Valepotriates From the Roots and Rhizomes of Valeriana jatamansi Jones as Novel N-Type Calcium Channel Antagonists

The roots and rhizomes of Valeriana jatamansi have long been used as folk medicine in Asia and usually named as “Zhizhuxiang” in Chinese for the treatment of abdominal distention and pain. However, its active ingredients and molecular targets for treatment of abdominal pain remain unrevealed. Inhibitors of Cav2.2 N-type voltage-gated calcium channels (VGCCs) are actively sought after for their potential in treating pain, especially chronic pain. As far as we know, the method used for seeking analgesic active ingredient from plant material has rarely been reported. The analgesic potentials of the EtOH extract (0.01 mg/ml) of the roots and rhizomes of V. jatamansi and its EtOAc, n-BuOH and H2O soluble parts (0.01 mg/ml, respectively) were tested herein on Cav2.2, using whole-oocyte recordings in vitro by tow-electrode voltage clamp. The results indicated that the EtOAc-soluble part exhibited the most potent inhibition of Cav2.2 peak current (20 mv). The EtOAc-soluble part was then subjected to silica gel column chromatography (CC) and giving 9 fractions. Phytochemical studies were carried out by repeated CC and extensive spectroscopic analyses after the fraction (0.01 mg/ml) was identified to be active and got seventeen compounds (1–17). All isolates were then sent for further bioactive verification (1 and 3 at concentration of 10 μM, others at 30 μM). In addition, the selectivity of the active compounds 1 and 3 were tested on various ion channels including Cav1.2, Cav2.1 and Cav3.1 VGCCs and Kv1.2, Kv2.1, Kv3.1 and BK potassium channels. The results indicated that compound 1 and 3 (an abundant compound) inhibited Cav2.2 with an EC50 of 3.3 and 4.8 μM, respectively, and had weaker or no effect on Cav1.2, Cav2.1 and Cav3.1 VGCCs and Kv1.2, Kv2.1, Kv3.1 and BK potassium channels. Compounds 1 and 3 appear to act as allosteric modulators rather than pore blockers of Cav2.2, which may play crucial role in attenuating nociception. The results of present research indicated that the ethnopharmacological utilization of V. jatamansi for relieving the abdominal distention and pain may mediate through Cav2.2 channel. Our work is the first demonstration of inhibition of Cav2.2 by iridoids, which may provide a fresh source for finding new analgesics.

The roots and rhizomes of Valeriana jatamansi have long been used as folk medicine in Asia and usually named as "Zhizhuxiang" in Chinese for the treatment of abdominal distention and pain. However, its active ingredients and molecular targets for treatment of abdominal pain remain unrevealed. Inhibitors of Ca v 2.2 N-type voltage-gated calcium channels (VGCCs) are actively sought after for their potential in treating pain, especially chronic pain. As far as we know, the method used for seeking analgesic active ingredient from plant material has rarely been reported. The analgesic potentials of the EtOH extract (0.01 mg/ml) of the roots and rhizomes of V. jatamansi and its EtOAc, n-BuOH and H 2 O soluble parts (0.01 mg/ml, respectively) were tested herein on Ca v 2.2, using whole-oocyte recordings in vitro by tow-electrode voltage clamp. The results indicated that the EtOAc-soluble part exhibited the most potent inhibition of Ca v 2.2 peak current (20 mv). The EtOAc-soluble part was then subjected to silica gel column chromatography (CC) and giving 9 fractions. Phytochemical studies were carried out by repeated CC and extensive spectroscopic analyses after the fraction (0.01 mg/ml) was identified to be active and got seventeen compounds (1-17). All isolates were then sent for further bioactive verification (1 and 3 at concentration of 10 µM, others at 30 µM). In addition, the selectivity of the active compounds 1 and 3 were tested on various ion channels including Ca v 1.2, Ca v 2.1 and Ca v 3.1 VGCCs and Kv1.2, Kv2.1, Kv3.1 and BK potassium channels. The results indicated that compound 1 and 3 (an abundant compound) inhibited Ca v 2.2 with an EC 50 of 3.3 and 4.8 µM, respectively, and had weaker or no effect on Ca v 1.2, Ca v 2.1 and Ca v 3.1 VGCCs and Kv1.2, Kv2.1, Kv3.1 and BK potassium channels. Compounds 1 and 3 appear to act as allosteric modulators rather than pore blockers of Ca v 2.2, which may play crucial role in attenuating nociception. The results of present research indicated that the INTRODUCTION N-type calcium channels (Ca v 2.2) are highly distributed at nerve terminals and on cell bodies of dorsal root ganglia (DRG) neurons, where they regulate the release of pain-related neuropeptides. Pharmacology and genetic studies showed that Ca v 2.2 inhibiting activity was an effective way to manage or alleviate symptoms of inflammatory, chronic and neuropathic pain (Kerr et al., 1988;Gohil et al., 1994;Saegusa et al., 2001Saegusa et al., , 2002McGivern and McDonough, 2004;McGivern, 2006;Altier et al., 2007;Trang et al., 2015). Ca v 2.2 antagonists have therefore long been considered as potential analgesics (Hazelhoff et al., 1982;Mathela et al., 2005;La et al., 2008;Yamamoto and Takahara, 2009;Pajouhesh et al., 2010;Schmidtkco et al., 2010;Swensen et al., 2014). Indeed, a synthetic form of ω-conotoxin MVIIA, a peptide toxin isolated from the marine cone snail, genus Conus, that potently inhibits Ca v 2.2, is in clinical use (marked as ziconotide) to treat severe and chronic pain (Schmidtkco et al., 2010). However, ziconotide can only be delivered intrathecally and even so has numerous serious adverse effects. Therefore, it is clearly desirable for development of new Ca v 2.2 antagonists.
Valeriana jatamansi Jones distributed widely in China and some other Asian countries (Mathela et al., 2005). The roots and rhizomes of this plant is a well-known herbal medicine "Zhizhuxiang" in the Chinese Pharmacopoeia for treating abdominal pain (Editorial Board of Chinese Pharmacopoeia, 2015). Several clinically used medicines for treating abdominal pain in the Chinese market, such as "Xiangguo Jianxiao Pian, " are prepared by Zhizhuxiang as a main component. However, the chemical basis and molecular targets of V. jatamansi Jones for the treatment of abdominal pain remain unrevealed.
Inspired by the analgesic effect of V. jatamansi Jones, bioassay-guided isolation and characterization of the active constituents of the plant that target Ca v 2.2 were carried out herein. Initially, the ethanol extract and the EtOAc, n-BuOH and H 2 O fractions of the plant were tested for their inhibition of Ca v 2.2. Subsequent function-guided phytochemical studies on the active EtOAc part resulted in the isolation and identification of thirteen iridoids (1-13) and four sesquiterpenes (14-17) (Figure 1), including two new valepotriates, namely jatamanvaltrate T and U (1-2). The inhibitory activities on Ca v 2.2 of all the compounds were tested. Among them, the new compound 1 and the main constituent 3 (8.45 g out of 32.5 kg of plant material) exhibited significant inhibitory effects on Ca v 2.2 with EC 50 values of 3.3 and 4.8 µM, respectively. In addition, compound 1 and 3 showed noticeable selectivity over Ca v 1.2, Ca v 2.1 and Ca v 3.1 VGCCs and Kv1.2, Kv2.1, Kv3.1 and BK potassium channels.

Plant Material
The roots and rhizomes of V. jatamansi were collected at Kunming, Yunnan Province, People's Republic of China, in July 2012, and identified by Prof. En-De Liu, and a voucher specimen (KUN NO. 0864803), has been deposited in State Key Laboratory of Phytochemistry and Plant Resource in West China, Kunming Institute of Botany, The Chinese Academy of Sciences, Kunming, China.

The Preparation of Crude Extract and Fractions
The air-dried roots (450 g) of V. jatamansi were powdered and extracted with 95% ethanol at room temperature (3L × four times, each time 24 h) with occasional shaking. The extracted was filtered and concentrated with a rotary evaporator at 45-50 • C, giving 75.5 g residues (The yield is approximate 16.8%). The ethanol extract (43.62 g) was suspended in distilled (600 ml) water and partitioned successively with EtOAc and n-BuOH (five times, each time 0.5 h), then evaporated on rotary evaporator to afford EtOAc-soluble part (12.73 g) and n-BuOH layer (5.38 g). The H 2 O-soluble part was passed through a macroporous absorbent resin (D-101) with H 2 O and ethanol, then concentrated under reduced pressure to obtain H 2 O layer (0.51 g). The EtOAc part was subjected to silica gel column chromatography eluted with a gradient of petroleum ether/acetone (1:0-0:1, v/v) to afford 9 fractions (Fr.1-Fr.9).

Extraction and Isolation
Based on the observed effect with the EtOAc portion on Ca v 2.2, further extraction and isolation of the EtOAc portion were carried out herein.

Electrophysiology
All experiments were performed at 20-22 • C. Whole-oocyte recordings by performed with two-electrode voltage clamp (TEVC). Electrodes were filled with KCl (3 mM) and had resistances of 0.3-1 M . The bath solution used to record calcium channel currents contained KCl (2 mM), BaCl 2 (1.8 mM), NaOH (50 mM), Ba(OH) 2 (40 mM) and HEPES (5 mM). pH 7.4 was adjusted with methanesulfonic acid and the solution was filtered. L-type (Ca v 1.2), N-type (Ca v 2.2), and P/Q-type (Ca v 2.1) calcium channel currents were evoked from a holding potential of −80 mV by 50-ms depolarizations ranging from −30 to 70 mV in 10-mV increments at 3-s intervals. Currents through T-type (Ca v 3.1) calcium channels were elicited by voltage pulses (50 ms) from −50 to 60 mV with a holding potential of −80 mV in 10-mV increments at 3-s intervals. Kv1.1 currents were recorded from a holding potential of −80 mV by 200-ms depolarizations ranging from −60 to 60 mV in 10-mV increments at 15-s intervals. Kv2.1 and Kv3.1 currents were obtained from a holding potential of −80 mV by 200-ms depolarizations ranging from −40 to 60 mV in 10-mV increments at 15-s intervals. BK channel currents were obtained from a holding potential 0f −50 mV by 60-ms depolarization ranging from 0 to 100 mV in 10-mV increments at 3-s intervals. All the currents were sampled and filtered, respectively.
In whole-cell recordings of HEK 293T cells, pipettes were fabricated from borosilicate glass (World Precision Instruments) using a micropipette puller (P-1000, Sutter Instrument), and were fire-polished to resistances of ∼3 MW. Whole-cell currents were elicited by 20-ms voltage steps from −60 to 80 mV with 10-mV increments, with a holding potential of −80 mV. Currents were amplified by Axopatch 200B and digitized by Digidata 1440A (Molecular Devices). Currents were low-pass filtered at 2 kHz and sampled at 10 kHz. The extracellular solutions contained (in mM) 5 CsCl, 10 BaCl 2 , 140 TEA.Cl, 10 Glucose and 10 HEPES. pH was adjusted to 7.4 with CsOH. The intracellular solutions contained   Figure S23). The EtOAc-soluble part (Supplementary Figure S24), which exhibited a stronger inhibitory activity, was subjected to silica gel column chromatography to produce 9 fractions. All of the fractions (0.01 mg/ml), together with the isolates (1-17) from the fractions, were tested for their inhibitory effects on Ca v 2.2, with a concentration of 10 µM (1 and 3) or 30 µM (2, and 4-17). Compounds 1 and 3 were further tested for their activities on Ca v 1.2, Ca v 2.1 and Ca v 3.1 VGCCs (10 µM) and on Kv1.2, Kv2.1, Kv3.1 and BK potassium channels (30 µM) (Supplementary Figure S25).
In a positive control experiment, the effect on Ca v 2.2 of 100 µM CdCl 2 and 0.2-1.0 µM ω-conotoxin MVIIA was tested by TEVC in Xenopus oocytes and by whole-cell recording in HEK 293T cells.

Data Analysis and Statistics
Data acquisition and analysis of the whole-oocyte recordings were carried out by using pClamp 10 (Molecular Devices Corporation, San Jose, CA, United States). Data fitting and statistical analyses were performed by PRISM 5.0 (GraphPad Software Inc., San Diego, CA, United States). EC 50 values and Hill slopes were determined by fitting the data points to a sigmoidal dose-response equation (Y = Min + (Max-Min)/(1+10ˆ((LogEC50-X) * n))), where Y is % Inhibition, X is the concentration of the compounds, Min is minimal inhibition, Max is maximum inhibition, and n is the Hill coefficient. All data were presented as mean ± SEM, and statistical analysis was performed using Student's t-test. P-values of < 0.05 were considered as significant, and levels of significance were marked by asterisks ( * * P < 0.01, * P < 0.05).

Functional Characterization of Isolated Compounds
The traditional use of the roots and rhizomes of V. jatamansi to treat abdominal pain led us to assume that some of their effects may be mediated through inhibition of Ca v 2.2. Therefore, we carried out a systematic functional assay of the various extracts, fractions and isolated compounds obtained from the herbal medicine. Initially, we found that the ethanol extract and its EtOAc-soluble layer (0.01 mg/ml) showed inhibitory effects of 36 and 49% on Ca v 2.2, respectively, while the n-BuOH and H 2 O layers (0.01 mg/ml) exhibited negligible activities ( Table 2). Of all the fractions from the ethyl acetate extract, fractions 4 and 6 exhibited strongest activities on Ca v 2.2, with 68.6 ± 3.1% and 60.1 ± 4.0% inhibition at 0.01 mg/ml, respectively ( Table 2). Further phytochemical isolation from these two fractions led to the discovery of two new valepotriates, jatamanvaltrate T and U (1-2) and fifteen known compounds (3-17). All of the isolates were evaluated for their inhibitory effects on Ca v 2.2. The new compound 1 (jatamanvaltrate T) and the main constituent 3 showed prominent inhibition of Ca v 2.2 (Table 2 and Figure 3). Other compounds produced no (10-17) or much weaker (2, 4-9) effects on Ca v 2.2, even at a concentration of 30 µM ( Table 2) (Supplementary Figure S26). Based on the bioassay results and the structure of these compounds, we postulate that the diene-type iridoid is a nucleus for the inhibitory activity on Ca v 2.2.
The inhibition of Ca v 2.2 by compounds 1 and 3 was dose-dependent, with an EC 50 of 3.3 µM (n = 5) and 4.8 µM (n = 3), respectively (Figure 3). Interestingly, the inhibition by both compounds was incomplete, plateauing at ∼65% at a near saturation concentration of 100 µM (Figure 3). This result suggests that compounds 1 and 3 act allosterically to modulate Ca v 2.2 gating rather than block channel conduction.
In a positive control experiment, we tested the effect of two known N-type VGCC inhibitors, the divalent cation cadmium (Cd 2+ ), which is a broad pore blocker of VGCCs, and the marine cone snail peptide toxin ω-conotoxin MVIIA, which is an N-type VGCC antagonist. As expected, both molecules inhibited Ca v 2.2 currents in Xenopus oocytes and HEK 293T cells (Supplementary  Figures S27, S28).
In summary, our study shows that valepotriates, one of the main constituents of the V. jatamansi, can inhibit Ca v 2.2 N-type VGCCs. This activity is consistent with the analgesic effect of V. jatamansi in alleviating abdominal distention and pain. Further bioassay-guided search in V. jatamansi may yield more compounds that target N-type VGCCs.

AUTHOR CONTRIBUTIONS
F-WD and H-HJ are the co-first authors responsible for making experiments, consulting literature, and writing article. LY, YG, and C-TZ are responsible for editorial assistance with this manuscript. DY, C-JY, and HL are responsible for participating in plasma sample preparation. JZ took part in the topic selection and the experiment advising. JY, YN, and J-MH are corresponding authors of this article will be responsible for conducting the research and all correspondence with the editorial and accept the consultation of the reader.