Schwarzinicine A inhibits transient receptor potential canonical channels and exhibits overt vasorelaxation effects

Abstract This study investigated the vasorelaxant effects of schwarzinicine A, an alkaloid recently reported from Ficus schwarzii Koord. Regulation of calcium homeostasis in vascular smooth muscle cells (VSMC) is viewed as one of the main mechanisms for controlling blood pressure. L‐type voltage‐gated calcium channel (VGCC) blockers are commonly used for controlling hypertension. Recently, the transient receptor potential canonical (TRPC) channels were found in blood vessels of different animal species with evidence of their roles in the regulation of vascular contractility. In this study, we studied the mechanism of actions of schwarzinicine A focusing on its regulation of L‐type VGCC and TRPC channels. Schwarzinicine A exhibited the highest vasorelaxant effect (123.1%) compared to other calcium channel blockers. It also overtly attenuated calcium‐induced contractions of the rat isolated aortae in a calcium‐free environment showing its mechanism to inhibit calcium influx. Fluorometric intracellular calcium recordings confirmed its inhibition of hTRPC3‐, hTRPC4‐, hTRPC5‐ and hTRPC6‐mediated calcium influx into HEK cells with IC50 values of 3, 17, 19 and 7 μM, respectively. The evidence gathered in this study suggests that schwarzinicine A blocks multiple TRPC channels and L‐type VGCC to exert a significant vascular relaxation response.

thus enabling further biological evaluation to be carried out (Lee et al., 2021).
The mainstay for treatment of hypertension is the use of L-type voltage-gated calcium channel (VGCC) blockers. Regulation of calcium homeostasis in vascular smooth muscle cells (VSMC) is viewed as one of the main mechanisms for controlling blood pressure. An important discovery of the transient receptor potential canonical (TRPC) family of proteins is that they were found to be present and functional in smooth muscle cells and endothelial cells of different animal species (Earley & Brayden, 2015).
Growing literature demonstrated that the molecular mechanisms of TRPC channels in modulating Ca 2+ dynamics are complex in nature.
For VSMCs contraction, a number of studies have suggested that TRPC-mediated Ca 2+ influx is through mechanisms involving direct Ca 2+ transport via the pore region of these channels. This mechanism is proposed to be via interaction with L-type VGCCs by altering the membrane potential or recruiting signalling messengers (Earley & Brayden, 2015;Kiselyov et al., 1998). Our research question was focusing on the direct inhibition of Ca 2+ entry during TRPC channels activation which leads to vasorelaxation. On the other hand, the expression of endothelial TRPC channels promotes vasorelaxation through a different pathwayassociation with BK Ca channel which induces membrane hyperpolarisation in VSMCs (Earley & Brayden, 2015;Kochukov, Balasubramanian, Noel, & Marrelli, 2012;Kwan et al., 2009;Martín-B ornez, Galeano-Otero, Del Toro, & Smani, 2020). It is suggested that these downstream relaxation effects following Ca 2+ influx prevent prolonged myogenic contraction.
The importance of these channels in the cardiovascular system has been clearly demonstrated in the last decade (Earley & Brayden, 2015;Wang et al., 2020). Changes to these TRPC channels have been linked to vascular pathologies such as hypertension and pulmonary oedema. The TRPC channels are a group of receptoroperated calcium-permeable non-selective cation channels of the TRP superfamily. They are activated by the phospholipase C (PLC) pathway (Wang et al., 2020) and inhibited by SKF96365 (Harteneck, Klose, & Krautwurst, 2011). Current evidence indicates that TRPC1, TRPC3, TRPC4 and TRPC6 channels are functionally important in regulating vascular contractility via the regulation of calcium dynamics (Kunichika et al., 2004;Liu, Xiong, & Zhu, 2014). TRPC1, 3, 4, 5 and 6 protein expressions were detected in various vascular beds, including cerebral and coronary arteries (Yip et al., 2004), Sprague-Dawley (SD) rat resistance arteries (Welsh, Morielli, Nelson, & Brayden, 2002), thoracic aorta (Facemire, Mohler, & Arendshorst, 2004) and embryonic rat thoracic aortic smooth muscle cells (A7r5 cells) (Jung, Strotmann, F I G U R E 1 (a) Chemical structure of schwarzinicine A, which was obtained naturally as a scalemic mixture (4:1 ratio of (+)/(À)). (b) Cumulative concentration-response curve of schwarzinicine A compared to vehicle control (DMSO), Pyr3, HC-070, nifedipine and SKF96365. Tissue relaxations were expressed as the percentage of phenylephrine-induced contraction. The data represent the mean values ± SEM of n number of animals. (c) Representative trace recording of schwarzinicine A in rat isolated aorta Schultz, & Plant, 2002). Interestingly, the expression of TRPC channels was found to increase in VSMCs isolated from hypertensive animals and patients (Chen et al., 2010;Dietrich et al., 2005). The development of the functional characterisation and pharmacology of these TRPC channels has been slow until recently, when a number of novel selective small molecules have been discovered (Wang et al., 2020).
Following these initial observations, the present study was designed to further investigate schwarzinicine A's potential role as a calcium channel modulator and to characterise its pharmacological effects using the rat isolated aorta as a native system. We hypothesised that schwarzinicine A modulates calcium dynamics across the cell membrane with the involvement of TRPC channels and possibly L-type VGCCs. were used in isometric tension recordings. SD rat was used as a model as it is an outbred stock which abundantly expresses TRPC 3, 4, 5 and 6 mRNA and protein in its aorta (Facemire et al., 2004). The aorta was excised in 2 mm ring and transferred into fresh Krebs-Ringer bicarbonate solution. The organ bath containing fresh isolated aortic tissues was assembled and tested for tissue viability as described in Loong, Tan, Lim, Mbaki, and Ting (2015).

| Effects of schwarzinicine A and other calcium channel inhibitors against rat isolated aortae
Endothelium-intact aorta was pre-constricted with 0.1 μM phenylephrine. The concentration of phenylephrine used elicited at least 70% submaximal of phenylephrine-induced tissue contraction. Once a stable tone was established, cumulative concentration response curves (CRCs) of concentration range from 0.1 nM to 30 μM were generated for schwarzinicine A. Each concentration was added every 7 min or when the tissue response had plateaued.

| Effects of schwarzinicine A on calcium chloride (CaCl 2 )-induced contraction in calcium-free Krebs solution
Endothelium-intact aorta was tested for tissue viability with 60 mM KCl twice in calcium-containing Krebs solution (Loong et al., 2015).
Then, the aorta was equilibrated in calcium-free Krebs solution for 30 min before incubating the tissues further with different concentrations of schwarzinicine A (3, 10, 30 μM) for another 30 min. These three concentrations were chosen based on the earlier generated CRC (see Figure 1b). Following a single application of 60 mM KCl to depolarise VGCC in calcium-free Krebs solution, a CRC to CaCl 2 (1 μM to 3 mM) was carried out.
The cells were grown in a humidified incubator with 5% CO 2 at 37 C overnight. The next day, cells were transfected using human TRPC6 cloned into pcDNA3 (stock 1.15 μg/μl) and jetPRIME ® transfection reagent (VWR, Lutterworth, UK) in a ratio of 1:3. TRPC6 cloned into pcDNA3 was obtained from Dr Melanie Ludlow (Beech Lab, Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds). Transfection mix was prepared by diluting 2 μg of hTRPC6 in 200 μl jetPRIME ® buffer and 6 μl of jetPRIME ® transfection reagent.
After a brief vortex, the mix was incubated at room temperature for 10 min before being added to the cells. The cells were incubated for 24 hours before being plated.
Schwarzinicine A and other DMSO-dissolved drugs were diluted ten10fold with descending DMSO content from 50% (10 mM), 25% (1 mM) to purified water (100 μM and below). The final concentration of DMSO did not exceed 0.15%, while methanol did not

| Data analysis
All data were expressed as mean ± SEM of n number of animals in isometric tension recording experiment. The relaxation responses were expressed as percentage inhibition of the respective contractile agonist-induced contraction. Maximum response (E max ) and EC 50 were obtained from non-linear regression fit curve. pEC 50 (negative logarithm of EC 50 ) values were also determined.
In the calcium imaging experiments, n is the number of DRG cells used. Neurones that exhibited an increase of less than 0.1 from basal ratio after the superfusion with KCl, were treated as non-viable and excluded from data collection. The mean 340/380 nm ratio represen- Statistical analyses were performed on experiments with at least n = 5. The study and analysis were carried out randomised, blinded, of equal group sizes unless stated in the respective results. Unpaired Student's t-test was used to compare between vehicle control and treatment groups. In multi-group analysis, one-way ANOVA followed by Dunnett's multiple comparisons test was used only when F in ANOVA was significant (p < .05). Data were analysed using GraphPad Prism Version 8.2.1 (La Jolla California USA). A probability of less than .05 (p < .05) was considered statistically significant, indicated by asterisks: *p < .05; **p < .01; ***p < .001, ****p < .0001.

| Schwarzinicine A exhibits overt vasorelaxation effect
Relaxation magnitude elicited by schwarzinicine A is significantly higher than all the calcium channel blockers tested in this study (see relative E max in Table 1). Vehicle control only evoked minimal tissue relaxation (Table 1). Interestingly, Schwarzinicine A has similar pEC 50 to the TRPC channel inhibitors SKF96365, Pyr3 and HC-070 (i.e., EC 50 values differ by less than an order of magnitude; Table 1) but is significantly less potent when compared with nifedipine (Table 1, Figure 1b). A representative trace of aortic relaxation elicited by schwarzinicine A is depicted in Figure 1c. 3.2 | Schwarzinicine A attenuates the contractile response to CaCl 2 in calcium-free Krebs solution Effects of schwarzinicine A on calcium CRC were tested against endothelium-intact aortic tissues. Aortic tissues treated with vehicle (0.165% DMSO v/v; E max : 92.4 ± 17.9%) and nifedipine served as the controls. In calcium-free Krebs solution, nifedipine totally abolished the contractile responses upon re-addition of CaCl 2 to the system (E max : À10.3 ± 5.7%; p < .0001 vs. vehicle control; Figure 2). The contractile response to CaCl 2 following pre-treatment with schwarzinicine A was concentration-dependent, with total abolishment observed in the presence of 30 μM schwarzinicine A (E max 3 μM: 72.4 ± 12.2%; p = .4857; 10 μM: 25.7 ± 5.7%; p = .0006; 30 μM: 2.2 ± 4.8%; p < .0001; vs. vehicle control). At 10 μM, nifedipine was much better at suppressing the calcium-induced responses when compared to schwarzinicine A.

| Schwarzinicine A attenuates KCl peak response in rat DRG neurones
Following the observation that schwarzinicine A attenuated the calcium-induced contractions in the rat aorta, calcium imaging experiments in DRG neurones were carried out to verify the role of schwarzinicine A in modulating calcium dynamics. In intracellular calcium imaging experiments, high potassium (30 mM KCl) produced a sharp peak in the 340/380 nm ratio with an average increment of 0.4 from basal level. Washing with superfusion buffer reversed this response from a depolarising state to basal level. The response of the second KCl addition after incubation with vehicle DMSO, serving as a HC-070 90.5 ± 1.9* (p = .0409) 0.7 6.9 ± 0.5 (p = .4426)

T A B L E 1 Vasorelaxation effects of schwarzinicine A and calcium channel blockers
Note: E max of each calcium channel blocker was expressed relative to the corresponding E max of schwarzinicine A and presented as the relative E max . Tissue relaxations were expressed as the percentage of phenylephrine-induced contraction. The data represent the mean values ± SEM of n number of animals. [One-way ANOVA followed by Dunnett's multiple comparison test, ****(p < 0.0001), *** (p < .001), **(p < .01) and *(p < .05) vs. schwarzinicine A.] control, was comparable to the initial KCl response (81.8 ± 0.8%, During the four-minute infusion of schwarzinicine A (30 μM), a slight decrease of 340/380 nm ratio was observed (refer to the circle in red in Figure 3b). Due to the preceding combination with schwarzinicine A, the second KCl peak response was significantly reduced to 55.2 ± 1.3% compared to that of the control KCl response (p < .0001, n = 183, Figure 3b,d).
The presence of L-type calcium channels upon depolarisation with high KCl was investigated using nifedipine. Nifedipine marginally attenuated the KCl peak response to 76.7 ± 2.3% compared to control (p < .05, Figure 3c,d). During the infusion with nifedipine, a gradual and slow basal calcium elevation, indicated by the 340/380 nm ratio, was observed (refer to the circle in red in Figure 3c). Interestingly, this observation was in contrast to what schwarzinicine A exhibited where the latter suppressed calcium movement into the cells. Subsequently, a further sharp peak increase was followed with the addition of combined KCl and nifedipine (Figure 3c). In this assay, schwarzinicine A exhibited a larger effect in suppressing calcium dynamics compared to nifedipine in the neuronal cells.  Figure 4a). At the highest concentration tested (100 μM), schwarzinicine A completely inhibited the influx of calcium through hTRPC6 in response to OAG. Our data reveal that the concentration required for half-maximal inhibition (IC 50 ) was 7 μM (pIC 50 5.1), with
Effects of schwarzinicine A on TRPC4 and TRPC5 channel activity were tested on hTRPC4 or hTRPC5-overexpressing (Tet+) HEK T-REx™ cells, using similar protocols to those described for hTRPC3 and hTRPC6. Potent and specific small-molecule activators of TRPC4 and 5 channels are available (e.g., the natural product (À)-englerin A) (Akbulut et al., 2015). However, in order to compare effects on vascular relaxation and TRPC channel modulation by schwarzinicine A, we favoured the use of the more physiologically-relevant TRPC4/5 activator S1P in these assays. S1P is an endogenous signalling molecule that has been shown to activate TRPC5 channels in human VSMCs . As S1P does not give calcium responses in HEK cells lacking TRPC4/5 expression , it is suitable for the profiling of potential TRPC4/5 inhibitors. did not fully inhibit responses to S1P. These data show that as well as inhibiting TRPC3/6 channels, schwarzinicine A can also inhibit both TRPC4 and TRPC5 channels.

| DISCUSSIONS
In an earlier report from this laboratory (Krishnan et al., 2020), we have shown the overt vasorelaxation effect of schwarzinicine A. This follow-up study was designed to determine the mechanism of action of its vasorelaxant effect with a specific focus on the mobilisation of intracellular calcium in VSMC. In some earlier reports, activation of TRPC channels has been proposed to induce endothelium-dependent vasorelaxation via activation of endothelial nitric oxide synthase (eNOS), production of nitric oxide (NO) or downstream hyperpolarisation through calcium-activated potassium (K Ca ) channels (Earley & Brayden, 2015;Kochukov et al., 2012;Kwan et al., 2009).
However, our findings excluded these vasorelaxant mechanisms of schwarzinicine A, including its involvement in the regulation of endothelium, NO, nucleotide second messengers and potassium channels.
The plausible effects of schwarzinicine A affecting the adrenoceptors have also been ruled out (see Supporting Information).
Schwarzinicine A exhibited the highest efficacy as a vasorelaxant amongst all compounds tested. It is interesting to note that its relaxation magnitude is significantly higher than that of nifedipine but with a lower potency recorded. In VSMC, L-type VGCC is widely known as the major calcium-permeable channel in the regulation of smooth muscle contraction. In addition, the expression of TRPC channels in reversed phenylephrine-induced tissue contraction, suggesting that it may inhibit downstream L-type VGCC and TRPC channels as these channels are involved in the vasocontraction induced by phenylephrine (Fransen et al., 2015).
When comparing the vasorelaxation profile of schwarzinicine A with SKF96365, Pyr3 and HC-070 (Just et al., 2018), all compounds share similar potency (pEC 50 values) in the micromolar range. The rank of vasorelaxant response (E max ) for these compounds is: schwarzinicine A > HC-070 > SKF96365 > Pyr3. Pyr3 has been reported as a selective TRPC3 channel inhibitor (IC 50 0.8 μM). HC-070 is a selective, nanomolar inhibitor of TRPC4/5 channels (IC 50 values of 46 nM for TRPC4 and 9.3 nM for TRPC5), but also inhibits TRPC3 channels at micromolar concentrations (IC 50 1.0 uM) (Just et al., 2018;Kiyonaka et al., 2009). SKF96365 has been characterised as a non-selective TRPC channel inhibitor (IC 50 : 5-25 μM) (Inoue et al., 2001;Merritt et al., 1990;Okada et al., 1998). In our experiments, the EC 50 value obtained for Pyr3 was similar to other studies, suggesting that Pyr3-induced relaxation in the rat aorta was mediated by blocking the TRPC3 channel expressed on the smooth muscle of the vasculature. On the other hand, the EC 50 obtained for HC-070 in the organ bath experiment is similar to its reported IC 50 for TRPC3/4/5 channels (Just et al., 2018). This suggests that the vasorelaxation effect of HC-070 is through inhibition of all these three channels. The role of TRPC3 channel appears to be dominant in vascular relaxation, compared to the role of TRPC4 and 5 channels, as no relaxation effect was observed at nanomolar concentrations of HC-070. However, we postulate that inhibition of TRPC3 alone could produce a partial but apparent vasorelaxation response. Non-selective inhibition of multiple TRPC channels as elicited by SKF96365 may be required to observe a maximal vascular response. Schwarzinicine A appears to be inhibiting multiple TRPC channels and L-type VGCCs, hence giving a large magnitude of relaxation compared to TRPCselective compounds and L-type VGCC blockers.
From the above results, the focus was then moved to verify the role of schwarzinicine A in regulating calcium mobilisation in rat aorta.
Depolarisation of the plasma membrane following depletion of external calcium mimics the condition of the store-operated calcium entry pathway. In a calcium-free environment, schwarzinicine A markedly suppressed the contractile response to the re-addition of CaCl 2 in a concentration-dependent manner, affecting calcium entry in a similar way to nifedipine, a potent L-type VGCC inhibitor. In rat DRG cells, KCl-stimulated VGCC via membrane depolarisation appears that Ltype VGCC-insensitive calcium influx seems to be dominant compared to the L-type VGCC-sensitive route. This is suggested as the incubation with nifedipine did not suppress KCl-induced contraction as much as schwarzinicine A. Most importantly, this finding proposed the notion that the inhibitory action of schwarzinicine A in calcium dynamics is not entirely the same as that of nifedipine but could involve a combination of different calcium channels. These observations indicate that schwarzinicine A affected calcium influx via a mechanism different from a pure L-type calcium channel blocker like nifedipine. However the evidence gathered did not rule out its inhibition on L-type VGCCs as KCl-induced depolarisation involved L-type VGCCs and store-operated TRP (Lopez et al., 2020;Ratz & Berg, 2006 information on their selectivity against other TRPC subtypes (Motoyama et al., 2018;Zhou et al., 2018). TRPC proteins are able to form heterotetrameric channels, particularly among TRPC1/4/5 and TRPC3/6/7, owing to their closely related homologues (Hofmann, Schaefer, Schultz, & Gudermann, 2002). Different TRPC tetramers can have distinct biophysical properties, pharmacology and function (Minard et al., 2018;Wang et al., 2020). Modulation of endogenous TRPC channels is complex and involves complex interplay with endogenous and dietary lipids (Svobodova & Groschner, 2016). Lipid modulation may be indirect; for example, S1P can activate channels (at least in part) through G-protein signalling . However, lipids may also act on TRPC channels directly. For example, DAGs have been shown to activate TRPC3/6/7 channels directly, independently of PKC activation (Beck et al., 2006;Hofmann et al., 1999), potentially via a proposed agonist binding site identified on TRPC6 (Bai et al., 2020). Lysophospholipids such as lysophosphatidylcholine (LPC) can modulate TRPC5 channel directly, potentially through direct interaction with the phospholipid bilayer or at lipid interaction sites of these channels (Flemming et al., 2006;Wright et al., 2020). Homotetramers of TRPC3 or TRPC6 are mostly formed in heterologous overexpression HEK 293 cells, however formation of heteromultimeric channels in vivo with different association patterns may alter their functional properties (Hofmann et al., 2002). TRPC3 was found to be upregulated in TRPC6-knockout mice when unexpected enhancement in vascular reactivity was detected (Dietrich et al., 2005), suggesting their functional roles may overlap to compensate for missing mechanisms. Despite possessing similar pore properties, they have different affinities towards calcium (Boulay et al., 1997;Zhu et al., 1996) which lead to distinct functional roles in regulating vascular smooth muscle contractility and other systems (Dietrich et al., 2005;Klein et al., 2020). Distinction between functions of TRPC3 and TRPC6 was supported by another study which demonstrated that TRPC3 mediates agonist-induced depolarisation in cerebral artery SMC, while TRPC6 plays a role in pressure-induced depolarisation, which potentiates myogenic tone (Reading, Earley, Waldron, Welsh, & Brayden, 2005). Therefore, the effect of schwarzinicine A on heteromeric TRPCs warrants further studies.
Future patch-clamp electrophysiology experiments and/or photoaffinity labelling would also be useful to provide further evidence on the link between schwarzinicine A and TRPC channel inhibition in smooth muscle cells (Bauer et al., 2020). Its high temporal resolution allows determination of drug reversibility, voltage-dependency, and binding site (extracellular or intracellular, or both).
In conclusion, schwarzinicine A as a novel phenylethylamine, has elicited an overt concentration-dependent relaxation response in the rat aorta. It displayed a calcium inhibitory property by blocking calcium entry in both the vascular function and calcium imaging measurement in rat DRG cells. This study also revealed schwarzinicine A as a non-selective TRPC channel blocker. Nevertheless, it is possible that schwarzinicine A inhibits more than one family of TRP channels.
Its ability to evoke marked vasorelaxation responses in aortic smooth muscle tissues could potentially be clinically relevant in cases of dysregulation of calcium-dependent vascular smooth muscle that is mediated through L-type VGCCs and TRPC channels. Finally, schwarzinicine A could be used as a potential candidate to facilitate the development of analogues to investigate the structure-activity relationships of TRPC channels.