Disease-associated mutations in TRPM3 render the channel overactive via two distinct mechanisms

Transient Receptor Potential Melastatin 3 (TRPM3) is a Ca2+ permeable non-selective cation channel activated by heat and chemical agonists such as pregnenolone sulfate and CIM0216. TRPM3 mutations in humans were recently reported to be associated with intellectual disability and epilepsy; the functional effects of those mutations however were not reported. Here we show that both disease-associated mutations of TRPM3 render the channel overactive, but likely via different mechanisms. The Val to Met substitution in the S4-S5 loop induced a larger increase in basal activity and agonist sensitivity at room temperature than the Pro to Gln substitution in the extracellular segment of S6. In contrast, heat activation was increased more by the S6 mutant than by the S4-S5 segment mutant. Both mutants were inhibited by the TRPM3 antagonist primidone, suggesting a potential therapeutic intervention to treat this disease.


INTRODUCTION
Transient Receptor Potential Melastatin 3 (TRPM3) is a Ca 2+ permeable, non-selective cation channel activated by heat (Vriens et al., 2011) and chemical activators such as the neurosteroid pregnenolone sulfate (PregS) (Wagner et al., 2008) and the synthetic compound CIM0216 (Held et al., 2015). TRPM3 is a well-established temperature sensor in peripheral sensory neurons of the dorsal root ganglia (DRG); its genetic deletion in mice leads to defects in noxious heat sensation as well as reduced inflammatory heat hyperalgesia (Vriens et al., 2011;Vandewauw et al., 2018). Inhibitors of TRPM3 also reduced both acute heat sensation and inflammatory heat hyperalgesia (Straub et al., 2013;Krugel et al., 2017). While TRPM3 -/mice show defects in noxious heat sensation, the channel shows increased activity well below the noxious range, when temperature is increased from 15 o C to 26 o C, with further increases in activity at 37 o C (Vriens et al., 2011). TRPM3 is also inhibited by activation of Gi-coupled receptors such as μ-opioid receptors and GABAB receptors in DRG neurons, and agonists of those receptors reduced nocifensive reactions to local injection of TRPM3 agonists (Badheka et al., 2017;Dembla et al., 2017;Quallo et al., 2017).
TRPM3 is also expressed in tissues other than peripheral sensory neurons, where its functional roles are not well understood. In pancreatic β-cells it was shown that application of the TRPM3 agonist PregS induced an increase in insulin secretion (Wagner et al., 2008), but TRPM3 -/mice showed no impairment in glucose homeostasis (Vriens et al., 2011). TRPM3 is also expressed in vascular smooth muscle cells, and PregS was reported to induce contractile responses in freshly isolated aorta, but the concentration required for this response was higher than the plasma levels of this compound (Naylor et al., 2010). TRPM3 is also expressed in various regions of the brain, where its functional role is essentially unexplored .
Very little is known about the physiological and pathophysiological roles of TRPM3 in humans. A recent paper showed that two missense mutations in TRPM3 are associated with a neurodevelopmental disorder with intellectual disability, hypotonia and epilepsy, pointing to important roles of this channel in the human brain (Dyment et al., 2019). Seven of the eight patients had a de novo Val to Met substitution in the S4-S5 loop, while one patient had a Pro to Gln substitution in the extracellular segment of S6. The effects of these mutations on channel function however were not reported.
Here we tested the functional effects of the two disease-associated mutations using electrophysiology and intracellular Ca 2+ measurements. We find that both diseaseassociated mutations render the channel overactive. Both mutants showed constitutive activity that was inhibited by the TRPM3 antagonist primidone. As primidone is a clinically used medication (Krugel et al., 2017), our findings offer potential clinical intervention to treat this channelopathy. We also find that the Val to Met substitution in the S4-S5 loop induced a larger left shift in the concentration response relationship to PregS and CIM0216 than the Pro to Gln substitution close to the pore-loop in the extracellular segment of S6. The increase in heat activation on the other hand was more pronounced in the S6 mutant. We conclude that both reported mutants of TRPM3 are gain of function, but the mechanism of increased channel activity is different for the two mutants.

RESULTS
Here we used, patch clamp electrophysiology and intracellular Ca 2+ measurements in HEK293 cells and two-electrode voltage clamp electrophysiology in Xenopus oocytes to study the effects of disease-associated mutations on TRPM3 function. TRPM3 has a large number of splice variants , but there is no information available which splice variants are expressed in the human brain, and relatively little is known about the functional differences between splice variants. To ensure that our results do not only apply to one variant, we used two commonly used and well-characterized splice variants of TRPM3 in our experiments. In HEK293 cells we expressed the human orthologue of the most studied mouse variant TRPM3α2 that was originally cloned from mouse brain (Oberwinkler et al., 2005;Vriens et al., 2011). In Xenopus oocytes we expressed the human TRPM3 splice variant we used in several previous studies (Badheka, 2015;Badheka et al., 2017), originally described in (Grimm et al., 2003) also called the TRPM31325 variant (Oberwinkler et al., 2005).
Most alternatively spliced exons are in the cytoplasmic N-terminus; thus, the numbering of mutated residues varies between splice variants. The more common S4-S5 segment mutant described as V837M (Dyment et al., 2019) corresponds to V990M in the hTRPM3 splice variant we expressed in Xenopus oocytes, and to V992M in the hTRPM3α2 variant we expressed in HEK293 cells. The S6 mutant P937Q in Dyment et al corresponds to P1090Q in the hTRPM3 variant we used in oocytes, and it is P1092Q in the hTRPM3α2 we used in HEK293 cells. Figure 1 Figure Supplement 1 shows the location of these residues in the TRPM3 sequence and putative structure.
First, we co-expressed the Ca 2+ indicator GCaMP6f (Chen et al., 2013) and the hTRPM3α2 isoform and its mutants in HEK293 cells and performed intracellular Ca 2+ measurements in a 96-well plate reader. We found that the concentration-response relationship to PregS (Fig. 1A-D) and CIM0216 (Fig. 1E-H) were left shifted in the mutant channels; V992M showed a much larger shift than P1092Q for both agonists. The V992M mutant also showed a larger increase in basal Ca 2+ levels than P1092Q at room temperature (21 o C), and the application of the TRPM3 antagonist primidone decreased basal Ca 2+ levels for both mutants in a concentration dependent manner . Primidone had no effect on basal cytoplasmic Ca 2+ levels in cells transfected with the wild type TRPM3 indicating negligible basal activity of the wild type channel at room temperature (Fig. 1I). Primidone robustly inhibited Ca 2+ signals evoked by EC50 concentrations of PregS both for wild type and mutant channels ( Figure 1  Next, we transfected HEK293 cells with the mutant and wild type hTRPM3α2 and used fura-2 Ca 2+ imaging to study the effects of acutely increased temperatures (Fig. 2). We first increased the temperature to 37 o C, followed by 10 μM primidone at room temperature to facilitate return of Ca 2+ to baseline. Then we applied 25 μM PregS (in the absence of primidone), and compared the Ca 2+ responses induced by 37 o C to that induced by PregS. In cells transfected with wild type TRPM3, the temperature-induced Ca 2+ response was, on average, 29.4% of that induced by PregS, for the V992M mutant it was 76.5%, whereas for the P1092Q mutant it was ~122.5% (Fig. 2D,E). In cells not expressing TRPM3 increasing temperature to 37 o C induced only negligible Ca 2+ signals (not shown).
Cytoplasmic Ca 2+ is an indirect measure of TRPM3 activity, thus next, we performed whole cell patch clamp experiments to compare currents induced by increased temperatures and by PregS. These measurements were performed in the absence of extracellular Ca 2+ to avoid indirect effects of increased cytoplasmic Ca 2+ such as Ca 2+ induced desensitization. We stimulated each cell with a temperature ramp from 23 o C to 36 o C followed by a saturating concentration of PregS (100 μM) at room temperature ( Fig. 3A-C). Since the current amplitudes induced by PregS were highly variable, presumably due to different expression levels of the channel (Fig. 3H), we normalized the currents induced by increased temperatures to those evoked by PregS and plotted these relative currents as a function of temperature ( Figure 3D-F). Figure G shows that the slope of the increase of these currents as a function of temperature were significantly steeper for the P1092Q mutant than for V992M, and both mutants were significantly steeper than the wild type.
Next, we expressed the wild type hTRPM31325 and its V990M and P1090Q mutants in Xenopus oocytes and performed full concentration response measurements with the TRPM3 agonist PregS. Consistent with our Ca 2+ measurements, the concentration response relationships for PregS were left-shifted for both mutants compared to wildtype, but the effect of the V990M mutation was much more pronounced than that of the P1090Q (Fig. 4A-D).
Both disease-associated mutations are de novo, and all known patients are heterozygous. To mimic heterozygous conditions, we co-injected oocytes with wild type cRNA and either mutant in a 1:1 ratio. The PregS dose response was still markedly left shifted for the V990M:TRPM3 combination, but it was only marginally shifted in the P1090Q:TRPM3 combination compared to WT TRPM3 ( Fig. 4E-G).
To assess basal current levels, we applied 50 μM primidone at room temperature (20-22 o C). Consistent with our Ca 2+ measurements, primidone evoked a significantly larger inhibition of basal activity in the V990M than in the P1090Q mutant both at 100 and -100 mV ( PregS were not different between the two mutants and wild type, at +100 mV, but the V990M mutants showed somewhat larger amplitudes at -100 mV than the wild type or the P1090Q mutant (Figure 4 Figure Supplement 1E). We also compared the basal current amplitudes before applying any stimuli, and found that oocytes expressing the V990M mutant showed higher currents than those expressing P1090Q, and both mutants had higher basal currents than wild type TRPM3 ( Figure 4   We also compared the currents induced by increased temperatures to those evoked by PregS in channels expressed in Xenopus oocytes ( Figure 4 Figure Supplement 2). We found that the ratio of currents induced by 30 o C over those induced by 50 μM PregS were significantly larger in the P1090Q mutant compared to V990M, and both mutants showed significantly larger current ratios than wild type TRPM3. Increasing temperature to 30 o C induced negligible currents in non-injected oocytes, but those currents became larger at higher temperatures (not shown), which prevented us from testing higher temperatures in this experimental setting.
TRPM3 has been shown to be inhibited by activation of Gi-coupled receptors via direct binding of Gβγ (Badheka et al., 2017;Dembla et al., 2017;Quallo et al., 2017). To test if the mutations alter receptor-induced inhibition, we co-expressed TRPM3 or its mutants with Gi-coupled muscarinic M2 receptors in Xenopus oocytes. Figure 4 Figure Supplement 3A-F show that when we applied acetylcholine (ACh) to stimulate M2 receptors, it evoked a ~50% inhibition of TRPM3 currents induced by 50 μM PregS. The inhibition of the P1090Q mutant was similar to wild type, but the V990M mutant was essentially not inhibited. To test whether the lack of inhibition was due to allosteric effects of the increased sensitivity to PregS, we stimulated the oocytes expressing the V990M mutant with 5 μM PregS where ACh induced a ~35% inhibition (Figure 4 Figure  Supplement 3C,E). We also tested inhibition at the EC50 of PregS for wild type (17 μM) and mutant channels, 0.6 μM for V990M and 7 μM for P1090Q. Figure 4 Figure Supplement 3G-K shows that wild type and V990M mutants were inhibited to a similar extent, the P1090Q mutant was inhibited somewhat more than the wild type channel. These data show that while the V990M mutation affects receptor-induced inhibition at high PregS concentrations, it is not likely to be the primary mechanism of its gain of function phenotype.

DISCUSSION:
Overall, our data show that disease-associated mutations in TRPM3 render the channel overactive. Both mutants showed basal activity even at room temperature, which was reduced by the TRPM3 antagonist primidone. Basal activity of the V992M mutant at room temperature was higher than that of the P1092Q mutant, but at 37 o C, the difference in basal activity between the two mutants became negligible. Given the increased constitutive activity of the mutants at body temperature, increased neuronal excitability and/or Ca 2+ -induced neuronal damage is a possible disease-causing mechanism. Primidone is a clinically approved antiepileptic drug; it is thought to exert its effects by being converted to barbiturate by the liver, but it crosses the blood brain barrier (Nagaki et al., 1999), and directly inhibits TRPM3 activity even below its therapeutic concentration (Krugel et al., 2017). Our data showing that primidone inhibited the basal activity of the mutant channels, suggests a potential therapy for this newly described channelopathy.
The V990/V992M mutation showed a larger increase in basal activity at room temperature, and also induced a larger left shift in the concentration response curves of the agonists PregS and CIM0216 than the P1090/1092Q mutation. This is expected if both mutations increase the stability of the open state of the channel, with the V990/992M mutation having a larger effect. If increased open state stability is the only explanation for the over-activity of the mutants, we would expect V990/992M to be also more sensitive to activation by increased temperatures than P1090/1092Q. This is however not what we observed. We consistently find that the P1090/1092Q mutant showed more pronounced activation by increased temperatures than the V990/992M mutant. This indicates that the mechanism of over-activity is different for the two mutants.
These two residues are in different locations; V990/992 is in the S4-S5 linker whereas P1090/1092 is in the outer portion of S6 ( Figure 1 Figure Supplement 1). The S4-S5 linker plays essential roles in channel gating, and has been shown to be a hotspot of disease-associated gain-of-function mutations in a number of TRP channels, including TRPV3, TRPV4, TRPM4 and TRPA1 (Hofmann et al., 2017). This channel segment also plays a role in binding of hydrophobic, or amphipathic ligands in TRPM channels (Huang et al., 2020). For example, in TRPM8 channels the S4-S5 linker is in direct contact with both the cooling agent icilin, and the menthol analog WS12 (Yin et al., 2019). While PregS is thought to activate TRPM3 by directly binding to the channel (Drews et al., 2014), its binding site in the channel is not known, and currently there is no structural information available for TRPM3. Because of the clear increase in basal activity of the V990/992M mutant, it is quite likely that the mutation primarily increased the stability of the open state and the decrease in EC50 of agonist activation is a consequence of the change in activation equilibrium constant (Colquhoun, 1998). It cannot be excluded however that the mutation also affected PregS binding concurrently, given the general role of this segment in ligand binding in TRP channels.
The membrane phospholipid phosphatidylinositol 4,5-bisphosphate (PIP2) is also required for PregS-induced TRPM3 activity (Badheka, 2015;Toth et al., 2015;Uchida et al., 2016). In TRPM8, the equivalent of V990 is two positions upstream from a residue that is in close contact with PIP2, and it is 4 and 7 positions downstream from two residues in close contact with icilin and the menthol analog WS12 (Yin et al., 2019) (Figure 1 Figure Supplement 1A). PIP2 is located adjacent to the menthol analogue WS12 and the cooling agent icilin in the TRPM8 structures, and menthol has been shown to allosterically affect PIP2 activation (Rohacs et al., 2005).Therefore, it is also possible that the V990/992M mutation affects PregS activation indirectly via PIP2. Increased Ca 2+ levels in cells expressing mutated channels may also modify cellular PI(4,5)P2 levels, which may alter channel activity. Differentiating between these possibilities will require future studies.
While the mechanism of temperature activation of TRP channels is not fully understood (Clapham and Miller, 2011;Islas, 2017;Arrigoni and Minor, 2018;Castillo et al., 2018), large scale unbiased mutagenesis studies on TRPV1 (Grandl et al., 2010) and TRPV3 (Grandl et al., 2008) show that mutations in the pore region and the outer portion of S6 in these channels selectively abolished heat-, but not agonist-induced channel activation. The P1090/1092Q mutation in TRPM3 is located in the outer portion of S6, and it had a stronger effect on heat activation than the V990/9992M mutation, therefore it is possible that the primary effect of the P1090/1092Q mutation is increasing heat sensitivity. Increased temperatures synergize with PregS in activating wild-type TRPM3 (Vriens et al., 2011), see also Figure 1 Figure Supplement 3B, therefore it is possible that the increased PregS sensitivity of the P1090/1090Q mutant is secondary to its increased heat activation.
Overall our data show that both disease associated mutations render TRPM3 overactive, but likely with different mechanisms. Table   Reagent , 2017). HEK293 cells were transfected with hTRPM3a2-GFP or its mutants using the Effectene reagent (Qiagen). Cells were loaded with 1 μM fura-2 AM (Invitrogen) for 40 min before the measurements at 37°C, and dual-excitation images at 340 and 380 nm excitation wavelengths were detected at 510 nm with a Roper Cool-Snap digital CCD camera. Measurements were conducted at room temperature in extracellular solution containing 137 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 10 mM HEPES and 10 mM glucose, pH 7.4. PregS, and primidone were applied with a gravity driven whole chamber perfusion system. Temperature stimulation was performed using a custom-built system as described earlier (Badheka et al., 2017) by pushing bath solution through a spiral tubing immersed in hot water using a 60 ml syringe while monitoring the temperature of the perfusion chamber using a CL-100 Warner Instruments temperature controller. The analogue signal from the CL-100 unit was fed into the Digidata digitizer and the temperature curve was collected in Clampex. Data analysis was performed using the Image Master 5 software (PTI).

Whole cell patch clamp experiments
HEK293 cells were transiently transfected with cDNA encoding the hTRPM3a2-GFP, or its mutants with 0.2 µg of constructs using the Effectene reagent (Qiagen) according manufacturer's protocol and were used in experiments 48-72 hours later. Measurements were carried out on GFP positive cells, in an extracellular solution containing (in mM) 137 NaCl, 5 KCl, 1 MgCl2, 10 HEPES and 10 glucose, pH 7.4. The intracellular solution contained (in mM) 140 potassium gluconate, 5 EGTA, 1 MgCl2, 10 HEPES, and 2 NaATP, pH 7.3. Patch clamp pipettes were prepared from borosilicate glass capillaries (Sutter Instruments) using a P-97 pipette puller (Sutter Instrument) and had a resistance of 2-4 MΩ. In all experiments after formation of GW-resistance seals, the whole-cell configuration was established and currents were recorded using a ramp protocol from -100 mV to +100 mV over 500 ms preceded by a -100 mV step for 100 ms; the holding potential was 0 mV, and this protocol was applied once every 1 second. The currents were measured with an Axopatch 200B amplifier, filtered at 5 kHz, and digitized through Digidata 1440A interface. In all experiments, cells that had a passive leak current over 100 pA were discarded. Data were collected and analyzed with the PClamp10.6 (Clampex) acquisition software (Molecular Devices, Sunnyvale, CA), and further analyzed and plotted with Origin 2019b (OrigiLab, Northampton, MA). Heat stimulation was performed as described for the Ca 2+ imaging experiments.

Two electrode voltage clamp experiments:
Xenopus laevis oocytes were prepared as described earlier (Badheka, 2015). All animal procedures were approved by the Institutional Animal Care and Use Committee at Rutgers New Jersey Medical School. In brief, frogs were anesthetized in 0.25% ethyl 3aminobenzoate methanesulfonate solution (MS222; Sigma-Aldrich); bags of ovaries were removed surgically from the anesthetized frogs. Individual oocytes were obtained by overnight digestion at 16°C in 0.2-0.3 mg/ml type 1A collagenase (Sigma-Aldrich), dissolved in a solution containing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2, and 5 μM HEPES, pH 7.4 (OR2 solution). The next day the collagenase containing solution was discarded and the oocytes were washed multiple times with OR2 solution. The oocytes were maintained in OR2 solution supplemented with 1.8 mM CaCl2, 100 IU/ml penicillin, and 100 μg/ml streptomycin at 16°C. cRNA was transcribed from the linearized human TRPM3 (hTRPM3) cDNA clone (Grimm et al., 2003), or its mutants in the pGEMSH vector using the mMessage mMachine kit (Thermo Fisher Scientific). cRNA (40 ng) was microinjected into individual oocytes, using a nanoliter-injector system (Warner Instruments). For combined injection of wild-type and mutant TRPM3 for Figure 4E-G, 40 μg total cRNA was injected in a 1:1 ratio. The V990M and P1090Q mutants were generated using the QuikChange II XL Site-Directed Mutagenesis Kit (Agilent Technologies). For the GPCR regulation of these mutants, we injected cRNA of human M2 muscarinic receptors together with TRPM3 or mutants at 1:1 ratio. Oocytes were used for electrophysiological measurements 48-72 h after microinjection. The hTRPM31325 clone in a mammalian expression vector was provided by C. Harteneck (Eberhard Karls University Tübingen, Tübingen, Germany), and it was subcloned into the pGEMSH oocyte vector using standard molecular biology techniques.
Two electrode voltage clamp experiments were performed as described (Badheka, 2015). In brief, oocytes were placed in extracellular solution (97 mM NaCl, 2 mM KCl, 1 mM MgCl2, and 5 μM HEPES, pH 7.4), and currents were recorded with thin-wall inner filament-containing glass pipettes (World Precision Instruments) filled with 3 M KCl in 1% agarose. Currents were measured with a ramp protocol from -100 to 100 mV once every 0.5 seconds with a GeneClamp 500B amplifier and analyzed with the pClamp 9.0 software (Molecular Devices). PregS, ACh and primidone were applied with a gravity driven whole chamber perfusion system. Temperature stimulation was performed the same way as for the whole cell patch clamp and Ca 2+ imaging experiments in HEK293 cells.

Statistics:
Statistical analysis was performed with Origin 2019b and Prism6. Data are plotted as mean ± SEM and scatter plots. No statistical method was used to predetermine sample sizes, but our sample sizes are similar to those generally employed by the field. Experiments were performed in random order. Data were analyzed with t-test, or oneway analysis of variance with Bonferroni's post hoc test, p values are reported in the figures.  Top: Sequence alignment of the outer part of S6 in TRPM channels, the P1090/1092 residue in TRPM3 is highlighted blue. Bottom: Sequence alignment of the S4-S5 loop of TRPM channels, the V990/992 residue in TRPM3 is labeled red, yellow highlights indicate residues in contact with the agonists icilin and/or WS12 in the TRPM8 structures (6nr3, 6nr4, 6nr2), the residue highlighted green is in contact with PI(4,5)P2 in TRPM8. B: the location of residues equivalent to V990/992 and P1090/1092 on the structure of TRPM4 (6bcj); one subunit is hidden for visibility. The V990/992 residue corresponds to V837 in (Dyment et al., 2019) while the P1090/1092 corresponds to P937.

Figure 1 -Figure Supplement 2: Primidone inhibits Ca 2+ responses induced by EC50 concentrations of PregS.
HEK293 cells were transfected with the Ca 2+ indicator GCaMP6f and the hTRPM3α2 or its mutants, and fluorescence was measured in a 96-well plate reader (Flexstation-3) at room temperature (~21 o C) as described in the method section. A-C: Cells were stimulated by PregS at its EC50 concentration for each construct, as indicated by the first arrow, then various concentrations of primidone were applied (second arrow). D: Hill1 fits of the concentration dependence of primidone; symbols show mean ± SEM from 5 or 6 wells, from two independent transfections. The IC50 values were 2.37 ± 0.39 μM wild type channels, 8.62 ± 1.42 μM for V992M and 3.52 ± 1.13 μM for P1092Q    The EC50 values were 17.1 ± 0.9 μM for wild type channels, 0.63 ± 0.05 μM for V990M and 6.94 ± 0.38 μM for P1090Q. E-F: Representative traces for measurements in oocytes injected with cRNA for wild type and mutant channels (1:1 ratio). G: Hill fits of the concentration dependence of the effects of PregS at 100 mV for wild type and mutant channel combinations; symbols represent individual oocytes from two different preparations. The EC50 values were 17.8 ± 1.04 μM for wild type channels, 3.83 ± 0.27 μM for V990M + WT and 12.2 ± 1.1 μM for P1090Q + WT.  Summary of the temperature responses expressed as a fraction of the currents induced by 50 μM PregS in the same measurements. F: Current amplitudes induced by increased temperatures and by PregS; symbols represent responses in individual oocytes from two different oocyte preparations; lines connect measurements in the same oocytes, bars are mean ± SEM. The current amplitudes induced by 50 μM PregS were, on average, smaller in both mutants than in wild type in these experiments, most likely because oocytes expressing these mutants showed substantially higher rate of death than those injected with wild type TRPM3, therefore the oocytes we used for electrophysiology were selected for lower expression. To be able to compare temperature-induced currents, we normalized current amplitudes to those induced by 50 μM PregS in panel E. Castillo, K., Diaz-Franulic, I., Canan, J., Gonzalez-Nilo, F., and Latorre, R. (2018). Thermally activated TRP channels: molecular sensors for temperature detection. Phys Biol 15, 021001.