Down-Modulation of Responses to Type I IFN Upon T Cell Activation

Purification and stimulation of naive CD4⁺ and CD8⁺ T cells

Human neonatal leukocytes were isolated from freshly collected heparinized neonatal blood by Ficoll-Paque density gradient centrifugation. CD4⁺/CD8⁺ T cells were purified by negative selection using a pan T isolation kit (Miltenyi Biotec, Auburn, CA). CD8⁺ T cells were then purified by positive selection with anti-CD8 microbeads (Miltenyi Biotec). The purity of the CD4⁺/CD45RA⁺ and CD8⁺/CD45RA⁺ T cells was >98%, as determined by flow cytometry. Naive CD4⁺ and CD8⁺ cells were stimulated with plate-bound anti-CD3 (TR66 (10)) and anti-CD28 Abs (BD PharMingen, San Diego, CA). When indicated, naive CD4⁺ cells were stimulated in the presence of 2.5 ng/ml IL-12 (Hoffmann-LaRoche, Nutley, NJ) and 200 ng/ml neutralizing anti-IL-4 Abs (BD PharMingen) for Th1 cultures; or 1 ng/ml IL-4 (BD PharMingen), 2 μg/ml neutralizing anti-IL-12 17F7 and 20C2 Abs (a gift from U. Gubler, Hoffman-LaRoche, Nutley, NJ), and 200 ng/ml neutralizing anti-IFN-γ (BD PharMingen) for Th2 cultures; or neutralizing anti-IL-4, anti-IL-12, and anti-IFN-γ Abs for Th0 cultures. Cells were washed on day 3 and expanded in RPMI medium supplemented with 10% FCS and 50 U/ml IL-2 (Chiron, Emeryville, CA). The specific polarized cytokine production was confirmed by single cell analysis of intracellular IFN-γ and IL-4 production. Briefly, T cells were collected 6 days after priming, washed, and stimulated for 2 h at 37°C with PMA (50 ng/ml; Sigma-Aldrich, St. Louis, MO) and ionomycin (1 μg/ml; Sigma-Aldrich). The cultures were incubated for an additional 2 h after adding brefeldin A (10 μg/ml; Sigma-Aldrich). Cells were then fixed with 4% paraformaldehyde, permeabilized with saponin, and stained with anti-human IFN-γ FITC and anti-human IL-4 PE (BD PharMingen). Cells were analyzed with a FACScan flow cytometer (BD Biosciences, Le Pont de Claix, France). A representative cytokine profile is shown in Fig. 3⇓A.

T cell proliferation assays

Naive CD4⁺ T cells were incubated in the presence or absence of 1 nM IFN-α2 for 20 h. Cells were then washed and stained with 2.5 μM CFSE (Molecular Probes, Eudene, OR) for 10 min at 37°C, before activation with anti-CD3/CD28 under polarizing or nonpolarizing conditions, as described above. Untreated cells were activated in the presence or absence of 1 nM IFN-α2 (see scheme in Fig. 1⇓). At times indicated, cells were analyzed for CFSE intensity as a function of cell division. Cells were analyzed with a FACScan flow cytometer (BD Biosciences). Dead cells were distinguished by their characteristics on the forward vs side scatter bit maps from the cytofluorograph. This method was in agreement with the analysis of 7-aminoactinomycin D staining (11). Anti-p27 Ab was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). In another set of experiments, anti-CD3/CD28-activated and polarized cells were harvested 2 days after stimulation (see scheme in Fig. 2⇓) and seeded in triplicates in the presence or absence of 1 nM IFN-α2 and in the presence of IL-2 (50 U/ml). Cell proliferation was assessed at the indicated times, by measuring [³H]thymidine incorporation. A total of 1 μCi of [³H]thymidine was added for 16 h to the cultures.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Effect of IFN-α2 on cell cycle entry of human naive T cells. A, Purified naive CD4⁺ T cells were preincubated (dashed line) or not (solid line) for 20 h with 1 nM IFN-α2, then labeled with CFSE and activated with mAbs to CD3/CD28, under Th1-polarizing conditions. IFN-α2 was constantly maintained in the medium of preincubated cells. Untreated cells were activated in the presence (dashed line) or absence (solid line) of IFN-α2. At 48 h (left panels) and 60 h (right panels) after activation, cells were harvested and analyzed for dilution of CFSE intensity by FACS. Numbers indicate the percentages of cells that entered each discrete cell division. Data represent one of three comparable experiments performed with cells from different donors. B, p27 levels in nonactivated and activated cells, treated or not with IFN-α2. Total lysates (10 μg) prepared from nonactivated naive cells incubated or not with IFN-α2 for 20 h (lanes 1 and 2) and from activated cells treated as described in A and harvested at 48 h (lanes 3, 4, and 5) and 60 h (lanes 6, 7, and 8) after activation were resolved by 7% SDS/PAGE. p27 levels were detected by immunoblotting with anti-p27. Equal protein loading was revealed by Red Ponceau staining.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Effect of IFN-α2 on the proliferation of activated T cells. Naive CD4⁺ cells were activated with mAbs to CD3/CD28 under Th2-polarizing (left panel) or Th0-nonpolarizing (right panel) conditions. After 2 days, cells were harvested and replated in the presence (dashed line) or absence (solid line) of 1 nM IFN-α2. Cell proliferation was assessed for the following 4 days, by measuring [³H]thymidine incorporated during a 16-h pulse. The data show the means and the SEs of triplicate cultures. For some of the data points, SEs were too small to be illustrated. Similar results were obtained with cells derived from two other donors and by directly counting viable cells, using trypan blue exclusion (data not shown).

Quantitative RT-PCR

Total RNA from cells treated with increasing doses of IFN-α2 and IFN-β (ranging from 1 pM to 1 nM) for different times was purified using the High Pure RNA Isolation kit (Roche Diagnostics, Mannheim, Germany). Reverse transcriptions were primed with oligo(dT) and performed using the murine leukemia virus reverse transcriptase from Invitrogen Life Technologies (Cergy Pontoise, France). Quantitative PCR assays were done at least in triplicates using the SYBR Green I technology on a LightCycler (Roche Diagnostics). The primer pairs used were: GAPDH forward, 5′-ACAGTCCATGCCATCACTGCC-3′, and reverse, 5′-GCCTGCTTCACCACCTTCTTG-3′; 6-16 forward, 5′-CATGCGGCAGAAGGCGGTAT3′, and reverse, 5′-CGACGGCCATGAAGGTCAGG-3′; MxA forward, 5′-ATCCTGGGATTTTGGGGCTT3′, and reverse, 5′-CCGCTTGTCGC TGGTGTCG-3′; IRF7 forward, 5′- GAGCCCTTACCTCCCCTGTTAT3′, and reverse, 5′-CCACTGCAGCCCCTCATAG-3′, 2′, 5′; oligoadenylate synthetase (25A69) forward, 5′-AACTGCTTCCGACAATCAAC-3′, and reverse, 5′-CCTCCTTCTCCCTCCAAAA-3′. Quantification standard curves were obtained using PCR products diluted in 10 μg/ml sonicated salmon sperm DNA. The specificity of PCR products was checked by melting curve analysis and DNA sequencing. Normalization of the IFN-stimulated gene (ISG) expression was done against GAPDH. Only ratios with a SE <0.2 log (95% confidence limits) were considered for the determination of induction levels.

Antiviral assay

Th1 and Th2 cells were treated or not with 1 nM IFN-α2 for 24 h. Cells were then infected with vesicular stomatitis virus (VSV) at a multiplicity of infection >1 for 1 h, then extensively washed and incubated for 8 h, a time corresponding to one virus replication cycle. Cells were lysed by three cycles of freezing/thawing, and the VSV-containing medium was titered on murine L cells for cytopathic effect by a limiting dilution assay.

FACS analysis and Abs

Surface type I IFN receptor 1 (IFNAR1) expression was monitored with 10 μg/ml of AA3 mAb (a gift of L. Runkel, Biogen, Cambridge, MA (12)). IFNAR2 was monitored with the MMHAR-2 mAb (PBL; Biochemical Laboratories, New Brunswick, NJ), followed by incubation with 10 μg/ml of biotinylated anti-mouse IgG Ab and streptavidin-PE (Jackson ImmunoResearch Laboratories, West Grove, PA). Surface IL-12Rβ1 expression was monitored with 5 μg/ml of the rat anti-human IL-12Rβ1 2B10 mAb (a gift of U. Gubler) (13), followed by incubation with biotinylated polyclonal anti-rat IgG Abs and with streptavidin-PE. Anti-CD69 PE and anti-CD25 FITC Abs were purchased from BD PharMingen. Anti-phospho-STAT1 were purchased from New England Biolabs (Beverly, MA), and anti-STAT2 from Upstate Biotechnology (Lake Placid, NY). Anti-phospho-STAT2 was a gift from D. E. Levy (New York University School of Medicine, New York, NY), and anti-STAT1 was a gift from C. W. Schindler (Columbia University, New York, NY).

Electrophoretic mobility shift assay

Gel shift assays were performed, as described previously (14). Briefly, naive and Th1/Th2 cells (2 × 10⁷ cells) were incubated or not with 1 nM IFN-α2 for 1 h, and nuclear extracts were prepared, as described previously (15). A total of 15 μg of nuclear extracts were incubated with a ³²P-labeled DNA probe corresponding to the ISG15 gene IFN-stimulated response element (5′-GATCCTCGGGAAAGGGAAACCGAAACTGAAGCC-3′), fractionated on a nondenaturing polyacrylamide gel, and autoradiographed.

IFN-α delays cell cycle entry of naive T cells

To evaluate the effect of IFN-α on T cell activation, we first studied whether IFN-α affected the entry of naive cells into the cell cycle. Human naive CD4⁺ cells were purified from neonatal leukocytes and incubated or not with 1 nM IFN-α2 for 20 h. Cells were then labeled with CFSE and activated using plate-bound anti-CD3 and anti-CD28 mAbs, under conditions allowing the establishment of Th1 cell lines (9). Cells that had not been pretreated with IFN were divided into two pools, one of which was activated in the presence of IFN (scheme in Fig. 1⇑A). At 48 h postactivation, the analysis of CFSE fluorescence dilution showed that cells pretreated with IFN had not entered into the cell cycle, while ∼24% of untreated cells were dividing (Fig. 1⇑A, top and middle panels). Conversely, the IFN treatment of cells at the time of activation did not affect their proliferation, and CFSE profiles were comparable to those obtained for untreated cells (Fig. 1⇑A, middle and bottom panels). Similar results were obtained with cells polarized under Th2 or nonpolarizing Th0 conditions (data not shown), suggesting that IFN acts independently of the presence of polarizing cytokines. The effect of IFN was reproducible, and the analysis of three independent donors showed inhibition of proliferation after IFN pretreatment, ranging from 50 to 100%. When CFSE fluorescence was analyzed at 60 h postactivation, the difference in the percentages of dividing cells between treated and untreated samples was reduced (Fig. 1⇑A, right panels), demonstrating that IFN does not block, but rather delays cell cycle entry.

The cyclin-dependent kinase inhibitor p27^Kip1 is a negative regulator of the cell cycle and, in particular, a candidate target of cytokine-mediated T cell growth regulation (16, 17). Therefore, we monitored the level of p27 within the same cell samples used for CFSE analysis. At 48 h postactivation, the p27 level was considerably reduced in both untreated cells and in cells treated with IFN at the time of activation, in comparison with resting naive cells (Fig. 1⇑B, compare lanes 4 and 5 with lane 1). Conversely, cells that had been treated with IFN before activation showed a consistent 2-fold increased level of p27 in comparison with untreated cells (Fig. 1B⇑, lane 3). At 60 h postactivation, the p27 level in treated and untreated cells was comparable (Fig. 1⇑B, lanes 6–8). Thus, the p27 content appears to correlate with the CFSE profiles observed in the different cell samples. We also compared p27 levels in nonactivated naive cells incubated or not with IFN for 20 h. No detectable increase was observed upon IFN treatment (Fig. 1⇑B, lanes 1 and 2).

Next, we investigated the effect of IFN-α2 on the ongoing proliferation of 2-day-activated CD4⁺ T cells. Naive precursors were activated with CD3/CD28 mAbs under Th2 or Th0 conditions. Two days postactivation, cells were replated in the presence or absence of IFN-α2, and proliferation was measured by [³H]thymidine incorporation. In contrast to the results obtained in naive cells, the proliferation of activated cells was not affected by IFN (Fig. 2⇑). Similar experiments performed with Th1, Th2, and Th0 cells from three independent donors yielded comparable results (data not shown).

Overall, these results demonstrated a different susceptibility of naive and effector T cells to the antiproliferative activity of IFN-α.

IFN-α/β responsiveness is down-modulated upon T cell activation

Type I IFN induce the expression of a large set of ISGs, whose promotors are activated by the trimeric ISG factor 3 (ISGF3) complex, composed by STAT1, STAT2, and IRF-9 (18). To evaluate the IFN responsiveness of T lymphocytes upon differentiation, we measured the transcriptional response of naive and effector T cells, by quantifying the level of expression of some ISGs. In pilot experiments, the accumulation of ISG mRNA in IFN-α-treated naive cells was found to peak at 4–6 h (data not shown). Naive cells were activated for 6 days under Th1-polarizing and Th0-nonpolarizing conditions (see the cytokine production profiles in Fig. 3⇓A). Naive and activated cells were incubated for 4 h with increasing doses of IFN-α2 or IFN-β, and total RNA was extracted. The levels of the ISG mRNA were quantified by real-time PCR. The GAPDH gene was chosen as the internal endogenous reference, because its expression was not affected by IFN treatment (data not shown). After normalizing ISG mRNA levels to the GAPDH level in each sample, the fold of induction of each gene was expressed as the ratio between treated and untreated samples in each cell subset. In Fig. 3⇓B, the fold of induction of the 6-16 gene was plotted as a function of IFN concentration. In naive cells, the 6-16 gene was induced 90-fold in response to as little as 1 pM of IFN-α2. Conversely, in both activated T cell populations, this gene was induced ∼10-fold. Importantly, this difference could not be overcome by higher doses of IFN. The quantification of the 6-16 mRNA induced by IFN-α2 and IFN-β in cells derived from independent donors showed similar results, confirming that the 6-16 gene was at least 10-fold less induced in activated cells than in naive precursors. The induction profile of three other ISGs (MxA, the 69-kDa 2′-5′ oligoadenylate synthetase, and IRF7) was studied and, depending on the gene, a 15- to 50-fold lower induction was observed in activated cells (Fig. 3⇓C). The phenomenon described above was not restricted to the CD4⁺ subset, because an average 10-fold reduced induction of the 6-16 mRNA by IFN-α2 was also observed in activated CD8⁺ cells, as compared with their naive precursors (Fig. 3⇓D).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

IFN-induced gene expression in naive and in activated T cells. A, Cytokine profiles of 6-day-activated Th1 (upper panel) and Th0 (lower panel) cells. Purified naive cells were activated with mAbs to CD3/CD28 under Th1-polarizing or Th0-nonpolarizing conditions. At day 6 after priming, cells were harvested, washed, and restimulated. The intracellular production of IFN-γ and IL-4 was analyzed by flow cytometry. B, Induction of the 6-16 mRNA by IFN-α2. Naive and 6-day-polarized Th1 cells were treated with increasing doses of IFN-α2 for 4 h, total RNA was extracted, and the 6-16 mRNA levels were measured by real-time PCR. The 6-16 transcripts were normalized to simultaneous quantification of GAPDH transcripts. The ratios between treated and untreated samples in each cell subset are shown, taking as 1 the ratio in untreated samples. C, Induction of the 6-16, MxA, 69-kDa 2′-5′ oligoadenylate synthetase (25A69), and IRF7 mRNAs by IFN-α2 and IFN-β. Naive and 6-day-polarized Th1 cells were treated with 1 nM IFN-α2 or 1 nM IFN-β for 4 h. The mRNA induction, calculated as in B, is presented here in log scale, due to the very low level in the activated subsets. D, 6-16 mRNA induction in naive and activated CD8⁺ cells. Purified naive CD8⁺ cells were activated with mAbs to CD3/CD28 for 6 days. Naive and activated cells were treated with increasing doses of IFN-α2 for 4 h. The induction of 6-16 mRNA was assessed as in A.

The transcriptional response to type I IFN was also measured at earlier stages of Th cell differentiation. Purified naive CD4⁺ cells from three independent donors were activated under Th1, Th2, or Th0 conditions. The 6-16 mRNA induction by IFN-α2 was measured in naive cells and in cells harvested at 24, 48, and 72 h after activation. Independently of the commitment to a specific Th subset, the reduction of the IFN response was observed at 24 h after activation and was maximal at 72 h, when most of the cells had started to proliferate (Fig. 4⇓A). Overall, these results suggested that the decreased transcriptional response to IFN of differentiated Th subsets was related to the activation/proliferation of the cells, rather than to a subset-specific polarization event. These results also raised the question as to whether the altered IFN-α response of effector cells was linked to the activation process itself, or whether it was associated to proliferation. To this end, we performed an experiment in which cells were monitored in parallel for surface expression of activation markers (CD69 and CD25), for CFSE fluorescence dilution, and for IFN responsiveness. These analyses were conducted on naive cells and on cells activated for 14 and 64 h under Th1-polarizing conditions. At 14 h postactivation, cells had not yet entered proliferation, while over 60% of them expressed CD69 and CD25 markers (Fig. 4⇓B). Importantly, these cells displayed a 12-fold reduced 6-16 mRNA accumulation as compared with naive cells. At 64 h postactivation, all cells were proliferating, and a further 2-fold reduction in the transcriptional response to IFN-α was measured. These results demonstrated that the decline in the IFN response occurred early upon T cell activation, before entry into proliferation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Induction of 6-16 mRNA by IFN-α2 at early times after activation. A, Neonatal naive CD4⁺ cells were purified from three independent donors and activated under Th1, Th2, or Th0 conditions. Nonactivated (time 0, naive cells) and activated cells were harvested after 24, 48, and 72 h; incubated for 4 h in the presence or absence of 1 nM IFN-α2; and analyzed for 6-16 and GAPDH transcripts. The ratio 6-16/GAPDH in the treated sample was normalized to the same ratio in the untreated sample and shown as the mRNA induction. B, Neonatal naive cells, purified from an independent donor, were activated under Th1 conditions. Cells were harvested after 14 and 64 h and incubated for 4 h in the presence or absence of 1 nM IFN-α2. The induction of 6-16 mRNA was assessed as in A. The expression of the activation markers CD69 and CD25 (right, upper, and middle panels) was monitored on nonactivated and 14-h-activated cells by staining with anti-CD69 PE and anti-CD25 FITC Abs. The rate of proliferation of nonactivated and 14 h- and 64 h-activated cells was measured by dilution of CFSE fluorescence (right, lower panel).

Activated T cells are sensitive to the antiviral effect of IFN-α

Because type I IFN play a determinant role in protecting cells from viruses, we asked whether, despite the lower magnitude of ISG induction, activated T cells were protected by IFN from viral infection. To answer this question, we tested whether IFN-α2 induced an antiviral state against VSV in Th1 and Th2 effector cells. The two populations were treated with IFN-α2 and exposed to VSV, and, after a single round of virus replication, the virus yield was measured. An IFN-α2-mediated reduction in the VSV yield was observed in both Th1 and Th2 cells (Fig. 5⇓), suggesting that the residual IFN response measured in these cells is sufficient to induce a complete antiviral response.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

IFN-α protects activated cells from VSV infection. Th1 and Th2 cells polarized for 6 days were left untreated or were treated with 1 nM IFN-α2, then infected with VSV (multiplicity of infection >1). The virus yield was assayed on L murine cells. Striped columns represent the residual virus after the infection step, open columns represent virus produced by untreated cells, and filled columns represent virus produced by IFN-treated cells.

The IFN-induced Janus kinase/STAT pathway is not impaired in polarized Th subsets

To investigate the mechanism underlying the decline in the transcriptional response to type I IFN observed in activated lymphocytes, we monitored the cell surface level of the IFN receptor chains, IFNAR1 and IFNAR2, and, in parallel, the level of the IL-12Rβ1 chain, in naive and in polarized cells at different times after activation. The level of IFNAR2 remained constant in the three cell subsets (data not shown), while the level of IFNAR1 declined after activation, so that 6-day-activated cells expressed nearly half the level present on precursor cells (Fig. 6⇓A). A rapid increase in the IL-12Rβ1 level was observed in both Th populations (Fig. 6⇓A, right panel). To evaluate the consequence of the IFNAR1 reduction, we measured the extent of activation of STAT1 and STAT2, in naive and in differentiated cells. The two forms of STAT1 (p91 and p84) were tyrosine phosphorylated by IFN-α2 in the three cell subsets (Fig. 6⇓B). The magnitude of phosphorylation was difficult to compare due to the increase of both STAT1 isoforms in Th1 cells (19). However, taking into account the relative amount of proteins, the percentage of phosphorylated p91 and p84 did not remarkably change in activated cells in comparison with naive precursors. Conversely, the STAT2 content remained constant and was therefore regarded as more indicative (Fig. 6⇓B). STAT2 was tyrosine phosphorylated by IFN-α2 to comparable levels, in a dose-dependent fashion and with similar kinetics, in the three cell subsets (Fig. 6⇓, B and C). No difference in the induced phosphorylation level of STAT3 could be observed (data not shown). Next, we measured, by EMSA, the activation of the transcriptional ISGF3 complex in nuclear extracts obtained from naive and differentiated cells that were treated for 1 h with IFN-α2. As shown in Fig. 6⇓D, a comparable level of activated ISGF3 was detected in the three cell subsets. Overall, these results demonstrate that the decline in the transcriptional response of Th1/Th2 to type I IFN is not due to a reduced sensitivity of these cells to the ligand, nor to impairment of receptor-generated signaling events, but rather may result from global changes occurring upon activation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

IFN-α2 signaling pathway in naive and activated Th cells. A, Surface level of IFNAR1 and IL-12Rβ1 during Th cell differentiation. The surface levels of IFNAR1 (left panel) and IL-12Rβ1 (right panel) were monitored by flow cytometry, at the indicated days after T cell activation (time 0). Shown is the median fluorescence intensity obtained in one of three independent experiments. B, Tyrosine phosphorylation level of STAT1 and STAT2 in naive, Th1, and Th2 cells. Naive precursors and 6-day-polarized Th1 and Th2 cells were treated for 30 min with the indicated doses of IFN-α2. Total cell lysates (40 μg for naive cells and 20 μg for Th1/Th2 cells) were resolved by 7% SDS-PAGE. Tyrosine-phosphorylated STAT1 and STAT2 were detected by immunoblotting with specific anti-phospho-Abs. Membranes were stripped and probed with the specific STAT Abs. The anti-STAT1 antiserum detects both p91 and p84 isoforms. C, Kinetics of STAT2 activation in naive and Th1 cells. Naive and 6-day-differentiated Th1 cells were treated for the indicated times with 50 pM IFN-α2. Lysates were analyzed as in B. D, Induction of the ISGF3 complex by IFN-α2. Naive, Th1, and Th2 cells were treated for 1 h with 1 nM IFN-α2. Nuclear extracts (15 μg) were subjected to EMSA, using a ³²P dsDNA-labeled probe corresponding to the IFN-stimulated response element of the ISG15 gene. When indicated, 10× cold probe was added as a competitor.