Ginger metabolites and metabolite-inspired synthetic products modulate intracellular calcium and relax airway smooth muscle

Abstract

Asthma affects millions of people worldwide and its prevalence is increasing. It is characterized by chronic airway inflammation, airway remodeling, and pathologic bronchoconstriction, and it poses a continuous treatment challenge with very few new therapeutics available. Thus, many asthmatics turn to plant-based complementary products, including ginger, for better symptom control, indicating an unmet need for novel therapies. Previously, we demonstrated that 6-shogaol (6S), the primary bioactive component of ginger, relaxes human airway smooth muscle (hASM) likely by inhibition of phosphodiesterases (PDEs) in the β-adrenergic (cyclic nucleotide PDEs), and muscarinic (phospholipase C, PLC) receptor pathways. However, oral 6S is extensively metabolized and it is unknown if the resulting metabolites remain bioactive. Here, we screened all the known human metabolites of 6S and several metabolite-based synthetic derivatives to better understand their mechanism of action and structure-function relationships. We demonstrate that several metabolites and metabolite-based synthetic derivatives are able to prevent Gq-coupled stimulation of intracellular calcium [Ca²⁺]_i and inositol trisphosphate (IP₃) synthesis by inhibiting PLC, similar to the parent compound 6S. We also show that these compounds prevent recontraction of ASM after β-agonist relaxation likely by inhibiting PDEs. Furthermore, they potentiate isoproterenol-induced relaxation. Importantly, moving beyond cell-based assays, metabolites also retain the functional ability to relax Gq-coupled-contractions in upper (human) and lower (murine) airways. The current study indicates that, although oral ginger may be metabolized rapidly, it retains physiological activity through its metabolites. Moreover, we are able to use naturally occurring metabolites as inspiration to develop novel therapeutics for brochoconstrictive diseases.

INTRODUCTION

Asthma affects over 300 million people worldwide, and its prevalence is increasing in many countries (1). Asthma is characterized by chronic airway inflammation, airway wall remodeling and narrowing, and pathologic bronchoconstriction and bronchial hyperresponsiveness (2, 3) making this disease a continuous treatment challenge (3–5). The presence of postoperative pulmonary complications (PPCs), such as bronchospasm, are frequent in patients with asthma and increase not only the length of ICU stay from 2 to 4.5 days (6) but also mortality from 0.2% to 8.3% (7). The social and economic burden associated with asthma is enormous, with more than 400 thousand deaths in 2016 (8) and billions of dollars lost annually in terms of lost wages and hospitalization expenses in the United States alone (9).

With the introduction of inhaled corticosteroids (ICS) and more specific and long-acting β-agonists, overall asthma mortality had decreased by 63% over a 20-year period (1985–2005), but in the past decade, mortality rates have plateaued (10). Even though strides have been made in curtailing asthma-related deaths, nearly half of asthmatics, due to poor symptom control, turn to complementary and alternative medicine (CAM) (11–14), including ginger (15). Ginger (Zingiber officinale) has been used for centuries and across cultures to treat various ailments (16). The major constituents of fresh ginger rhizome are gingerols, with 6-gingerol, 8-gingerol, and 10-gingerol (6G, 8G, and 10G, respectively) being the most abundant (17). Upon preparation of dried ginger, gingerols rapidly dehydrate to the corresponding shogaols (Fig. 1A). In many reports, 6-shogaol (6S) has been shown to have enhanced biological activities compared with 6G (17). In the past decade, there has been a concerted effort to elucidate the exact effects and mechanisms of action of the various active constituents in plant-based remedies, including ginger that is used in CAM. This is not surprising given that many of our earliest medications, such as morphine, aspirin, and penicillin, are purified and unmodified natural products (18). Even today natural products continue to dominate and often overshadow purely laboratory-based drug discovery, by being able to provide a more diverse and chemically complex architecture (19). For example, from 1940 to 2014 ∼75% of all FDA-approved small molecule anticancer drugs are natural products or natural product-derived (20).

Figure 1.

Ginger constituents, their human metabolites, and synthetic derivatives are grouped according to shared chemical architecture elements. A: a natural ginger phytochemical 6-shogaol (6S) is the primary component of ginger powder after thermal dehydration of 6-gingerol (6G). The other major compounds, 8-shogaols and 10-shogaols (8S and 10S), products of dehydration of 8-gingerols and 10-gingerols, respectively, are also shown. B: the structures of the human metabolome of 6S as potential therapeutics for bronchoconstriction are shown. 6S (gray inset) undergoes various in vivo biotransformations: cysteine conjugation (red), reduction (purple), (thio)etherification (blue), and oxidation (green) reactions. C: additional compounds have been synthesized, such as aspirinate derivatives of 6G (GAS) and 6S (SAS), and various fluorinated and deoxygenated products of the parent metabolite M14. n = number of carbon atoms.

To this effect, our laboratory has previously demonstrated that phytochemicals isolated from ginger (6G, 8G, and 6S) can relax acetylcholine (ACh)-induced human airway smooth muscle (hASM) contraction by a reduction of intracellular calcium concentrations [Ca²⁺]_i (21). Further studies showed that 6G, 8G, and 6S potentiate β-agonist-induced relaxation of hASM, most likely by inhibiting phospholipase C (PLC), a key enzyme in the transduction of the extracellular signal (e.g., ACh), through muscarinic G protein-coupled receptors, to release sarcoplasmic reticulum (SR) stores of calcium that then lead to ASM constriction (22). Similar findings regarding [Ca²⁺]_i attenuation have been reported in murine ASM, albeit with crude methanol extracts from fresh ginger (23). In addition, we showed in cell-free assays that these compounds inhibit phosphodiesterases (PDEs), thereby preventing the breakdown of cAMP and promoting relaxation (22).

Although individual constituents of ginger, like 6S, have shown promise as potential drug targets, they are metabolized rapidly (24–28), and it is unclear if the resulting compounds retain any biological activity of their parent molecules. Until recently, the components of the 6S metabolome have largely been unknown. However, over the past several years, a diverse set of end-products of 6S metabolism has been identified, comprising of cysteine adducts (M1, M2, M4, and M5) (29–31), reduction products (M6, M8, M9, and M11) (30), (thio)ethers (M7, M10, and M12) (30), and oxidation products (M14 and M15) (32) (Fig. 1B). In addition, recently, several metabolite-inspired synthetic derivatives of 6S (GAS, SAS, M14-1, M14-2, M14-4, M14-8, and M14-9) (Fig. 1C) have been reported to have an enhanced activity over their corresponding parent compounds as anticancer drugs (33), and as protective agents against oxidative damage (34).

Although our prior studies have shown that 6S is able to relax hASM by inhibition of PLC-mediated intracellular calcium increase and PDE-mediated breakdown of cAMP (21, 22), orally administered 6S is metabolized quickly. Herein, we wanted to determine if 6S metabolites (Fig. 1B) retained the ability of the parent compound to modulate the influx of cytosolic calcium, and, furthermore, to functionally relax hASM tissue. Moreover, we hypothesized that some metabolites or metabolite-inspired synthetic derivatives (Fig. 1C) will show enhanced bioactivity compared with the parent compounds, opening opportunities for the development of novel drugs to help treat bronchial hyperreactivity.

MATERIALS AND METHODS

Human Tissue

Airway smooth muscle strips were acquired from tracheal discards from deidentified healthy donor lungs used in transplant surgery, which were deemed not human subjects research by the Institutional Review Board of Columbia University (IRB-AAAF2560).

Animals

All animal studies were approved by the Columbia University Institutional Animal Care and Use Committee. Eight- to 12-wk-old C57BL/6J mice of both sexes (Jackson Laboratories, Bar Harbor, ME) were used for all experiments.

Materials and Reagents

All materials were obtained from Sigma Aldrich (St. Louis, MO) unless otherwise specified. All cell culture reagents and supplies were purchased from Thermo Fisher Scientific (Waltham, MA) unless otherwise noted.

Shogaols and Metabolites

6-Shogaol was obtained from Dalton Pharmaceuticals (Toronto, ON, Canada). 8-shogaol and 10-shogaol derivatives were purified from ginger extract as previously reported (35). The shogaol metabolites and synthetic derivatives were synthesized following published methods: GAS (33), SAS (36), M1 (30), M2 (29, 30), M4-M12 (30), M14 (32), M14-1 (34), M14-2 (34), M14-4 (34), M14-8 (34), M14-9, and M15 (32, 34).

Cultured Cells

Immortalized hASM cells (a kind gift from Dr. William Gerthoffer at the University of Nevada, Reno, NV) lentiviral-transfected to express the human M3 muscarinic receptor (M3R) (22), and primary nonasthmatic hASM cells (a kind gift from Dr. Reynold Panettieri at the University of Pennsylvania, Philadelphia, PA) were prepared as described previously (37, 38) and cultured in M199 media (Gibco, Grand Island, NY) with fetal bovine serum (FBS, 10%), insulin transferrin selenium (ITS) (1.72 µM insulin, 0.069 µM transferrin, and 0.04 µM sodium selenite), epidermal growth factor (EGF, 0.25 µg/mL, R&D Systems, Inc., Minneapolis, MN), fibroblast growth factor (FGF, 1 µg/mL, R&D Systems, Inc., Minneapolis, MN), and antibiotics (100 U/mL penicillin G, 100 µg/mL streptomycin, and 0.25 µg/mL amphotericin B). The cells were cultured in 5:95% CO₂:air at 37°C in a humidified incubator until the study day. Immortalized hASM cells were used for experiments with ACh, as primary hASM cells do not retain the expression of muscarinic receptors. Primary hASM cells were used for experiments with bradykinin and histamine.

Intracellular Calcium Assays

Cells (immortalized hASM, primary hASM) were grown to confluence in black-walled, clear, and flat-bottomed Nunc 96-well plates. Immediately before experiments cells were exposed to basal M199 media without FBS or other additives for 48 h, washed with 100 µL fluorescence buffer four times (FB: 10 mM HEPES, 1 mM CaCl₂, 0.5 mM MgCl₂, 1 mM Na₂HPO₄, 5 mM KCl, 145 mM NaCl, 5 mM glucose, pH 7.4 at 37°C), then loaded with a Ca²⁺-specific fluorophore, Fura-2 AM (5 µM) with Pluronic F-127 (0.1%) (both from Life Technologies, Grand Island, NY) for 30 min in FB at 37°C and 5:95% CO₂:air. The cells were washed again with FB three times and incubated for an additional 30 min at 37°C and 5:95% CO₂:air. The cells were then pretreated (10 min) with vehicle (0.1% DMSO), PLC inhibitor U73122 (2.5–10 µM, MedChemExpress, Monmouth Junction, NJ), or a shogaol derivative (50 or 10 µM). Calcium fluorescence was recorded in the FlexStation 3 microplate reader (Molecular Devices, Sunnyvale, CA) (λ_ex = 340 and 380 nm, λ_em = 510 nm, cutoff at 495 nm) to determine baseline readings for 80 s followed by a challenge with acetylcholine (ACh), bradykinin, or histamine (each at 10 µM) with fluorescence readings for a total of 300–400 s. Raw values were expressed as a ratio of stimulated (f) to baseline (o) fluorescence (F) at (340/380) nm excitation:

$$\frac{F(340/380)f}{F(340/380)o}$$

These values were then normalized to the maximal vehicle response to get the percentage (%) of peak vehicle response and to allow for easy multicompound comparison. The intracellular calcium ([Ca²⁺]_i) dose responses to 6S (0.1–500 µM) were obtained in a similar manner in immortalized hASM stably expressing the muscarinic M3 receptor.

Inositol Trisphosphate Synthesis Assays

hASM cells [immortalized hASM cells stably transfected to express the M3R (22, 39), or primary hASM from healthy and asthmatic donors] were grown to confluence in 24-well plates and loaded overnight with ³H-myoinositol (6.4 µCi/mL; 44 Ci/mmol) (Perkin Elmer, Boston, MA) in inositol- and serum-free DMEM at 37°C and 5:95% CO₂:O₂ as described previously (40, 41). Following washing with Hank’s buffer with 10 mM LiCl, cells were pretreated for 15 min with PLC inhibitor U73122 (2.5–10 µM), shogaol derivatives (10–100 µM), or vehicle (0.1% DMSO) at 37°C, and then either stimulated with 10 µM ACh or bradykinin (30 min) or left unstimulated (basal response). The reaction was terminated by the addition of cold MeOH. Following a CHCl₃ extraction, newly synthesized ³H-inositol phosphates were separated from ³H-inositol by column chromatography. The amount of newly synthesized ³H-inositol phosphates [in counts per million (cpm)] was expressed as a ratio of the amount of ³H-inositol in each well that accounts for any differences in cell density or ³H-inositol incorporation into the plasma membrane.

MTT Cell Proliferation Assays

Vybrant MTT Cell Proliferation Assay Kit (Molecular Probes, Eugene, OR) was used to assess the cellular toxicity of shogaol compounds. MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] is a water-soluble compound that undergoes conversion to an insoluble formazan in the presence of oxidoreductases that are present in living cells. The company’s protocol was followed. Briefly, immortalized hASM cells stably overexpressing M3R were grown to confluence in phenol red-free Dulbecco’s Modified Eagle’s Medium/F12 media (Gibco, Grand Island, NY) with 10% FBS and antibiotics in clear and flat-bottomed Nunc 96-well plates (Thermo Fisher Scientific, Waltham, MA). Cells were treated with shogaol compounds (10 µM), H₂O₂ (100 mM, positive control), or DMSO (0.1%, vehicle) for 1 h, and then washed twice with media. MTT (1 mM) was then added and cells were incubated × 3 h (37°C, 95:5% air-CO₂) at which time all but 25 µL were removed. DMSO (50 µL) was used to solubilize any formazan crystals that may have formed. Cells were incubated × 10 min (37°C, 95:5% air-CO₂), mixed, and solubilized formazan absorbance was measured at 540 nm in the FlexStation 3 microplate reader (Molecular Devices, Sunnyvale, CA).

Human Myograph Studies

Human airway smooth muscle (hASM) tissue was carefully dissected, and the epithelia were removed. hASM strips were suspended under isotonic force (1.5 g) in organ baths and equilibrated in Krebs–Henseleit (KH) buffer (118 Mm NaCl, 5.6 mM KCl, 0.5 mM CaCl₂, 0.24 mM MgSO₄, 1.3 mM NaH₂PO₄, 25 mM NaHCO₃, 5.6 mM d-glucose, pH 7.4) for 1 h at 37°C. KH buffer was bubbled with a gas mixture of O₂:CO₂ (95:5%), and buffer was exchanged every 15 min for 1 h during equilibration of tracheal rings at 1.5 g resting tension. All buffers included indomethacin (10 µM) to block the synthesis of endogenous prostaglandinoids. Following equilibration, three ACh dose-response studies (100 nM to 1 mM) were performed with extensive buffer exchanges between and after these cycles. Pyrilamine (10 µM), MK571 (10 µM), and tetrodotoxin (1 µM) were added to buffers for 15 min to inhibit contractile contributions from endogenous histamine, leukotriene D4, and nerves, respectively. Strips were then contracted to their individually calculated EC₅₀ and allowed to reach plateau force. Each hASM strip was then treated with either vehicle (0.1% DMSO) or M6 (100 µM). The magnitude of relaxation was assessed at 30 min following treatment. Values reported are the percent of remaining contractile force of ACh EC₅₀-induced contraction at 30 min, with the force obtained after the ACh EC₅₀ plateau contraction representing 100% of muscle force and the baseline tension representing 0%.

Mouse Precision-Cut Lung Slices Preparation

Mouse precision-cut lung slices (PCLS) were prepared from C57BL/6J wild-type mice of both sexes as described in detail previously (42–44). Briefly, mice were euthanized after an intraperitoneal injection of 100 mg/kg pentobarbital sodium and the trachea was cannulated. The lungs were inflated with 1.3 mL of 2% low-melting temperature agarose in HBSS at 37°C. Lung lobes were cut into 130 µm slices using a tissue slicer (Compresstom VF-300; Precisionary Instruments, San Jose, CA). The collected lung slices were incubated overnight at 37°C in low-glucose DMEM (Gibco, Grand Island, NY) supplemented with antibiotics in a cell culture incubator with 10% CO₂.

Peripheral Lung Luminal Airway Diameter Measurements

Contractile response of peripheral airways was measured as previously described (45). In brief, lung slices were mounted in a custom-made perfusion chamber and the airway luminal diameter change was observed using phase-contrast microscopy, and images were recorded using a charge-coupled device (CCD) camera and image acquisition software (Video Savant; IO Industries, London, ON, Canada). Lung slices were continuously superfused with HBSS buffer allowing for the introduction and removal of contractile mediators (i.e., methacholine) or pretreatments (i.e., shogaol compounds or 0.1% DMSO vehicle) without halting perfusion.

Mouse Myograph Studies

A multiwire myograph system (DMT, Ann Arbor, MI) was used to measure tracheal ring ASM tension. C57BL/6J male mice were euthanized after an intraperitoneal injection of 100 mg/kg sodium pentobarbital. The tracheas were dissected free of connective tissue and cut in half. Each half-closed tracheal ring was mounted in an organ bath between two pins with one side attached to a force transducer and filled with Krebs–Henseleit (KH) buffer (115 Mm NaCl, 2.5 mM KCl, 1.91 mM CaCl₂, 2.46 mM MgSO₄, 1.38 mM NaH₂PO₄, 25 mM NaHCO₃, 5.56 mM d-glucose, pH 7.4) that was maintained at 37°C and continuously bubbled with O₂:CO₂ (95:5%). After equilibration at 5-mN resting tension for 1 h in KH buffer with buffer exchanges every 15 min, contractile force was continuously recorded using AcqKnowledge 3.7.3 software (Biopac Systems, Goleta, CA). The rings were contracted three times with increasing concentrations of ACh (0.1 µM to 1 mM) with extensive buffer exchanges between and after these cycles. Each ring was then contracted to its individually calculated EC₅₀ concentration with ACh. When the contraction reached a plateau in muscle contractile force, tissues were pretreated with DMSO (0.1%, control), 6-shogaol (6S), or the known PDE inhibitors, 3-isobutyl-1-methylxanthine (IBMX) or roflumilast (Tocris Bioscience, Minneapolis, MN) (50 µM each), followed by increasing concentrations of isoproterenol (0.1 nM to 10 µM). Slopes of tension over time (g/min) were calculated and normalized to control slope (DMSO) at each isoproterenol concentration from the same experiment.

Statistical Analysis

The values for cellular, human tissue, and mouse tissue data were analyzed using one-way ANOVA with repeated measures. The Bonferroni correction was applied for multiple comparisons. Statistical significance was established at P < 0.05 unless otherwise noted, and all values are expressed as means ± standard error of measurement (SEM).

RESULTS

Structure-Activity Relationship Screen Based on the 6-Shogaol Human Metabolome Reveals Potent Inhibitors of PLC-Mediated Increases in [Ca²⁺]_i

Prior studies have suggested that hASM relaxation by ginger phytochemicals is caused by inhibition of phospholipase C (PLC)-mediated increases in [Ca²⁺]_i (21). Herein, we measure the effect of 6S metabolites on ACh-stimulated [Ca²⁺]_i. Immortalized hASM cells stably overexpressing the muscarinic M3 receptor (M3R) were used because primary hASM cells do not retain the expression of muscarinic receptors to allow for assays with ACh. The cells were loaded with the fluorescent ratiometric calcium indicator, Fura-2 AM (5 µM). Baseline calcium fluorescence was recorded and did not vary appreciably between wells. The hASM cells were pretreated with shogaol derivatives for 10 min at 50 µM (Fig. 2A) or 10 µM (Fig. 2B) followed by stimulation with acetylcholine (ACh, 10 µM). In cells receiving vehicle (0.1% DMSO), ACh significantly increased [Ca²⁺]_i (Fig. 2C). However, pretreatment of hASM cells with ginger constituents, 6S and 8S, but not 10S, significantly reduced ACh-induced increases in intracellular calcium compared with vehicle (Fig. 2A), similar to our prior studies (21). Moreover, many of the human 6S metabolites (M2, M6, M7, M8, M9, M10, M12, M14, and M15), several of the synthetic compounds inspired by the metabolite M14 (M14-1, M14-4, and M14-8), and the aspirinate derivative based on 6G (GAS), also significantly attenuated ACh-stimulated calcium increases (Fig. 2A). On the other hand, certain metabolites and synthetic derivatives (M1, M4, M5, M11, M14-2, and M14-9) or the aspirinate derivative based on 6S (SAS) did not significantly attenuate increases in intracellular calcium fluorescence compared with the vehicle control.

Figure 2.

Intracellular calcium [Ca²⁺]_i response (Fura-2 AM, 5 µM) to ACh stimulation (10 µM) in the presence of shogaols, metabolites, or synthetic derivatives at 50 µM (A) or 10 µM (B), or vehicle (0.1% DMSO) in immortalized hASM cells overexpressing the muscarinic M3 receptor (M3R). Certain metabolites are able to attenuate Ca²⁺ increases after exposure to ACh. Values are reported as percent of maximal vehicle response. C: representative tracings. ***P < 0.001 and *P < 0.05 compared with vehicle (0.1% DMSO), $P < 0.001 compared with 6S, #P < 0.001 compared with M14, n = 6–8 replicates. ACh, acetylcholine; hASM, human airway smooth muscle.

In subsequent experiments, pretreatment of hASM cells at a lower concentration (10 µM) with a selection of the most promising compounds that reduced [Ca²⁺]_i at 50 µM (Fig. 2A) identified the synthetic metabolite-based derivative, M14-4 with enhanced ability to inhibit ACh-induced [Ca²⁺]_i increase compared with 6S and its parent compound M14 (Fig. 2B). Responses are normalized to a vehicle (0.1% DMSO).

6-Shogaol Reduces Acetylcholine-Induced [Ca²⁺]_i Response in a Dose-Dependent Manner

To examine whether the disruption of ACh-induced [Ca²⁺]_i increase is dose-dependent, immortalized hASM cells stably overexpressing M3R were loaded with the calcium-specific fluorophore, Fura-2 AM (5 µM). The cells were then pretreated with a range of 6S concentrations (0.1–500 µM) followed by stimulation with ACh (10 µM). ACh-induced increases in intracellular calcium concentrations were clearly attenuated in a dose-dependent manner with increasing concentrations of 6S (EC₅₀ = 17.32 ± 1.85 µM, n = 3 replicates) (Fig. 3).

Figure 3.

Dose-response of [Ca²⁺]_I (Fura-2 AM, 5 µM) to ACh stimulation (10 µM) in the presence of 6S (0.1-500 µM) in immortalized hASM cells overexpressing M3R. Values are reported as percent of maximal vehicle (DMSO 0.1%) response, n = 3 replicates. ACh, acetylcholine; hASM, human airway smooth muscle.

6-Shogaol and Its Metabolites Also Attenuate Bradykinin-Induced and Histamine-Induced [Ca²⁺]_i Response in Primary hASM Cells

To demonstrate that the effect on intracellular calcium is not only limited to one cell type or one Gq-coupled receptor (e.g., M3R), primary hASM cells were loaded with the ratiometric calcium dye, Fura-2 AM (5 µM) and pretreated (10 min) with selected shogaol derivatives (50 µM). For comparison, parallel experiments with immortalized hASM cells stably overexpressing M3R were also performed as already described. Baseline fluorescence was measured before stimulation with ACh (immortalized hASM cells, Fig. 4A), bradykinin (primary hASM cells, Fig. 4B), or histamine (primary hASM cells, Fig. 4C) at 10 µM. In both types of cells (immortalized and primary hASM), and when calcium response was elicited with three different Gq-coupled ligands (ACh, bradykinin, or histamine), all of the tested compounds (6S, M2, M6, and M14) significantly reduced the increase in [Ca²⁺]_i fluorescence signal compared with vehicle (0.1% DMSO) (Fig. 4).

Figure 4.

[Ca²⁺]_i response (Fura-2 AM, 5 µM) to either ACh (A), bradykinin (B), or histamine (C) stimulation (10 µM) in the presence of shogaol derivatives (50 µM) or vehicle (0.1% DMSO) in immortalized hASM cells overexpressing M3R (A) or primary hASM cells (B and C). Values are reported as percent of maximal vehicle response. ***P < 0.001 compared with vehicle (0.1% DMSO, n = 4–5 replicates). ACh, acetylcholine; hASM, human airway smooth muscle.

Shogaol Derivatives Inhibit ACh-Stimulated Inositol Phosphate Synthesis

Given that the calcium experiments showed that shogaol derivatives can modulate calcium response to ACh, we next tested whether they also reduced production of inositol trisphosphate (IP₃), which is a well-characterized surrogate for PLCβ activity and is its downstream second messenger product. Upon activation of PLCβ (via ACh-induced stimulation of M3R that allows for Gα_q proteins to activate PLCβ) PIP₂ is hydrolyzed to IP₃ and DAG. IP₃ is then able to bind to IP₃ receptors (IP3R) on sarcoplasmic reticulum (SR) causing the release of SR-stored Ca²⁺ into the cytoplasm and eventually leading to contraction. Using radiolabeled inositol, we demonstrate that ACh-stimulated PLCβ-mediated IP₃ synthesis is attenuated by pretreatment of cells with 6S, metabolite M6, and synthetic derivative M14-4 compared with vehicle (0.1% DMSO) (Fig. 5A). These results correspond to the same patterns that were observed with calcium fluorescence assays (Fig. 2A). Moreover, pretreatment with metabolites M5 and M11, and synthetic derivative M14-9 does not diminish IP₃ synthesis compared with vehicle (Fig. 5A), which is also in agreement with the results from [Ca²⁺]_i fluorescence assays (Fig. 2A). Finally, metabolite treatment does not affect the basal levels of IP3 synthesis (Fig. 5B).

Figure 5.

Inositol trisphosphate (IP₃) synthesis in immortalized hASM cells overexpressing M3R. A: cells were treated with shogaol derivative (100 µM) or vehicle (0.1% DMSO) for 15 min then stimulated with ACh (10 µM) for 30 min followed by measurement of IP₃ synthetic products. B: cells were left untreated (basal, 0.1% DMSO) or treated with M5 (100 µM) for 30 min followed by measurement of IP₃ synthetic products without ACh stimulation. Values are reported as percent of maximal vehicle (A) or basal (B) response. **P < 0.01 compared with vehicle/basal response, n = 4–8 replicates. ACh, acetylcholine; hASM, human airway smooth muscle.

Shogaol Derivatives Are Not Toxic to Cells

To show that the decrease in IP₃ synthesis and [Ca²⁺]_i signals are due to the test compound exposure and not cell death, we screened selected compounds (6S, M6, M9, M14-4, and M14-9) at 50 µM and cellular exposure of 1 h. Under these conditions, none of the 6S metabolites or synthetic derivatives demonstrate any significant cellular toxicity (measured by the commercially available MTT assay) in contrast to the positive control (100 mM H₂O₂) as compared with vehicle (0.1% DMSO) (Fig. 6). 6-Shogaol causes minimal cell toxicity (24%) after 1 h of exposure. Most of our cell-based and tissue assays are 10–40 min long.

Figure 6.

MTT-based toxicity studies with shogaol derivatives in primary nonasthmatic hASM cells. Cells were treated for 1 h with shogaol derivatives (50 µM), vehicle (0.1% DMSO), or H₂O₂ (100 mM) as a positive control, followed by exposure to MTT (1 mM). Compounds that were toxic to cells (e.g., positive control) prevented the conversion of the MTT to formazan, which was quantified by the measurement of formazan-specific absorbance (540 nm). Values are reported as percent of formazan remaining in vehicle-treated cells. ****P < 0.0001 compared with vehicle, n = 10 replicates. Data are represented as means ± SE. hASM, human airway smooth muscle.

The Human Metabolites of 6-Shogaol Relax Upper Airways (Human) and Lower Airways (Mouse) Ex Vivo

Because metabolites retain biological activity similar to that of the parent compound, 6S, in cell-based studies, we next tested their ability to relax precontracted muscle in upper and lower airways. Normal human tracheas (upper airway tissue) with epithelia denuded were suspended in the organ bath and contracted with acetylcholine (ACh) at approximately half-maximal effective concentration (∼EC₅₀) that was determined for each tracheal sample. Upon stable force generation, M6, the major human metabolite of 6S, was added to the buffer. DMSO (0.1%) was used as vehicle control (Fig. 7A). After 30 min of exposure, M6 significantly relaxes an ACh EC₅₀ contraction in hASM strips compared with vehicle (36 ± 13% of force remaining compared with 97 ± 7.1%, P < 0.01, n = 7 strips from 3 patients) (Fig. 7B).

Figure 7.

Ex vivo human airway smooth muscle contractile force measurements. Strips of human tracheal airway smooth muscle suspended in organ baths were contracted with an EC₅₀ concentration of ACh and were then subjected to vehicle (0.1% DMSO) or the primary human metabolite of 6-shogaol, M6 (100 µM). A: representative tracing. B: summary bar graph of percent of muscle force remaining at 30 min after exposure. **P < 0.01 compared with vehicle (0.1% DMSO, n = 7 strips from 3 patients).

In addition to relaxing upper airways, 6S and selected metabolites also relax lower airways. Precision-cut lung slices (PCLS) of C57BL/6J mice of both sexes were prepared as previously described (44). The airways were contracted with EC₅₀ of methacholine (MCh) and then exposed to vehicle (DMSO 0.1%), 6S, M6, M9, M14-4, and M14-9 (negative control). The parent compound, 6S, relaxes MCh-contracted lower airways (Fig. 8A) in a dose-dependent fashion with an EC₅₀ of 27.7 µM (Fig. 8B). When compared with 30 μM (near the EC₅₀ of 6S), 6S and its metabolites M6 and M9 were able to relax MCh-contracted lower airways equally well and very effectively (Fig. 8C). The synthetic derivative, M14-4, based on the naturally occurring metabolite M14 was much less potent. As expected from cellular studies of calcium (Fig. 2) and inositol phosphate synthesis (Fig. 5), M14-9 also had a very minimal effect on relaxation (Fig. 8C).

Figure 8.

6-Shogaol (6S) and metabolites relax mouse lower airways. A: a representative tracing (left) and photographs of airways (right) in a typical experiment where resting lower airway (1) contracts when exposed to MCh (100 nM) (2), relaxes when perfused with 6S and metabolites M6 and M9 (not shown) in the continuous presence of MCh (3), and recontracts when 6S is washed out with the solution containing only MCh (4). The images were taken at the times indicated by arrows and numbers in the trace. B: lower airways contracted with MCh (100 nM) are relaxed by 6S in a dose-dependent manner (1–300 µM) giving an EC₅₀ of 27.7 µM. C: to compare the efficacy of relaxation among different shogaol derivatives, the slices were contracted with MCh (100 nM) and then exposed to 30 µM 6S, M6, M9, M14-4, or M14-9 in experiments similar to that shown in A. 6S and metabolites M6 and M9, but not synthetic derivatives M14-4 and M14-9, relax airways almost completely. Airway relaxation is expressed as a percentage of ginger derivative-induced increase in luminal area compared with the MCh-induced decrease in luminal area in the same experiment. ***P < 0.001 compared with DMSO, ns = not significant compared with DMSO, n = 9 slices from three mice of both sexes. Data are represented as means ± SE. MCh, methacholine.

6-Shogaol and Its Main Human Metabolite M6 Prevent Recontraction of Isoproterenol-Relaxed Mouse Tracheal Rings and Potentiate β-Agonist-Induced Relaxation of Mouse ASM, Likely by Inhibiting PDE

Phosphodiesterases (PDEs) are endogenous enzymes that degrade cAMP, a second messenger that activates protein kinase A (PKA) leading to muscle relaxation. PDE and PLC belong to the same superfamily of enzymes that hydrolyze phosphodiester bonds. Previously, our laboratory has shown that 6S inhibits recombinant PDE4D (22) suggesting that ginger constituents may lead to ASM relaxation by inhibiting both intracellular calcium surge (by inhibiting PLC) and cAMP degradation (by inhibiting PDE). Functionally, when ACh-contracted mouse tracheal rings are relaxed with a β-agonist like isoproterenol, the tissue quickly regains a portion of its contractile tone (Fig. 9A, DMSO controls), which we hypothesize is due to degradation of cAMP by endogenous PDEs. Here, we show that cotreatment with 6S is able to prevent muscle recontraction, similar to cotreatment with 3-isobutyl-1-methylxanthine (IBMX) (Fig. 9A) and roflumilast, nonspecific and PDE4-specific PDE inhibitors, respectively (Fig. 9B). Similarly, cotreatment with M6, the major human metabolite of 6S, prevented muscle recontraction (Fig. 9B). On the other hand, the synthetic compound derived from metabolites M14, M14-9, did not prevent recontraction, which is in agreement with cell-based data where this compound did not prevent ACh-induced calcium surge (Fig. 2) or inositol phosphate synthesis (Fig. 5). This data lends functional support to our prior acellular assays with purified PDE enzyme.

Figure 9.

A bioactive component of ginger, 6-shogaol (6S), and its human metabolites prevent ASM recontraction after β-agonist-induced relaxation and potentiate β-agonist action. A: representative wire myograph tracings of isoproterenol relaxation (0.1 nM to 10 µM) and subsequent recontraction (DMSO vehicle control, black), which is blocked by 50 µM 6S (gray, left) and IBMX, a PDE inhibitor (gray, right). B: the comparison of recontraction slopes (open arrows in A) after pretreatment of mouse tracheal rings with DMSO (0.1%, control), IBMX (50 µM, nonselective PDE inhibitor), roflumilast (50 µM, selective PDE4 inhibitor), 6S (50 µM), M6 (50 µM, main metabolite of 6S), or M14-9 (50 µM, synthetic-derivative based on metabolite M14) followed by exposure to isoproterenol (100 nM). 6S and M6 prevent ASM recontraction, similar to known PDE inhibitors, whereas M14-9 does not, much like it does not prevent ACh-induced calcium rise or inositol phosphate synthesis. Slopes were normalized to DMSO control, n = 6–21 tracheal rings from 3–10 mice per group, *P < 0.05, **P < 0.01, ***P < 0.001, ns = not significant, all compared with DMSO. C: in ASM tracheal rings contracted with an EC₅₀ concentration of ACh, isoproterenol elicited a concentration-dependent relaxation. Tracheal rings were pretreated with DMSO (0.1%), M6 (50 µM), or M14-9 (50 µM) for 10 min, followed by exposure to isoproterenol (0.1 nM to 10 µM). M6 significantly potentiated isoproterenol-induced relaxation and caused a leftward shift in the relaxation curve (EC₅₀ = 7.9 nM) compared with the vehicle (0.1% DMSO, EC₅₀ = 22.9 nM), whereas M14-9 did not (EC₅₀ = 23.5 nM). Curves were fit with a four-parameter sigmoid to determine half-maximal effective concentrations (EC₅₀), n = 6–9 tracheal rings per group from 3–5 mice of both sexes. Data are represented as means ± SE. ACh, acetylcholine; ASM, airway smooth muscle; IBMX, 3-isobutyl-1-methylxanthine; PDE, phosphodiesterase.

Moreover, treatment of mouse tracheal rings with M6 potentiated isoproterenol-induced relaxation at each concentration of isoproterenol and caused a leftward shift in the overall relaxation curve (EC₅₀ = 7.9 nM) compared with the vehicle (0.1% DMSO, EC₅₀ = 22.9 nM) (Fig. 9C). This is in agreement with our previously published results with 6S, which was also shown to have a similar effect on β-agonist-induced relaxation (EC₅₀ = 1.1 nM) compared with DMSO vehicle (EC₅₀ = 28.5 nM) (22). On the other hand, the synthetic derivative M14-9, based on the metabolite M14, did not affect isoproterenol-induced relaxation (EC₅₀ = 23.5 nM) compared with the vehicle, as was predicted from cell-based assays where the compound also had no effect on calcium surge (Fig. 2) and inositol phosphate synthesis (Fig. 5) in response to ACh, or on airway relaxation (Fig. 8C).

DISCUSSION

Asthma is a chronic disease affecting millions worldwide and is on the rise as more people move to urban areas (46). It is associated with worse outcomes, especially if poorly managed, in various settings [such as perioperatively (47)] and illnesses [such as COVID-19 (48)]. Nearly half of the patients with asthma report poor symptom control with current therapies, and many turn to complementary and alternative medicines, such as ginger. Previously, we have reported that several specific components of ginger [e.g., 6-shogaol (6S)] are able to relax airway smooth muscle (ASM), but the bioavailability of orally ingested ginger components is limited because they are metabolized rapidly. In the current study, we demonstrate that metabolites of 6-shogaol attenuate receptor-Gq protein-mediated intracellular calcium and inositol phosphate increases, likely by inhibiting phospholipase (PLC) β, an important mediator of airway smooth muscle contraction. Furthermore, we demonstrate that metabolites retain the functional ability to relax upper (human tracheal muscle) as well as lower (mouse PCLS) airways, much like the parent 6-shogaol parent compound.

To facilitate the design of novel therapeutics, important insights into the structure-activity relationship (SAR) for inhibiting receptor-Gq protein-mediated intracellular calcium increases were obtained from fluorescent calcium assays with 23 distinct compounds: three ginger phytochemicals (6S, 8S, and 10S), 13 human metabolites of 6S (M1, M2, M4-M12, M14, and M15), two aspirinate derivatives (GAS and SAS), and five synthetic derivatives based on the metabolite M14 (M14-1, M14-2, M14-4, M14-8, and M14-9). Increasing the length of the aliphatic side chain hinders the activity of shogaols (6S > 8S ≫ 10S) (Fig. 10), which correlates with previously published reports with 6-gingerols, 8-gingerols, and 10-gingerols (21). Cysteine conjugation of 6S is inactivating, except for M2, which has a primary amine (vs. an amide in M5) and a ketone (vs. alcohol in M1) (Fig. 1B). These properties may allow the formation of more stable hydrogen bonds. The presence of secondary alcohol instead of a ketone imparts better activity (compare M9 with M11, and M12 with M10), also likely due to the ability to participate in hydrogen bonding. The identity of the atom added across the α,β-unsaturated ketone of 6S also makes a difference. For example, the ether M7 has a decreased activity compared with the corresponding thioether M10. In addition, masking of the para-hydroxy group with a benzoate derivative (GAS and SAS) renders the resulting molecules less active compared with parent compounds.

Figure 10.

Summary of structural determinants that are important for activity based on structure-activity relationship (SAR) studies performed by measuring the inhibition of intracellular calcium increases after stimulation with acetylcholine by 6S and its derivatives. Color-coded to match the groupings in Fig. 1. *Abbreviations: EDG, electron-donating group; EWG, electron-withdrawing group. †Listed are compounds that demonstrate the given structural element and their effect on activity. Compounds with stronger inhibition of intracellular calcium surge after acetylcholine stimulation are listed first.

The subgroup of synthetic molecules derived from M14 shows that the correct placement of electron-withdrawing groups (EWGs) such as trifluoromethyl (CF₃) moieties is important. Improved activity is achieved with the meta-CF₃ (M14-1) compared with ortho-CF₃ (M14-8) in the otherwise electron-donating group (EDG)-devoid phenyl backbone. Moreover, EDGs found on the parent compound (meta-methoxy and para-hydroxy), improve activity. If both phenyl EDGs are removed and replaced with an ortho-CF₃ EWG (M14-9) activity is dramatically reduced compared with M15. Halogenation of potential drug targets is often used in medicinal chemistry to improve drug properties, but it is not always clear that it is a useful transformation. Here, ortho-fluorination (M14-2) is deleterious (compare with M14-1). Finally, the planarity of the side chain may confer additional beneficial properties. The activities of M14 and M14-8, both with an additional double bond, are greater compared with M15 and M14-9, respectively.

Given these structural insights, it is not surprising then that the most potent inhibitor of intracellular calcium rise is M14-4. It has the same arrangement of EDG as the parent compound (meta-methoxy and para-hydroxy), but also the EWG (CF₃) in the optimal meta position. In more selective calcium assays at lower concentrations (Fig. 2B), we were able to show that M14-4 is superior to all the inhibitors that showed activity at higher concentrations including its parent compound, M14, and closely related cousin, M14-1.

Modulation of intracellular calcium by 6S metabolites is not limited to ASM isolated from a single source or one Gq-coupled receptor. Selective compounds (6S, M2, M6, and M14) are able to inhibit the increase of [Ca²⁺]_i in primary hASM cells derived from nonasthmatics stimulated with bradykinin or histamine, as well as in immortalized hASM cells overexpressing the M3 muscarinic receptor (M3R) stimulated with acetylcholine. This decrease in Gq-stimulated [Ca²⁺]_i upon exposure to shogaol derivatives is not due to shogaol-induced toxicity, according to our assays with MTT. Furthermore, the Gq-stimulated [Ca²⁺]_i increase is dose-dependent with an EC₅₀ ∼ 17 µM for 6S.

There was good congruency between those metabolites that inhibited Gq-mediated [Ca²⁺]_i and those metabolites that inhibited inositol phosphate synthesis (i.e., 6S, M6, M9, and M14-4). Similarly, there was concordance between calcium and IP3 assays with M5, M11, and M14-9 in that they did not inhibit increases in either of these second messengers induced by bradykinin receptor activation. Inositol trisphosphate (IP₃) and diacylglycerol (DAG) are the products of PLC activity and IP3 binds to a receptor on the sarcoplasmic reticulum to induce intracellular calcium release, so these findings are consistent with inhibition of PLC activity, which we have previously demonstrated with the parent compound 6S.

Phospholipase C belongs to a superfamily of enzymes that cleave phosphodiester linkages. Other important members of this superfamily are cyclic nucleotide phosphodiesterases (PDEs), which are of central importance in airways relaxation. Previously, we have shown that 6S is able to inhibit not just PLC, but also recombinant PDE4D (22). PDEs degrade cAMP, a second messenger classically associated with β-adrenoceptor airway smooth muscle relaxation. Thus, inhibition of PLC and PDE leads to bronchorelaxation through complementary pathways (Fig. 11). The ability of a metabolite of 6S, M6, to inhibit PDE in murine airway smooth muscle is demonstrated in isolated tracheal rings. The dose-dependent β-agonist relaxation of acetylcholine-contracted murine tracheal rings is commonly accompanied by a rapid partial recontraction of the airway smooth muscle, which is thought to be due to the rapid degradation of cAMP by PDEs. This partial recontraction is reflected by an upslope in the muscle force after each concentration of β-agonist. We demonstrated that this upslope of recontraction is attenuated in the presence of 6S and M6 as well as two classic inhibitors of PDEs, IBMX and roflumilast. Moreover, the presence of M6 caused a left shift in the isoproterenol-mediated relaxation, again consistent with PDE inhibition. These findings with M6 are consistent with our previously published results with 6S, which inhibited purified PDE4D (22). Given the similarity of the inhibition profile between 6S and M6 for PLC and PDE inhibition, it is likely that additional metabolites will retain the ability to inhibit PDE as well.

Figure 11.

Simplified depiction of the components of adrenergic (left) and muscarinic (right) pathways involved in smooth muscle constriction/relaxation with PLC and PDE as the likely targets of shogaol derivatives (green structure). Muscarinic pathway: briefly, an extracellular ligand (e.g., ACh) binds to M3R causing a conformational change that allows Gα_q to activate PLC. Activated PLC hydrolyzes membrane-bound PIP₂ into DAG (a second messenger in the PKC pathway) and IP₃. IP₃ is able to bind to IP3R on SR membranes, causing the release of SR-stored Ca²⁺. An increase in cytosolic [Ca²⁺] activates the CaM-MLCK complex, which is then able to phosphorylate MLC, leading to contraction. Adrenergic pathway: upon agonist binding, βAR undergoes a conformational change that allows Gα_s to activate AC, which is then able to convert ATP to cAMP. cAMP is a potent activator of PKA that phosphorylates MLCP and turns it on. Activated MLCP can then remove the phosphate moiety from MLC thereby setting off relaxation. The most likely targets of M6, M14-4, and other active shogaol derivatives (illustrated as a simplified backbone structure in green) are PDE and PLC, based on our prior work. AC, adenylyl cyclase; ACh, acetylcholine; βAR, β2 adrenergic receptor; CaM, calmodulin; DAG, diacylglycerol; IP₃, inositol trisphosphate; IP₃R, IP₃ receptor; MLC, myosin light chain; MLCK, MLC kinase; MLCP, MLC phosphatase; PDE, phosphodiesterase E; PIP₂, phosphatidylinositol-4,5-bisphosphate; PKA, protein kinase A; PKC, protein kinase C; PLC, phospholipase C; SR, sarcoplasmic reticulum.

Although several metabolites and metabolite-inspired synthetic compounds showed promise in cell-based calcium and IP3 assays, cellular activity does not always translate into physiologically significant activity. We tested the most promising compounds for their ability to relax precontracted airway smooth muscle. We demonstrate that the major human metabolite of 6S (M6) retains the biological activity of the parent compound and is able to relax acetylcholine-precontracted human tracheal muscle strips. In addition, M6 and M9, along with the parent compound 6S, can relax methacholine-precontracted peripheral mouse airways in precision-cut lung slices. However, M14-4, which was the most potent in cellular calcium assay screen, is not able to relax precontracted airway smooth muscle as well as other metabolites or 6S, underscoring the importance of evaluating promising targets in functional assays. We speculate that this is most likely because M14-4 does not penetrate the intact tissue when delivered in a superfusate over the tissue, in comparison with the direct application of the compound to a cell monolayer in the calcium studies. Nonetheless, these results support the feasibility of designing compounds that are based on a novel architecture that would allow us to develop new drugs for bronchoconstriction. Given our cell-based and tissue-based studies, the most promising compounds in the currently available arsenal are M6 and M9, which could be further derivatized to enhance their activity by increasing planarity or addition of thiol moiety in the β position compared with the alcohol. Furthermore, although M14-4 did not have an effect in tissue-based studies, it is possible that changes to the backbone (such as the reduction of the ketone) would increase its permeability and possibly even further enhance its activity in both cell and tissue studies.

In conclusion, we report the first comprehensive examination of the complex human metabolites of 6-shogaol in airway smooth muscle and demonstrate that they retain biological activity both in terms of their cellular mechanism of action and functional assays in intact airway smooth muscle. We have been able to identify a novel synthetic compound (M14-4) based on the metabolite M14 with greatly improved ability to inhibit acetylcholine-induced intracellular calcium increase compared with the parent metabolite and the original ginger phytochemical, 6-shogaol. Although this compound is not able to relax ASM as well as 6S or other metabolites, these experiments pave the way for the development of new synthetic compounds that are inspired by endogenous metabolic pathways. This may lead to novel molecules with improved bronchodilating properties that can modulate enzymes in complementary pathways, and it demonstrates the power and utility of using natural products to guide our drug discovery.

GRANTS

This research was supported by the National Institutes of Health (NIH) National Heart, Lung, and Blood Institute Grant HL122340 (to C.W.E.), National Institute of General Medical Sciences Grants GM065281 (to C.W.E.) and T32GM008464 (to E.L.), the Stony Wold-Herbert Fund Inc. (to E.L.), and the Virginia Apgar Scholars Program (to E.L.) at the Department of Anesthesiology, Columbia University.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

E.L. and C.W.E. conceived and designed research; E.L., J.F.P.-Z., Y.Z., Y.Z., S.S., and C.W.E. performed experiments; E.L., J.F.P.-Z., Y.Z., and C.W.E. analyzed data; E.L., J.F.P.-Z., Y.Z., S.S., and C.W.E. interpreted results of experiments; E.L., J.F.P.-Z., and C.W.E. prepared figures; E.L. and C.W.E. drafted manuscript; E.L., J.F.P.-Z., and C.W.E. edited and revised manuscript; E.L., J.F.P.-Z., Y.Z., Y.Z., S.S., and C.W.E. approved final version of manuscript.