Albumax I and RPMI-1640 were purchased from Fisher Scientific^® (Leicestershire, UK). All solvents used were HPLC-grade or higher quality and purchased from Sigma-Aldrich (St. Louis, Missouri USA). Radiolabeled isoprenic precursors [1-(n)-³H] geranylgeranyl pyrophosphate triammonium salt {[1-(n)-³H]-GGPP; 14 Ci/mmol}, [1-(n)-³H] phytol (20 Ci/mmol), and {[1-(n)-³H] phytyl-PP ([³H]-phytyl-PP; 20 Ci/mmol) were obtained from Amersham-Pharmacia Biotech (Buckinghamshire, UK). Adenosyl-L-methionine, S-[methyl-³H] ([³H]-SAM, 82 Ci/mmol) and [1-(n)-³H] FPP triammonium salt ([³H]-FPP 23 Ci/mmol) were purchased from Perkin Elmer^® (Waltham, Massachusetts, EUA). PK, α-tocopherol, γ-tocopherol, phytol, phytyl-PP, GGPP, UQ-8, UQ-9 and MK-4 pure standards were purchased from Sigma-Aldrich. Saponin, hypoxanthine, gentamycin sulfate, D-sorbitol, glucose, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and another reagent not cited here were also purchased from Sigma-Aldrich.

P. falciparum clone 3D7 was cultured in vitro by the classic culture methodology reported by Trager and Jensen (Trager and Jensen, 1976) employing RPMI-1640 medium completed supplemented with Albumax I (0.5%) in 75 cm³ cell culture flasks (Radfar et al., 2009). The pH was adjusted to 7.4, and a gas mixture of 5% CO₂, 5% O₂, and 90% N₂ was employed. Culture synchronization at ring stages was performed with 5% (w/v) D-sorbitol solution as described by Lambros & Vanderberg (Lambros and Vanderberg, 1979). Parasite development was monitored by Giemsa-stained thin smear microscopy. Blood was leuco-depleted, and PCR (for mycoplasma) and optic microscopy were used to monitor culture contamination (Rowe et al., 1998).

Synchronous cultures of P. falciparum were labeled at ring stages with 15 μCi/ml [³H]-GGPP, [³H]-FPP, [³H]-phytyl-PP or [³H]-phytol or 2.5 μCi/ml [³H]-SAM in complete RPMI 1640 medium for 16-18 h. After the labeling period, infected erythrocytes containing late trophozoite or schizont stages were isolated by magnetic separation (see below) or purified discontinuous Percoll^® (see below). For all metabolic labeling experiments, approximately 4x10⁸ infected erythrocytes were used.

Cultures at the schizont stage were purified by a magnetic column (MACS separation column, Manual Cell Separation) (Ribaut et al., 2008). Briefly, the column was equilibrated and washed at room temperature with RPMI-1640 complete medium. The culture was centrifuged, and the pellet was suspended in complete culture medium (1:10 v/v) and loaded into the column. Then, the column was washed with RPMI-1640 complete medium, and the retained parasites were collected (Ribaut et al., 2008). Infected red blood cells were centrifuged at 600 x g for 5 min, and the pellet was frozen in liquid nitrogen for further analysis.

Ring (1-20 h after reinvasion), trophozoite (20-30 h after reinvasion) and schizont (30-45 h after reinvasion) forms were purified with a 40/70/80% discontinuous Percoll^® gradient (10,000 x g, 30 min, 25°C) (Braun-Breton et al., 1986). In the fraction containing the ring stage and uninfected erythrocytes (80%), parasites were isolated by treatment with 0.1% (w/v) saponin for 5 min followed by two washes with PBS, pH 7.2, at 10,000 x g for 10 min. The fractions containing trophozoite stages or schizont stages were centrifuged at 600 x g for 5 min, and the pellets were frozen in liquid nitrogen for further analysis.

The parasitophorous vacuole/erythrocyte-free parasites for phytol phosphorylation assays were obtained by saponin lysis (Christophers and Fulton, 1939). For this purpose, erythrocytes were suspended in 30 mL of PBS (30 mM Na₂HPO₄, 6 mM KH₂PO₄, pH 7.4, 120 mM NaCl) with 2 g/L glucose and 0.2% (v/v) saponin at 4°C and centrifuged at 700 x g at 4°C. The resulting pellet was washed twice in 30 ml of complete cold RPMI-1640 medium. Finally, the pellet was frozen in liquid nitrogen for posterior analysis.

Lyophilized infected erythrocytes (1 x 10¹⁰) were suspended in 4 mL of deionized water in glass tubes. Cell lysis was carried out at 4°C by ultrasonication with 4 pulses of 5 seconds with 10% intensity and 15 second intervals between them. The proteins were precipitated by adding 800 µL of ethanol. After mixing for 1 min, extraction was performed three times with 8 mL of hexane. The sample was shaken for 1 min and centrifuged at 2700 x g for 10 min at 4°C (Ischebeck et al., 2006). The hexane extract of erythrocytes infected with P. falciparum was concentrated under nitrogen flow and analyzed by RP-HPLC.

The lysis of 1 x 10¹⁰ erythrocytes infected with P. falciparum was achieved by sonication as described for PK and phytol extraction. The extraction of phytyl-PP was carried out three times with 6 mL of n-butyl alcohol saturated with water. The solution was centrifuged at 3000 x g for 10 min at 4°C. The n-butyl alcohol phases were combined and evaporated under nitrogen stream and analyzed by TLC.

The samples were dissolved in a small volume of n-butyl alcohol saturated with water and applied on a silica 250-PA TLC plate (Mallinckrodt Baker, Griesheim, Germany). The plate was developed with isopropyl alcohol/ammonia (32%)/H₂O (6:3:1 v/v/v). Phytyl-PP was identified by retention factor (R _f ) and coelution with a commercial standard (Ischebeck et al., 2006).

The methodology for measuring phytol phosphorylation activity in parasite extracts was carried out as previously reported (Ischebeck et al., 2006). Approximately 4x10⁹ saponin-purified free schizonts were resuspended in a solution containing 100 mM Tris-HCl buffer, pH 7.5, 1 mM dithiothreitol (DDT); 1 mM MgCl₂; 1 mM isoascorbate; 1 mM KCl; 0.1% albumin serum; and a protease inhibitor cocktail purchased from Sigma. Then, parasites were successively frozen and thawed 4 times in liquid nitrogen. Subsequently, parasites were lysed three times by 10 second sonications at 30 second intervals. Then, samples were centrifuged at 100,000 x g for 30 min at 4°C (Ischebeck et al., 2006), yielding a membrane fraction (pellet) and a cytoplasmic fraction (supernatant).

Both phases were separated, and 300 μL of reaction buffer was added (50 mM MgCl₂; 20 mM NaF; 50 mM Tris-HCl, pH 8; 20 mM sodium orthovanate; 1 mM DTT; 50 mM CaCl₂; 100 pmol [³H]-phytol). Finally, ATP, CTP, UTP and GTP were added to a final concentration of 200 mM each. After 30 min incubation at 37°C, the reaction was stopped by adding 300 μl of saturated n-butanol (1:1 v/v). The sample was then vortexed and centrifuged at 10,000 x g for 10 min. The supernatant was collected, and the extraction procedure was repeated twice again by adding the same volume of saturated butanol (1:1 v/v). The three collected supernatants were transferred to a single glass tube and dried under a nitrogen stream. For chromatographic analysis, the sample was suspended on the RP-HPLC initial mobile phase (system II, see below) using phytol, phytyl-P and phytyl-PP as internal standards (Ischebeck et al., 2006).

The stationary phase was a Phenomenex Luna^® C18 column (250 mm x 4.6 mm x 5 μm) (Phenomenex^®, CA, USA) coupled to a pre-C18 column (Phenomenex^®, CA, USA). A UV-vis or a 170 diode array detector (DAD) (Gilson^®, Villiers-le Bel, France) and an FC203B fraction collector were used. The software used for data processing was the Trilution™ LC 3.0 System Software. Three different elution systems were used to analyze the samples.

System I: A linear gradient with methanol (solvent A) and acetonitrile (solvent B) was used at a flow rate of 1 mL/min. The ratio was initially 50% (B), which increased to 70% (B) in 30 min and was then maintained for 20 min. The fractions were collected once per minute. Pure standards of UQ-8, UQ-9, MK-4 and PK were co-injected. As the intraerythrocytic stages of P. falciparum biosynthesize UQ-8, UQ-9 and MK-4, the pure standards of these compounds were injected in this system to show that they do not co-elute with PK and tocopherol. UQs were monitored at 290 nm, MK-4 at 270 nm and PK at 245 nm.

System II: The system consisted of a linear gradient elution with 25 mM NaHCO₃ (A) and acetonitrile (B). The gradient started at 30% (B) and increased to 0% (B) in 20 min, was maintained for 19 min and returned to initial conditions in two minutes. The flow rate was set at 1 mL/min, the fractions were collected once per minute, and internal standards were monitored at 214 nm.

System III: An isocratic system with methanol/ethanol (1:1 v/v) was used at a flow rate of 0.5 mL/min. Samples were collected once per minute. The pure standards of UQ-8, UQ-9 and MK-4 were injected into this system to assess whether they co-eluted with PK and were monitored at the wavelengths described above.

An electron impact gas chromatography-mass spectrometer (GC-MS) with a Trace GC and Y2K ion trap PolarisQ MS System (Finnigan, ThermoQUEST Inc., San Jose, CA) with a data analysis program (Xcalibur version 1.3) was used. The device was equipped with a TR-1MS column (30 m, 0.25 mm, 0.25 μm, Thermo Scientific, USA, CA). The injector temperature was set at 220°C, and there was a splitless liner. An oven temperature of 100°C was initially maintained for 2 min and then increased to 300°C at a rate of 25°C/min. The temperature was maintained for 10 min and then reduced to its initial value. The transfer line was maintained at 260°C, and the helium flow was 1.5 mL/min. The MS was operated in positive mode with an ion source at 200°C. The mass range monitored was m/z 40 to 500 (full scan). Vitamin K1 and its derivatives were identified by the ions at m/z 450 for PK (Sussmann et al., 2016), m/z 452 for KH₂ and m/z 480 for methylated hydrophylloquinone.

System I: The analysis was performed in an LCQ Duo™ MS (ThermoFinnigan, EUA) coupled with Nano RP-HPLC (LC-Packings, model Ultimate) operated with a flux of 250 µL/min. The solvent was 80% acetonitrile with 0.2% lithium iodide. Electrospray ionization (ES) was used, and the MS was operated in positive mode with a potential of 4.5 kV and a capillary temperature of 180°C. The mass spectra were processed with Xcalibur^® 2.0 software (Copyright^© Thermo Electron Corporation, USA).

System II: The analysis was performed with a MicrOTOF-QII MS (Bruker^®). The equipment was operated in negative mode with capillary voltage of 3500 V, N₂ as a mixture gas at 2 bar, N₂ as a drying gas at 4 L/min and a drying temperature of 180°C. Funnel 1 was an RF funnel with 400 Vpp, a quadrupole energy of 6 eV, collision energy of 12 eV and the hexapole RF was operated at 200 Vpp. In tandem MS (MS/MS), collision energies of 0 to 20 eV were used.

System III: The analysis was performed with a High-Resolution LC-QToF - maXis II™ (Bruker^®). The column was a Shimadzu XR ODSIII (150x2 mm) with flux 0.2 ml/min. The solvent was 60% acetonitrile and 40% methanol at 40°C.

The equipment was operated with atmospheric pressure chemical ionization (APCI) in positive mode with capillary voltage of 3500 V, N₂ as a mixture gas at 3 bar, N₂ as a drying gas at 4 L/min and a drying temperature of 250°C. Funnel 1 was an RF funnel with 400 Vpp, a quadrupole energy of 7 eV, collision energy of 6 eV and the hexapole RF was operated at 700 Vpp.

Each extract obtained from infected erythrocytes, uninfected erythrocytes and standard were filtered and injected into the LC-QToF equipment. Prior to the injection of the samples, system suitability tests and blanks were carried out between each of them.

Electrochemical measurements were performed at 298 ± 1 K in a thermostatic cell with CH 660I equipment. A BAS MF2012 glassy carbon working electrode (geometrical area 0.071 cm²), a platinum wire auxiliary electrode and an Ag/AgCl (3 M NaCl) reference electrode were used in a conventional three-electrode arrangement. Cyclic and square wave voltammetry (CV and SWV, respectively) were used as detection modes. Eventually, derivative convolution of the data was performed to increase the peak resolution. Thin films of vitamin K1 and lyophilized blood samples were prepared on GCEs following a previously reported procedure (Doménech-Carbó et al., 2013b) by pipetting 10 µL of the liquid vitamin or 20 µL of a dispersion (1 mg/mL) of blood in ethanol and allowing the solvent to evaporate in air. As a result, a uniform, fine coating of the solid was adhered to the working electrode. Aqueous 0.10 M potassium PBS at physiological pH that had been previously degasified by bubbling argon for 10 min was used as a supporting electrolyte.

The stabilization of hydrophylloquinone was achieved by di-O-methylation as described (Sussmann et al., 2016). Briefly, 4x10⁸ infected erythrocytes labeled with [³H]-phytyl-PP were lysed by ultrasonication in a Branson sonicator at 4°C with three pulses of 5 second duration with 10% potency and 10 second intervals between them. Then, 50 mg of potassium hydroxide and 60 μL of methyl iodide (CH₃I) were added. The reaction was stirred overnight under a nitrogen atmosphere at room temperature in the dark. The reaction mixture was partitioned between water (2 mL) and dichloromethane (3 × 2 mL). The organic phase was evaporated under nitrogen stream, and the residue was submitted to RP-HPLC analysis and scintillation counting.

For atovaquone (ATO) IC₅₀ value estimation, parasites were incubated for 48 _h in the absence or presence of several concentrations of PK: (i) 0.010 µM was used as the first concentration of ATO, which reached 0.000156 µM by serial dilution (1:1/vol:vol), and (ii) 5.63 µM was used as the first concentration of PK, which reached 0.09 µM by serial dilution (1:1/vol:vol). For MK-4 IC₅₀ value estimation, parasites were incubated for 48 _h in the absence or presence of MK-4; 16 µM was the first of several concentrations of MK-4 used.

Assays were performed in 96-well plates starting at 2% parasitemia in the ring stage and 2% hematocrit as described (Desjardins et al., 1979). Parasite growth was monitored after 48 h by SYBR Green I^® DNA staining (Smilkstein et al., 2004). For this purpose, 100 μL of culture was incubated in darkness and at room temperature after adding 100 μL of Syber Green I^® 2/10,000 (v/v) in lysis buffer [20 mM Tris, pH 7.5; 5 mM EDTA; 0.008% saponin (w/v); 0.08% Triton X-100 (v/v)]. After 1 h, fluorescence was measured with a POLARstar Omega fluorometer^® (BMG Labtech^®, Ortenberg, Germany) with excitation and emission bands centered at wavelengths of 485 and 530 nm, respectively. Inhibition of parasite growth was analyzed in relation to the logarithm of the concentration using a nonlinear regression (dose-response slope/variable sigmoid equation) by using GraphPad Prism^® software. Only experiments showing R-squared values (R²) ≥ 0.95 were accepted. All experiments were performed at least three times with three technical triplicates for each experiment.

Oxygen consumption assays were performed in cultures at the trophozoite/young schizont stages (Murphy et al., 1997). Red blood cells were centrifuged at 1500 x g for 5 min at 4°C and lysed with 30 mL of 0.03% saponin in PBS at 25°C for 5 min. The obtained sample was centrifuged at 1500 x g for 5 min at 4°C and washed in respiration buffer (125 mM sucrose, 65 mM KCl, 10 mM HEPES-KOH, pH 7.2, 5 mM MgCl₂, 2 mM KH₂PO₄, 0.5 mM EGTA). Free parasites were suspended to 1x10⁹ cells/ml in respiration buffer, and the samples were utilized immediately for oxygen consumption measurements with a high-resolution oxygraph (Oxygraph-2k Oroborus Instruments, Innsbruck, Austria). ATO, PK, carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP) or antimycin A (AntA) additions were performed during the assay as described in the results.

Radioactive fractions with retention times coincident with the standards α-tocopherol, γ-tocopherol and phylloquinone (PK) were observed in schizont stages of P. falciparum when [³H]-phytol ( Figure 1A ), [³H]-GGPP ( Figure 1B ), [³H]-FPP ( Figure 1C ), [³H]-SAM ( Figure 1D ), [³H]-phytyl-PP ( Figure 4 ) were used as precursors and analyzed by reverse-phase high-performance liquid chromatography (RP-HPLC) using solvent system I. Using another RP-HPLC method (solvent system III) peaks with retention times coincident with the standards MK-4 and PK in the three stages of intraerythrocytic P. falciparum labeled with [³H]-GGPP were identified (supplemental data, Figure S1A ). The greatest incorporation of the radioactive precursor was detected in the schizont stage. Similar results were found in the schizont stage when [³H]-FPP was used as a precursor and samples were analyzed by the system III RP-HPLC method (supplemental data, Figure S1B ). The incorporation of [³H]-FPP and [³H]-GGPP into phytylated quinones suggests that parasites biosynthesize de novo phytyl-PP. Moreover, the use of [³H]-SAM shows that the probable PK and tocopherols were methylated as expected (Threlfall et al., 1968; Gaudillière et al., 1984).

The presence of PK in the parasite was confirmed by two methods of mass spectrometry. First, using liquid chromatographic separation and high-resolution mass spectrometry with atmospheric pressure chemical ionization LC-APCI-QToF. Chromatographic signal ( Figure 2A ) detected at 20.3 min in infected erythrocytes with molecular ion mass [M-H]⁺ of m/z 451.3611 Da was only 2.2 ppm different from the signal presented by phylloquinone standard, m/z 451.3601 Da ( Figure 2B ). Uninfected erythrocytes showed no signal ( Figure 2C ). Second, using the liquid chromatography/electrospray ionization mass spectrometry (LC/ESI) system I. The molecular ion peaks at m/z 456 and 439 Da corresponded to [M + Li] and [M + Li-H₂O], respectively, considering the molecular formula C₃₁H₄₆O₂ (supplemental data, Figure S2 ). The fragmentary ions at m/z 204, 221 and 245 Da were identified as the aromatic moieties of PK (supplemental data, Figure S2A ). A similar profile was detected in infected erythrocytes (1 x 10¹⁰ parasite extracts) (supplemental data, Figure S2B ) but with 3.5-fold higher intensity than the same amounts of uninfected erythrocytes (supplemental data, Figure S2C ), comparing peak area and total ion count (TIC) for each sample.

The presence of PK was not detected in 100 mL of lyophilized complete culture medium extracted as described for the parasites ( Figure S2D ), suggesting that this vitamin is inherent to the complex erythrocyte-parasite. Although vitamin K1 is not biosynthesized by humans, when provided in the diet, it can be transported by plasma lipoproteins and found in the bloodstream (Shearer, 1992; Shearer et al., 2012).

The structure of PK determined by mass spectrometry was further analyzed by electrochemical studies to confirm its presence and the oxidation state of this compound in infected erythrocytes. The proposed methodology explores the electroactive character of both vitamin K1, which is dominated by a quinone/hydroquinone redox interconversion (Munir et al., 2012; Webster, 2012), and Fe-heme, where the Fe(III)/Fe(II) redox pair has been widely studied (Lexa et al., 1988; Chen et al., 1990; Tieman et al., 1990; Jeon and Bruice, 1992; Sagara et al., 1998; de Groot et al., 2005). The electrochemistry of vitamin K3 was studied by Zhu and Li (Zhu and Li, 1999). Here, the voltammetry of microparticles (VMP), a technique that enables the determination of the voltammetric response of sparingly soluble solid materials (Doménech-Carbó et al., 2013a), was applied to vitamin K1 films deposited on glassy carbon electrodes (GCEs) and microparticulate deposits of uninfected blood and infected erythrocytes following reported procedures (Doménech-Carbó et al., 2013b).

Figure 3 compares the square wave voltammetry of vitamin K1, uninfected blood and infected erythrocytes in contact with phosphate-buffered saline (PBS) when the potential is scanned in the negative direction. Vitamin K1 displayed a unique cathodic signal (C₁) at −0.52 V ( Figure 3A ), which can be assigned to the proton-assisted reduction of the quinone unit, whereas for uninfected blood, a reduction signal at −0.20 V (C₂) was recorded ( Figure 3B ). This signal appears as a single, undifferentiated peak, which can be attributed to the reduction of Fe(III)-heme to Fe(II)-heme. Remarkably, the voltammograms for P. falciparum-infected erythrocytes showed both peaks C₁ and C₂ ( Figure 3C ), thus denoting the presence of PK accompanying Fe-heme to some extent forming a hemoglobin-phylloquinone complex (Zhu and Li, 1999).

When the potential is scanned in the positive direction, the vitamin K1 displays an anodic signal at 0.0 V (A₁) corresponding to the anodic counterpart of the process C₁. Similarly, the uninfected erythrocytes ( Figure 3D ) yielded peak A₂ at -0.20 V, corresponding to the reversal of the process C₂. Remarkably, the voltammogram of the infected erythrocytes ( Figure 3E ) showed overlapping peaks at -0.30 (A₃), -0.20 (A₂), 0.00 (A₁) and +0.10 V (A₄). Peak A₃ can in principle be attributed to the oxidation of a Fe(II)-heme form differing from the pristine Fe(II)-heme produced in uninfected erythrocytes conceivably the farnesylated species generated by action of P. falciparum. However, the multiple anodic peak profile suggests that other reduced species were generated in the infected erythrocytes.

To determine whether probable PK P. falciparum can be found in the reduced form, schizonts stages were metabolically labeled with [³H]-phytyl-PP. The di-O-methylation of hydrophylloquinone was performed as previously described to stabilize the compound’s redox state (Sussmann et al., 2016). The presence of di-O-methylated hydrophylloquinone was confirmed by RP-HPLC using system I. The identity of derivatized hydrophylloquinone was confirmed by gas chromatography/mass spectrometry (GC/MS) analysis as described (Sussmann et al., 2016). The radioactive fractions of the parasites coincide with the α-tocopherol, methylated hydrophylloquinone and PK standards, but were not detected in uninfected blood, demonstrating that the phytylated molecule is not biosynthesized by erythrocytes ( Figure 4 ).

Once the metabolite, probable PK, was characterized in both redox states in the parasite P. falciparum, we further investigate the biosynthesis of the isoprenoid quinone side chain. For this purpose, parasites were labeled with [³H]-GGPP as described. Extracts were obtained using the phytyl-PP extraction protocol and analyzed by RP-HPLC system II. The metabolic profile obtained showed radioactive incorporation in fractions containing a phytyl-PP co-injected standard ( Figure 5 ) indicating the formation of phytyl-PP having GGPP as a precursor.

Identification of phytyl-PP in the parasite was also performed by mass spectrometry (HRESIMS). For this purpose, 5x10⁹ infected and uninfected cells and the corresponding amount (50 mL) of lyophilized complete culture medium were used. The crude extract was submitted to thin-layer chromatography (TLC) purification as described (Ischebeck et al., 2006), and the bands corresponding to phytyl-PP were scraped, extracted with saturated butanol (1:1 v/v) and analyzed by direct injection into the LC/MS using system II ( Figure 6 ). The phytyl-PP was detected in infected erythrocytes based on the ion at m/z 455.2324 Da corresponding to [M-H]^- with an error of 2 ppm comparing to calculated formulae for C₂₀H₄₂P₂O₇ (calc. m/z 455.2333 Da) and with the same isotopic profile as the standard. The sodium adduct [M+Na-2H]^- was also detected.

Another question we faced was whether parasites could also produce phytyl-PP via phytol phosphorylation. For this purpose, we followed the methods of Ischebeck et al. and Valentin et al. in previous studies (Ischebeck et al., 2006; Valentin et al., 2006). The plasmodial phosphorylation of phytol was detected ( Figure 7 ) as described for Arabidopsis (Ischebeck et al., 2006; Valentin et al., 2006). Phytol kinase activity requires phosphate donors such as ATP, CTP, UTP, or GTP. In most organisms, various isoprenoid phosphorylation reactions are mediated by CTP (Heller et al., 1992; Inoue et al., 1994; Bentinger et al., 2007).

The antiplasmodial effect of 0.6 nM atovaquone {ATO, trans-2-[4-(4-chlorophenyl} cyclohexyl]-3-hydroxy-1,4-naphthalenedione) was investigated ( Figure 8A ). After treatment with different concentrations of PK, the parasite proliferation was recovered. After treatment with 1.41 µM PK, the effect of ATO was almost nullified, with growing rates similar to the control (untreated parasites). With this result, it was decided to expand the experiment by estimating the ATO IC₅₀ value at different PK concentrations. The interference of PK in the antimalarial effect of ATO was confirmed by treatment with 2.81 and 5.63 µM PK, which significantly increased the ATO IC₅₀ value by 2.6- and 3-fold, respectively ( Figure 8B ). It is important to note that PK by itself and at the concentrations used in our experiments does not affect parasite proliferation or the oxygen concentrations in the oxygraph (data not shown).

Since a supposed role of MK-4 in the electron transport chain had already been indicated (Tonhosolo et al., 2010), we also tried to perform tests to recover the effect of ATO with this metabolite. However, we found that when administered exogenously, MK-4 had an antimalarial effect, with an IC₅₀ of 1.18 ± 0.23 µM ( Figure S3A ). At 0.25 µM concentration, MK-4 had an antiplasmodial effect; therefore, we used concentrations of 20 and 50 nM for recovery tests. As seen in Figure S3B , MK-4 does not recover the antiplasmodial effect of ATO at the tested concentrations. Since PK attenuated ATO, oxygen consumption assays were used to investigate whether this vitamin could directly interfere with the mechanism of action of ATO in vitro.

The rates of O₂ consumption were measured after the addition of erythrocyte-free schizonts and gave the total cellular oxygen consumption rate (R’) ( Figure 9A ). The parasites were treated with ATO (up to a final concentration of 1 nM). ATO is an inhibitor of mitochondrial ETS complex III and therefore, as expected, reduced the oxygen consumption. Afterward, PK was titrated up to 30 µM. Our results demonstrate that PK triggered an oxygen consumption rate similar to R’ at 20 µM and started to decrease the rate at above 30 µM ( Figure 9B ). An additional titration with carbonyl cyanide-4-(trifluoromethoxy) phenylhydrazone (FCCP) was performed to obtain the optimum concentration of the protonophore that triggers the maximum oxygen flux/electron transfer through the ETS (E/ET capacity). It was verified that at 30-35 µM FCCP, the oxygen consumption slope reached E ( Figure 9B ), ensuring mitochondrial integrity and proper functioning of all ETS components. Furthermore, after treatment with 50 µM antimycin A (AntA), was possible to estimate that approximately 18% of R’ corresponded to residual oxygen consumption (ROX).