Proteomic affinity chromatography
Setup is summarized in Supplementary Table 1. GNF179- and Compound X-coated beads were mixed with P. falciparum protein lysates (extracted using low 0.02% or standard 0.8% detergent) and subsequently competed with free GNF179; mixed human cell extracts were also tested. Ten samples were analyzed in parallel (TMT 10-plex) to generate values for the affinity of the beads to the bound proteins (“depletion” values, four samples) and to generate IC50 values (six samples) in a single experiment. Sample 1 and 2 represent the vehicle control; sample 3 and 4 were done in the same way, but while the beads were discarded after the first incubation step the extract was incubated with fresh beads to measure how much protein could rebind to the fresh beads (was depleted from the extract by first bead-binding). Apparent dissociation constants were determined by considering the protein depletion by the beads. Sample 5-10 were used to generate IC50 values by adding GNF179 over a range of concentrations (maximum of 20µM, 1:3 dilutions). MS/MS was utilized for quantification. MS1 score analysis was done using Prism 10.
Strain construction
For S. cerevisiae transformation, 8 OD600 early-log phase cells (0.8 OD600 mL-1) were washed twice with cell-grade water and then twice with lithium acetate buffer (100mM lithium acetate, 10mM Tris-HCl pH 7.5, 1mM EDTA). Approximately 100µg boiled salmon sperm DNA and 500ng PCR-generated linear DNA were added, followed by six-fold volume of 40% v/v polyethylene glycol in lithium acetate buffer. The sample was incubated at 30°C for 30 minutes, treated with 10% v/v DMSO, and subsequently heat-shocked at 42°C for 15 minutes. Cells were sedimented and resuspended in 1mL YPD (1% yeast extract, 2% peptone, 2% dextrose), allowing overnight recovery at 30°C before plating on YPD+Zeocin (50µg mL-1) agar plates. SDS-PAGE and subsequent Western blot with mouse anti-myc (Invitrogen, R950-25, 1:1000) or rabbit anti-TAP (Invitrogen, CAB1001, 1:1000) antibodies detected the proteins of interest; rabbit anti-ScATP2 (gifted by the Subramani Lab at UCSD, 1:2000) was included as a loading control. HRP-linked goat anti-mouse (Bio-Rad, 1706516, 1:5000) and goat anti-rabbit (Bio-Rad, 1706515, 1:5000) antibodies were the secondary antibodies used for chemiluminescence. The mutant strains used in this study can be found in Supplementary Table 5.
The three linear DNA constructs transformed into S. cerevisiae were created using oligos summarized in Supplementary Table 6. Myc-tagged full-length ScSEY1 was generated using oligos oKC1 and oKC2 on pPICZ-C plasmid (gifted by the Subramani Lab). Myc-tagged truncated ScSEY1(S437*) was generated using oligos oKC3 and oKC2 on the same plasmid. Oligos oKC4 and oKC5 were used on genomic DNA (gDNA) from the ScSEY1-myc mutant to obtain the additional ScSEY1 gene construct that was transformed into BY4741+ScSEY1-TAP, which is from the Yeast-TAP Tagged ORF Library (Ghaemmaghami et al., 2003).
For K. phaffii (gifted by the Subramani Lab) transformation to express P. vivax SEY1, 50 OD600 early-log phase cells were resuspended in 5mL YPD containing 20mM HEPES pH 8.0 and 25mM 1,4-Dithiothreitol and then incubated at 30°C for 15 minutes (with orbital rotation). After, cells were washed three times with cold sterile, and then with cold 1M sorbitol; the cells were incubated in 500µL cold 1M sorbitol for one hour. Exactly 100µL of cells were mixed with 500ng linearized plasmid DNA. After 10 minutes on ice, the sample was placed in a 2mm cuvette and electroporated (50μF, 200Ω, 7500kV cm-1). Cells were plated on SD+CSM-His (0.17% yeast nitrogen base, 0.5% ammonium sulfate, 2% dextrose, 0.077% CSM-His) agar plates. Strains underwent quality assurance using SDS-PAGE and subsequent Western blot with mouse anti-myc and HRP-linked goat anti-mouse antibodies.
The plasmid DNA transformed into K. phaffii was generated by amplifying myc-tagged codon-optimized PvSEY1, which was ordered from IDT, with oligos oKC6 and oKC7. The pIB2 plasmid (gifted by the Subramani Lab) was digested with EcoRI and KpnI restriction enzymes. Gibson assembly (NEB, E2611) was used to combine these two gene fragments. The complete plasmid was linearized with StuI restriction enzyme.
For transformation of Rosetta 2 DE3 (Novagen, 71400) and DH5-alpha strains of E. coli (NEB, C2987), 50μL of chemically competent cells were mixed with 10ng circular plasmid or 10% of a Gibson reaction. Cells were incubated on ice for 30 minutes before heat-shocking at 42°C for 30 seconds. After incubation on ice for five minutes, cells were added to 950μL SOC and incubated at 37°C for an hour (with shaking at 225rpm). Cells were plated on LB+Ampicillin (0.5% yeast extract, 1% tryptone, 1% NaCl, 100µg mL-1 ampicillin) plates. Plasmids were extracted using miniprep kits (Promega, A1460) and sequenced by Plasmidsaurus for accuracy.
For construction of the PfSEY1-smV5Tet line, 25μg homology-directed repair (HDR) plasmid was linearized by digestion, purified, and then co-transfected with 20μg guide RNA (gRNA) plasmid (containing Streptococcus pyogenes Cas9) into P. falciparum 3D7 parasites (synchronized as schizonts) using the Amaxa 4D system. Parasites were cultured with 500nM anhydrotetracycline (aTc) from the onset of transfection. One day post transfection, 2.5nM WR99210 (Jacobus Pharmaceuticals) drug pressure was applied.
HDR (pAM115) and gRNA plasmids (pAM116, pAM117, pAM118): 3’- and 5’-homology regions (HR) of PfSEY1 were amplified from P. falciparum 3D7 gDNA using oligos oJDD6152/6153 and oJDD6146/6147, respectively. A codon-altered region for the last 88 amino acids was generated by IDT and amplified using oligos oJDD6148/6149. The smV5Tet-Dozi-DHFR drug cassette region and pGEM backbone were amplified from pPG03 (Gurung et al., 2024) using oligos oJDD6150/6151 and oJDD6154/6155, respectively. Fragments were joined using the NEBridge Golden Gate method (with NEB BsaI-HF v2) by New England Biolabs. To generate the three PfSEY1 gRNA plasmids, oligos corresponding to each guide were annealed and ligated into BpiI-digested pRR216 (Rudlaff et al., 2019), which encodes SpCas9 and a U6 guide cassette. For pAM116, oligos oJDD6156/6157 were used. For pAM117, oligos oJDD6158/6159 were used. For pAM118, oligos oJDD6160/6161 were used (Supplementary Table 6).
Protein extraction
Protein lysate containing functional ScSEY1-myc was extracted from S. cerevisiae, while protein lysate containing functional PvSEY1-myc was extracted from K. phaffii. Approximately 400 OD600 early-log phase cells were washed twice with sterile water and incubated in 8mL zymolyase buffer (500mM KCI, 5mM MOPS-KOH pH 7.2, 10mM Na2SO3, 5000U Zymolyase-100T) for 30 minutes at 30°C (orbital rotation). Cells were then pelleted and resuspended in 4mL homogenization buffer (5mM MES, 1M sorbitol, 5mM NaF, 20mM EDTA, 1mM PMSF, 1x protease inhibitor cocktail) and lysed using a Dounce homogenizer. Cell debris was removed by centrifuging at 1,000g for 10 minutes. ScSEY1 was solubilized in 0.8% digitonin, while PvSEY1 protein was solubilized in 1% Triton X-100, for two hours at 4°C (with orbital rotation). Solubilized proteins were clarified by centrifugation at 100,000g for 20 minutes. Proteins were detected via Western blot using mouse anti-myc, rabbit anti-TAP, and rabbit anti-ScATP2 antibodies, along with the corresponding HRP-linked secondary antibodies.
Full-length functional PvSEY1-myc-His protein was extracted from E. coli. Approximately 1L of early-log phase cells were harvested after induction with 1mM IPTG at 16°C for 16 hours (with shaking at 160rpm). Cells were lysed in 20mL lysis buffer (1% lysozyme, 1% protease inhibitor cocktail, 1% Triton X-100, 1mM PMSF, 20mM imidazole, 2500U endonuclease in B-PER reagent) at 4°C for two hours (with orbital rotation). The cell lysate, cleared by chilled centrifugation at 20,000g for 20 minutes, was incubated with Ni-NTA beads at 4°C for 16 hours (with orbital rotation) before passing them through a column. The beads were washed three times with wash buffer (500mM NaCl, 20mM Tris-HCl pH 7.9, 20mM imidazole). Then, proteins were eluted off the beads using two column volumes of elution buffer (500mM NaCl, 20mM Tris-HCl pH 7.9, 300mM imidazole). Protein samples were concentrated using Pierce protein concentrators (Thermo Scientific, 88502) and quantified via Qubit (Invitrogen, Q33211). Proteins were detected via Western blot, using mouse anti-myc and HRP-linked goat anti-mouse antibodies.
For quantifying PfSEY1-smV5 abundance in knockdown experiments, whole cell lysates were extracted from P. falciparum cells that were harvested from red blood cells via saponin lysis. The parasites were boiled in Laemmli buffer containing beta-mercaptoethanol for five minutes; cell debris was removed via centrifugation at max speed for five minutes. Protein concentration was measured using Qubit before SDS-PAGE and Western blot with mouse anti-V5 (Invitrogen, R960, 1:1000), mouse anti-PfHSP70 (Genscript, SC1320A 1:1000), and goat anti-mouse antibodies.
IC50 determination
For proteomic affinity chromatography experiments, inhibition of intraerythrocytic P. falciparum growth was determined by a modified in-vitro [3H]-hypoxanthine incorporation method (Desjardins et al., 1979). Briefly, a culture of parasitized red blood cells (0.5% parasitemia with over 70% of total parasites in ring stage, 2.0% hematocrit) in RPMI-1640, 5% AlbuMAX, and 5μM hypoxanthine was exposed to drug serial dilutions. 96-well plates (Costar, #3894) were incubated for 24 hours at 37°C, 5% CO2, 5% O2, and 95% N2. After, [3H]-hypoxanthine was added, and plates were incubated for an additional 24 hours. Then, plates were harvested on glass fiber filters (Wallac, #1450-421) using a cell harvester (Tomtec 96, PerkinElmer). Filters were dried, and MeltiLex A melt-on scintillator sheets (PerkinElmer #1450-441) were used to determine [3H]-hypoxanthine incorporation. Radioactivity was measured using a MicroBeta counter (PerkinElmer). Data were normalized using the incorporation by the positive control (infected red blood cells without drug). IC50 values were determined using Excel and Grafit 5. Human biological samples (placenta in pull-downs, red blood cells in IC50 determination) were sourced ethically, and their use in the research was in accordance with terms from an IRB/REC-approved protocol.
For yeast experiments, a single colony was grown in YPD overnight at 30°C to early-log phase. Exactly 0.0005 OD600 of cells were grown in 100µL YPD and exposed to a ten-point dose response of GNF179 or antifungals. After incubation at 30°C for 16 hours (with shaking at 500rpm), OD600 was measured on a plate reader (BioTek Synergy). Percent growth was normalized against the vehicle-only condition (1% DMSO) and plotted against log10 compound concentration using Prism 10, generating curves and IC50 measures. Experiments were conducted in 96-well format, with at least two biological and technical replicates. Background strains tested include GM, a mutant S. cerevisiae strain devoid of 16 ABC multidrug transporter (Suzuki et al., 2011), S. cerevisiae BY4741, and K. phaffii GS115 (Supplementary Table 5).
For experiments with P. falciparum PfSEY1-smV5Tet, parasites were cultured in human O+ red blood cells at 2% hematocrit (BioIVT, Hicksville, NY) in prewarmed parasite growth media (RPMI-1640 with L-glutamine and 25mM HEPES, supplemented with 2g L-1 NaHCO3, 2.6g L-1 AlbuMAX II, 13.6mg L-1 hypoxanthine, and 50mg L-1 gentamicin). Selection of the mutant was maintained with 2.5nM WR99210, and normal wild-type levels of PfSEY1 were maintained with 500nM aTc. Parasitemia was assessed by examining thin blood smears fixed with methanol and subsequently stained with Giemsa under an upright microscope (Olympus CX33). Asynchronous parasites in the presence of aTc and WR99210 were treated with a 12-point dose response of GNF179 (0.003-500nM) in 384-well V-bottom plate format. After 72 hours, growth media was removed and 2µM Syto-61 (Invitrogen, S11343) was added. Plates were read using a flow cytometer (BD FACSCanto) 20 minutes later. Data was visualized using FlowJo (BD Biosciences) and analyzed with Prism 10, generating curves and IC50 values. Experiments were conducted in biological and technical replicates and normalized against the DMSO-only condition.
Whole genome sequencing
Yeast cells were grown overnight to early-log phase and gDNA was extracted using the DNeasy Blood and Tissue kit, following manufacturer’s protocol (Qiagen, 69504). DNA was diluted to 300ng mL-1 for DNA library preparation with Nextera XT (Illumina, FC-131-1024), barcoding with DNA/RNA UD Indexes Set A (Illumina, 20027213), and purification with AMPure XP beads (Beckman, A63880). Resulting purified tagmented DNA was pooled and submitted to the IGM Genomics Center at UCSD for Illumina MiSeq sequencing. Paired-end raw sequence reads were aligned to K. phaffi GCF_000027005.1 reference genome using BWA-MEM (Li & Durbin, 2009) and preprocessed following standard GATK version 3.5 protocols (McKenna et al., 2010; Van der Auwera et al., 2013). SNVs and indels were called with GATK HaplotypeCaller and filtered based on the following exclusion criteria: quality score (QUAL) <500, filtered depth (DP) <7.
Thermal shift assay
Solubilized protein lysate from 400 OD600 early-log phase yeast cells was divided into three microcentrifuge tubes – one containing 20mM GNF179, another containing 20mM artemisinin, and a third containing DMSO. Each tube was then further divided into six PCR tubes, one for each temperature condition (40°C-65°C). Each tube was heated for two minutes, and then chilled for two minutes. Centrifugation at 100,000g for 20 minutes in 4°C removed denatured protein aggregates. Analysis of the supernatants was done by SDS-PAGE and subsequent Western blot with mouse anti-myc and rabbit anti-ScATP2 antibodies. Intensities were quantified using ImageJ and analysis was done using Prism 10.
Molecular docking
To generate the theoretical PvSEY1 structure, protein sequences were processed in ColabFold v1.5.1 (Mirdita et al., 2022), using CaSEY1 (RCSB: 5CA9) as a template. The highest scoring model was then analyzed using MGLTools (Eberhardt et al., 2021; Trott & Olson, 2010). The protein structure was stripped of water molecules, then given Kollman charges and polar hydrogen atoms. Molecular structure for GTP or KAF156 was then uploaded. Autogrid 4 was executed around the GTPase domain or the C-terminal tail of predicted PvSEY1 structure. Autodock 4 was then used to fit the ligand into these regions, generating a score for each position. PyMol was used to create protein models, noting hydrogen bonds between ligand and amino acid residues within five angstroms.
Surface plasmon resonance
Full-length P. knowlesi SEY1 (His-tagged) purchased from MyBioSource was processed on a Biacore 3000 by Profacgen. PkSEY1-His ligand was immobilized on NTA sensor chips (washed with 350mM EDTA and 500mM NiSO4 prior) and washed with 0.05% surfactant P20 in 1x PBS to reach 1300 resonance units (RU). Various concentrations of the GNF179 analyte were then injected using the following parameters: 120 second contact time, 0.5 second dissociation contact time, 30µL min-1 flow rate. To obtain the KD, data were analyzed using BIAevaluation software.
GTPase assay
GTPase activity was measured by quantifying free phosphate production using the EnzChek Phosphate assay kit (Invitrogen, E6646). The 100µL reactions comprise 50µL reaction buffer (20% glycerol, 2mM EDTA, 4mM β-mercaptoethanol, 5mM MgCl2, 50mM HEPES, 200mM KCl), 20µL MESG, and 1µL PNP; the remaining 29µL is composed of the following: GTP, compound or vehicle, 1µM enzyme. A360 measurements were taken after 30 minutes on a plate reader (BioTek Synergy), kept at 37°C. Free phosphate generated was plotted against increasing GTP or compound concentration. Assays were done in 96-well format, with biological and technical replicates. Analysis was done using Prism 10.
Ultrastructural expansion microscopy
P. falciparum 3D7 parasites expressing Cas9 were tightly synchronized using a combination of density separation by 60% Percoll gradient and subsequent selective osmotic lysis after two hours of incubation to select for newly invaded parasites. Cultures were then treated with 25nM DMSO or GNF179 for 16 hours before samples at approximately 1% hematocrit were settled on poly-d-lysine-coated 12mm round coverslips for 30 minutes at 37°C, and then fixed by 4% PFA in PBS for 20 minutes at 37°C. Coverslips were washed three times with PBS before treating with 500µL of 1.4% v/v formaldehyde and 2% v/v acrylamide in PBS. Samples were incubated overnight at 37°C. Gelation, denaturation, staining, and expansion of the gels were performed as previously described (Liffner & Absalon, 2021). Gels were stained with rabbit anti-PfBIP (gifted by the Dvorin Lab at Harvard Medical School, 1:2000) for ER or rabbit anti-PfERD2 (gifted by the Muralidharan Lab at the University of Georgia, 1:2000) for Golgi. Secondary antibody staining with goat anti-rabbit Alexa Fluor 488 (Thermo Scientific, A-11034, 8µg mL-1), NHS Ester Alexa Fluor 405 (Thermo Scientific, A30000, 8 µg mL-1), and SYTOX Deep Red (Thermo Scientific, S11380, 1µM) were used to stain proteins and DNA. Stained gels were imaged using a Zeiss LSM900 AxioObserver with an Airyscan 2 detector and 63x Plan-Apochromat objective lens with numerical aperture of 1.4. Images were analyzed using Zen Blue 3.5 software (Zeiss, Oberkochen, Germany) using 3D Airyscan processing at moderate filter strength. Images shown are maximum intensity projections of between 10 and 30 z-slices of the entire image. Nucleolus and ER area was calculated by taking the area of the projection that is showing the largest surface area of the respective signal. Distance of Golgi to nucleus was calculated by measuring the 3D distance of the closest point of the Golgi signal to that of the nuclear signal. Analysis was done on Prism 10.
Parasite growth assay
A culture of P. falciparum PfSEY1-smV5Tet grown under 2.5nM WR99210 selection pressure and 500nM aTc was washed with prewarmed parasite growth media three times to remove aTc. Quantitative growth assays were performed in 96-well V-bottom plates using asynchronous parasites set up in triplicate and cultured with 5, 25, 50 or 500nM aTc. All cultures, which started at 1% parasitemia, were checked daily by blood smears.