Differential Expression and Function of Phosphodiesterase 4 (PDE4) Subtypes in Human Primary CD4⁺ T Cells: Predominant Role of PDE4D

Reagents, inhibitors, primer/probe sets, Abs, and siRNA

Percoll, Sepharose A, and G, 5,8-[³H]cAMP, and [³H]thymidine were obtained from Amersham Biosciences (GE Healthcare). 4-(2-Aminoethyl)-benzenesulfonyl fluoride and complete mini protease inhibitor (EDTA-free) were from Roche. Medium and solutions used for culturing of CD4⁺ T cells were purchased from Invitrogen Life Technologies. All other chemicals not specifically mentioned were of analytical grade and were obtained from Sigma-Aldrich. The PDE4 inhibitor RP73401 (piclamilast, WO9212961) was synthesized at the chemical facilities of Altana Pharma. The PDE4 inhibitor rolipram was purchased from Sigma-Aldrich.

Forward/reverse primer and probe sets and TaqMan universal PCR master mix for quantitative PCR were purchased from Applied Biosystems. The designed primer and probe sets were as follows: 18S rRNA (X03205) forward, CGGCTACCACATCCAAGGAA, reverse, GCT GGAATTACCGCGGCT, probe VIC-TGCTGGCACCAGACTTGCCCTC; PDE4A (NM_006202) forward, GTGGCTCCGGATGAGTTCTC, reverse, GGGCTGCTGTGGCTTACAG, probe FAM-CCGGGAGGAATTCGTGGT-minor groove binder (MGB); PDE4B (NM_002600) forward, AGCAGCACAAAGACGCTTTGT, reverse, TCAGTCTCTCCCAGGGAATCTC, probe FAM-TGATTGATCCAGAAAAC-MGB; PDE4C (NM_000923) forward, ACTCTGGAGGAGGCAGAGGAA, reverse, AGGCAACTCCAAGGCCTCTT, probe FAM-AAGAGACAGCTTTAGCC-MGB; PDE4D (NM_006203) forward, GGCAGGGTCAAACTGAGAAATT, reverse TGACTGCCACTGTCCTTTTCC, probe FAM-TAGAGGAAGATGGTGAGTCAG-MGB. The PDE4 subtype primer/probe sets have been designed to bind at the 3′ end of the open reading frame of each subtype, at positions where individual splice variants of one subtype are identical.

For the stimulation of human primary CD4⁺ T cells, anti-CD3 mAb (Orthoclone OKT-3; Janssen-Cilag) and anti-CD28 mAb (clone CD28.2; Immunotech) were used. The rabbit polyclonal Abs AC55 (anti-PDE4A) and K118 (anti-PDE4B) as well as the murine mAb m3S1 (anti-PDE4D), all applied for immunoprecipitations, were used as described (32). Purified mouse myeloma IgG1 as isotype control was purchased from Zymed (Invitrogen Life Technologies). For immunoblotting experiments, rabbit polyclonal Abs raised against PDE4A, PDE4B, and PDE4D were used as detailed elsewhere (33). Mouse monoclonal anti-β-actin Abs were purchased from Sigma-Aldrich. siRNAs were from Dharmacon (Lafayette): PDE4A_si (D-007647-04, GUAACAGCCUGAACAACUC), PDE4B_si (D-007648-04, GAAAGAGACCUCCUAAAGA), PDE4D_si (D-004757-06, GAACUUGCCUUGAUGUACA), PDE4D_si-2 (D-004757-05, GAAAUCAAGUGUCAGAGUU). As control, nontargeting NEG_si (more than four mismatches to every known human gene, D-001210-01) was used.

Isolation and cytometrical analysis of primary CD4⁺ T cells

Research conducted for this study with human material has been approved by the local ethics committee. For the isolation of human peripheral CD4⁺ T cells by negative selection, 250 ml of blood from normal, healthy donors were taken by antecubital venipuncture, 0.3% (w/v) citrate-treated, and diluted 1.6-fold with PBS before centrifugation at room temperature for 10 min at 220 × g. The lower phase was layered on a Percoll gradient (density 1.077 g/ml) and subject for centrifugation for 25 min at 800 × g. The interphase containing the PBMC was washed twice in PBS containing 2 mM EDTA and was then resuspended on ice in PBS, 0.5% (v/v) heat-inactivated FCS, 2 mM EDTA and prepared for magnetical separation using the autoMACS system (Miltenyi Biotec), as described by the manufacturer (CD4⁺ T Cell Isolation kit II, Program Deplete; Miltenyi Biotec). For the isolation of naive (CD45RA⁺) and memory (CD45RO⁺) T cells, CD4⁺ T cells of one donor were separated into two parts, and each part was labeled with anti-CD45RA or anti-CD45RO microbeads (Miltenyi Biotec), respectively, for magnetic separation.

For flow cytometrical analysis (Coulter Epics XL-MCL; Beckman Coulter), 2–5 × 10⁵ cells in 100 μl of staining buffer (PBS, 2% (v/v) heat-inactivated FCS) were labeled with Abs directed against surface markers CD4 and CD25 (Immunotech). Surface marker-specific Abs were used in combination with their respective isotype controls. The dye 7-aminoactinomycin D (7-AAD) was used to check for viability (Immunotech).

Transfection and stimulation of CD4⁺ T cells

Transfection of human peripheral, freshly isolated CD4⁺ T cells (5 × 10⁶ cells/transfection) was conducted using the nucleofector kit for unstimulated human T cells according to the instructions of the manufacturer (nucleofector device I, program U-14; Amaxa) (34). At standard conditions, final siRNA concentrations were 1.5 μM in a total volume of 100 μl. Immediately after nucleofection (NF), cells were recovered with 2.5 ml of prewarmed complemented medium (RPMI 1640 medium, 10% (v/v) heat-inactivated FCS, 1% (v/v) l-glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin), transferred to 6-well plates (Corning Costar), and incubated in a humidified incubator at 37°C/5% CO₂. For control experiments, nucleofection was performed without siRNA (= NF control) or with nontargeting siRNA (= NEG_si). After 24 h of culturing, transfected cells were checked for viability by flow cytometry and diluted to a concentration of 1 × 10⁶ cells/ml. To test the nucleofection efficiency, different concentrations of FITC-labeled nontargeting siRNA (BLOCK-iT Fluorescent Oligo, Sequitur; Invitrogen Life Technologies) were transfected.

For functional analysis, CD4⁺ T cells were stimulated via the CD3 chain of the TCR and the costimulatory molecule CD28, similar to a protocol described elsewhere (24). To coat 96-well microtiter plates (Microtest tissue culture plate, Falcon; BD Biosciences) with anti-CD3 mAbs, the plates were incubated for 2.5 h at 37°C/5% CO₂ with 50 μl of a 6 or 60 μg/ml solution of anti-CD3 in PBS, corresponding to 0.3 or 3 μg anti-CD3/well, respectively. A total of 2 × 10⁵ cells in a total volume of 200 μl was used per well. For costimulation, soluble anti-CD28 mAb in a volume of 10 μl was added to each well to obtain final concentrations of ∼0.3 or ∼3 μg of anti-CD28/ml, respectively.

Optionally, the PDE4 inhibitor RP73401 (piclamilast) was included at a final concentration of 1 μM. Stock solutions of RP73401 in DMSO were diluted 1/1000 (v/v) to result in a final DMSO concentration of 0.1% (v/v) in the assay. Cells were incubated at 37°C/5% CO₂, for several time periods, as indicated. Every condition was performed at least in duplicates.

RNA isolation, cDNA synthesis, and quantitative PCR

A total of 0.1–1 × 10⁷ CD4⁺ T cells were washed in PBS, lysed with 350 μl of RNA lyses tissue buffer containing 1% 2-ME, and homogenized with QIAshredder spin columns (Qiagen). Total RNA was isolated using RNeasy mini columns according to the instructions of the manufacturer (Qiagen). RNA concentration was determined by spectrophotometric analysis using a SpectraMax 190 (Molecular Devices) or a NanoDrop ND-1000. The average yield was determined to be ∼0.8 μg of total RNA per 1 × 10⁶ unstimulated human primary CD4⁺ T cells. Reverse transcription was performed with 0.5–1 μg of RNA using AMV reverse transcriptase (Roche).

Quantitative PCR were performed on an ABI Prism 7900 HT sequence detection system (Applied Biosystems). Primer and probe sets have been tested for specificity and were optimized to have similar maximal amplification efficiencies (95 ± 3%, data not shown) in duplex reactions (i.e., combined detection of 18S rRNA and a target gene). Standard quantitative PCR were performed in a total volume of 25 μl in triplicates, each containing 10 ng of cDNA as template, 50 nM of primer and probe detecting 18S rRNA, and 900 nM/200 nM primer/probe detecting the target gene in 1× TaqMan universal PCR master mix (Applied Biosystems). The thermal cycler protocol consisted of a 2-min period at 50°C (uracil removal for PCR carryover protection), a 10-min period at 95°C (denaturation of native DNA), and of 40 cycles of 15 s at 95°C (denaturation of PCR product) and 1 min at 60°C (annealing/extension). As controls for DNA contamination, total RNA and water were taken as templates. For control experiments, plasmids were used containing conserved C-terminal parts of the subtypes PDE4A, PDE4B, PDE4C, and PDE4D. Quantification of PCR experiments was based on the comparative ΔΔ cycle threshold (C_t) method as described elsewhere (35) using either unstimulated cells or nucleofection control cells as calibrator.

Immunoprecipitation and PDE activity measurement

Cells were washed in PBS and resuspended in lysis buffer (50 mM Tris-HCl (pH 7.4), 10 mM sodium fluoride, 1 mM EDTA, 0.2 mM EGTA, 10 mM sodium pyrophosphate, 250 mM sodium chloride, 5% glycerol, 1 mM 4-[2-aminoethyl]-benzenesulfonyl fluoride, 1 tablet/10 ml complete protease inhibitor, 1 mM sodium orthovanadate, 1% Nonidet P-40 (Calbiochem), and 15 mM 2-ME). After sonication, the lysate was centrifuged at 4°C for 20 min at 16,000 × g. The protein concentration in the supernatant was determined using Bio-Rad protein assay dye reagent concentrate according to the instructions of the manufacturer (Bio-Rad). Immunoprecipitation and subsequent activity assays for the determination of both total PDE activity and PDE4 inhibitor-insensitive PDE activity (in the presence of 10 μM rolipram) were conducted in accordance with a method described earlier (28). Alternatively, cells were resuspended in homogenization buffer (20 mM Tris-HCl (pH 7.6), 140 mM sodium chloride, 3.8 mM potassium chloride, 1 mM EGTA, 1 mM magnesium chloride, 0.5 mM 4-[2-aminoethyl]-benzenesulfonyl fluoride, 10 μM leupeptin, 2 mM benzamidine, 10 μM pepstatin A, 5 μM trypsin inhibitor, 1 mM 2-ME). After sonication, the lysate was centrifuged at 4°C for 20 min at 1000 × g. Total PDE and PDE4 inhibitor-insensitive PDE activity (in the presence of 1 μM piclamilast) were assayed as described previously (24). Briefly, 5–20 μg of protein were diluted in 20 mM Tris-HCl (pH 7.8) to a final volume of 150 μl, either including 0.1% DMSO or PDE4 inhibitor. After preincubation at 37°C for 5 min, assay mixture (final 20 mM Tris-HCl (pH 7.8), 5 mM magnesium chloride, 0.5 μM cAMP, 100 μM EGTA, and ∼30,000 cpm/assay [³H]cAMP) was added and the samples were incubated at a final volume of 200 μl at 37°C for 20 min. The reaction was stopped with 0.2 N HCl (4°C) on ice. Following incubation at 37°C for 15 min with 25 μg of 5′-nucleotidase (Crotalus atrox snake venom; Sigma-Aldrich), the assays were loaded on Poly-Prep chromatography columns (Bio-Rad) filled with 1 ml of QAE Sephadex A-25 (Sigma-Aldrich). [³H]Adenosine was eluted with 2 ml of 30 mM ammonium formiate (pH 6.0), mixed with 3 ml of scintillation fluid (Lumasafe Plus; Lumac), and measured for total radioactivity in a scintillation counter (Beckman Coulter). Results were corrected for blank values (measured in the presence of denatured protein) and the PDE4 activity was calculated as the difference of total PDE activity and PDE4 inhibitor-insensitive PDE activity. In all PDE assays, the amount of cAMP hydrolyzed was below 25% of the original substrate concentration.

Detection of PDE4 subtypes by immunoblotting

Cells were washed in PBS and resuspended in lysis buffer (150 mM sodium chloride, 5 mM EDTA, 50 mM Tris-HCl (pH 7.4), 0.1% sodium azide, 0.5% Triton X-100, 0.5 mM 4-[2-aminoethyl]-benzenesulfonyl fluoride, 10 μM leupeptin, 2 mM benzamidine, 10 μM pepstatin A, 5 μM trypsin inhibitor) at ∼1.5 × 10⁷ cells per 100 μl of lysis buffer. Protein determination, SDS-PAGE, and immunodetection of PDE4 subtypes were performed as detailed elsewhere (33). Modifications were the use of 50 μg of protein per lane, the transfer to nitrocellulose membranes, a semidry blotting technique (Bio-Rad), and the immunodetection with ECL advance Western blotting detection reagent (Amersham Biosciences). Detection of β-actin was used as loading control.

Functional analysis of CD4⁺ T cells

As parameter for T cell function after anti-CD3/CD28-induced stimulation, cytokine concentrations of IL-2, IFN-γ, and IL-5 in the culture medium were measured. Supernatants were removed after 24 or 72 h, pooled, and stored at −80°C until specific commercially available enzyme-linked immunosorbent assays (IL-2, IFN-γ, and IL-5 assay kits; Immunotech) were performed. Due to the high variability of cytokine levels from different blood donors and the comparable narrow linear range of the ELISA standard curve, appropriate dilution factors have been determined before the analysis. Dilutions were performed in the buffers equipped in the assay kits. All cytokines were determined from the pool fraction in duplicate in accordance with the manufacturer’s instructions (Immunotech).

The incorporation of [³H]thymidine was taken as a parameter to determine CD4⁺ T cell proliferation. After 48 h of stimulation, 10 μl of [³H]thymidine (0.2 μCi/well) were added to the cells in duplicates for a period of 18 h until cell transfer and radioactivity measurement, as detailed elsewhere (24).

Data analysis and software

Averages are presented as mean ± SD. For statistical analysis, ANOVA (ANOVA, GraphPad Prism 4.0; GraphPad Software) was used. Densitometric analysis of immunoblots was performed with the Advanced Image Data Analyzer (AIDA 3.22; Raytest). The primer and probe sets for quantitative PCR analysis were designed using Primer Express 2.0 (Applied Biosystems).

PDE4 subtypes have distinct mRNA expression patterns in CD4⁺ T cells and are differentially up-regulated upon anti-CD3/CD28 stimulation

For quantitative RT-PCR experiments, we designed primer/probe sets specific for each of the four PDE4 subtypes. To monitor the amount of template in different reactions, we used 18S rRNA as internal standard, which we had shown not to be differentially expressed upon T cell stimulation (data not shown). In validation experiments, we tested the primer/probe sets for the PDE4 subtypes in combination with the primer/probe set detecting 18S rRNA (duplex PCR). Because all PDE4 subtype primer/probe sets showed similar maximal efficiencies in standardized PCR (95 ± 3%, data not shown), we concluded that PDE4 expression patterns could also be directly compared. In unstimulated CD4⁺ cells, we detected mRNA expression of the PDE4 subtypes PDE4A, PDE4B, and PDE4D. Although the PDE4C primer/probe set is functional in control experiments, we never detected expression of PDE4C in CD4⁺ cells, regardless of whether the cells were stimulated or not. The highest mRNA expression level was found for PDE4D (∼12-fold higher than PDE4A) followed by PDE4B (∼8-fold higher than PDE4A) and PDE4A (set to 1 as reference) (Fig. 1⇓A). We then investigated whether the PDE4 subtypes are differentially expressed upon stimulation. Therefore, cells were stimulated with anti-CD3/CD28 for different periods of time. For each gene, the expression in unstimulated cells (time point 0 h) was used as reference point. Upon stimulation, PDE4A mRNA was steadily up-regulated and had a ∼15-fold higher expression after 120 h (Fig. 1⇓B). PDE4B mRNA was up-regulated ∼2-fold within the first 16 h, returned to basal level after 48 h, and further decreased at later times, indicating a transient up-regulation of this subtype (Fig. 1⇓C). PDE4D mRNA expression slightly decreased within the first 24 h after stimulation, but was then steadily up-regulated and reached a ∼4-fold higher expression after 120 h, indicating a long-term up-regulation of this subtype (Fig. 1⇓D).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

PDE4A, PDE4B, and PDE4D mRNA expression in untreated and anti-CD3/CD28 stimulated human primary CD4⁺ T lymphocytes. Cells were separated from whole blood as described in Materials and Methods and were subject for total RNA isolation and cDNA transcription. Quantitative PCR was performed using ribosomal 18S rRNA as internal standard (detected after 10–13 cycles) and specific primer/probe sets for PDE4A, PDE4B, PDE4C, and PDE4D. By comparing ΔC_t values, the expression level of individual PDE4 subtypes was reported relative to PDE4A, which was detected after 27–30 cycles (ΔC_t ∼17) (A). After stimulation with 3 μg anti-CD3/well and 3 μg anti-CD28/ml, time-dependent mRNA expression for PDE4A (B), PDE4B (C), and PDE4D (D) was measured relative to untreated conditions (t = 0 h). PDE4C transcripts were not detected after 40 cycles. Results are expressed as mean ± SD of four donors. Significance of differences is indicated: ∗, p < 0.05; ∗∗, p < 0.01; B–D, compared with untreated cells (t = 0 h).

To obtain an overall picture of the PDE4 subtype expression after T cell stimulation, we combined the different expression levels (Fig. 1⇑A) with the time courses of each individual subtype (Fig. 1⇑, B–D), using the expression level of PDE4A mRNA in unstimulated cells as reference for all other data points (Fig. 2⇓A). Due to the considerably high PDE4B mRNA levels in untreated CD4⁺ T cells, the ∼2-fold up-regulation after 16 h stimulation lead to the predominant expression of this subtype at 16 and 24 h. From 48 h onward, the induction of PDE4D and the decrease of PDE4B resulted in PDE4D being the predominant subtype. PDE4A mRNA displayed the lowest expression level up to 48 h after stimulation. Due to the steady up-regulation of this subtype and the down-regulation of PDE4B at later time points, the expression level of PDE4A exceeded that of PDE4B at 120 h but still did not reach the high expression level of PDE4D.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Time-dependent mRNA expression profiles of PDE4 subtypes in CD4⁺ T cells by different anti-CD3/CD28 stimulation. The quantitative mRNA expression levels of PDE4A, PDE4B, and PDE4D in untreated cells (shown in Fig. 1⇑A) were combined with the respective time courses (shown in Fig. 1⇑, B–D). All data were expressed relative to the PDE4A expression level in untreated cells (t = 0 h). Thereby, the complete quantitative mRNA expression pattern of the PDE4 subtypes was obtained for cells stimulated with 3 μg anti-CD3/well and 3 μg anti-CD28/ml (A). The same sets of experiments were performed with cells stimulated with 0.3 μg of anti-CD3/well and 0.3 μg of anti-CD28/ml (B), with 3 μg of anti-CD3/well alone (C), and with 0.3 μg anti-CD3/well alone (D). Results in A are expressed as mean ± SD of four donors. Results in B–D are the average of two donors.

The induction of PDE4 subtypes upon stimulation is independent of the anti-CD3-CD28 ratio

To examine whether the distinct regulation of individual PDE4 mRNA levels is dependent on the anti-CD3:anti-CD28 ratio and subsequently on the costimulatory anti-CD28 stimulus, we activated CD4⁺ T cells with 0.3 μg of anti-CD3/well in combination with 0.3 μg of anti-CD28/ml (Fig. 2⇑B), with 3 μg of anti-CD3/well alone (Fig. 2⇑C), or with 0.3 μg of anti-CD3/well alone (Fig. 2⇑D). All conditions resulted in similar time-dependent expression patterns as were found for the strong stimulation with 3 μg of anti-CD3/well plus 3 μg of anti-CD28/ml shown in Fig. 2⇑A, hereby confirming the long-term induction of PDE4A and PDE4D and the transient up-regulation of PDE4B. For all additional experiments related to PDE4 subtype expression, either on mRNA, enzyme activity, or protein level, we used this latter stimulation condition.

PDE4 subtypes are similarly regulated both in naive and memory CD4⁺ T cells

Because CD4⁺ T cells isolated from human whole blood still represent a rather heterogeneous population of cells, we investigated whether different subpopulations of CD4⁺ cells may regulate the PDE4 subtype expression specifically. Therefore, we studied the expression patterns separately in naive CD4⁺ T cells characterized by the surface marker CD45RA and in memory CD4⁺ T cells expressing the marker CD45RO (36). Due to the isolation strategy (see Material and Methods), we obtained CD45RA⁺CD4⁺ T cells (>95% pure) and CD45RO⁺CD4⁺ T cells (>98% pure), whereas cells coexpressing CD4, CD45RA, and CD45RO were sorted out. In unstimulated naive CD4⁺ T cells, we found slightly lower levels of PDE4A, PDE4B, and PDE4D transcripts compared with unstimulated memory CD4⁺ T cells (Fig. 3⇓, A–C). After anti-CD3/CD28 stimulation, memory CD4⁺ T cells showed a slightly faster up-regulation of PDE4A (Fig. 3⇓A) and PDE4D (Fig. 3⇓C), as well as a more pronounced transient up-regulation of PDE4B (Fig. 3⇓B) compared with naive CD4⁺ T cells. However, these differences in the regulation of individual PDE4 subtypes between naive and memory CD4⁺ T cells were not statistically significant. Because the expression profiles of PDE4 subtypes both in naive and memory CD4⁺ T cells were found to be very similar, all following experiments were performed with CD4⁺ cells that were not further separated.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

mRNA expression profiles of PDE4 subtypes in CD45RA⁺ (naive) and CD45RO⁺ (memory) human primary CD4⁺ T lymphocytes. CD4⁺ cells were further purified by negative selection using either anti-CD45RA or anti-CD45RO microbeads. Cells were stimulated with 3 μg of anti-CD3/well and 3 μg of anti-CD28/ml for different periods of time. Subsequent experiments were performed as outlined in the legend of Fig. 1⇑. Expression profiles of PDE4A (A), PDE4B (B), and PDE4D (C) from both T cell subpopulations are shown. Results are expressed relative to untreated conditions in naive T cells (t = 0 h), as mean ± SD of four donors. Differences between naive and memory cells were statistically not significant.

PDE4 subtype activity resembles mRNA expression pattern

Changes in mRNA levels do not necessarily result in changes of the respective protein levels. Therefore, we investigated whether the induction of PDE4 subtypes observed in quantitative PCR experiments translates in an increase in PDE4 enzyme activity. Upon stimulation, total cAMP-hydrolizing PDE activity (mean ± SD = 37 ± 16 pmol/min/mg in resting cells) was augmented in a time-dependent manner (Fig. 4⇓A), resulting in a significantly higher activity 72 and 120 h after stimulation. This increase was mainly due to an increase in PDE4 activity, whereas PDE4-independent enzyme activity initially decreased 24 h after stimulation and recovered to baseline levels during the time period studied.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

cAMP hydrolyzing phosphodiesterase activity in untreated and anti-CD3/CD28 stimulated human primary CD4⁺ T lymphocytes. Cells were stimulated with 3 μg of anti-CD3/well and 3 μg of anti-CD28/ml for different periods of time. A, Total cAMP-PDE hydrolyzing activity and PDE4 inhibitor-insensitive PDE activity (= PDE4 independent) in T cell lysates were measured as detailed in Materials and Methods. PDE4 activity was determined by subtracting the PDE4 independent activity from the total activity. Data are expressed as mean ± SD of three to five donors. ∗, p < 0.05; ∗∗∗, p < 0.001 (compared with untreated cells, t = 0 h). B, Cell lysates were immunoprecipitated with Abs recognizing PDE4A (AC55), PDE4B (K118), and PDE4D (m3S1) (32 ). The PDE activity recovered in the immunoprecipitation pellet was measured as detailed in Materials and Methods. Shown is one representative experiment of three. For the experiment, CD4⁺ cells of three donors were independently stimulated, but pooled for lysis and immunoprecipitation.

To ascertain which PDE4 subtypes contribute to this up-regulation, an immunoprecipitation strategy was applied with Abs specific for PDE4A, PDE4B, and PDE4D. After precipitation, the samples were assayed for PDE4 cAMP hydrolyzing activity. In unstimulated CD4⁺ T cells (time point 0 h in Fig. 4⇑B), the highest activity was measured in samples precipitated with the PDE4D-specific Ab. Whereas PDE4B activity was somewhat lower than PDE4D activity, PDE4A contributed least to the total PDE4 activity. However, after anti-CD3/CD28 stimulation, PDE4A activity was greatly induced starting at 72 h and reached a 4-fold higher activity at the end of the time period examined (Fig. 4⇑B). PDE4B activity was transiently up-regulated, peaking with a 3.4-fold higher activity at 24 h followed by a decrease. After 120 h, PDE4B activity was similar to the level observed in untreated cells. At this time point, PDE4A activity exceeded the PDE4B activity. Compared with untreated cells, PDE4D activity remained relatively constant during the first 48 h after stimulation, and increased at 72 and 120 h after stimulation totaling in a 1.5-fold up-regulation. Taken together, the pattern of PDE4 subtype-specific enzyme activities qualitatively closely resembled the mRNA expression profile.

The up-regulation of short PDE4 splice variants may account for the induction of PDE4 activity

To elucidate which splice variants are expressed in untreated CD4⁺ T cells and may account for the measured changes in mRNA levels and protein activities after stimulation, we used Abs recognizing the conserved C termini of the PDE4 subtypes to perform immunoblotting experiments (Fig. 5⇓, A and B). In unstimulated T cells, a polyclonal Ab directed against PDE4A detected immunoreactivity at ∼130 kDa and with a very low density below 100 kDa, which can be attributed to long (PDE4A4, PDE4A10, and/or PDE4A11) and short (PDE4A1) PDE4A forms, respectively. Contrary to our mRNA and enzyme activity measurements, the intensity of the immunodetected signals did not show a long-term increase after stimulation with anti-CD3/CD28. A polyclonal Ab directed against PDE4B gave rise to two predominant bands. One band migrated slightly above 100 kDa and might be attributed to PDE4B long forms (PDE4B1 and/or PDE4B3). The high immunoreactivity in unstimulated cells decreased after stimulation during the examined time period (0–120 h). The other band, migrating at ∼75 kDa (might be attributed to PDE4B2) was also present in unstimulated cells, but was strongly induced, peaking 24–48 h after stimulation. At later time points, a reduction of this band was observed. Considering the transient induction of the subtype PDE4B measured by PCR and activity assays, the immunodetected down-regulation of long PDE4B splice variants is compensated by a pronounced up-regulation of the short splice variant and may result in a net up-regulation between 24 and 48 h after stimulation. A polyclonal Ab directed against PDE4D detected several bands of different molecular weights. One set of bands was in the range of ∼60–75 kDa and likely corresponds to the short forms PDE4D6 (∼59 kDa), PDE4D2 (∼68 kDa), and PDE4D1 (∼72 kDa). These short forms were up-regulated in a time-dependent manner, with the middle band, assumingly PDE4D2, showing the most pronounced up-regulation. The other set of bands was above ∼85 kDa and may be the long splice variants of PDE4D (PDE4D3, PDE4D4, PDE4D5, PDE4D7, PDE4D8, PDE4D9). These bands showed low immunoreactivities that were not, or only slightly, changed after stimulation. The unchanged immunodetection of β-actin as loading control confirmed that the observed regulation of PDE4 subtypes was not due to different amounts of protein loaded onto the gel. For quantification, densitometric analysis of PDE4 long and short forms and of β-actin was performed and expressed relative to untreated cells (Fig. 5⇓B). As indicated above, both PDE4B and PDE4D short forms are up-regulated, whereas PDE4B and PDE4D long forms are down-regulated or unaffected, respectively.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Detection and quantification of PDE4 splice variants in untreated and anti-CD3/CD28-stimulated human primary CD4⁺ T lymphocytes by immunoblotting and densitometric analysis. Cells were stimulated with 3 μg of anti-CD3/well and 3 μg of anti-CD28/ml for different periods of time. Cells were lysed and 50 μg of protein of the 1000 × g supernatant were separated by SDS-PAGE. A, Immunoblots detecting PDE4A, PDE4B, and PDE4D splice variants. Immunodetection was performed using rabbit polyclonal Abs recognizing PDE4A, PDE4B, or PDE4D (33 ). β-Actin was used as loading control. Shown is one representative of two experiments with cells from three donors that have been independently stimulated, but pooled for lysis and immunoblotting. B, Densitometric analysis of PDE4 long and short forms and β-actin, each expressed relative to untreated conditions. PDE4A short forms were not analyzed quantitatively due to the very faint immunodetected signal. Shown are mean values of two independent experiments, each with cells from three donors that have been independently stimulated, but pooled for lysis and immunoblotting.

Validation of siRNA-mediated PDE4 subtype mRNA and activity knockdown in CD4⁺ T cells

In the following set of experiments, we addressed the question whether the specific expression patterns of PDE4 subtypes correlates with a respective relevance for different T cell functions. Because PDE4 inhibitors so far available are rather nonselective for PDE4 subtypes, we used a siRNA-based approach to selectively knockdown the different PDE4 subtypes. Although we used a comparatively high stimulation condition (3 μg of anti-CD3/well and 3 μg of anti-CD28/ml) for all PDE4 expression and activity studies to detect maximal effects in the regulation of PDE4 subtypes (Figs. 1–5⇑⇑⇑⇑⇑), we used a weaker stimulation condition (0.3 μg of anti-CD3/well and 0.3 μg of anti-CD28/ml) for all functional studies and therefore also for all siRNA validation experiments. This weaker stimulation was chosen because we had observed that full stimulation decreases the efficacy of PDE4 inhibitors (data not shown) and might thus also decrease functional effects of PDE4 subtype-specific siRNAs. To transfect human primary CD4⁺ T cells with siRNAs, we used the Amaxa NF technique (34). In a series of extensive control experiments, we tested this siRNA-based strategy for unspecific, cytotoxic effects. The nucleofection procedure alone (without siRNA) produced a considerable loss of cells (37 ± 10%), but had only a minor effect (5 ± 3%) on cell mortality of recovered cells, as measured by 7-AAD staining 24 and 48 h after nucleofection (Table I⇓). The release of cytokines was also affected by the nucleofection procedure, compared with untreated cells (45 ± 7%). In contrast, the expression of surface markers, such as CD4 and CD25 measured by flow cytometry, as well as cell proliferation was not changed by the nucleofection procedure. Although the nucleofection procedure alone had substantial effects on some of the measured parameters, the application of siRNA had no additional effects on any of the parameters investigated (see last column in Table I⇓).

The transfection efficiency was tested with FITC-labeled nontargeting siRNA and resulted in ∼94% FITC-positive cells (Fig. 6⇓A). From sets of siRNAs specific for each PDE4 subtype, we selected those individual siRNAs that were most effective in down-regulating the respective mRNA (data not shown). Different siRNA concentrations were tested to determine the optimal concentration needed for mRNA knockdown (Fig. 6⇓B). In unstimulated CD4⁺ cells, individual siRNAs directed against PDE4A, PDE4B, and PDE4D concentration dependently induced specific mRNA knockdowns of each respective PDE4 subtype 24 h after transfection compared with nucleofection controls (NF control, cells transfected without siRNA) (Fig. 6⇓B). Maximal mRNA knockdown was achieved with 1.5 μM siRNA. Notably, the knockdown efficacy for every individual siRNA was within the same range of ∼59–65%. The siRNAs target sequences that are highly conserved between the different PDE4 subtypes. Although each siRNA is capable to efficiently down-regulate the corresponding mRNA, none of the used siRNAs did affect the highly homologous mRNAs of the other PDE4 subtypes. Additionally, nontargeting siRNA used as a negative control had no effect on the expression of PDE4A, PDE4B, and PDE4D mRNA (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Validation of siRNA-mediated knockdown of PDE4 subtypes in human primary CD4⁺ T lymphocytes. A, Transfection efficiency, tested by using 2 μM FITC-labeled siRNA. Uptake of siRNA was measured by flow cytometry. B, Concentration-dependent siRNA-mediated mRNA knockdown in unstimulated CD4⁺ T cells 24 h after nucleofection. Data are shown relative to each nucleofection (NF) control, as one representative of two experiments. C, Time-dependent siRNA-mediated mRNA knockdown after stimulation with 0.3 μg of anti-CD3/well and 0.3 μg of anti-CD28/ml. Results are expressed relative to the respective NF control as the mean ± SD of three to six donors. Significant differences between NF control and siRNA-transfected cells are indicated (∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; compared with NF control of each time point). D, PDE activity knockdown, induced by PDE4 subtype-specific siRNAs, in CD4⁺ T cells 24 and 72 h after anti-CD3/CD28 stimulation. Results are expressed as cAMP-PDE activity in percent of the NF control, as mean ± SD of three donors. Significant differences between NF control and siRNA-transfected cells are indicated (∗∗, p < 0.01; ∗∗∗, p < 0.001; compared with NF control).

To test whether the PDE4 subtype-specific siRNAs were also effective after stimulation, transfected cells were left unstimulated for 24 h to allow siRNA-mediated mRNA depletion before the addition of the anti-CD3/CD28 stimulus. Subsequently, the mRNA knockdown effects were determined at 24, 48, and 72 h after stimulation and were compared with the NF controls (Fig. 6⇑C) taken at the respective time point. Significant mRNA knockdowns were achieved up to 72 h after stimulation with all applied siRNAs, although the siRNA-mediated knockdown effects started to diminish. To ascertain that the mRNA knockdown translates into a decrease in protein level, we performed cAMP-PDE activity assays with lysates from siRNA-transfected CD4⁺ T cells (Fig. 6⇑D). Lysates from T cells treated only with the nucleofection procedure (NF control) were used as reference. Twenty-four hours after stimulation, a marginal reduction of total cAMP-PDE activity and PDE4 activity (mean ± SD = 10 ± 13%, NS) was achieved with siRNA directed against PDE4A. siRNAs directed against PDE4B and PDE4D significantly reduced total cAMP-PDE activity and PDE4 activity (PDE4B-siRNA: 28 ± 5%, PDE4D-siRNA: 51 ± 11%). To evaluate whether the knockdown effects on PDE4 activity were maintained for a longer time period, we repeated the PDE activity measurements at 72 h. At this latter time point, the reduction of PDE4 enzyme activity by the different siRNAs were similar to the 24-h time point (PDE4A-siRNA: 18 ± 9%, PDE4B-siRNA: 27 ± 14%, PDE4D-siRNA: 46 ± 7%). Considering that the individual PDE4 subtypes contribute quite differently to the overall PDE4 activity, it is expected that PDE4A siRNA will have the smallest effect on PDE4 activity whereas PDE4B and PDE4D siRNAs will have larger effects even when the individual subtypes are diminished to the same extend. Indeed, the collective effects of all individual PDE4 siRNAs sum up to ∼90% both after 24 and 72 h, indicating a pronounced efficacy of the siRNAs to down-regulate their respective subtype. Taken together, the strategy to selectively knockdown individual PDE4 subtypes in primary CD4⁺ T cells was well tolerated, specific, and efficient. Because enzymatic activity was effectively reduced 72 h after stimulation but mRNA levels started to recover at the same time, we considered the siRNA approach to be suitable for the application on functional T cell assays in a time period of up to 72 h.

Anti-CD3 and anti-CD28 stimulation induces the release of cytokines, which can be suppressed by PDE4 inhibition

To determine how human primary CD4⁺ T cells functionally respond to anti-CD3/CD28 stimulation, we measured IL-2, IFN-γ, and IL-5 secretion at different times in supernatants of stimulated cells (Fig. 7⇓A). In untreated cells (time point 0 h), IL-2, IFN-γ, as well as IL-5 levels were below the detection limit. After stimulation, IL-2 was strongly induced with maximal levels 24 h after stimulation, but returned to basal levels within 5 days. Compared with IL-2 secretion, IFN-γ and IL-5 levels were substantially induced at later time points, but were from 24 h onward steadily up-regulated and were maximal at the end of the examined time period (134 h).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Time course of functional T cell parameters after anti-CD3/CD28 stimulation and suppression of T cell functions by complete PDE4 inhibition. A, Time-dependent cytokine secretion of CD4⁺ T cells (IL-2, IFN- γ, IL-5) after stimulation with 0.3 μg of anti-CD3/well and 0.3 μg of anti-CD28/ml. Supernatants were collected and cytokine concentrations were determined by ELISA. Results are expressed as mean ± SD of four donors. B, Suppression of CD4⁺ T cell functions by the panPDE4 inhibitor RP73401 (piclamilast). Cells treated with the nucleofection procedure alone (NF control) were incubated for the indicated period of time after anti-CD3/CD28 stimulation either in the presence of 0.1% DMSO or additionally in the presence of 1 μM RP73401. Data are shown as mean ± SD (%) of four to six donors. Significance of differences is indicated: ∗∗, p < 0.01; ∗∗∗, p < 0.001; compared with NF control (+0.1% DMSO).

To investigate how T cell functions are affected by PDE4 inhibition, we treated CD4⁺ cells with the very potent panPDE4 inhibitor RP73401 (piclamilast), which does not discriminate between distinct PDE4 subtypes (24), and stimulated the cells for the indicated time periods (Fig. 7⇑B). To compare the effects of RP73401 with the effects of PDE4 subtype-specific siRNAs, we used cells treated with the nucleofection procedure for the RP73401 experiments, but did not apply siRNA. To adjust the control conditions for DMSO concentrations, we treated transfected cells with 0.1% DMSO and set the measured cytokine level and proliferation rate of these cells to 100%. A total of 1 μM RP73401 (a maximum concentration that guarantees PDE4 selectivity, data not shown) significantly inhibited anti-CD3/CD28 induced IL-2 release (∼39 ± 12% inhibition, 24 h after stimulation), IFN-γ release (∼52 ± 11% inhibition, 72 h after stimulation), and IL-5 release (∼43 ± 13% inhibition, 72 h after stimulation) (Fig. 7⇑B). Under these conditions, proliferation was also significantly inhibited, but to a lesser extent (∼13 ± 4% inhibition, 66 h after stimulation).

Individual PDE4 subtypes have different functional significance in human primary CD4⁺ T cells

To ascertain which individual PDE4 subtype(s) mediate the inhibitory effect of a panPDE4 inhibitor, PDE4 subtype-specific siRNAs were nucleofected into CD4⁺ T cells. After a 24-h resting period, transfected cells were stimulated with anti-CD3/CD28. A total of 1.5 μM nontargeting siRNA was used as control for statistical analysis (NEG_si in Fig. 8⇓). The functional impact of individual siRNAs on cytokine release and on cell proliferation is reported as a percentage of the RP73401 effect, which can be assumed to represent the maximal effect and was set therefore to 100%. Comparable to Fig. 7⇑B, cells treated with the nucleofection procedure alone were used to determine these maximal effects. Twenty-four hours after stimulation, siRNA targeting PDE4A showed no significant inhibitory effect on IL-2 synthesis, however, knockdown of either PDE4B or PDE4D significantly suppressed anti-CD3/CD28 induced IL-2 release to an extent close to that observed with RP73401 (Fig. 8⇓A). Because substantial amounts of IFN-γ and IL-5 were only measured at later time points, the functional impact of PDE4 subtype-specific siRNAs on these latter cytokines was measured 72 h after stimulation. siRNA targeting PDE4A had no significant inhibitory effect on IFN-γ release. siRNA targeting PDE4B had a slightly higher inhibitory effect, however, siRNA targeting PDE4D showed the most pronounced inhibitory effect on IFN-γ synthesis similar to RP73401 (Fig. 8⇓B). Likewise, IL-5 synthesis was not significantly affected by the application of siRNAs targeting PDE4A and PDE4B, but was significantly inhibited by knockdown of PDE4D, to an extent close to that obtained with RP73401 (Fig. 8⇓C). Remarkably, the combined application of siRNAs targeting PDE4A, PDE4B, and PDE4D (PDE4_all_si in Fig. 8⇓), produced the most pronounced inhibitory effect on cytokine release. The relatively weak inhibitory effect of RP73401 on T cell proliferation (compare Fig. 7⇑B) was mimicked by PDE4D knockdown, whereas siRNA targeting PDE4A and PDE4B were hardly effective (Fig. 8⇓D). Furthermore, the addition of the latter two siRNAs to PDE4D siRNA did not further increase the inhibitory effect on T cell proliferation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

Suppression of CD4⁺ T cell functions by PDE4 subtype-specific siRNAs. A total of 1.5 μM nontargeting siRNA (NEG_si) or PDE4 subtype-specific siRNAs were nucleofected into human primary CD4⁺ T cells. Cells were left in culture for 24 h and were then stimulated with 0.3 μg of anti-CD3/well and 0.3 μg of anti-CD28/ml. The siRNA-mediated suppression of IL-2 secretion after 24 h (A), IFN-γ secretion after 72 h (B), IL-5 secretion after 72 h (C), and proliferation ([³H]thymidine incorporation) after 66 h (D) is depicted in percent of the RP73401 effect. This maximally expected effect was determined using cells treated with the nucleofection procedure alone. NEG_si, nontargeting control siRNA; PDE4_all_si, combined application of PDE4A_si, PDE4B_si, and PDE4D_si. Results are expressed as mean ± SD of four to six donors. Significance of differences is indicated: ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; compared with 1.5 μM NEG_si.