Discordant effects of interleukin-2 on viral and immune parameters in human immunodeficiency virus-1-infected monocyte-derived mature dendritic cells

Abstract

Use of interleukin-2 (IL-2) in the immunotherapy of human immunodeficiency virus (HIV) has frequently resulted in the restoration of CD4 lymphocyte counts but not of virus-specific responses. We reasoned that the absence of reconstituted functional immune parameters could be related to the inability of IL-2 to correct HIV-induced dysfunctions in antigen-presenting cells. In this study, we used in vitro-differentiated monocyte-derived macrophages (MDMs) and mature dendritic cells (MDDCs), acutely infected with primary HIV-1 isolates, to analyse the effects of IL-2 on virus replication, co-receptor expression, and cytokine or chemokine release. Stimulation of MDMs with IL-2 had no measurable effect on HIV-1 replication, on cytokine secretion, or on CD4 and CXCR4 gene expression. Moreover, although a significant down-regulation of CCR5 mRNA expression could be repeatedly detected in MDMs, this IL-2-mediated effect was not of substantial magnitude to affect virus replication. On the other hand, IL-2 stimulation of MDDCs dramatically increased HIV-1 replication and this effect was highly evident on low-replicating, CXCR4-dependent isolates. Nevertheless, the HIV-enhancing activity of IL-2 in MDDCs was not accompanied by any measurable change in cytokine or chemokine release, in virus receptor and co-receptor mRNA accumulation, or in the surface expression of a battery of receptors implicated in virus entry, cell activation or costimulatory function. Taken together, these findings point to a role for IL-2 in inducing virus purging from dendritic cell reservoirs but indicate no relevant potential of the cytokine in restoring defective elements of innate immunity in HIV infection.

INTRODUCTION

Use of highly active antiretroviral therapy (HAART) has considerably improved the management of patients with human immunodeficiency virus-1 (HIV-1) and can frequently result in long-term virus control. However, it has become evident that HAART alone is not efficient in eliminating virus reservoirs and does not always lead to a substantial immune reconstitution [1–3]. Several immunotherapeutic approaches have been proposed to enhance the restoration of CD4 lymphocyte numbers and function [4,5]. Among these approaches, interleukin (IL)-2 has been extensively evaluated in patients with HIV-1 and has repeatedly demonstrated efficacy in inducing a significant rise in CD4-cell counts [4–7]. Nevertheless, the capacity of IL-2 to restore HIV-specific responses is not quite evident [6,7], and a correlation between the IL-2-induced increase in CD4-cell counts and measurable clinical benefit has yet to be demonstrated.

Besides the well-documented effects of IL-2 on lymphocytes and natural killer (NK) cells, other cell populations of the innate immune system, including monocytes and dendritic cells, are known to express functional IL-2 receptors and to respond to the cytokine signal [8–12]. Although monocyte activation by IL-2 is known to result in the up-regulation of proinflammatory cytokine release and in the phosphorylation of selected kinases, the responses of differentiated macrophages to the cytokine are less obvious [13–16]. Moreover, the ability of IL-2 to regulate the function of dendritic cells has been a matter of controversy. Thus, whereas mouse dendritic cells were reported to respond to IL-2 by increased IL-12-dependent interferon-γ (IFN-γ) production [11] and by enhanced CD25-dependent antigen uptake [17], the inducible IL-2 receptor on dendritic cells, using CD25-null mice, was found to be unnecessary for dendritic cell development, for T-cell stimulation, and for the regulation of lymphocyte responses [18]. On the other hand, mouse dendritic cells, but not macrophages, were recently shown to secrete IL-2 in response to Gram-negative bacteria, and this response is believed to confer unique T-cell stimulatory capacity to dendritic cells [19].

Studies addressing the effects of IL-2 on HIV-1-infected CD4 lymphocytes have clearly established a role for the cytokine in up-regulating virus replication and in increasing the expression of HIV-1 co-receptors CCR5 and CXCR4 [20–22]. Moreover, treatment of HIV-1 patients with IL-2 was found to inhibit the expression and activation of transcription factors known to repress the virus long-terminal repeat [23]. These findings could explain the rise in plasma viraemia among patients receiving IL-2 without HAART. However, neither the immune effects of IL-2 on human antigen-presenting cells (APCs), nor the ability of IL-2 to regulate HIV-1 replication in these virus reservoirs, have been investigated in detail. Therefore, in this study, we used monocyte-derived macrophages (MDMs) and mature dendritic cells (MDDCs), acutely infected with primary HIV-1 isolates, to characterize the regulatory effects of IL-2 on viral replication. Furthermore, the capacity of IL-2 to induce the release of a battery of cytokines and chemokines, in MDM and MDDC cultures, and to modulate expression of the relevant cell receptors, were examined. Our results demonstrated that despite a general absence of measurable immune activation of APCs by IL-2, the presence of the cytokine in infected MDDC cultures leads to potent up-regulation of viral replication. This effect could not be observed in MDM cultures, was independent of virus tropism, and could not be linked either to the induction of HIV-1-enhancing cytokines or to the inhibition of HIV-suppressive β-chemokines.

MATERIALS AND METHODS

Reagents and antibodies

Recombinant human IL-2, IL-4, tumour necrosis factor-α (TNF-α) and granulocyte–macrophage colony-stimulating factor (GM–CSF) were purchased from R & D Systems (Abingdon, UK). All reagents were verified for the absence of endotoxin contaminant (<6 × 10⁻² U/ml) by using the Limulus amebocyte lysate assay (Biowhitakker France, Fontenay-sous-Bois, France). Anti-CD4-conjugated phycoerythrin (PE) (13B8.2), anti-CD14-conjugated PE (RMO52), anti-CD3-conjugated fluorescein isothiocyanate (FITC) (UCHT1), anti-human leucocyte antigen-DR (HLA-DR)-conjugated PE (B.8.12.2), anti-CD25-conjugated PE (B1.49.9), and anti-CD83-conjugated PE (HB15a) monoclonal antibodies (MoAbs) were purchased from Immunotech (Beckman Coulter, Marseille, France). Anti-CD86-conjugated PE (B70/B72), anti-CD80-conjugated PE (L307.4), anti-CD122-conjugated PE (Mik-β2), anti-CCR5-conjugated PE (2D7), anti-CXCR4-conjugated PE (12G5), and their isotype-matched controls, were obtained from Becton-Dickinson Pharmingen (Rungis, France).

Recovery of primary HIV-1 isolates and analysis of co-receptor specificity

The recovery of HIV-1 isolates from peripheral blood mononuclear cells (PBMCs) of infected subjects was performed in our laboratory, as described previously [24,25]. Three isolates with different cell tropism were used in this study, and the co-receptor usage of each isolate was determined using the GHOST cell infectivity assay, as reported previously [25,26]. The isolates were CHR4, CHR1 and D2, presenting, respectively, macrophage (M), dual and T-cell line (T) tropism. Although CHR1 is dual-tropic and could use either CCR5 (R5) or CXCR4 (X4) receptors in the GHOST cell infectivity assay, the ability of this isolate to infect primary APCs was found to be X4 dependent, as determined by the capacity of stromal-cell derived factor-1α– but not a mixture of macrophage inflammatory protein (MIP)-1α, MIP-1β and regulated on activation, normal, T-cell expressed, and secreted (RANTES) – to inhibit virus entry and replication [24]. In addition to these three primary isolates, certain infection assays were also performed using the laboratory-adapted M-tropic HIV-1 Bal, or two other primary isolates – HIV-1/0572 (M-tropic) and HIV-1 D4 (dual-tropic).

Preparation of MDMs and MDDCs

MDMs and MDDCs were generated during a 7-day culture period from adherent monocytes purified from PBMCs of healthy donors, as described previously [24,27,28]. Briefly, MDMs were generated following culture in RPMI-1640 containing 10% human AB serum (Etablissement de Transfusion Sanguine, Lille, France). The same medium, supplemented with recombinant human GM–CSF, IL-4 and TNF-α, was used to differentiate monocytes into MDDCs. At the end of the differentiation period,> 90% of the MDMs were CD14⁺ and MDDCs were found to represent mature dendritic cells, as judged by morphologic (adherent cells with fine membrane projections) and phenotypic (CD14⁻, CD3⁻, high levels of CD80 and CD86,> 40% CD83⁺ and > 60% CD4⁺) criteria. The use of TNF-α to drive the differentiation of monocytes into mature dendritic cells has been repeatedly associated with maximal surface expression of certain costimulatory molecules and with adherence to tissue-culture plates [24,28]. It should be emphasized that the absence of CD3⁺ lymphocyte contaminants (<1%) was routinely verified prior to the use of MDMs and MDDCs in the infection assays.

HIV-1 infection, in vitro, and evaluation of viral replication

Cultures were set up in 24-well plates (Falcon, Le Pont de Claix, France) at 5 × 10⁵ cells/ml, and were then acutely infected by a 2-h exposure to HIV-1 at a dose corresponding to 10 000 counts per minute (c.p.m.) of reverse transcriptase activity/10⁶ cells [28]. Immediately after infection and throughout the whole culture period, triplicate wells were maintained in the presence or absence of 100 U/ml of recombinant human IL-2 (R & D Systems), unless indicated otherwise. Viral replication was evaluated by measuring the levels of p24 protein in culture supernatants [25,28]. For the calculation of means, p24 levels below the detection limit of the assay (7·5 pg/ml) were given arbitrarily a value of 10 pg/ml.

Reverse transcription–polymerase chain reaction (RT–PCR) detection of virus receptor expression

DNase-treated RNA extracts (4–100 ng) from MDMs and MDDCs were amplified using rTth polymerase (Perkin-Elmer, Emeryville, CA), in the presence of specific primer pairs, to detect CD4, CCR5, CXCR4 and the housekeeping gene, β-actin, as internal control. The primers used and the RT–PCR protocol were as described previously [27,29]. The PCR products were electrophoresed through a 1·8% agarose gel and stained with ethidium bromide for ultraviolet visualization. Imaging systems (Image Master ID prime; Amersham Pharmacia Biotech, Orsay, France) were used to quantify the level of mRNA accumulation of each receptor after normalization to that of β-actin [27,28]. This semiquantitative RT–PCR analysis was performed by calculating the ratio of the receptor RNA band volume over that of β-actin, in the linear phase of the amplification [27]. Results of the expression level in treated samples are presented as per cent of the mRNA expression level in untreated cultures, which was assumed to be 100%.

Levels of secreted cytokines

Quantification in culture supernatants of the levels of TNF-α, IL-6, IL-10, IFN-γ, MIP-1α, MIP-1β and RANTES was performed by using commercially available enzyme-linked immunosorbent assay (ELISA) kits (R & D Systems), according to the manufacturer's instructions.

Flow cytometry analysis

To assess surface-receptor expression on MDMs and MDDCs, 2 × 10⁵ cells were incubated for 30 min at 4°C with specific MoAbs, or with isotype-matched controls, diluted 1 : 100 in phosphate-buffered saline (PBS) containing 2% heat-inactivated fetal calf serum (FCS) (Sigma, Saint Quentin Fallavier, France). Following two washes in PBS, cells were resuspended, fixed in 1% paraformaldehyde and analysed using a flow cytometer (FACSCalibur; Becton-Dickinson, Rungis, France). Live cells were gated on their forward and side light scatter characteristics, and the percentage of positive cells and the mean fluorescence intensity (MFI) were recorded. In some experiments, exclusion of dead cells was verified by propidium iodide staining.

Statistical analysis

All results were expressed as mean values ± standard error of the mean (s.e.m.). The Student's t-test for paired data or the non-parametric Wilcoxon matched-pairs test were used to determine the statistical significance of all reported results. P-values of <0·05 were considered to be statistically significant.

RESULTS

Regulation of HIV-1 replication by IL-2

To determine the effect of stimulating HIV-1-infected MDMs and MDDCs with IL-2 on the level of viral replication, cultures were maintained for 12–14 days postinfection in the presence or absence of 100 U/ml of the cytokine. Viral p24 levels were measured in culture supernatants and results from three independent experiments, using cells generated from three different donors, are shown in Fig. 1. Infection of MDMs with the M-tropic CHR4 isolate resulted in easily detectable p24 levels (3765 ± 1673 pg/ml), and viral replication was not significantly modified in parallel cultures that were maintained in the presence of IL-2 (5279 ± 3548 pg/ml; P > 0·05). In contrast, IL-2-stimulated MDDCs were observed to produce a 3·7–6·1-fold increase in p24 levels as compared with unstimulated cultures (Fig. 1a), and this increase was of statistical significance (P < 0·05). Similarly, the dual-tropic CHR1 isolate replicated equally well in MDM cultures that were stimulated or unstimulated with IL-2 (Fig. 1b). However, although CHR1 presented a low replication level in primary MDDCs, the presence of IL-2 in the infected cultures dramatically up-regulated the replication of this HIV-1 isolate (Fig. 1b). On the other hand, no detectable p24 release could be observed in any MDM culture infected with the D2 strain, thereby confirming the T-tropic nature of this isolate (Fig. 1c). Nevertheless, although the replication of HIV-1D2 in unstimulated MDDCs was weak or below the detection limit, stimulation of MDDCs with IL-2 induced a very high replication level in cultures from the three tested donors (Fig. 1c). Thus, it seems evident that IL-2 has no stimulatory effect on HIV-1 replication in MDMs but could potently drive the replication of M-, dual- and T-tropic isolates in MDDCs. These results were confirmed in additional experiments using MDMs and MDDCs that were infected with the laboratory-adapted strain HIV-1 Bal or with other M-tropic (HIV-1/0572) and dual-tropic (HIV-1D4) primary isolates (data not shown).

Fig. 1.

Effects of interleukin-2 (IL-2) on human immunodeficiency virus-1 (HIV-1) replication in monocyte-derived macrophages (MDMs) and mature dendritic cells (MDDCs). Following a 7-day differentiation period, MDMs and MDDCs obtained from three different donors were acutely infected with primary HIV-1 isolates and were then maintained for 12–14 days in the presence or absence of IL-2 (100 U/ml). Levels of virus p24 protein were evaluated in the supernatants of cultures infected with M-tropic CHR4 (a), dual-tropic CHR1 (b), and T-tropic D2 (c) strains. Different symbols reflect experiments on cells from different donors and horizontal bars represent the means.

Effect of IL-2 on cytokine and chemokine release

To determine whether IL-2 could regulate the release of cytokines and chemokines in HIV-1 Bal-infected MDMs and MDDCs, supernatants were collected 24–72 h following stimulation and assayed for the content of different cytokines and chemokines. Results from five and four independent experiments, respectively, using MDM and MDDC cultures, are shown in Fig. 2 and reflect the detectable levels after a 24- or 48-h stimulation period. The levels of the HIV-suppressive β-chemokines MIP-1β, MIP-1α and RANTES, or of the HIV-enhancing cytokines IL-6, IL-10 and TNF-α, were identical in unstimulated and IL-2-stimulated MDMs (Fig. 2a) or MDDCs (Fig. 2b). This was the case at all time-points tested and the levels tended to decrease with time, reaching very low values after a 72-h culture period (data not shown). Note that the higher levels of TNF-α detected in MDDCs, as compared with MDMs, was related to the presence of exogenously added cytokine. On the other hand, no detectable levels of IFN-γ were observed in any of the tested supernatants from either MDM or MDDC cultures, regardless of whether or not they were stimulated with IL-2. These results indicate that the HIV-enhancing effect of IL-2 in MDDCs is independent of its capacity to regulate the release of cytokines or chemokines which could either drive or control viral replication. Moreover, to ensure that the lack of measurable cytokine and chemokine release following IL-2 stimulation was not related either to the concentration of IL-2 used or to HIV-1 infection, we examined, in five additional experiments, the profile of cytokine/chemokine release in uninfected MDMs and MDDCs that were stimulated with 100 or 500 U/ml of IL-2. Once again, the spontaneous release of MIP-1α, MIP-1β, RANTES, IL-6, IL-10 and TNF-α was not modified in any of the cultures treated with either concentration of IL-2 (data not shown). Therefore, the absence of IL-2-mediated effects on cytokine and chemokine release in HIV-1-infected MDMs and MDDCs could not be explained by the state of infection or by the dose selected for stimulation.

Fig. 2.

Levels of secreted chemokines and cytokines in human immunodeficiency virus-1 (HIV-1)-infected monocyte-derived macrophage (MDM) and mature dendritic cell (MDDC) cultures that were stimulated or unstimulated with 100 U/ml of interleukin (IL)-2. Supernatants were collected after 24 and 48 h of stimulation and evaluated for the content of macrophage inflammatory protein (MIP)-1β, MIP-1α, regulated on activation, normal, T-cell expressed, and secreted (RANTES), IL-6, IL-10 and tumour necrosis factor-α (TNF-α), using commercially available enzyme-linked immunosorbent assay (ELISA) kits. Results represent the mean values ± standard error of the mean (s.e.m.) of five independent experiments with MDMs (a) and of four experiments with MDDCs (b).

Modulation of receptor expression by IL-2

To determine the effects of IL-2 on the gene expression level of CD4, CCR5 and CXCR4 in the two populations of APCs, total RNA extracts, obtained 2 and 24 h after stimulation, were analysed by RT–PCR. Results from six independent experiments, shown in Fig. 3, revealed the absence of any significant effect of stimulating MDMS or MDDCs with IL-2 on the level of mRNA accumulation of either CD4 or CXCR4. However, whereas stimulation of MDMs with IL-2 significantly down-regulated CCR5 gene expression, at the two time-points tested, no enhancing or inhibitory effect on CCR5 gene expression by the cytokine could be demonstrated in MDDCs (Fig. 3). This suggests that the observed effects of IL-2 on viral replication in MDDCs could not be mediated by up-regulation of gene expression of the major virus receptor CD4, or of either of the two co-receptors CCR5 and CXCR4.

Fig. 3.

Effects of interleukin-2 (IL-2) on CD4, CCR5, and CXCR4 gene expression in monocyte-derived macrophages (MDMs) and mature dendritic cells (MDDCs). Cultures were incubated for 2- and 24 h, either unstimulated or stimulated with 100 U/ml IL-2. RNA extracts were then subjected to reverse transcription–polymerase chain reaction (RT–PCR) analysis to detect the levels of gene expression of CD4, CCR5, CXCR4, and β-actin (as a housekeeping gene). Changes in the expression levels were calculated as percentage of baseline expression in unstimulated cells taken as 100% (indicated by the dotted horizontal line). Results represent the mean values ± standard error of the mean (s.e.m.) of six independent experiments using cells from different donors. *P < 0·05 versus baseline expression in unstimulated cultures.

Because the absence of an effect at the level of gene expression of a receptor does not necessarily preclude a potential effect on cell-surface expression, we analysed (using flow cytometry) the ability of IL-2 to regulate, in MDDCs, the surface expression of CD4, CCR5 and CXCR4. In addition, we addressed the potential effects of IL-2 on the expression of a battery of receptors, including IL-2 receptor chains α (CD25) and β (CD122), activation markers and costimulatory molecules. Using MDDCs from four separate donors, stimulated or not stimulated with 100 and 500 U/ml of IL-2 for 6, 24 and 48-h, no significant modulation in the expression of any of the studied receptors was observed. A representative set of results, from one of the four tested donors, is shown in Fig. 4 to indicate the level of expression of various receptors following 24 h of culture in the presence or absence of IL-2 (500 U/ml). Similar results were noted when the expression of the costimulatory receptor, CD80, was tested (data not shown). These findings further confirm the lack of IL-2-mediated effects on the expression of virus receptors in MDDCs, and point to the absence of relevant phenotypic changes induced by the cytokine in this population of APCs. Moreover, the lack of detectable CD3⁺ cells, together with the identical percentages of CD4⁺ cells among unstimulated and IL-2-stimulated cultures (Fig. 4), clearly rule out potential contamination with CD4 lymphocytes to be at the origin of the observed IL-2 stimulatory effect on virus replication in MDDCs.

Fig. 4.

Expression of various receptors in mature dendritic cells (MDDCs) following 24 h of culture in the presence or absence of interleukin-2 (IL-2) (500 U/ml). Cells, from one representative donor, were subjected to fluorescence-activated cell sorter (FACS) analysis to detect the expression on the surface of virus receptors, CD3, IL-2 receptors, costimulatory molecules, and activation markers. Histograms illustrate the level of positive staining with a specific antibody in unstimulated (shaded) and IL-2-stimulated (unshaded) cells. Dotted unshaded histograms demonstrate staining with the isotype-matched antibody control.

DISCUSSION

IL-2 is a cytokine with pleiotropic regulatory effects on different leucocyte subpopulations and is considered to be the primary T-cell growth factor [30]. Moreover, IL-2 has been shown to enhance various T-cell responses and has been used in vivo as an immunotherapeutic agent for purposes such as immune reconstitution and tumour reduction [30]. These well-described activities of IL-2 have led to its evaluation in HIV-1-infected subjects regarding restoration of the numbers of CD4 lymphocytes, which are dramatically reduced following infection. Despite an impressive recovery of CD4-cell numbers in patients treated with IL-2, the adaptive immune responses to the virus could not be restored [6,7]. This may be explained, at least in part, by the inability of the cytokine to correct persistent dysfunctions in innate immune cells, and which are believed to be at the origin of the defective virus-specific responses [31,32]. Therefore, in this study, we addressed the effects of IL-2 on in vitro-generated APCs with the aim of characterizing a potential role of the cytokine in regulating HIV-1 replication in MDMs and MDDCs. Moreover, we determined, using certain immune parameters, the ability of human cells of the innate immune system to respond to IL-2.

Stimulation of HIV-1-infected MDMs with IL-2 had no measurable effect on viral replication. This was the case for both CCR5-dependent M-tropic and CXCR4-dependent dual-tropic strains. In addition, the absence of productive infection of MDMs with a primary T-tropic isolate was not modified in the presence of IL-2. These results are concordant with previous reports indicating the ability of CXCR4 – using primary dual-tropic but not T-tropic HIV-1 strains – to productively infect MDMs [33,34]. Furthermore, the absence of a measurable effect of IL-2 on M-tropic virus replication in MDMs has also been observed under similar conditions where the cytokine was added to cultures after the period of infection [35]. However, MDMs differentiated for 10 days in the presence of IL-2 were found to become less permissive to a subsequent infection with HIV-1, and this correlated with a reduced expression of CD4 and CCR5 in the IL-2-differentiated MDMs [35]. Although we have observed a significant down-regulation of CCR5 gene expression in MDMs, 2 and 24 h following IL-2 stimulation, no down-regulation of CD4 expression could be noted, and the effect on CCR5 expression was not of a sufficiently high magnitude to cause a significant change in viral replication. Thus, it could well be that although stimulation of MDMs with IL-2, postinfection, may not affect HIV-1 replication, the differentiation of monocytes into MDMs in the presence of IL-2 could result in cells with reduced expression of virus receptors, rendering them less susceptible to infection or to virus spread in culture. On the other hand, HIV-1-infected MDDCs that were maintained in the presence of IL-2 presented significantly higher viral replication as compared with unstimulated cells. This effect was observed independently of virus tropism, and tended to attain higher magnitude with isolates that weakly replicated in MDDCs. Importantly, HIV-1 strains that presented an undetectable p24 level in MDDCs (T-tropic and, to a lesser extent, dual-tropic strains) were found to achieve a high replication level in cultures maintained with IL-2. This suggests the ability of IL-2 to enhance HIV-1 replication in productively infected cells and to purge the virus in latently infected or in virus-low-replicating MDDCs. A similar effect of IL-2 on the induction of HIV-1 replication in latently infected CD4 T cells has been described, and this effect was found to become dramatically magnified in the presence of the inflammatory cytokines, IL-6 and TNF-α[36]. Thus, it is reasonable to explain the differences in the effects of IL-2 on HIV-1 replication among the two cell populations studied by the presence of exogenously added TNF-α and/or GM–CSF in MDDC, but not in MDM, cultures. These proinflammatory cytokines, together with low amounts of endogenously released IL-6, may be sufficient to sustain a low level of viral replication [37,38], and this appears to be considerably up-regulated in the presence of IL-2. Future studies, using IL-2-containing cocktails of cytokines to stimulate HIV-1-infected MDMs, may reveal the proinflammatory mediators or growth factors that could synergize with IL-2 in inducing virus replication in macrophages. Nevertheless, the possibility that IL-2 could selectively transduce signals in MDDCs, leading to the regulation of cellular factors implicated in viral replication, cannot be ruled out. This is also substantiated by the fact that IL-2 signalling pathways are multiple, cell dependent, and able to result in the activation or inhibition of several classes of messenger molecules and transcription factors that could regulate HIV-1 replication [39,40].

Analysis of cytokine and chemokine release in IL-2-stimulated MDMs or MDDCs revealed the absence of any measurable effect in both types of HIV-infected APCs. The spontaneously released levels of β-chemokines, IL-6, IL-10 and TNF-α were found to remain constant following IL-2 stimulation. This indicates that IL-2 signalling in MDMs and MDDCs does not lead to the regulated expression of certain cytokines or chemokines which could, respectively, enhance or suppress HIV replication [20,36,37,41]. These findings are in agreement with previously published data on the absence of IL-2-mediated induction of cytokines and chemokines in HIV-infected MDMs [35]. Moreover, the observed inability of IL-2 to induce cytokine release in uninfected cells suggests that differentiated human APCs, in contrast to undifferentiated monocytes [13], do not respond to IL-2 signalling by producing increased levels of secreted cytokines or chemokines. Such differences in the responsiveness to IL-2 between blood monocytes and differentiated macrophages have been previously observed, although the basis for this difference has not been revealed [16]. It could well be that, during differentiation, MDMs express a lower number of IL-2 receptors than monocytes [42] and, consequently, become less responsive to the cytokine signal. Alternatively, the reported differences between monocytes and macrophages in activating certain transcription factors could account, at least in part, for the altered responsiveness of MDMs to IL-2 [43].

It has previously been reported that stimulation of MDMs with IL-2 resulted in a down-regulation of CCR5 gene expression without any measurable effect on CXCR4 mRNA accumulation [27]. We have confirmed these findings and extended them to show the absence of a cytokine effect on the expression of the major virus receptor, CD4, in MDMs. The observed effect on CCR5, although attaining statistical significance, was not of a sufficient magnitude to allow substantial inhibition of virus infection and replication. On the other hand, stimulation of MDDCs with IL-2 presented no significant effect on the gene expression of CD4, CCR5 and CXCR4. This was also confirmed at the level of cell-surface expression using flow cytometry analysis. These findings strongly suggest that the HIV-enhancing effect of IL-2 in MDDCs could not be mediated by its capacity to up-regulate the expression of virus receptors. Importantly, stimulation (for up to 48 h) of MDDCs with the cytokine had no measurable effect on the level of surface expression of the IL-2 receptors CD25 and CD122, of activation markers (HLA-DR), and of costimulatory molecules (CD40, CD80 and CD86). Therefore, it becomes evident that IL-2 presents no capacity to regulate surface receptor expression in MDDCs and, in particular, those receptors that are critical for the antigen-presenting function of these cells.

In conclusion, we have presented data to show that in vitro-generated mature MDDCs are infectable with HIV-1 primary isolates which use either CCR5 or CXCR4 as co-receptors. Stimulation of infected MDDCs with IL-2 triggered a potent up-regulation of virus replication that was highly evident in cultures infected with low-level replicating strains. This effect of IL-2 was not accompanied by any significant modification of the capacity of MDDCs either to release cytokines and chemokines or to express a battery of receptors implicated in virus entry, cell activation and antigen presentation. The absence of measurable immune effects of IL-2 on both MDMs and MDDCs suggests that the cytokine would have a minimal role, if any, in correcting functional defects of APCs that are commonly induced following HIV-1 infection [31,32]. Moreover, the HIV-enhancing activity of IL-2 in MDDCs may lead to increased HIV dissemination, although such an effect could be useful for inducing virus purging from these reservoir cells. The implications of these findings would need to be addressed in clinical studies targeting the evaluation of immunotherapy in HIV disease.