Hemozoin‐induced activation of human monocytes toward M2‐like phenotype is partially reversed by antimalarial drugs—chloroquine and artemisinin

Abstract Plasmodium falciparum malaria is the most severe form of malaria with several complications. The malaria pigment‐hemozoin (Hz) is associated with severe anemia, cytokine dysfunction, and immunosuppression, thus making it an interesting target for developing new strategies for antimalarial therapy. Monocytes (MO) in circulation actively ingest Hz released by Plasmodium parasites and secrete pro‐ and anti‐inflammatory cytokines. M1 and M2 types represent the two major forms of MO/macrophages (MQ) with distinct phenotypes and opposing functions. Imbalance in the polarization of these types is reported in many infectious diseases. Though the association of Hz with immunosuppression is well documented, its role in activation of MO in context of M1/M2 phenotypes remains to be addressed. We report here that natural Hz drives human MO toward M2‐like phenotype as evidenced by the expression of M2 signature markers. Hz‐fed MO showed elevated transcript and secreted level of IL‐10, CCL17, CCL1, expression of mannose‐binding lectin receptor (CD206), and arginase activity. Hz attenuated HLA‐DR expression, nitric oxide, and reactive oxygen species production, which are the features of M1 phenotype. Our data also implicate the involvement of p38 MAPK, PI3K/AKT, and NF‐κB signaling pathways in skewing of Hz‐fed MO toward M2‐like type and suppression of mitogen‐stimulated lymphocyte proliferation. Importantly, antimalarial drugs—chloroquine and artemisinin—partially reversed activation of Hz‐induced MO toward M2‐like phenotype. Considering the limitations in the current therapeutic options for malaria, we propose that these drugs may be re‐examined for their potential as immunomodulators and candidates for adjunctive treatment in malaria.


| INTRODUC TI ON
Malaria is one of the leading causes of morbidity affecting around 212 million individuals and 4, 29,000 deaths in endemic regions according to WHO 2015. Plasmodium falciparum causes the most severe form of malaria with complications that include cerebral malaria, pulmonary edema, acute renal failure, or severe anemia (Buffet et al., 2011;Das, 2008;Medana & Turner, 2006;Mohan, Sharma, & Bollineni, 2008). P. falciparum malaria is frequently associated with immunosuppression and increased secondary infections (Hu, 2013;Ihekwereme, Esimone, & Nwanegbo, 2014;Orf & Cunnington, 2015;Troye-Blomberg & Perlmann, 1988). Immunosuppression enhances parasitemia and reduces immune responses to not only malaria parasite proteins but other pathogens as well. Accumulating evidence suggests that a combination of parasite and host factors is involved in the pathogenesis of severe malaria.
MO are heterogeneous innate immune cells that play a pivotal role in primary response to pathogens, tissue homeostasis, inflammation, resolution, and repair (Shi & Pamer, 2011). MO are highly plastic and polarize toward M1 phenotype in the presence of inflammatory environment such as LPS and IFNγ, while the anti-inflammatory M2 phenotype is driven by TH2 cytokines IL-4, IL-13, IL-10, and TGFβ. IL-12-producing M1 phenotype is associated with antiparasitic and tumor-resistance capabilities, and IL-10-producing M2 phenotype is related to wound healing, immune regulation, and resolution of inflammation (Mantovani et al., 2004;Martinez & Gordon, 2014;Sica & Mantovani, 2012).
Multiple signaling pathways including JAK/STAT, PKC/ERK, and PI3K/AKT/mTOR are involved in the maintenance of M2 phenotype.
While accumulating experimental and clinical data support the significant contribution of Hz in immunosuppression, its role in driving the MO toward specific MO phenotypes and the implications therein remains to be unraveled. In the present study, we aimed to (a) investigate the effect of Hz in the activation of human MO toward M2 phenotype, (b) examine the signaling pathways involved in the process, and (c) explore the potential of antimalarial drugschloroquine (CHQ) and artemisinin (ART)-in reversion of Hz-driven activation of MO.

| Chemicals and reagents
Sterile tissue culture plastics were purchased from corning. RPMI 1640 media was purchased from Sigma. The parasite and MO cultures were ensured to be mycoplasma free and endotoxin low (<0.125). Inhibitors SB203580 (SB), parthenolide (PAR), LY294002 (LY), and drugs CHQ diphosphate and ART were purchased from Sigma.
Parasite cultures with >25% rings were synchronized by 5% aqueous sorbitol and cultured for 50 h until the rupture of mature schizonts and release of Hz. Culture supernatants containing Hz were carefully aspirated and collected and centrifuged, and the blackish-brown Hzcontaining pellet was washed three times with sterile ice cold PBS (pH 7.2). Pigment isolated from different batches of parasite cultures was pooled, aliquoted, and stored in PBS at −20°C. The total heme content in isolated Hz was determined as previously described (Polimeni et al., 2012). Hz isolated from the supernatants of synchronized parasite cultures after schizont rupture is mostly devoid of parasite material. P. falciparum DNA was not detected in natural Hz (at concentration used in our experiments) on agarose gel (Supporting Information Figure S1).

| Isolation of MO (CD14 + )
The study for use of human blood MO was approved by the ethics committee of National Centre for Cell Science, Pune, India. Buffy coats were obtained from healthy donors recruited for blood collection at government-recognized blood bank, Pune. Informed consent from donors was not essential as discarded buffy coats obtained after plasmapheresis were used. Peripheral blood mononuclear cells (PBMCs) were isolated from buffy coats by Histopaque 1077. CD14 + MO were enriched from PBMCs by positive selection using magnetic beads (MACS) (Miltenyi Biotec, Germany), according to the manufacturer's instructions. The purity of MO obtained was 85%-90% CD14 + and viability >95% (Supporting Information Figure S2). Adherent MO for phagocytosis was obtained as previously described with some modifications (Polimeni et al., 2012). Briefly, 2 × 10 6 MO/2 ml was seeded per well in six-well tissue culture plates and incubated for 1 h at 5% CO 2 in RPMI 1640 medium supplemented with 25 mM HEPES, 2 mM glutamine, 100 mg/ml streptomycin, 100 units/ml Penicillin G, and 10% fetal bovine serum (Sigma St Louis, MO). Cells were washed with RPMI 1640 medium, and the adherent MO obtained was incubated at 37°C overnight. The adherence step applied after MACS and before phagocytosis further enriched the MO population to contain >95% MO. Culture medium was replaced with fresh medium before the start of experiments.

| Phagocytosis of Hz and in vitro cell culture experiments
Phagocytosis of Hz and latex particles was carried out as previously described (Polimeni et al., 2012). Parasite Hz (50 μg/ml) and latex beads (0.114 μm diameter; 10 μl of a 100-fold dilution of the 2.5% V/V) were exposed to 10 6 MO. The dose of pigment used was based on Hz amounts detected during P. falciparum infection and previous in vitro and in vivo studies (Deshpande & Shastry, 2004;Jaramillo et al., 2005). The latex beads were used as phagocytic control . The plates were incubated in 5% CO 2 incubator at 37°C for 2 h to allow phagocytosis. Un-phagocytized Hz and beads were removed by media washes, and the cells were incubated for indicated time intervals. In cell signaling experiments, the MO were pretreated with different pharmacological pathway inhibitors, SB203580 (10 μM), parthenolide (10 μM), and LY294002 (10 μM) 1 h prior to phagocytosis. Antimalarial drugs-ART (20 μM) and CHQ (20 μM)-were added 2 h after exposing the MO to Hz pigment.

| Flow cytometry
The MO fed with Hz and latex beads were collected after 24 h by supplemented with 2% FBS, and data were acquired on BD FACS Canto II (BD Biosciences) and analyzed using FACS DIVA software and FlowJo software (BD Biosciences). All parameters measured were based on SSC versus MO-specific CD14 + gated cells.

| RNA isolation and cDNA synthesis
The MO fed with Hz/latex were collected after 12 h and stored in Trizol reagent (Life Technologies, Inc., Gaithersburg, MD) at −80°C until RNA isolation. Total RNA was extracted using the Purelink RNA kit (Ambion, USA) following the manufacturer's protocol of Trizol plus Purelink columns. The purified total RNA was resuspended in RNase-free water and quality was analyzed using Nanodrop (ND-1000 spectrophotometer). 600 ng of RNA was reverse transcribed with ImProm-II™ Reverse Transcription System (Promega) as per manufacturer's instructions.

| Quantitative real-time polymerase chain reaction (qPCR)
qPCR analysis of mRNA expression of M1 and M2 phenotype-specific marker genes in Hz and latex fed MO was performed using Power syber green (Thermo Fisher Scientific) and gene-specific primers. The primers related to M1/M2 phenotype for real-time PCR analysis were designed using NCBI primer 3 Software and were custom synthesized from Sigma and IDT technologies. Oligo sequences are listed in Supporting Information Table S1. The cycling parameters were as follows: Initial denaturation at 95°C for 10 min to activate the AmpliTaq Gold DNA polymerase followed by 95°C for 15 s, 60°C for 60 s, and then 72°C for 30 s for a total of 40 cycles. The comparative threshold cycle (C t ) value for GAPDH was used to normalize loading variations in the real-time PCR. The extent of mRNA expression for a given condition was represented by the relative value to the lower DDCT between those estimated for the different conditions: the control DDCT value was fixed at 1.0. The specificity of PCR was confirmed by melting curve analysis. Real-time PCR was done on thermocycler Applied Biosystems 7300 (Applied Biosystems).

| Western blot analysis
The MO fed with Hz were washed with PBS and lysed with equal volumes of RIPA buffer containing 1× protease inhibitor cocktail on ice and then centrifuged for 10 min at 16,000 × g at 4°C. Lysates with equal amounts of protein were separated by 10% SDS-PAGE and then transferred onto PVDF membrane (Millipore, Bedford, MA, USA), blocked with 5% Milk in TBS containing 0.1% Tween 20, and incubated overnight at 4°C with primary antibodies. Phospho-AKT, phospho-P38, phospho-p65 (NF-κB), total p65 (NF-κB) were purchased from Abcam; total P38 from Santa Cruz Biotechnology; AKT from Cell signaling technology, IκBα from Santa Cruz Biotechnology, and GAPDH from Sigma.
The membranes were exposed to HRP-labeled anti-rabbit IgG secondary antibodies (1:5,000) (BioRad) for 1 h at RT and detected by enhanced chemiluminescence detection system and visualized using GE image quant 4000 chemiluminescence system (GE Healthcare Lifesciences).

| Lymphocyte proliferation assay
Adherent MO were prepared and exposed to Hz and latex as described above, and autologous lymphocytes previously col-
The samples were centrifuged at 13,000 × g for 10 min to remove insoluble material. The samples were incubated with substrate buffer containing arginine buffer and manganese (Mn) solution for reaction to occur for 2 h at 37°C along with appropriate sample blank and urea standard in separate wells. The reaction was terminated by adding urea reagent and incubation for 1 h. The absorbance was recorded at 430 nm and arginase activity was calculated as units/L.

| NO estimation
The NO production in the culture supernatants of Hz and latex fed MO was measured as stable end product of NO, that is, nitrite using the Griess reagent. Briefly, 100 μl of cell culture supernatant was added to 96-well plate in triplicates, followed by 50 μl sulphanilamide and 50 μl N-1-napthylethylenediamine dihydrochloride.
The plate was incubated in dark for 15 min at 37°C. Absorbance at 540 nm was measured by microplate reader, and nitrite concentration was estimated using a sodium nitrite standard curve. RPMI1640 media without phenol red dye was used for this assay.

| ROS detection assay
Reactive oxygen species (ROS; including superoxide, hydrogen peroxide, and other reactive oxygen intermediates [ROI]) was estimated by H 2 O 2 -sensitive probe H 2 DCFDA (Thermofisher scientific) by flow cytometry. Briefly, Hz and latex fed MO were incubated with H 2 DCFDA (10 μg/mL) and PE-anti-human CD14 at RT for 20 min in the dark. PE-labeled anti-human CD14 was used to gate MO.
Fluorescent DCF in the MO emitting green fluorescence and proportional to intracellular ROS was acquired on BD FACS Canto II, and data were analyzed using FACS DIVA software.

| Statistical analysis
All values (expressed as M ± SEM) were obtained from three or more independent experiments. Comparison between groups was done with the one-way ANOVA (Bonferroni correction). Statistical comparisons were made as indicated, significance levels: *p < 0.05, **p < 0.01, ***p < 0.001. GraphPad Prism6 software was used for statistical analysis.

| Hz induces M2-like phenotype in peripheral blood-derived-human MO
The effect of Hz on activation of MO toward the M2 phenotype was examined in CD14 + sorted cells and analyzed for a panel of M1-and M2-associated markers. IL-10, a well-characterized antiinflammatory cytokine implicated in M2 polarization, was elevated at transcript level following Hz exposure (Figure 1a). Consistent with our previous work (Deshpande & Shastry, 2004), a strong induction of secreted and cellular IL-10 was observed (Figure 1b and Supporting Information Figure S3a). The transcript and secreted level of IL-12p70 was assessed in Hz-fed MO, and treatment with LPS + IFNγ was used as positive control. Interestingly, Hz phagocytosis did not alter the IL-12p70 level. As expected, the production of IL-12 was increased in MO exposed to LPS + IFNγ (Supporting Information Figure S3b,c). M1 phenotype is also induced by microbial products through TLR ligands (Murray & Wynn, 2011;Murray et al., 2014;Wang, Liang, & Zen, 2014). Therefore, additional experiments were conducted to examine TLR engagement in Hz-driven MO activation. MO exposed to natural Hz were analyzed by flow cytometry and western blotting for expression of TLR9. β-Hematin (sHz) was used as a reference for flow cytometry analysis (Coban et al., 2005). As shown in Supporting Information Figure S4a,b, natural Hz had no effect on TLR9 expression, sug- Results are M ± SEM from at least three independent donors. Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001 in comparison with the control and latex as determined by one-way ANOVA (Bonferroni test) addition, Hz phagocytosis elevated transcript level and secretion of M2b-associated cytokines-TNFα, IL-1 β, and IL-6 (Supporting Information Figure S5a,b). The amount of Hz in the range of 25-75μg/ml significantly elevated the secreted levels of IL-10, CCL17, CCL1, and surface expression of CD206 in MO (Supporting Information Figure S5c). Interestingly, CXCL13 and CD163 (markers related to M2c phenotype) were unaltered under similar experimental conditions (Supporting Information Figure S6).
MO polarization endows subtypes with specific biochemical and functional properties. M1 MO are characterized by elevated production of NO and ROI that are essential for clearance of pathogens.
Conversely, M2 MO lack the ability to generate oxidative intermediates, synthesis of ornithine and polyamines through the arginase pathway. We found that Hz-fed MO robustly elevated arginase activity ( Figure 1f) and significantly reduced the NO (Figure 1g

| Hz facilitates the M2-like phenotype via p38 MAPK, PI3K-AKT, and NF-κB pathways
The role of p38 MAPK, PI3K-AKT, and NF-κB pathways has been implicated in IL-10 synthesis and M2 polarization; however, their relevance in Hz-induced M2-like phenotype is not reported. In this study, exposure to Hz led to a significant surge in the phosphoryla-  (Figure 2f). The levels of TNFα, IL-1β, and IL-6 were markedly reduced with Hz in the presence of SB203580 and parthenolide. In contrast, IL-1β and IL-6 cytokines were further elevated in presence of LY294002 (Supporting Information Figure S8a,b).
LY294002 can act on other pathways along with PI3K kinases that may be responsible for the regulation of IL-6 and IL-1β (Gharbi et al., 2007). To confirm the role of PI3K-AKT pathway, we used another inhibitor-wortmannin in similar experiments. Wortmannin effectively reduced the level of IL-1β, IL-6, and IL-10 in supernatants of MO fed with Hz, signifying the role of PI3K-AKT pathway in inhibiting M2-like phenotype (Supporting Information Figure S8c).  were based on previous studies on MO (Jang, Choi, Byun, & Jue, 2006;Wang et al., 2011). ART in concentrations ranging from 5 to 400 μM did not affect the viability of MO as assessed for cytotoxicity by MTT assay (Supporting Information Figure S9a). In our study, we found that treatment of MO preexposed to Hz with ART resulted in dramatic reduction of IL-10 levels ( Figure 3a Figure S9c). The Hz-induced NF-κB activation was inhibited by ART as indicated by decreased p65 phosphorylation and increased IκBα in MO (Supporting Information Figure S9d

| D ISCUSS I ON
P. falciparum infection manifests itself as varying degrees of severity in malaria, inclusive of uncomplicated, mild, severe, and cerebral malaria. Variable susceptibility to the parasite is governed by a constant conflict between immunoprotection and immunopathology.  (Mantovani et al., 2004), it is likely that Hz might facilitate recruitment of CCR4 and CCR8-bearing TH2 cells and regulatory T cells of TH2 immunity which contribute to immunosuppression. It is noteworthy in this context that regulatory T cells suppress T-cell responses in malaria (Hisaeda et al., 2004). Furthermore, Hz induces production of M2b-specific cytokines-TNFα, IL-1β, and IL-6 in human MO. These data corroborate with reports demonstrating the presence of inflammatory cytokines IL-1β, TNFα, and IL-6 along with high level of anti-inflammatory IL-10 in plasma of malaria patients (Perera et al., 2013;Weinberg et al., 2016).
Functionally, M1-type MO are characterized by elevated production of NO and ROI that are essential for clearance of pathogens (Mantovani et al., 2004). Conversely, M2-type MO express high arginase activity which is due to production of arginase-1, the enzyme that converts arginine to ornithine and urea causing depletion of arginine and reduction of NO (Biswas & Mantovani, 2012;Galván-Peña & O'Neill, 2014;Martinez & Gordon, 2014). In malaria, depletion of NO results in low NO bioavailability that affects adherence of parasites to endothelium and is associated with severity of the disease.
High arginase activity and an inverse relation between disease severity and NO production are recently reported in malaria (Weinberg et al., 2016). In this context, MO fed with Hz demonstrate elevated arginase activity compared with untreated and latex-ingested MO.
These data along with the findings from studies that plasmodium parasites (Olszewski et al., 2009) and lysed infected erythrocytes (Reiter et al., 2002) release arginase suggest that cumulative elevation of arginase activity from these sources might contribute to M2like activity. Recent studies have reported Hz-induced production of ROS (Jaramillo et al., 2005) and NO (Jaramillo, Gowda, Radzioch, & Olivier, 2003) in MO. In contrast to these reports, we found that Hz significantly reduced production of NO and ROS levels compared with untreated and latex-ingested MO. The difference in the results may be attributed to the nature of Hz used; while the earlier studies were performed with delipidized Hz and β-hematin, our experiments were conducted with natural Hz. In addition, mitogen-stimulated lymphocyte proliferation was significantly suppressed in the presence of Hz-fed MO compared with control and latex-ingested MO.
This observation is in agreement with the study that malaria patients have suppressed T-and B-cell functions (Zander & Butler, 2013).
The signaling pathways p38 MAPK, PI3K-AKT, and NF-κB have been implicated in IL-10 synthesis (Saraiva & O'Garra, 2010) and M2 polarization (Zhang et al., 2010) - . In this study, Hz F I G U R E 5 Proposed model for hemozoin (Hz)-induced M2-like monocytes (MO) during infection. Hz released after schizont rupture is ingested by circulating MO in blood, activates PI3K-AKT, NF-κB, and p38 MAPK pathways, resulting in increased expression of M2 markers IL-10, CD206, arginase activity. Specifically, Hz induces M2a (CCL1) and M2b (CCL17, TNFα, IL-1β, and IL-6) phenotype that regulate TH2 responses. Hz attenuates HLA-DR expression, nitric oxide production, and reactive oxygen species-features of M1 phenotype. Antimalarial drugs-artemisinin and chloroquine-partially reverse the Hz-induced M2-like phenotype activated p38 MAPK, PI3K-AKT, and NF-κB pathways and pharmacological inhibitors for these pathways dramatically down-regulated the expression and secretion of IL-10, thus, over-riding the induction triggered by Hz exposure. In line with this observation, the inhibitors effectively reduced Hz-induced M2-like phenotypic, biochemical, and functional properties. While these findings implicate the role of p38 MAPK, PI3K-AKT, and NF-κB pathways, we cannot rule out the possibility of other signaling pathways in Hz-driven activation of MO toward M2 phenotype. Since Hz does not induce a specific phenotype subset but drives the MO toward M2a and M2b types, it may be postulated that Hz stimulates atypical activation mediated through PI3K-AKT, p38 MAPK, and NF-κB pathways.
We next aimed to identify strategies that could aid in reversing the effect of Hz in driving the MO to an M2-like phenotype.
In this regard, we conducted experiments with CHQ and ART for three reasons; first, they are reported to inhibit pathways crucial for polarization to M2 type; second, they are widely used for treatment of malaria; and third, they possess immunomodulatory properties. ART have been reported to decrease IL-10, IL-1β, IL-6, and TNFα levels (Shakir, Hussain, Javeed, Ashraf, & Riaz, 2011;Wu et al., 2016) through MAPK (Wang et al., 2009), NF-κB (Prato, Gallo, Giribaldi, Aldieri, & Arese, 2010;Wang et al., 2011), and PI3K-AKT (Kim et al., 2013) pathways. CHQ has also been demonstrated to act through MAPK (Kono et al., 2008;Weber, Chen, & Levitz, 2002) and NF-κB (Long, Liu, Wang, Zhou, & Zheng, 2015;Yang et al., 2013) pathways and decrease IL-1β, IL-6, and TNFα in mononuclear phagocytes in different diseases (Jang et al., 2006;Long et al., 2015). In another study, CHQ has been shown to be an effective anticancer drug in mice by inhibiting tumor resistant MQ (M2 MQ) and decreasing TGFβ and IL-10 production. The effect was accompanied by decreased myeloid-derived suppressor cells (MDSC), Tregs, and increasing CD8 + T cells in tumor milieu . The ability of these drugs to inhibit crucial pathways and production of cytokines/chemokines that are signature for M2 type MO has extended their use for other parasitic diseases, rheumatic diseases, and cancer. In our study, the drugs were found to be more effective in reducing the expression of phenotypic markers-cytokines, chemokines, and surface marker-CD206 associated with M2-like phenotype than reversing altered biochemical and functional properties. ART did not reverse arginase activity in Hz-fed MO. Our finding is supported by an earlier report that ART did not significantly alter arginase activity in leishmania donovani-infected MQ though it enhanced protective immune responses by reversing NO levels (Sen, Ganguly, Saha, & Chatterjee, 2010). These observations are not surprising as CHQ and ART are known for their antimalarial action through multiple modes on their target, but their mechanism of action as immunomodulators is not clearly elucidated. The M2 phenotype is previously related to abnormal autophagy and angiogenesis (Li et al., 2016). We speculate that CHQ, a known autophagy inhibitor, may attenuate Hz-induced M2-like MO activation by inhibiting autophagy. Previously, falciparum malaria pathogenesis has been related to upregulated angiogenic factors VEGF, Ang-2, and sFLT-1 along with dysregulated and excessive immune responses (Muehlenbachs, Mutabingwa, Edmonds, Fried, & Duffy, 2006;Silver, Higgins, McDonald, & Kain, 2010). The antiparasitic action of ART is accompanied with disruption of parasite proteins, alteration of mitochondrial functions, and angiogenesis (Ak, 2015; Shakir et al., 2011). In relation to ability of M2 phenotype to stimulate angiogenesis, ART may attenuate Hz-induced M2-like phenotype by altering the release of angiogenic factors.
The schematic diagram ( Figure 5) summarizes the Hz-induced alterations in M1-and M2-associated markers, the pathways involved and partial reversal by CHQ and ART in human MO. Hz released after schizont rupture is readily ingested by circulating MO in blood; activates PI3K-AKT, NF-κB, and p38 MAPK pathways resulting in increased expression of M2 markers; and drives the MO toward M2 (M2a and M2b)-like phenotype. Antimalarial drugs-ART and CHQ-partially reverse the process of activation of MO toward M2-like phenotype. In conclusion, the findings from this study add a new dimension to the understanding of mechanism/s underlying Hz-induced immunosuppression by demonstrating its ability to drive the MO toward M2-like phenotype. Though antimalarial resistance poses a major challenge and discourages the continuation of CHQ and ART, we propose that these drugs may be re-examined for their potential as immunomodulators and candidates for adjunctive treatment in malaria.

ACK N OWLED G M ENTS
We thank, Dr. S.C. Mande, Director, National Centre for Cell Sciences

CO N FLI C T O F I NTE R E S T
Research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.