β-Chemokine production in CD40L-stimulated monocyte-derived macrophages requires activation of MAPK signaling pathways

www.elsevier.com/locate/jnlabr/ycyto Cytokine 23 (2003) 53–63

b-Chemokine production in CD40L-stimulated monocyte-derived macrophages requires activation of MAPK signaling pathways Paola Di Marzioa,*, Barbara Sherryb, Elaine K. Thomasc, Giovanni Franchinb, Helena Schmidtmayerovaa, Michael Bukrinskya,d a

Department of Molecular Pathogenesis of HIV, The Picower Institute for Medical Research, 350 Community Drive, Manhasset, NY 11030, USA b Department of Cytokine Biology, The Picower Institute for Medical Research, 350 Community Drive, Manhasset, NY 11030, USA c Immunex Corporation, 51 University Street, Seattle, WA 98101, USA d Department of Microbiology and Tropical Medicine, The George Washington University, 2300 Eye Street, Washington, DC 20037, USA Received 27 October 2002; received in revised form 10 May 2003; accepted 16 May 2003

Abstract CD40 ligand is a cell surface molecule on CD4+ T cells that interacts with its receptor, CD40, on antigen presenting cells to mediate humoral and cellular immune responses. Our previous studies demonstrated that a trimeric soluble form of CD40L (CD40LT) activates macrophages to produce b-chemokines and decrease CCR5 and CD4 cell surface expression, thus inducing resistance to HIV-1 infection. However, the mechanism(s) by which CD40LT mediates these effects in primary macrophages remains unclear. In this report, we demonstrate that CD40LT induces synthesis of b-chemokines through the activation of MAPK signaling pathways. Treatment of macrophages with CD40LT results in a rapid activation of p38 and ERK1/2 mitogen-activated protein kinases. Inhibitors of these MAPKs blocked b-chemokine production, while protein kinase A and C inhibitors had little or no effect. We also provide evidence that CD40LT stimulates b-chemokine production directly, as well as indirectly via a TNF-a-dependent mechanism. At the early time points, CD40LT directly stimulated b-chemokine production, whereas at later time points the effect was mediated to some extent by TNF-a. In conclusion, our results suggest that CD40–CD40L interactions are important for the activation of monocyte-derived macrophage antiviral response affecting both viral replication and the recruitment of immune cells. Ó 2003 Elsevier Ltd. All rights reserved. Keywords: CD40L; MAPKs; Macrophages

1. Introduction CD40 receptor (CD40) is a 45–50 kDa protein that is a member of the tumor necrosis factor receptor (TNFR) superfamily. Interaction of CD40 with its ligand, CD40 ligand (CD40L), has been shown to play a crucial role in cellular and humoral immune response. For example, CD40 promotes priming of T-helper type I cells, and interaction between T-cell CD40L and B-cell CD40 leads to B-cell proliferation and differentiation into antibodysecreting plasma cells [1]. CD40 is expressed on numerous cell types besides B cells, including monocyte/macro* Corresponding author. Center of Immunology and Inflammation, North Shore-LIJ Research Center, 350 Community Drive, Manhasset, NY 11030, USA. E-mail address: [email protected] (P. Di Marzio). 1043-4666/03/$ - see front matter Ó 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S1043-4666(03)00186-8

phages, endothelial cells, dendritic cells, fibroblasts, vascular smooth muscle cells and neurons [2,3] whereas CD40L is expressed on activated CD4+ T cells, platelets and mast cells [4–6]. Early studies of CD40 focused on its role in co-stimulation of B cell proliferation and immunoglobulin isotype switching [7]. A broader role for CD40 signaling was revealed when other cell types, in particular macrophages, were found to express CD40. Signaling through CD40 in monocyte/macrophages results in up-regulation of co-stimulatory molecules and rescue from apoptosis [8–10]. In these cells, CD40– CD40L interaction also induces the production of proinflammatory cytokines, such as IL1a/b, IL-6, IL-8, IL-12 and TNF-a, which are important regulators of both humoral and cellular immune responses [8,11]. In addition, CD40L stimulates the production by macrophages of a group of chemoattractant cytokines, the

54

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

b-chemokines, such as, monocyte chemotactic protein 1 (MCP-1), macrophage inflammatory protein-1a and -1b (MIP-1a, MIP-1b) and regulated upon activation, normal T expressed and secreted (RANTES) [12–14]. b-Chemokines play an important role in the development of immune responses by stimulating the migration of immune cells to the appropriate tissues or compartments within tissues. Following the initial observation of anti-HIV activity of MIP-1a, MIP-1b and RANTES [15], it was found that CCR5, one of the receptors for the b-chemokines MIP-1a, MIP-1b and RANTES, also functions as the principal co-receptor for M-tropic strains of HIV-1 [16,17] and numerous studies demonstrated the role of these proteins in the pathogenesis of HIV-1 infection [18,19]. It is well known that macrophages play a key role in HIV transmission and pathogenesis, but the specific molecular mediators of the macrophage-associated immune response against HIV-1 are not well characterized. Because we showed that macrophages secrete bchemokines after CD40L ligation, we wished to investigate the signaling pathway(s) induced by CD40L. It has been reported that CD40-mediated responses in monocytes depend on the induction of protein tyrosine kinase (PTK), which is required for both the activation of inflammatory cytokine synthesis and rescue from apoptosis [20,21]. Recently it was reported that in human monocytes and THP-1 cells the CD40 ligation activates two of the mitogen-activated protein kinases (MAPKs), extracellular regulated kinase 1/2 (ERK 1/2) and Jun N-terminal kinase 1 (JNK), but not p38 [22,23]. The ERK 1/2 pathway was shown to be directly involved in the CD40-mediated secretion of the proinflammatory cytokines, such as TNF-a, IL-6 and IL-8. In our previous studies, we demonstrated that a trimeric soluble form of CD40L, CD40LT, activates monocyte-derived macrophages (MDMs) to produce high levels of b-chemokines and to reduce CCR5 and CD4 cell surface expression [13]. However, the mechanism of this activity has not been defined, and it has not been determined whether chemokine upregulation was a direct effect of CD40 signaling or was mediated by some intermediate factor. Here we show that in MDMs CD40 ligation by CD40LT activates ERK1/2 and p38, but not JNK1/2. This activation directly upregulates secretion of the b-chemokines MIP-1a, MIP-1b and RANTES, and also of the pro-inflammatory cytokine TNF-a. 2. Results

chemokines, MIP-1a, MIP-1b and RANTES [13]. We now asked the question whether CD40LT regulates bchemokine production at a transcriptional or posttranscriptional level. To address this question, we first performed a time course experiment to determine the earliest time of b-chemokine release after CD40LT stimulation. MDMs were treated with CD40LT (1 lg/ ml) for 1, 3, 6 and 18 h and left untreated. Cell culture supernatants were collected and tested for the presence of b-chemokines and the proinflammatory cytokine, TNF-a, by specific ELISAs. As shown in Fig. 1A, untreated MDM released low amounts of MIP-1a, MIP-1b and TNF-a. The addition of CD40LT to MDM markedly enhanced bchemokine and TNF-a production. MDM released MIP1a and MIP-1b as early as 1 h after CD40LT treatment. CD40LT-induced b-chemokine release was biphasic. bChemokine levels increased at 3 h and did not change after 6 h. After 18 h of CD40LT stimulation, however, the bchemokine production was markedly increased again. In contrast, RANTES was hardly detectable at 3 h and increased at 6 and 18 h. TNF-a production steadily increased over the course of the treatment. Interestingly, MIP-1a and MIP-1b production decreased after 18 h, while RANTES levels remained elevated up to three days after stimulation (data not shown). Furthermore, MDM from all tested donors secreted b-chemokines in response to CD40LT concentrations over 0.5 lg/ml. No significant induction was observed at lower doses (data not shown). To determine whether CD40LT regulates b-chemokine production at the transcriptional or post-transcriptional level, we analyzed b-chemokine-specific mRNA. As shown in Fig. 1B, CD40LT induced accumulation of MIP-1a, MIP-1b and, to a lesser extent, RANTES mRNA as early as 1 h after stimulation. This result suggested that CD40LT-stimulated transcription of the b chemokine genes. In support of this assumption, actinomycin D (Act D), an RNA polymerase II inhibitor, completely blocked accumulation of b-chemokine mRNAs (Fig. 2B) and proteins (Fig. 2A). To exclude the possibility that the effect of CD40LT was mediated by expression of an intermediate factor, we tested whether cycloheximide (CHX), an inhibitor of protein synthesis, would block CD40LT-induced bchemokine mRNA accumulation. As expected, CHX blocked MIP-1a and MIP-1b protein synthesis (Fig. 2A). In contrast, it induced a marked increase in the steady state level of b-chemokine transcripts (Fig. 2B). These results indicate that CD40LT directly activates transcription of MIP-1a, MIP-1b and RANTES genes, apparently via a signaling cascade.

2.1. b-Chemokine production by CD40LTstimulated MDM is regulated at the transcriptional level

2.2. Late-stage CD40LT-induced b-chemokine production is partially mediated by TNF-a

Previously, we had demonstrated that CD40LTstimulated macrophages secrete high levels of the b-

While results described in the previous section indicated that b-chemokine production early after

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

55

Fig. 1. Time course of CD40LT-induced b-chemokine production in MDMs. (A) Secretion of b-chemokines by MDMs stimulated with CD40LT. Freshly isolated monocytes from healthy donors were cultured (1
106 cells/ml) in medium containing M-CSF (2 ng/ml). After seven days of culture the cells were treated with CD40LT (1 lg/ml) or medium alone. Supernatants were harvested at 1, 3, 6 and 18 h as indicated. The concentration of MIP-1a, MIP-1b, RANTES and TNF-a in the supernatants was measured by ELISA. One representative experiment out of two is shown. (B) Induction of b-chemokine gene expression by CD40LT in MDM. Total cellular RNA was extracted from MDM stimulated with CD40LT (1 lg/ml) for 1 or 3 h as indicated above. MIP-1a, MIP-1b and RANTES mRNA expression was analyzed by RT-PCR assay as described in Section 4. PCR experiments were run for 25 cycles for MIP-1a and MIP-1b and for 30 cycles for RANTES. Amplification of GAPDH was used as control for the amount of cDNA in each sample. A representative experiment out of three is shown.

CD40LT stimulation does not involve synthesis of an intermediate factor, they did not exclude the possibility that such factor, in particular TNF-a, could contribute to the late-stage synthesis of b-chemokines. To test this possibility, we treated MDM with CD40LT (1 lg/ml) in the presence of antibodies to TNF-a (10 lg/ml) for 3 or 18 h. As shown in Fig. 3A upper panel, MDM stimulated with CD40LT for 3 h secreted MIP-1a and MIP-1b, but not RANTES. Interestingly, CD40LTinduced b-chemokine production was not reversed by antibody to TNF-a. However, after 18 h of CD40LT treatment partial inhibition of b-chemokine production was observed. The addition of anti-TNF-a antibody led to a 30% decrease of MIP-1a and MIP-b and 50% decrease of RANTES secretion (Fig. 3A lower panel). In a control experiment, treatment of MDM with TNF-a induced MIP-1a and MIP-1b secretion by MDM in a dose-dependent manner, and this induction was blocked by the anti-TNF-a antibody (Fig. 3B). Taken together, these results suggest that the initial b-chemokines production is directly stimulated by CD40LT,

whereas at later time points, TNF-a contributes to increase of b-chemokine production. In our previous publication, we demonstrated that CD40LT-mediated activation of MDM decreased cell surface expression of CCR5, one of the main HIV-1 coreceptors [13]. Since CD40LT induces TNFa production [24], we analyzed the potential role of TNF-a in CD40LT-induced CCR5 downregulation. As shown in Fig. 4A, TNF-a antibody did not revert the CD40LTinduced downregulation of CCR5 (and CCR1, another MIP-1a receptor) measured by reduction of [125I]MIP1a binding. To further confirm the role of TNF-a in CD40LT-mediated CCR5 downregulation we determined the expression of CCR5 by flow cytometry. As shown in Fig. 4B antibody to TNF-a were not able to reverse the CCR5 downregulation induced by CD40LT treatment. In contrast, a cocktail of chemokine antibodies (anti-MIP-1a [5 lg/ml], anti-MIP-1b [5 lg/ml] and anti-RANTES [2 lg/ml]) was able to reverse the effect. This result suggests that CD40LT-induced CCR5 dowregulation is mediated by b-chemokines.

56

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

Fig. 2. CD40LT regulates b-chemokine production by MDM at the transcriptional level. (A) Cycloheximide and Act D block b-chemokine production in CD40LT-stimulated MDM. Macrophages were prepared as described in Fig. 1. After seven days in culture, the cells were treated with cycloheximide (CHX) (10 lg/ml) or Act D (1 lM) for 30 min before adding CD40LT (1 lg/ml). After 3 h supernatants were harvested and the concentration of MIP-1a, MIP-1b and RANTES in the supernatants was measured by ELISA. A representative experiment out of three is shown. (B) Effect of CHX and Act D on the expression of b-chemokine mRNA in MDM. Total cellular RNA was extracted from MDM stimulated with CHX (10 lg/ml) or Act D (1 lM) for 30 min. After 30 min the cells were stimulated with CD40LT (1 lg/ml) for 1 h. MIP-1a, MIP-1b and RANTES mRNA expression was analyzed by RT-PCR assay as described in Section 4. One representative experiment out of two is shown.

2.3. CD40LT upregulates p38 and ERK 1/2, but not JNK, in MDM

2.4. CD40LT-induced b-chemokine production is linked to p38 and ERK 1/2 activation

As a first step in analyzing the signal transduction mechanism(s) leading to b-chemokine production by CD40LT-stimulated MDM, we investigated whether MAPK pathways (ERK, JNK and p38) are activated by CD40LT. MDM were stimulated with CD40LT for various time points, lysed and subjected to immunoblot analysis for phosphorylated ERK 1/2, JNK and p38. As shown in Fig. 5A, CD40LT treatment resulted in a rapid activation of p38 and ERK 1/2. Phosphorylation of p38 (a measure of p38 activation) was detected 15 min after stimulation, and started to decrease after 30 min. A similar pattern of activation/phosphorylation was observed for ERK 1/2 MAPK, except that a secondary increase in phosphorylation was seen between 3 and 18 h (Fig. 5A). This increase was likely due to secretion of an autocrine factor such as TNF-a. In contrast to p38 and ERK 1/2, analysis of JNK 1/2 did not reveal any activation of this MAPK in CD40LT-treated MDM. As shown in Fig. 5B, CD40LT activates p38 and ERK1/2 in a dose-dependent manner. Activation of p38 was observed at CD40LT doses as low as 0.1 lg/ml, whereas higher CD40LT concentration (1 lg/ml) was required for ERK 1/2 activation. These results suggest that the p38-dependent pathway might be the primary signal transduction mechanism initiated by CD40LT in MDM.

Given the high level of activated p38 and ERK 1/2 in CD40LT-stimulated MDM, we tested whether activation of these signaling pathways was responsible for bchemokine production. Toward this end, we used specific inhibitors of p38 and ERK 1/2 activation, SB202190 and PD98059, respectively. MDM were pretreated with SB202190 and PD98059 individually (10 lM each) or in combination (5 lM each) for 1 h prior to CD40LT stimulation. Culture supernatants were collected for analysis following 3 and 18 h incubations in the presence of inhibitors. As shown in Fig. 6, MIP-1a and MIP-b production was partially blocked by both SB202190 and PD98059 3 h after CD40 ligation, and was completely shut off after 18 h of incubation with inhibitors. In contrast, TNF-a production was completely blocked at 3 h by a combination of p38 and ERK1/2 inhibitors. One possible explanation of this result was that longer incubation with the inhibitors was required to fully block the MAPK activity. To test this assumption, MDMs were pretreated with PD98059 and SB202190 (individually or in combination) for 24 and 4 h prior to CD40 stimulation. Longer pre-treatment time only partially decreased bchemokine production measured 3 h after CD40 ligation (data not shown). This result indicates that the early release of b-chemokines only partially depends on

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

57

MAPK signaling, while the late response is fully dependent on p38 and ERK 1/2. Such interpretation is consistent with the role of TNF-a in the late, but not the early, release of b-chemokines by CD40LTtreated macrophages. To further characterize the signaling events involved in the induciton of b-chemokines by CD40L we tested protein kinases A (H-89 2.5 lM; chelerythrine chloride 2 lM) and C (Ro31-8425, 1 nM) inhibitors. These inihibitors did not block CD40Linduced b-chemokine production (data not shown). Taken together these results suggested that Erk1/2 and p38 are the major signaling pathways responsible for bchemokine production in CD40L-stimulated MDM.

3. Discussion

Fig. 3. Antibody to TNF-a inhibits late-stage b-chemokine release from CD40LT-stimulated MDM. (A) Macrophages were treated with CD40LT (1 lg/ml) in the presence or absence of MAb against TNF-a or MAb isotype control (10 lg/ml). After 3 and 18 h of culture, supernatants were harvested and the concentration of MIP-1a, MIP-1b and RANTES was measured by ELISA. b-Chemokine production is presented as percentage  SE of expression relative to control (CD40L treated cells) taken as 100%. Results are the mean of three independent experiments. (B) Antibody to TNF-a blocks TNF-a-induced bchemokine secretion. MDM were left untreated or treated for 3 h with different doses of TNF-a (ranging from 0.1 to 10 ng/ml) in the presence or absence of MAb against TNF-a (10 mg/ml). Supernatants were harvested and tested for the presence of MIP-1b by ELISA.

Mononuclear phagocytes, and in particular macrophages and dendritic cells, are among the first to encounter HIV-1 in the body [25–27]. Through their effector functions, these cells contribute to the innate immune response to viral infection. An important component of this activity is secretion of factors with anti-HIV activity, such as interferons and b-chemokines [28,29]. In contrast, some of these factors have also been reported to have a stimulating effect on HIV replication and cell-mediated transmission [30–32]. However, the role of b-chemokines in HIV-1 pathogenesis in vivo remains to be elucidated. Macrophages produce b-chemokines upon interaction with T cells during antigen presentation. An important component of this interaction is the contact between CD40 on macrophages and CD40L (CD154) expressed on activated T lymphocytes; this contact determines to a large extent signaling exchange between these two cell types. While the capacity of CD40 to induce b-chemokine production by macrophages has been previously reported, the signaling events which trigger this response have not been determined. The present study was undertaken to elucidate the mechanisms of b-chemokine production in MDMs stimulated with a trimeric soluble form of CD40 ligand. Our initial experiments demonstrated that induction of MIP-1a, MIP-1b and RANTES in MDMs stimulated with CD40LT occurs at the level of transcription. This conclusion was supported by RT-PCR analysis and by experiments with RNA polymerase II inhibitor, Act D, which completely blocked b-chemokine production in CD40LT-stimulated MDM. This was an important issue to establish, as some agents (e.g. bacterial endotoxin) can induce production by macrophages of certain factors (e.g. TNF-a) at the post-transcriptional level [33]. Previous reports have shown that TNF-a can induce the secretion of b-chemokines, downregulate CCR5 and reduce HIV infection [12,34,35]. In addition, antibody to TNF-a was shown to block CD40L-stimulated b-chemokine production [12], suggesting that the latter

58

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

Fig. 4. CD40LT-mediated downregulation of CCR5 on MDM is mediated by TNF-a. (A) After seven days of culture, MDM were treated with CD40LT (1 lg/ml) in the presence or absence of MAb against TNF-a or MAb isotype control (10 lg/ml) for 3 h. At the end of incubation, cells were washed and incubated immediately with (125I)MIP-1a for 2 h at 4  C. Results are presented as percentage  SE of MIP-1a binding (after subtraction of non-specific background control) relative to control (untreated cells) taken as 100%. Results are the mean of two independent experiments. (B) MDM were stimulated with CD40LT (1 lg/ml) for 3 h in the presence or absence of anti-TNF-a antibody (10 lg/ml) or a cocktail of b-chemokine antibodies (antiMIP-1a (5 lg/ml), anti-MIP-1b (5 lg/ml) and anti-RANTES (2 lg/ml) or MAb isotype control (10 lg/ml). The cells were then immunostained for the bchemokine receptor CCR5 (CCR5-PE) and analyzed by flow cytometry. The results are presented as percentage  SE of mean fluorescence intensity (MFI; after subtraction of isotype control) relative to control (untreated cells) taken as 100%. Results are the mean of two independent experiments.

effect might be secondary to CD40L-induced secretion of TNF-a. In this report, we provide evidence for both direct and indirect, TNF-a-mediated, b-chemokine production. At the early time points after CD40LT stimulation (up to 3 h), the antibody to TNF-a did not block production of b-chemokines, whereas a partial block was observed at later time points (18 h after CD40LTstimulation). Furthermore, the antibody to TNF-a did not revert CD40LT-induced CCR5 downregulation. These results suggest that CD40LT-induced b-chemokine production is biphasic. At early time points CD40LT acts directly to induce b-chemokine synthesis, while at later time points, when TNF-a secretion reaches a threshold level, this cytokine becomes the major driving force behind b-chemokine production. The biphasic nature of this response may be important within the context of host anti-HIV responses. The initial release of b-chemokines by MDM is rapid and peaks within 3 h after CD40LT stimulation. b-Chemokines released during this period may protect MDM and adjacent CD4+ T cells from being infected by HIV. The additional increased b-chemokine production detected at later time points may act to further enhance the host immune response by continuing to elicit inflammatory

and immune cells to the site of infection. b-Chemokines produced by CD40L-stimulated MDM may regulate the course of infection either by blocking specific co-receptor usage in uninfected cells or through their chemoattractive effect on the immune cells. CD40L-induced chemokine/cytokine production combined with the role of CD40L/CD40 in stimulating CTL activity in the absence of CD4+ T cells [36,37] may play an important role in vivo in anti-HIV immunity. While these b-chemokine activities are host protective, some reports suggested that these same mediators may contribute to the pathogenesis of HIV infection by recruiting uninfected immune cells and by enhancing HIV replication [30–32]. The biphasic bchemokine response to CD40L stimulation may be detrimental also in other pathological conditions, in particular autoimmune diseases, where increased production of cytokines/chemokines can exacerbate the development of the inflammatory process [38]. The present study evaluates the potential role of MAPK family members as downstream mediators of CD40 signaling in MDMs. Thus far, no conclusive studies dealing with CD40-mediated responses in macrophages have been reported. Most published

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

59

Fig. 5. Analysis of MAPK phosphorylation in response to CD40LT signaling. (A) After seven days of culture, MDM were either left unstimulated or stimulated with CD40LT (1 lg/ml) for 0.25, 0.5, 1, 3 and 18 h as indicated. After stimulation the cells were washed, lysed and analyzed on SDSPAGE followed by Western blotting using a polyclonal antibody directed against the dually phosphorylated ERK 1/2 (44 and 42 kDa), p38 (38 kDa), SAP/JNK1 (54 kDa) and SAP/JNK2 (46 kDa). To control for protein loading, antibodies recognizing the total ERK 1/2, p38 and JNK 1/2 were used. One representative experiment out of four is shown. (B) Dose response of CD40LT-induced ERK 1/2 and p38 phosphorylation in MDM. The cells were stimulated with different concentrations of CD40LT (ranging from 0.1 to 5 lg/ml) or left unstimulated for 30 min. After stimulation the cells were washed, lysed and analyzed on SDS-PAGE followed by Western blotting as described above. Band density is represented below Western immunoblots. Representative of two experiments.

60

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

Fig. 6. CD40-dependent b-chemokine secretion requires p38 and ERK 1/2 activity. MDM were preincubated for 1 h with PD98059 (10 lM), SB202190 (10 lM) or a combination of both (5 lM each inhibitors) prior to stimulation with CD40LT (1 lg/ml) or medium alone. After 3 and 18 h supernatants were collected and analyzed for b-chemokine and TNF-a secretion by ELISA. Representative of four independent experiments.

analyses focused on monocytes, which are undifferentiated cells with poorly defined physiological activity. Thus, it had been shown that ERK 1/2 kinases mediate signaling by CD40 in primary human peripheral blood monocytes and production of IL-1b and TNF-a [23]. Consistent with this finding, Pearson and colleagues demonstrated that the primary CD40-dependent MAPK

pathway activated in monocytes and THP1 promonocytic cell line was the ERK 1/2 pathway [22]. Furthermore, they showed that the CD40-dependent secretion of pro-inflammatory cytokines, IL-6, IL-8 and TNF-a was also mediated by ERK 1/2. Our analysis demonstrated that CD40 signaling in MDM involves also p38 MAPK. Both ERK 1/2 and p38 were rapidly phosphorylated

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

after CD40 cross-linking, whereas over the same time period phosphorylation of JNK, another member of the MAPK family, was not increased over background. This observation appears to contradict recently published results of Suttles et al. [23] who reported that CD40 activates ERK 1/2, but not p38, in primary monocytes. Furthermore, Hacker’s group demonstrated that Tolllike receptor (TLR)/interleukin 1 receptor (IL-1R) signaling pathway activated tumor necrosis factor receptor-associated factor 6 (TRAF6) leading to activation of the c-jun NH2 terminal kinase in peritoneal macrophages [39]. The discrepancy between our results and those of Sutter’s and Hacker’s group may reflect the fact that CD40 signaling in monocytes is different from that in fully differentiated macrophages and peritoneal macrophages. This would be consistent with a large body of work demonstrating that these two cell types differ substantially in many functional activities, including signaling responses to stimuli [40,41]. Furthermore, we have shown that CD40-dependent secretion of MIP-1a, MIP-1b and RANTES is linked to activation of both ERK 1/2 and p38. Specific inhibitors of p38 and MEK 1/2 activity (compounds SB202190 and PD98059, respectively) partially reduced b-chemokine production in response to CD40LT over a 3 h period, and completely blocked it after 24 h. Therefore, the capacity of CD40LT to induce b-chemokines in MDM is partially mediated by ERK 1/2- and p38 MAPK-dependent signaling pathways during the early response of MDM to CD40LT stimulation, and fully dependent during the late response. Given that the early b-chemokine production is TNF-a-independent and most likely reflects direct activation of b-chemokine gene transcription by CD40-dependent signaling (see above), it would be important to determine additional signaling pathways contributing to b-chemokine activation. The total dependence of b-chemokine production on ERK 1/2 and p38 in the late phase of response might be explained by the principle role of these MAPKs either in CD40-induced TNF-a production or in TNF-adependent signaling. Taken together, studies presented in this report identify important components of the signal transduction pathway involved in b-chemokine production by macrophages stimulated through the CD40 receptor. These results will provide the basis for future analysis of anti-HIV activity of macrophages and understanding their role in innate immunity. In addition, the data shown here are important for understanding pathogenesis of autoimmune diseases, where the disregulated production of monocyte/macrophage-derived inflammatory cytokines has been well documented [38]. For example, it has been shown that blockade of inflammatory cytokine signaling can decrease symptoms of rheumatoid arthritis [42]. The ability to pharmacologically disrupt upstream events in the CD40 pathway

61

would provide a therapeutic advantage in treatment of such diseases.

4. Materials and methods 4.1. Isolation and culture of primary human macrophages Peripheral blood mononuclear cells from healthy donors undergoing leukopheresis were separated on a Ficoll-Hypaque (Amersham Pharmacia Biotech Inc., Piscataway, NJ) gradient. Monocytes were allowed to adhere to plastic for 2 h at 37  C in RPMI 1640 supplemented with 10% heat-inactivated human serum (Biowhittaker, Walkersville, MD). Adherent cells were extensively washed and maintained for additional 24 h in medium supplemented with macrophage-colony stimulating factor (M-CSF; 2 ng/ml). Adherent monocytes were washed, removed from the flask by gentle scraping, seeded onto 24-well plates at a density of 1
106 cells/well, and cultured for seven days in the presence of M-CSF. After seven days, the cells were 99.5–99.8% macrophages as determined by cytochemical staining for non-specific esterase (a-Naphthyl Acetate Esterase Kit, Sigma, St Louis, MO). These cells are referred to as monocyte-derived macrophages, MDM. All cell culture reagents were endotoxin free.

4.2. Chemokine and cytokine assays Cell culture supernatants were obtained from MDM stimulated with CD40LT (1 lg/ml) (Immunex Corp., Seattle, WA) at different time points. Supernatant samples were collected, centrifuged for 5 min at 13,000
g and stored at 80  C. Concentrations of MIP-1a, MIP-1b, RANTES and TNF-a were measured using commercial ELISA Kits (R&D Systems, Minneapolis, MN). In some assays, MDM were pretreated for 1 h with p38 or ERK1/2 inhibitors, SB202190, 4-(4fluorophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole; or PD98059, 2-(29 amino-3-methoxyphenyl) oxanaphthalen-4-one (Calbiochem, San Diego, CA), respectively, at 10 lM each prior to the addition of CD40LT. No cytotoxicity was observed at this concentration as judged by MTT and Trypan blue exclusion assays. The chemokine and cytokine ELISA limits of detection were: 1.5 pg/ml for MIP-1a, 4 pg/ml for MIP1b, 8 pg/ml for RANTES and 4.5 pg/ml for TNF-a.

4.3. Radiolabeled chemokine binding assay For chemokine binding assay, MDM were exposed to different treatments and then washed before carrying

62

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

out binding experiments. For each sample, 0.5
106 cells were incubated with 0.5 nM of [125I]MIP-1a (NEN Life Science Products, Boston, MA) for 2 h at 4  C in a binding buffer comprising D-MEM and 1% BSA. Cells were washed twice with cold PBS, lysed in RIPA buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 2 mM EDTA, 1% Nonidet P-40, 0.5% deoxycholic acid and 1% SDS) and then counted in a gamma counter (Gamma Track 1193; TM Analytic, Brandon, FL) [43]. Non-specific background binding was determined by counting radioactivity bound in the presence of a 100fold molar excess of unlabeled MIP-1a. 4.4. Immunofluorescence assay MDM (1
106 cells/well) were washed with PBS and incubated with washing buffer (PBS 1
, 0.1% sodium azide, 20% human serum) at room temperature for 20 min. The cells were then incubated for 30 min at room temperature with the appropriate dilution of anti-CCR5 (clone 2D7) and an isoptype-matched IgG control Mab (to evaluate the level of non-specific binding) labeled with phycoerythrin (Becton Dickinson, Mountain View, CA). After three washes with cold washing buffer, the cells were fixed in PBS containing 4% formaldehyde and gently scraped. The cells were analyzed by fluorescenceactivated cell sorting (FACS) on a Becton Dickinson FACSort. For each determination 10,000 cells were analyzed. Fluorescence was analyzed using the CellQuest software (Becton Dickinson). 4.5. RT-PCR analysis of chemokine mRNA Total RNA was prepared using RNeasy (Qiagen Inc., Valencia, CA), treated with RNAse-free DNAse (Roche Molecular Biochemicals, Indianapolis, IN) and used as template in RT-PCR. RNA was reverse transcribed in a 20 ll reaction containing 0.1 lg of total RNA, 0.1 lg of oligo(dT), 200 U of Superscript reverse transcriptase (GIBCO-BRL, Rockville, MA) and 0.2 lM each of dATP, dCTP, dGTP and dTTP. After a 1 h incubation at 37  C, 1 : 10, 1 : 100 and 1 : 1000 dilutions of the cDNA product were prepared and amplified as described previously [44]. To control for contamination with genomic DNA, parallel amplifications were performed in the absence of reverse transcriptase. These were uniformly negative. The PCR products were separated by electrophoresis through 1.5% agarose gel. The sequence of the GAPDH (gyceraldehyde 3-phosphate-dehydrogenase) has been previously described [44]. The sequences for the primers used to amplify bchemokine are as follows: MIP-1a–b consensus senseGTC TGT GCT GAT CCC AGT GA MIP-1a, antisense-TTG TCA CCA GAC GCG GTG TG; MIP-1b, antisense-GGA CAC TTA TCC TTT GGC TA and RANTES, sense-CCG CGG CAG CCC TCG

CTG TCA TCC, antisense-CAT CTC CAA AGA GTT GAT GTA CTC C. 4.6. Immunoblot analysis MDM (1
106 cells/ml) were treated with appropriate stimuli and lysed and proteins from cell lysates (5 lg) were separated by SDS-PAGE, as described previously [45]. Proteins were transferred to a PDVF membrane (Hybond-P; Amersham Pharmacia Biotech Inc., Piscataway, NJ) and blocked for 1 h at room temperature in PBS/10% non-fat dry milk. Membranes were washed and probed with the appropriate antibody diluted 1 : 10,000 in PBS/1% non-fat dry milk for 1 h at room temperature (Phospho Plus Antibody Kit, New England Biolabs, Beverly, MA). Specific bands were revealed by incubating membranes for 1 h at room temperature with horseradish peroxidase-labeled goat anti-rabbit immunoglobulin diluted 1 : 5000, followed by enhanced chemiluminescence (ECL kit; Amersham Pharmacia Biotech Inc., Piscataway, NJ). Densitometry analysis was performed for all immunoblots using GS-700 Imaging Densitometer with Quantity One software (Bio-Rad, Hercules, CA). Acknowledgements We thank Carol A. Amella for technical assistance, R. Mariani and all the members of the Bukrinsky’s laboratory for helpful advice and support. This work was supported by funds from The Picower Institute for Medical Research (to PDM) and NIHR01 AI29110-12 (to BS). References [1] Grewal IS, Flavell RA. The role of CD40 ligand in costimulation and T-cell activation. Immunol Rev 1996;153:85–106. [2] Clark LB, Foy TM, Noelle RJ. CD40 and its ligand. Adv Immunol 1996;63:43–78. [3] Tan J, Town T, Mori T, Obregon D, Wu Y, DelleDonne A, et al. CD40 is expressed and functional on neuronal cells. EMBO J 2002;21:643–52. [4] Castle BE, Kishimoto K, Stearns C, Brown ML, Kehry MR. Regulation of expression of the ligand for CD40 on T helper lymphocytes. J Immunol 1993;151:1777–88. [5] Gauchat JF, Henchoz S, Mazzei G, Aubry JP, Brunner T, Blasey H, et al. Induction of human IgE synthesis in B cells by mast cells and basophils. Nature 1993;365:340–3. [6] Henn V, Slupsky JR, Grafe M, Anagnostopoulos I, Forster R, Muller-Berghaus G, et al. CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 1998; 391:591–4. [7] Banchereau J, Bazan F, Blanchard D, Briere F, Galizzi JP, van Kooten C, et al. The CD40 antigen and its ligand. Annu Rev Immunol 1994;12:881–922. [8] Kiener PA, Moran-Davis P, Rankin BM, Wahl AF, Aruffo A, Hollenbaugh D. Stimulation of CD40 with purified soluble gp39

P. Di Marzio et al. / Cytokine 22 (2003) 53–63

[9]

[10]

[11] [12]

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

induces proinflammatory responses in human monocytes. J Immunol 1995;155:4917–25. Malik N, Greenfield BW, Wahl AF, Kiener PA. Activation of human monocytes through CD40 induces matrix metalloproteinases. J Immunol 1996;156:3952–60. Tian L, Noelle RJ, Lawrence DA. Activated T cells enhance nitric oxide production by murine splenic macrophages through gp39 and LFA-1. Eur J Immunol 1995;25:306–9. Stout RD, Suttles J. The many roles of CD40 in cell-mediated inflammatory responses. Immunol Today 1996;17:487–92. Cotter RL, Zheng J, Che M, Niemann D, Liu Y, He J, et al. Regulation of human immunodeficiency virus type 1 infection, beta-chemokine production, and CCR5 expression in CD40Lstimulated macrophages: immune control of viral entry. J Virol 2001;75:4308–20. Di Marzio P, Mariani R, Lui R, Thomas EK, Landau NR. Soluble CD40 ligand induces beta-chemokine production by macrophages and resistance to HIV-1 entry. Cytokine 2000;12:1489–95. Kornbluth RS, Kee K, Richman DD. CD40 ligand (CD154) stimulation of macrophages to produce HIV-1-suppressive betachemokines. Proc Natl Acad Sci U S A 1998;95:5205–10. Cocchi F, DeVico AL, Garzino-Demo A, Arya SK, Gallo RC, Lusso P. Identification of RANTES, MIP-1 alpha, and MIP-1 beta as the major HIV-suppressive factors produced by CD8 þ T cells. Science 1995;270:1811–5. Deng H, Liu R, Ellmeier W, Choe S, Unutmaz D, Burkhart M, et al. Identification of a major co-receptor for primary isolates of HIV-1. Nature 1996;381:661–6. Dragic T, Litwin V, Allaway GP, Martin SR, Huang Y, Nagashima KA, et al. HIV-1 entry into CD4+ cells is mediated by the chemokine receptor CC-CKR-5. Nature 1996;381: 667–73. Garzino-Demo A, DeVico AL, Conant KE, Gallo RC. The role of chemokines in human immunodeficiency virus infection. Immunol Rev 2000;177:79–87. Loetscher P, Moser B, Baggiolini M. Chemokines and their receptors in lymphocyte traffic and HIV infection. Adv Immunol 2000;74:127–80. Poe JC, Wagner DHJ, Miller RW, Stout RD, Suttles J. IL-4 and IL-10 modulation of CD40-mediated signaling of monocyte IL1beta synthesis and rescue from apoptosis. J Immunol 1997; 159:846–52. Suttles J, Evans M, Miller RW, Poe JC, Stout RD, Wahl LM. T cell rescue of monocytes from apoptosis: role of the CD40– CD40L interaction and requirement for CD40-mediated induction of protein tyrosine kinase activity. J Leukoc Biol 1996;60: 651–7. Pearson LL, Castle BE, Kehry MR. CD40-mediated signaling in monocytic cells: up-regulation of tumor necrosis factor receptorssociated factor mRNAs and activation of mitogen-activated protein kinase signaling pathways. Int Immunol 2001;13:273–83. Suttles J, Milhorn DM, Miller RW, Poe JC, Wahl LM, Stout RD. CD40 signaling of monocyte inflammatory cytokine synthesis through an ERK1/2-dependent pathway. A target of interleukin (IL)-4 and IL-10 anti-inflammatory action. J Biol Chem 1999; 274:5835–42. Alderson MR, Armitage RJ, Tough TW, Strockbine L, Fanslow WC, Spriggs MK. CD40 expression by human monocytes: regulation by cytokines and activation of monocytes by the ligand for CD40. J Exp Med 1993;178:669–74. Ho DD, Rota TR, Hirsch MS. Infection of monocyte/macrophages by human T lymphotropic virus type III. J Clin Invest 1986;77:1712–5. Meltzer MS, Gendelman HE. Mononuclear phagocytes as targets, tissue reservoirs, and immunoregulatory cells in human immunodeficiency virus disease. Curr Top Microbiol Immunol 1992; 181:239–63.

63

[27] Salahuddin SZ, Rose RM, Groopman JE, Markham PD, Gallo RC. Human T lymphotropic virus type III infection of human alveolar macrophages. Blood 1986;68:281–4. [28] Kinter A, Arthos J, Cicala C, Fauci AS. Chemokines, cytokines and HIV: a complex network of interactions that influence HIV pathogenesis. Immunol Rev 2000;177:88–98. [29] Poli G, Biswas P, Fauci AS. Interferons in the pathogenesis and treatment of human immunodeficiency virus infection. Antiviral Res 1994;24:221–33. [30] Chang TL, Gordon CJ, Roscic-Mrkic B, Power C, Proudfoot AE, Moore JP, et al. Interaction of the CC-chemokine RANTES with glycosaminoglycans activates a p44/p42 mitogen-activated protein kinase-dependent signaling pathway and enhances human immunodeficiency virus type 1 infectivity. J Virol 2002;76:2245–54. [31] Kelly MD, Naif HM, Adams SL, Cunningham AL, Lloyd AR. Dichotomous effects of beta-chemokines on HIV replication in monocytes and monocyte-derived macrophages. J Immunol 1998; 160:3091–5. [32] Schmidtmayerova H, Sherry B, Bukrinsky M. Chemokines and HIV replication. Nature 1996;382:767. [33] Han J, Beutler B. The essential role of the UA-rich sequence in endotoxin-induced cachectin/TNF synthesis. Eur Cytokine Netw 1990;1:71–5. [34] Bailer RT, Lee B, Montaner LJ. IL-13 and TNF-alpha inhibit dual-tropic HIV-1 in primary macrophages by reduction of surface expression of CD4, chemokine receptors CCR5, CXCR4 and postentry viral gene expression. Eur J Immunol 2000;30:1340–9. [35] Lane BR, Markovitz DM, Woodford NL, Rochford R, Strieter RM, Coffey MJ. TNF-alpha inhibits HIV-1 replication in peripheral blood monocytes and alveolar macrophages by inducing the production of RANTES and decreasing CC chemokine receptor 5 (CCR5) expression. J Immunol 1999;163:3653–61. [36] Diehl L, den Boer AT, Schoenberger SP, van d V, Schumacher TN, Melief CJ, et al. CD40 activation in vivo overcomes peptideinduced peripheral cytotoxic T-lymphocyte tolerance and augments anti-tumor vaccine efficacy. Nat Med 1999;5:774–9. [37] Schoenberger SP, Toes RE, van d V, Offringa R, Melief CJ. T-cell help for cytotoxic T lymphocytes is mediated by CD40–CD40L interactions. Nature 1998;393:480–3. [38] Sacca R, Cuff CA, Ruddle NH. Mediators of inflammation. Curr Opin Immunol 1997;9:851–7. [39] Hacker H, Vabulas RM, Takeuchi O, Hoshino K, Akira S, Wagner H. Immune cell activation by bacterial CpG-DNA through myeloid differentiation marker 88 and tumor necrosis factor receptor-associated factor (TRAF) 6. J Exp Med 2000; 192:595–600. [40] Chow CW, Demaurex N, Grinstein S. Ion transport and the function of phagocytic cells. Curr Opin Hematol 1995;2:89–95. [41] Hart PH, Bonder CS, Balogh J, Dickensheets HL, Donnelly RP, Finlay-Jones JJ. Differential responses of human monocytes and macrophages to IL-4 and IL-13. J Leukoc Biol 1999;66:575–8. [42] Moreland LW, Baumgartner SW, Schiff MH, Tindall EA, Fleischmann RM, Weaver AL, et al. Treatment of rheumatoid arthritis with a recombinant human tumor necrosis factor receptor (p75)-Fc fusion protein. N Engl J Med 1997;337:141–7. [43] Franchin G, Zybarth G, Dai WW, Dubrovsky L, Reiling N, Schmidtmayerova H, et al. Lipopolysaccharide inhibits HIV-1 infection of monocyte-derived macrophages through direct and sustained down-regulation of CC chemokine receptor 5. J Immunol 2000;164:2592–601. [44] Di Marzio P, Tse J, Landau NR. Chemokine receptor regulation and HIV type 1 tropism in monocyte-macrophages. AIDS Res Hum Retroviruses 1998;14:129–38. [45] Di Marzio P, Choe S, Ebright M, Knoblauch R, Landau NR. Mutational analysis of cell cycle arrest, nuclear localization and virion packaging of human immunodeficiency virus type 1 Vpr. J Virol 1995;69:7909–16.