ANTI-INFLAMMATORY CYTOKINES IN HIV INFECTION

Anti-inflammatory cytokines, like IL-4, IL-10, IL-13, IL-19, IL-20 and TGF-β appear later than inflammatory cytokines and increase the expression of decoy receptors, limit the production of pro-inflammatory cytokines and chemokines (mandatory for leukocyte recruitment) and cytotoxic processes.

IL-4 and IL-10 are the prototypic anti-inflammatory cytokines that downregulate tissue inflammation by switching off pro-inflammatory effects of Th1 cytokines such as IFN-α and TNF.

IL-4 is produced by CD4+ Th cells, mast cells [228] and macrophages upon recognition of extracellular pathogens [229]. Upon TLR-mediated activation of DC, naive T cells release IL-4 serving as growth factor for the expansion and activation of Th2 cells [230] secreting IL-4, IL-5 and IL-6 [231]. Th2 cytokines inhibit the development of Th1 cells and, consequently, cell-mediated immune responses [232]. On the other side, IL-4 has also been reported to induce over-expression of DC-SIGN on in vitro differentiated DC [233], thus boosting both the ability of DC to interact with pathogens and the above described mechanism.

Concerning HIV infection, IL-4 can up-regulate CXCR4 and down-regulate CCR5, thus potentially playing a role in the so called “phenotypic switch” from R5 to X4 HIV-1 strains, an event occurring at late stage of disease in approximately 50% of individuals infected with clade B HIV-1 [234-239]. In vitro, IL-4 has been reported to induce virus replication on freshly isolated monocytes but to inhibit in MDM, depending on its ability to inhibit the levels of pro-inflammatory cytokines in the two cellular systems [240].

IL-10 is secreted by DC and Th2 cells [241]. This cytokine shuts off T cell activation because of its ability to inhibit the production of pro-inflammatory cytokines and chemokines, as well as the expression of DC- costimulatory molecules [242].

IL-10 has been reported to upregulate in vitro CXCR4 expression and X4 HIV infection of DC [243]. However, this effect did not alter the efficiency of viral transmission to autologous CD4⁺ T cells, since this process involves DC-SIGN and not conventional viral receptors, such as CXCR4 [243]. In a similar way, IL- 10 has been founf to enhance CCR5 expression on freshly isolated monocytes and thus their susceptibility to R5 infection [234]. However, other reports have shown that IL-10 inhibits in vitro HIV infection in macrophages [244,245].

As reported for other cytokines, depending on the cellular models and experimental conditions (in particular levels of exogenously added IL-10) both inhibitory and inductive effects on HIV replication have been reported. Indeed, in MDM inhibition of viral replication was correlated to the prevention of the synthesis and release of endogenous pro-inflammatory cytokines, such as TNF-α and IL-6 [246], whereas lower concentrations of IL-10 induced HIV replication, an effect that has been correlated to the cooperation with the released TNF-α and IL-6, as demonstrated in the chronically infected U1 cells [247-249].

IL-10 levels has been found elevated in lymph nodes of HIV-infected individuals [55,250,251], although progression to AIDS is characterized by decreased expression an event that has been correlated with the expression of an IL-10 promoter variant [252]. In SCID mice reconstituted with human cells, such as hu-PBL- SCID and thy/liv-SCID-hu-mice, IL-10 levels have been correlated to the activation state of human cells

[253]. However, IL-10 has been shown to induce up-regulation of MHC-I on CD4+CD8+ human thymocytes

[254] and to inhibit IFN-γ production and HIV replication [244,255].

IL-13 is produced by mononuclear phagocytes and its decreased plasma levels in HIV+ individuals have been associated with the progression of disease [256]. In fact, effective antiretroviral therapy, such as HAART+IL- 2 increased IL-13 plasma levels [257]. Unfortunately, side-effect such as intestinal epithelial atrophy has been reported in SIV-infected macaques treated with IL-13 [258].

IL-13 has been reported to inhibit HIV replication via indirect mechanisms, such as downregulation of CD4 [259], CCR5 [238,259] and CXCR4 [238,256,259-261] in MDM or HIV expression by induction of an endogenous IL-1β∕IL-1 receptor antagonist balance in favor of the latter in chronically infected U1 cells stimulated with LPS and GM-CSF [262]. On the other hand, IL-13 did not inhibit HIV replication in activated T cells [48,263].

Finally, IL-13 has been reported to upregulated DC-SIGN expression on monocytes [264,265], although it did influence the ability of monocytes to transmit the virus T cells [264].

IL-19 and IL-20 belong to the anti-inflammatory cytokines of the IL-10 family, and are produced mostly by monocytes during inflammation [266]. They act predominantly on non immunological tissue and organs, such as the skin, lungs and various internal organs, including reproductive organs [266]. No information is available on their potential involvement in HIV infection, although upregulated expression of both cytokines has been reported following stimulation of mammary and amniotic epithelial cells with extracellular Tat [267].

TGF-β is a potent anti-inflammatory cytokine, secreted by hemopoietic, endothelial, and connective tissue cells. Different from other cytokines, TGF-β is produced as inactive cytokine and it became activated by proteolytic cleavage once secreted [268]. Active TGF-β influences cell cycle, differentiation, wound healing, angiogenesis and apoptosis, and its potency and multiple activities are likely also due to the fact that it is stored by the extracellular matrix.

In vivo, increased levels of TGF-β have been reported in the sera of AIDS patients with in the percentage and levels of expression) were observed for CXCR4 [304]. Thus, gut mucosa appears to provide a cellular environment highly susceptible to HIV infection, in particular for R5 strains. In fact, in vitro R5 infection of gut mucosa resulted in increased replication when compared to either PBMC [304] or tonsils [305]. On the other hand, it is of note that CD4+ cells resident in the gut mucosa are also infectable in vitro with the HIV-1 X4 strain [303-305], whereas no data about in vivo X4 infection are available [306]. In fact, this tissue probably represent the main barrier preventing X4 but favouring R5 infection [307]. Thus, it could be argued that other factors independently of cellular phenotype are involved.

In fact, over the cellular phenotype the extracellular environment and in particular the cytokine/chemokine milieu exerts a profound effect on HIV infection and replication. In fact, the cytokine microenvironment in GALT and vagina mucosa influence HIV subtype localization; high levels of IL-2, IL-7 and IL-15 in GALT [308] and of TNF-α[223] in vaginal mucosa favor infection of HIV clade C [224], possibly in virtue of the transcription binding sites (3 sites for NF-kB and 1 site for AP1) forming its LTR promoter.

Regarding cytokines, gut and viral pathogenesis many findings have been reported either in SIV infected animal and HIV+ individuals. Although the time points at which biopsies are collected are very different in the two systems (i.e., primary infection in SIV infected animals, chronic infection in HIV+ individuals), most of the findings reported in SIV infected animals have then been reproduced in humans. Regarding cytokines and SIV infection, it has been demonstrated that upon SIV infection viral replication and cytokines expression occurs at the same time points [223,309] in the different tissues, although with different levels of cytokine expression [223]. In particular, that mucosal tissue of genital tract express highest levels than systemic lymphoid tissues, and the gut even less IFN-I than LN [223].

Similarly to SIV, colon biopsy of chronically infected and under antiretroviral treatment HIV+ individuals (compared to healthy controls) were characterize to have higher levels of chemokine receptors CCR5 and CXCR4 [311], as well as of CCL5/RANTES, CCL3/MIP-1a and CCL4dMIP-1β [311,312], IL-2 [312], TNF- α [312], IL-12 [312], IL-4 [312], IFN-γ [312], CXCL10 [313] and IL-1α [313]. Another similitude with SIV+ animals was the fact that high levels of CCL5/RANTES in gut were associated with high viral loads [311], suggesting that b-chemokines are not an in vivo correlate of immune-protection. In this regard, CCR5 ligands can stimulate the replication of X4 viruses in activated primary T cells [314,315], as well as other chemokines of both the CC (such as CCL2/MCP-1) [315,316] and CXC (for example CXCL8/IL-8 and CXCLl/Gro-a) families can stimulate HIV replication in both T cells and MDM [49,50].

In the last years an in vitro system for the culture of gut biopsy/explant has been developed [305], providing important clues about acute HIV infection of this tissue. In fact, it has been shown that GALT supports very well infection with both R5 and X4 strains [305]. On the other hand, GALT does not express b-chemokines upon HIV infection, as otherwise observed in tonsils [305]. However, several immunomodulatory cytokines were upregulated upon HIV infection, such as GM-CSF, TGF-β, IL-16, IP-10, MIG, IL-10 [305]. These information suggest that in GALT acute HIV infection (at least in this in vitro system) does not increase the expression of CCR5-binding chemokines, thus explaining, at least in part, why gut mucosa is more prone than tonsillar tissue to R5 infection [305]. In conclusion, the cellular phenotype (activated memory T cells), the high percentage of CD4+ T positive for CCR5 expression, the high levels of CCR5 and the inability of HIV to induce CCR5-binding chemokines may explain why GALT is highly prone to R5 infection; these characteristics are then translated in the high cellular depletions observed after R5 infection in the gut, not/poorly observed in lymphoid tissues [305]. The above characteristics do not explain why acute X4 infection in vivo normally does not show up, a finding reported also in vitro by using intestinal lamina propria isolated mononuclear cells stimulated with CD2/CD28 showing no productive infection with X4 strain [317]. Probably other additional mechanisms may serve as gatekeepers restricting X4 infection in vivo [307]. Related to the pattern of induced cytokines it is also important to observe that the fold of induction of GM- CSF, TGF-β, IL-16, IP-10, MIG, IL-10 is always much higher that those observed in lymphoid tissue[305]; this characteristic may be important for explaining the high percentage of CD4+ cell depletion observed in gut [305,318], exceeding the number of infected cells [305]. Moreover, GM-CSF plays a role as T cells and monocyte-macrophage activator, whereas IL-10 and TGF-β are potent immunosuppressor factors, possibly contributing to cell depletion. On the other side, GM-CSF, TGF-β, IL-16, IP-10, MIG, IL-10 also have direct role in the control of viral replication [153]. Of note it is the fact that most of the chemokines induced in the gut of HIV+ individuals are also characteristic of chronic inflammations, such as ulcerative colitis and Chrohn’s disease (i.e., CCL2/MCP-1, CCL3/MIP-1a, Eotaxin, IP-10, IL-8) [319], characterized by recruitment from blood of leukocytes infiltrating the intestinal mucosa [319]. Thus, a superimposable mechanism might also be present in HIV infected individuals, leading to the continous recruitment of target cells after the acute phase (at which time point most of the resident CD4+ cells are depleted) and thus contributing to the steady-state level of viremia. Although an important innate immune response is evident during the acute HIV infection in the gut, it is also important to note that this immunoresponse is not properly functional, as evidenced by the lack of perforin expression from resident CTL [312].

During the last years many new information became available about HIV infection of GALT and vagina, shifting the scientific interest on the primary and chronic infection of mucosae. The above reports clearly show that either SIV or HIV infection induce a broad and early immune activation in the gut mucosae. Therefore, therapeutical strategies reducing inflammation (i.e., glucocorticoids and antioxidants) at mucosae must be evaluate in the regimen of anti-HIV therapies. The study of this approach is facilitated by the existence of in vitro system for the culture of gut biopsy/explant [305], recently used to evaluate toxicity and efficacy of microbicides [301,320]. Moreover, gut mucosa, as well as vaginal tissue, represent also an important anatomical site where to induce immunological responses, then translated to the systemic immunity.

Lymphoid tissue. Over GALT, the primary site of infection and CD4⁺ T cells depletion during acute HIV infection [321], draining lymph nodes represent the main reservoir of HIV infection, with the virus carried by DC [322]. Virions released via the lymphatic system spread the infection throughout the body, whereas into the lymph node virions bind to follicular dendritic cells (FDC) [292]. Highlitghtin the relevance of virus­bearing FDC is the fact that these cells have been proposed to be responsible of the second-phase of the biphasic plasma viral decay [323].

Another lymphoid organ targeted by HIV is thymus. Altered thymopoiesis is due to direct infection thymocytes [324], as well as indirect cytopathic mechanisms. This latter mechanism is mediated by the aktered cytokine milieu in the stromal of thymus. This cytokine milieu derives from apoptosis of uninfected cells [324] and mediates disruption of stromal architecture and thymocytes depletion [325]. The impairment of thymic function ends in a diminished capacity of replenish the lymphocyte pool, therefore contributing to the CD4⁺ T cells depletion observed during the disease progression [326]. In fact, the immune restoration observed in HIV⁺ children under HAART [327] or HiV⁺ adults under HAART plus IL-2 [328] has been described to be in part thymus dependent.