In the major human fungal pathogen Candida albicans, potassium (K⁺) channels fine-tune ionic balance under stressful environmental conditions, contributing to colonization of the human host. Two-pore domain, outwardly rectifying potassium (TOK) channels, uniquely found in fungi, remain insufficiently characterized despite early evidence implicating them in diverse intracellular processes essential for cellular growth and viability, suggesting their potential as antifungal targets. Here, we describe the first atomic resolution structure of a fungal potassium channel - TOK1 from C. albicans (CaTOK) - revealing a unique architecture defined by eight transmembrane helices and a membrane topology distinct from all other known K⁺ channel classes. The first four transmembrane helices form a tetraspanin-like bundle – the TOK auxiliary subunit-like channel (TALC) domain – that can sense external stimuli and has unexpected structural homology to auxiliary subunits of human neuronal ion channels. The TOK1 pore features an inner helical gating mechanism with ‘up’ and ‘down’ conformations analogous to those of mammalian dimeric K⁺ channels. The K⁺ selectivity filter of CaTOK exhibits atypical ion coordination at elevated K⁺ concentrations. Finally, a structured cytosolic C-terminal bundle directly interacts with the TOK1 pore helices and TALC domain, establishing an intramolecular network likely important for maintaining overall channel architecture and modulating gating. These findings provide a structural framework for understanding TOK channel activity and lay the groundwork for future studies on fungal ion homeostasis, pathogenicity, and therapeutic development.

Construct Design

The CaTOK sequence was amplified from pMAX vectors described in Manville et al. (2024) and subcloned into a pFastBac vector (Gibco) containing a Human Rhinovirus 3C protease cleavage site (HRV 3C) and a Twin-Strep-tag (2x-strep) on the C-terminus for baculovirus expression. The CaTOK sequence was adjusted to correct for alternative CTG codon usage in C. albicans (CaTOK), which encodes for a serine instead of a leucine and was not accounted for in the original clone. The final plasmid, pFB CaTOK2x-strep, included two leucine to serine substitutions at residues 350 and 385 was subsequently used for structural studies (NCBI: XP_712629.2). Characterization of CaTOK mutants was designed in the original CaTOK pMAX plasmid (CaTOK Xenopus codon usage) and synthesized by Genscript. To evaluate whether the leucine to serine substitution at residues 350 and 385 altered the CaTOK biophysical properties previously reported, we subcloned the corrected, untagged CaTOK sequence into pMAX (CaTOK Candida codon usage) and performed two-electron-voltage-clamp recordings. The corrected CaTOK displayed nearly identical biophysical properties, mixed selectivity for Na+ and K⁺, and response to external acidification.

Multiple Sequencing Alignments

Sequences used were gathered from the NCBI BLASTP (protein-protein search) database using CaTOK primary sequence as a bait across known pathogenic fungal species or obtained from previously reported literature. Sequences were verified using AlphaFold 3 generated models of the structure for the 2P/8TM consensus and selectivity filter motifs. SnapGene software was used to perform alignments using Clustal Omega sequence alignment algorithm.

Baculovirus Generation, Expression, and Purification of CaTOK

For bacmid generation, the pFastBac CaTOK2x-strep was transformed into DH10Bac E. coli (Thermo Fisher) and selected by streaking on blue-white colony selection plates. Bacmid DNA was amplified, purified by ethanol precipitation, and transformed into Sf9 cells using CellFectin II (Thermo Fisher) and ESF 921 insect cell culture media (Expression Systems) in a six-well plate for baculovirus generation. The supernatant was collected after a 4 day incubation at 27°C containing baculovirus (P0) and filtered. P0 virus was immediately used to infect a 50 mL suspension of SF9s in log-phase to generate higher titer P1 virus. After 4 days of expression at 27°C and 120 RPM, P1 virus was collected, filtered, and stored with 5% FBS (Gemini Bio Products). A final amplification of the baculovirus was generated (P2) in high quantity (200-500 mL) and stored with 5% FBS. For protein expression and purification, P2 virus (4% of total cell suspension volume) was used to infect SF9 cells in a 1 L culture. Infected cells were incubated for 48 hrs at 27°C and 120 RPM. Cells were then harvested at 600 x g, flash-frozen and stored at -80°C. Frozen cell pellets were thawed and resuspended in solubilization buffer containing 150 mM KCl, 100 mM Tris-HCl pH 7.5, 1 mM EDTA, 1:50 dilution of Protease Inhibitor Cocktail Set III, EDTA-free (MilliporeSigma), 1 mM 4-(2-Aminoethyl)-benzenesulfonylfluoride hydrochloride (AEBSF, Gold Biotechnology), 1 mM benzamidine hydrochloride monohydrate (benzamidine, Gold Biotechnology), and 4 mM DNase I (Gold Biotechnology). Resuspended pellets were Dounce homogenized and spun at 200,000 x g for 2 hours at 4°C to isolate the membrane pellet. The pellet was resuspended in solubilization buffer containing 10 mM n-dodecyl-β-D-maltopyranoside (DDM, Anatrace) and 1 mM cholesteryl hemisuccinate (CHS, Anatrace) and spun at 40,000 x g for 45 minutes at 4°C. Lysate was then syringe filtered using 0.45 µm filter and poured over a gravity column containing a 1 mL bed volume of Strep-Tactin XT 4 Flow Resin (IBA Lifesciences) three times. Wash buffer containing 150 mM KCl, 100 mM Tris-HCl pH 7.5, and 1 mM DDM was used to wash the column twice. Elution buffer containing 75 mM KCl, 1x BXT Buffer (IBA Lifesciences), 1 mM DDM, and 0.1 mM CHS was pipetted up and down on the closed column and left at room temperature for 1 hour. Eluent was collected and two additional washes with elution buffer were concentrated using a 100 kDa concentrator (MilliporeSigma). Sample was spin filtered using 0.45 µm cellulose acetate Spin-X filter (Costar) and further purified by size exclusion chromatography (SEC) using a Superdex 200 Increase 10/300 GL (Cytiva) in a SEC buffer containing 150 mM KCl, 20 mM HEPES pH 7.0, 1 mM DDM, and 0.1 mM CHS. This final step of purification was also performed to achieve buffer exchange of CaTOK into KCl and HEPES pH 7.0, allowing for structural analysis of CaTOK under elevated K⁺ conditions at neutral pH.

Cryo-EM Sample Preparation and Data Collection

Purified CaTOK_2x-strep (3 µL) was pipetted onto a R1.2/1.3, 400 mesh copper Quantifoil carbon grids after being glow discharged at 15 mA for 1 minute using a PELCO easiGlow (Ted Pella). Vitrification of the sample was performed using a Vitrobot Mark IV (set to 4°C, 100% humidity, 2.5 s blot time, -5 blot force, 5 s wait time) (Thermo Fisher) into liquid ethane. Grids were screened and imaged using a 300 keV Titan Krios Microscope with a high-brightness field emission gun (XFEG), Gatan Image Filter BioQuantum energy filter, and Gatan K3 camera at the Janelia Research Campus in Ashburn, VA. Images were collected with a spherical aberration of 2.7 mm, super resolution magnification of 105,000 x with a super-resolution pixel size of 0.4135 Å, defocus range of -0.8 to -2 µm, energy filter slit width of 20 eV, and total e- dose of 50 e- per Å2 (Table 1). Two CaTOK cryo-EM datasets were collected from separate grids that were prepared with an identical sample and combined before data processing.

Cryo-EM Data Processing

Motion correction was performed in Relion 5.0 using GPU-based MOTIONCOR255. Contrast Transfer Function (CTF) was computed for micrographs using CTFFIND4.1, and subsequent automated particle picking was carried out using Topaz. Majority of data processing was performed in CryoSPARC V4.5. To determine the oligomeric state of CaTOK, we performed the initial three-dimensional refinement without symmetry imposed (‘C1 symmetry’). From this initial electron density map (at 3.2 Å resolution), we observed clear two-fold symmetry within the map, further demonstrating that CaTOK assembles as a dimer. Subsequent refinement jobs were performed using C2 symmetry. During an initial 3D classification strategy to potentially identify conformational states of the channel (sorting into 3 classes), we isolated the ‘up’ and ‘down’ states of helix S8 but also observed a clear dissociation of the C-terminal bundle from a subset of particles. To improve the resolution of the C-terminal bundle and minimize the inclusion of these particles in the final reconstruction, we performed a second 3D classification strategy, this time sorting into 10 classes. For this analysis, we identified three classes with well-resolved C-terminal bundle density and combined these to determine the overall structure of CaTOK at 2.6 Å resolution. Within these three classes, we again observed distinct movement in helix S8, isolating ‘up’ and ‘down’ states, which we refined independently to 2.8 Å and 3.1 Å resolution, respectively. 

Model Building, Refinements, and Structural Analysis

An AlphaFold 3-predicted model of CaTOK was used as a starting template for alignment and model building in ChimeraX and Coot using the sharpened final average map. This model was further refined using phenix.refine with the unsharpened final average map. Model refinement was assessed using MolProbity. The CaTOK ‘up’ and ‘down’ conformations were built and refined using the final average model and sharpened ‘up’ and ‘down’ state electron density maps, following the same refinement procedure as described. Figures were generated using Pymol, ChimeraX, Chimera, Adobe Illustrator, and Biorender.com. Pore radius measurements were calculated using HOLE.

Immunoblotting and Crosslinking Experiments

For crosslinking experiment to assess oligomeric state, purified CaTOK_2x-strep was incubated with 2.5 mM, 7.5 mM, and 25 mM bismaleimidohexane (BMH, Thermo Fisher) for 1 hour at room temperature. The reaction was subsequently quenched by adding 20 mM DTT and incubating for 15 minutes. For immunoblotting, samples were mixed with 1x Laemmli sample buffer with 250 mM DTT, separated by SDS-PAGE on a 4-20% Mini-PROTEAN TGX Gel (Bio-Rad), and transferred to a polyvinylidene difluoride (PVDF) membrane. CaTOK_2x-strep was detected using a mouse monoclonal anti-strep tag primary antibody (1:10,000 dilution, IBA Lifesciences) and a goat anti-mouse secondary antibody conjugated with horseradish peroxidase (1:10,000 dilution, 806 Jackson ImmunoResearch). HRP-conjugated antibodies were detected using Western ECL Substrate solution (Bio-Rad) and the resulting chemiluminescent signal was imaged with an iBrightFL7000 (Thermo Fisher).

Vesicle Reconstitution and Potassium Flux Assays

CaTOK_2x-strep was purified using the same methodology as described above, but with a SEC buffer containing 3 mM n-decyl-β-D-maltopyranoside (DM, Anatrace). The purified CaTOK_2x-strep was then reconstituted into lipid vesicles containing a 3:1 lipid mixture of POPE (1-palmitoyl-2-oleoyl813 sn-glycero-3-phosphocholine, Avanti Research) and POPG (1-palmitoyl-2-oleoyl-sn-3-phospho814 (1’-rac-glycerol, Avanti Research) solubilized in 8% n-Octyl-β-D-maltopyranoside (OM). The final protein:lipid ratio of 1:100 (w/w) was used for reconstitution, resulting in a final concentration of approximately _{0.1 mg/mL CaTOK}2x-strep~ and 10 mg/mL lipids. An empty vesicle control was alsoprepared without the addition of protein. Detergent was removed by dialysis over 3 days using a buffer containing 150 mM KCl and 10 mM HEPES pH 7.0. The pH was adjusted using N-methyl D-glucamine (NMDG) to minimize the presence of small inorganic cations. The dialysis buffer was exchanged into fresh buffer every 24 hours. After dialysis, samples were aliquoted, flash-frozen in liquid nitrogen and stored at -80°C. The potassium flux assay was adapted from the method described by Miller and Long, 2012 and other such assays. In brief, vesicles were thawed, sonicated for 5 seconds and diluted 1:50 in a flux assay buffer (FAB) containing 150 mM NMDG-Cl, with 10 mM HEPES pH 7.0 (as prepared above), 0.5 mg/mL bovine serum albumin, and 4 µM 9-amino-6-chloro-2-methoxyacridine (ACMA, MilliporeSigma). This creates a potassium gradient across the vesicles as there is no additional potassium added to the external buffer. Data were collected using a Perkin Elmer Envision Multimode Plate Reader with Alpha 680nm laser and EnVision 1.14 836 software. Measurements were collected as technical triplicates at 30 second intervals for a total of 870 seconds for the experiment with excitation and emission set to 410 nm and 490 nm. A final concentration of 1 µM Carbonyl cyanide m-chlorophenyl hydrazone (CCCP, MilliporeSigma) was added 120 seconds after sample was incubated into FAB and pipetted gently. A final concentration of 20 nM Valinomycin (MilliporeSigma) was added at 630 seconds to release potassium from vesicles. Relative fluorescence was calculated as the fluorescence at time t divided by the fluorescence at time 0. Technical replicates were averaged to give a mean fluorescence at each time point. Data were graphed using GraphPad Prism 10.5.

Electrophysiology

Channel subunit cRNA preparation and Xenopus laevis oocyte injection.

We generated cRNA transcripts encoding TOK channels from Candida albicans (wild-type and mutant), by in vitro transcription using the T7 mMessage mMachine kit (Thermo Fisher Scientific), after vector linearization, from cDNA sub-cloned into plasmids (pMAX) incorporating Xenopus laevis β-globin 5’ and 3’ UTRs flanking the coding region to enhance translation and cRNA stability. Mutant CaTOK channel cDNAs were generated by Genscript (Piscataway, NJ, USA). We injected defolliculated stage V and VI Xenopus laevis oocytes (Xenopus1, Dexter, MI, US) with 1-10 ng of CaTOK cRNA and incubated the oocytes at 16 ^oC in ND96 oocyte storage solution containing penicillin and streptomycin.

Two-electrode voltage clamp (TEVC)

We performed TEVC at room temperature using an OC-725C amplifier (Warner Instruments, Hamden, CT) and pClamp 9.2 software (Molecular Devices, Sunnyvale, CA) 1-2 days after cRNA injection. Oocytes were placed in a small-volume oocyte bath (Warner) and viewed with a dissection microscope. Oocytes expressing CaTOK channels were recorded in 4 mM extracellular KCl bath solution containing (in mM): 96 NaCl, 4 KCl, 1 MgCl2, 1 CaCl2 and 10 HEPES (pH 7.5). Pipettes were of 1-2 MΩ resistance when filled with 3 M KCl. Current-voltage relationships were recorded from a holding potential of -80 mV, with prepulse voltages elicited from -80 to +40 mV at 10 mV intervals, followed by a -30 mV tail pulse. Peak outward current for each voltage step was measured at the end of the depolarizing pulse, typically at +40 mV, corresponding to the plateau phase of CaTOK outward rectification. We plotted normalized tail currents for CaTOK versus prepulse voltage and fitted with a single Boltzmann function. where g is the normalized tail conductance, A1 is the initial value at -∞, A2 is the final value at +∞, V_1/2 is the half-maximal voltage of activation and Vs the slope factor. We analyzed data using Clampfit 11.2 (Molecular Devices) and GraphPad Prism software (GraphPad, San Diego, CA, 877 USA), stating values as mean ± SEM. Statistics were calculated using a One-Way ANOVA with Brown-Forsythe and Welch ANOVA tests. Additionally, Dunnett's T3 multiple comparison test with individual variance for each comparison was conducted.