Malaria is still a major global public health problem with 207 million cases resulting in more than 400,000 deaths annually (Murray et al., 2012; World Health Organization, 2015). The most severe complication, termed cerebral malaria (CM), a neuropathology induced primarily by Plasmodium falciparum, is the main cause of death in human malaria (Murray et al., 2012; World Health Organization, 2015). Current therapeutic strategies of CM are limited to anti-parasitic drugs, typically administered late during infection. These pharmacological interventions lack efficacy in many cases, and the mortality rate of CM, even after treatment, remains at 10–20% (Stockman, 2012; Oluwayemi et al., 2013). Better understanding of the parasitological and immunological events leading to the development of CM could aid in the development of improved therapeutic options to treat the disease.

Multiple cell types, including monocytes, macrophages, natural killer cells and CD8⁺ T cells are sequestered within the brain at the onset of experimental CM (ECM) in mice. ECM is a serious neurological syndrome that has many of the clinical and pathological features of human CM in susceptible strains of mice infected with Plasmodium berghei ANKA (PbA) (Amante et al., 2010). In this regard, there is mounting evidence suggesting that CD8⁺ T cells are the main effector cells in the development of ECM (Boubou et al., 1999; Belnoue et al., 2002; Nitcheu et al., 2003; Potter et al., 2006; Rénia et al., 2006; Suidan et al., 2008; Hunt et al., 2010; Claser et al., 2011; Haque et al., 2011; Shaw et al., 2015). Because of ethical limitations, it is difficult to determine whether CD8⁺ T cells are involved in the pathogenesis of human CM. But in a systematic post-mortem study of the brains of Malawian children with CM, few CD8⁺ T cells were observed intravascularly in distended capillaries (Dorovini-Zis et al., 2011), which is not inconsistent with the PbA mouse model where the relatively small numbers of sequestered CD8⁺ T cells are difficult to observe by histology (Belnoue et al., 2002). Thus, the key role of CD8⁺ T cells and other immune cells in human disease has been a topic of heated debate. The cellular mechanisms implicated in the damage to the blood–brain barrier seem to involve the degranulation of Granzyme B, perforin, and proinflammatory cytokines such as interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α) and lymphotoxin-α (LT-α) (Grau et al., 1991; Engwerda et al., 2002; Nitcheu et al., 2003; Potter et al., 2006; Suidan et al., 2008; Claser et al., 2011; Haque et al., 2011). However, evidence of the induction of CD8⁺ T cells specific to blood-stage Plasmodium antigens was described only recently (Lau et al., 2011; Howland et al., 2013). Because MHC I-restricted epitopes of Plasmodium antigens during blood-stage malaria were not known, transgenic lineages of parasites expressing model epitopes, for which T-cell receptor (TCR) transgenic mice are available, were generated to study the immune response of antigen-specific CD8⁺ T cells (Lundie et al., 2008; Miyakoda et al., 2008). These studies revealed that antigens of blood-stage P. berghei parasites are captured and cross-presented by CD8α⁺ dendritic cells to induce activation, proliferation, and effector function of parasite-specific CD8⁺ T cells (Miyakoda et al., 2008; Lundie et al., 2008). In addition, they confirmed that parasite-specific cells are sequestered in the brain and are pathogenic to the host by inducing CM (Lundie et al., 2008; Miyakoda et al., 2008; Howland et al., 2015b). However, the mechanisms that induce the pathogenic activity of parasite-specific CD8⁺ T cells during the blood-stage of Plasmodium infection remain poorly understood.

Angiotensin II (Ang II) is a renin–angiotensin system (RAS) effector molecule, which exerts its actions via AT₁ receptors (AT₁R) and AT₂ receptors (AT₂R), which have been reported to mediate contrasting functions (Basso and Terragno, 2001). Initially, it was thought that the main physiological role of Ang II was to control blood pressure through the regulation of vascular tonus and electrolytic balance (Basso and Terragno, 2001). However, studies have shifted the attention toward its nonclassic effects, and Ang II has been proposed to be central in the inflammatory aspects of different diseases (Bush et al., 2000; Donadelli et al., 2000). Previously, our group and others have demonstrated that T cells express a functional RAS that produces and responds to Ang II mainly via AT₁R (Kunertradek et al., 1994; Nataraj et al., 1999; Inoue et al., 2006; Guzik et al., 2007; Jurewicz et al., 2007; Hoch et al., 2008; Platten et al., 2009; Silva-Filho et al., 2011, 2013, 2015, 2016; Zhang et al., 2012). AT₁R expression is upregulated in polyclonal T cells during the blood-stage of PbA infection, and it stimulates the production of perforin and migration/sequestration of polyclonal CD8⁺ T cells in the brain. In turn, CD8⁺ T cells promote cerebral edema, cognitive impairment, and lethal disease (Silva-Filho et al., 2011, 2013). In contrast, more recently, we showed that AT₁R signaling induces expansion but dampens the activation and exhaustion of antigen-specific CD8⁺ T cells during the effector response to whole-parasite immunization (Silva-Filho et al., 2016). Also, effector cells lacking AT₁R generate a higher number of memory cells, which is an important factor to limit parasite development in the liver (Silva-Filho et al., 2016). So, it seems that the intrinsic role and function of AT₁R in CD8⁺ T cells could be stage dependent, decreasing the protective response during immunization with liver-stage parasites but stimulating the harmful response during infection with blood-stage parasites. Indeed, inhibition of AT₁R only in Plasmodium-specific CD8⁺ T cells promoted better control of blood parasitemia and improved survival of mice against the lethal disease induced by blood-stage malaria (Silva-Filho et al., 2016). In agreement, pharmacological blockade of AT₁R protected mice against ECM and increased survival (Silva-Filho et al., 2013; Gallego-Delgado et al., 2016). Nonetheless, this study characterized the effects of AT₁R during PbA infection by using only pharmacological approaches, the effects of which may not always be due to a specific receptor blockade. Thus, there is no clear evidence whether the intrinsic role of AT₁R in CD8⁺ T cells differs following immunization vs. infection. Addressing this issue is important for better delineation of strategies targeting the AT₁R to improve the protective memory CD8⁺ T cells or to inhibit harmful pro-inflammatory CD8⁺ T-cell responses.

Here, we examined the mechanisms of action of AT₁R specifically in the Plasmodium-specific CD8⁺ T-cell response during blood-stage malaria, by using wild-type (WT; AT₁R^+/+) or AT₁R^−/− CD8⁺ T cells from ovalbumin (OVA)-specific TCR transgenic mice (OT-I), which allow the tracking of the antigen-specific CD8⁺ T cell response (Lundie et al., 2008; Miyakoda et al., 2008; Cockburn et al., 2010; Chen and Zavala, 2013; Silva-Filho et al., 2016). In parallel, to complement the genetic approach, we also tested the effects of two drugs that block the AT₁R response, losartan (AT₁R antagonist) or captopril (inhibitor of Ang II production), in mice that received WT OT-I cells. For the infection, we used transgenic PbA expressing the C-terminal fragment of OVA (amino acids 150–386) fused to the N-terminus of PbA heat shock protein 70 (Miyakoda et al., 2008). This region in the OVA comprehends the epitope SIINFEKL, which allows the response of the SIINFEKL-specific CD8⁺ T cell population (OT-I cells) to be monitored as the population responding to the blood-stage Plasmodium antigen (Miyakoda et al., 2008).

Our results revealed that blockage of AT₁R by both genetic and pharmacological tools impaired priming of Plasmodium-specific CD8⁺ T cells, as observed by reduced expression of activation markers such as CD69, CD160, and CD44. AT₁R promotes upregulation of interleukin-2 receptor α chain (IL-2Rα) and interleukin-2 (IL-2) production, important for the clonal expansion of antigen-specific CD8⁺ T cells. Moreover, there is a reduction in the expression of the integrin LFA-1 (CD11a), and the chemokine receptors CCR5 and CXCR3 in AT₁R^−/− OT-I cells. The lack of AT₁R signaling also reduced exhaustion as observed by lower expression of CTLA-4 and LAG-3. Consequently, AT₁R inhibition decreased cytokine production and degranulation by Plasmodium-specific CD8⁺ T cells. Our results demonstrate that, different from the effector response to whole parasite immunization, Ang II, via AT₁R, stimulates the Plasmodium-specific CD8⁺ T-cell response following infection with blood-stage malaria. This brings new perspectives to the protective effect of inhibition of AT₁R (Silva-Filho et al., 2013, 2016; Gallego-Delgado et al., 2016) and advances the knowledge of the mechanisms involved in the pathogenic activity of parasite-specific CD8⁺ T cells during blood-stage malaria. Moreover, it adds novelty to the functions attributed to the RAS in malaria pathogenesis.

Accumulating evidences show that CD8⁺ T cells are the main effector cells in the development of CM in murine models (Boubou et al., 1999; Belnoue et al., 2002; Nitcheu et al., 2003; Potter et al., 2006; Rénia et al., 2006; Suidan et al., 2008; Hunt et al., 2010; Claser et al., 2011; Haque et al., 2011; Shaw et al., 2015). It is known that Ang II binds to two different receptors, AT₁R and AT₂R, but we verified that only AT₁R is upregulated in CD8⁺ T cells and mediates the effects of Ang II during PbA infection (Silva-Filho et al., 2011, 2013, 2016). AT₁R acts as a co-stimulatory receptor for T-cell activation, promoting perforin expression and sequestration of CD8⁺ T cells in the brain, a cellular mechanism involved in cerebral edema and behavioral impairment during the blood stage of PbA infection, as well as in other models of diseases (Guzik et al., 2007; Platten et al., 2009; Silva-Filho et al., 2011, 2013; Zhang et al., 2012). In this work, using a transgenic parasite lineage expressing the model antigen ovalbumin (OVA-PbA) (Miyakoda et al., 2008), we further evaluated the role of AT₁R expressed in blood-stage Plasmodium-specific CD8⁺ T cells. We demonstrated that Ang II/AT₁R is a stimulatory axis for activation, exhaustion, migration and effector function of Plasmodium-specific CD8⁺ T cells. These observations suggest the role of AT₁R in stimulating a harmful CD8⁺ T-cell response that accelerates lethal CM (Silva-Filho et al., 2013, 2016; Gallego-Delgado et al., 2016) and expand the knowledge of the RAS in malaria pathogenesis.

Previous studies using transgenic lineages of parasites expressing model epitope in combination with TCR transgenic mice (Lundie et al., 2008; Miyakoda et al., 2008), revealed that antigens of blood-stage P. berghei parasites are captured and cross-presented by CD8α⁺ dendritic cells and stimulate antigen-specific CD8⁺ T cells response (Lundie et al., 2008; Miyakoda et al., 2008; Lau et al., 2011; Howland et al., 2013, 2015a). In addition, highly activated parasite-specific CD8⁺ T cells are sequestered in the brain and recognize antigens presented by endothelial cells (Grau et al., 1991; Boubou et al., 1999; Engwerda et al., 2002; Nitcheu et al., 2003; Potter et al., 2006; Suidan et al., 2008; Claser et al., 2011; Haque et al., 2011; Shaw et al., 2015). Recently, it was identified that Vβ8.1⁺ CD8⁺ T cells recognize the H-2D^b-restricted epitope SQLLNAKYL from glideosome-associated protein 50 from PbA (Howland et al., 2013). These cells are enriched in the brain during ECM (Boubou et al., 1999; Belnoue et al., 2002) and they are able to damage the blood–brain barrier and mediate CM (Howland et al., 2013, 2015a). In human CM, because of ethical limitations, the key role of CD8⁺ T cells in human disease is difficult to determine. However, there is evidence that CD8⁺ T cells play an important role. In a systematic post-mortem study of the brains of Malawian children with CM, few CD8⁺ T cells were observed intravascularly in distended capillaries (Dorovini-Zis et al., 2011), similar to mice infected with PbA mouse, where the relatively small numbers of sequestered CD8⁺ T cells are difficult to observe by histology (Belnoue et al., 2002). CXCL10, whose primary function is to mediate the migration of activated CD8⁺ T cells, has been associated with CM. CXCL10 was the only biomarker quantified in post-mortem serum, that was statistically different between Ghanaian children who had died of CM vs. malarial anemia (Armah et al., 2007). In Indian patients, high levels of CXCL10 were also associated with a higher risk of CM mortality (Jain et al., 2008; Wilson et al., 2011). In addition, a single nucleotide polymorphism in the CXCL10 gene promoter was associated with increased gene expression and a higher risk of CM (Wilson et al., 2013). The discovery of blood-stage P. falciparum MHC class I epitopes would clarify whether pathogenic CD8⁺ T cells are induced and important during human CM.

Here, using known pharmacological tools we confirmed that AT₁R is important for the clonal expansion of the effector Plasmodium-specific CD8⁺ T cells during blood-stage malaria. It has been demonstrated that IL-2 signaling is important for CD8⁺ T-cell proliferation as shown in different models (Nataraj et al., 1999; Inoue et al., 2006; D'Cruz et al., 2009). Thus, as a correlated mechanism, we verified that AT₁R increases IL-2R expression and IL-2 production by parasite-specific CD8⁺ T cells. This effect seems to involve Ang II-induced NADPH oxidase-mediated reactive oxygen species generation and activation of calcineurin phosphatase (Nataraj et al., 1999; Inoue et al., 2006). In addition to expansion, AT₁R has also been reported to modulate T-cell activation (Nataraj et al., 1999; Guzik et al., 2007; Jurewicz et al., 2007; Hoch et al., 2008; Platten et al., 2009; Silva-Filho et al., 2011). Here we observed, using both genetic and pharmacological approaches, that AT₁R induces higher activation and exhaustion of Plasmodium-specific CD8⁺ T cells during blood-stage malaria. The higher expression of CD69 and CD44 in AT₁R-sufficient OT-I cells indicates that AT₁R confers a higher capacity to respond to antigen, which could contribute to the exhausted phenotype. This also could be correlated with the higher IL-2 production by WT OT-I cells, because CD69 signaling upregulates IL-2 expression via NFAT and AP-1 transcription factors (D'Ambrosio et al., 1993). In turn, CD160 expression increases in cells expressing high levels of CD44 and producing more IFN-γ (Tsujimura et al., 2006). In agreement, CD44 and CD160 are upregulated to a greater degree in AT₁R-sufficient CD8⁺ T cells, which also produce higher amounts of IFN-γ. -γ. Along with the lower activation, there is a lower percentage of AT₁R-deficient OT-I cells expressing the exhaustion molecules CTLA-4 and LAG-3. Usually, exhausted T cells lose the ability to produce cytokines such as IL-2, IFN-γ and TNF-α and to degranulate (Wherry and Kurachi, 2015). Here, although AT₁R^−/− OT-I cells are less exhausted, they produce lower amounts of cytokines under re-stimulation. This difference could be because AT₁R^−/− OT-I cells are also less activated.

We also demonstrated the importance of AT₁R in CCR5 and CXCR3 upregulation in Plasmodium-specific CD8⁺ T cells. This could correlate with the AT₁R-induced migration and sequestration of T cells into the brain microvasculature during ECM (Silva-Filho et al., 2013; Howland et al., 2015a). In this regard, CCR5 expression and their ligand RANTES/CCL5 are upregulated in the brain of patients with CM (Sarfo et al., 2004). In addition, in CCR5-deficient mice, there is a lower number of CD8⁺ T cells accumulated in the cerebral microvasculature and 80% of mice are resistant to PbA-mediated ECM (Belnoue, 2003). Another chemokine receptor involved in ECM pathogenesis is the CXC chemokine receptor 3 (CXCR3) (Hansen et al., 2007; Campanella et al., 2008; Miu et al., 2008; Van den Steen et al., 2008; Nie et al., 2009). CXCR3 expression is upregulated only in T cells from CM-susceptible but not CM-resistant mice, indicating that CXCR3 expression correlates with disease severity (Van den Steen et al., 2008). In agreement, 70–80% of CXCR3^−/− mice are resistant to PbA-mediated CM (Campanella et al., 2008; Miu et al., 2008). Thus, the upregulation of the integrin LFA-1 (CD11a), and the chemokine receptors CCR5 and CXCR3 in AT₁R-sufficient Plasmodium-specific CD8⁺ T cells could be the mechanisms involved in AT₁R-induced sequestration of CD8⁺ T cells in the inflamed tissues during ECM (Silva-Filho et al., 2013; Howland et al., 2015a). The quality of the T-cell response, verified by the ability to produce multiple cytokines and degranulate when in contact with the target cell, is another important parameter for consideration of the effector response (Sandberg et al., 2001; Lichterfeld, 2004; Seder et al., 2008; Thiers, 2008). During blood-stage malaria, degranulation of Granzyme B, perforin, and exacerbated levels of proinflammatory cytokines such as TNF-α and INF-γ are involved with the damage to the blood–brain barrier (Grau et al., 1991; Boubou et al., 1999; Engwerda et al., 2002; Nitcheu et al., 2003; Potter et al., 2006; Suidan et al., 2008; Claser et al., 2011; Haque et al., 2011; Shaw et al., 2015). Here, we verified that along with the higher activation phenotype, WT OT-I cells produce higher amounts of IL-2, TNF-α, and IFN-γ and degranulate in comparison to AT₁R^−/− OT-1 cells.

The present results are similar to previous studies showing that Ang II, via AT₁R, increases T-cell activation, differentiation into effector cells, adhesion, and migration, leading to the infiltration of T cells in different organs (Nataraj et al., 1999; Inoue et al., 2006; Guzik et al., 2007; Jurewicz et al., 2007; Crowley et al., 2008; Hoch et al., 2008; Platten et al., 2009; Silva-Filho et al., 2011, 2013; Zhang et al., 2012). However, it is important to highlight the observed contrasts between the current findings and our results using immunization with radiation-attenuated Plasmodium sporozoites (Silva-Filho et al., 2016). For instance, CD69, CD160, CD44, LAG3, and CTLA-4 expression on day 7 is unaffected in the absence of AT₁R following immunization (Silva-Filho et al., 2016), but here we showed that the expression of these proteins is reduced by day 6 following infection. These data suggest that there are biologically important context-specific differences in the intrinsic role of AT₁R in CD8⁺ T cells. Likewise, HO-1 plays distinct roles at different stages of the Plasmodium life cycle. During the liver stage, upregulation of HO-1 leads to an increase in parasite liver load (Epiphanio et al., 2008), whereas in mice injected with PbA-infected red blood cells, the establishment of experimental CM is suppressed (Pamplona et al., 2007). Malaria is a very complex disease, with multiple arms of immune system working together and modulating one another. Thus, intrinsic differences between the two experimental models, such as the absolute expression level of the antigen, the location and mechanisms of antigen presentation, are important factors in determining antigen-specific immune responses that could explain such discrepancies (Bagot et al., 2004; Lin et al., 2014). In addition, divergences may also occur due to: (1) the life-long deletion of AT₁R in T cells might lead to compensatory changes in other genes and related signals, modulating T-cell populations or other properties of specific T cells; (2) AT₁R can exhibit dual signaling in which G-protein activation leads to deleterious effects, whereas Gα(q) protein-independent/β-arrestin–dependent pathways promote beneficial effects (Zhang et al., 2012); (3) the AT₁R can heterodimerize with other angiotensin or bradykinin receptors, affecting downstream signaling pathways. In addition, AT₁R activates multiple downstream signals important to induce pro-inflammatory transcription factors, T-cell activation, proliferation, chemotaxis and cytokine production (Sinclair et al., 2008; Smith-Garvin et al., 2009; Balakumar and Jagadeesh, 2014). However, the AT₁R-induced signaling pathways behind these processes are not yet known. Thus, the predominance of a set of signaling pathways induced by AT₁R could diverge during the T-cell response following different immune contexts. Future studies will investigate how AT₁R signaling leads to transcription of the T-cell molecules following infection vs. immunization, which may indicate additional levels of complexity in the role of AT₁R in the interplay between host and pathogen.

Previously, it was suggested that increased levels of Ang II have a beneficial effect against malaria-induced pathology in mouse models. A significant reduction in blood parasitemia was observed in mice infected with PbA treated with a supraphysiological concentration of Ang II (Gallego-Delgado et al., 2015). A moderate reduction in the establishment of CM and decreased incidence of brain hemorrhage followed by a modest increase in survival were also verified (Gallego-Delgado et al., 2015). Apparently, these results differ from the current study and our previous works (Silva-Filho et al., 2013, 2016). However, because Ang II is quickly metabolized forming different biologically active peptides, the amelioration of mice could be attributed to increased levels of plasma Ang-(1–7) (Saraiva et al., 2011; Silva et al., 2016a,b). Recently, the same group verified that pharmacological blockade of the AT₁R or stimulation of AT₂R protected mice against CM, reduced cerebral hemorrhages and increased survival (Gallego-Delgado et al., 2016). In contrast, AT₂R-deficient mice were more susceptible to CM. In agreement with this study and our previous works (Silva-Filho et al., 2013, 2016), these observations confirm that Ang II receptors could influence the outcome of experimental CM; protection can be achieved by blockade of AT₁R or activation of AT₂R, whereas the opposite effect is observed by activation of AT₁R when AT₂R is deleted. Thus, in response to infection during blood-stage malaria, AT₁R induces the formation of a larger effector population of Plasmodium-specific CD8⁺ T cells with a higher capacity to migrate to inflamed tissues, translated by the higher expression of integrin and chemokine receptors, and higher cytokine production. These observations show that inhibition of AT₁R signaling restricts Plasmodium-specific CD8⁺ T-cell function. In addition, because the level of antigen-specific T-cell recruitment to the brain is governed largely by the magnitude of splenic T-cell priming rather than by secondary differences in brain-localized T-cell migratory cues (Lin et al., 2014; Howland et al., 2015a), these data suggest how inhibition of AT₁R could increase the resistance of mice against CM, promoting survival, improving cognitive parameters, and reducing cerebral edema (Silva-Filho et al., 2013, 2016; Gallego-Delgado et al., 2016). Together, these data bring new contributions to the mechanisms involved in the pathogenic activity of CD8⁺ T cells during the blood-stage of Plasmodium infection and to the functions of the RAS in malaria pathogenesis. Repurposing of Ang II modulators, such as AT₁R antagonists or ACE inhibitors, as adjunctive treatment for CM is a potential therapeutic possibility (Silva-Filho et al., 2013; Gallego-Delgado et al., 2016; Silva et al., 2016b).

JS performed all the experiments; JS, CC, and AP conceived the project, and JS and AP wrote the manuscript. All authors read and approved the final version.

This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (www.cnpq.br): 1. 57.3695/2008-3 AS, 2. 57.3767/2008-4 CC, 3. 471771/2013-9 CC, 4. 456997/2014-8 AS, and Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (www.faperj.br): 1. E-26/110.551/2010 CC, 2. 111681/2013 CC, 3. E-26/102.170/2013 AS, 4. E-26/201.197/2014 CC.