Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The expression and secretion of cytokines by activated T lymphocytes is a strictly regulated process, involving interactions between hematopoetic cells and nonhematopoetic cells (1). T lymphocytes stimulated with mitogen or a combination of antibodies against CD2, CD3, and CD28 synthesize and secrete a variety of cytokines, including interleukin (IL)-3, IL-4, granulocyte-macrophage colony-stimulating factor (GM-CSF), and interferon- (IFN-) (2). Obviously, a balanced cytokine expression requires both stimulatory and inhibitory signals. An increase in intracellular cyclic adenosine monophosphate (cAMP) levels might be stimulatory or inhibitory in the regulation of cytokine expression and is dependent on the cell type studied. In general, cAMP initiates an inhibitory signal and diminishes the responsiveness to exogenous activators (7).

Under physiological conditions, increase of intracellular cAMP can be achieved by binding of -adrenoceptor (AR) agonists to their specific receptors. This results in activation of membrane-bound adenylyl cyclase and subsequent increase of intracellular cAMP (7, 10, 11). -Adrenoceptors are present in several cell types, including T lymphocytes (12, 13). It has been reported that AR agonists have an inhibitory effect on T-lymphocyte activation (14). Furthermore, it has been suggested that the outcome of Th1 or Th2 cytokine profiles are differentially regulated by the adenylyl cyclase system, the Th2-derived cytokines (IL-4, IL-5) being less inhibitable by cAMP elevations (15). In this respect, it is interesting that both AR number and adenylyl cyclase responses are reduced after an allergen challenge-induced asthmatic attack (18). Thus, allergen exposure may contribute to the Th2 type immunoresponse as observed in allergic asthma (19). In addition, concanavalin A (Con A)-induced IL-3, GM-CSF, and IL-4 mRNA accumulation and protein secretion are diminished by simultaneous activation of adenylyl cyclase (5, 6).

It has been demonstrated that the AR family consists of three subtypes designated ₁, ₂, and ₃, which share about 50% sequence homology (10, 20). In rat and human adipocytes it has been shown that they are distinct in their ability to generate cAMP, the ₃AR being less potent to generate cAMP (21). -Adrenoceptors in T lymphocytes have been classified as ₁ and ₂ using radioligand displacement experiments (13). At present, it is unknown whether T lymphocytes have ₃-adrenoceptors as well.

In this study we investigated the involvement of these AR subtypes in the regulation of four T lymphocyte- derived cytokines that are important in the regulation of the asthmatic inflammatory response (IL-3, IL-4, IFN-, and GM-CSF) by using various AR agonists and antagonists. The data showed that downregulation of IL-3, IFN-, and GM-CSF mRNA accumulation and protein secretion of GM-CSF are selectively mediated by the ₂AR and are related to the intracellular accumulation of cAMP induced by the agonist, whereas IL-4 mRNA accumulation was not sensitive to ₂AR stimulation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Preparation of Cells

Peripheral blood cells were obtained from healthy volunteer platelet donors, and mononuclear cell suspensions were prepared by Ficoll-Hypaque (Lymphoprep; Nycomed, Oslo, Norway) density-gradient centrifugation. T lymphocytes were isolated by 2-aminoethylisothiouronium bromide-treated sheep red blood cell (SRBC) rosetting. The SRBC were lysed with 155 mmol/liter NH₄Cl, 10 mmol/liter KHCO₃, 0.1 mmol/liter ethylenediamine tetraacetic acid (EDTA) and the remaining cell preparations contained more than 98% T lymphocytes as assessed by flow cytometric analysis after staining with an anti-CD2 monoclonal antibody (Becton Dickinson, Sunnyville, CA) and less than 1% CD14-positive cells (Becton Dickinson). After isolation, T lymphocytes were cultured overnight at 37°C in RPMI 1640 medium (Flow, Rockville, MD) containing 10% fetal bovine serum (FBS; Hyclone, Logan, UT) supplemented with 100 U/ml penicillin, 100 µg/ml streptomycin, 2 mmol/liter L-glutamine and 6 ng/ml colistine.

Stimulation

Cells (5 × 10⁶/ml) were incubated in RPMI 1640/FBS medium for 6 h with 25 µg/ml Con A (Calbiochem, La Jolla, CA) as described previously (5, 6), in the absence or presence of several AR (ant)agonists. Sodium metabisulfite (10⁵ M) was added to the medium as an antioxidant. ()Isoproterenol hydrochloride (Sigma, St. Louis, MO) was used as a nonselective AR agonist at concentrations of 10⁴ to 10⁸ mol/liter. Fenoterol hydrobromide (Sigma) was used as a ₂AR selective agonist at concentrations of 10⁵ and 10⁶ mol/liter. Timolol maleate (gift from Merck, Sharp and Dohme, Haarlem, The Netherlands) was used as a nonselective AR antagonist at a concentration of 10⁶ mol/liter. Atenolol hydrochloride (Sigma) was used as a ₁AR selective antagonist at a concentration of 3 × 10⁷ mol/liter. ICI 118,551 hydrochloride (gift from Zeneca, Macclesfield, UK) was used as a ₂AR antagonist at a concentration of 10⁶ mol/liter. BRL 37344 (gift from Smith Kline Beecham, Epsom, UK) was used as a selective ₃AR agonist at variable concentrations.

Extraction of mRNA and Analysis

Total cellular RNA was isolated by the guanidinium isthiocyanate/cesium chloride method (22). Twelve micrograms of total RNA was electrophoresed in 2.2 mol/liter formaldehyde, 1.1% agarose gels, and blotted onto nylon membranes (Hybond N⁺; Amersham, Buckinghamshire, UK) (23). cDNA probes were labeled with [-³²P]dCTP (3,000 Ci/mol; Amersham) by the random hexamer priming method (24). The following cDNA probes were used: (1) the 0.3-kb EcoRI/HindIII insert of human IL-3 cDNA purified from the phM1 plasmid (gift from Dr. H. Burger, TNO, Rijswijk, The Netherlands); (2) the 0.3-kb EcoRI/ NheI insert of human IL-4 cDNA (gift from Dr. S. Narula, Schering Plough, Bloomfield, NJ); (3) the 0.6-kb EcoRI/ HindIII insert of human IFN- cDNA purified from the pSP65 plasmid (gift from Dr. C. B. Wilson, University of Washington, Seattle, WA); (4) the 0.45-kb, EcoRI/AvaII insert of human GM-CSF cDNA (gift from Dr. S. Gillis, Immunex, Seattle, WA); and (5) the EcoRI linearized pBR322 plasmid containing a 7.8-kb human 28S cDNA insert.

Hybridization was performed at 65°C for 18 h in 0.5 mol/liter Na₂HPO₄, pH 7.2, 1 mmol/liter EDTA, and 7% sodium dodecyl sulfate (SDS). Membranes were washed once in 2× saline sodium citrate (SSC), 0.1% SDS; once in 1× SSC, 0.1% SDS; and finally in 0.3× SSC, 0.1% SDS for 20 min at 65°C. The membranes were exposed to Kodak X-omat XAR films (Eastman Kodak, Rochester, NY) at 80°C using an intensifying screen. Quantification of mRNA levels was performed by densitometry using a Gel Scan laser densitometer (Pharmacia LKB, Uppsala, Sweden).

Reverse Transcription (RT) and Polymerase Chain Reaction (PCR)

Reverse transcription. Total cellular RNA, 1 µg, was resuspended in 15 µl diethylpyrocarbonate (DEPC; Sigma)- treated H₂O and incubated at 68°C for 10 min. After they were cooled on ice the samples were spun briefly (20, 2,000 × g) in a microcentrifuge. Next, 10 µl of 10 × RT mix (135 mM Tris-HCl, pH 8.3, 204 mM KCl, 27 mM MgCl₂, 0.24 mg/ml bovine serum albumin (BSA), 5.4 mM of each dNTP, 14.3% glycerol) and 3 U of reverse transcriptase (Promega Corporation, Madison, WI) were added to the samples. The samples were spun briefly (20, 2,000 × g) and incubated at 37°C for 60 min.

Polymerase chain reaction. A specific primer pair for the ₃-adrenergic receptor was synthesized on a Gene Assembler Plus DNA synthesizer (Pharmacia) and purified using NAP 10 columns (Pharmacia). The primers shared no sequence homology with the ₁ or ₂ adrenergic receptor. The sense primer was 5' CCTGTGCACCTTGGGTCTCA 3' (position 907-927), the antisense primer was 5' TCGAGCCGTTGGCAAAGCCT 3' (position 1239-1219).

Measurement of cAMP Production

T lymphocytes were washed twice in Tris buffer containing 120 mmol/liter NaCl, 1 mmol/liter MgCl₂, 5 mmol/liter KCl, 0.6 mmol/liter CaCl₂, 25 mmol/liter Tris (hydroxymethyl-amino-ethane), 5 mmol/liter glucose, and 0.1 mmol/liter human albumin, adjusted to pH 7.4 with HCl. Cells were suspended in RPMI 1640/FBS medium to a final density of 2.0 × 10⁶ cells/ml. Stimulation of cAMP production was performed as described before (25). In short, the samples were preincubated with 1-methyl-3-isobutylxantine (0.5 mmol/ liter) for 10 min. After preincubation, the samples were stimulated for 10 min with Con A (25 µg/ml) in the absence or presence of the AR agonist, without or with an antagonist. Reactions were terminated by adding 2 N HCl-0.1 mol/liter EDTA followed by incubating the samples at 80°C for 10 min. After centrifugation of precipitated protein, the samples were neutralized by CaCO₃ and cAMP was measured using an enzyme immunoassay (Biotrak, Amersham, UK) as specified by the manufacturers. cAMP concentrations are expressed as pmol cAMP/10⁶ T lymphocytes.

Measurement of GM-CSF Protein

Human T lymphocytes (10⁶/ml) were stimulated with Con A (25 µg/ml) plus isoproterenol (10⁴ to 10⁸ mol/liter) in the absence and presence of timolol (10⁶ M) for 24 h. Vitality of cells was controlled by the trypan blue exclusion method. Secreted protein was quantified in cell-free supernatants using a human GM-CSF enzyme-linked immunosorbent assay (ELISA) kit (Genzyme Corp., Cambridge, MA) as recommended by the manufacturers.

Data Analysis

pEC₅₀ (log EC₅₀) values were derived from the individual agonist concentration response curves. pK_B values of the antagonists were derived from the EC₅₀ values obtained in the absence and presence of the antagonist according to pK_B = log{[antagonist]/(DR1)} (26). Statistical analysis was performed using Student's t test for paired observations. Statistically significant changes are reported when P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

To study the involvement of AR subtypes in the regulation of IL-3, IL-4, GM-CSF, and IFN- mRNA expression, human T lymphocytes were incubated for 6 h with Con A, in the absence and presence of isoproterenol or fenoterol with or without selective or nonselective AR antagonists. As shown in Figure 1, Con A-induced accumulation of IL-3, IFN-, and GM-CSF mRNA was found to be dose-dependently inhibited by isoproterenol. IFN- mRNA accumulation was most sensitive to the AR agonist, followed by GM-CSF and IL-3. Log EC₅₀ values for inhibition by isoproterenol of Con A-induced IFN-, GM-CSF, and IL-3 mRNA accumulation as calculated from the individual plots were found to be 6.32 ± 0.03, 6.13 ± 0.19, and 5.96 ± 0.28 (mean ± SEM; n = 3). In contrast, IL-4 mRNA accumulation was not significantly inhibited by isoproterenol. The dose-dependent inhibition of IFN-, GM-CSF, and IL-3 mRNA accumulation induced by isoproterenol was strongly antagonized by simultaneous addition of the selective ₂AR antagonist ICI 118,551 (10⁶ M) (Figure 1); from the parallel shifts to the right, pK_B values of 7.82 ± 0.31, 7.71 ± 0.09, and 7.37 ± 0.16, respectively, were derived. As depicted in Figure 2, simultaneous addition of atenolol (0.3 × 10⁶ M) to Con A-activated T lymphocytes (i.e., under selective blockade of ₁AR) did not antagonize at all the isoproterenol-induced inhibition of mRNA accumulation for IFN-, GM-CSF, and IL-3, indicating that ₁AR was not involved.

View larger version (62K): [in this window] [in a new window]   View larger version (26K): [in this window] [in a new window]   Figure 1.   Dose-response curves (A) of the effect of isoproterenol (ISO) on the accumulation of Con A-induced IFN-, GM-CSF, IL-3, and IL-4 mRNA. Northern blot analysis of total cellular RNA extracted from T lymphocytes after 6 h of stimulation in the absence (solid squares) and presence (solid triangles) of 106 M ICI 118,551 (ICI). mRNA levels of three independently performed experiments were quantified by densitometric scanning and normalized to the respective 28S signals. Representative mRNA data are shown in B.

View larger version (41K): [in this window] [in a new window]   Figure 2.   The effect of two concentrations of isoproterenol (ISO) and fenoterol on the accumulation of Con A-induced IFN-, GM-CSF, and IL-3 mRNA. Northern blot analysis of total cellular RNA extracted from T lymphocytes after 6 h of stimulation in the absence and presence of 0.3 × 106 M atenolol. The corresponding bar diagrams represent relative mRNA values of a representative experiment after normalization to the respective 28S signals.

Next, fenoterol was added to the T-lymphocyte cultures as a selective ₂AR agonist at concentrations of 10⁵ and 10⁶ M. As shown in Figures 2 and 3, these concentrations of fenoterol strongly inhibited the Con A-induced IFN-, GM-CSF, and IL-3 mRNA accumulation. Simultaneous addition of atenolol (0.3 × 10⁶ M) had no effect (Figure 2), whereas timolol (10⁶ M)able to antagonize strongly both ₁AR and ₂ARmarkedly decreased the fenoterol-induced inhibition of IFN-, GM-CSF, and IL-3 mRNA accumulation. Fenoterol did not significantly affect IL-4 mRNA accumulation at these concentrations (data not shown). The foregoing data indicate that isoproterenol and fenoterol exert their inhibitory effects on IFN-, GM-CSF, and IL-3 mRNA accumulation via the ₂AR without involvement of ₁AR.

View larger version (30K): [in this window] [in a new window]   Figure 3.   The effect of two concentrations of fenoterol on the accumulation of Con A-induced IFN- (filled bars), GM-CSF (hatched bars), and IL-3 (open bars) mRNA. Northern blot analysis of total cellular RNA extracted from T lymphocytes after 6 h of stimulation in the absence and presence of 106 M timolol. The corresponding bar diagrams represent relative mRNA values of a representative experiment after normalization to the 28S signals.

Next, we questioned whether or not the expression of these cytokines could be modulated by the selective ₃AR agonist BRL 37344, because we were able to detect ₃AR mRNA in T lymphocytes by RT-PCR (Figure 4A). As shown in Figure 4B, the accumulation of Con A-induced IL-4, IFN-, and GM-CSF mRNAs was not affected by BRL 37344, used at concentrations up to 10⁴ M. Similar results were observed for IL-3 mRNA accumulation (data not shown).

View larger version (29K): [in this window] [in a new window]   Figure 4.   Ethidium bromide-stained agarose gel (A) demonstrating high levels of RT-PCR amplified product with the predicted 332 bp size. Total RNA obtained from four different donors was isolated from Con A-stimulated T lymphocytes, reverse transcribed, and amplified in the presence of 3AR specific primers (lanes 1-4). Lane 5, 100-bp marker. Effect of BRL 37344 (B) on the accumulation of IFN-, GM-CSF, and IL-4 mRNA. Northern blot analysis of total cellular RNA extracted from T lymphocytes activated with Con A in the absence and presence of different concentrations (as indicated) of BRL 37344. The 28S signal shows comparable levels of RNA in each lane.

To study whether the observed mRNA data have biological significance, GM-CSF protein secretion was measured in supernatants of T lymphocytes (10⁶ cells/ml) obtained from four individual donors. As depicted in Figure 5A, the data for GM-CSF mRNA accumulation were reflected in the protein secretion data. Isoproterenol showed a dose-dependent inhibitory effect on the secretion of Con A-induced GM-CSF protein, which was strongly antagonized by timolol (10⁶ M). Unstimulated T lymphocytes did not secrete detectable levels of GM-CSF protein (< 0.1 ng/ml). Con A-stimulated T lymphocytes secreted 6.9 ± 0.4 ng/ml GM-CSF protein. In the presence of Con A plus isoproterenol (10⁶ M, 10⁵ M, and 10⁴ M), a significant reduction in GM-CSF protein was observed: 3.8 ± 0.9 ng/ ml (P < 0.02), 2.6 ± 0.8 ng/ml (P < 0.02) 2.7 ± 0.4 ng/ml (P < 0.01), respectively, whereas lower concentrations of isoproterenol (10⁷ and 10⁸ M) did not have significant effects. The pEC₅₀ value of isoproterenol turned out to be 6.38 ± 0.17 (n = 4). In the presence of timolol (10⁶ M) a marked shift to the right was observed, yielding a pK_B value of 8.23 ± 0.37 (n = 4). Atenolol (0.3 × 10⁶ M) did not antagonize the inhibitory effect of isoproterenol on the Con A-induced production of GM-CSF protein (data not shown).

View larger version (13K): [in this window] [in a new window]   Figure 5.   Dose-response curves (A) for isoproterenol on Con A-induced GM-CSF protein secretion in the absence (solid squares) and presence (solid triangles) of 106 M timolol. GM-CSF was measured in supernatants of T lymphocytes (106/ml) harvested after 24 h of stimulation. Values are the means ± SEM of four independent experiments. Dose-response curves (B) for isoproterenol-induced cAMP production (expressed as pmol/106 T cells) in Con A-activated T lymphocytes in the absence (solid squares) and presence (solid triangles) of 106 M timolol. Values are presented as the means ± SEM of five independent experiments.

In addition, we determined the accumulation of intracellular cAMP in T lymphocytes obtained from five individual donors in relation to the inhibitory effect of isoproterenol on Con A-induced GM-CSF secretion. T lymphocytes stimulated with Con A alone demonstrated a basal level of cAMP (0.41 ± 0.10 pmol). Concentrations of 10⁶ M, 10⁵ M, and 10⁴ M isoproterenol significantly enhanced accumulation of intracellular cAMP levels (1.51 ± 0.26 pmol, P < 0.05; 2.17 ± 0.24 pmol, P < 0.02; 1.76 ± 0.16 pmol, P < 0.02), whereas lower concentrations did not. Again, timolol (10⁶ M) strongly antagonized the isoproterenol-induced cAMP accumulation, which resulted in a parallel shift to the right of 2.6 log units (Figure 5B); the pK_B value of timolol amounted to 8.54 ± 0.38 (n = 5). The concentration-dependent accumulation of cAMP by isoproterenol and the antagonism by timolol were inversely related to the effect of isoproterenol on GM-CSF protein secretion (compare Figure 5B with 5A). In contrast, the ₃AR agonist BRL 37344 did not significantly affect cAMP accumulation: even in the presence of 10⁴ M BRL 37344 the level of generated cAMP did not significantly differ from basal levels (Figure 6).

View larger version (16K): [in this window] [in a new window]   Figure 6.   Effect of increasing concentrations of isoproterenol and BRL 37344 on the production of cAMP (expressed as pmol/ 106 T cells) in Con A-activated T lymphocytes. The data are presented as means ± SEM of five independent experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It is now known that T lymphocytes express both ₁ and ₂ adrenoceptors on their membranes (13). The effects of several cytokines on the expression of these receptors have been studied extensively (27). On the other hand, although we have previously shown that IL-3 and GM-CSF mRNA accumulation and protein secretion can be inhibited by isoproterenol (6), there is lack of information regarding the identity of AR subtype(s) involved. In this study we have characterized the AR subtype(s) involved in the regulation of four T lymphocyte-derived cytokines: IL-3, IL-4, IFN-, and GM-CSF. We observed that the nonselective AR agonist isoproterenol dose dependently diminished the Con A-induced accumulation of IL-3, GM-CSF, and IFN- mRNA. IL-3, GM-CSF, and IFN- mRNA levels were also inhibited by the selective ₂AR agonist fenoterol, whereas the highest concentration of isoproterenol (10⁴ M) had no significant effect on IL-4 mRNA accumulation. These observations correspond with similar observations showing that the Th2-derived cytokine IL-4 is less sensitive to cAMP elevations compared with IFN-, which is a Th1-derived cytokine (15). However, it recently was demonstrated that the sensitivity for cAMP is dependent on the mode of activation of the T lymphocytes. IL-4 mRNA accumulation was not inhibited by prostaglandin E₂ (PGE₂) in A23187 plus phorbol myristate acetate (PMA)-stimulated T lymphocytes, whereas IL-4 mRNA was dose dependently decreased by PGE₂ in Con A- or anti-CD3 plus anti-CD28-stimulated T lymphocytes (5).

The observed downregulation of the studied cytokines was selectively mediated by the ₂AR. Evidence was derived from the experiments carried out with selective AR antagonists. ICI 118,551a selective ₂AR antagonist (20)was able to antagonize the inhibitory effect of isoproterenol on the accumulation of IFN-, GM-CSF, and IL-3 mRNA with pK_B values of 7.82 ± 0.31, 7.71 ± 0.09, and 7.37 ± 0.16, respectively. Although these values are intermediate between the pA₂ values reported for ₂AR in guinea pig tracheal smooth muscle (8.72) and ₁AR guinea pig atria (7.19) (30), it should be realized that binding of lipophilic -adrenoceptor antagonists like ICI 118,551 to serum proteins present in the incubation medium may strongly diminish the free drug concentration that will lower pK_B or pA₂ values (31). Moreover, coincubation with the (very hydrophilic) selective ₁AR antagonist atenolol had no effect at all on the inhibition of isoproterenol or fenoterol. Therefore, the observed downregulation of the Con A-induced cytokine expression by isoproterenol and fenoterol is selectively mediated by the ₂AR, whereas apparently no ₁AR is involved.

In addition, we also demonstrated the occurrence of ₃AR mRNA in T lymphocytes by using the RT-PCR technique. Although significant levels of ₃AR mRNA were detectable in T lymphocytes from four different donors, we did not find functional activity of the ₃AR in the cytokine regulation of T lymphocytes. A possible explanation for this observation might be that ₃AR mRNA in T lymphocytes is, at most, partially translated into protein. Besides, stimulation by the selective ₃AR agonist BRL 37344 did not result in increased intracellular cAMP levels. Interestingly, human adipocytes, in contrast to rat adipocytes, also have a very low responsiveness for BRL 37344 with respect to cAMP accumulation as well as lipolysis, indicating the marginal role of ₃AR in human adipocyte lipolysis compared with ₁AR and ₂AR (32).

In conclusion, our data present evidence that AR stimulation exerts an important negative regulatory feedback control mechanism in the expression of T-cell- derived cytokines, which is solely mediated by the ₂AR subtype. In asthmatic patients, a reduced responsiveness of the cAMP-dependent signaling pathway has been observed after an allergen challenge (18, 33). As a consequence of this disturbed inhibitory control mechanism, an augmented expression of IL-3, GM-CSF, and IFN-, which are potent mediators in the development and course of airway inflammation, should be envisaged.