Modulation of T-type Ca2+ channels by Lavender and Rosemary extracts

Medicinal plants represent a significant reservoir of unexplored substances for early-stage drug discovery. Of interest, two flowering Mediterranean plants have been used for thousands of years for their beneficial effects on nervous disorders, including anxiety and mood. However, the therapeutic potential of these plants regarding their ability to target ion channels and neuronal excitability remains largely unknown. Towards this goal, we have investigated the ability of Lavender and Rosemary to modulate T-type calcium channels (TTCCs). TTCCs play important roles in neuronal excitability, neuroprotection, sensory processes and sleep. These channels are also involved in epilepsy and pain. Using the whole-cell patch-clamp technique, we have characterized how Lavender and Rosemary extracts, as well as their major active compounds Linalool and Rosmarinic acid, modulate the electrophysiological properties of recombinant TTCCs (CaV3.2) expressed in HEK-293T cells. Both the methanolic and essential oil extracts as well as the active compounds of these plants inhibit Cav3.2 current in a concentration-dependent manner. In addition, these products also induce a negative shift of the steady-state inactivation of CaV3.2 current with no change in the activation properties. Taken together, our findings reveal that TTCCs are a molecular target of the Lavender and Rosemary compounds, suggesting that inhibition of TTCCs could contribute to the anxiolytic and the neuroprotective effects of these plants.


Introduction
Medicinal plants have been identified and used throughout human history [1]. Because of their ability to synthesize a wide variety of chemical compounds (alkaloids, polyphenolics, terpenoids, fatty acids and lipids, etc.) either for their normal development or against stressful and threatening conditions, they have been suggested to be an interesting pharmaceutical industry. Moreover, because of the potential side and adverse effects of synthetic drugs, scientists interested in drug discovery have turned their attention to herbal medicines as effective lead compounds for the management of health ailments including inflammatory, cardiovascular and neurological disorders [2,3]. It's worth noting that 49% of the new chemical drugs that PLOS

Methanolic extraction protocol
Fresh plant materials were dried at 40˚C during 48 to 96 hours, then homogenized to fine powder by grinding and sieving until the stabilization of weight. 20 g of dried plant materials were extracted with 200 ml of pure methanol and kept on a rotary shaker for maceration for a total duration of 72h. Thereafter, the extracts were filtered and evaporated to dryness in Rotavapor 1 vacuum (60 rpm at 40˚C). The final extracts were stored at 4˚C for further studies.

Plant essential oil and active principles
Essential oils of Lavandula steachas, Lavandula angustifolia Miller and Rosmarinus officinalis were purchased from Vitalba (Sartène, France). Rosmarinic acid (RA) and Linalool were purchased from Sigma-Aldrich.

Cell culture and transfection protocols
HEK-293T cells stably expressing the Ca V 3.2 channels isoform (kindly provided by Dr. E. Perez-Reyes, University of Virginia) were cultivated in Dulbecco's Modified Eagle's Medium supplemented with GlutaMax, 400μg/ml G418 (Life Technologies) and 10% fetal bovine serum (Invitrogen). In some experiments, HEK-293T cells transfection with plasmids expressing human Ca V 3 constructs was performed using jet-PEI (QBiogen) with a DNA mix containing 0.5% of a GFP encoding plasmid and 99.5% of the Cav3 constructs. Two days after transfection, HEK-293T cells were dissociated with Versene (Invitrogen) and plated at a density of~35x10 3 cells in 35 mm Petri dish for electrophysiological recordings performed the following day.

Electrophysiological recordings
Whole-cell calcium currents were recorded at room temperature using an Axopatch 200B amplifier (Molecular Devices). For recording macroscopic T-type calcium currents, the extracellular solution contained the following (in mM): 135 NaCl, 20 TEACl, 2 CaCl 2 , 1 MgCl 2 , and 10 HEPES (pH adjusted to 7.25 with KOH,~330 mOsm). Borosilicate glass pipettes have a typical resistance of 1.5-2.5 MOhm when filled with the internal solution containing the following (in mM): 140 CsCl, 10 EGTA, 10 HEPES, 3 Mg-ATP, 0.6 GTPNa, and 3 CaCl 2 (pH adjusted to 7.25 with KOH~315 mOsm). Recordings were filtered at 2 kHz. During the Ca V 3.2 current recordings, the chamber was constantly perfused (~100 μl/min) with the control or with the drug solutions using a gravity-driven homemade perfusion device. Data were analyzed using the pCLAMP9 (Molecular devices) and GraphPad Prism (GraphPad) softwares. The doseresponse curves were obtained from fitting data to the Hill

Statistical analysis
Results are presented as the mean ± SEM, and n is the number of cells used. Statistical significance was evaluated by Student's unpaired t-test ( Ã P<0.05, ÃÃ P<0.01 and ÃÃÃ P<0.001)

Modulation of Ca V 3.2 calcium channels by medicinal plant methanolic extracts
In a first set of experiments, we tested the ability of several Mediterranean medicinal plants to modulate TTCCs. These experiments were performed using recombinant Ca V 3.2 channels. Ca V 3.2 channel modulation was determined by measuring the T-type current in whole cell configuration on cells stepped from -80 to -30 mV following superfusion of the extracts at a concentration of 30 μg/ml. Fig 1 illustrates the efficacy of four methanolic plant extracts by showing typical Ca V 3.2 current trace recordings and the corresponding time plots. Application of Lavandula stoechas (LS) inhibited significantly Ca V 3.2 channels (Fig 1A and 1B). The average current inhibition induced by 30 μg/ml of LS was 85% (I/Ictrl = 15 ± 5.2% p<0.01, n = 6). LS developed a fast inhibitory effect that did not readily reverse upon wash-out. Similarly, a 42% inhibition (I/Ictrl = 68 ± 2% p<0.01, n = 6) was obtained following application of the methanolic extract of Rosmarinus officinalis (RO) on Ca V 3.2 currents (Fig 1C and 1D). On the contrary, no significant inhibition was obtained after the application of Ricinus cummunis (RC) (Fig 1E and 1F, I/Ictrl = 95.7 ± 1.5%, n = 7) and Citrullus colocynthis (CC) extracts ( Fig  1G and 1H, I/Ictrl = 95.4 ± 2.1%, n = 6). These data led us to further investigate the efficacy of Lavandula and Rosmarinus species to modulate Ca V 3.2 channels.

Ca V 3.2 channel inhibition by Lavandula species and Linalool is concentration-dependent
Next, we characterized the effect of two Lavandula species essential oils; Lavandula stoechas (LS), Lavandula angustifolia Miller (LA) and their active principle Linalool. Ca V 3.2 current recordings were performed during application of increasing concentrations of LS, LA and Linalool (Fig 2). T-type current inhibition by these three compounds was concentration-dependent. Analysis of the dose-response curve after treatment with LS essential oil revealed IC 50 values of 16.9 ± 2.9 μg/ml (n = 7) with a Hillslope value of 0.9 ± 0.1 (Fig 2A). TTCCs were also inhibited by serial concentrations of LA Miller essential oil solutions (Fig 2B)

Effects of Lavandula essential oils and Linalool on Ca V 3.2 channel activation
Inhibition of Ca V 3.2 channels by Lavender and its natural constituent Linalool may be related to specific modifications in Ca V 3.2 channel gating properties. Hence, we investigated whether TTCC inhibition by Lavender could be related to change in channel availability or activation properties. Toward this goal, the inhibitory effect of these natural compounds on Ca V 3.2 current was studied for a wide range of depolarizing test potentials (TPs) from -80 to +10 mV. Representative Ca V 3.2 current traces before and after the application of 30 μg/ml LS are shown in Fig 3A (top and bottom panels respectively), as well as the corresponding current-voltage (I-V) curves ( Fig 3B). These average I-V curves show that 30 μg/ml of LS inhibit the amplitude of Ca V 3.2 currents similarly at all membrane potentials ( Fig 3B, n = 6). Moreover, application of LS did not significantly shift the activation curve of Ca V 3.2 channels. The V 0.5 for activation was -53.8 ± 0.4 mV for control condition and -54.9 ± 0.4 mV during LS application, respectively, revealing no significant change in steady-state activation in the presence of LS (n = 6, p = 0.11, Fig 3B). In addition, fitting of the individual current traces, as presented in Fig 3A, revealed that neither activation nor inactivation kinetics of Ca V 3.2 channels were changed after Lavender treatment (Fig 3C and 3D). Similar results were obtained with 50 μg/ml LA Miller essential oil (Fig 3E, n = 5) and 100 μM Linalool (Fig 3F, n = 6). Inhibition of Ca V 3.2 currents following treatment with LA Miller essential oil and Linalool was conserved over the complete range of test potentials. Also, the threshold potential for activation and the membrane potential of the maximum peak current were the same before and during the application of LA Miller, with no significant shift in the activation curve of Ca V 3.2 currents (V 0.5Ctrl = -54.8 ± 0.4 mV and V 0.5LAMiller = -54.1 ± 0.3 mV, n = 5, p = 0.19), and Linalool (V 0.5Ctrl = -50.5 ± 0.7 mV and V 0.5Linalool = -52.3 ± 0.8 mV, n = 6, p = 0.14).

Effects of Lavandula on Ca V 3.2 current steady-state inactivation
We next examined whether Lavender and its constituents could modify the steady-state inactivation properties of Ca V 3.2 channels. A representative family of Ca V 3.2 currents evoked by the protocol designed to measure steady-state inactivation is depicted in Fig 4A before (top traces) and during the application of 30 μg/ml LS (bottom traces). LS (30 μg/ml) produced a depressant action of the maximal conductance of Ca V 3.2 channels as well as a significant hyperpolarizing shift of steady-state inactivation from -75.8 ± 1.1 mV in control conditions to -81.3 ± 1.0 mV in LS (n = 6, p<0.05, Fig 4B). Similar results were obtained after application of LA Miller (50 μg/ml, Fig 4C) and Linalool (100 μM:~15 μg/ml, Fig 4D) with a significant shift in the steady-state inactivation from -75.9 ± 0.7 mV to -83.1 ± 0.8 mV for LA Miller (n = 5, p<0.001) and from -74.6 ± 0.7 mV to -81.3 ± 0.6 mV for Linalool (n = 6, p<0.001). These data demonstrate that Lavender constituents inhibit Ca V 3.2 current by decreasing the maximal conductance of Ca V 3.2 channels and induce a negative shift of Ca V 3.2 steady state-inactivation curve, suggesting that Lavender compounds could interact with the inactivated state of TTCCs. In addition, we have investigated the modulation of Cav3.2 current in the presence of 100 μM Linalool at three different frequencies of stimulation (1, 0.2 and 0.033 Hz). These experiments (see Fig 5) show that, although the percentage of inhibition at the steady-state was not significantly different in all three conditions ( Fig 5A-5C), the time course of inhibition was significantly faster in experiments done at 1 Hz (time for 50% inhibition~5 s, Fig 5D), compared to slower frequencies: 0.2 Hz (~15 s) and 0.033 Hz (~40 s). These data, support further statedependent inhibition of Cav3.2 channels by Linalool.

Inhibition of Ca V 3.2 calcium channels by Rosmarinus officinalis and Rosmarinic acid
Next, we studied the effects of Rosmarinus officinalis (RO) and its active principle Rosmarinic acid (RA), a caffeic acid ester compound, on the modulation of Ca V 3.2 channels. The Ca V 3.  Effect of Rosmarinus officinalis and Rosmarinic acid on Ca V 3.2 channel activation and inactivation Similar to that described for Lavender compounds, we then studied the effect of RO and RA on Ca V 3.2 activation and inactivation properties. Analysis of current traces and I-V curves revealed that RO (50 μg/ml) inhibited Ca V 3.2 currents at all tested potentials without changing the steady-state activation properties (V 0.5Control = -52.8 ± 0.4 mV, V 0.5RO = -53.5 ± 0.5 mV, n = 7, Fig 7A). Similar results were obtained after treatment with 50 μM RA (V 0.5Control = -51.8 ± 0.6 mV, V 0.5RA = -50.6 ± 1.3 mV, n = 6, p = 0.4, Fig 7C). To further elucidate the blocking mechanisms of RO and RA, steady-state inactivation was determined in the absence and presence of these natural substances. These experiments showed that treatment with 50 μg/ml RO both reduced the maximal conductance of Ca V 3.2 channels and negatively shifted the midpoint of voltage-dependence of inactivation for Ca V 3.2 towards negative potential (V 0.5Control = -73.7 ± 0.9 mV, V 0.5RO = -77.8 ± 1.1 mV, n = 7, p<0.05, Fig 7B). The application of 50 μM (~18 μg/ml) of RA induced a shift towards more negative membrane potentials (V 0.5Control = -77.4 ± 0.8 mV, V 0.5RA = -82.4 ± 1.2 mV, n = 6, p<0.01, Fig 7D). In addition, activation and inactivation kinetics were unchanged after either RO or RA treatment of Ca V 3.2 channels.

Lavandula steochas and Rosmarinus officinalis preferentially bind to the inactivated state of T-type calcium channels
A growing body of reports suggested that TTCC blockers bind to / stabilize the inactivated state of these channels [38,41,42]. Indeed, the negative shift induced by Lavender and Rosemary in the channel availability suggests that these natural compounds preferentially bind to the inactivated state of Ca V 3.2 channels, thus shifting the equilibrium away from states from which channels can open [43]. To evaluate further whether Lavender and Rosemary compounds bind to the inactivated state of TTCCs, we have measured the inhibition of Ca V 3.2 currents by LS and RS at HPs -100 and -80 mV (Fig 8, upper graphs). If the effects of LS and RO on channel inactivation would contribute significantly to the inhibition of the Ca V 3 channels, then applying LS and RO to cells voltage-clamped at potentials significantly more negative than -80 mV would produce less current inhibition. As expected, the inhibition by LS (20 μg/ ml) was significantly more pronounced when cells were held at HP -80 mV (70 ± 4.4%, n = 6) than at HP -100 mV (51 ± 5.3%, n = 6, p<0.05). Similar data were obtained after the application of RO (20 μg/ml) with 41.3 ± 1.7% of inhibition at HP -80 mV and 23.4 ± 1.7% inhibition at HP -100 mV (n = 6 p<0.01). The efficacy of washout was examined for the two extracts at HPs -100 and -80 mV (Fig 8, lower graphs) and, conversely, washout appeared significantly more efficient at HP -100 mV (LS Washout = 91.6 ± 1.8%, RO Washout = 87.7 ± 3.0%) than at HP -80 mV (LS Washout = 32.6 ± 3.1%), RO Washout = 51.7 ± 2.5%, p<0.001). Taken together, the results suggest that these natural compounds preferentially bind to, and stabilize, Ca V 3.2 channels in the inactivated state.

Discussion
In and Citrullus colocynthis, we have identified that Lavender and Rosemary compounds could significantly inhibit the Ca V 3.2 TTCCs in a concentration-dependent manner. Importantly, Lavender and Rosemary are widely used medicinal plants. Second, our results provide evidence that their active principles, Linalool and Rosmarinic acid respectively, also inhibit Ca V 3.2 channels. Third, we report that these compounds induce a negative shift in the steadystate inactivation properties and we show that their inhibitory effect on Ca V 3.2 channels is significantly enhanced in the range of physiological membrane potential (HP -80 mV), compared to more negative potential (HP -100 mV). Taken together, our findings support a pharmacological modulation of TTCCs by Lavender and Rosemary and we suggest that TTCC inhibition by these natural components may contribute to the neuroprotective and anticonvulsant activities of these medicinal plants.
Lavender (Lavandula stoechas, Lavandula angustifolia Miller) and Rosemary (Rosmarinus officinalis) all inhibit the amplitude of Ca V 3.2 current in a dose-dependent manner. The IC 50 values of L. stoechas, L. angustifolia Miller and R. officinalis were estimated to be 16.9, 34.1 and 53.5 μg/ml respectively, suggesting that Ca V 3.2 channels are more sensitive to Lavandula species and in particular to Lavandula stoechas. Furthermore, Lavender and Rosemary also inhibit the other TTCC isoforms, Ca V 3.1 and Ca V 3.3. The IC 50 values for LA Miller inhibition were 26.1 ± 4.8 μg/ml (n = 7) for Ca V 3.1 and 86.2 ± 18.1 μg/ml for Ca V 3.3 (n = 7). Interestingly, the percentage of inhibition obtained after treatment of the various TTCCs with 10 μg/ml of LA Miller (31% for Ca V 3.1, 16% for Ca V 3.2 an 10% for Ca V 3.3) is similar to that obtained after treatment of HVA, P/Q-type calcium channels with Silexan (25%), a patented active substance produced from L. angustifolia flowers by steam distillation and consisting of the main active constituents linalool and linalyl acetate [10]. Altogether, our data extend previous electrophysiological studies describing the effect of Lavender and its active principle Linalool on other voltage-gated calcium channels [10,44]. Importantly, we report for the first time to our knowledge, inhibition of voltage-gated calcium channels, in particular TTCCs, by Rosmarinus officinalis and its active principle Rosmarinic acid.
Linalool is a monoterpene compound reported to be the major component of Lavender essential oil. It has been reported to trigger glutamate activation in response to NMDA receptors modulation in the cerebral cortex [45] and reduces acetylcholine release at mouse neuromuscular junction by modifying nicotinic receptors kinetics [46], suggesting possible pathways in sedative and anticonvulsant effects in mice [47,48]. Other studies to investigate the molecular mechanisms associated with linalool therapeutic use revealed that Linalool could interact with voltage-gated channels, in particular voltage-gated calcium channels [10,44]. Narusuye et al.   after treatment with linalool and linalyl acetate concentrations (1 μM), suggesting potent anxiolytic properties of linalool via modulation of voltage-dependent calcium channels [10]. Our electrophysiological study confirms the inhibition of voltage-gated calcium channels, specifically TTCCs, by Linalool. Linalool attenuates Ca V 3.2 currents in a dose-dependent manner. The IC 50 for Linalool inhibition of TTCCs estimated to be 84 μM is found to be lower than the IC 50 obtained for the inhibition of other ionic channels in different preparations. As an example, the IC 50 values of Linalool blockade for the voltage-gated sodium is estimated to be around 560 μM [49], suggesting that Linalool is more potent in inhibiting Ca V 3.2 channels.
Rosmarinic acid has been shown to exert neuroprotective effect against antioxidative stress and excitotoxicity and to possess anxiolytic/antidepressive-like effects [20, 50]. The mechanism by which RA exerts its anti-inflammatory effects is not well understood, although it has been shown that RA inhibits lipoxygenase [51] and cyclooxygenase activity [52], block complement activation [53] and T-cell antigen receptor (TCR)-mediated signaling [54]. Whether Rosmarinic acid could modulate ion channels, especially voltage-gated channels, was currently Lavender and Rosemary extracts modulate T-type Ca 2+ channels unknown. Our study therefore reveals that TTCCs, are inhibited by Rosmarinus officinalis and Rosmarinic acid in a dose-and voltage-dependent fashion. Consequently, TTCCs may therefore represent a novel molecular target for Rosmarinic acid, although further experiments are needed to characterize the efficacy of Rosmarinic acid to possibly modulate other ion channels.
Inhibition of TTCCs is highly dependent on their inactivation state. Analysis of the biophysical properties of Ca V 3.2 channels before and after Lavender (Lavandula stoechas, Lavandula angustifolia Miller and Linalool) and Rosemary (Rosmarinus officinalis and Rosmarinic acid) treatments showed that these natural compounds not only reduced the maximal conductance of Ca V 3.2 channels but also shifted the steady-state inactivation properties towards more negative membrane potentials without having effect on the activation properties. Our study describes that inhibition of Ca V 3.2 channels by Lavender and Rosemary was significantly enhanced for HP -80 mV, compared to HP -100 mV (Fig 7). Indeed, the blocking effect was more efficient at a HP mimicking resting's membrane potential, at which a large fraction of TTCCs are inactivated [42,55]. This suggests an interesting mechanism by which Lavender and Rosemary could attenuate the cell excitability by decreasing intracellular calcium concentration and inducing sedative and/or anticonvulsant-like effects, as well as other various therapeutic effects such as neuroprotective properties. Our results showing that these compounds negatively shift the inactivation state suggest that these natural compounds interact with inactivated TTCCs and stabilize them in the inactivated state. This is reminiscent to that reported for phenylalkylamines and dihydropyridines that bind preferentially to the inactivated state of L-type calcium channels (HVA), conferring tissue-selectivity of these drugs that are useful as antihypertensive and antiarrhythmics treatments [56,57].
Compounds selective on TTCCs could have unexpected therapeutical utility, particularly to treat the various disease states in which TTCCs are up-regulated. For instance, up-regulation of Ca V 3.2 channels was observed in both cardiac myocytes and chromaffin cells maintained under chronic hypoxic conditions [58,59]. Cav3.2 channel overexpression was also found associated to neuroendocrine differentiation of prostate cancer cells [60]. Importantly, TTCCs represent novel interesting molecular targets for pain and epilepsy [31, 33, 43, [61][62][63]. Inhibition of TTCCs has been reported to play an important role in the therapeutic action of many drugs [64]. For example, Gomora et al. confirmed the hypothesis that the blockade of TTCCs may underlie the therapeutic usefulness of succinimide antiepileptics [65]. In the same context, Tringham and coworkers have identified two high affinity TTCC blockers that were able to attenuate burst firing of thalamic reticular nucleus neurons in the Genetic Absence Epilepsy Rats from Strasbourg (GAERS) [66]. Blockade of TTCCs is suggested to be useful in a wide variety of neurological disorders such as neuropathic pain [67,68]. Indeed, Jagodic and collaborators have demonstrated that TTCCs are significantly upregulated in small dorsal root ganglion (DRG) during chronic constriction injury (CCI)-induced neuropathy [69]. Therefore, the inhibition of TTCCs by NNC 55-0396, a selective TTCC inhibitor [70] was suggested to be useful in decreasing pre-and postsynaptic transmission and the neuronal activity in anterior cingulate cortex after a CCI leading to the attenuation of neuropathic pain [71]. ABT-639 is a peripherally acting TTCC blocker that selectively inhibits TTCCs in a dose-dependent manner. In preclinical studies, oral administration of ABT-639 was reported to alleviate nociceptive and neuropathic pain in rat models [72]. However, phase 2 clinical studies using microneurography, a relevant technique that assesses abnormal spontaneous activity in C-nociceptors as a marker for spontaneous pain, revealed that administration of ABT-639 100 mg twice daily did not reduce neuropathic pain in diabetic patients [73,74]. Interestingly, Z944, a potent selective blocker of TTCCs (50-160 nM) was shown to be effective in reducing pain in preclinical models as well as in human patients [75].
Other studies have also suggested TTCCs as interesting molecular targets for natural compounds. Eugenol, a local analgesic used in clinical dentistry that naturally present in cloves (Syzygium aromaticum) modulates TTCCs in a dose-dependent fashion with IC 50 of 500 μM. The depressant effect of Eugenol on TTCCs was suggested to inhibit action potentials and the neuronal conduction of sensory signals in TG neurons leading to eugenol pain-relieving action [38]. Furthermore, Ross et al. have demonstrated that Δ9-tetrahydrocannabinol and cannabidiol, the most prevalent biologically active constituents of Cannabis sativa, inhibit recombinant as well as native TTCCs [39]. Interestingly, Cannabidiol is currently under development as an antiepileptic drug [76]. It is likely that attenuation of TTCC conductance causes the decrease in neurotransmitter release mediated by these compounds contributing to the wellknown psychoactive actions of cannabinoids, as well as the anti-nociceptive and anticonvulsant properties [39, [76][77][78].

Conclusion
Our data show that Lavender and Rosemary extracts efficiently inhibit TTCCs by preferentially binding to inactivated channels. Altogether, this study demonstrates that TTCCs represent a novel molecular target for Lavender and Rosemary likely to be involved in some of the Mediterranean medicinal plants' therapeutic use.