An altered production of cytokines in HIV-infected patients may contribute to the disease progression and pathogenesis of HIV infection in several ways [1,2]. The levels of pro-inflammatory cytokines such as tumour necrosis factor (TNF)-α, interleukin (IL)-1, and IL-6 are elevated in HIV-infected individuals and may induce virus replication [3,4]. In contrast, cytokines such as interferon (IFN)-α, IFN-β, and IL-16 have been reported to suppress virus replication [5–7]. Moreover, other cytokines such as IL-2, IL-4, IL-10, IL-12, IL-18, transforming growth factor-β, and IFN-γ have been shown to have dual roles either enhancing or suppressing HIV expression, depending on experimental conditions [8–20]. Furthermore, chemokines and their receptors are also important modulators of HIV replication. CCR5 and CXCR4 are demonstrated to be co-receptors for entry of HIV to the host cell, whereas their corresponding chemokine ligands may block virus entry [21,22]. In addition, cytokine dysregulation also appears to contribute to immunodeficiency and clinical manifestation in HIV-infected patients by mechanisms such as enhanced T-cell apoptosis, induction of endothelial cell dysfunction and increased oxidative stress.
The adenylate cyclase–cAMP–protein kinase A (PKA) pathway plays an important role in the regulation of immune function [23,24]. It has been shown that cAMP – through PKA type I – abolishes the T-cell activation induced via the T-cell receptor (TCR)/CD3 complex by activation of c-terminal Src kinase, thereby inhibiting signalling at the level of Src kinases [25–27]. We have earlier reported that cAMP levels are elevated in T cells from HIV-infected patients and that a PKA type I selective antagonist reverses T-cell immune function in these patients, making PKA type I an interesting drug target for immune modulation in HIV infection [28,29].
In this study, we analysed the level of cytokine gene expression in unstimulated T cells from HIV-infected patients on highly active antiretroviral therapy (HAART) compared to normal T cells. Furthermore, we studied the regulation of cytokine and cytokine-related genes by cAMP agonist and antagonist treatment of anti-CD3 activated T cells to understand better the effects of the cAMP/PKA pathway on cytokine networks in HIV infection.
Patients and blood donors
Heparinized blood samples were obtained from 10 HIV-seropositive patients (eight men and two women; mean age 47 years (range, 29–65) years) on HAART, who had been stable in virological and immunological parameters for > 6 months prior to the study and had no inter-current infections. Patients had CD4 T-cell counts of 403 ± 44 (mean ± SEM) × 106/l, CD4/CD8 ratios of 0.44 ± 0.11 (mean ± SEM), and median viral loads of 145 HIV RNA copies/ml (range, 1–10 000; distribution: < 50, n = 4; > 50 < 400, n = 1; > 400, n = 5) HIV RNA copies/ml. Healthy blood donor samples (whole blood) were obtained from Ullevaal University Hospital Blood Center, Oslo, Norway.
Peripheral blood CD3 T cells were purified by negative selection as described previously  and checked for purity by flow cytometry (> 95% CD3 T cells). T cells suspended in RPMI-1640 medium containing 2 mM L-glutamine, 1% non-essential amino acids, 1 mM sodium pyruvate, 100 U penicillin/streptomycin supplemented with 10% heat-inactivated foetal calf serum (GibcoBRL, Paisley, UK), were incubated in flat-bottomed 96-well plates (Costar, Cambridge, Massachusetts, USA) in a CO2 incubator at 37°C in the presence of mouse IgG anti-CD3 antibodies (clone SpvT3b; 10 ng/ml; Zymed Laboratories, South San Francisco, California, USA) with cross-ligation of the TCR/CD3 complex by addition of magnetic beads coated with sheep anti-mouse IgG (Dynal beads M-450 Sheep anti-mouse IgG, Dynal, Oslo, Norway) at a cell : to : bead ratio of 1 : 1. Cells were harvested after 3, 6, 12 and 24 h. When used, cAMP agonist (8- CPT-cAMP; 300 μM; BioLog Life Science, Bremen, Germany) and PKA type I selective cAMP antagonist (Rp-8-Br-cAMPS; 1000 μM; BioLog) were added to cultures 30 min before stimulation.
Purification of poly A+ mRNA
Poly A+ mRNA was purified from pooled T-cell lysates (HIV patients: n = 10; healthy controls: n = 3) by using Dynabeads mRNA DIRECT Kit (Dynal) according to the manufacturer's instructions.
cDNA array hybridization
For synthesis of 33P-labelled cDNA, cDNA Labelling and Hybridization Kit and Human Cytokine-Specific Primers (R&D Systems, Abingdon, Oxon, UK) were used. Hybridization was performed according to the manufacturer's instructions. Briefly, Cytokine cDNA Expression Array Membranes (R&D Systems) were pre-hybridized for 2 h at 65°C in the presence of 100 μg/ml salmon testes DNA (R&D Systems). Subsequently, cDNA probe was allowed to hybridize to the array membranes overnight at 65°C. To visualize the hybridization signals, Super Signal Phosphor Screens (Packard, Pangbourne, Berkshire, UK) and Cyclone Phosphor-Imager (Packard) were used.
Individual hybridization signals were identified and quantified densitometrically using Phoretix Array software. Background was subtracted from each spot based on the signal intensity of negative control spots (normalized to the signal intensity of β-actin). When duplicate spots, representing one gene, were more than 1.3 × divergent, the highest signal was excluded. Signals that were twofold or greater in intensity level compared to background were included.
Assay variability and biological variation
To determine the intra-assay variability, negatively selected peripheral T cells from a healthy blood donor were stimulated by cross-ligation of the TCR/CD3 complex for 6 h followed by poly A+ mRNA isolation and subsequent cDNA probe synthesis. The cDNA probe was divided into four aliquots and hybridized to four different cytokine gene expression macro array membranes simultaneously. We found that the average SEM was 1.0% for empty spots (defining the background) and 22.1% for specific signals (n = 4). Similarly, to examine the biological variation between individuals, we looked at the level of expression of cytokine related genes in peripheral T cells from three different healthy blood donors in parallel under all four experimental conditions used in the study and we found that the average SEM as a percentage of the mean across the whole dataset was 33% between donors, which also included the assay variability. Following this examination, we pooled RNA from three donors for comparison with pooled patient samples in order to average signals in both groups, as patient samples were too small to allow individual assessment.
Real-time quantitative RT–PCR
Sequence specific PCR primers and TaqMan probes were designed using the Primer Express software version 1.5 (Applied Biosystems, Foster City, California, USA). Quantification of mRNA was performed using the ABI Prism 7700 (Applied Biosystems) as suggested by the manufacturer. Briefly, 2 ng mRNA was reverse transcribed using TaqMan Reverse Transcription reagents (Applied Biosystems). TaqMan and SyBr Green assays were performed using 2 × TaqMan Universal Master Mix and 2 × SyBr Green Universal Master Mix (Applied Biosystems), respectively. β-actin gene expression was analysed using Human β-actin TaqMan Pre-developed Assay Reagents (Applied Biosystems) and used to adjust for unequal amounts of mRNA.
Northern blot analysis for β-actin
T cells from three healthy blood donors were left untreated or stimulated by cross-ligation of TCR/CD3 complex for 6 h in the absence and presence of cAMP agonist or cAMP antagonist. Next, 20 μg of extracted RNA was subjected to Northern blot analysis as described elsewhere  using a DNA probe for β-actin. The level of gene expression of β-actin was unchanged at the four different conditions used (data not shown).
Time kinetics of cytokine and cytokine-related genes following anti-CD3 T-cell activation
T cells from three blood donors were untreated or stimulated by cross-ligation of the TCR/CD3 complex for 3, 6, 12 and 24 h and cytokine gene expression was analysed. Array data showed a biphasic cytokine profile with maximal numbers of regulated genes observed after 6 h of anti-CD3 T-cell stimulation (Fig. 1); 6 h stimulation was thus used for further studies. The fact that the maximal number of regulated genes was observed at 6 h is due to an overlap between early (3–6 h) and late (6–24 h) regulated genes at this time.
The level of expression of cytokine genes in unstimulated T cells from HIV-infected patients on HAART
Next, we explored the level of spontaneous expression of cytokine and cytokine-related genes in unstimulated T cells from HIV-infected patients on HAART compared to normal T cells by screening for the number of genes that were hyper- and hypo-activated in the two cell populations. Array results showed that 45% (169/375) of cytokine-related genes were expressed at twofold or higher levels in unstimulated T cells from HIV-infected patients compared to levels in normal T cells (Table 1). Genes with 10-fold higher expression levels in unstimulated patient T cells are listed in Table 2. Interestingly, 33% (56/169) of the genes with elevated spontaneous expression in patient T cells were hypo-responsive to anti-CD3 stimulation, even though a significant regulation occurred in healthy controls. Moreover, data from real-time RT–PCR showed that the spontaneous expression levels of seven out of nine selected genes from the array data (based on regulation/role in HIV infection; TNF-α, CCR5, CXCR4, amphiregulin, I-309, 4-1BB, macrophage inflammatory protein (MIP)-1α, MIP-1β, and lymphotoxin-β) were between 1.5- and 3.6-fold higher in unstimulated T cells from HIV-infected patients compared to healthy individuals, whereas two genes were expressed at similar levels in both patient and control T cells, supporting the observations made with the array analysis (Fig. 2).
Cyclic AMP-regulated cytokine genes in anti-CD3 stimulated T cells from HIV-infected patients on HAART
The cAMP/PKA type I pathway has been shown to be hyper-activated in T cells from HIV-infected patients [28,29]. To examine the effect of modulation of the cAMP signalling pathway on cytokine and chemokine networks, T cells from HIV-infected patients on HAART and healthy blood donors were triggered via the TCR/CD3 complex in the presence and absence of cAMP agonist or cAMP antagonist followed by macro array analysis.
Overall, array data showed that a greater number of genes were regulated by cAMP agonist in activated T cells from HIV-infected patients compared to healthy individuals (Table 3). Furthermore, the majority of these genes was down-regulated whereas few genes were up-regulated by cAMP agonist in anti-CD3 activated patient T cells. Notably, the reverse pattern was observed in controls, where the majority of regulated genes went up in response to cAMP agonist. Certain genes were oppositely regulated by cAMP agonist and cAMP antagonist, indicating some level of tonic regulation by endogenous cAMP in patient cells (labelled by an asterisk in Table 4). This included MIP-1β (down-regulated by agonist) and CXCR4 and amphiregulin (up-regulated by agonist). Moreover, other groups of genes such as chemokines (e.g., I-309, MIP-1α, MIP-3α, RANTES and SDF-1), cytokines (e.g., IFN-γ) and TNF superfamily (e.g., lymphotoxin β and TNF-α) were down-regulated by cAMP agonist, but not up-regulated by cAMP antagonist in anti-CD3 activated patient T cells (Table 4).
A set of cAMP-regulated genes, as described above, was selected for further quantification by real-time RT–PCR. Data from this analysis showed that CXCR4 and CCR5 were strongly up-regulated by cAMP agonist in anti-CD3 activated T cells from both HIV-infected patients and from healthy controls (Fig. 3a and b). Interestingly, the gene expression level of CCR5 was nearly fourfold higher in unstimulated T cells from HIV-infected patients compared to controls. Amphiregulin was profoundly up-regulated following cAMP agonist treatment and markedly down-regulated by cAMP antagonist in anti-CD3 activated T cells from both patients and controls (Fig. 3c). Moreover, MIP-1β was up-regulated upon anti-CD3 stimulation, whereas it was strongly down-regulated in T cells from both healthy and HIV-infected patients treated with cAMP agonist (Fig. 3d). TNF-α was up-regulated upon T-cell activation whereas strongly down-regulated by cAMP agonist in anti-CD3 activated T cells from both healthy and HIV-infected individuals. Interestingly, the gene expression level of TNF-α was almost threefold higher in unstimulated T cells from HIV-infected patients compared to healthy control T cells (Fig. 3e). Activation of T cells from healthy and HIV-infected individuals resulted in a down-regulation of lymphotoxin-β gene expression levels, which were reduced even further following cAMP agonist treatment (Fig. 3f).
HIV infection induces multiple changes to cytokine mRNA expression, and circulating CD4 T cells in HIV infection have been shown to have an increased mRNA level on a per cell basis for cytokines such as IL-10, TNF-α and IFN-γ . It was recently reported that normal proliferating peripheral blood mononuclear cells (PBMC) infected in vitro with the T-cell tropic laboratory strain of HIV-1 (RF) showed an overall increase in the number of genes expressed post-infection , supporting our observations that 45% of cytokine related genes were expressed at twofold or higher levels in the CD3 T cells of HAART-treated patients compared to healthy blood donors and suggesting that patient cells are in a pre-activated state. Moreover, our data showed that 33% of the genes that were hyper-activated in unstimulated patient T cells were hypo-responsive upon activation compared to control T cells, which suggests that patient T cells are pre-activated in vivo, and could be refractory to further activation. The patients included in the study are on HAART and more profound differences might have been seen if patients with higher viral load were studied. Despite this, a marked spontaneous activation of T cells from HIV patients even on HAART was observed that could contribute to the persistent T cell dysfunction in these patients. Notably, our data does not correlate directly with CD3 T cells from healthy individuals, as the CD4 : CD8 ratios would vary between patient and normal T cells.
In this study we found that the cAMP/PKA type I pathway markedly influenced several chemokines and chemokine receptors at the mRNA level. In HIV-infected patients, members of the chemokine/chemokine receptor family constituted 26% of the genes regulated by cAMP agonist/antagonist, which constitutes 12.5% of the genes on the macro array. The chemokine receptors CCR5 and CXCR4 were expressed at higher levels in unstimulated T cells from HIV-infected patients compared to control T cells, and were up-regulated by cAMP agonist, indicating that the higher gene expression levels in patient T cells could be due to the elevated level of cAMP in these cells. Interestingly, it was recently demonstrated that a CRE element within the CXCR4 promoter regulates CXCR4 levels in response to changes in cAMP signalling . Notably, the gene expression of the cognate ligand (SDF-1) of CXCR4 was reduced by cAMP.
The gene expression of chemokines with an anti-HIV effect was down-regulated by cAMP agonist in activated T cells from HIV-infected patients, e.g., RANTES and MIP-1β. Cyclic AMP agonist treatment down-regulated the MIP-1β levels, whereas cAMP antagonist slightly increased the gene expression level of MIP-1β, suggesting a regulation of gene expression via an endogenously active PKA type I pathway. Lower levels of RANTES have been reported in patients compared to healthy individuals . This could be due to increased cAMP levels in patients as cAMP agonist down-regulated the gene expression of RANTES. The effects of cAMP on chemokines may be relevant in HIV infection as reports state that MIP-1β and RANTES block viral entry and inhibit viral replication [22,35]. Again, a receptor and its cognate ligand(s) were oppositely regulated by cAMP: cAMP induced the gene expression of CCR5, whereas the gene expression of MIP-1β and RANTES were decreased. However, the protein levels of these components of the cytokine network could be comparably different.
Cyclic AMP agonist also down-regulated a set of inflammatory cytokines. There was an increased gene expression of TNF-α in unstimulated HIV-infected T cells compared to normal T cells, consistent with other reports where the levels of TNF-α mRNA in PBMC from HIV-infected patients were highly increased [31,34,36,37]. Furthermore, we observed a down-regulation of TNF-α by cAMP agonist, which indicates that the increased levels of TNF-α in T cells from HIV-infected patients are not due to elevated levels of cAMP, whereas cAMP antagonist moderately up-regulated TNF-α in anti-CD3 activated patient T cells. Down-regulation by cAMP agonist was also observed for IFN-γ, which could be of potential importance in HIV infection, as this cytokine is associated with viral suppression and lack of disease progression . However, IFN-γ has been reported to induce HIV replication as well .
Amphiregulin belongs to the epidermal growth factor (EGF) family and has been found to be present in a large number of both malignant and benign colon tumours, as well as in normal colon tissue . Our data showed that amphiregulin mRNA was expressed at higher levels in unstimulated T cells from HIV-infected patients compared to T cells of normal controls, and was strongly up-regulated when activated T cells were treated with cAMP agonist, whereas a markedly reduced gene expression was seen with cAMP antagonist. This suggests that this EGF-like protein is under control of endogenous cAMP levels. However, the role of amphiregulin in T cells and in HIV infection will have to be examined in more detail in future studies.
In conclusion, our data show that cytokine networks are pre-activated in T cells from HIV-infected patients compared to controls, and indicate that a set of cytokine genes in activated patient T cells is under cAMP-mediated tonic regulation, i.e., affected by endogenous hyper-activation of the cAMP/PKA type I pathway. The use of a PKA type I selective cAMP antagonist may have a positive effect on cytokine networks in patient T cells by normalizing gene expression levels and providing a stronger antiviral response. Future studies will show how the activation status of the cytokine network and regulation by the cAMP/PKA type I pathway is affected in various T cell subsets (e.g., CD4+CD25+, CD4+CD25−).
The authors thank G. Josefsen and B. Lunden for skilful technical assistance.
Sponsorship: Supported by The Program for Advanced Studies in Medicine, The Research Council of Norway, The University of Oslo (EMBIO programme), Novo Nordic Research Foundation Committee, Anders Jahre's Foundation and European Union RTD grant no. QLK3-CT-2002-02149.
1. Graziosi C, Gantt KR, Vaccarezza M, Demarest JF, Daucher M, Saag MS, et al
. Kinetics of cytokine expression during primary human immunodeficiency virus type 1 infection. Proc Natl Acad Sci USA
2. Khoo SH, Pepper L, Snowden N, Hajeer AH, Vallely P, Wilkins EGL, et al
. Tumour necrosis factor c2 microsatellite allele is associated with the rate of HIV disease progression. AIDS
3. Fauci AS. Host factors and the pathogenesis of HIV-induced disease. Nature
4. Poli G. Cytokines and the human immunodeficiency virus: from bench to bedside. Eur J Clin Invest
5. Baier M, Werner A, Bannert N, Metzner K, Kurth R. HIV suppression by interleukin-16. Nature
6. Poli G, Orenstein JM, Kinter A, Folks TM, Fauci AS. Interferon-alpha but not AZT suppresses HIV expression in chronically infected cell lines. Science
7. Vieillard V, Cremer I, Lauret E, Rozenbaum W, Debre P, Autran B, et al
. Interferon beta transduction of peripheral blood lymphocytes from HIV-infected donors increases Th1-type cytokine production and improves the proliferative response to recall antigens. Proc Natl Acad Sci USA
8. Akridge RE, Reed SG. Interleukin-12 decreases human immunodeficiency virus type 1 replication in human macrophage cultures reconstituted with autologous peripheral blood mononuclear cells. J Infect Dis
9. Al Harthi L, Roebuck KA, Landay A. Induction of HIV-1 replication by type 1-like cytokines, interleukin (IL)-12 and IL-15: effect on viral transcriptional activation, cellular proliferation, and endogenous cytokine production. J Clin Immunol
10. Angel JB. Improving immune function and controlling viral replication in HIV-1-infected patients with immune-based therapies. AIDS Read
11. Bailer RT, Holloway A, Sun J, Margolick JB, Martin M, Kostman J, et al
. IL-13 and IFN-gamma secretion by activated T cells in HIV-1 infection associated with viral suppression and a lack of disease progression. J Immunol
12. Biswas P, Poli G, Orenstein JM, Fauci AS. Cytokine-mediated induction of human immunodeficiency virus (HIV) expression and cell death in chronically infected U1 cells: do tumor necrosis factor alpha and gamma interferon selectively kill HIV-infected cells? J Virol
13. Choi HJ, Dinarello CA, Shapiro L. Interleukin-18 inhibits human immunodeficiency virus type 1 production in peripheral blood mononuclear cells. J Infect Dis
14. Kinter AL, Poli G, Fox L, Hardy E, Fauci AS. HIV replication in IL-2-stimulated peripheral blood mononuclear cells is driven in an autocrine/paracrine manner by endogenous cytokines. J Immunol
15. Kinter AL, Bende SM, Hardy EC, Jackson R, Fauci AS. Interleukin 2 induces CD8+ T cell-mediated suppression of human immunodeficiency virus replication in CD4+ T cells and this effect overrides its ability to stimulate virus expression. Proc Natl Acad Sci USA
16. Poli G, Kinter AL, Justement JS, Bressler P, Kehrl JH, Fauci AS. Transforming growth factor beta suppresses human immunodeficiency virus expression and replication in infected cells of the monocyte/macrophage lineage. J Exp Med
17. Poli G, Kinter AL, Justement JS, Bressler P, Kehrl JH, Fauci AS. Retinoic acid mimics transforming growth factor beta in the regulation of human immunodeficiency virus expression in monocytic cells. Proc Natl Acad Sci USA
18. Pugliese A, Gennero L, Vidotto V, Speranza F, Tambini R, Torre D. Interleukin-18 enhances HIV-1 production in a human chronically-infected T cell line (H9-V). Cell Biochem Funct
19. Valentin A, Lu W, Rosati M, Schneider R, Albert J, Karlsson A, et al
. Dual effect of interleukin 4 on HIV-1 expression: implications for viral phenotypic switch and disease progression. Proc Natl Acad Sci U S A
20. Weissman D, Poli G, Fauci AS. IL-10 synergizes with multiple cytokines in enhancing HIV production in cells of monocytic lineage. J Acquir Immune Defic Syndr Hum Retrovirol
21. Berger EA, Murphy PM, Farber JM. Chemokine receptors as HIV-1 coreceptors: roles in viral entry, tropism, and disease. Annu Rev Immunol
22. Kalinkovich A, Weisman Z, Bentwich Z. Chemokines and chemokine receptors: role in HIV infection. Immunol Lett
23. Kammer GM. The adenylate cyclase-cAMP-protein kinase A pathway and regulation of the immune response. Immunol Today
24. Torgersen KM, Vang T, Abrahamsen H, Yaqub S, Tasken K. Molecular mechanisms for protein kinase A-mediated modulation of immune function. Cell Signal
25. Skalhegg BS, Landmark BF, Doskeland SO, Hansson V, Lea T, Jahnsen T. Cyclic AMP-dependent protein kinase type I mediates the inhibitory effects of 3',5'-cyclic adenosine monophosphate on cell replication in human T lymphocytes. J Biol Chem
26. Skalhegg BS, Tasken K, Hansson V, Huitfeldt HS, Jahnsen T, Lea T. Location of cAMP-dependent protein kinase type I with the TCR-CD3 complex. Science
27. Vang T, Torgersen KM, Sundvold V, Saxena M, Levy FO, Skålhegg BS, et al
. Activation of the COOH-terminal Src kinase (Csk) by cAMP-dependent protein kinase inhibits signaling through the T cell receptor. J Exp Med
28. Aandahl EM, Aukrust P, Skalhegg BS, Müller F, Froland SS, Hansson V, et al
. Protein kinase A type I antagonist restores immune responses of T cells from HIV-infected patients. FASEB J
29. Aandahl EM, Aukrust P, Muller F, Hansson V, Tasken K, Froland SS. Additive effects of IL-2 and protein kinase A type I antagonist on function of T cells from HIV-infected patients on HAART. AIDS
30. Dahle MK, Knutsen HK, Tasken KA, Pilz R, Tasken K. Cyclic AMP regulates expression of the RI alpha subunit of cAMP-dependent protein kinase through an alternatively spliced 5’ UTR. Eur J Biochem
31. Fan J, Bass HZ, Fahey JL. Elevated IFN-gamma and decreased IL-2 gene expression are associated with HIV infection. J Immunol
32. Vahey MT, Nau ME, Jagodzinski LL, Yalley-Ogunro J, Taubman M, Michael NL, et al
. Impact of viral infection on the gene expression profiles of proliferating normal human peripheral blood mononuclear cells infected with HIV type 1 RF. AIDS Res Hum Retroviruses
33. Cristillo AD, Highbarger HC, Dewar RL, Dimitrov DS, Golding H, Bierer BE. Up-regulation of HIV coreceptor CXCR4 expression in human T lymphocytes is mediated in part by a cAMP-responsive element. FASEB J
34. Alonso K, Pontiggia P, Medenica R, Rizzo S. Cytokine patterns in adults with AIDS. Immunol Invest
35. Cohen DE, Walker BD. Human immunodeficiency virus pathogenesis and prospects for immune control in patients with established infection. Clin Infect Dis
36. Poli G, Fauci AS. The effect of cytokines and pharmacologic agents on chronic HIV infection. AIDS Res Hum Retroviruses
37. Vyakarnam A, McKeating J, Meager A, Beverley PC. Tumour necrosis factors (alpha, beta) induced by HIV-1 in peripheral blood mononuclear cells potentiate virus replication. AIDS
38. Saeki T, Stromberg K, Qi CF, Gullick WJ, Tahara E, Normanno N, et al
. Differential immunohistochemical detection of amphiregulin and cripto in human normal colon and colorectal tumors. Cancer Res