P. falciparum glycosylphosphatidylinositol (GPI) is the first factor that was identified as a malaria parasite PAMP (88). Structurally, parasite GPI comprises a heterogenous group of molecules consisting of triacylated phosphatidylinositol linked to the glucosamine moiety of glycan having four mannose residues and a glucosamine residue (89–91). The structural heterogeneity of malaria GPI is due to the variation in length and level of unsaturation in acyl residues present at different positions of the phosphatidylinositol moiety. This structural heterogeneity in the lipid moieties has no bearing on immune response-inducing activity of GPI. This is evident from the fact that the sn-2 lyso GPI, obtained by the removal of acyl moiety at the sn-2 position of parasite GPI, could efficiently induce cytokine responses similar to the unmodified parasite GPI. The biosynthesis of GPI is essential for the survival of parasites as GPI anchors several proteins of merozoites to the plasma membrane that are involved in erythrocyte invasion (92–94). In the absence of GPI anchoring, these proteins are not expressed on the surface and hence merozoites cannot invade erythrocytes. Malaria parasites synthesize GPI in several folds excess over the actual amounts needed for anchoring proteins to the surface of merozoites and thus, significant amounts of GPI remain not linked to proteins (90). The GPI molecules that are not linked to proteins are exposed on the cell surface and thus are likely targeted by the innate immune system.

The identification that parasite GPI is a malaria inflammatory response-inducing pathogenicity factor was based on the observation that the purified parasite GPI could induce strong pro-inflammatory responses by Mφs (88). When administered to mice, GPI induced symptoms that resembled the systemic clinical conditions of malaria, including pyrexia, cachexia, hypoglycemia, and TNF-α-induced sepsis. In several subsequent studies, GPI was shown to induce a wide range of immune responses, including the production of TNF-α and IL-1 by Mφs, expression of nitric oxide synthase by Mφs and endothelial cells, and the expression of ICAM-1, VCAM-1, and E-selectin by leukocytes and endothelial cells through the activation of several signaling events (95–97). Consistent with the property of GPI in inducing malaria-like symptoms, immunization of mice with a synthetic glycan portion of GPI produced anti-GPI antibodies, and immunized mice infected with P. berghei ANKA, an experimental cerebral malaria model, were protected from cerebral malaria (98). Further, the presence of anti-GPI antibodies in people in malaria endemic was associated with a significant protective immunity against malaria illnesses (99, 100).

Subsequent studies have shown that P. falciparum GPI activates Mφs through the induction of an outside-in signaling, without binding to plasma membrane or being internalized by the cells, but instead by recognition through weak interactions on the cell surface (101). The intact structure of GPI is essential for bioactivity as neither the carbohydrate moiety nor the triacylated phosphatidylinositol lipid portion can induce immune responses. Further studies have shown that the parasite GPI induced production of pro-inflammatory cytokines by Mφs occurs through recognition mainly by TLR2-TLR1 heterodimer and to a much lesser extent by TLR4 (102–105); Figure 1 and Table 2. Sensing of GPI by TLR2-TLR1 leads to the activation of ERK, p38, JNK MAPK, and NF-κB signaling pathways, which differentially contribute to the production of various inflammatory mediators (103). Thus, malaria parasite GPI is mainly a TLR2-activating PAMP.

A large body of accumulated data over the past two decades demonstrates that microbial (bacteria, viruses, and parasites) DNA, and self-DNA in some pathological situations, function as PAMP and are recognized by TLR9 in endosomes and by DNA sensors in the cytosol (56–60). In the case of malaria, the first demonstration that TLR9 senses malaria parasites and induces immune responses was by Pinchyangukul et al. (123). They showed that a soluble component in the schizont extract of P. falciparum that was heat labile and precipitable with ammonium sulfate (the description agrees with the active component being protein-DNA complex) induces cytokine and chemokine responses by human plasmacytoid DCs (pDCs) and mouse DCs through the activation of TLR9-MyD88 signaling pathway. Subsequently, it was shown that the TLR9 signaling-inducing malaria factor is DNA (107, 108, 124). TLR9 specifically recognizes the unmethylated CpG motifs of DNA (68, 69). The genomic DNA of P. falciparum and P. vivax contain, respectively, ~300 and ~2,000 CpG motifs (109). The higher content of CpG motifs is likely responsible for the strong fever-inducing ability of P. vivax compared to P. falciparum (125).

Malaria parasite DNA enters the endosomes of the innate immune cells, such as Mφs and DCs, through phagocytic uptake of IRBCs, merozoites, the nuclear material of parasites, DNA-protein-hemozoin complex, and DNA-containing immune complexes (107, 108, 124, 126). The endosomes are then fused to lysosomes to form phagolysosomes. In the acidic environment of phagolysosomes, DNA is released and recognized by TLR9, leading to the activation of MAPK and NF-κB signaling pathways and cytokine and chemokine responses (Figure 1 and Table 2). In addition to TLR9, parasite DNA is recognized by cytosolic DNA sensors upon the release of phagolysosomal contents into the cytosol. In the cytosol, several distinct cytosolic sensors can potentially recognize DNA (Figure 1). Thus far, it has been demonstrated that two cytosolic PRRs sense parasite DNA: (i) cGAS sensing dsDNA and inducing the activation of STING-TBK1-IRF3 signaling and downstream production of type I IFNs (44, 110), and (ii) AIM2 recognizing dsDNA, resulting in the activation of inflammasome and caspase 1. The activated caspase 1 converts pro forms of IL-1 and IL-18 into active IL-1 and IL-18. Robust production of IL-1 and IL-18 in response to malaria infection requires, in addition to inflammasome signaling, TLR (mainly TLR9 and TLR7) mediated production of pro-IL-1 and pro-IL-18. P. falciparum genomic DNA contains >80% AT nucleotides. The AT-rich motifs of parasite DNA form loop structures, which have been shown to induce type I IFN response through STING-TBK1-IRF3 signaling (109). In some viruses, the AT-rich loop motifs of DNA are transcribed by RNA polymerase III to form double-stranded RNA containing 5′-triphosphate that induces type I IFNs through RIG-I-MAVS-IRF3 signaling (110). However, in the case of P. falciparum DNA, pol III-dependent recognition of AT-rich motifs of DNA seems to be not involved in cytosolic sensing as deficiency in pol III had no effect on immune responses to the parasite DNA (109).

The innate immune system senses RNA of both the liver and the blood stage parasites. In the liver stage, as described in the section Liver Stage Parasite Sensing above, RNA is recognized exclusively in the cytosol by MDA5 (Figure 1) (79, 80). In contrast, in the blood stage infection, mouse parasite RNA is recognized by TLR7 in phagolysosomes of DCs, leading to type I IFN production (Figure 1). In fact, studies in a mouse model of P. chabaudi-infection showed that type I IFN production is the earliest cytokine response (24 h postinfection) during the blood stage malaria infection (127). Parasite RNA enters endosomes of Mφs and DCs upon the uptake of parasites/parasite components, inducing type I IFN response through TLR7 signaling (81, 106, 128). Type I IFNs thus produced prominently contribute to the upregulation of pro-inflammatory cytokines, such as IFN-γ and IL-12, during the early stages of infection (106). RNA of the blood stage mouse parasites is also sensed by cytosolic MDA5, leading to type I IFN production (81).

While it is clear that RNA induced TLR7 signaling plays an important role in early type I IFN production in mouse malaria, it remains unclear whether or to what extent RNA of human parasites is able to induce type I IFN response. This is because, although RNA of human malaria parasite P. falciparum has been reported to induce cytokine responses through TLR7 recognition, the reported activity appears to be very low (106). Further studies are needed to determine whether or to what extent P. falciparum RNA is immunostimulatory.

Hemozoin is a hydrophobic, crystalline insoluble polymer of heme formed during the digestion of hemoglobin by parasites in the digestive vacuoles to use the released amino acid as a food source (45, 129–131). In parasites, hemozoin is associated with lipids and proteins, and is released into the blood circulation when merozoites are egressed from matured schizonts. Hemozoin by itself is an inert material and appears to have no specific receptor for recognition, but it influences the innate immune responses to malaria parasites in several ways; reviewed in Olivier et al. (45). Studies have reported that hemozoin is a carrier of malaria parasite DNA into endosomes for TLR9 recognition (107). Although how and where parasite DNA associates with hemozoin have not been specifically studied in vivo, there exist several possibilities. Malaria merozoites egressed from the matured schizonts have a half-life of <5 min (132), and many merozoites cannot invade red blood cells within this short period time. These uninvaded merozoites are likely to be lysed, releasing DNA that may complex with hemozoin via the associated proteins. Additionally, Mφs and neutrophils undergo apoptotic death after ingesting IRBCs (133), releasing the degraded parasite materials, including DNA, RNA, hemozoin and other components. The released DNA can complex with hemozoin. However, hemozoin is not obligatory for the entry of parasite DNA into endosomes. The parasite nuclear material, DNA containing immune complexes, merozoites, and IRBCs are also taken up by phagocytic cells (107, 108, 124, 126). It is difficult to quantify in vivo the extent to which DNA enters endosomes through phagocytic uptake of infected erythrocytes, merozoites and nuclear materials compared to hemozoin-bound DNA in inducing immune responses.

Since hemozoin is an inert, indigestible material, it curtails the ability of Mφs to induce innate immune responses to malaria. Upon uptake of IRBCs or hemozoin, Mφs become immunosuppressive and dysfunctional due to damages caused by oxidative burst (134). Monocytes and Mφs that ingest parasite IRBCs undergo apoptotic death with little or negligible release of cytokines, although cytokines are expressed to certain extent (133). Phagocytic internalization of large amounts of whole parasites and non-digestible hemozoin renders Mφs nonfunctional because of phagolysosomal acidification and apoptotic death (133, 135). Hemozoin is known to inhibit the differentiation and maturation of human monocyte-derived DCs as well (136, 137). However, certain subsets of inflammatory Mφs, such as spleen red pulp Mφs, CD169⁺ Mφs, and CD16⁺ monocytes produce cytokines in response to malaria parasites (126–128). It is not known whether hemozoin significantly alters the capacity of these cells to produce cytokines and chemokines.

The effect of hemozoin on DC function remains unclear, reviewed in Wykes and Good (138). A previous study has shown that hemozoin-internalized DCs localized in T cell areas of the spleen in P. chabaudi-infected mice and that T cell activated by these DCs lacked effector function (139). These results suggest that hemozoin considerably compromises DC function. However, in a later study, DCs efficiently produced DNA-TLR9 signaling-induced cytokines in response to DNA bound to natural hemozoin and thus it was suggested that hemozoin enhances the activity of DNA by facilitating its entry to endosomes (107). It remains unclear whether this is the case in infected host. On the other hand, it has been shown that subsequent to the uptake of parasite components and the activation of TLR7 and TLR9 signaling, hemozoin destabilizes phagolysosomal membrane, leading to the release of nucleic acid and hemozoin into the cytosol (111). The released parasite components are sensed by cytosolic PRRs, leading to several signaling pathways (44, 81, 111–113), including: (i) cGAS-STING-TBK1-IRF3 signaling by dsDNA; (ii) MDA5-MAVS-TBK1-IRF3 signaling by RNA; (iii) STING signaling mediated by AT-rich motifs through sensing by an unidentified receptor; (iv) AIM2 inflammasome signaling induced by parasite DNA, and (v) NLRP12 inflammasome activation via an unidentified mechanism (82). In addition, hemozoin also activates NLRP3 inflammasome (45, 111–113); Figure 1 and Table 2. The activation of NLRP3 inflammasome by malarial hemozoin is mediated through the activation of Lyn and Syk tyrosine kinases (45). Synthetic hemozoin also activates Lyn and Syk signaling and, depending on the morphology and particle size, induces distinct immune response. It appears that hemozoin has no specific PRR, but it induces a danger signal (112, 113); hence, hemozoin is a DAPM. While nucleic acid-TLR-mediated signaling results in the synthesis of large amounts of pro-IL-1, inflammasome signaling and activation of caspase 1 lead to the cleavage of pro-IL-1 to the active, fever-inducing IL-1. Thus, hemozoin plays a significant role in the production of IL-1, contributing to fever induction in infected people. Besides hemozoin, parasite biomass, and purified merozoites that are devoid of hemozoin can potentially activate inflammasome and caspase 1 as it is known that even inert materials such as alum, asbestos, silica, uric acid crystals and cholesterol particles, activate NLRP3 inflammasome (45).

Uric acid is the final oxidation product of purine metabolism and is released in large amounts by dying cells. Under physiological conditions, uric acid exists as a monoionic urate and forms insoluble monourate crystal (140). Uric acid is an important antioxidant in plasma and can induce protective anti-inflammatory responses in vascular and other diseases (141, 142). However, excessive formation of uric acid promotes pathogenic conditions, such as severe and chronic inflammatory arthritis, gout, and certain metabolic syndromes (141, 142). The pathogenic role of uric acid is due to the activation of NLRP3 inflammasome, which results in caspase 1 activation and conversion of pro-IL-1 into active IL-1 (114–116); Figure 1 and Table 2. During malaria, a large number of parasite-ingested Mφs, neutrophils and other immune cells die, which likely release high levels of uric acid. Importantly, a significant amount of uric acid is formed during purine nucleotide metabolism by parasites and accumulates as a precipitate in IRBCs (143). Large amounts of hypoxanthine also accumulate in parasite IRBCs. The uric acid precipitate and hypoxanthine are released to the blood upon schizont burst. In the blood stream, hypoxanthine is oxidized to uric acid. In agreement with its pathogenic role, high levels of uric acid are found in the blood circulation of patients having severe malaria (144).

CD36 is a multifunctional class B scavenger receptor that binds diverse types of ligands and pathogens, facilitating internalization by cells (67, 118, 149). Many cell types, including Mφs, DCs, platelets, and endothelial cells, express CD36. CD36-mediated phagocytosis of pathogens and pathogenic molecules activates Src/Syk kinases, leading to the activation of ERK, p38 and Jun members of the MAPK signaling pathways and NF-κB (67, 118, 119, 150) (Figure 2). Signals of these pathways synergize with TLR and inflammasome signaling, contributing to innate immune responses (119, 156). In malaria, CD36 plays several roles, including (i) mediating the sequestration of P. falciparum in vascular capillaries through the binding of adhesive PfEMP1 expressed on the surface of IRBCs, (ii) phagocytic uptake of parasites, and (iii) enhancing innate immune responses (120, 157–162). Studies have shown that CD36-induced MAPK signaling contributes to the production of TNF-α in mouse malaria infection and modulates parasite GPI-induced cytokine responses in mouse Mφs and human blood DCs (120, 161, 163). In P. falciparum endemic regions, single nucleotide polymorphisms in the Cd36 gene have been linked to protection from cerebral and other severe malaria (164, 165). A more recent study in P. yoelii-infected mice demonstrated that CD36 significantly contributes to cytokine responses by innate immune cells, and the upregulation of MHCII expression, phagocytic activity, Th1 responses and expression of complement and Fc receptors (117). Overall, these responses lead to decreased parasite burden in infected host. It is possible that, similar to CD36, other scavenger receptors, such as MARCO modulate innate immunity to malaria to a certain extent.

Endothelial protein C receptor (EPCR), which is also known as activated protein C (APC) receptor, is a transmembrane glycoprotein in vasculatures that binds protein C and promotes thrombin-thrombomodulin complex-mediated protein C activation (166). Under normal conditions, thrombomodulin on the endothelial surface binds thrombin and activates protein C to APC; this process is strongly promoted by EPCR. APC is detached from EPCR and inactivates blood coagulation factors Va and VIIIa, thereby exerting anticoagulant effect. APC binding to EPCR confers cytoprotective roles, such as antiapoptotic, anti-inflammatory and barrier stabilization responses (153–155). In malaria, EPCR binds certain members of PfEMP1 expressed on the surface of IRBCs and contributes to severe malaria (17, 167–169). Although the downstream events of EPCR-PfEMP1 binding-mediated signaling remain to be understood, it appears that the binding results in the loss of EPCR and thrombomodulin from the endothelial cell surface by shedding, leading to decreased protein C activation and compromised EPCR-mediated protection (Figure 2). This may reduce anti-inflammatory responses and increase thrombin production, blood coagulation, pro-inflammatory responses, endothelial cell apoptosis, leading to the loss of barrier function. The cumulative effects of these responses may contribute to endothelial cell damage, promoting cerebral and other severe malaria illnesses.

ICAM-1 mediates the sequestration of P. falciparum in the brain through the binding of PfEMP1 on the surface of IRBCs to endothelial cells (15, 16, 25). The sequestration of parasites in vascular capillaries of the brain induces inflammatory responses, resulting in the infiltration of cytotoxic effector cells and endothelial barrier damage. These processes have been implicated in the development of cerebral malaria, but the experimental evidence is controversial. An earlier study has reported that interaction with ICAM-1 contributes to increased serum TNF-α in cerebral malaria model of P. berghei ANKA-infected mice and that ICAM-1 deficient mice survive >15 days compared to 6–8 days in wild type (WT) mice (170). However, a later study reported that ICAM-1 is dispensable for cerebral malaria pathogenesis (171). NK cells cross-talk with myeloid cells through LFA-1 binding to ICAM-1, producing IFN-γ (172). Similarly, T cells are also known to interact with endothelial cells through LFA-1 mediated binding to ICAM-1, producing IFN-γ. Studies have shown that, upon binding to its ligands, ICAM-1 is activated by phosphorylation, inducing Src-PI3K-Akt, Ca2+, Rho and MAPK signaling (121, 122) (Figure 2). It is likely that these signals contribute to immune responses to malaria; thus far this aspect has not been examined.

In malaria, like in most pathogenic infections, the innate immune system functions as the first line of defense by controlling parasite growth and regulating the development of adaptive immunity (14, 75–77). As discussed in section Liver Stage Parasite Sensing, during the liver stage malaria infection, parasite-infected hepatocytes produce type I IFNs through cytosolic sensing of RNA. This cytokine response contributes to the killing of parasite-infected hepatocytes by NKT cells, exposing parasite components. The antigen-presenting cells, primarily DCs and inflammatory Mφs, can potentially recognize the exposed parasite factors. In addition, sporozoites injected by infected mosquitoes that could not enter the blood circulation remain in dermis and die. In addition, some sporozoites that enter blood circulation may not invade hepatocytes and die. These dead parasites are likely sampled by DCs and inflammatory Mφs, leading to TLR- and inflammasome-mediated immune responses. The activated antigen-presenting cells have potential to modulate immune responses to the blood stage infection. However, because the parasite load in liver in natural infections is very low, the innate immune responses are likely to be also very low. Thus, the responses produced by antigen presenting cells against liver stage parasites may not exert a significant influence on immune responses induced by the blood stage parasites. However, in hyper endemic areas, repetitive infections in humans may induce immune tolerance in antigen-presenting cells that may modulate immunity to the blood stage infection to a certain extent.

During the blood stage infection, unlike the liver stage, parasites grow exponentially through repetitive erythrocytic cycles, rapidly accumulating the biomass. This leads to an efficient induction of innate immune responses. Early during the blood stage infection, DCs and Mφs are key first responders of the innate immune system. However, as noted above, Mφs upon internalization of infected erythrocytes, merozoites or hemozoin become immunosuppressive and dysfunctional (133–137); unable to secrete cytokines and chemokines. It appears that the primary role of Mφs during the early stages of blood stage malaria infection is to control parasite growth through phagocytic clearance. In contrast to Mφs, DCs efficiently produce cytokines and chemokines in response malaria parasites, and effectively interact with cells of the innate and adaptive immune system. Thus, DCs play key roles in the initiation and regulation of innate and adaptive immunity to malaria. Also, it should be noted that some subsets of Mφs having certain features of DCs, such as CD11c⁺ spleen red pulp Mφs and CD169⁺ inflammatory Mφs, produce cytokines (127, 133, 173).

As noted in section RNA, during the blood stage malaria infection, type I IFNs are the earliest (24 h postinfection) cytokines produced through the activation of TLR7-MyD88 and IRF7 signaling (174). Early type I IFN response is produced primarily by the spleen red pulp Mφs and pDCs in P. chabaudi-infected mice (106, 127). Two recent studies by using lethal malaria model of P. yoelii YM infection, in which parasites grow rapidly to attain peak parasitemia of ~80% by 6 days postinfection, have also showed TLR7-dependent peak levels of type I IFN production at 24–36 h postinfection (81, 128). This early cytokine response is mediated through the coordination of TLR7- and cytosolic sensing mechanisms: TLR7-MyD88-IRF7 signaling in endosomes, and DNA-cGAS-STING-TBK1-IRF3/IRF7 and RNA-MDA5-MAVS-TBK1-IRF3 signaling in the cytosol (81). It has been found that SOCS1 expressed in response to cytosolic nucleic acid sensor-mediated signaling significantly downregulates TLR7-mediated type I IFN response. As such, deficiency in SOCS1 results in markedly high levels of type I IFNs, providing resistance against high parasite burden-dependent lethality (81). Additionally, it has been shown that CD169⁺ Mφs activated through STING-mediated signaling migrate to bone marrow, where they interact with pDCs to induce type I IFN production through TLR7-MyD88 signaling (128). Interaction of classical DCs with pDCs is also important in pDCs producing cytokine responses to malaria parasites (38, 108). It appears that Mφs and neutrophils that internalize parasites undergo pyroptosis (111), exposing parasite components, which are taken up by classical DCs and CD169⁺ Mφs, and activated through STING-mediated cytosolic signaling. The early type I IFN response triggers the infiltration of immune cells to the blood that may subsequently localize to the spleen. In contrast, in P. berghei ANKA-infected mice, relatively high level of IFN-α was seen at 4 days postinfection with low or negligible IFN-α production during 1–3 days postinfection (175). Collectively, the above results indicate that different strains of malaria parasites differentially induce type I IFN. That is, the earliest type I IFN production (24–36 h postinfection) by P. chabaudi and P. yoelii is mediated mainly through parasite RNA-induced TLR7 signaling (81, 106, 128), whereas such RNA-mediated early type I IFN response is not readily apparent in P. berghei ANKA (175). The production of type I IFNs at later stages of P. berghei ANKA infection likely involves a different mechanism. In other studies, cGAS sensing of the AT-rich motifs of P. falciparum DNA in the cytosol could induce type I IFN production through the STING-TBK1-IRF3/IRF7 signaling pathway, independent of TLR9-MyD88 signaling (109, 110). This could be the mechanism through which P. berghei ANKA induces significant levels of type I IFN at 4 days pi (175).

Early type I IFN response to malaria infection contributes to the suppression of antiparasitic immunity and promotes cerebral and other severe malaria illnesses under certain situations. This is evident from the observations that deficiency in IFNαR in P. yoelii YM infection resulted in increased IFN-γ levels, significantly decreased parasitemia and resistant to parasite burden-dependent death (128). Also, in P. berghei ANKA-infected mice, which produce low levels of early type I IFN response but produce a significant amount at a somewhat later stage (4 days postinfection), deficiency in IFNαR or treatment with anti-IFNαR1 antibody resulted in increased IFN-γ production, increased number of IFN-γ-positive NK and CD4⁺ T cells in the spleen, and significantly decreased parasitemia and protection from cerebral malaria (175, 176). From these results, it is evident that type I IFNs suppress the production of IFN-γ and thus anti-parasitic function under certain conditions. By contrast, in a different situation, high levels of type I IFN production at early stages of P. yoelii YM infection promoted antiparasitic immunity. In this case, blocking of SOCS1, a suppressor of cytokine signaling 1, expression resulted in high levels of type I IFN production, leading to increased IFN-γ production and reduced parasitemia, protecting mice from parasite burden-dependent death (81). Consistent with the observations of the latter study (81), daily treatment of P. berghei ANKA-infected mice with recombinant IFN-α, which to a certain extent resembles a situation of high levels of type I IFN production at early stage of infection, also significantly increased IFN-γ production by splenic CD8⁺ T cells, markedly reducing parasitemia and preventing cerebral malaria (177). The results of these two studies suggest that type I IFNs promote IFN-γ-dependent anti-parasitic immunity, providing resistance against severe malaria. On the other hand, treatment of P. berghei ANKA-infected mice with recombinant IFN-β resulted in reduced TNF-α and IFN-γ production, decreased expression of ICAM-1 and CXCL9 in the brain, reduced CXCR3 expression by T cells and T cell infiltration to the brain, thereby significantly preventing cerebral malaria and increasing mice survival (178). Thus, type I IFNs play contrasting roles in malaria in a context-dependent manner and also dependent on type I IFN isomeric composition.

Compared to the results of two studies outlined above (175, 176), wherein deficiency in IFNαR contributed to increased IFN-γ response and decreased parasitemia in P. berghei ANKA-infected mice, Palomo et al. observed contrasting immune responses (179), although in all these studies mice were protected from cerebral malaria. Palomo et al. found that, compared to infected WT mice, P. berghei ANKA-infected mice deficient in IFNαR had similar parasitemia, reduced expression of IFN-γ and granzyme B by T cells, decreased expression of T cell-attracting chemokine CXCL9, and low infiltration of CXCR3⁺-expressing CD8 T cells to the brain. A notable difference in these studies is that the latter study used GFP knock-in transgenic parasite clone, which appeared to have slower growth rate than the WT parasites used in the former study. These differences may alter the dynamics of immune responses. Nevertheless, the results agree with the notion that type I IFNs play distinct roles under different malaria conditions.

Overall, the available data indicate that type I IFNs play disparate roles in malaria infection depending on the timing and amount of production, relative levels of IFN-α, IFN-β and perhaps other isomers, and compositions of cellular and cytokine milieu during the progression of infection, and parasite strains. A recent review provides a comprehensive and up-to-date account on the role and contrasting effects of type I IFNs in malaria (180). Type I IFNs are pleiotropic cytokines and as such, they induce a wide range of effects on innate and adaptive immune cells during various pathogenic infections, contributing to either protection against infection or pathogenesis (181–183). These differential effects are likely dependent on the levels of type I IFNs; low levels at early stages of infection mediate cell-mediated immunity, but high levels cause immunosuppression (182). Thus, it is not surprising that type I IFNs produced during malaria infection also induce a wide range of cellular effects involving a complex interplay between various cell types.

In addition to producing type I IFNs, DCs produce a wide range of pro-inflammatory cytokines, including TNF-α, IL-12, and IL-6, and chemokines, such as CXCL1, CXCL2, CCL2, CCL5, CXCL9, and CXCL10 in response to malaria parasites, and play crucial roles in malaria immunity and pathogenesis (34, 38, 108, 184–186). Type I IFNs prime DCs for efficient cytokine and chemokine production and activate NK, NKT, γδ T, and T cells to induce IFN-γ and other inflammatory responses (182). IL-12 produced by DCs activates NK cells to induce the production of IFN-γ, which promotes Th1 and effector T cell responses (187). The augmented production of IFN-γ contributes to an efficient parasitemia control by priming Mφs and neutrophils for increased phagocytic activity and thus parasite clearance (188, 189). IFN-γ also contributes to cerebral and other severe malaria clinical conditions under certain situations, such as parasite sequestering in vital organs (187, 188, 190). Chemokines, on the other hand, promote the recruitment of immune cells to mount effective cell-mediated anti-parasitic effects (191). However, these responses also contribute to severe pathology (37, 192, 193). By and large, the initial innate immune responses are mainly aimed at controlling parasite growth by potentiating antiparasitic cell-mediated immunity. However, since pro-inflammatory responses contribute to pathogenesis (14, 33–38), as infection progresses, the function of DCs switches from pro-inflammatory and Th1-inducing to anti-inflammatory and Th2-inducing phenotypes (39, 40, 194). Eventually, balanced pro- and anti-inflammatory and Th1/Th2 responses prevent pathogenesis and promote protective humoral immunity to malaria (39–42); imbalanced responses contribute to pathogenesis. Thus, innate immune responses contribute to either protective immunity or pathogenesis in a context dependent manner.

DNA and RNA play prominent roles in mouse malaria immunity and pathogenesis; under certain conditions TLR2 and TLR4 also play important roles. As such, studies in various mouse models have demonstrated that TLR9, TLR7, TLR4, and TLR2 play important roles in malaria immunity and cerebral, placental and other severe malaria pathology (43, 44, 81, 102, 103, 106–108, 124, 128, 174, 195–201). Gene polymorphism studies in endemic areas that assessed the role of TLRs in malaria have linked TLR9, TLR4, and TLR2 to either susceptibility or resistance to malaria (202–214). While the involvement of TLR9, TLR4, and TLR2 in malaria immunity/pathology is evident from studies in both mouse malaria models and in humans (43, 106, 195, 197, 198, 200–203, 205, 210), thus far none of the studies in endemic areas have revealed the association of TLR7 with human malaria immunity/pathology. It is unclear whether studies in endemic areas that have assessed association of TLRs have investigated the role of TLR7 in malaria susceptibility/resistance. However, since P. falciparum RNA appears to have very low immunostimulatory activity, it is possible that TLR7 has either minor or no role in human malaria. Accumulated evidence indicates that TLRs prominently contribute malaria immunity and pathology, whereas other signaling mechanisms such as inflammasome and adherence-mediated signaling may play minor roles (172, 215).

The contribution of pro-inflammatory responses produced by the innate immune system to protective immunity and pathogenesis is primarily context dependent. In non-cytoadherent parasites, such as P. yoelii 17XNL strain and P. chabaudi that exhibit slow growth rates during the initial phase of infection, pro-inflammatory responses are protective by facilitating anti-parasitic effector function and cell-mediated immunity. In contrast, in the case of cytoadherent parasites, such as P. berghei ANKA that sequesters in brain, lungs, liver and adipose tissues, and P. berghei NK65 that sequesters in lungs and liver, pro-inflammatory responses are pathogenic by promoting cytotoxicity in effector cells, which cause organ damage. In this regard, the observations made in mouse models parallel those of vivax and falciparum malaria in humans. In P. falciparum infection, strong pro-inflammatory responses contribute to an effective control of parasitemia which otherwise results in fulminant infection (41, 42). However, despite this beneficial effect, pro-inflammatory responses promote effector cell function, which contributes to organ damage and severe illnesses. In contrast, in the case of P. vivax, which is relatively slow growing and does not cytoadhere except at low levels in lungs and the placenta (125), despite causing a number of clinical conditions, fatal illnesses are relatively rare compared to P. falciparum.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.