the immediate Jak3 tyrosine phosphorylation upon IL-2 addi- tion. PHA treated T-lymphocytes were treated with IL-2 alone or together with IFN-a for 10 and 30 min, whereafter Jak3 was immunoprecipitated and tyrosine phosphoryla- tion detected by Western blotting. In contrast to the abrogation of the IL-2 dependent up-regulation of Jak3, IFN-a did not inhibit the rapid tyrosine phosphorylation at any of the timepoints investigated (Fig. 2B).

Total cellular RNA was isolated using Trizol Reagent according to the instructions of the manufacturer (Life Technologies), and the RNA concentration was measured spectrophotometricly. Twenty micrograms of total RNA from each sample was subjected to electrophoresis in a 1.2% agarose formaldehyde gel. Blotting was performed as described [4]. Northern filters (Hybond C-extra, Amersham) were hybridized with the cDNA probes encoding IL-2Ra and pim-1, and probes were labelled as previously described [4]. EtBr staining of rRNA bands was used to ensure equal RNA loading.

IFN-a effects on Stat5 DNA-binding

Quantification of apoptosis using Annexin V staining

Redistribution of plasma membrane phosphatidylserine is a marker of apoptosis and was assessed using Annexin V FLUOS and Propidium Iodide staining kit according to the manufacturer’s protocol (Boehringer Mannheim, Mannheim Germany).

Although the role of Stat5 in T-lymphocyte proliferation is still somewhat controversial, recent studies indicate that Stat5 is indeed required for a proper proliferative response to IL-2 [15]. To characterize the effect of IFN-a in the onset of IL-2 signalling and Stat5 binding, PHA activated cells were treated with IL-2 alone, IFN-a alone or IL-2 together with IFN-a. Nuclear extracts were prepared from these cells and analysed by EMSA, by using a labelled IFP-53 GAS

Quantitation of band intensity

For scanning of the films a Luminescent image analyzer LAS-1000 plus (Fuji Film Co., Ltd) was used. Hybridiza- tion signals were quantified by using IMAGE GAUGE Ver. 3.12 by Fuji Film Co., Ltd.

R E S U L T S

IFN-a inhibits IL-2 dependent up-regulation of Jak3, but not its transient phosphorylation

Purified primary T-lymphocytes from healthy donors, were treated with PHA for three days, and subsequently with IL-2 alone or IL-2 together with IFN for up to 72 h. Using the stimulation protocol described in the materials and methods section, the initial PHA stimulation provides an activation signal, but alone, is clearly not sufficient to induce proliferation in this setting, in agreement with previous data

Fig. 1. E(cid:128)ects of IFN-a on apoptosis as measured by annexin V and propidium iodide staining. PHA activated T-cells were treated with IL-2 alone or IL-2 together with IFN-a for 48 h. The percentage of apop- totic (annexin V positive) cells in the samples is shown. Annexin V staining is represented on the x-axis and propidium iodide staining is represented on the y-axis.

32 S. Erickson et al.

(Eur. J. Biochem. 269)

(cid:211) FEBS 2002

Hours

24

48

0

A

IFN-a

–

+

–

+

–

depending on celltype and type of stimulation [23,24]. Again, cotreatment with IFN-a for 48 h resulted in a significantly decreased Stat5 bandshift as well as supershift. Thus, IFN-a clearly inhibited the binding of Stat5a/b to the IFP53 element at this timepoint. Total levels of Stat5 protein on the other hand, were not changed by cotreatment with IFN-a, as demonstrated by Western blotting for this protein (data not shown).

Minutes

10

30

B

IFN-a

–

+

–

+

W:P-Tyr

Jak3

Jak3

W:Jak3

Recently an IL-2 responsive element in the IL-2Ra promoter has been described. This region contains a consensus and a nonconcensus GAS motif (GAS-c/GAS-n), which has been shown to be essential for proper IL-2 induced IL-2Ra transcription [25]. IL-2 stimulated prolifer- ating T-lymphocytes have also recently been shown to contain Stat1, 3, 4 and 5 DNA binding activity to this GAS element upon costimulation with IFN-a [26]. To further characterize the effect of IFN-a on the binding activity also to this element, further EMSAs were performed using this GAS motif (GAS-c/GAS-n) as a probe. IL-2 alone induced one major complex already after 0.5 h. IFN-a treated cells also expressed one single complex, albeit with a broader migration pattern, implicating the comigration of two separate bandshifts. Co-treated cells also expressed both of these complexes 0.5 h after stimulation (Fig. 4A). At later timepoints (6 h and thereafter), complex formation was significantly decreased in IL-2 treated and cotreated cells, and only the slowly migrating form was detectable. IFN-a treated cells alone contained no complexes at these later timepoints (Fig. 4A). Furthermore, as observed using IFP- 53 GAS as a probe, IFN-a clearly inhibited the complex formation compared to IL-2 treated cells alone at the two latest timepoints (24 and 48 h). To verify the complex composition, supershifts with a set of different Stat anti- bodies were performed. In the IL-2 treated cells Stat5 was the only antibody generating a supershift shifting the majority of the complex (Fig. 4B). In the extracts from IL-2/IFN-a cotreated cells, incubation with Stat1, Stat3, Stat4 and Stat5 antibodies all resulted in supershifts (Fig. 4B), however, the Stat1 antibody generated the most prominent shift. In extracts from IFN-a treated cells alone Stat1 shifted the absolute majority of the complex, however, the complexes also contained very small amounts of Stat3 and 4 binding (Fig. 4B). Taken together these data confirm the findings using the IFP53 GAS element, that is, IFN-a cotreatment inhibited the prolonged IL-2 dependent Stat5 DNA binding activity but not the immediate one. Furthermore, identical data on the effects of IFN-a on Stat5 activity were also obtained using an IRF-1 GAS element (data not shown).

Effects of IFN-a on Stat5-induced transcription

IL-2 dependent

DNA-element that is known to bind Stat5 with high affinity [22]. EMSA analysis revealed that untreated and PHA treated T-cells contained no DNA binding activity to this element. Stimulation of the cells with exogenous IL-2 was repeatedly found to induce a biphasic IFP-53 GAS binding activity, with the appearance of two clear bandshifts already 0.5 h after stimulation (Fig. 3A). The shifts gradually decreased at 6–24 h, with a reappearance of the strong shift at 24–48 h (Fig. 3A). Co-treatment with IFN-a did not significantly alter the initial wave of IFP-53 GAS binding activity, whereas it severely inhibited the reappearance of bandshifts during the second wave of IL-2 induced binding at 24–48 h (Fig. 3A). Furthermore, IFN-a cotreatment also caused the appearance of a novel bandshift in-between the two IL-2 induced bands (Fig. 3A). The composition of the DNA-binding complexes from T-cells stimulated with IL-2 and/or IFN-a for 0.5 and 48 h, as analysed by supershifts with specific antibodies. The IL-2 induced complexes were intact following incubation with Stat1 antibodies, while antibodies recognizing both Stat5a and b shifted the entire complex (Fig. 3B). Antibodies for Stat5a and Stat5b, respectively, partly shifted the complex (data not shown). In the IFN-a cotreated sample, the additional intermediate band was completely shifted using antibodies recognizing Stat1. In other studies, Stat1 has sometimes been found to form a slightly faster moving complex than Stat5. The reason for this discrepancy is probably due to the formation of Stat5 complexes containing different Stat5 isoforms

IL-2 induced T-cell proliferation is associated with the induction of several IL-2 responsive genes such as IL-2Ra, pim-1 and c-myc. IL-2Ra and pim-1 expression have also been shown to be regulated by Stat5 [26,27]. We have previously shown that c-myc the up-regulation in T-cells is not abrogated by cotreatment with IFN-a, but rather slightly enhanced [16]. To analyse whether the inhibitory effect of IFN-a on IL-2 induced Stat5 DNA binding correlated with an altered expression of Stat5 inducible genes, Northern blotting and hybridization using IL-2R-a and pim-1 as probes was performed. Untreated cells did not show IL-2Ra or pim-1 expression,

Fig. 2. E(cid:128)ects of IFN-a on IL-2 induced Jak3 protein expression and phosphorylation. (A) Immunoblot analysis of Jak3 expression was analysed in quiescent (day 0) or PHA activated T-cells treated with IL-2 or IL-2/IFN-a for 24 or 48 h. (B) PHA activated T-cells were treated with IL-2 alone or IL-2 together with IFN-a for 10 and 30 min Whole protein extracts were immunoprecipitated with polyclonal Jak3 antibodies followed by sequential immunoblotting with anti-phosp- hotyrosine and anti-Jak3 Ig.

(cid:211) FEBS 2002

Interferon-a inhibits Stat5 induced transcription (Eur. J. Biochem. 269) 33

a a a a

a a a a

A

N F I / 2 - L I

2 - L I

N F I

2 - L I

N F I

2 - L I

N F I

2 - L I

N F I

N F I / 2 - L I

N F I / 2 - L I

N F I / 2 - L I

Hours

6

24

48

0 PHA

0.5

Stat 5

Stat 1

Stat 5

h 8 4

h 5 . 0

a a

B

h 5 . 0

N F I / 2 - L I

h 8 4 2 - L I

A H P

h 5 . 0 2 - L I

A H P

N F I / 2 - L I

N F I

a

0

0

1 t a t S –

5 t a t S –

1 t a t S –

5 t a t S –

1 t a t S –

5 t a t S –

1 t a t S –

5 t a t S –

1 t a t S –

5 t a t S –

Antibody

a a a a a a a a a a

-

-

-

-

-

-

-

-

-

Stat 5

Stat 1

Stat 5

genes. T-cells stimulated with IL-2 for 24 and 48 h expressed continued high levels of IL-2R-a and pim-1. IFN-a cotreated cells exhibited partially reduced expression levels of IL-2R-a compared to IL-2 treated cells alone (Fig. 5), with a 30–40% reduction at 24 h and around 20% reduction at 48 h, as measured by scanning densitometry.

however, PHA treatment for 3 days induced expression of both these genes. IL-2 stimulation further increased the expression of both genes reaching a plateau between 6 and 24 h (Fig. 5). IFN-a did not inhibit the early IL-2 (2 and 6 h) induced expression and at these timepoints, IFN-a treated cells alone maintained some expression of both

Fig. 3. Stat5 binding to the IFP-53 GAS element. (A) Nuclear extracts from quiescent (day 0) or PHA activated T-cells treated with, IL-2 (0.5–48 h), IL-2/IFN-a (0.5–48 h) and IFN-a (0.5–48 h) treated T-cells were prepared and incubated with a 32P-labelled IFP-53 GAS element. DNA binding activity was analysed by EMSA. (B) Nuclear extracts from quiescent (day 0), PHA (day3), IL-2 (0.5 and 48 h), IL-2/IFN-a (0.5 and 48 h) and IFN- a (0.5 h) treated T-cells were prepared and incubated for one hour with Stat1 and Stat5 antibodies, followed by binding to a 32P-labelled IFP-53 GAS element. DNA binding activity was analysed by EMSA.

34 S. Erickson et al.

(Eur. J. Biochem. 269)

(cid:211) FEBS 2002

A

Na F I / 2 - L I

2 - L I

Na F I / 2 - L I

Na F I

2 - L I

Na F I

2 - L I

Na F I

2 - L I

Na F I

2 - L I

Na F I

Na F I / 2 - L I

Na F I / 2 - L I

Na F I / 2 - L I

48

6

4

Hours

0 PHA

0.5

2

3

Stat 5 Stat 1

B

0 PHA

IL-2 0.5h

IL-2/IFNa

0.5h

IFNa

0.5h

1 t a t S –

3 t a t S –

4 t a t S –

5 t a t S –

1 t a t S –

3 t a t S –

4 t a t S –

5 t a t S –

1 t a t S –

3 t a t S –

4 t a t S –

5 t a t S –

Antibody

-

-

-

-

-

Stat 5 Stat 1

a a a a a a a a a a a a

The effects of IFN-a on pim-1 expression on the other hand were minor. IFN-a treated cells alone did not express either of the genes after 24 h of treatment. The effects of cotreatment with IFN-a on IL-2 induced Stat5 DNA- binding activity thus partially seem to translate into effects on Stat5 dependent transcription of the IL-2Ra gene.

Effects of IFN-a on the PtdIns3K pathway

Recently, the PtdIns3K pathway has been linked to the activation of the basic cell cycle machinery in primary T-lymphocytes [28]. IL-2 induced activation of this pathway leads to down-regulation of the CKI p27, and up-regulation of D-type cyclins as well as activation of E2F driven transcription. Furthermore, IL-2 induced PtdIns3K activa- tion involves phosphorylation of Akt (protein kinase B). In order to find out whether IFN-a inhibits the activation of the PtdIns3K pathway, Akt phosphorylation in IL-2 stimulated T-cells was analysed by immunoblotting. Anti- bodies recognizing serine 473 phosphorylated Akt revealed that unstimulated T-cells expressed detectable levels of phosphorylated Akt. IL-2 stimulation only slightly increased the expression and IFN-a did not inhibit the phosphorylation at any timepoint (data not shown). Anti- bodies recognizing threonine 308 phosphorylated Akt gave an identical result (data not shown).

the receptors as well as intracellular signalling molecules such as the Jak and Stat molecules. Type I interferons signal through a similar set of Jak/Stat molecules but with completely different cellular effects. IFN-a is a well-estab- lished antiproliferative cytokine, both in malignant and nonmalignant cells. As mentioned above, we have recently shown that IFN-a inhibits IL-2 induced proliferation in PHA-activated T-lymphocytes, by abrogating the activation of the cell cycle machinery [16]. The present study investi- gates whether this effect may be mediated by cross-talk between the signalling molecules in the IL-2 and IFN-a pathways. We found that IFN-a did not affect the immediate IL-2 mediated phosphorylation of Jak3, the immediate Stat5 DNA binding or the early induction of Stat5 induced genes. However, Jak3 protein expression was severely inhibited by IFN-a after 24 h of exposure as well as the prolonged Stat5 DNA-binding. Furthermore, IFN-a partially inhibited the prolonged expression of the Stat5 responsive gene, IL-2Ra, although the reduction was not as prominent as the effect of IFN-a on Stat5 DNA-binding. The reason for the discrepancy between the effects of IFN-a on Stat5 DNA binding and the steady state mRNA levels of Stat5 responsive genes, is unclear at present. It is most likely due to regulation of such genes also by other transcription factors than the ones investigated in the present study. This is also supported by the fact that low mRNA levels of these genes appear well before there is any measurable Stat5 DNA binding activity in the cells (Figs 3,4 and 5).

D I S C U S S I O N

Induction of proliferation in human primary T-lymphocytes depends on cytokine stimulation both in vivo and in vitro, and IL-2 is a central molecule in this response. However, also IL-4, IL-7, and IL-15 all promote T-cell proliferation in a manner similar to IL-2 [8]. These cytokines share the cc in

The links between IL-2 signalling and activation of the cell cycle in T-cells are poorly understood. Activation of the PtdIns3K pathway in the human IL-2 dependent T-cell line, Kit225, induces CyclinD3 expression and down-regulation of p27, resulting in pRb phosphorylation and E2F driven transcription [28]. In this system the Stat5 and MEK/ERK2

Fig. 4. Stat binding to IL-2Ra GAS-c/GAS-n. (A) Nuclear extracts from quiescent (day 0) or DHA activated T-cells treated with, IL-2 (0.5–48 h), IL-2/IFN-a (0.5–48 h) and IFN-a (0.5–48 h) were prepared and incubated with a P32 IL-2Ra GAS-c/GAS-n labelled GAS ele- ment. DNA binding activity was analysed by EMSA. (B) Nuclear extracts from quiescent (day 0), IL-2 (0.5 h), IL-2/IFN-a (0.5 h) and IFN-a (0.5 h) -treated T-cells were prepared and incubated for one hour with Stat1, Stat3, Stat4 and Stat5 antibodies, followed by bind- ing to a P32 IL-2Ra GAS-c/GAS-n labelled GAS element. DNA binding activity was analysed by EMSA.

(cid:211) FEBS 2002

Interferon-a inhibits Stat5 induced transcription (Eur. J. Biochem. 269) 35

-

-

-

-

a a a a

-

-

-

-

2 - L I

N F I

2 - L I

N F I

2 - L I

N F I

2 - L I

N F I

N F I / 2 - L I

N F I / 2 - L I

N F I / 2 - L I

N F I / 2 - L I

a a a a

2

6

24

48

0

PHA

Hours

IL-2Ra

Pim-1

18S rRNA

28S rRNA

pathways seem dispensable for cell cycle activation. How- ever, peripheral T-lymphocytes from Stat5 double-knock- out mice (Stat5a–/–, Stat5b–/–) are profoundly deficient in their ability to proliferate in response to IL-2, whereas this is not the case for the Stat5a and Stat5b single knockouts [15]. Also, the incapability of the Stat5a/b deficient peripheral T-cells to respond to IL-2 stimulation was not to be rescued by supraphysiological concentrations of IL-2, which would bypass the requirement for the expression of IL-2Ra, suggesting a direct role of Stat5 for IL-2 induced cell cycle progression of peripheral T-lymphocytes [15].

induction of CyclinD2, CyclinD3, CyclinE, Cyclin A and Cdk6. However, downregulation of p27 and induction of Cdk2 was not affected in the Stat5a/b deficient mice [15]. Addition of IFN-a together with IL-2 to PHA-activated T-cells seems to result in a broader effect on the cell cycle machinery with inhibited induction of CyclinD3, CyclinE, Cdk2 and Cdk6 as well as an abrogated down-regulation of p27 [16]. It is therefore likely that IFN-a inhibits several pathways downstream of the IL-2 receptor to generate this prominent effect. Activation of the PtdIns3K pathway results in Akt (protein kinase B) phosphorylation and, as discussed above, a partial activation of the cell cycle machinery. As IFN-a has been demonstrated to regulate this pathway in some systems [29], we analysed whether IFN affected the IL-2 dependent phosphorylation of Akt. Quiescent T-cells exhibited quite a clear level of Akt-phos- phorylation and IL-2 only slightly enhanced this phospho- rylation. We could not detect any clear differences in

The finding that IFN-a inhibited the IL-2 induced expression of Jak3 protein levels and prolonged Stat5 DNA-binding and transcription in T-cells may be of importance in contributing to the antiproliferative pheno- type and inability to activate the cell cycle machinery in these cells. Stat5a/b deficient T-cells exhibited extensive defects in IL-2 induced cell cycle activation, with impaired

Fig. 5. IFN-a e(cid:128)ects on IL-2 induced IL-2Ra and pim-1 expression. Analysis of IL-2Ra (A) and pim-1 (B) regulation in quiescent (day 0), and PHA (day 3), IL-2 (2–48 h), IL-2/IFN-a (2–48 h) and IFN-a (2–48 h) -treated cells by Northern blot analysis. 18S and 28S rRNA serves as a control for equal loading.

36 S. Erickson et al.

(Eur. J. Biochem. 269)

(cid:211) FEBS 2002

phosphorylation levels in the IFN cotreated cells. Based on these experiments it is unlikely that IFN inhibits cell cycle activation by inhibiting the PtdIns3K pathway, however, additional experiments need to be performed to characterize this pathway greater in detail.

3. Tiefenbrun, N., Melamed, D., Levy, N., Resnitzky, D., Ho(cid:128)man, I., Reed, S.I. & Kimchi, A. (1996) Alpha interferon suppresses the cyclin D3 and cdc25A genes, leading to a reversible G0-like arrest. Mol. Cell. Biol. 16, 3934–3944.

4. Sangfelt, O., Erickson, S., Einhorn, S. & Grander, D. (1997) Induction of Cip/Kip and Ink4 cyclin dependent kinase inhibitors lines. Oncogene 14, by interferon-alpha in hematopoietic cell 415–423.

One possible explanation for the almost complete inhi- bition of the IL-2 induced proliferation could be that IFN-a inhibited the formation of a functional IL-2 receptor. However, based on our findings this seems unlikely. First, IFN-a did not inhibit the immediate IL-2 dependent Stat5 DNA binding or Jak3 phosphorylation. Secondly, we have previously shown that the induction of the oncogene c-myc is not inhibited by cotreatment with IFN-a, rather IFN seemed to synergize with IL-2 to enhance c-myc expression [16]. Taken together these data strongly indicate that the IL-2 receptor is also functional in the presence of IFN-a.

5. Sangfelt, O., Erickson, S., Castro, J., Heiden, T., Gustafsson, A., Einhorn, S. & Grander, D. (1999) Molecular mechanisms under- lying interferon-alpha-induced G0/G1 arrest: CKI-mediated regu- lation of G1 Cdk-complexes and activation of pocket proteins. Oncogene 18, 2798–2810.

6. Erickson, S., Sangfelt, O., Castro, J., Heyman, M., Einhorn, S. & Grander, D. (1999) Interferon-alpha inhibits proliferation in human T lymphocytes by abrogation of interleukin 2-induced changes in cell cycle-regulatory proteins. Cell Growth Di(cid:128)er. 10, 575–582. 7. Taniguchi, T. (1995) Cytokine signaling through nonreceptor protein tyrosine kinases. Science 268, 251–255. 8. Nelson, B.H. & Willerford, D.M. (1998) Biology of the interleu- kin-2 receptor. Adv. Immunol. 70, 1–81.

9. Minami, Y., Nakagawa, Y., Kawahara, A., Miyazaki, T., Sada, K., Yamamura, H. & Taniguchi, T. (1995) Protein tyrosine kinase Syk is associated with and activated by the IL-2 receptor: possible link with the c-myc induction pathway. Immunity 2, 89–100. 10. Higuchi, M., Asao, H., Tanaka, N., Oda, K., Takeshita, T., Nakamura, M., Van Snick, J. & Sugamura, K. (1996) Dispens- ability of Jak1 tyrosine kinase for interleukin-2-induced cell growth signaling in a human T cell line. Eur. J. Immunol. 26, 1322–1327.

11. Miyazaki, T., Kawahara, A., Fujii, H., Nakagawa, Y., Minami, Y., Liu, Z.J., Oishi, I., Silvennoinen, O., Witthuhn, B.A., Ihle, J.N., et al. (1994) Functional activation of Jak1 and Jak3 by selective association with IL-2 receptor subunits. Science 266, 1045–1047.

12. Kawahara, A., Minami, Y., Miyazaki, T., Ihle, J.N. & Taniguchi, T. (1995) Critical role of the interleukin 2 (IL-2) receptor gamma- chain- associated Jak3 in the IL-2-induced c-fos and c-myc, but not bcl-2, gene induction. Proc. Natl Acad. Sci. USA 92, 8724–8728.

In this study, we repeatedly found a biphasic activation of Stat5 in response to IL-2 (Figs 3A and 4A). One explanation for the decrease in Stat5 DNA binding between 3 and 24 h could be the expression of SOCS (suppressors of cytokine signalling). SOCS-3 has been reported to be rapidly and transiently induced in T-lymphocytes following IL-2 stim- ulation, inhibiting Stat5 phosphorylation by interacting with Jak1 and IL-2Rb [30]. Furthermore, IFN-a was also found to induce SOCS-3 [31]. One possible explanation for the low levels of Stat5 DNA binding in the IFN-a cotreated cells could thus be that SOCS-3 levels do not return to basal levels and by that inhibit further Stat5 phosphorylation. Another possibility is that the suppression of Jak3 expression (Fig. 1A) in the IFN treated cells affects the phosphorylation of the IL-2Rb chain and thereby inhibits the Stat5 activity. It has been noted that activation of Stats may also contribute to the pathogenesis of human leukaemia. Constitutive Stat activation has been reported in cell lines derived from patients with acute myelogenous leukaemia, acute lymphoblastic leukaemia, chronic myelogenous leukaemia and T-cell lymphoma [32–35]. Several of these malignancies respond favourably to IFN treatment, an effect that could theoretically be mediated via an effect on Stat5 activation. A further elucidation of the interaction between IFN-signalling and Stat5 activation in such cells may thus also have implications for the understanding of IFNs antitumor activity in these diseases.

A C K N O W L E D G E M E N T S

13. Liu, X., Robinson, G.W., Gouilleux, F., Groner, B. & Hennighausen, L. (1995) Cloning and expression of Stat5 and an additional homologue (Stat5b) involved in prolactin signal transduction in mouse mammary tissue. Proc. Natl Acad. Sci. USA 92, 8831–8835.

14. Teglund, S., McKay, C., Schuetz, E., van Deursen, J.M., Stravopodis, D., Wang, D., Brown, M., Bodner, S., Grosveld, G. & Ihle, J.N. (1998) Stat5a and Stat5b proteins have essential and nonessential, or redundant, roles in cytokine responses. Cell 93, 841–850.

15. Moriggl, R., Topham, D.J., Teglund, S., Sexl, V., McKay, C., Wang, D., Ho(cid:128)meyer, A., van Deursen, J., Sangster, M.Y., Bunting, K.D., Grosveld, G.C. & Ihle, J.N. (1999) Stat5 is required for IL-2-induced cell cycle progression of peripheral T cells. Immunity 10, 249–259.

We are grateful for the excellent technical assistance of Ms Elisabet Anderbring, Ann-Charlotte Bjo¨ rklund, Marika Ylisela¨ and Teija Westerlund. Dr Martin Corcoran is thanked for careful reading of the manuscript. This work has been supported by the Swedish Cancer Society, the Cancer Society of Stockholm, King Gustaf Vth Jubilee Fund, Alex and Eva Wallstro¨ ms Foundation Medical Research Council of the Academy of Finland, the Technology Development Center of Finland, the Sigrid Juselius Foundation and the Finnish Cancer Foundations.

R E F E R E N C E S

16. Erickson, S., Sangfelt, O., Castro, J., Heyman, M., Einhorn, S. & Grander, D. (1999) Interferon-alpha inhibits proliferation in human T lymphocytes by abrogation of interleukin 2-induced changes in cell cycle-regulatory proteins. Cell Growth Di(cid:128)er. 10, 575–582.

1. Isaacs, A. & Lindenmann, J. (1957) Virus interference. I. The 17. Erickson, S., Sangfelt, O., Heyman, M., Castro, J., Einhorn, S. & Grander, D. (1998) Involvement of the Ink4 proteins p16 and p15 in T-lymphocyte senescence. Oncogene 17, 595–602. interferon. Proc. R. Soc. Lond. Series B 147, 258–267.

18. Firpo, E.J., Ko(cid:128), A., Solomon, M.J. & Roberts, J.M. (1994) Inactivation of a Cdk2 inhibitor during interleukin 2-induced 2. Grander, D. & Einhorn, S. (1998) Interferon and malignant disease – how does it work and why doesn’t it always? Acta Oncol. 37, 331–338.

(cid:211) FEBS 2002

Interferon-a inhibits Stat5 induced transcription (Eur. J. Biochem. 269) 37

proliferation of human T lymphocytes. Mol. Cell. Biol. 14, 4889–4901.

28. Brennan, P., Babbage, J.W., Burgering, B.M., Groner, B., Reif, K. & Cantrell, D.A. (1997) Phosphatidylinositol 3-kinase couples the interleukin-2 receptor to the cell cycle regulator E2F. Immunity 7, 679–689. 19. Matikainen, S., Ronni, T., Hurme, M., Pine, R. & Julkunen, I. (1996) Retinoic acid activates interferon regulatory factor-1 gene expression in myeloid cells. Blood 88, 114–123.

29. Pfe(cid:128)er, L.M., Mullersman, J.E., Pfe(cid:128)er, S.R., Murti, A., Shi, W. & Yang, C.H. (1997) STAT3 as an adapter to couple phosphati- dylinositol 3-kinase to the IFNAR1 chain of the type I interferon receptor. Science 276, 1418–1420. 20. Nelson, B.H., Lord, J.D. & Greenberg, P.D. (1996) A membrane- proximal region of the interleukin-2 receptor gamma c chain suf- ficient for Jak kinase activation and induction of proliferation in T cells. Mol. Cell. Biol. 16, 309–317.

30. Cohney, S.J., Sanden, D., Cacalano, N.A., Yoshimura, A., Mui, A., Migone, T.S. & Johnston, J.A. (1999) SOCS-3 is tyrosine phosphorylated in response to interleukin-2 and suppresses STAT5 phosphorylation and lymphocyte proliferation. Mol. Cell. Biol. 19, 4980–4988. 21. Nelson, B.H., McIntosh, B.C., Rosencrans, L.L. & Greenberg, P.D. (1997) Requirement for an initial signal from the membrane- proximal region of the interleukin 2 receptor gamma (c) chain for Janus kinase activation leading to T cell proliferation. Proc. Natl Acad. Sci. USA 94, 1878–1883.

31. Brender, C., Nielson, M., Ropke, C., Nissen, M.H., Svejgaard, A., Billestrup, N., Geisler, C. & Odum, N. (2001) Interferon-alpha induces transient suppressors of cytokine signalling expression in human T cells. Exp. Clin. Immunogenet. 18, 80–85. 22. Eilers, A., Baccarini, M., Horn, F., Hipskind, R.A., Schindler, C. & Decker, T. (1994) A factor induced by di(cid:128)erentiation signals in cells of the macrophage lineage binds to the gamma interferon activation site. Mol. Cell. Biol. 14, 1364–1373.

32. Gouilleux-Gruart, V., Gouilleux, F., Desaint, C., Claisse, J.F., Capiod, J.C., Delobel, J., Weber-Nordt, R., Dusanter-Fourt, I., Dreyfus, F., Groner, B. & Prin, L. (1996) STAT-related tran- scription factors are constitutively activated in peripheral blood cells from acute leukemia patients. Blood 87, 1692–1697. 23. Meinke, A., Barahmand-Pour, F., Wo¨ hrl, S., Stoiber, D. & Decker, T. (1996) Activation of di(cid:128)erent Stat5 isoforms contri- butes to cell-type restricted signaling in response to interferons. Mol. Cell. Biol. 16, 6937–6944.

24. Azam, A., Lee, C., Strehlow, I. & Schindler, C. (1997) Function- ally distinct isoforms of STAT5 are generated by protein pro- cessing. Immunity 6, 691–701.

33. Weber-Nordt, R.M., Egen, C., Wehinger, J., Ludwig, W., Gouilleux-Gruart, V., Mertelsmann, R. & Finke, J. (1996) Constitutive activation of STAT proteins in primary lymphoid and myeloid leukemia cells and in Epstein–Barr virus (EBV)- related lymphoma cell lines. Blood 88, 809–816.

25. John, S., Robbins, C.M. & Leonard, W.J. (1996) An IL-2 response element in the human IL-2 receptor alpha chain promoter is a composite element that binds Stat5, Elf-1, HMG-I (Y) and a GATA family protein. EMBO J. 15, 5627–5635.

34. Shuai, K., Halpern, J., ten Hoeve, J., Rao, X. & Sawyers, C.L. (1996) Constitutive activation of STAT5 by the BCR-ABL oncogene in chronic myelogenous leukemia. Oncogene 13, 247–254. 35. Zhang, Q., Nowak,

26. Matikainen, S., Sareneva, T., Ronni, T., Lehtonen, A., Koskinen, P.J. & Julkunen, I. (1999) Interferon-alpha activates multiple STAT proteins and upregulates proliferation-associated IL-2Ral- pha, c-myc, and pim-1 genes in human T cells. Blood 93, 1980–1991.

I., Vonderheid, E.C., Rook, A.H., Kadin, M.E., Nowell, P.C., Shaw, L.M. & Wasik, M.A. (1996) Activation of Jak/STAT proteins involved in signal transduction pathway mediated by receptor for interleukin 2 in malignant T lymphocytes derived from cutaneous anaplastic large T-cell lymphoma and Sezary syndrome. Proc. Natl Acad. Sci. USA 93, 9148–9153. 27. Mui, A.L., Wakao, H., Kinoshita, T., Kitamura, T. & Miyajima, A. (1996) Suppression of interleukin-3-induced gene expression by a C-terminal truncated Stat5: role of Stat5 in proliferation. EMBO J. 15, 2425–2433.