- Top of page
- Materials and methods
The events occurring at the time of the initial encounter of HIV-1 with the immune system are understood to be critical in terms of later prognosis [1,2]. Initiation of antiretroviral therapy (ART) during primary HIV-1 infection (PHI) can decrease viral diversification and saturation of reservoirs, and preserve virus-specific CD4 T helper responses. It has also been reported to decrease rapid clinical progression . All these factors may be very important in the setting of improved ARTs and use of immune modulation .
Measurements of cytokine and chemokine levels have been useful in the understanding of HIV-1 pathogenesis. Viral replication is regulated by a complex network of both HIV-1-suppressive [e.g. interferon (IFN)-α, interleukin (IL)-10, macrophage inflammatory protein (MIP)-1α, MIP-1β and RANTES] and inductive [e.g. tumour necrosis factor (TNF)-α, IL-1β, IL-6, IL-15 and monocyte chemotactic protein (MCP)-1] cytokines and chemokines . Although there are conflicting reports regarding the role of certain cytokines during the course of the infection, their contribution to the state of generalized immune activation is well acknowledged [6–8]. The aim of this study was to characterize the pattern of immune activation using analysis of cytokines and chemokines in very early infection, in order to correlate their levels to virological outcome both at the time of PHI and during an analytical treatment interruption (ATI).
- Top of page
- Materials and methods
Immune activation is an important factor in the pathogenesis of chronic HIV-1 infection [12,13]. Disease stage and VL correlate with the level of immune activation . However, the role of immune activation has not been determined during the very early stage of the infection which is associated with intense viral replication and major loss of some of the components of the immune system [1,15,16]. Plasma HIV-1 RNA of untreated subjects tends to reach a set-point 1–2 months after PHI  and is considered to be a prognostic marker of disease progression . In our study, a clear correlation was seen between VL values at PHI and those after ATI. This suggests that the events immediately after HIV transmission and during the first few days after onset of PHI symptoms influence viral kinetics in later infection.
All subjects showed an intense state of immune activation during PHI, the degree of which correlated with VL, and was higher in the nonresponder group. The only exception was for MCP-1, the levels of which were higher during PHI among responders. We found that the cytokine/chemokine pattern during the PHI phase was associated with the type of virological outcome after treatment cessation; thus, subjects who had lower levels of VL and of all soluble factors at PHI (with the exception of MCP-1) also had lower VL after ART cessation.
It has been suggested that events after ART cessation mimic those of the PHI period. Thus, in general viral kinetics follows at least partially the pattern seen during PHI, and clinical symptoms similar to those seen at PHI may occur [15,18]. However, in our study clear differences were seen in the cytokine profile after cessation of ART compared with the PHI phase. We found that in the ATI phase responders had significantly elevated levels of IFN-γ, MIP-1β and MCP-1 and significantly lower levels of IL-15 and eotaxin compared with nonresponders. This shift in the cytokine profile suggests that there may be better preservation of innate and adaptive immunity in subjects with a lower degree of immune activation during PHI. Betts et al.  recently suggested that the presence of polyfunctional HIV-1-specific CD8 T cells characterized by strong cytokine (IFN-γ, TNF-α and IL-2) and chemokine (MIP-1β) production is negatively correlated with disease progression. Our data for the plasma compartment are consistent with this hypothesis. The analysis of the cytokine profile at the time of the VL peak during PHI and after ART cessation also shows that levels of IFN-γ, TNF-α, IL-10, MIP-1α, MIP-1β, RANTES and eotaxin were significantly higher during PHI than after ATI. This finding was probably related to the substantially higher VL during PHI [20,21].
A beneficial effect of increased levels of β-chemokines during PHI [22,23] was not confirmed in our study. RANTES, MIP-1α and MIP-1β followed the general cytokine pattern. However, changes were observed during the ATI phase, with a tendency towards increased chemokine levels in responders and a negative correlation with VL for MCP-1, MIP-1α and MIP-1β. Thus, the chemokines studied here may have different impacts on viral replication depending on the phase of HIV-1 infection considered [24,25].
We found that plasma levels of IL-15 and eotaxin were elevated in nonresponders during ATI. Plasma IL-15 has been previously reported to be increased in subjects on ART  as well as in those with successful structured treatment interruption (STI) outcomes . In contrast, our data do not support a beneficial effect of IL-15 after cessation of ART in terms of virological set-point.
Increased plasma levels of eotaxin have been found in several inflammatory diseases [27–29]. This protein is not only a potent chemoattractant of eosinophils but also contributes to the recruitment of immune cells following viral infection . Eotaxin is a ligand for chemokine (C-C motif) receptor 3 (CCR3), which can serve as an HIV-1 co-receptor in at least in vitro settings [31,32]. Furthermore, genes for eotaxin, MCP-1 and MCP-3 are localized to the same chromosomal region, which has been linked to a role in the modulation of HIV-1 transmission, probably by activating the immune system . Interestingly, Promadej-Lanier et al. have shown that peak systemic levels of eotaxin coincide with initial detection of viral RNA in macaques challenged with vaginal simian/human immunodeficiency virus (SHIV) inoculate. The uniform high eotaxin levels in nonresponders during both the primary and the ATI phases strongly suggest that this chemokine does contribute to immune activation, rather than having a beneficial effect in vivo.
The cellular sources of the chemokines and cytokines studied here have not been explored. Both their active release by immune cells and their release during cell death may contribute to the elevated plasma levels observed during PHI. The initial phase of HIV-1 infection is indeed associated with a rapid decline in CD4 T cells as a result of cell death caused by necrosis and apoptosis [16,35]. Necrotic cell death (direct viral cytopathic effect) is associated with inflammation/immune activation and with the recently described involvement of the intracellular protein high mobility group box protein 1 (HMGB1) in this process [36–38].
In conclusion, our study adds important information in terms of the pattern of cytokine and chemokine production during early HIV-1 infection and after cessation of ART. We have shown that the degree of immune activation is highest during PHI and is associated with VL. Further investigations are required to clarify the role of ART in the change in cytokine/chemokine pattern observed after treatment interruption and its potential impact on the long-term virological outcome.