Electrophysiological characterization of a schistosome transient receptor potential channel activated by praziquantel

Abstract

Ion channels have proved to be productive targets for anthelmintic chemotherapy. One example is the recent discovery of a parasitic flatworm ion channel targeted by praziquantel (PZQ), the main clinical therapy used for treatment of schistosomiasis. The ion channel activated by PZQ – a transient receptor potential ion channel of the melastatin subfamily, named TRPM_PZQ – is a Ca²⁺-permeable ion channel expressed in all parasitic flatworms that are PZQ-sensitive. However, little is currently known about the electrophysiological properties of this target that mediates the deleterious action of PZQ on many trematodes and cestodes. Here, we provide a detailed biophysical characterization of the properties of Schistosoma mansoni TRPM_PZQ channel (Sm.TRPM_PZQ) in response to PZQ. Single channel electrophysiological analysis demonstrated that Sm.TRPM_PZQ when activated by PZQ is a non-selective, large conductance, voltage-insensitive cation channel that displays distinct properties from human TRPM paralogs. Sm.TRPM_PZQ is Ca²⁺-permeable but does not require Ca²⁺ for channel gating in response to PZQ. TRPM_PZQ from Schistosoma japonicum (Sj.TRPM_PZQ) and Schistosoma haematobium (Sh.TRPM_PZQ) displayed similar characteristics. Profiling Sm.TRPM_PZQ responsiveness to PZQ has established a biophysical signature for this channel that will aid future investigation of endogenous TRPM_PZQ activity, as well as analyses of endogenous and exogenous regulators of this novel, druggable antiparasitic target.

Graphical Abstract

1. Introduction

Over a billion people worldwide are at risk for infection by various neglected tropical diseases (NTDs). One of the most burdensome of these diseases is schistosomiasis (Colley et al., 2014; McManus et al., 2018), caused by infection by parasitic flatworms known as schistosomes (Schistosoma spp.). These blood flukes cause disease due to their prolific egg-laying capacity, as eggs deposited in host tissues trigger a robust inflammatory and fibrotic response that propels a suite of pathological morbidities including periportal fibrosis, portal hypertension and hepatospleanomegaly as well as sequelae encompassing ascites, weakness, undernutrition and growth retardation (Colley et al., 2014; McManus et al., 2018).

Schistosomiasis has been targeted for elimination as a public health problem by 2030 in the most recent World Health Organization (WHO) ‘road map’ (W.H.O., 2020). Inherent to this strategy is the continued efficacy of praziquantel (PZQ), the key clinical drug used for treating schistosome infections. PZQ causes a rapid, spastic paralysis of schistosomes, with concomitant surface damage that mediates in vivo vulnerability of these blood flukes to immune attack (Brindley and Sher, 1990). PZQ has served for over four decades as the essential clinical therapy for treating schistosomiasis and is effective against many other parasitic flatworm infections (Andrews et al., 1983). Moreover, as a cheap, efficacious drug – it has proved a stalwart of mass drug administration campaigns to decrease the incidence of schistosomiasis.

Surprisingly, the parasite target of PZQ has remained undefined since its discovery in the 1970s (Gonnert and Andrews, 1977; Thomas and Gonnert, 1977). A promising candidate was recently identified, unmasked as a transient receptor potential ion channel of the melastatin subfamily (TRPM), named TRPM_PZQ (Park et al., 2019). TRPM_PZQ is present in all parasitic flatworms that show sensitivity to PZQ. TRPM_PZQ is a large TRP channel (~1.5 and 2 times longer than human TRPM2 and human TRPM8), and it is potently and stereoselectively activated by PZQ (channel sensitivity ranges from tens to hundreds of nanomolar (Park et al., 2019; Park et al., 2021)). TRPM_PZQ sensitivity to PZQ correlates with the sensitivity of different parasites to PZQ, most dramatically shown by the insensitivity of Fasciola spp. TRPM_PZQ to PZQ owing to variation of a single amino acid residue within the transmembrane ligand binding pocket (Park et al., 2021). Fasciola spp. infections are known to be refractory to PZQ.

Since Schistosoma mansoni TRPM_PZQ (Sm.TRPM_PZQ) was only recently discovered (Park et al., 2019), there is much to learn about the properties of this channel and how it functions during the parasitic lifecycle. Such insight also extends to other members of the schistosome TRP channel family, which are also not presently well defined. While some clues may come from extrapolation from the properties of human TRP channels, which have been thoroughly studied (Vangeel and Voets, 2019; Huang et al., 2020; Koivisto et al., 2022), applying a vertebrate-centric classification to parasitic flatworm TRPs is anthropomorphic. Parasitic flatworm TRPM channels must be characterized individually to resolve their properties and understand how their function is customized to the unique physiological demands of the parasitic lifecycle. Additionally, Sm.TRPM_PZQ is a validated, druggable target, providing further impetus to characterize the regulation of this ion channel. In conjunction with drug screening initiatives (Chulkov et al., 2021), this effort will hopefully realize mechanistically characterized, small molecule tools for modulating parasitic flatworm biology, as well as novel chemotypes that may show promise as therapeutic leads.

In this paper, we perform the first, detailed electrophysiological characterization of Sm.TRPM_PZQ properties at the single channel level in response to the application of PZQ. This yields a biophysical signature of the channel that will be useful for resolving the activity of Sm.TRPM_PZQ in worms exposed to PZQ, and comparison with TRPM_PZQ responses evoked by endogenous stimuli throughout the schistosome lifecycle.

2. Materials and methods

2.1. Materials

All chemicals were from Sigma (St. Louis, MO, USA). The PZQ enantiomers, (R)-PZQ and (S)-PZQ, were resolved according to the methods of Woelfle et al. (2011).

2.2. Cell culture

All cell culture reagents were from Thermo Fisher Scientific (Waltham, MA, USA). HEK293 cells (ATCC CRL-1573.3) were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), penicillin-streptomycin-glutamine (PSG, penicillin (100 units/ml)), streptomycin (100 μg/ml), and L-glutamine (290 μg/ml)). Total RNA from adult male and female Schistosoma haematobium (Egyptian Strain, NR-31801) was sourced from the Schistosomiasis Resource Center at the Biomedical Research Institute (Rockville, MD, USA). The full-length sequence of S. haematobium TRPMPZQ (Sh.TRPMPZQ) was determined from S. haematobium total RNA using reverse transcription-PCR (RT-PCR) and 5’-RACE. 5′-RACE was performed using the SMARTer RACE 5′3′ Kit (Takara Bio USA, Mountain View, CA, USA). The full-length sequences for the schistosome TRPM_PZQ orthologs are detailed in Rohr et al. (2022). For expressing schistosome TRPM_PZQ channels, codon-optimized cDNAs (Genscript, Piscataway, NJ, USA) for Sm.TRPM_PZQ, Schistosoma japonicum TRPM_PZQ (Sj.TRPM_PZQ), or Sh.TRPM_PZQ were subcloned into the mammalian expression vector pcDNA3.1 between the NotI and EcoRI sites, and individual plasmids were transiently transfected into HEK293 cells using Lipofectamine-2000.

2.3. Electrophysiology

HEK293 cells were plated on glass coverslips and the following day co-transfected with plasmids encoding a GFP and Sm.TRPM_PZQ, Sj.TRPM_PZQ or Sh.TRPM_PZQ. Prior to recording, coverslips were secured within a recording chamber mounted on an Olympus BX51WI upright microscope. A multiClamp 700B amplifier and Digidata 1440A digitizer (Molecular Devices, San Jose, CA, USA) were used for electrophysiological recordings. Signals were passed through an 8-pole Bessel low pass filter at 1 kHz and sampled at 10 kHz. Data analysis was performed using Clampfit 10 software (Molecular Devices). Patch pipettes were made of borosilicate glass (BF150-110-10, Sutter Instrument, Novato, CA, USA) pulled on a vertical puller (Narishige, Amityville, NY, USA, Model PC-10) to a resistance of 8–10 MΩ for single channel recordings or 5–8 MΩ for ‘whole-cell’ configuration. Ionic selectivity reversal potential was measured as previously described (Owsianik et al., 2006). The solutions (bath solution and pipette solution) for single channel recordings were symmetrical and contained: 145 mM NaCl, 10 mM HEPES, 1 mM EGTA at pH 7.4, unless specified otherwise. The bath solution was replaced for ion substitution experiments as specified in the relevant figure legends. All recordings were performed at room temperature. For assays investigating channel pore size, the bath solution was perfused with different molecular weight polyethylene glycols (PEGs, 10% w/v). Hydrodynamic radii (R_H) were calculated using R_H = 0.06127(MW)^0.3931nm (Krasilnikov, 2002), where MW represented the molecular weight of the polymer in Da (Sabirov and Okada, 2004). Cation-anion selectivity was determined from transport numbers (t₊ [cations] and t₋ [anions], where t₊+t₋=1) derived from a modified Goldman-Hodgkin-Katz’s equation: $t_{+}=[C_1-C_2\text{exp }(V_{rev}^{\mathit{\text{Corr}}}F/RT)]/[(C_1-C_2)(1+\text{exp }(V_{rev}^{\mathit{\text{Corr}}}F/RT))]$, where C₁ and C₂ represent electrolyte concentrations in the bath and pipette; $V_{rev}^{\mathit{\text{Corr}}}$, reversal potential (V_rev) corrected to the liquid junction potential (LJP, (Barry, 1994)); R, universal gas constant; T, the absolute temperature in Kelvin; and F, Faraday’s constant (Alcaraz et al., 2004). Monovalent cation permeability (P_X) relative to sodium permeability (P_Na) was estimated using: $P_X\text{:}{P}_{Na}=\text{exp }(V_{rev}^{\mathit{\text{Corr}}}F/RT)$ where P represents permeability of ion species X (Na, Cs, K or Rb). The bath (XCl) and pipette (NaCl) solutions were symmetrical (145mM). Divalent cation selectivity was derived from P_Ca:P_Na = ([Na⁺]_i/(4[Ca²⁺]_e))exp (V_revF/RT)(exp (V_revFRT) + 1), where [Na⁺]_i and [Ca²⁺]_e represent Na⁺ (internal) and Ca²⁺ (external) concentrations (Owsianik et al., 2006). Before experimentation, the Ca²⁺-to-Na⁺ liquid junction potential was adjusted using MultiClamp 700B. Statistical comparisons were made after ANOVA using Tukey’s range test, performed in Origin (OriginLab Corporation).

2.4. Fluorescence measurements of membrane potential

A fluorescent membrane potential (E_m) dye (Molecular Devices, R8042) was used to measure membrane potential by detecting changes in fluorescence in real-time using a Fluorescence Imaging Plate Reader (FLIPR^TETRA, Molecular Devices) instrument. A loading buffer containing a ‘blue’ E_m dye was prepared at 2x by dissolving the contents of a reagent vial with 10 mL of assay buffer (1x HBSS, 20 mM HEPES, pH 7.4). An equal volume of loading buffer was added to each well on a multi-well plate. For a 96-well plate, 50 μL of loading buffer was added to 50 μL of cells and media. Dye-loaded plates were incubated for 30 min at 37°C, 5% CO₂, with no wash step after loading, prior to reading on a FLIPR^TETRA instrument. After 20 s of basal read, fluorescence changes were monitored for 10 min at λ_ex=510–545 nm, λ_em=565–625 nm after (±)-PZQ addition. HEK293 cells not expressing Sm.TRPM_PZQ showed no fluorescence response to (±)-PZQ using either a Ca²⁺-sensitive dye (fluo-4) or the E_m dye. In Sm.TRPM_PZQ expressing cells, changes in membrane potential evoked by (±)-PZQ were reported as a ratio, F/F₀ (where “F” represents fluorescence at a specific time and “F₀” represents fluorescence at time = 0). For Ca²⁺, imaging assays were performed using the fluorescent Ca²⁺ indicator ‘Fluo-4 Direct’ (Invitrogen). HEK293 cells were loaded with 2x Fluo-4 Direct dye supplemented with probenecid (2.5 mM) for 30 min at 37°C, followed by an additional 30 min at room temperature in the dark. After dye loading, the fluorometric Ca²⁺ assay was performed at room temperature using methods previously detailed (Park et al., 2019).

3. Results

3.1. Single channel properties of S. mansoni TRPM_PZQ

For electrophysiological analyses, HEK293 cells were co-transfected with cDNA encoding GFP (control), or plasmids encoding GFP and Sm.TRPM_PZQ, and whole cell currents evoked by (±)-PZQ were recorded after 24–48 h. Whereas no responses to (±)-PZQ were observed in GFP-control cells, (±)-PZQ (10 μM) activated an inward current in Sm.TRPM_PZQ expressing cells (Fig. 1A). The inward currents observed in Sm.TRPM_PZQ expressing cells were blocked by subsequent addition of La³⁺ (LaCl₃, 10 mM) as previously shown (Park et al., 2019).

Fig. 1.

Activation of the Schistosoma mansoni ion channel, Sm.TRPM_PZQ, by enantiomers of praziquantel (PZQ). (A) Representative whole-cell transmembrane current recorded from HEK293 cells co-transfected with either GFP and Sm.TRPM_PZQ (Sm.TRPM_PZQ), or GFP alone (Ctrl, purple). Traces show continuous recording before and after the addition of racemic PZQ ((±)-PZQ, 10 μM), and finally in the presence of LaCl₃ (10 mM, grey bar). Pipette and bath solutions were: 140 mM CsMeSO, 1 mM EGTA, 10 mM HEPES-CsOH, pH 7.2 (pipette solutions) and 145 mM NaCl, 10 mM HEPES-NaOH, pH 7.4 (bath solution). The recording voltage was −60 mV. Also shown is the cumulative dataset of whole-cell peak current measurements (mean ± S.E.M., n=6), *** P<0.01. (B) Representative traces evoked by responses to (±)-PZQ in HEK293 cells expressing GFP alone (Ctrl) or GFP and Sm.TRPM_PZQ in response to (±)-PZQ (10 μM then 100 μM) or DMSO (0.1% v/v). (C and D) Representative traces for single channel Sm.TRPM_PZQ activity evoked by (C) (R)-PZQ (10 μM) and (D) (S)-PZQ (100 μM). In (C) (S)-PZQ (10 μM, blue) did not evoke Sm.TRPM_PZQ activity. However, subsequent addition of (R)-PZQ (10 μM, red) to the same patch triggered channel activity. Highlighted regions are expanded to show examples of single channel fluctuations evoked by the different ligands. Recordings were made in the ‘inside-out’ configuration at +40mV. Bath and pipette solutions were symmetrical (145 mM CsCl, 10 mM HEPES-CsOH, 1 mM EGTA, pH 7.4). Arrows indicate the time point of the indicated drug addition to the bath.

For analyses of single channel activity, transmembrane currents were recorded from individual patches excised from GFP-positive HEK293 cells expressing Sm.TRPM_PZQ. Recordings were made under symmetrical conditions (pipette and bath solutions: 145 mM CsCl, 10 mM HEPES-CsOH, 1 mM EGTA, pH 7.4). In excised ‘inside-out’ patches from HEK293 cells transfected with GFP alone, there were no changes in current upon addition of racemic PZQ ((±)-PZQ, 10 μM or 100 μM) to the bath solution (Fig. 1B, top panel). However, in cells expressing Sm.TRPM_PZQ, addition of (±)-PZQ (10 μM) caused clearly distinguishable step-like fluctuations of the transmembrane current (Fig. 1B, bottom panel). Channel activity rapidly increased in response to (±)-PZQ and remained stable over the lifetime of the patch (≤10 min) with little sign of decay or inactivation. Low spontaneous basal activity of the channel was observed in cells expressing Sm.TRPM_PZQ. Application of vehicle (0.1% v/v DMSO) alone did not change transmembrane conductance (Fig. 1B). The effects of the PZQ enantiomers (R)-PZQ and (S)-PZQ were also assessed. Fig. 1C shows that whereas (S)-PZQ (10 μM) failed to evoke a response, addition of (R)-PZQ (10 μM) to the same excised patch elicited single channel fluctuations similar to the activity evoked by (±)-PZQ. However, higher concentrations of (S)-PZQ (100 μM) evoked Sm.TRPM_PZQ mediated currents (Fig. 1D). The data are consistent with the reported stereoselectivity of PZQ enantiomers resolved at Sm.TRPM_PZQ in experiments using fluorescent Ca²⁺ indicators, as well as the known effects of the individual enantiomers of PZQ on schistosome worms (Park et al., 2019).

Single channel recordings were also made from Sm.TRPM_PZQ expressing cells in a ‘cell-attached’ configuration. Here, the behavior of Sm.TRPM_PZQ depended on the polarity of the voltage applied to the patch. Fig. 2A shows the difference between channel activity under symmetrical recording conditions (145 mM NaCl, 10 mM HEPES, 1 mM EGTA, pH 7.4 solution) in the presence of (±)-PZQ (10 μM) at either +80 mV or −80 mV. At the positive voltage, clearly distinguishable step-like fluctuations were observed, while at the negative voltage the signal from the same patch was noisier and single Sm.TRPM_PZQ channel activity was poorly distinguishable. The poorer signal resolution in ‘cell-attached’ mode impaired analysis of the Sm.TRPM_PZQ single channel I-V relationship in ‘cell-attached’ mode at negative voltages. Consistent with this observation, when a ‘cell-attached’ patch was pulled from a cell, the resolution of single channel activity in the ‘inside-out’ configuration improved within minutes at the negative holding potential (−80 mV, Fig. 2B).

Fig. 2.

Single channel resolution of the Schistosoma mansoni ion channel, Sm.TRPM_PZQ, in different recording modes. (A) Comparison of recordings of Sm.TRPM_PZQ activity in the ‘cell-attached’ configuration at +80 mV and −80 mV in response to (±)-PZQ recorded under symmetrical conditions (145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, (±)-PZQ (10 μM), pH 7.4 in both the bath and pipette solutions). (B) Representative recording from a HEK293 cell expressing Sm.TRPM_PZQ was initially performed in ‘cell-attached’ mode and then using the ‘inside-out’ configuration, with the arrow showing the timing of patch excision from the cell surface. Once the patch is removed from the cell and the inner part of the membrane is no longer exposed to cytoplasm, the single channel fluctuation signal became better resolved within several minutes. Insets: enlargements of Sm.TRPM_PZQ single channel fluctuations in the ‘inside-out’ configuration at −80 mV (red dashed region).

3.2. Properties of TRPM_PZQ from other schistosome species

The properties of Sm.TRPM_PZQ were then compared with the properties of the TRPM_PZQ orthologs from S. japonicum (Sj.TRPM_PZQ) and S. haematobium (Sh.TRPM_PZQ). These three species of schistosomes are responsible for the majority of human disease worldwide (Colley et al., 2014; McManus et al., 2018). Single channel activity recordings were made from HEK293 cells in ‘cell-attached’ mode under symmetrical, Ca²⁺-free recording conditions (145 mM NaCl, 10 mM HEPES, 1 mM EGTA, pH 7.4) at +60 mV. Addition of (±)-PZQ (10 μM) to the bath solution triggered single channel activity in expressing each of these wild type TRPM_PZQ constructs (Fig. 3A). Analysis of the current-voltage (I-V) relationship showed a linear relationship with regression analysis (R²>0.99), yielding an average slope conductance of 112±3 pS for Sh.TRPM_PZQ, 132±4 pS for Sj.TRPM_PZQ, and 133±7 pS for Sm.TRPM_PZQ (mean ± S.E., Fig. 3B). In a small number of patches, lower conductance fluctuations were occasionally resolved, similar to observations with human TRP channels (Fernandez et al., 2011; Canul-Sanchez et al., 2018). Calculations of channel open probability P_open, yielded similar values of P_open=0.57±0.10 for Sh.TRPM_PZQ, P_open=0.42±0.16 for Sj.TRPM_PZQ, and P_open=0.39±0.17 for Sm.TRPM_PZQ (Fig. 3C). From these single channel recordings, mean dwell time values (Fig. 3D) and dwell time distributions for TRPM_PZQ channel fluctuations evoked by (±)-PZQ were analyzed (Fig. 3E). The mean open time for each of the TRPM_PZQ orthologs was indicative of a relatively long-lived open state under these recording conditions (mean open time = 65±2 ms for Sh.TRPM_PZQ,110±4 ms for Sj.TRPM_PZQ and 87±11 ms for Sm.TRPM_PZQ, Fig. 3D). These dwell time distributions were each fitted by a single exponential (time constants of 30±1 ms, 40±1 ms and 16±0.1 ms for Sh.TRPM_PZQ, Sj.TRPM_PZQ, and Sm.TRPM_PZQ), indicating TRPM_PZQ activation is independent of the gating of neighboring TRPM_PZQ channels (Fig. 3E). Overall, TRPM_PZQ from each of these schistosome species displayed similar electrophysiological characteristics (linear IV relationship, conductance, and open probability).

Fig. 3.

Electrophysiological characterization of orthologs of the ion channel, TRPM_PZQ, in other schistosome species. (A) Representative current traces for Schistosoma haematobium TRPM_PZQ, Sh.TRPM_PZQ (blue), Schistosoma japonicum TRPM_PZQ, Sj.TRPM_PZQ (red), or Schistosoma mansoni TRPM_PZQ, Sm.TRPM_PZQ (black), showing single channel fluctuations elicited by (±)-PZQ (10 μM). Recordings were made in Ca²⁺-free solution (145 mM NaCl, 10 mM HEPES, 1 mM EGTA, pH 7.4) at +60 mV in a ‘cell-attached’ configuration. (B) Current (I)-voltage (V) relationships for Sh.TRPM_PZQ, Sj.TRPM_PZQ or Sm.TRPM_PZQ single channel activity from (A). Linear regression analysis yielded average slope conductance values reported in the text. (C) Sh.TRPM_PZQ, Sj.TRPM_PZQ and Sm.TRPM_PZQ single channel steady-state open probability (P_open, mean ± S.D.) was recorded under identical conditions. No statistical difference was found (P<0.05) between the open probabilities of the channels. (D) Dwell time (τ, mean ± sem) Sh.TRPM_PZQ (4228 events), Sj.TRPM_PZQ (2246 events) and Sm.TRPM_PZQ (889 events) recorded under the same conditions. *P<0.05; **P<0.01; ***P<0.001. (E) Dwell time distribution histograms of Sh.TRPM_PZQ, Sj.TRPM_PZQ and Sm.TRPM_PZQ channels recorded in ‘inside-out’ mode. A single exponential fitting of the data (R² >0.98) yielded exponential constants reported in section 3.2

3.3. Schistosome TRPM_PZQ does not require Ca²⁺ for channel activation

These single channel recordings (Figs. 1–3) were all performed in a recording solution free from addition of divalent ions, and in the presence of 1 mM EGTA (i.e. Ca²⁺-free conditions). This is an important point as human TRPM2 (Hs.TRPM2), with which TRPM_PZQ shares domain homology (Park and Marchant, 2020), requires Ca²⁺ as an obligatory co-agonist needed for channel gating (Csanady and Torocsik, 2009). As an alternative approach to demonstrate the Ca²⁺-independence of Sm.TRPM_PZQ, we used a fluorescent membrane potential dye to resolve Sm.TRPM_PZQ activation in intact cells. At resting potentials, the membrane potential dye binds to the cell membrane and enters the cell on depolarization, resulting in an increase in fluorescence, mimicking a membrane potential depolarization response. This fluorescent, cell-based assay is therefore an indirect measurement of membrane potential response.

(±)-PZQ activation of Sm.TRPM_PZQ was studied in intact HEK293 cells using either this fluorescent membrane potential dye (E_m) or a fluorescent Ca²⁺ indicator (Fluo-4 direct). In either case, baseline fluorescence was recorded for 20 s before stimulation with various concentrations of (±)-PZQ (0.01–3 μM). In the presence of Ca²⁺, (±)-PZQ-evoked a dose-dependent signal resolved using either reporter dye (Fig. 4A and B). However, whereas responses resolved by Fluo-4 were abolished in the absence of extracellular Ca²⁺ (using media containing 1 mM EGTA, Fig. 4C), (±)-PZQ-evoked fluorescence changes were still apparent using the membrane potential dye (Fig. 4D). There were no fluorescence changes evident in untransfected HEK293 cells in response to (±)-PZQ (0.01–3 μM, Fig. 4E). Sm.TRPM_PZQ activity can therefore be resolved in the presence of Ca²⁺ (using a fluorescent Ca²⁺ indicator) or by using a membrane potential dye that resolves cellular depolarization following Sm.TRPM_PZQ activation in the absence of Ca²⁺.

Fig. 4.

Activity of the Schistosoma mansoni ion channel, Sm.TRPM_PZQ, resolved using different fluorescence reporter assays. (A and B) Fluorescence traces depicting a concentration-dependent increase in fluorescence after stimulation of Sm.TRPM_PZQ expressing HEK293 cells with varying concentrations of (±)-PZQ in normal HBSS (with Ca²⁺) solution, resolved using either (A) a Ca²⁺ indicator (fluo-4 direct) or (B) a membrane potential dye (E_m reporter). (C and D) Responses to (±)-PZQ in Sm.TRPM_PZQ expressing cells in Ca²⁺-free HBSS solution (supplemented with 1 mM EGTA) resolved with (C) fluo-4 and (D) the membrane potential dye. (E) Representative fluorescence traces from control, untransfected HEK293 cells incubated with the E_m reporter in response to varying concentrations of PZQ (0.1 to 100 μM).

3.4. Ion permeation properties of Sm.TRPM_PZQ

While Sm.TRPM_PZQ is a Ca²⁺-permeable channel under physiological conditions (Park et al., 2019, 2021; Chulkov et al., 2021), there has been no quantitative characterization of the TRPM_PZQ ion permeation signature. Using an ‘inside-out’ configuration, we measured the I-V relationship for Sm.TRPM_PZQ under various electrolyte gradients. First, Sm.TRPM_PZQ single channel fluctuations were evoked by (±)-PZQ (10 μM) under symmetrical solutions (145 mM KCl, 10 mM HEPES, 1 mM EGTA, pH 7.4) with 10 mV/s voltage ramps applied to obtain the I-V relationship. Then, the bath solution was substituted with a 0.1x KCl gradient (14.5 mM KCl, 10 mM HEPES, 1 mM EGTA, 10 μM PZQ, pH 7.4) and the I-V relationship was rerecorded. Fig. 5A shows a shift in the I-V plot towards negative voltages under a 0.1x KCl gradient. The reversal voltage, V_rev, was close to zero for recordings made in symmetrical solutions and −76.2±1.8 mV under the 0.1x KCl gradient (mean ± S.D., n=5). Reversal potential calculations (see section 2) yield t₊=1.05 and t₋=−0.05, indicating a virtually ideal Sm.TRPM_PZQ selectivity to cations over anions.

Fig. 5.

Examination of the permeation profile of the Schistosoma mansoni ion channel, Sm.TRPM_PZQ. (A) Representative current traces (voltage ramp at 10 mV/s) of Sm.TRPM_PZQ activity evoked by (±)-PZQ (10 μM, bath solution) from an ‘inside-out’ recording from patches excised from Sm.TRPM_PZQ expressing HEK293 cells. Recording solutions were: black, symmetrical: 145 mM KCl, 10 mM HEPES, 1 mM EGTA, pH 7.4; red, bath solution replaced with (in) 14.5 mM KCl, 10 mM HEPES, 1 mM EGTA, 0.01 mM PZQ, pH 7.4 solution. Inset: Sm.TRPM_PZQ single channel activity without (black) and with (red) electrolyte concertation gradient. (B) Representative current traces (voltage ramp at 10 mV/s) showing activity of Sm.TRPM_PZQ in response to (±)-PZQ (10 μM, bath) recorded in an ‘inside-out’ configuration. Pipette solutions were: 145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, pH 7.4 or 100 mM CaCl₂, 10 mM HEPES-NaOH, pH 7.4. Bath solutions were: 145 mM X⁺ (Na⁺, Cs⁺, K⁺, Rb⁺, or Na⁺)Cl, 10 mM HEPES-X⁺OH, 1 mM EGTA, (±)-PZQ (10 μM), pH 7.4.

Sm.TRPM_PZQ permeability ratios for different cations were then assessed. Currents were measured during a voltage ramp (10 mV/s) from an ‘inside-out’ patch configuration, where the bath solution contained different monovalent cation species (bath solution, 145 mM X⁺Cl (where X⁺ represents Na⁺, Cs⁺, K⁺, or Rb⁺, 10 mM HEPES-X⁺OH, 1 mM EGTA, 10 μM (±)-PZQ, pH 7.4). The pipette solution was 145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, pH 7.4. Fig. 5B shows representative I-V traces recorded under different bath salt solutions compositions. Cation currents flowed both outward and inward without displaying significant shifts to negative or positive voltages. Reversal voltage (V_rev) data are shown in Table 1. Permeability ratios were calculated as P_Na : P_Cs : P_K : P_Rb = 1: 1.26 : 1.24 :1.28. Lastly, P_Ca:P_Na was estimated using a similar protocol with solutions: (pipette) 100 mM CaCl₂, 10 mM HEPES-NaOH, 1 mM EGTA, pH 7.4; (bath) 145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, 10 μM (±)-PZQ, pH 7.4. The calculated P_Ca:P_Na was 0.75. Collectively, these results demonstrate that Sm.TRPM_PZQ acts as a non-selective cation channel, permeable to both monovalent and divalent cations.

Table 1.

Reversal voltages (V_rev, mean ± S.D.) of the current-voltage plots displayed in Fig. 6. Corresponding concentrations of the major electrolytes (in mM) in the recording pipette and bath solutions are shown.

Pipette / bath solution (mM)	Vrev (mV)
145 KCl / 14.5 KCl	−74.02 ±1.03
145 NaCl / 145 NaCl	0.74 ± 0.30
145 NaCl / 145 CsCl	−0.41 ± 5.2
145 NaCl / 145 KCl	2.15 ± 5.26
145 NaCl / 145 RbCl	3.38 ± 2.93
100 CaCl2 / 145 NaCl	0.49 ± 4 08

Next, to estimate the size of the Sm.TRPM_PZQ pore, Sm.TRPM_PZQ activity was recorded in the presence of a range of polyethylene glycol (PEG) polymers (Krasilnikov, 2002; Sabirov and Okada, 2004). In the absence of PEG, addition of (±)-PZQ (10 μM) to ‘inside-out’ patches resulted in clearly resolvable single channel responses (Fig. 6A). Perfusion of different molecular weight PEGs (10% w/v, symmetrical recording conditions at −60 mV) resulted in changes in channel opening behavior (Fig. 6A). The largest PEG species (molecular weight ≥3,000 Da) did not change the dwell time, with responses identical to PEG-free solution. However, beginning with PEG1000, channel activity increasingly displayed ‘flickering’ behavior and a significant decrease in the dwell time was observed. This effect was most pronounced with PEG400, although the single channel current amplitude remained unaffected. In the presence of PEG200, a reduction in the single channel unitary current was observed. The effect of the smaller PEG species on channel opening was dependent on the polarity of the recording. For example, with the bath solution supplemented with PEG400 (10% w/v), well resolved single channel activity was seen at +60 mV (where cations flow from the pipette to the grounded bath solution) matching the profile of single channel activity in the absence of PEG. However, when the polarity was reversed (−60 mV), channel open state fluctuations were considerably greater (Fig. 6B). A summary of these assays, showing changes in mean open time and single channel current in the presence of the different sized PEGs, are shown in Fig. 6C and D. Overall, relating these observations to the hydrodynamic radius (R_H, in nm) of the different PEG compounds, we conclude that PEG ≤1000 DA interfered with ion flow through Sm.TRPM_PZQ, and PEG200 (R_H=~0.45 nm) cause channel blockade.

Fig. 6.

Effect of differentially sized polyethylene glycol (PEG) species on the activity of the Schistosoma mansoni ion channel, TRPM_PZQ.. (A) Activity of Sm.TRPM_PZQ expressed in HEK293 cells, recorded in an ‘inside-out’ configuration at −60mV under symmetrical conditions (145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, pH 7.4). Similar recordings were then made in the presence of 10% w/v PEG of differing molecular weights (200–10,000 Da) exposed to the inner side of the channel. (B) Representative traces of Sm.TRPM_PZQ single channel activity clamped at either positive (black, +60 mV) or negative voltage (blue, −60 mV) recorded in an ‘inside-out’ configuration under symmetrical conditions (145 mM NaCl, 10 mM HEPES-NaOH, 1 mM EGTA, pH 7.4). The bath solution was supplemented with 10% w/v PEG400 and (±)-PZQ (10 μM). (C and D) Mean open time (C) and single channel current (D) of Sm.TRPM_PZQ measured (mean ± S.E.) in the presence of different molecular weight PEG compounds.

4. Discussion

Here, we provide the first characterization of the electrophysiological properties of Sm.TRPM_PZQ, revealing several features of this channel that hold relevance for understanding the action of PZQ against schistosomes. We show that, when activated by PZQ, Sm.TRPM_PZQ displays distinct channel properties from many of the human TRPM channels, and these properties are conserved with the other schistosome TRPM_PZQ orthologs (Sh.TRPM_PZQ and Sj.TRPM_PZQ). Importantly, PZQ-activated TRPM_PZQ acts as a high conductance, non-selective cation channel, competent to trigger cellular Ca²⁺ signals and exert a profound, long-lasting membrane depolarization in cells where it is endogenously expressed. Each of these conclusions are discussed further in the following paragraphs.

First, Sm.TRPM_PZQ displays properties unique from the human TRPM paralogs, including Hs.TRPM2 and Hs.TRPM8, which display some domain homology with Sm.TRPM_PZQ (Park and Marchant, 2020). Unlike Hs.TRPM2, Ca²⁺ does not function as a co-agonist at Sm.TRPM_PZQ and unlike Hs.TRPM8, Sm.TRPM_PZQ displays a linear I-V relationship (Fernandez et al., 2011; Himmel et al., 2020; Huang et al., 2020). This distinctiveness underscores the specialization of flatworm TRP channels along distinct evolutionary trajectories – almost mosaic-like in terms of combining properties resolved in vertebrate TRPs – and highlights the danger of extrapolating characteristics from their vertebrate ‘counterparts’. While the properties of schistosome TRP channels are not currently well defined, experimental analyses to date have revealed distinctive properties and pharmacology (Bais et al., 2018). This is because TRPM diversification occurred independently between these evolutionary lineages, with the TRPM1–8 expansion being specific to vertebrates (Himmel et al., 2020). Parasitic flatworm TRP channels must therefore be characterized individually to resolve their properties and understand how they are customized to meet the specific sensory signaling requirements of the parasitic lifecycle.

Second, the similar functional properties of the schistosome TRPM_PZQ orthologs are not unexpected, as all schistosome species are sensitive to PZQ (Andrews et al., 1983; Chai, 2013), and recent work has demonstrated that the PZQ binding site of TRPM_PZQ is completely conserved between Schistosoma spp. TRPM_PZQ (Rohr et al., 2022). Likely these biophysical properties are shared with TRPM_PZQ orthologs in other parasites, although detailed characterization of TRPM_PZQ in other parasitic flatworms is currently lacking. We note the activity of schistosome TRPM_PZQ was poorly resolved at negative voltages in a ‘cell-attached’ configuration (Fig. 2). In this recording configuration, cytoplasmic constituents are potentially dragged into the pore interfering with the single channel signal. Whether TRPM_PZQ has any such endogenous inhibitors will require additional investigation. This effect was reproduced in the PEG polymer assays, where PEG species added to the bath side of ‘inside-out’ patches caused poorly resolved currents at negative potentials (Fig. 6). The blockade at negative voltages underscores the cation selectivity of the channel, as if TRPM_PZQ were anion permeable, the decrease in signal resolution would also have been observed at positive voltages. Estimated pore sizes calculated from PEG assays (Fig. 6) are consistent with the pore size of a TRPM2 channel resolved in the open state (0.26 nm, (Huang et al., 2018)). In addition, all of the schistosome TRPM_PZQ channels studied here displayed voltage independence, behaving as ‘ohmic’ channels where channel activity is a linear function of transmembrane voltage (Fig. 3B). This property likely relates to the highly conserved sequence WNxxD found in the S3 transmembrane domain of TRPM channels (Winking et al., 2012). For Hs.TRPM2 (which is voltage-independent) this sequence is WNKLD, matching an identical S3 sequence that is found in Sm.TRPM_PZQ.

Lastly, and most importantly, these experiments show that Sm.TRPM_PZQ behaves as a large conductance (Fig. 3, ~130 pS in 145 mM NaCl), non-selective cation channel (Fig. 5) when activated by PZQ that shows little desensitization in either single channel recordings (Fig. 1) or intact cell functional assays (Fig. 4; Park et al., 2019, 2021; Chulkov et al., 2021; Rohr et al., 2022). Relative to the behavior of voltage-operated Ca²⁺ channels (Ca_v) measured in vertebrate systems (unitary currents of 0.18 pA at −40 mV under physiological conditions, (Rubart et al., 1996)), a single opening of Sm.TRPM_PZQ in response to PZQ will likely effect membrane depolarization and Ca²⁺ influx at a considerably higher rate than activation of multiple Ca_vs. Further, TRPM_PZQ activity will not only trigger cellular Ca²⁺ signals but also potentiate Ca²⁺ influx via downstream engagement of Ca_v channels that are activated by the subsequent membrane depolarization. The strong depolarizing stimulus that engages downstream Ca_v channels provides an explanation for data that has previously shown the involvement of neuronal Ca_v channels as regulators of PZQ efficacy in vivo (Nogi et al., 2009; Zhang et al., 2011; Chan et al., 2017). Knockdown of Ca_v channels will attenuate the in vivo action of PZQ by attenuating this subsequent membrane depolarization. These properties of TRPM_PZQ underpin this target’s vulnerability to chemotherapeutic attack – as the channel behaves as an ‘excitotoxic’ stimulus when engaged by PZQ. Activation of only a few channels would effect a considerable depolarizing stimulus. Such characteristics should also tweak our curiosity as to how TRPM_PZQ activity is endogenously dampened to prevent damage to during normal, physiological activity in schistosome nerves (and possibly muscles) where TRPM_PZQ is endogenously expressed. Low expression levels of TRPM_PZQ (Wendt et al., 2020), expression changes throughout the parasite lifecycle, minimal spontaneous openings of the channel (Fig. 1), the presence of endogenous cytoplasmic inhibitors (Fig. 2) as well as other aspects of channel regulation may all act to attenuate ion flux at physiological membrane potentials. Additional insight will derive from unmasking the endogenous route(s) to TRPM_PZQ activation in parasitic flatworms. One such regulator is addressed in the companion paper (Chulkov et al., 2022).

Finally, we note that the large amplitude of PZQ-evoked signals mediated by TRPM_PZQ is of great benefit for high throughput screening approaches, which require a good signal to noise for assay miniaturization (Chulkov et al., 2021). TRPM_PZQ activity can be resolved in intact cells by fluorescent reporters that sense changes in cytoplasmic Ca²⁺ (Park et al., 2019; Chulkov et al., 2021; Park et al., 2021) or membrane depolarization (Fig. 4). Optimization of such voltage-sensing reporters will assist the functional profiling of other parasitic flatworm ion channels that are impermeable to Ca²⁺. The large size of TRPM_PZQ signals also supports electrophysiological analyses at single channel resolution as an orthogonal approach for validating the action, and mechanism of channel regulation, of novel TRPM_PZQ ligands (Chulkov et al., 2021).

Highlights.

• TRPM_PZQ is characterized in Schistosoma mansoni using single channel electrophysiology

• TRPM_PZQ is a voltage-independent, large conductance, non-selective cation channel

• TRPM_PZQ activity can also be resolved using a membrane potential reporter dye