Elimination of hepatic rodent plasmodium parasites by amino acid supplementation

iScience Article Elimination of Hepatic Rodent Plasmodium Parasites by Amino Acid Supplementation Patrı ´cia Meireles, Daniela Bra ´s, Diana Fontinha, A ˆngelo F. Chora, Karine Serre, Anto ´nio M. Mendes, Miguel Prude ˆncio antoniomendes@medicina. ulisboa.pt (A.M.M.) mprudencio@medicina. ulisboa.pt (M.P.) HIGHLIGHTS RKV supplementation leads to the elimination of hepatic P. berghei parasites An overall activation of the innate immune system mediates parasite elimination MyD88 is a key player in the elimination of hepatic P. berghei parasites Meireles et al., iScience 23, 101781 December 18, 2020 ª2020 The Authors. https://doi.org/10.1016/ j.isci.2020.101781 ll OPEN ACCESS iScience Article Elimination of Hepatic Rodent Plasmodium Parasites by Amino Acid Supplementation Patrı ´cia Meireles, 1 Daniela Bra ´s, 1 Diana Fontinha, 1 A ˆngelo F. Chora, 1 Karine Serre, 1 Anto ´nio M. Mendes, 1, * and Miguel Prude ˆncio 1,2, * SUMMARY Plasmodium parasites, causative agents of malaria, scavenge host nutrients to sustain their intracellular replication. Modulation of the host’s nutritional status can potentially help control infection by limiting the parasite’s access to nutrients, or by boosting the immune system. Here, we show that dietary supplementation of mice employing a combination of arginine (R) with two additional amino acids, lysine (K) and valine (V), termed RKV, significantly decreases Plasmodium liver infection. RKV supplementation results in the elimination of parasites at a late stage of their development in the liver. Our data employing genetic knockout mouse models and in vivo depletion of specific cell populations suggest that RKV supplementation boosts the host’s overall innate immune response, and that parasite elimination is dependent on MyD88 signaling in immune cells. The immunostimulatory effect of RKV supplementation opens a potential role for dietary supplementation as an adjuvant for prophylaxis or immunization strategies against Plasmodium infection. INTRODUCTION Malaria is an infectious disease that remains a major cause of morbidity and mortality worldwide, for which new cost-effective interventions are urgently needed (WHO, 2019). Plasmodium parasites, the causative agents of malaria, are transmitted by female Anopheles mosquitoes as sporozoites, which are deposited under the mammalian host’s skin and home to the liver through the circulatory system. After traversing several cells, sporozoites productively invade hepatocytes, inside which they develop into exoerythrocytic forms containing thousands of merozoites. The end of the liver stage of Plasmodium infection is marked by the release of these newly formed parasites into the bloodstream, where they invade red blood cells, and initiate the symptomatic, erythrocytic stage of the disease (Prudencio et al., 2006). Numerous studies suggest that poor nutritional status or nutrient deficiencies increase a population’s vulnerability to infections (Schaible and Kaufmann, 2007;Jones and Berkley, 2014). That is also the case for malaria, for which it is well established that host deficiencies in several micronutrients (e.g., vitamin A and zinc) can exacerbate malaria, and that modulating parasite access to other nutrients, such as glucose, vitamin B5, and choline, can have a significant impactonparasitegrowthand,consequently,ondisease (Kirk and Saliba, 2007;Mancio-Silva et al., 2017;Counihan et al., 2017;Shankar, 2000;Caulfield et al., 2004). Dietary supplementations employing various nutrients, such as Coenzyme Q10, Vitamin C, Vitamin D, iron, Arg, tetrahydrobiopterin (BH4), or folate, among others, have been shown to directly impact Plasmodium erythrocytic stages (Nyariki et al., 2019;Qin et al., 2019;Wu et al., 2018;Castberg et al., 2018;Goheen et al., 2017;Awasthi et al., 2017;Alkaitis and Ackerman, 2016;Meadows et al., 2015). Interestingly, cysteamine has been shown to potentiate the activity of anti-malarial drugs, like artemisinins (Moradin et al., 2016), opening a potential new pathway to using nutrient supplementation to improve malaria treatment. Despite numerous studies to understand how different nutrients may affect Plasmodium infection, their usefulness as modulators of disease remains largely unexplored. Conversely, little is known about the effects of dietary supplementation on the liver stage of Plasmodium infection. Dietary supplementation of n-3 fatty acids in the form of fish oil has been shown to inhibit P. berghei hepatic development (Vreden et al., 1995). Also, the administration of a high-fat diet to mice highly impaired Plasmodium liver infection leading to parasite elimination, an effect associated with increased expression of oxidative stress-related genes (Zuzarte-Luis et al., 2017). Interestingly, iron 1 Instituto de Medicina Molecular Joa ˜o Lobo Antunes, Faculdade de Medicina, Universidade de Lisboa, Av. Prof. Egas Moniz, 1649-028 Lisbon, Portugal 2 Lead Contact *Correspondence: antoniomendes@medicina. ulisboa.pt (A.M.M.), mprudencio@medicina. ulisboa.pt (M.P.) https://doi.org/10.1016/j.isci. 2020.101781 iScience 23, 101781, December 18, 2020 ª2020 The Authors. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 1 ll OPEN ACCESS supplementation has yielded contradictory results in what concerns its impact on Plasmodium liver infection. While one study has suggested that it promotes hepatic parasite development (Goma et al., 1996), another, more recent, study reported a hepcidin-dependent decrease in hepatic parasite numbers following iron supplementation (Ferrer et al., 2016). Thus, a more comprehensive understanding on the impact of dietary alterations on the liver stage of Plasmodium infection is clearly warranted. Arg (R) is involved in many metabolic pathways, including the synthesis of nitric oxide (NO), which plays an important role in the killing of invading pathogens, and the synthesis of polyamines via the arginase pathway, which, in turn, can support pathogen growth (Wanasen and Soong, 2008;Das et al., 2010). The competition between these two pathways has been shown to dictate the outcome of infections by Trypanosoma spp., Leishmania spp.,Toxoplasma gondii,Shistosoma mansoni,Candida albicans,Helicobacter pylori, and Plasmodium spp. (reviewed in (Das et al., 2010;Phillips, 2018)). Arg is the only amino acid-based dietary supplementation that has been evaluated in the context of malaria. Its administration was reported to increase the circulating levels of Arg in P. berghei-andP. yoelii-infected mice, leading to enhanced NO production (Martins et al., 2012;Zhu et al., 2012). This, in turn, was shown to significantly impact the pathology associated with the blood stage of Plasmodium parasites, reversing cerebrovascular constriction in P. berghei-infected mice displaying signs of experimental cerebral malaria (Ong et al., 2018). Yet, while some studies employing P. yoelii-infected BALB/c and P. berghei-infected C57BL/6 mice report a significant decrease in parasitemia and an improvement in the survival of the animals following Arg supplementation (Zhu et al., 2012;Ong et al., 2018), these effects were not observed in another study employing the latter model (Martins et al., 2012). Therefore, although Arg supplementation has been shown to be beneficial for some aspects of the malaria pathology, the role for this amino acid in the context of the disease remains to be clearly defined. Our previous studies have shown that Arg uptake plays an essential role in the Plasmodium parasite’s intrahepatic development and maturation (Meireles et al., 2017). In the liver, Arg is taken up by the infected hepatocytes through the host cell’s SLC7A2-encoded transporters and is metabolized primarily by the parasite’s own arginase pathway to secure the biosynthesis of polyamines which are crucial for its development (Meireles et al., 2017). This observation led us to hypothesize that the liver stage of Plasmodium infection might be impaired by a dietary supplementation that would significantly alter the parasite’s metabolism of Arg. To investigate this, we aimed at blocking the polyamine synthesis pathway of both the parasite and the host cell by providing the amino acids lysine (Lys, K) and valine (Val, V), which are well-known inhibitors of the arginase enzyme (Hunter and Downs, 1945). The blockage of Arg metabolism for polyamine synthesis is expected to channel the use of this amino acid as a substrate of iNOS, boosting NO production and potentially impacting Plasmodium development in the liver. We, therefore, designed an amino acid supplementation regimen named RKV, which combines Arg, R with Lys, K and Val, V and employed the rodent P. berghei parasite in combination with different mouse strains to investigate RKV’s impact on Plasmodium liver infection. Our results show that RKV supplementation leads to a significant elimination of hepatic Plasmodium parasites, likely through the action of the innate immune system, and in an MyD88-dependent manner. RESULTS RKV Dietary Supplementation Impairs P. berghei Hepatic Infection To assess the possibility of modulating hepatic infection by Plasmodium through dietary supplementation, we sought to increase the bioavailabilty of Arg (R) as a physiological substrate for the synthesis of nitric oxide (NO), which is a key mediator of immune responses (Lee et al., 2017;Roth, 1992;Bogdan, 2001). To achieve this, we supplemented the drinking water of C57BL/6J mice with 2.5% (w/v) of Arg and the arginase inhibitors Lys (K) and Val (V), either individually or in combinations of equal concentrations (RV, KV, RK, and RKV). Mice were provided with supplemented water ad libitum for 4 weeks, while non-supplemented sterilized water was provided to control (Ctrl) mice, following which all animals were infected by intravenous (iv) injection of luciferase-expressing rodent P. berghei sporozoites. Our quantitative real-time polymerase chainreaction(qRT-PCR)resultsshowthatRKV supplementation significantly decreased P. berghei liver load 46 hr post-infection (hpi) by 65 G31%, while supplementation with either the individual components or with any combination of two of the amino acids that make up the RKV formulation did not have a significant impact on liver infection (Figure 1A). This observation is in complete agreement with our bioluminescence analysis of infected mouse livers, which indicated a 63 G37% decrease in the hepatic load of RKVll OPEN ACCESS 2iScience 23, 101781, December 18, 2020 iScienc e Article supplemented mice relative to untreated controls (Figures S1A and S1B). Of note, we also showed that RKV supplementation impacts liver infection by P. yoelii, another rodent malaria parasite, to an extent similar to that observed for P. berghei (70 G12% decrease; Figure S1C), indicating that this phenotype is not speciesspecific, at least among rodent malaria parasites. When P. berghei infection was allowed to proceed to the blood, no significant differences between RKV-supplemented and control mice were observed in terms of pre-patency time or survival from experimental cerebral malaria (Figures S1D and S1E), as expected from a <90% difference in liver parasite load between the two groups of mice (Siddiqui et al., 2015). Having established that RKV supplementation can significantly impact Plasmodium liver infection, we then sought to determine the minimum period of dietary supplementation required for this effect to be observed, by varying the duration of supplementation prior to infection. Our results show that one week of RKV supplementation is sufficient to observe a significant decrease in hepatic infection by P. berghei (42 G32%), an effect that is even more pronounced after 4 weeks of dietary supplementation (68 G 25%), and which appears to plateau thereafter (53 G28% reduction in comparison to non-supplemented controls after 6 weeks of supplementation; Figure 1B). RKV Dietary Supplementation Leads to the Elimination of Late Liver Stage Parasites Having shown that RKV leads to a marked decrease in the liver load of P. berghei-infected mice, we asked whether this reduction resulted from a decrease in the number of hepatic parasites and/or an impairment of their intra-hepatic growth. To evaluate both possibilities, liver sections from P. berghei-infected Ctrl and RKV-supplemented mice were collected 46 hpi and analyzed by immunofluorescence microscopy. Our results show a marked decrease in the number of parasites per liver area of RKV-supplemented mice relative to controls (Figure 2A), as well as a smaller but statistically significant reduction in parasite size (Figure 2B). The decreased number of parasites suggests that RKV supplementation may lead to either a decrease in hepatocyte invasion by the parasite, or to an elimination of parasites developing in the liver of RKV-supplemented mice. To investigate this, livers from Ctrl and RKV-supplemented mice were collected, and parasite load was assessed by qRT-PCR at different times following injection of P. berghei sporozoites. Our results show that the establishment of infection in the liver of RKV-supplemented mice is indistinguishable from that of Ctrl mice, indicating that the parasite’s ability to invade and infect hepatocytes is not affected by Figure 1. RKV Supplementation Increases Mammalian Host Resistance to Plasmodium Liver Infection (A) The drinking water of C57BL/6J WT mice was supplemented with 2.5% (w/v) of single amino acids Arg (R), Lys (K), and Val (V) (left), or with different combinations of two (RK, KV, and RK) or three (RKV) of the same amino acids (right), for 4 weeks prior to infection with P. berghei sporozoites. Liver parasite load was assessed at 46 hpi by qRT-PCR. Pool of R3 independent experiments. (B) The drinking water of C57BL/6J WT mice was replaced by RKV-supplemented water on the day of P. berghei sporozoite injection (0 d), or 1 day, 1 week, 4 weeks, or 6 weeks before infection. Forty-six hpi, livers were collected and liver parasite load was assessed by qRT-PCR. Mice drinking non-supplemented water were used as controls. Pool of 2–5 independent experiments. Statistical significances assessed by Kruskal-Wallis with post-test Dunn applied in (A) and Oneway ANOVA with post-test Dunett in (B) with *p < 0.05 and ***p < 0.001. Significant differences are indicated in yellow. See also Figure S1. ll OPEN ACCESS iScience 23, 101781, December 18, 2020 3 iScienc e Article this dietary supplementation. Notably, parasite load in the livers of RKV-supplemented mice is lower than that of Ctrl mice only from 42 hpi onward, suggesting that dietary supplementation leads to the elimination of liver parasites at a late stage of their hepatic development (Figure 2C). Impairment of Liver Stage Development Is due to a Direct Effect of Lys (K) Dietary Supplementation on P. berghei Parasites In order to investigate the direct impact of amino acid supplementation on hepatic Plasmodium parasites, Huh7 cells, a human hepatoma cell line, were infected with luciferase-expressing P. berghei parasites in the presence of high concentration of either the individual components or with the various combinations of amino acids that make up the RKV formulation. Our results show that supplementation of the culture medium with both Lys alone and the KV combination lead to a marked decrease in hepatic infection in vitro, which is stronger than that observed with RKV (Figure 3A). To further investigate this, we infected mouse primary hepatocytes with GFP-expressing P. berghei parasites, which enable independently assessing the number of infected cells and the extent of parasite development inside these cells by flow cytometry (Prudencio et al., 2008). Our ex vivo data clearly show that supplementation with either Lys alone or the KV combination significantly reduces the number of infected hepatocytes at 46 hpi by 63 G18% and 50 G14%, respectively (Figure 3B), while also decreasing intra-hepatic parasite development by 67 G 19% and 65 G19% (Figure 3C). Also of note, supplementation with Arg alone had no impact on the number of infected hepatocytes (Figure 3B) but markedly increased parasite development at 46 hpi (Figure 3C). Collectively, these results indicate that, both in vitro and ex vivo, Lys exerts an inhibitory effect on Plasmodium hepatic infection, whereas Arg enhances development, through direct effects on the parasite. Conversely, in mice, co-supplementation of Arg with Lys and Val, but not with Lys alone, leads to a striking decrease in hepatic parasite numbers (Figure 1A), indicating that the elimination of liver parasites in vivo occurs in a mammalian organism-dependent manner and cannot be explained solely on the basis of its direct effect on the parasite’s metabolism. These observations suggest a potential role for the inflammatory or immune responses on the inhibition of liver infection in the context of RKV supplementation. RKV Dietary Supplementation Does Not Induce Liver Damage or Metabolic Inflammation To assess the impact of RKV supplementation on the health status of the mice and investigate a potential metabolic inflammation of the liver, we started by analyzing several in-life and biochemical parameters in Figure2. RKVSupplementationLeadstotheActiveEliminationofHepaticP. berghei Parasites (A and B) The drinking water of C57BL/6J WT mice was supplemented with the RKV combination for 4 weeks prior to infection with P. berghei sporozoites and the number (A) and size (B) of liver parasites was assessed by immunofluorescence microscopy at 46 hpi. (C) The livers of Ctrl and RKV-supplemented mice infected with P. berghei sporozoites were collected at the indicated timepoints, and parasite liver load was assessed by qRT-PCR. Pool of 2–4 independent experiments with error bars representing SEM. Significant differences established by unpaired t test (A and B) or Kruskal-Wallis with post-test Dunn (C) with ***p < 0.001. Significant differences are indicated in yellow. ll OPEN ACCESS 4iScience 23, 101781, December 18, 2020 iScienc e Article RKV-supplemented and Ctrl mice. Our results revealed no significant differences between the two groups of mice in terms of mouse weight and water consumption, or the array of plasma parameters analyzed (Table 1). Furthermore, histological analyses of liver sections from Ctrl and RKV-supplemented mice revealed no alterations in the liver architecture of the latter (Figure S2A), and similar scores of hepatocellular damage and liver inflammatory cell infiltration for both (Figures S2B and S2C). Finally, we analyzed the potential induction of an oxidative stress response in the livers of Ctrl and RKV-supplemented mice by quantifying 16 oxidative stress-related genes, including heme oxygenase-1 (HO-1, encoded by Hmox1), an enzyme that has been shown to be upregulated during hepatic Plasmodium infection (Epiphanio et al., 2008). Our qRT-PCR data show that the expression of neither of those genes is altered in RKV-supplemented mice in comparison to Ctrl mice (Figure S2D). Overall, our data suggest that a 4-week regimen of RKV dietary supplementation does not bear significant toxicity to the mice or negatively impact their health status. The Effect of RKV Dietary Supplementation on Hepatic Infection Is Mediated by the Host Immune System but Not Dependent on NO Production The rationale for formulating the RKV dietary supplementation arose from the hypothesis that the addition of Lys and Val to an Arg-based dietary supplementation might inhibit arginase activity, consequently channeling the available Arg toward NO production by immune cells, and ultimately leading to parasite elimination. To directly test this hypothesis, we compared the liver infection loads of Ctrl and RKV-supplemented Nos2 / mice, which cannot produce NO via iNOS. Our results show that RKV dietary supplementation leads to a reduction in liver parasite load for Nos2 / mice similar to that observed in WT mice (Figure 4A). We further observed that the expression of iNOS in P. berghei-infected, RKV-supplemented, WT mice is similar to that of their non-supplemented counterparts (Figure S3A). Overall, these data indicate that parasite elimination upon RKV dietary supplementation does not depend on an increase in iNOSmediated NO production. Next, we evaluated whether the observed hepatic parasite elimination could be mediated by an immune response elicited or boosted by the RKV dietary supplementation. First, we assessed the impact of RKV-supplementation on Ifnar /- mice, which lack the type-I interferon receptor, to evaluate the role of type I-IFN innate immune responses in the observed decrease in hepatic parasite survival. This innate response has been shown to peak at around 42 hpi following injection of Plasmodium sporozoites and to control liver infection (Liehl et al., 2014,2015;Miller et al., 2014). Our data showed that the absence of type-I IFN signaling does not abolish the decrease in liver parasite load consistently Figure 3. Lys (K) Supplementation Directly Inhibits Hepatic P. berghei Parasites In Vitro and Ex Vivo (A) In vitro cultured Huh7 cells were infected with luciferase-expressing P. berghei parasites and incubated in medium with the supplementation of either single amino acids, Arg (R), Lys (K), and Val (V), or any combinations of two or three of these ammino acids. Overall infection load was assessed by bioluminescence at 48 hpi. (B and C) Mouse primary hepatocytes were incubated with either single amino acids Arg (R), Lys (K), and Val (V), or any combinations of two or three of these ammino acids, prior to infection with GFP-expressing P. berghei parasites. Flow cytometry analysis was used to quantify the relative proportion of infected cells at 48 hpi by assessing the number of GFP + cells (B) as well as parasite development inside hepatocytes by determining the geometric mean of the GFP signal intensity (C). Pools of 2–3 independent experiments with error bars representing SEM. Significant differences established by Kruskal-Walliswithposttest Dunn with *p < 0.05, **p < 0.01 and ***p < 0.001. Significant differences are indicated in yellow. ll OPEN ACCESS iScience 23, 101781, December 18, 2020 5 iScienc e Article observed in RKV-supplemented mice (Figure 4B). Moreover, similar expression levels for several Interferonstimulated genes (ISGs), namely Ifit1, Ifi44, Usp18, Ifit3 and Irf7, were observed in the livers of Ctrl and RKVsupplemented mice at various time points after infection (Figure S3B).Hence,eventhoughatype-IIFNmediated immune response is active in RKV-supplemented mice, it does not appear to be responsible for the parasite elimination observed. Second, mice that were previously subjected to lethal irradiation, and therefore completely ablated of their immune system (Greenberger and Epperly, 2009;Paix et al., 2018), were supplemented with RKV, in parallel with non-irradiated and non-supplemented Ctrl mice. Our results show that irradiation of RKV-supplemented mice before infection completely abolishes the reduction in liver parasite load observed in supplemented, non-irradiated mice, clearly suggesting an implication of the immune system in parasite elimination upon RKV supplementation (Figure 4C). MyD88 Signaling Is Essential for RKV Dietary Supplementation-Mediated Elimination of Hepatic Parasites To dissect the components of the immune system directly involved in the mechanism of parasite elimination triggered by RKV dietary supplementation, we used a combination of genetic KO mice and specific antibodies to assess the impact of different immune cell populations in the observed phenotype. As most of the cells that compose the innate immune compartment are of the myeloid lineage, we started by employing MyD88 / mice, a mouse strain that lacks a crucial adaptor molecule involved in signal transduction after recognition of pathogens by innate receptors, such as Toll-like receptors (TLRs), which are essential for the function of myeloid cells (Warner and Nunez, 2013;Arnold-Schrauf et al., 2015;Akira and Takeda, 2004). Our results show that, in the absence of MyD88, the reduction in liver parasite load that is typically observed in RKV-supplemented mice is completely abolished, suggesting that MyD88 signaling is essential for parasite elimination (Figure 5A). To ascertain whether the MyD88 signaling process responsible for this effect occurred in hepatocytes or in myeloid cells, we performed a similar experiment employing Alb-Cre.MyD88 f/f and LysM-Cre.MyD88 f/f mice, two mouse strains that lack MyD88 specifically in hepatocytes and in myeloid cells, respectively. Our results showed an impairment of P. berghei hepatic infection in either Alb-Cre.MyD88 f/f or LysM-Cre.MyD88 f/f mice similar to that observed in their WT counterparts (MyD88 f/f mice; Figure S4A). Although these results strongly indicate that hepatocytes do not play Parameter Ctrl RKV p value Mouse weight (g) a 22.48 G1.97 23.45 G0.71 0.3527 (ns) Daily water intake (mL) b 4.36 G0.33 4.41 G0.56 0.9767 (ns) Serum parameters c ALT (U per L) 20.33 G2.31 14.67 G2.52 0.077 (ns) ALP (U per L) 92.60 G20.66 100.73 G29.10 0.070 (ns) AST (U per L) 100.00 G5.00 111.33 G10.79 0.400 (ns) GGT (U per L) 1.33 G0.58 1.33 G0.58 0.792 (ns) Total protein (g per dL) 3.83 G0.21 3.93 G0.38 0.700 (ns) Total bilirrubin (mg per dL) 0.01 G0.01 0.04 G0.02 0.110 (ns) BUN (mg per dL) 43.03 G10.66 44.95 G0.92 1.000 (ns) Creatinine (mg per dL) 0.22 G0.03 0.23 G0.05 1.000 (ns) Table 1. Mouse Weight, Average Water Intake, and Plasma Biochemistry of Ctrl and RKV-Supplemented Mice Data are represented as mean GSD. p values were determined using the non-parametric two-tailed Mann-Whitney test. See also Figure S2. ns, not significant; ALT, alanine aminotransferase; ALP, alkaline phosphatase; AST, aspartate aminotransferase; GGT, gamma-glutamyltransferase; BUN, blood urea nitrogen. a Mouse weight on the day of infection. N = 2 independent experiments. b Average daily water intake per mouse on the fourth week of supplementation. N > 3 independent experiments. c Concentration in the plasma at the time of liver collection (45-46 hpi). N = 3 independent experiments. ll OPEN ACCESS 6iScience 23, 101781, December 18, 2020 iScienc e Article a crucial role in the observed phenotype, a similar conclusion cannot be taken with certitude in the case of the myeloid compartment, as there is evidence that this KO strategy may not be fully efficacious for the various myeloid cell populations (Clausen et al., 1999;Abram et al., 2014). Therefore, we decided to confirm these results by employing alternative methods to deplete the different myeloid cell populations, including phagocytic cells, such as Kupffer cells and macrophages, as well as neutrophils and monocytes. We started by administering liposome-encapsulated clodronate to RKV-supplemented mice in order to completely eliminate Kupffer cells and strongly reduce the presence of monocytes/macrophages in the liver, as shown by the reduction of expression of the Clec4f, CD68 and F4/80 markers in the liver (Figure S4B). Importantly, our results showed that depleting phagocytes does not abolish the reduction in liver parasite load typically observed upon RKV supplementation, suggesting that these cells are not involved in the process of RKV-mediated hepatic parasite elimination (Figure 5B). Next, we assessed the involvement of both neutrophils and monocytes, independently or in combination, on the mechanism of parasite elimination by RKV supplementation. Our results show that neither of these innate immune cells play a critical role in the observed reduction of parasite survival. RKV supplementation of Genista mice, a mouse model that lacks mature neutrophils (Ordonez-Rueda et al., 2012), as well as of mice injected with the monocytedepleting anti-CCR2 antibody, displayed a reduction in liver parasite load similar to that observed in RKVsupplemented WT mice, thus excluding neutrophils and monocytes from playing an essential role in the mechanism of RKV-mediated parasite elimination (Figures 5BandS4C). Furthermore, administration of the anti-Gr-1 antibody confirmed and expanded these results showing that even the simultaneous depletion of both neutrophils and monocytes does not abolish the reduction in liver parasite load observed upon RKV supplementation (Figures 5BandS4D). Collectively, these results suggest a crucial role for MyD88 signaling in the process of RKV-dependent parasite elimination and exclude hepatocytes and the main myeloid cell populations as key players in this mechanism. RKV-Dependent Hepatic Parasite Elimination Results from the Coordinated Action of Various Host Innate Immune Cell Populations Having excluded the involvement of the most abundant myeloid cell populations from the mechanism of parasite elimination by RKV supplementation, we proceeded to investigate the possible involvement of lymphoid cells in this process. To this end, we employed Rag2 / mice, which lack B, T, natural killer (NK) T cells and gd T cells (Shinkai et al., 1992;Cording et al., 2018), and Rag2 / g / mice, which, in addition to these cells, also lack NK cells and innate lymphoid cells (ILCs) (Mazurier et al., 1999;Cording et al., 2018). Our results showed Figure 4. RKV-Dependent Parasite Elimination Is Immune-Mediated but Does Not Rely on NO Production Ctrl and RKV-supplemented mice were allowed to drink ad libitum for 4 weeks, after which they were infected with P. berghei sporozoites and parasite liver load was assessed by qRT-PCR 46 hpi. (A) Nos2 / mice were employed to assess whether NO production via iNOS is involved in RKV-dependent hepatic parasite elimination. (B) Ifnar /- mice were employed to determine whether the RKV-dependent parasite elimination mechanism is dependent on a boost of the type-I IFN response. Pool of >3 independent experiments. (C) WT mice were irradiated with 900 rad one day prior to infection with P. berghei sporozoites. Pool of 2 independent experiments. Error bars representing SEM. Significant differences established by two-tailed Mann-Whitney test (A) or Unpaired t test (B and C) with *p < 0.05 and ***p < 0.001. Phenotype reversion shown in yellow. See also Figure S3. ll OPEN ACCESS iScience 23, 101781, December 18, 2020 7 iScienc e Article that while RKV-supplemented Rag2 / mice display a reduction in hepatic parasite load similar to that observed in RKV-supplemented WT mice, RKV-supplemented Rag2 / g / mice display a liver parasite load similar to that of their non-supplemented Ctrl counterparts (Figure 5C). These results suggest a possible role of NK cells and/or ILCs in the mechanism of RKV-mediated impairment of liver infection, excluding the main adaptive lymphocyte populations from an involvement in parasite elimination by RKV supplementation. Thus, we next assessed the specific contribution of NK cells to this phenotype, through the administration of the depleting antiNK1.1 antibody to RKV-supplemented and Ctrl mice. Our results showed that depleting approximately 90% of the NK cells in the liver (defined as NK1.1 + TCRb  cells; Figure S4E) does not abolish the reduction in the liver parasite load typically observed upon RKV supplementation, excluding NK cells as the sole player in the RKVrelated impairment of hepatic infection (Figure 5D). Figure 5. MyD88 Signaling Is Essential for the Multidimensional Stimulation of the Host’s Innate Immune System Induced by RKV Supplementation All Ctrl and RKV-supplemented mice were allowed to drink ad libitum for 4 weeks, after which they were infected with P. berghei sporozoites and parasite liver load was assessed by qRT-PCR 46 hpi. (A) MyD88 / mice were employed to assess the role of innate immunity and myeloid cells in hepatic parasite elimination. (B) Phagocytes were depleted through the administration of liposome-encapsulated clodronate 2 days before sporozoite injection. The role of neutrophils on RKV-dependent parasite elimination was investigated employing Genista mice, which lack mature neutrophils. Monocytes were depleted by the daily injection of anti-CCR2 antibody from day 2today 1 post-infection. Finally, anti-Gr1 was administered to WT mice 2 hr after sporozoite injection, in order to deplete both neutrophils and monocytes simultaneously. Anti-CCR2: 1 experiment; All others: pools of 3 independent experiments. (C) WT, Rag2 / and Rag2 / ɣ / mice were supplemented with RKV for 4 weeks before infection with P. berghei sporozoites. Rag2 / mice lack all the adaptive lymphoid populations while Rag2 / ɣ / mice also lack NK cells and ILCs, which are innate immune populations. Pools of >3 independent experiments. (D) NK cells were depleted through the administration of anti-NK1.1 antibody, 1 day before infection with P. berghei sporozoites. Pool of 3 independent experiments. (E) Ctrl and RKV-supplemented Rag2 / mice were injected with anti-Thy1.2 antibody, 1 day before infection with P. berghei sporozoites, to deplete ILCs. One experiment. All panels: Error bars represent SEM. Significant differences established by unpaired t test (A, D, and E) or Two-tailed Mann-Whitney test (B and C) with *p < 0.05, **p < 0.01 and ***p < 0.001. Phenotype reversion shown in yellow. See also Figure S4. ll OPEN ACCESS 8iScience 23, 101781, December 18, 2020 iScienc e Article SUPPLEMENTAL FIGURES AND TABLES Figure S1 (Related to Figure 1)– RKV supplementation significantly impacts liver parasite load but does not affect blood stage infection or mouse survival. The drinking water of C57BL6 WT mice was supplemented with the RKV combination for 4 weeks prior to infection with luciferaseexpressing P. berghei (A, B, D and E) or with P. yoelii (C) sporozoites. (A) Liver parasite load was assessed at 46 hpi by bioluminescence. Pool of 3 independent experiments. (B) Representative image of bioluminescence assessment of liver infection in Ctrl and RKV-supplemented mice. (C) Liver parasite load in P. yoelii -infected mice was assessed at 46 hpi by qRT-PCR. One experiment. (D) Blood parasite load in Ctrl and RKV-supplemented mice, assessed by a bioluminescence assay. Representative experiment out of 2 independent experiments. (E) Survival of Ctrl and RKV-supplemented mice after infection. Pool of 2 independent experiments. (A and C) Two-tailed Mann-Whitney, (D) Two-way ANOVA with post-test Bonferroni, (E) Log-Rank Mantel-Cox test. Error bars represent SEM. ** p < 0.01 and *** p < 0.001. Figure S2 (Related to Table 1)– RKV supplementation does not cause liver toxicity or an increased inflammatory state. (A) Representative microphotographs of liver sections from Ctrl and RKV-supplemented mice stained with HE. Insets show a normal aspect and organization of the liver cells. (B) Hepatocellular damage score of livers from Ctrl and RKV-supplemented mice. (C) Hepatic inflammatory cell infiltration score in livers from Ctrl and RKV-supplemented mice at 39 hpi. (B and C) One experiment with 5 mice per group. (D) Expression of several oxidative stress-related genes in whole livers of Ctrl and RKV-supplemented mice at 46 hpi. Pool of 3 independent experiments. Twotailed Mann-Whitney test. Error bars represent SEM. Figure S3 (Related to Figure 4)– The effect of RKV supplementation on liver parasite load is not dependent on increased expression of iNOS or of interferon-stimulated genes (ISGs) (A) Expression of iNOS in whole livers of Ctrl and RKV-supplemented mice at 46 hpi. Pool of 3 independent experiments. Unpaired t-test. (B) Expression of five ISGs in RKV-supplemented and Ctrl mice at different timepoints during P. berghei liver infection. Pool of 2 – 4 independent experiments. KruskalWallis with post-test Dunn’s. Error bars represent SEM. Figure S4 (Related to Figure 5)– RKV-dependent parasite elimination is dependent on the immune system but is not singly mediated by either hepatocytes, macrophages, monocytes, neutrophils, NK cells or ILCs. (A) Alb-Cre.MyD88f/f, which lack MyD88 specifically on hepatocytes, LysM-Cre.MyD88f/f, which lack MyD88 specifically on myeloid cells, and the corresponding littermate control mice (MyD88f/f) were allowed to drink Ctrl and RKV supplemented water before infection with Plasmodium sporozoites. Liver load was assessed at 46 hpi. Pool of > 3 and 2 independent experiments, for Alb-Cre.MyD88f/f and LysM-Cre.MyD88f/f, respectively. (B) The efficiency of the depletion of phagocytes by clodronate administration was assessed at 46 hpi by qRT-PCR, by quantifying the expression of Clec4f (a marker of Kupffer cells), CD68 (a marker of monocytes and macrophages) and F4/80 (a marker of macrophages) in whole livers. Pool of 2 independent experiments. (C) The efficiency of the depletion of monocytes by anti-CCR2 administration was assessed at 46 hpi by flow cytometry. After extraction, liver leukocytes were stained with LIVE/DEAD Fixable Aqua Dead Cell Staining kit, anti-CD11b, anti-CD11c, anti-Ly6C and anti-Ly6G. Monocytes were defined as Ly6C+ Ly6Gcells inside the CD11b+ CD11cpopulation. One experiment. (D) The efficiency of the depletion of neutrophils and monocytes by anti-Gr1 administration was assessed at 46 hpi by flow cytometry. Total liver leukocytes were stained with LIVE/DEAD Fixable Aqua Dead Cell Staining kit, anti-CD11b, anti-CD11c, anti-Ly6C and anti-Ly6G. Neutrophils were defined Ly6G+ Ly6Ccells and monocytes as Ly6C+ Ly6Gcells, both inside the CD11b+ CD11cpopulation. Representative experiment out of 3 independent experiments. (E) The efficiency of the depletion of NK cells by antiNK1.1 administration was assessed at 46 hpi by flow cytometry. Liver leukocytes were stained with LIVE/DEAD Fixable Aqua Dead Cell Staining kit, anti-NK1.1 and anti-TCRβ. NK cells were defined as NK1.1+ TCRβ - cells inside the total live population. Representative experiment out of 3 independent experiments. (F) The efficiency of the depletion of ILCs by administration of anti-Thy1.2 to Rag2-/- mice was assessed at 46 hpi by flow cytometry. Total liver leukocytes were stained with LIVE/DEAD Fixable Aqua Dead Cell Staining kit, anti-CD45, anti-CD3ε, anti-Gr1, anti-CD11b, anti-CD11c, anti-B220, antiTer119 and anti-CD127. ILCs were defined as LineageCD127+ cells inside the live CD45+ population. Lineage: anti-CD3ε, anti-Gr1, anti-CD11b, anti-CD11c, anti-B220 and anti-Ter119. One experiment. Error bars represent SEM. (A, C and F) Two-tailed Mann-Whitney test; (B, D and E) Unpaired t-test. ** p < 0.01 and *** p < 0.001. Table S1 (Related to Figures 1-5)- List of primer sequences used for gene expression quantification. Gene forward primer (5’ - 3’) reverse primer (5’ - 3’) Pb18S AAGCATTAAATAAAGCGAATACATCCTTAC GGAGATTGGTTTTGACGTTTATGTG Hprt TTTGCTGACCTGCTGGATTAC CAAGACATTCTTTCCAGTTAAAGTTG Ifit1 CCTTTACAGCAACCATGGGAGA GCAGCTTCCATGTGAAGTGAC Ifi44 TCGATTCCATGAAACCAATCAC CAAATGCAGAATGCCATGTTTT Usp18 CGTGCTTGAGAGGGTCATTTG GGTCGGGAGTCCACAACTTC Ifit3 CTGAACTGCTCAGCCCACAC TGGACATACTTCCTTCCCTGA Irf7 CTTCAGCACTTTCTTCCGAGA TGTAGTGTGGTGACCCTTGC Hamp CCTATCTCCATCAACAGATG AACAGATACCACACTGGGAA Hmox1 GTCTCTGCAGGGGCAGTATC TGCTCGAATGAACACTCTGG Prdx1 GTTGGCCGCTCTGTGGATGAGAT ATCACTGCCAGGTTTCCAGCCAGC Prdx2 GTTCTCCGGCCTAGGGCTCTCTC GCCGGAGGCCATGACTGCGTG Prdx5 GCTCCGTGCATCGACGTGCT CTCCCACCTTGATCGGGGCCA Prdx6 CACCACGGGCAGGAACTTTGATG TCACGCTCTCTCCCTTCTTCCAGT Nox2 TGCAGTGCTATCATCCAAGC CTTTCTCAGGGGTTCCAGTG Nox4 TCAGGACAGATGCAGATGCT CTGGAAAACCTTCCTGCTGT Sod1 TACTGATGGACGTGGAACCC GAACCATCCACTTCGAGCA Sod2 GCTTGATAGCCTCCAGCAAC ACTGAAGTTCAATGGTGGGG Cat TCAGGGCCGCCTTTTTGCCT ACTCGAGCGCGGTAGGGACA Txrnd1 ATGGACAGTCCCATCCCGGGA GCCCACGACACGTTCATCGTCT Txn1 TGCTACGTGGTGTGGACCTTGC TCTGCAGCAACATCCTGGCAGT Srnx1 AGTAGTAGTCGCCACCCTGG AGAGCCTGGTGGACACGAT Gsta2 TTGAAGTAGTGAAGCACGGG ATTGGGAGCTGAGTGGAGAA Gpx1 CAATGTAAAATTGGGCTCGAA GTTTCCCGTGCAATCAGTTC Gsr ATCGTGCATGAATTCCGAGT GGTGGTGGAGAGTCACAAGC Clec4f TGAGTGGAATAAAGAGCCTCCC TCATAGTCCCTAAGCCTCTGGA Cd68 AGCTGCCTGACAAGGGACACT AGGAGGACCAGGCCAATGAT F4/80 CCCCAGTGTCCTTACAGAGTG GTGCCCAGAGTGGATGTCT TRANSPARENT METHODS Chemicals RPMI 1640, RPMI 1640 without arginine, William’s E, PBS pH 7.4, trypsin, FBS, non-essential amino acids, penicillin/streptomycin, glutamine, HEPES pH 7, liver perfusion medium (LPM) and liver digestion medium (LDM) were purchased from Gibco/Invitrogen. L-arginine hydrochloride, L-lysine and L-valine were purchased from FisherScientific. All other chemicals were obtained from Sigma, unless otherwise specified. Cells Huh7 cells were cultured in RPMI 1640 medium supplemented with 10% v/v FBS, 1% v/v nonessential amino acids, 1% v/v penicillin/streptomycin, 1% v/v Glutamine and 1% v/v HEPES, pH 7 and maintained at 37 °C with 5% CO2. Mouse primary hepatocytes were cultured in William’s E medium supplemented with 4% v/v FBS and 1% v/v penicillin/streptomycin and maintained at 37 °C with 5% CO2. Mice All animals used in this study were housed in the facilities of Instituto de Medicina Molecular João Lobo Antunes (iMM, Lisbon, Portugal), with a maximum of five animals per cage, and free access to water and food. C57BL/6J wild-type (WT) mice were purchased from Charles River Laboratories (L’Arbresle, France). Nos2-/- mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). Ifnar-/-, MyD88-/- and Rag2-/- experimental mice, and AlbCre and LysM-Cre breeders were purchased from Instituto Gulbenkian de Ciência (IGC, Lisbon, Portugal). All other mouse strains were obtained from breedings established at iMM’s rodent facility. Four weeks old male mice were used in all experiments that employed exclusively C57BL/6J WT mice. Both male and female 4 to 8 weeks old mice were used in experiments that employed genetically deficient mice. In the latter experiments, 4 to 8 weeks old WT or littermate male and/or female mice were used to match gender and age of the genetically deficient mice. All animal experiments were performed in strict compliance to the guidelines of iMM´s animal ethics committee (ORBEA) and the Federation of European Laboratory Animal Science Associations (FELASA). In the experiments in which the infection was allowed to proceed to the blood, animals were humanely euthanized at the first behavioral signs of onset of experimental cerebral malaria (ECM). Parasites Sporozoites were isolated from the salivary glands of infected female Anopheles stephensi mosquitoes bred at iMM’s insectary facility, prior to being employed in infections. A GFP/luciferase-expressing P. berghei ANKA parasite line (676m1cl1 line) was used in all experiments (Ploemen et al., 2009), with the exception of the flow cytometry experiments, in which a GFP-expressing P. berghei ANKA parasite line (259cl1 line) was employed (FrankeFayard et al., 2004) and the P. yoelii experiment, in which a GFP-expressing P. yoelii (strain 17XNL) was employed (Ono et al., 2007). Isolation of mouse primary hepatocytes Mouse primary hepatocytes were isolated using a modified two-step perfusion protocol followed by a Percoll purification step, as previously described (Goncalves et al., 2007, Liehl et al., 2014, Meireles et al., 2017). Mice were euthanized by CO2 inhalation and immediately processed for cannulation of the portal vein using a 26-gauge needle, followed by the sectioning of the inferior vena cava (IVC) to allow the fluid to drain. The liver was perfused with liver perfusion medium (LPM), followed by liver digestion medium (LDM). Intermittent clamping of the IVC was performed during LDM perfusion to improve tissue digestion. After digestion, the liver was excised and the cells were liberated, sequentially filtered through a 100 μm and a 70 μm cell strainer and spun at 50×g for 3 min. The pellet was resuspended in William’s Medium E with 10% v/v of FBS, carefully overlaid on a 60% v/v Percoll solution (1:1) and spun at 750×g for 20 min, without break, at 20 °C. Viable hepatocytes deposited in the pellet were washed with William’s E Medium with 10% v/v FBS, spun at 50×g for 3 min and resuspended in complete William’s E Medium (supplemented with 4% v/v FBS and 1% v/v penicillin/streptomycin). Hepatocytes were then plated at a density of 1.0 × 105 in 24-well plates. Viability and yield were assessed by trypan blue staining. In vitro and ex vivo amino acid supplementation In the conditions with physiological concentrations of Arg, the medium of Huh7 cells or mouse primary hepatocytes was replaced by Arg-free medium supplemented with 100 µM Arg. In the conditions in which there is supplementation of Arg, the approximate supraphysiological concentration of this amino acid which is normally present in RPMI, 1 mM, was added. Supplementations with Lys and Val, were performed by adding 20 mM of each of these amino acids to medium with physiological (K, V and KV) or supraphysiological concentrations of Arg (RK, RV, RKV). These concentrations of Lys and Val were chosen because they have been shown to completely inhibit arginase activity in rat primary hepatocytes at physiological concentrations of Arg (Lerzynski et al., 2006). Overall in vitro infection by luminescence Overall hepatic infection was determined by measuring the luminescence intensity in Huh7 cells infected with the above referred GFP/luciferase-expressing P. berghei line, as previously described (Ploemen et al., 2009). Huh7 cells (1.0 × 104 per well) were seeded in 96-well plates the day before infection. Sporozoite addition was followed by centrifugation at 1800xg for 5 min. Medium was replaced approximately 2 hpi by the appropriate medium. Parasite infection load was measured 48 hpi by a bioluminescence assay (Biotium) using a multiplate reader Infinite M200 (Tecan). Cell viability was assessed by the CellTiter-Blue assay (Promega) according to the manufacturer’s protocol. Quantification of P. berghei parasite numbers and development by flow cytometry Intracellular parasite numbers and development were assessed by determining the percentage of GFP+ cells and by measuring the intensity of the GFP signal of mouse primary hepatocytes infected with a GFP-expressing P. berghei line at 48 hpi, as previously described (Prudencio et al., 2008). Primary hepatocytes (1.0 × 105 per well) were infected with 5.0 × 104 sporozoites one day after being plated and the medium was replaced by the appropriate medium 2 h after infection. Cells were collected for flow cytometry analysis at 48 hpi and analyzed on a Becton Dickinson FACSCalibur. Data acquisition and analysis were carried out using the CELLQuest (version 3.1.1 f1, Becton Dickinson) and FlowJo (version 6.4.7, FlowJo) software packages, respectively. In vivo amino acid supplementation The drinking water of 4 to 8 weeks old mice, was replaced by sterilized water containing 2.5% (w/v) L-arginine hydrochloride (FisherScientific), 2.5% (w/v) L-lysine (FisherScientific), 2.5% (w/v) L-valine (FisherScientific) or a combination of the three amino acids (RKV). The mice were allowed to drink ad libitum for 4 weeks, unless otherwise specified. Non-supplemented, sterilized water was provided to Ctrl mice. The supplementation was maintained until completion of the experiment. In vivo treatments C57BL/6J WT mice were lethally irradiated (900 rad) in an Irradiator Gammacell ELAN 3000, one day before P. berghei sporozoite injection. For depletion of phagocytic cells, 200 µL of liposome-encapsulated clodronate (Clodronate Liposomes) were injected i.v. 2 days before infection. Mice injected with 200 µL of liposome-encapsulated PBS were used as controls. Monocytes were depleted by the i.p. injection of 20 µg of anti-CCR2 antibody (clone MC-21; kindly provided by Matthias Mack (Mack et al., 2001)) daily from day -2 to day 1 post-infection. Two hundred and fifty µg of anti-Gr1 (clone RB6-8C5; BioXCell) were injected i.p. 2 h after infection, to deplete both neutrophils and monocytes. For NK cell depletion, 150 µg of antiNK1.1 antibody (clone PK136; BioXCell) were injected i.p. 1 day prior to infection. Finally, to deplete ILCs, Rag2-/- mice were injected i.p. with 200 µg of anti-Thy1.2 antibody (clone 30H12;