Extracellular vesicles have been studied extensively in last decade in the context of malaria. They were first described as a potential factor in malaria pathogenesis especially cerebral malaria, a lethal severe symptom. The number of extracellular vesicles were increased in human samples, murine model as well as in vitro studies indicating that these cell-derived vesicles play important roles in the disease [13, 14, 18–22, 25]. EVs cargo various biomolecules such as proteins, nucleic acids (DNA, RNA, and miRNAs) participating in cell-to-cell communication both within species and cross talk between human host and malaria parasites [25–29, 41, 44, 58].
In the present study, we isolated EVs from patients from Thailand presenting with uncomplicated malaria and demonstrated that some miRNAs can be detected in higher abundance in these EVs. The selected miRNAs included hsa-miR–451a, hsa-miR–15b–5p, hsa-miR–16–5p, hsa-let–7a–5p and hsa- miR–150–5p that were previously analysed in the context of malaria. These miRNAs were analysed by qRT-PCR. We selected hsa-miR–451a as endogenous control in this study as its Cq values were the most stable among all groups, according to its standard deviation value. Moreover, previous studies demonstrated that hsa-miR–451a was highly expressed in both uninfected and parasitized red blood cells [35] and its expression was independent from the intraerythrocytic development of the malaria parasite [36]. Also, recent studies were able to detect hsa-miR–451a in EVs generated in in vitro cultures of P.falciparum [44–46]. In contrast, Chamnanchanunt et al. found that hsa-miR–451a was down-regulated in P.vivax-infected patient plasma [47]. An interesting study by Lamonte et al. showed that this miRNA was highly expressed in the red blood cell with hemoglobin S (sickle cell) and also showed that it translocated to the parasite resulting in parasite growth inhibition [41]. Other studies demonstrated that EVs cargo hsa-miR–451a could be internalized to target the parasites and diminish the parasite burden [41, 46]. Furthermore, in an in vitro study, red blood cell derived EVs containing hsa-miR–451a and human Argonaute 2 (Ago2) were shown to be internalized by endothelial cells. This miR-Ago2 could down regulate the expression of CAV–1 and ATF2 resulting in endothelial cells alteration which is a plausible factor contributing vascular dysfunction in cerebral malaria [45]. As various studies showed that EVs that cargo hsa-miR–451a could be taken up by various cells in the context of malaria. It would therefore be interesting to enumerate the number of EVs and compare it with the change in abundance of hsa-miR–451a.
We also analysed the relative expression of hsa-miR–16–5p and found no change in the abundance of this miRNA in both EVs isolated from P.vivax-infected patients and P.falciparum-infected patients when compared to uninfected individuals. This is consistent with two previous studies that hsa-miR–16–5p are highly expressed in both P.falciparum-infected erythrocytes and normal erythrocyte. Thus, the relative expression of this miRNA might not be modulated during infection. However, our results differ from a study that found down-regulation of hsa-miR–16–5p in P.vivax-infected patients [47]. Interestingly, a study from our group using an experimental cerebral malaria model (P.berghei ANKA) found an up-regulation of miR–16 in plasma-derived microvesicles [50] suggesting that further studies are needed to elucidate the expression of this miRNA during complicated and uncomplicated malaria infection. Indeed, it is important to note that in our current study none of the P.falciparum-infected patients were diagnosed with severe or cerebral malaria, and were all classified as uncomplicated malaria. Next, we determined the relative expression of hsa-let–7a which was found in infected erythrocytes [36]. The relative abundance of hsa-let–7a was significantly increased in both P.vivax-infected and P.falciparum-infected patients compared to uninfected controls. Our results are in accordance with previous studies showing that this miRNA could be detected in EVs and might be derived from parasitized red blood cells [44, 45]. It is noteworthy that the overall expression of this miRNA is higher in the malaria-infected patients (regardless of the parasite) than in uninfected donors. Several studies demonstrated that hsa-let–7a plays a role in host-parasite interaction. This miRNA in complex with Ago2 [44] could be detected in P. falciparum andpotentially targets the Plasmodium gene: Rad54 [40]. In addition, hsa-let–7a expression was low in erythrocyte-derived EVs in a P.falciparum in vitro experiment but another member in the let7 family: hsa-let–7b was higher in the EVs fraction compared to infected or uninfected erythrocytes. Interestingly, miRNAs profiling from in vitro P.falciparum infected EVs also showed that hsa-let–7a, hsa-let–7i, hsa-let–7g and hsa-let–7f were highly expressed [45]. Some of these miRNAs might derive from parasitized red blood cells as they were found highly enriched in previous analyses [36]. In addition, a study in experimental cerebral malaria demonstrated an increased expression of let–7i in brain tissues which might link to cerebral malaria pathogenesis [42]. This implies that miRNAs within the let–7 family might play a role in the parasite biology, malaria pathogenesis and further studies should be done to elucidate their functions.
Plasmodium vivax is the most widely distributed human malaria parasite. To date, several clinical complications have been reported in P.vivax-infected patients such as severe thrombocytopenia, acute renal failure, acute respiratory distress syndrome as well as cerebral malaria [59, 60]. Moreover, it can cause relapse infection as it can “hibernate” in the patient’s liver in the hypnozoite stage. Later, this dormant parasite will re-invade and cause disease in patients. Understanding the biology of this species is yet needed to improve diagnosis especially of latent parasites and drug development. Studying the role of EVs, miRNAs and EVs-derived miRNAs in the context of P.vivax infection could be a way to improve these parameters. In the present study, we found an up-regulation of two miRNAs, hsa-miR–15b–5p and hsa- miR–150–5p in the EVs-derived plasma in the P.vivax-infected patients but not in the P.falciparum-infected patients, warranting further work on these miRNAs. An analysis of miRNAs from whole blood of adult imported falciparum malaria showed a down regulation of hsa- miR–150–5p [51]. However, the authors did not analyse samples from vivax malaria. Kaur et al. have identified a potential biomarker for differential diagnosis between uncomplicated and complicated vivax malaria, hsa-miR–7977. In addition, an increase in this miRNA was predicted to be involved in malaria pathogenesis through the Transforming Growth Factor Beta (TGF-β) signaling pathway [51]. Recently, a study in non-human primate (Aotus lemurinus lemurinus) confirmed that bone marrow is an important reservoir for gametocytogenesis and proliferation of P.vivax [61]. A study in bone marrow aspirate of human diagnosed with P.vivax showed an aberrant expression of miRNAs in CD71 positive erythroid cells during infection, hsa-miR–150 and hsa-miR–16 were down-regulated while hsa-miR–144 was increased. However, hsa-miR–451a, the most abundant erythroid miRNA, remained stable [62]. In experimental cerebral malaria (ECM), an investigation in brain tissue found a higher expression of miR–150 while in the microvesicles from ECM mice showed non statistically different change of the miR–150 [42, 50]. Nonetheless, our study only investigated the plasma from the patients before drug administration and we did not follow the clinical manifestation after treatment. Taken together, these data suggest that hsa-miR–150–5p might be an essential miRNA involved in malaria infection.
In this study, the relative expression of hsa-miR–15b–5p was higher in EVs from P.vivax-infected patients. To date, there is no evidence about the altered expression of this miRNA which might link to vivax malaria. However, this miRNA is present in P.falciparum in vitro and its abundance was decreased following infection. Also, it was potentially predicted to form a chimeric fusion with ring-infected erythrocyte surface antigen (RESA) [41]. Even though the relative expression of this miRNA was not statistically significant in P.falciparum-infected group, we noticed that it was slightly higher when we compared it with the uninfected group. We hypothesized that we could detect a higher expression of hsa-miR–15b–5p in the EVs as a result of red blood cells vesiculating during infection. Mammalian Target of Rapamycin (mTOR) has been described as a master regulator of an innate and mainly adaptive immunity. By targeting mTOR, an accumulation of parasitized red blood cells and lymphocytes in brain was reduced in experimental cerebral malaria as a rapamycin might block CD8+ T cell differentiation [63]. A study in P.vivax-infected patients showed a decrease in TH17 population while regulatory T cell (Treg) remained stable [64]. Interestingly, miR–15b/16 could down regulate the mTOR and induce Treg cells differentiation [65]. Additionally, a study in multiple sclerosis showed miR–15b could suppress TH17 differentiation and reduced the disease pathogenesis [66]. This suggests a role for hsa-miR–15b in immune regulation, nevertheless, the function of hsa-miR–15b is yet to be understood. Importantly, EVs cargo biomolecules that could modulate recipient immune cells, suggesting a potential role for EVs in the malaria immunity [26, 30]. To explain our result, the rising abundance of hsa-miR–15b–5p in the vivax patients might serve as an immune regulator in response to the infection and might have a protective effect on the disease pathogenesis. We therefore suggest a further functional determination of this miRNA in association with malaria biology and pathobiology.
In order to provide knowledge about dysregulated miRNAs of interest, target prediction was determined. Genes targeted by hsa-miR–150–5p, hsa-miR–15b–5p and hsa-let–7a–5p associate with liver stage of the parasites (HGF, HSPG, LRP),, Toll-like receptor signaling pathway (TLR2/4, MYD88), immunological interaction (IL18, IL10, IL6, TGF-β, GCSF, TNF-α) and cell adhesion molecules (CD36, LFA1, ICAM1, TSP, CD40L, CD40) which might be involve in cerebral malaria. Targeted genes were searched against KEGG pathway database. Several pathways might be important in the context of malaria. Importantly, the adherens junction and the transforming growth factor (TGF)-β were found enriched by the targeted genes of 3 dysregulated miRNAs. An earlier study also demonstrated that up-regulated of miR–19a–3p and miR–19b–5p were associated with TGF-β pathway [43]. However, our study did not include these miRNAs. It is therefore interesting for the next study to include these miRNAs to define the combinatorial effect on the pathway. TGF- β is an anti-inflammatory cytokine that have been described in many studies in the context of malaria. Serum TGF- β level was lower in the falciparum malaria patients, particularly in severe malaria groups and its level had an impact on disease severity as well as clinical outcome [67–70]. These are consistent with studies in murine malaria model that demonstrated the inverse correlation between TGF- β levels, timing of release and severity of the disease such as cerebral complications [71–75]. In addition, a very recent study in controlled human malaria infection (CHMI) demonstrated that serum levels of this anti-inflammatory cytokine were significantly decreased in CHMI and simultaneously coincided with elevations of the level of D-dimer, Von Willebrand factor, IL–6, IFN-γ, numbers of platelets (thrombocytopenia) but not parasitaemia [76]. It is important to note that thrombospondin, a major activator of TGF- β (THBS1, THBS2)[77], was possibly modulated by up-regulated miRNAs in this study. This suggests the potential relationship among up-regulated miRNAs, thrombospondin level, and TGF- β.
Another enriched pathway is adherens junctions that are possibly regulated by hsa-miR–150–5p, hsa-miR–15b–5p and hsa-let–7a–5p. Blood-brain barrier (BBB) is a vital compartment of central nervous system as it separates the CNS from surrounding environment. Adherens junctions in endothelial cells participate to the forming and maintaining of the integrity of the BBB. A number of studies also showed that some miRNAs might regulate this type of junction. For example, the down regulation of Vascular Endothelium Cadherin (VE-Cadherin) was affected by overexpression of miR–101 and this lead to HIV-associated neurological disorder [78]. Also, the overexpression of miR–142–3p repressed the expression of VE-Cadherin and impaired vascular integrity in zebrafish [79]. Similarly, in murine experimental cerebral malaria, the authors postulated the roles of overexpressed miR–19a–3p and miR–19b–5p in this pathway as well [43]. Knowing the roles of miRNAs in the context of malaria particularly cerebral malaria pathogenesis is paramount as it might lead to an adjunctive therapy. For instance, inhibition of miR–27 could prevent vascular leakage associated with ischemia [80]. However, no study of the dysregulated miRNAs analysed in our work, which are in association with this pathway, has been performed in human malaria. More studies are therefore needed to fill this gap.
Circulating miRNAs have been studied and proposed as diagnostic biomarkers in many infectious diseases including malaria. Most studies on infectious diseases have detected human miRNAs such as those in tuberculosis [81], hepatitis B [82], schistosomiasis [83] while some studies investigated microbial miRNAs as biomarkers as well [57, 84–86]. Despite the fact that Plasmodium spp. lacks RNA interference machinery and its own miRNAs [35, 39], human derived miRNAs were demonstrated as promising biomarkers [47, 51]. For example, in human malaria, hsa-miR–16 and hsa-miR–451a were proposed to be biomarkers for P.vivax infection diagnosis [47]. In our study, we evaluated the potential of miRNAs isolated from EVs in malaria patients for the first time. The ROC and AUC were performed. The calculated p-values from AUC analysis indicated that hsa-miR–150–5p, hsa-miR–15b–5p might be used as biomarkers for vivax malaria whereas hsa-let–7a–5p might be used to test for both vivax and falciparum malaria. However, the sensitivity and specificity were not much higher. We suggest further analysis of these miRNAs because the number of patient samples were relatively low in this study. Furthermore, as mentioned earlier, our study selected the miRNAs based on previous studies that investigated these miRNAs in the context of malaria. A more in-depth study is needed if we aim to develop new biomarkers. For instance, profiling miRNA using microarrays or next-generation sequencing will allow an evaluation of all miRNAs present in the EVs. Also, these miRNAs should be analysed in the patients after they recover from the disease. It might be useful in the context of vivax malaria as this species can cause relapse infection. In addition, these markers should be further analysed and compared in the different groups of falciparum malaria patients such as uncomplicated and severe malaria. These will be useful if patients can be early predicted the chance of developing severe malaria beforehand.