Learn more: PMC Disclaimer | PMC Copyright Notice
Associated Data
Abstract
Cannabinoids are a major class of compounds produced by the plant Cannabis sativa. Previous work has demonstrated that the main cannabinoids cannabidiol (CBD) and tetrahydrocannabinol (THC) can have some beneficial effects on pain, inflammation, epilepsy, and chemotherapy-induced nausea and vomiting. While CBD and THC represent the two major plant cannabinoids, some hemp varieties with enzymatic deficiencies produce mainly cannabigerolic acid (CBGA). We recently reported that CBGA has a potent inhibitory effect on both Store-Operated Calcium Entry (SOCE) via inhibition of Calcium Release-Activated Calcium (CRAC) channels as well as currents carried by the channel-kinase TRPM7. Importantly, CBGA prevented kidney damage and suppressed mRNA expression of inflammatory cytokines through inhibition of these mechanisms in an acute nephropathic mouse model. In the present study, we investigate the most common major and minor cannabinoids to determine their potential efficacy on TRPM7 channel function. We find that approximately half of the tested cannabinoids suppress TRPM7 currents to some degree, with CBGA having the strongest inhibitory effect on TRPM7. We determined that the CBGA-mediated inhibition of TRPM7 requires a functional kinase domain, is sensitized by both intracellular Mg⋅ATP and free Mg2+ and reduced by increases in intracellular Ca2+. Finally, we demonstrate that CBGA inhibits native TRPM7 channels in a B lymphocyte cell line. In conclusion, we demonstrate that CBGA is the most potent cannabinoid in suppressing TRPM7 activity and possesses therapeutic potential for diseases in which TRPM7 is known to play an important role such as cancer, stroke, and kidney disease.
Introduction
Cannabis sativa has a rich history as a plant-based medicine, used to treat conditions like chronic pain, nausea, anxiety, insomnia, anorexia, and muscle spasms.1,2 It is also used in the treatment of chronic disorders such as autoimmune and neurodegenerative diseases,3 cancer,4,5 multiple sclerosis,6–8 and epilepsy.9–11 This broad therapeutic efficacy underscores the importance of elucidating the underlying mechanisms behind the diverse benefits of cannabis. Compositionally, the cannabis plant is dominated by a large class of natural products known as phytocannabinoids, which possess anti-inflammatory and analgesic properties.12,13 Like the plant itself, the phytocannabinoids act through a multitude of mechanisms. These include G-protein-coupled receptors such as cannabinoid receptors type 1 (CB1) and type 2 (CB2) or CB1/CB2-independent pathways.14 In addition, cannabinoids have been found to affect other biological targets, including members of the transient receptor potential (TRP) family of ion channels.15,16
Cannabigerolic acid (CBGA) is the acidic form cannabigerol (CBG) and represents the biosynthetic precursor to a vast number of cannabinoids, including tetrahydrocannabinol (THC) and cannabidiol (CBD). Like CBD, cannabigerolic acid (CBGA) has been found to have anticonvulsant effects in mouse models of epilepsy.17 CBGA has been reported to be a dual PPARα/γ agonist with the potential to treat metabolic disorders such as diabetes and dyslipidemia.18 Preclinical findings show that CBG, the decarboxylated version of CBGA, reduces intraocular pressure, possesses antioxidant, anti-inflammatory, and antitumoral activities, and has antianxiety, neuroprotective, dermatological, and appetite-stimulating effects.19 Both CBGA and CBG have neuroprotective properties in Parkinson’s disease model20,21; however, the mechanisms of action and pharmacological effects remain to be elucidated in more detail.
TRPM7 is a member of the transient receptor potential melastatin (TRPM) family of ion channels and is composed of a 6 transmembrane spanning ion channel domain and a functional serine/threonine kinase domain at the protein’s intracellular C-terminus.22 TRPM7 conducts a range of divalent cations such as Ca2+, Mg2+, Zn2+, Ni2+, Mn2+, and Co2+ into cells via the channel domain.23,24 The TRPM7 protein is expressed ubiquitously in mammals, where it is critical for regulating both organismal and cellular Mg2+ homeostasis. While TRPM7 knock-out is embryonic lethal,25 dysregulated TRPM7 activity represents an important factor in a variety of diseases, including neuronal death in ischemia,26 cardiac atrial fibrillation,27 and renal diseases.28–30 TRPM7 also plays a key role in several fibrotic disorders, including cardiac,31 hepatic,32 pulmonary,33 and renal fibrosis.34 TRPM7 also drives cell proliferation, cell growth, apoptosis, and differentiation, thereby contributing to hyperproliferative diseases such as cancer.22,35,36
TRPM7 is suppressed by physiological amounts of intracellular adenosine triphosphate (Mg⋅ATP) and Mg2+.37–39 Inhibition of TRPM7’s enzymatic activity by genetic modification of its endogenous kinase domain reduces the sensitivity of TRPM7 to its intracellular inhibitors Mg2+ and Mg·ATP.40 The binding of Mg⋅ATP to the kinase domain is required to phosphorylate other protein targets, whereas TRPM7 channel domain function is not required for enzymatic activity of the TRPM7 kinase function.
Previous work has demonstrated the ability of cannabinoids to modulate calcium signaling through various TRP channels15 and calcium release-activated calcium (CRAC) channels.41 We recently reported that CBGA was the most potent cannabinoid suppressing store-operated calcium entry (SOCE) and IL-2 production in T cells.41 Building upon this initial discovery, we showed that CBGA also decreased IL-2 production and ameliorated kidney inflammation in an acute nephropathic mouse model.28 Additionally, CBGA reduced kidney fibrosis in a progressive renal fibrotic model.28 Encouraged by this in vivo result, we explored the mechanism behind the renal-protective properties of CBGA. We showed a novel mechanism by which CBGA reduces kidney disease at least in part owing to its inhibitory action on TRPM7, indicating the potential therapeutic impact of CBGA on both acute and chronic kidney disease. In this communication we characterize the mechanistic interaction between TRPM7 and CBGA and report that the inhibitory action of CBGA on TRPM7 requires its kinase domain to remain functional and is modulated by both intracellular and Mg·ATP, free Mg2+, and free Ca2+.
Materials and Methods
Cell Culture
Tetracycline (Tet)-inducible HEK293-TREx cells, stably transfected with human TRPM7 (hTRPM7), were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco) containing 10% fetal bovine serum (FBS, Gibco), zeocin (0.4 mg/mL, Invitrogen), and blasticidin (5 µg/mL, Invitrogen). HEK293 cells stably transfected with the phosphotransferase activity-deficient point mutation in the hTRPM7 kinase domain, hTRPM7/K1648R and hTRPM7/G1799D, were cultured in the same medium. Tet-inducible human TRPM6 + human TRPM7 HEK293-TREx cells were maintained in DMEM medium containing 10% FBS, zeocin (0.4 mg/mL), blasticidin (5 µg/mL), and hygromycin (0.4 mg/mL, Chemi Cruz). Overexpression was induced by adding 1 µg/mL tetracycline (Gibco) to the culture medium. Patch-clamp experiments were performed 18-21 h post-tetracycline induction. HEK293-TREx cells were transiently transfected with green fluorescent protein (GFP)-tagged human TRPM6 plasmid using Lipofectamine 2000 (Invitrogen). Whole-cell patch-clamp experiments were carried out 26 h post-transfection. Expression of human TRPM6 constructs was identified by green fluorescence. U266 cells were cultured in RPMI-1640 medium (Gibco) containing 10% FBS. Cells were maintained at 37°C under 5% CO2 conditions.
Whole-cell Patch-Clamp Recording
Patch-clamp experiments were performed in the whole-cell configuration. Currents were elicited by a ramp protocol from −100 to +100 mV over 50 ms acquired at 0.5 kHz and a holding potential of 0 mV. Outward current amplitudes over the course of the experiment were extracted at +80 mV, inward currents were extracted at -80 mV, and plotted versus time. Data were normalized to cell size measured immediately after whole-cell break-in (pA/pF). Capacitance was measured using the automated capacitance cancellation function of the EPC-9 (HEKA, Germany). All values are given as mean ± SEM. Patch pipettes (Sutter Instrument, CA) were pulled and polished, had a tip resistance of 2-3 MΩ when filled with internal solution. Standard extracellular solution contained (in m m): 140 NaCl, 2.8 KCl, 2 MgCl2, 1 CaCl2, 10 HEPES-NaOH, and 11 Glucose (pH 7.2, 300 mOsm). Standard pipette-filling solution for screening of cannabinoids on inducible hTRPM7/HEK293 cells contained (in m m): 130 K-glutamate, 8 NaCl, 1 MgCl2, 10 EGTA, and 10 HEPES-KOH. For other patch-clamp experiments using hTRPM7/HEK293 cells, internal cellular solution contained (in m m): 140 Cs-glutamate, 8 NaCl, 10 Cs-BAPTA, 10 HEPES-CsOH. Na-ATP, MgCl2, and CaCl2 were supplemented with desired concentrations of [Mg⋅ATP]i, [Mg2+]i, and [Ca2+]I as calculated with WebMaxC (https://somapp.ucdmc.ucdavis.edu/pharmacology/bers/maxchelator/webmaxc/webmaxcS.htm). For the experiments of CBGA preincubation and TRPM6/TRPM7, internal cellular solution contained (in m m): 130 Cs-glutamate, 8 NaCl, 10 Cs-BAPTA, 10 HEPES-CsOH, 1 Na-ATP, and 1.764 MgCl2. Cells were incubated in extracellular solution containing 1.8 µm CBGA or acetonitrile as a control, patch-clamp was started after 10 mins and continued by 60 mins. For intracellular CBGA patch-clamp experiments, each concentration of CBGA was added in the same internal cellular solution to the experiment of CBGA preincubation. For transiently transfected TRPM6, internal solution contained (in m m): 130 Cs-glutamate, 8 NaCl, 10 Cs-BAPTA, and 10 HEPES-CsOH. For U266 cells, internal solution contained (in m m): 140 Cs-glutamate, 8 NaCl, 1 MgCl2, 10 EGTA, and 10 HEPES-CsOH. All internal solutions were adjusted to pH 7.2, 300 mOsm. Cannabinoids were diluted to each concentration with the extracellular solution used for patch-clamp experiments. Acetonitrile was diluted in extracellular solution and used as a control application.
High-Throughput Screening Bioassay
U266 cells were incubated in fresh culture medium supplemented with 2 µm Fura-2-AM (Invitrogen), 0.1% Pluronic F-127 (Sigma-Aldrich), and 2 m m Probenecid (Sigma) and incubated at 37°C under 5% CO2 for 1 h. Cells were washed 3 times with solution containing (in m m): 140 NaCl, 2.8 KCL, 10 HEPES-NaOH, 11 Glucose (pH 7.2, 300 mOsm) after Fura-2-AM incubation, fresh solution was added, 50000 cells were plated in each well of black 96-well plates with clear bottom (Greiner, Austria), and transferred to a kinetic plate reader (Hamamatsu FDSS-7000EX, Hamamatsu Photonics KK, Japan). Baseline fluorescence was measured for 1 min before application of cannabinoids (0, 1, 3, 5, 10, and 30 μm, respectively). After 10 min incubation with cannabinoids, 0.5 m m MnCl2 was applied. Data acquisition continued for another 340 s. Fluorescence was excited at 360 nm, and emission was measured at 510 nm to visualize intracellular Mn2+ quench of Fura-2. Data were normalized to the point before application of MnCl2 (650-659 s), and Mn2+ quench was measured at 661-690 s into the experiment. The 0 μm cannabinoid trace was set as maximum quench (100% TRPM7 activity) as a vehicle control.
Results
Cannabigerolic Acid is a Potent TRPM7 Inhibitor
Cannabinoids have previously been shown to modulate members of the TRP channels.4,15,16,42 We previously reported that TRPM7 is involved in the progression of kidney damage and fibrosis30 and recently found that CBGA confers a reno-protective effect at least partially through blocking of the TRPM7 ion channel.28 We therefore performed a whole-cell patch-clamp screen of 22 pure cannabinoids at a concentration of 10 µm against the TRPM7 channel inducibly overexpressed in HEK293 cells. Figure 1 shows the strong inhibition of TRPM7 in response to various cannabinoids applied extracellularly for 120 s. Figure 1A illustrates the averaged temporal development of TRPM7 outward currents at +80 mV derived from high-resolution outwardly rectifying membrane currents evoked by voltage ramps spanning −100 to +100 mV over 50 ms (Figure 1C). The outward currents under our experimental conditions are carried by monovalent Cs+, whereas the much smaller inward currents are carried by divalent cations Ca2+ and Mg2+. Application of 10 µm CBGA completely blocked both the outward currents (Figure 1A) as well the much smaller inward currents as illustrated in a typical individual cell for the + 80 and −80 mV (Figures 1A and B).
Cannabinoids are potent inhibitors of TRPM7 currents. The inhibitory effect of cannabinoids on TRPM7 current was assessed in whole-cell patch-clamp recordings on hTRPM7 overexpressed in HEK293 cells. In all experiments, the intracellular solution contained free Mg2+ at 780 µm with no added Mg·ATP. Cannabinoids were applied extracellularly at 10 µm from 140 to 260 s (black bars). Average TRPM7-mediated outward currents at +80 mV were extracted from ramp currents evoked by voltage ramps (−100 to +100 mV) delivered at 0.5 Hz and plotted as a function of time. The data points were normalized to the time point of cannabinoid application (140 s). (A) Inhibition of TRPM7 outward currents by cannabigerolic acid (CBGA) at +80 mV. (B) Outward and inward currents (at +80 and −80 mV; note the different y-axes scales) taken from a typical cell taken from (A), illustrating that CBGA inhibited both inward and outward currents. (C) High-resolution currents as a function voltage evoked by voltage ramps of −100 to +100 mV from the same cell as in (B) before (140 s) and after application of CBGA (260 s). (D) Average TRPM7 outward currents at +80 mV (n = 3-4) exposed to various cannabinoids (only cannabinoids with strong inhibition illustrated). (E) Bar graph comparing the remaining TRPM7 currents at the end of each cannabinoid application (260 s).
For convenience reasons and due to their larger amplitudes, we analyzed the effects of 22 cannabinoids on TRPM7 outward currents. Examples of the inhibitory effects of various cannabinoids applied at 10 µm are illustrated in Figure 1D and the bar graph in Figure 1E quantifies the efficacy of the various cannabinoids by plotting the relative amplitudes of TRPM7 outward currents at 260 s into the experiment. The acidic versions of cannabinoids (except CBDVA) inhibited TRPM7 activity by more than 75% (Figure 1E; blue bars) and most of the decarboxylated cannabinoids had no significant effect, except for CBG which had a modest inhibition of 43% (Figure 1E; red bars). However, 2 varinol forms of cannabinoids, cannabigerovarin (CBGV) and cannabidivarin (CBDV) strongly suppressed TRPM7 current activity by 81% and 93%, respectively.
Since CBGA and CBDV exhibited the strongest inhibition to TRPM7, we next performed experiments to assess their concentration-response behavior on TRPM7 (Figure 2). At a concentration of 10 µm, CBDV completely suppressed TRPM7 currents (Figure 2A; purple circles), whereas 3 µm CBDV inhibited 40% (Figure 2A; blue circles) and 1 µm resulted in 11% inhibition (Figure 2A; red circles). The dose-response curve fitted to these data yielded a half-maximal inhibitory concentration (IC50) of CBDV of 3.4 ± 0.2 µm (Figure 2C; black). Similarly, 1 µm CBGA suppressed 35% of TRPM7 (Figure 2B; red circles), 0.3 µm inhibited TRPM7 by 26% (Figure 2B; green circles), and the calculated IC50 was 1.8 ± 0.6 µm (Figure 2C; red circles). Taken together, these data suggest that CBDV and CBGA provide a strong and rapid block of TRPM7 currents in a dose-dependent manner and therefore emerge as the most potent inhibitors of TRPM7.
Cannabigerolic acid (CBGA) and cannabidivarin (CBDV) are a potent TRPM7 inhibitors. The inhibitory effect of CBDV and CBGA on TRPM7 current was assessed in whole-cell patch-clamp recordings on hTRPM7 overexpressed in HEK293 cells. The levels of free Mg2+ and Mg·ATP (if present) are indicated in the graphs. Cannabinoids were applied extracellularly at the indicated concentrations from 140 to 260 s (bars). Average TRPM7-mediated outward currents at +80 mV were extracted from ramp currents evoked by voltage ramps (−100 to +100 mV) delivered at 0.5 Hz and plotted as a function of time. The data points were normalized to the time point of cannabinoid application (140 s). (A) CBDV concentrations of 1 µm (n = 7), 3 µm (n = 12), and 10 µm (n = 3) were applied. Standard Ringer without cannabinoid contained acetonitrile as vehicle and was used as control (n = 3). (B) Cannabigerolic acid concentrations of 0.3 µm (n = 24), 1 µm (n = 6), 3 µm (n = 6), and 10 µm (n = 3) were applied. (C) Concentration-response curves of CBDV and CBGA on TRPM7 currents. Data points were obtained from the normalized currents in panels (A) and (B) at 260 s. Data fits yielded IC50 values for CBDV of 3.4 ± 0.2 µm and for CBGA of 1.8 ± 0.6 µm. (D) Cells were preincubated with either acetonitrile as vehicle control (n = 16) or 1.8 µm CBGA (n = 13) for 10-60 min prior to whole-cell recordings. (E) Cannabigerolic acid concentrations of 0.3 µm (n = 7), 0.6 µm (n = 8), and 1 µm (n = 6) were applied intracellularly. Standard intracellular solution without cannabinoid contained acetonitrile as vehicle and was used as control (n = 5). (F) Concentration-response curve of intracellular CBGA on TRPM7 currents. Data points were obtained from the normalized currents in panel (E) at 200 s. Data fit yielded an IC50 value for CBGA of 407 ± 46 n m.
We noticed that the inhibition of TRPM7 by low doses of CBGA was slow and accompanied by a slow and small increase in cell membrane capacitance 0.3-0.6 pF at concentrations of 5-10 µm (data not shown), indicating that the compound might partition into and possibly cross the plasma membrane to inhibit TRPM7. In the absence of a steady-state inhibition during the 120 s acute extracellular application of CBGA, the cannabinoid’s true IC50 value could well be lower if applied for longer times and/or administered intracellularly. To test this notion, we preincubated cells with the previously calculated IC50 concentration of 1.8 µm CBGA for 10-60 min and then recorded TRPM7 currents in patch-clamp recordings. This resulted in a stronger than 50% inhibition (Figure 2D), suggesting that longer incubations allowed CBGA to approach steady state, but may still represent an underestimate as in whole-cell recordings with CBGA-free intracellular solutions CBGA could escape through the patch pipette and gradually alleviate the block. Therefore, we performed additional experiments in which we perfused CBGA directly into the cytosol by including it in patch pipette. Indeed, perfusing cells with increasing concentrations of CBGA potently suppressed TRPM7 currents, resulting in an IC50 value 407 n m (Figure 2F), suggesting that CBGA inhibits TRPM7 at submicromolar concentrations from the cytosolic side.
Inhibitory Effect of CBGA is Selective for TRPM7 Over TRPM6
TRPM7 shares significant structural and functional characteristics with its sister channel TRPM622,43,44 and both can even form heteromeric assemblies with distinct properties.45–47 TRPM6 and TRPM7 have distinct tissue distributions in that TRPM7 is ubiquitously expressed,48 whereas TRPM6 is primarily expressed in the kidney and intestine, where it plays a significant role in magnesium reabsorption.44 Both proteins are divalent-selective ion channels39,47,49,50 and they both possess a functional kinase domain.43,51 Their main functional differences are divergent sensitivities to free Mg2+ and Mg·ATP. Thus, TRPM7 is negatively regulated by physiological levels of Mg·ATP,39 whereas TRPM6 is largely insensitive to the adenine nucleotide complex46 and heteromeric channels that include TRPM6 are much less sensitive to Mg·ATP.45,46 Conversely, TRPM6 is exquisitely sensitive to even low micromolar free Mg2+ levels,46 whereas TRPM7 is regulated by hundreds of micromolar levels.37,39 To determine CBGA’s selectivity for TRPM7, we assessed the effects of CBGA on TRPM6 channels transiently overexpressed in HEK293 cells.46 TRPM6 currents activated extremely rapidly when perfused with Mg2+-free intracellular solutions, likely due to the diffusional escape of cytosolic Mg2+ ions.46 Applying 10 µm CBGA was completely ineffective in inhibiting fully activated TRPM6 currents (Figure 3A), whereas this concentration of CBGA completely blocks TRPM7 currents (see Figure 2B).
Homomeric TRPM6 and heterormeric channels of TRPM7 and TRPM6 are less sensitive to cannabigerolic acid (CBGA) inhibition. The same experimental protocol as in Figure 2B but recorded in HEK293 cells transiently overexpressing TRPM6 or stably transfected with heteromeric TRPM7 and TRPM6 channels. (A) TRPM6 currents in transiently transfected HEK293 cells evoked by perfusing cells with 0 Mg2+ and Mg·ATP. The data points were normalized to the time point before 10 µm CBGA (n = 10) application (200 s). Standard Ringer without cannabinoid contained acetonitrile as vehicle and was used as control (n = 9). (B) Heteromeric TRPM6/TRPM7 currents in stably transfected HEK293 cells. The data points were normalized to the time point before CBGA application (140 s). Concentrations of 0.3 µm (n = 7), 1 µm (n = 7), 5 µm (n = 8), and 10 µm (n = 10) were applied from 140 to 260 s (bar). Standard Ringer without cannabinoid contained acetonitrile as vehicle and was used as control (n = 5). (C) The concentration-response curve of CBGA on heteromeric TRPM7/TRPM6 currents (IC50 = 8 ± 2.5 µm). Data points for the curve were obtained from the normalized currents in panel (A) at 260 s.
Unfortunately, TRPM6 channels cannot be recorded under the same experimental conditions as TRPM7 owing to their high sensitivity to free Mg2+, so that these experiments were conducted with intracellular solutions devoid of free Mg2+ and Mg·ATP. As will be shown below, the absence of Mg2+ and Mg·ATP also greatly diminishes the efficacy of CBGA to inhibit TRPM7. For this reason, we investigated the effects of CBGA on heteromeric TRPM7/TRPM6 channels stably overexpressed in HEK293 cells and recorded with intracellular solutions containing 700 µm Mg2+ and 860 µm Mg·ATP. As illustrated in Figure 3B, increasing concentrations of CBGA inhibited these heteromeric channels, but with significantly lower potency than previously determined for TRPM7 (see Figures 2B and C), resulting in an IC50 value of 8 µm (Figure 3C) and suggesting that TRPM6 is much less or not at all susceptible to CBGA inhibition.
Inhibitory Effect of CBGA Requires a Functional Kinase Domain
The inhibitory effect of CBGA on TRPM7 is consistent with the previously described suppression observed in our acute nephropathic mouse model study.28 Given the unique bifunctional properties of the protein (possessing both an ion channel function and protein kinase activity), we investigated whether CBGA inhibits TRPM7 via its endogenous kinase domain. Utilizing 2 HEK293 cell lines overexpressing phosphotransferase activity-deficient versions of human TRPM7 carrying point mutations K1648R or G1799D in the kinase domain.40 Lysine residue K1648 is essential for Mg·ATP binding and its replacement with arginine K1648R results in a complete loss of a TRPM7 kinase activity.40 Similarly, the glycine residue G1799 is thought to coordinate the Mg2+ ion of adenine nucleotides and its mutation to aspartic acid G1799D also results in loss of kinase activity. We examined the inhibitory effect of CBGA on both these mutants’ currents. While 10 µm CBGA completely blocked TRPM7 currents in TRPM7/WT overexpressing HEK293 cells (92.4%, Figure 2B; purple), it inhibited the TRPM7/K1648R mutant cell line by just 36% (Figure 4A; purple). Moreover, CBGA concentrations of 0.3, 1, and 3 µm showed no significant inhibition of TRPM7 currents (Figure 4A; green, red, and blue circles, respectively), suggesting that an active kinase domain is required for CBGA inhibition. The IC50 of CBGA in the TRPM7/K1648R mutant was 20.2 ± 12.9 µm (Figure 4B), much higher than the TRPM7/WT value of 1.8 ± 0.6 µm and currents recorded from HEK293 cells overexpressing the other phosphotransferase-deficient mutant (G1799D) were completely unaffected by 5 µm CBGA (Figure 4C). Based upon these findings, we conclude that the potent CBGA-mediated inhibition of TRPM7 currents requires a functional TRPM7 kinase domain.
The inhibitory effect of cannabigerolic acid (CBGA) on TRPM7 requires a functional TRPM7 kinase domain. The same experimental protocol as in Figure 2B but recorded in HEK293 cells stable transfected with the phosphotransferase activity-deficient point mutations K1648R and G1799D. The levels of free Mg2+ and Mg·ATP are indicated in the graphs. (A) The data points were normalized to the time point before cannabinoid application (140 s). Concentrations of 0.3 µm (n = 13), 1 µm (n = 12), 3 µm (n = 7), 5 µm (n = 7), and 10 µm (n = 7) were applied from 140 to 260 s (bar). Standard Ringer without cannabinoid contained acetonitrile as vehicle and was used as control (n = 7). (B) The concentration-response curve of CBGA on K1648R TRPM7 currents (IC50 = 20.2 ± 12.9 µm). Data points for the curve were obtained from the normalized currents in panel (A) at 260s. (C) The same experimental protocol as in Figure panel (A) but recorded in HEK293 cells stable transfected with the phosphotransferase activity-deficient point mutation G1799D. The data points were normalized to the time point before cannabinoid application (140 s). CBGA was applied at 5 µm (n = 7) and standard Ringer without cannabinoid contained acetonitrile as vehicle and was used as control (n = 5).
Inhibition of TRPM7 By CBGA is Sensitized By Mg2+ and Mg⋅ATP
Both Mg⋅ATP and Mg2+ are known to bind to the TRPM7 kinase domain and are critical for its phosphotransferase activity. Both also regulate TRPM7 channel function as they cause a reduction in TRPM7 channel activity.37,39,40 As illustrated in Figure 4, CBGA was less effective at inhibiting TRPM7 currents in the TRPM7/K1648R point mutant cell line and completely ineffective in the TRPM7/G1799D mutant, prompting us to investigate whether the CBGA-mediated inhibition of TRPM7 was affected by intracellular levels of free Mg2+ and/or Mg⋅ATP. Figure 5A shows that at a physiological intracellular concentration of 780 µm free Mg2+, 5 µm CBGA blocked 50% of TRPM7 currents with either 1 (Figure 5A; green) or 2 m m Mg⋅ATP (Figure 5A; blue) in HEK293 cells overexpressing hTRPM7/WT. However, omitting Mg⋅ATP dramatically limited the inhibitory effect of 5 µm CBGA on TRPM7 to just 6.8% (Figure 5A; black circles).
Cannabigerolic acid (CBGA) inhibits TRPM7 currents in an Mg⋅ATP- and Mg2+-dependent manner. (A) The effect of Mg⋅ATP on suppression of TRPM7/WT currents by CBGA. The internal solution contained 780 µm intracellular free Mg2+ and additionally Mg·ATP concentrations of 0 m m (n = 6), 1 m m (n = 6), and 2 m m (n = 6). A 5 µm CBGA was applied from 140 to 260 s (bar). The data points were normalized to the time point before CBGA application (140 s). (B) The effect of free intracellular Mg2+ on the inhibition of TRPM7/WT currents by CBGA. The internal solution contained 1 m m Mg⋅ATP and additionally free Mg2+ concentrations of 0 µm (n = 7), 240 µm (n = 6), 780 µm (n = 6), and 1 m m (n = 7). A 5 µm CBGA was applied from 140 to 260 s (bar). (C) Same experimental protocol as in (B) but the internal solution contained 0 m m Mg⋅ATP and additionally free Mg2+ concentrations of 0 µm (n = 4), 240 µm (n = 6), 780 µm (n = 6), and 1 m m (n = 7). A 5 µm CBGA was applied from 140 to 260 s (bar). (D) Concentration-response curves of free Mg2+ on TRPM7/WT currents without CBGA (EC50 = 883 ± 251 µm), on TRPM7/WT exposed to 5 µm CBGA in WT (EC50 = 878 ± 96 µm), and on TRPM7/K1648R mutant (red; EC50 = 2.2 ± 0.7 m m). Data points for the curves were obtained from the normalized currents in panels (B) and (E) at 260 s and at 140 s for WT control without CBGA. (E) The effect of free intracellular Mg2+ on the inhibition of K1648R currents by CBGA. The internal solution contained 1 m m Mg⋅ATP and additionally free Mg2+ concentrations of 0 µm (n = 5), 240 µm (n = 7), 780 µm (n = 7), and 1 m m (n = 6). A 5 µm CBGA was applied from 140 to 260 s (bar). (F) Same experimental protocol as in (E) but the internal solution contained 0 m m Mg⋅ATP and additionally free Mg2+ concentrations of 0 µm (n = 5), 240 µm (n = 6), 780 µm (n = 6), and 1 m m (n = 7). A 5 µm CBGA was applied from 140 to 260 s (bar).
We next determined the inhibitory efficacy of 5 µm CBGA in the presence of various concentrations of free Mg2+ in the presence of a constant intracellular Mg⋅ATP concentration of 1 m m (Figure 5B). Free Mg2+ by itself will inhibit TRPM7 in a concentration-dependent manner (Figure 5D; black) with an IC50 of 883 ± 251 µm. The application of CBGA under these conditions did not significantly alter this relationship. In the presence of 5 µm CBGA, physiological levels of 780 µm free Mg2+ inhibited 50% of TRPM7 currents (Figure 5B; green), and 1 m m free Mg2+ slightly enhanced the inhibition of TRPM7 currents to 57% (Figure 5B; red). Conversely, lowering the concentration of free Mg2+ to 240 µm greatly diminished the inhibitory effect of CBGA to just 26% (Figure 5B; orange). The IC50 of intracellular free Mg2+ in the presence of 5 µm CBGA was calculated to be 878 ± 96 µm (Figure 5D; blue) in HEK293 cells overexpressing TRPM7/WT. This is very similar to the inhibition by free Mg2+ in the absence of CBGA, suggesting that a constant CBGA concentration does not significantly alter the Mg2+-sensitivity of TRPM7.
To further probe the Mg2+-dependence of CBGA, we examined whether CBGA requires Mg⋅ATP for Mg2+-dependent TRPM7 inhibition. Here, we omitted 1 m m Mg⋅ATP from the intracellular medium and assessed the dose-response behavior at varying concentrations of intracellular free Mg2+. Figure 5C illustrates that the inhibitory effect of 5 µm CBGA on TRPM7 was largely lost in the presence of both 780 µm (Figure 5C; green) and 240 µm free intracellular Mg2+ (Figure 5C; orange). However, CBGA’s inhibitory effect on TRPM7 was partially restored at 1 m m free Mg2+, at which concentration TRPM7 currents were again inhibited by 32% when exposed to 5 µm CBGA (Figure 5C; red). The above results demonstrate that the inhibitory effect of a constant dose of CBGA on TRPM7 is enhanced and sensitized by the presence of intracellular Mg2+ and Mg·ATP in a dose-dependent manner.
To further support the notion that the binding of Mg⋅ATP and/or Mg2+ to the kinase domain of TRPM7 is crucial to enable CBGA-mediated modulation of TRPM7 currents, we used HEK293 cells overexpressing the kinase-deficient TRPM7/K1648R point mutant. Here, 5 µm CBGA was less effective at suppressing TRPM7 currents in the kinase mutant cells when intracellular levels of free Mg2+ were progressively reduced from 1 m m (26%, Figure 5E; red), to 780 µm (19%, Figure 5E; green), and 240 µm (7%, Figure 5E; orange). Additionally, there was no inhibitory effect in Mg2+-free conditions in the presence of 1 m m Mg⋅ATP in the TRPM7/K1648R overexpressing HEK293 cells (Figure 5E; black). Here, the TRPM7/K1648R mutant cells provided a higher IC50 value of intracellular free Mg2+ of 2.2 ± 0.7 m m (Figure 4D; red) as compared to 878 ± 96 µm in TRPM7/WT cells. Not surprisingly, under intracellular Mg⋅ATP-free conditions, the inhibitory effect of CBGA on TRPM7 currents was essentially absent, regardless of the free Mg2+ concentration in TRPM7/K1648R overexpressing HEK293 cells (Figure 5F). Taken together, these data suggest that the CBGA-mediated inhibition of TRPM7 currents not only requires a functional kinase domain, but also needs Mg⋅ATP and/or free Mg2+ to be effective.
[Ca2+]i Reduces the Inhibitory Effect of CBGA on TRPM7
We have previously reported that TRPM7 regulates SOCE52 and have recently demonstrated that CBGA suppresses IL-2 production in T cells through SOCE inhibition.41 Therefore, we assessed the effect of intracellular calcium on CBGA’s ability to inhibit TRPM7. The absence of intracellular free Ca2+ allowed 10 µm CBGA to completely block TRPM7 currents (Figure 6A; black) in hTRPM7/WT overexpressing cells. However, the presence of physiological free Ca2+ of 75 n m reduced the inhibitory effect of CBGA on TRPM7 to 48% (Figure 6A; blue), and this effect was exacerbated at higher free Ca2+ concentrations, where CBGA-mediated inhibition was reduced to 36% at 150 n m free Ca2+ (Figure 6A; red) and 33% at 300 n m free Ca2+ (Figure 6A; green). The EC50 of intracellular free Ca2+ concentration was 71 ± 3 n m (Figure 6B), right around resting levels of free Ca2+.
Calcium interferes with TRPM7 inhibition by cannabigerolic acid (CBGA). (A) The effect of intracellular free Ca2+ on TRPM7 inhibition by CBGA. The internal solution was buffered to free Ca2+ concentrations of 0 n m (n = 6), 75 n m (n = 7), 150 n m (n = 7), and 300 n m (n = 7) using 10 m m BAPTA and appropriate concentrations of CaCl2. A 10 µm CBGA was applied from 140 to 260 s (bar). The data points were normalized to the time point before CBGA application (140 s). (B) The concentration-response curve of intracellular calcium to inhibit TRPM7 currents by 10 µm CBGA (EC50 = 71 ± 3 n m). Data points for the curve were obtained from the normalized currents in panel (A) at 260 s. (C) The same experimental protocol as in (A) using 0 n m (n = 7) and 150 n m free Ca2+ (n = 6) in K1648R TRPM7 overexpressing HEK293 cells.
Next, we investigated this effect of intracellular Ca2+ on HEK293 cells overexpressing the kinase-deficient K1648R mutant of TRPM7 in response to 10 µm CBGA. Figure 6C demonstrates that intracellular Ca2+-free conditions were less efficient at reducing the inhibitory effect of 10 µm CBGA on TRPM7 in TRPM7/K1648R overexpressing cells (29%, Figure 6C; black) compared to TRPM7/WT overexpressing HEK293 cells (69%, Figure 6A; black). Unlike the situation in wild-type cells, increasing intracellular free Ca2+ to 150 n m in these TRPM7/K1648R overexpressing HEK293 cells maintained a 29% inhibition and failed to significantly affect the CBGA-mediated inhibition of TRPM7 (Figure 6C; red). Taken together, these results suggest that the efficacy of CBGA-mediated inhibition of TRPM7 decreases when intracellular free Ca2+ increases and, similarly to the free Mg2+ modulation, this effect appears to occur through the TRPM7 kinase domain.
CBGA Inhibits Native TRPM7 Currents in B Lymphocytes
The results presented above demonstrate that CBGA inhibits currents through TRPM7 channels heterologously overexpressed in HEK293 cells. We next investigated natively expressed, endogenous TRPM7 currents in the B lymphocyte myeloma cell line U266, where TRPM7 has previously been shown to be one of the main channels involved in multiple myeloma dissemination.53 We assessed the inhibitory effect of CBGA on native TRPM7 activity in U266 cells in both in cell populations using fluorescence quenching of Fura-2 and in individual cells using the whole-cell patch-clamp technique. Since TRPM7 channels are constitutively active at low levels and conduct divalent cations such as Ca2+, Mg2+, Mn2+, and Zn2+, TRPM7 channel function can be assayed using Mn2+ as a surrogate for ion transport through TRPM7,50,54 although cells have additional pathways that allow Mn2+ influx. Permeation of Mn2+ results in the fluorescence quenching of the calcium indicator Fura-2.50,54,55 Using a kinetic plate reader, we monitored the real-time Mn2+ quenching of Fura-2 by various concentrations of CBGA in U266 cells. We used the TRPM7 inhibitor NS859330,52,56 as a positive control, which at 10 µm suppressed 51% of the native TRPM7 activity in U266 cells. CBGA dose-dependently inhibited Mn2+ quenching of Fura-2 (Figure 7A) with an IC50 of 4.8 ± 1.1 µm (Figure 7B). Next, we assessed the inhibitory effect of CBGA on native TRPM7 currents in individual cells using the whole-cell patch-clamp technique. Here, 5 µm CBGA significantly blocked TRPM7 currents by 67% (Figures 7C and D; blue) and 10 µm CBGA completely blocked TRPM7 currents (Figures 7C and D; red) compared to vehicle control application (Figures 7C and D; black). These data suggest that CBGA can suppress native TRPM7 channel activity at low micromolar concentrations.
Cannabigerolic acid (CBGA) inhibits native TRPM7 currents in U266 cells. (A) U266 cells were used to investigate the inhibitory efficacy of CBGA in TRPM7-specific bioassay using fluorescence quenching of Fura-2 by Mn2+ influx (1 µm GdCl3 was used to block Mn2+ entry through SOCE in all wells). Cells were were pretreated with various concentrations of CBGA for 10 mins (bar) and then exposed to 0.5 m m external MnCl2 (bar) to induce fluorescence quenching. NS8593 was used as positive control to suppress TRPM7. (B) Concentration-response curve of CBGA on Mn2+-mediated fluorescence quench in U266 cells. Data points for the curve were obtained from the normalized Fura-2 quench in (A) at 1000 s. Data fit yielded a half-maximal inhibitory concentration IC50 of 4.8 ± 1.1 µm. (C) Cannabigerolic acid inhibition of native TRPM7 currents in U266 cells using patch-clamp recordings. Cannabigerolic acid at concentrations of 5 µm (n = 11) and 10 µm (n = 10) were applied from 120 to 320 s (bar). Standard Ringer’s solution without cannabinoid contained acetonitrile as vehicle and was used as control (n = 6). Data points were normalized to the time point before application of CBGA (120 s). (D) Bars reflect TRPM7 current amplitudes obtained from panel (C) at the end of CBGA application (320 s).
Discussion
We have previously reported an interplay between TRPM7 and SOCE, where TRPM7 regulates SOCE through its kinase domain.52 In addition, we have found that TRPM7 expression is upregulated in inflammatory renal damage and kidney fibrosis, and pharmacological intervention with the TRPM7 inhibitors NS859330 or CBGA28 provide a reno-protective effect. In the present study, we identify and characterize the mechanism of action and the conditions that modulate the inhibitory effects of CBGA on TRPM7. This inhibitory effect requires a functional TRPM7 C-terminal kinase (Figures 4 and 5) and is concentration-dependently enhanced by intracellular Mg2+ and Mg⋅ATP acting at the kinase domain’s Lys-1648 residue (Figure 5). We also determined that intracellular Ca2+ interferes with the efficacy of CBGA on TRPM7 activity (Figure 6), possibly by competing with Mg2+ at the kinase locus.
The presently most potent and selective TRPM7 inhibitor is waixenicin A, a marine-derived xenicane diterpene isolated from the soft coral Sarcothelia edmondsoni. Waixenicin A activity is regulated by intracellular free Mg2+, and its potency reduced in TRPM7/K1648R overexpressing cells that harbor a point mutation in kinase domain.54 This suggests that waixenicin A suppresses TRPM7 currents through Mg2+-dependent mechanism at the kinase domain or a location that interacts with this domain. Our results show that CBGA appears to possess similar properties to waixenicin A in that CBGA synergizes with intracellular Mg2+ and Mg·ATP to block TRPM7 channel activity. We propose that intracellular free Mg2+ and Mg⋅ATP facilitate binding and coordination of the CBGA molecule at the kinase domain itself or some allosteric site that interacts with the kinase domain and ultimately results in inhibition of TRPM7 channel activity. While we have demonstrated a functional interplay of cannabinoids with Mg2+ and Mg·ATP and kinase domain mutations, we cannot rule out potential interactions and binding sites outside the kinase domain, including the possibility that the autophosphorylation status of the channel might be relevant.51,57,58 Given the fact that a single point mutation in the cytosolic kinase domain essentially obliterates CBGA’s ability to block TRPM7 indicates that a relatively nonspecific and more general membrane-delimited effect is unlikely. Thus, our data provide good circumstantial evidence for the requirement of a functional kinase activity rather than a direct demonstration that CBGA binds to the kinase domain at its catalytic sites, although this remains an intriguing scenario. This awaits additional biochemical/structural work akin to that provided for other agonists and antagonists of TRPM7.59
We previously reported CBGA inhibits SOCE by virtue of blocking CRAC channels subsequently suppressing IL-2 production in vitro and impeding the production of numerous inflammatory cytokines in vivo in mouse kidney.41 Since TRPM7 can also modulate SOCE via its kinase activity, CBGA could have a dual effect on calcium influx by inhibiting SOCE directly and indirectly via TRPM7. Moreover, since the present study establishes that high intracellular Ca2+ interferes with the efficacy of CBGA’s inhibitory action of TRPM7 activity (Figure 5), there may be yet another synergistic effect of CBGA in play. As CBGA can inhibit SOCE and reduce intracellular Ca2 concentrations when stimulated, the reduction of intracellular Ca2+ would facilitate CBGA’s suppression of TRPM7 channels and further suppress SOCE,52 thereby providing a positive feedback loop to reduce SOCE-mediated inflammation. These interconnected inhibitory effects may be the underlying mechanisms that endow CBGA with strong anti-inflammatory properties.
Although both CBD and CBG appear to have anti-inflammatory effects in vivo60 ), we found that CBGA, but not CBD, suppressed mRNA expression of inflammatory cytokines and TRPM7 protein in a cisplatin-induced model of acute kidney injury.28 Interestingly, CBD was generally ineffective in reducing most inflammatory markers and the combination of CBGA + CBD treatment reduced most mRNA levels similarly to CBGA treatment, although the suppressive effect was somewhat less in some cytokines, possibly indicating negative cooperativity effects of these cannabinoids when combined.28 In patch-clamp experiments, we found that CBD didn’t suppress TRPM7 currents28 and in the present study, we found that the additional presence of 10 µm CBD did not significantly affect the IC50 of CBGA on TRPM7 currents. Interestingly, CBD nevertheless seemed to limit the total extent of CBGA-mediated inhibition to ∼85%, which could help explain its apparent reduction of CBGA’s inhibitory efficacy on TRPM7 in vivo in the nephropathic model.
Our manuscript provides data on the effects of CBGA, a compound that is generally regarded as safe and consumed by people as a nutraceutical. Its potency under physiological free Mg2+ and Mg·ATP conditions (intracellular IC50 of 407 n m) is higher than most other TRPM7 inhibitors (eg, NS895356 and VER15500861), with the possible exception of waixenicin A, which has an IC50 value in the low nanomolar range and exhibits a similar regulation by Mg2+ and nucleotides as CBGA54 or CCT128930,62 which inhibits TRPM7 with IC50 values of 0.63 to 0.86 µm independently of intracellular free Mg2+ levels.62 Interestingly, CCT128930 is a known ATP-competitive protein kinase B inhibitor, although its inhibitory activity on TRPM7 was proposed to involve residues near the TRPM7 selectivity filter rather than the kinase domain.62 While some reports have claimed submicromolar IC50 values for NS8953 and VER155008 as well, they were either obtained through indirect measurements of Ca2+ entry induced by naltriben or under nonphysiological levels of 0 Mg2+ and MgATP in the presence of HEDTA (completely divalent-free). In intact cells under physiological conditions, where cytosolic levels of Mg2+ and Mg·ATP are in the hundreds of micromolar or several millimolar, NS8953 and VER155008 typically require several micromolar concentrations to exhibit significant effects.30,56 Cannabigerolic acid exhibits submicromolar potencies when allowed longer exposure to gain cytosolic access or is administered intracellularly, where it inhibits with an IC50 of 407 n m (see Figures 2D–-FF).
While 10 µm CBGA is needed to achieve complete block, the IC50 is submicromolar when administered intracellularly and 1.8 µm when applied acutely, extracellularly, and for 120 s. Note that in our in vivo study on kidney disease,28 we observed efficacy at 10 mg/kg i.p. injections. Pharmacokinetic data indicate that CBGA at that dose injected i.p. can reach Cmax values 64 µg/mL in plasma,63 which is equivalent to 180 µm with a plateau of 20 µg/mL over 2 h. So even if one assumed poor bioavailability of 99% bound to serum albumins, one might still achieve low micromolar levels in plasma. Given the translocation into the cytosol and potential accumulation and retention, one might achieve effective levels of CBGA that can account for the observed biological effects,28 although another study used CBGA at 30-100 mg/kg in their anticonvulsant assays.17 While therapeutically effective dosing of CBGA is unknown, the maintenance dose of another cannabinoid (cannabidiol or CBG; Epidiolex) in humans is 25 mg/kg/d. Aside from its potential therapeutic value, CBGA appears to be a polyspecific inhibitor that targets CRAC channels and SOCE with submicromolar IC50 from the extracellular side41 and TRPM7 with similar potency from the cytosolic side. Interestingly, both of these targets interact with each other in that TRPM7 plays a role in refilling intracellular stores and modulates CRAC channel activity via its kinase domain52 and conversely, Ca2+ entry via CRAC channels would modulate the efficacy of CBGA in suppressing TRPM7 (see Figure 6).
In summary, CBGA can now be regarded as a potent submicromolar inhibitor of both SOCE via CRAC channels on the extracellular side as well as calcium and magnesium entry via TRPM7 channels on the cytosolic side. Inhibiting CRAC channels can have therapeutic potential in various medical conditions, primarily those involving immune system dysfunction or calcium dysregulation and particularly those in which lymphocytes play a prominent role. Some of the medical conditions in which CRAC channel inhibitors might be therapeutically useful include autoimmune and allergic disorders,64,65 transplant rejection,66 pancreatic diseases,67 chronic pain,68,69 and various cancers.70–72 The additional inhibitory effects of CBGA on TRPM7 can not only potentiate the suppression of SOCE through the regulation of CRAC channels via its kinase domain,52 but also offer additional therapeutic areas in which TRPM7 plays an important role of its own. This includes cardiac hypertrophy, arrhythmias, and vascular smooth muscle contraction,27,73–75 neurological disorders such as Alzheimer’s, Parkinson’s, epilepsy stroke, and ischemia,26,76–82 kidney disorders,28,30,83,84 and various cancers.23,36,85–89
Contributor Information
Sayuri Suzuki, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA.
Clay Wakano, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA.
Mahealani K Monteilh-Zoller, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA.
Aaron J Cullen, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA.
Andrea Fleig, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA. University of Hawaii Cancer Center, 651 Ilalo St., Honolulu, HI 96813, USA. John A. Burns School of Medicine, University of Hawaii, 651 Ilalo St., Honolulu, HI 96813, USA.
Reinhold Penner, Center for Biomedical Research, The Queen’s Medical Center, 1301 Punchbowl St., Honolulu, HI 96813, USA. University of Hawaii Cancer Center, 651 Ilalo St., Honolulu, HI 96813, USA. John A. Burns School of Medicine, University of Hawaii, 651 Ilalo St., Honolulu, HI 96813, USA.
Author Contributions
S.S. and A.F. performed patch-clamp experiments and analyses. S.S., C.W., and M.M. performed high-throughput screening experiments and analyses. A.C. prepared all cannabinoids. All authors wrote, edited, and reviewed the manuscript.
Funding
This work was supported by the National Center for Complementary and Integrative Health of the National Institutes of Health under award number R01AT011162 (R.P.), by Hamamatsu/Queen’s PET Imaging LLC (S.S.), and by the National Institute of Neurological Disorders and Stroke under award number R61NS124922 (A.F.). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Conflict of Interest
The authors declare that they have no conflicts of interest with the contents of this article. A.F. holds the position of Editorial Board Member for Function and is blinded from reviewing or making decisions for the manuscript.
Data Availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
