Article pubs.acs.org/jnp
Dietary Acetylenic Oxylipin Falcarinol Differentially Modulates GABAA Receptors Marta Magdalena Czyzewska,*,† Lukasz Chrobok,† Alan Kania,† Magdalena Jatczak,† Federica Pollastro,‡ Giovanni Appendino,‡ and Jerzy Wladyslaw Mozrzymas†,§ †
Laboratory of Neuroscience, Department of Biophysics, Wroclaw Medical University, ul. Chalubinskiego 3, 50-368 Wroclaw, Poland Dipartimento di Scienze del Farmaco, Università del Piemonte Orientale, Via Bovio 6, 28100 Novara, Italy § Department of Animal Molecular Physiology, Institute of Zoology, Wroclaw University, Cybulskiego 30, 50-205 Wroclaw, Poland ‡
ABSTRACT: The dietary oxylipins falcarinol (1a) and falcarindiol (1b) trap thiols by direct nucleophilic addition to their diyne system, but despite this, only falcarinol (1a) is a reversible agonist of cannabinoid receptors, providing a rationale for comparing their activity also on other neuronal targets. Because GABAA receptors (GABAARs) are exquisitely sensitive to polyacetylenic oxylipins in terms of either potentiation (falcarindiol, 1b) or inhibition (oenanthotoxin, 2a), the activity of 1a was investigated on synaptic (α1β2γ2L) and extrasynaptic (α1β2δ and α1β2) subtypes of GABAARs. Falcarinol (1a) significantly enhanced the amplitude of currents mediated by α1β2γ2L receptors, but this effect was associated with a use-dependent block. Conversely, α1β2 receptors were inhibited without any sign of use-dependent block for the entire range of concentrations tested (1−10 μM). Interestingly, responses mediated by α1β2δ receptors, showing no or very little macroscopic desensitization, were strongly potentiated by 1a, exhibiting a fading reminiscent of macroscopic desensitization. When compared to the activity of falcarindiol (1b), falcarinol (1a) showed a higher affinity for GABAARs and, overall, a substantially different profile of pharmacological action. Taken together, the present data support the view that modulation of GABAARs might underlie the insecticidal and sedative activity of falcarinol (1a).
F
alcarinol (1a) is a major dietary polyacetylenic oxylipin. It is common in food plants from the family Apiaceae [carrot (Daucus carota L.), celery (Apium graveolens subsp. dulce (Mill.) Schübl. & G. Martens), parsnip (Pastinaca sativa L.), coriander (Coriandrum sativum L.) and parsley (Petroselinum crispum (Mill.) Fuss)], and also occurs in medicinal plants from the family Araliaceae, like ginseng (Panax ginseng L.) and ivy (Hedera helix L.).1,2 Falcarinol, a bitter compound, was long considered an undesirable dietary toxin, as cogently documented by its alternative name “carotatoxin”.3 However, over the past three decades, this view has been revised substantially in the wake of the discovery of the antinociceptive4 and potential anticancer5 properties of this compound. Current interest in falcarinol (1a) was triggered by distinct lines of research that exemplify the pleiotropic profile of bioactivity of this polyacetylene. Thus, while carotenoid-rich plants like carrots have been related epidemiologically to a preventive activity on cancer, clinical trials of β-carotene as a chemopreventive agent have failed, and it has been suggested that the epidemiological correlations that spurred these studies would better fit the biological profile of falcarinol (1a), an inhibitor of cancer resistance proteins as well as a selective cytotoxin,6 rather than that of β-carotene.3 On the other hand, falcarinol (1a) was also identified as an inverse agonist of the cannabinoid receptor CB1, capable of selectively alkylating the anandamide binding site, increasing the expression of chemokines in the skin, and overall inducing pro-allergic effects in the © XXXX American Chemical Society and American Society of Pharmacognosy
skin.7 As an electrophilic agent, falcarinol (1a) can trap thiol groups in a covalent way8 and is considered the causative agent of contact dermatitis from ivy,9 a growing problem in several European countries as well as in the U.S.10 The reaction of falcarinol (1a) and falcarindiol (1b) with thiols proceeds by direct addition to the triple bond and leads to a mixture of diastereomeric vinyl sulfides (Scheme 1).8,11 This electrophilic behavior underlies the CB1 reverse-agonist activity of 1a7 as well as its pro-allergenic properties, and might well represent Received: August 1, 2014
A
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
nature, it might cross the blood-brain barrier and enter the brain, where, apart from neuritogenic activity12 and the interaction with CB1,7 it could also interact with GABAA receptors (GABAARs), which are exquisitely sensitive to polyacetylenic compounds.13−15 Thus, oenanthotoxin (2a) and dihydrooenanthotoxin (2b), the poisonous acetylenic constituents of Oenanthe species, inhibit GABAergic responses in a dose-dependent manner (EC50 value in the low micromolar range) by a complex mechanism comprising allosteric modulation of agonist binding and desensensitization, with some contribution of the open channel block.13,14 Minor structural changes greatly modify the modulatory effects on GABAARs, with allosteric effects and open channel block being strongly dependent on polarity,15 as shown by the observation that the related polyacetylene falcarindiol (1b) enhances rather than inhibits current responses to low GABA concentrations (Scheme 1).16 Falcarindiol (1b) and falcarinol (1a) react in the same way with thiols,8,11 but show different biological profiles, since
Scheme 1. Reaction of Falcarinol (1a, R = H) and Falcarindiol (1b, R = OH) with Thiols
the hallmark of its biological profile. In humans, falcarinol is rapidly absorbed after oral administration.3 Due to its lipophilic
Figure 1. Current responses mediated by α1β2γ2L receptors are affected by falcarinol (1a) by a complex mechanism comprising initial upregulation and progressive accumulation of the use-dependent inhibition. (A) Typical superimposed current traces evoked by 3 μM GABA under control conditions (black line) and in the presence of 1 μM 1a (gray line, first response to GABA in the presence of 1a). (B) Superimposed traces of currents evoked by 3 μM GABA in control conditions (black line) and in the presence of 3 μM 1a (gray line, additionally sweep numbering is provided). Note the use-dependent current inhibition. (C) Statistics of relative values of current amplitude for considered range of 1a concentrations. Gray bins present the saturation of the 1a effect (observed upon recordings of several current responses), while white bins present the relative value of amplitude for the first sweep. (D) Normalized and superimposed initial phases of currents evoked by 3 μM GABA in control conditions 1a (black line) or in the presence of 1 μM 1a (dark gray line) and 10 μM 1a (gray line). (E) Statistics of relative values of current rise time for all tested 1a concentrations. (F) Normalized and superimposed currents evoked by 3 μM GABA under control conditions (black line) or in the presence of 3 μM 1a. Deepening of current fading in the presence of 1a is shown by comparison of the 1st, 5th, and 6th sweeps (note the saturation of this effect in 5th and 6th sweeps). (G) Statistics of the extent of fading recorded under control conditions (black column) and in the presence of 1a within the considered concentration range. Gray bins present mean values of the saturated 1a effect, whereas white bins present fading of the first sweep. Asterisks indicate statistical significance (p < 0.05). B
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 2. Falcarinol (1a) downregulates the amplitude of currents mediated by α1β2 receptors and affects the receptor kinetics. (A) Superimposed current responses under control conditions (black line) and in the presence of 1 μM of 1a. (B) Statistics of relative values of current amplitudes measured in the presence of the range considered of 1a concentrations. (C) Normalized and superimposed traces of the initial phases of current responses under control conditions (black line) and in the presence of 0.1 μM 1a (gray line) and 5 μM 1a (dark gray line). (D) Statistics of relative rise time of currents in the presence of all 1a concentrations tested. (E) Normalized and superimposed current traces recorded under control conditions (black line) and in the presence of 5 μM 1a (dark gray line). (F) Statistics of fading calculated as (Imax − Iend)/Imax. The black column presents the mean value for fading under control conditions. Asterisks indicate statistical significance (p < 0.05).
−40 mV. The mean value of current amplitude mediated by α1β2γ2L receptors was −643.82 ± 8.77 pA (n = 58). Low concentration of 1a (0.1−1 μM) significantly increased the amplitude of evoked currentsthe relative values were the following: 1.18 ± 0.05 (p < 0.05), 1.29 ± 0.06 (p < 0.05), and 1.41 ± 0.09 (p < 0 0.05) for 0.1, 0.3, and 1 μM 1a, respectively (Figure 1A,C). The first GABA application (first sweep) in the presence of 3 μM 1a had a markedly larger amplitude than in control conditions (Figure 1B,C) but at 5 μM a reversal of current enhancement (for the first sweep) was observed (Figure 1C). Interestingly, in addition to this biphasicity of falcarinol (1a) concentration-dependence, an apparently usedependent effect of this compound was revealed, because in consecutive sweeps, a profound and progressive reduction of current amplitude was seen (Figure 1B,C). The saturation of this use-dependent effect was achieved after 5 to 6 sweeps. Notably, as shown in Figure 1F, after the first pulse in the presence of 1a, subsequent current responses were characterized by increased current fading (see also below), further suggesting a use-dependent block. The relative amplitude values determined for the first sweep were 1.76 ± 0.25, n = 6 (p < 0.05) and 1.17 ± 0.21, n = 8 (p > 0.05) for 3 and 5 μM 1a, respectively. After saturation of the effect of 1a on the current amplitude, for responses mediated by α1β2γ2L, the relative values were 0.68 ± 0.12, n = 8 (p < 0.05) and 0.40 ± 0.05, n = 8 (p < 0.05) for 3 and 5 μM 1a, respectively. In the presence of a high concentration of 1a (10 μM), the current amplitude showed a tendency to decrease already at the first sweep (Figure 1C), whereas at subsequent ones, a profound current reduction was observed, reaching saturation at the second response (for the first sweep, the relative amplitude value was
falcarindiol is completely inactive in assays of CB1 binding7 and also displays lower cytotoxic activity than falcarinol.17 These differences highlight the role of the molecular milieu for covalent ligands 18 and provide a rationale for investigating the action of falcarinol on GABAA receptors. GABAARs are responsible for inhibition in the adult mammalian central nervous system and have a pentameric structure, with as many as 19 subunits identified so far, namely, α1 − α6, β1 − β3, γ1 − γ3, ε, θ, ρ1 − ρ3, δ, and π.19 Subunit composition has been shown to determine their functional and pharmacological properties.20,21 In particular, two major forms of GABAergic inhibition have been described, referred to as phasic (synaptic) and tonic (extrasynaptic) and mediated by different receptor types, for instance, α1β2γ2L and α1β2δ (with some contribution from α1β2), respectively.22,23 In the present report, the differential modulation of these two forms of inhibition by falcarinol (1a) is described.
■
RESULTS AND DISCUSSION Falcarinol (1a) was isolated in a very large yield (ca. 1% on dry weight basis) from the edible tubers of earth chestnut (Bunium bulbocastanum L., Apiaceae), a plant popular in Europe before the introduction of potatoes from the New World,24 with the seeds used as a spice (black cumin) in Asia. To assess the impact of 1a on GABAergic currents, responses mediated by receptors implicated in tonic and phasic inhibition were investigated. First, the effect was examined on the most abundant GABAA receptor in the central nervous system (CNS), α1β2γ2L, which plays a crucial role in phasic inhibition. For this purpose, whole-cell responses to 8 s application of 3 μM GABA were recorded at the holding membrane voltage of C
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 3. Falcarinol (1a) upregulates the amplitude of currents mediated by α1β2δ receptors and affects their kinetics. (A) Superimposed traces of currents evoked by 3 μM GABA alone (black line) and in the presence of 3 μM of 1a. (B) Statistics of relative values of current amplitude. (C) Normalized and superimposed traces of current onset measured in control conditions (black line) in the presence of 0.1 μM of 1a (gray line) and in the presence of 5 μM of 1a (dark gray line). (D) Statistics of relative values of current onset (rise time). (E) Normalized and superimposed of current traces evoked by 3 μM GABA alone (black line) or in the presence of 5 μM 1a (gray line). (F) Statistics of fading calculated as (Imax − Iend)/ Imax. The black column presents the mean value for fading under control conditions. Asterisks indicate the statistical significance (p < 0.05).
conductance with respect to α1β2γ2L receptors, and it has been proposed that these receptors may play a role in mediating the tonic GABAergic conductance.25 Thus, the impact of 1a on this type of GABAARs was also investigated. The mean amplitude of currents mediated by α1β2 receptors (evoked by 3 μM GABA) was −576.40 ± 10.9 pA (n = 5). Interestingly, in contrast to α1β2γ2L receptors, 1a, at all tested concentrations, reduced current amplitude in a dose-dependent manner (relative values: 0.86 ± 0.05, n = 7, p < 0.05 for 0.1 μM; 0.81 ± 0.03, n = 8, p < 0.05 for 0.3 μM; 0.50 ± 0.03, n = 14, p < 0.05 for 1 μM; 0.20 ± 0.01, n = 8, p = 0.05 for 3 μM; 0.30 ± 0.05, n = 10, p < 0.05 for 5 μM; 0.06 ± 0.01, n = 6, p < 0.05 for 10 μM), and this effect was not associated with any apparent use-dependence (Figure 2A,B). Under control conditions, the time course of currents mediated by α1β2 receptors was characterized by a faster rise time (121.89 ± 9.66 ms, n = 45) with respect to α1β2γ2L receptors, and this difference was significant (p < 0.05). However, α1β2 receptor-mediated currents were characterized by a profound fading (Figure 2A,E), reflected by a large value of the fading coefficient 0.67 ± 0.03 (n = 34), which was significantly larger (p < 0.05) in comparison to that determined for α1β2γ2L receptors (0.35 ± 0.02, n = 40). Falcarinol (1a) showed a concentration-dependent effect on the current onset, accelerating it at low (relative rise time 0.70 ± 0.05, n = 7, p < 0.05 for 0.1 μM 1a and 0.82 ± 0.05, n = 8, p < 0.05 for 0.3 μM 1a, Figure 2C,D) and slowing it down at high concentrations (relative rise time 2.16 ± 0.38, n = 10, p < 0.05 for 5 μM 1a; 1.52 ± 0.12, n = 4, p < 0.05 for 10 μM 1a, Figure 2C,D). At 1 and 3 μM, the effect on the current onset was not significant, although at 3 μM, a clear trend to slow it down was seen (1.02 ± 0.15, n = 11, p > 0.05 for 1 μM 1a; 1.43 ± 0.36, n = 5, p > 0.05 for 3 μM 1a, Figure 2D). The effect of falcarinol on
0.52 ± 0.18, n = 6, p > 0.05; for the second sweep, the mean relative value was 0.08 ± 0.03, n = 6, p < 0.05; the third sweep had an amplitude undistinguishable from the second one). The effect of 1a on the amplitude of currents mediated by the α1β2γ2L receptor was reversible, ruling out a dock-and-lock covalent mechanism of activity mediated by the alkylation of cysteine groups. The time course of the α1β2γ2L receptor-mediated currents was described by determination of the current rise time and fading parameters (see Experimental Section). Under control conditions, the mean values for the rise time and fading coefficient were 164.99 ± 9.57 ms (n = 50) and 0.35 ± 0.02 (n = 40), respectively. Falcarinol (1a), at low and medium concentrations (0.1−5 μM), had no effect or showed a slightly prolonged rise time of evoked currents (the relative values for 0.1, 0.3, 1, 3, 5 μM of 1a respectively: 0.90 ± 0.05, n = 8, p > 0.05; 1.25 ± 0.10, n = 7, p < 0.05; 1.21 ± 0.12, n = 18, p > 0.05; 0.99 ± 0.16, n = 6, p > 0.05; 0.79 ± 0.11, n = 8, p > 0.05). Falcarinol (1a) at high concentration (10 μM) clearly reduced the rise time of current responses (relative value: 0.19 ± 0.07, n = 3, p < 0.05) (Figure 1D,E). Acceleration of current onset was accompanied by increased fading except for the lowest concentrations of 1a (under control conditions 0.35 ± 0.02, n = 40; mean values for effect at its saturation: 0.21 ± 0.06, n = 9, p < 0.05 for 0.1 μM 1a; 0.32 ± 0.08, n = 8, p > 0.05 for 0.3 μM; 0.50 ± 0.04, n = 19, p < 0.05 for 1 μM; 0.85 ± 0.02, n = 8, p < 0.05 for 3 μM; 0.85 ± 0.03, n = 8, p < 0.05 for 5 μM; 0.97 ± 0.03, n = 6, p < 0.05 for 10 μM 1a; mean values for the first sweep: 0.78 ± 0.02, n = 7, p < 0.05 for 3 μM 1a, 0.59 ± 0.05, n = 8, p < 0.05 for 5 μM 1a, 0.86 ± 0.04, n = 5, p < 0.05 for 10 μM 1a, Figure 1F,G). It is known that the α1 and β2 subunits may coassemble and form α1β2 receptors characterized by higher affinity and lower D
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
strongly dependent on the identity of the subunit completing the pentamer comprising two α1 and two β2 subunits. For the α1β2γ2L receptors, which play a crucial role in phasic inhibition and in being also the most abundant GABAAR in the CNS, low falcarinol (1a) concentrations enhanced current amplitude elicited by nonsaturating GABA, similar to previous observation for falcarindiol (1a) in primary hippocampal neurons.16 This effect may suggest an upregulation of agonist binding rate, but such a mechanism would predict also acceleration of the current onset that has not been observed in the present experiments. Most likely, 1a may affect not only binding, but also gating (e.g., the opening/closing rate), although this possibility would require a detailed receptor kinetic study beyond the scope of this current investigation. Interestingly, high concentrations of 1a induced a use-dependent inhibition of α1β2γ2L receptors, strongly suggesting the open channel block. Observation of cumulative amplitude reduction, shortening of current rise time correlated with amplitude decrease and increased current fading were all compatible with the open channel block mechanism. Interestingly, the use-dependent blockade observed for compound 1a closely resembled the effect described for oenanthotoxin (2a) on current responses elicited by low GABA.14 In addition, it was found that oenanthotoxin analogues characterized with a low hydrophobicity also induced a use-dependent block.15 Thus, the open channel block might be a peculiar feature of several polyacetylenes acting on GABAA receptors containing a γ2L subunit. Taken altogether, the present data indicated that in the case of α1β2γ2L receptors, falcarinol (1a) may exert an allosteric modulation affecting probably both binding and gating properties and such actions are accompanied by an open channel block. In contrast to α1β2γ2L GABAARs, α1β2 receptors were inhibited by 1a with no sign of an open channel block. A relatively potent amplitude inhibition is associated with a biphasic effect on the rise time and a weak impact on current fading (Figure 2). Again, such effects appear incompatible with a pure modulation of receptor binding properties but the precise mechanism of action of falcarinol (1a) on these receptors requires further studies. In contrast to α 1 β 2 GABAARs, α1β2δ receptors, which are believed to be more representative for mediating tonic currents, were potentiated strongly by medium and high doses of 1a. No signs of usedependent inhibition were observed, although the trend of growth of relative increase in current amplitude appeared to reverse at high concentrations of 1a (Figure 3). Interestingly, amplitude increase in the presence of falcarinol is accompanied by appearance of marked current fading, which was absent under control conditions. This finding is intriguing as some δ subunit-containing receptors have been described as “nondesensitizing”. However, the lack of macroscopic desensitization does not provide sufficient evidence for the absence of microscopic desensitization (i.e., that the desensitized conformation is not available for the receptor).27,28 Indeed, in the presence of 3α,21-dihydroxy-5α-pregnan-20-one (THDOC), responses of δ subunit-containing receptors greatly increased and pronounced fading appeared revealing the receptor ability to desensitize.29 Similar observations lacking macroscopic desensitization could be revealed by a high potency agonist has been described for α1β1γ2L receptors with mutation at the agonist binding site (α1F64).30 In contrast to α1β2γ2L and α1β2, the effect of falcarinol (1a) on α1β2δ receptors appeared irreversible. The nature of this interaction is not clear. It cannot
current fading was to moderately reduce the fading coefficient in a dose-dependent manner while at low concentrations of 1a (0.1−1 μM) there was no effect (0.65 ± 0.06, n = 7, p > 0.05 for 0.1 μM 1a; 0.60 ± 0.03, n = 8, p > 0.05 for 0.3 μM 1a; 0.64 ± 0.02, n = 12, p > 0.05 for 1 μM 1a). At higher doses (3,10 μM), however, this effect became significant (0.42 ± 0.04, n = 6, p < 0.05 for 3 μM; 0.54 ± 0.08, n = 10, p > 0.05 for 5 μM 1a; 0.45 ± 0.02, n = 5, p < 0.05 for 10 μM 1a, Figure 2E,F). As already mentioned, α1β2 receptors have been shown to be putatively extrasynaptic.25 However, it is widely accepted that δ subunit-containing receptors play a crucial role in mediating tonic inhibition.23,26 Therefore, it was decided to extend the current studies to an extrasynaptic receptor containing δ subunit. To minimize the number of changes in the subunit composition, α1β2δ receptors were chosen as a representative extrasynaptic GABAAR. Under control conditions, application of 3 μM GABA elicited responses with an average amplitude of −174.34 ± 29.51 pA (n = 49). The 10−90% rise time for these currents in control conditions was 141.19 ± 4.47 ms (n = 36). As expected, currents mediated by α1β2δ receptors showed minimal fading, consistent with weak macroscopic desensitization. Indeed, the fading coefficient was close to zero (Figure 3E) as differences between Imax and Iend values (methods) were at the level of current noise. Low concentrations of falcarinol (1a) (0.1−1 μM) either had no effect or slightly increased the amplitude of evoked currents (relative values: 1.04 ± 0.05, n = 7, p > 0.05 for 0.1 μM 1a; 1.46 ± 0.07, n = 15, p < 0.05 for 0.3 μM 1a; 1.26 ± 0.08, n = 7, p > 0.05 for 1 μM, Figure 3B). Interestingly, at medium and high concentrations, falcarinol profoundly increased the amplitude of current responses (relative values: 3.20 ± 0.58, n = 9, p < 0.05 for 3 μM 1a; 8.31 ± 1.74, n = 7, p < 0.05 for 5 μM 1a; 5.68 ± 1.97, n = 4, p < 0.05 for 10 μM 1a, Figure 3A,B). In contrast to observations made for α1β2γ2L and α1β2 receptors, this effect was irreversible in the time scale of the recordings made (after tens of minutes of rinsing, only a slight washout of effect of 1a was observed). Falcarinol, at low concentration, except for 1 μM, slowed down the current onset and had no effect at medium and high concentrations (relative values for 10−90% rise time: 1.69 ± 0.27, n = 6, p < 0.05 for 0.1 μM 1a; 1.87 ± 0.18, n = 8, p < 0.05 for 0.3 μM 1a; 1.05 ± 0.05, n = 8, p > 0.05 for 1 μM 1a; 1.50 ± 0.15, n = 5, p > 0.05 for 3 μM 1a; 1.78 ± 0.33, n = 5, p > 0.05 for 5 μM 1a; 1.39 ± 0.13, n = 4, p > 0.05 for 10 μM 1a, Figure 3C,D). At low concentrations (0.1−1 μM), 1a had no effect on current fading, which remained at the detection level as under control conditions (Figure 3F). However, surprisingly, at higher concentrations, a strong upregulation of current amplitude (Figure 3E,F) was accompanied by the appearance of a marked fading (0.17 ± 0.05, n = 9, p < 0.05 for 3 μM 1a; 0.37 ± 0.07, n = 7, p < 0.05 for 5 μM 1a; 0.47 ± 0.04, n = 4, p < 0.05 for 10 μM 1a, Figure 3F). The major finding of this study was that falcarinol (1a) exerted differential effects on GABAA receptor subtypes involved in phasic and tonic forms of GABAergic inhibition. Strikingly, the effect on each GABAAR subtype investigated was characterized by a substantially different mechanism. Indeed, responses mediated by the α1β2γ2L receptors were potentiated with a concomitant use-dependent block, with the α1β2 receptors inhibited and the α1β2δ receptors irreversibly potentiated (within a recording period up to 40 min), with no sign of the use-dependent block. The present data thus have demonstrated that the effect of 1a on GABAA receptors is E
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
using the conditions developed by Christensen and Kreutzmann.36 Plant Material. Bulbs of Bunium bulbocastanum were collected around Ceva (Italy) in August 2010. The plant material was identified by Dado Luciano and Renzo Salvo, and a voucher specimen (number 03/10), is kept at the Dipartimento di Scienze del Farmaco, Università del Piemonte Orientale, Novara, Italy. Extraction and Isolation. Powdered dried tubers of Bunium bulbocastanum (27.6 g) were extracted with acetone (2 × 250 mL) to afford, after removal of the solvent, 600 mg of a brownish oil, that was purified by gravity-column chromatography on silica gel (25 mL) using petroleum ether-EtOAc (9:1) as eluant, to afford 280 mg (1%) of 1a as a colorless oil, identified by comparison with an authentic sample available from previous studies (optical rotation, IR, and 1H NMR).12 Before biological evaluation, 1a was further purified by flash chromatography on silica gel (hexane-EtOAc 9:1 as eluant). The product so obtained (purity 96% by HPLC) was dissolved in DMSO and the solution was stored frozen until assaying.13 Cell Culture and Transfection. Human embryonic kidney 293 (HEK 293) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Life Technologies) supplemented with 10% fetal bovine serum (Life Technologies), 50 U/mL streptomycin and 50 U/mL penicillin (Life Technologies) at 37 °C in a humidified incubator with 5% CO2 in Nunc flasks. Before transfection, the cells were detached from the flasks and replated on Petri dishes (Becton, Dickinson Co.) coated with 1 μg/mL poly-D-lysine. A standard calcium phosphate precipitation method was used for cell transfection. Cells were cotransfected with rat GABAAR subunits cloned in pCMV vectors with plasmid that encoded human CD4. DNA samples of rat GABAA receptor subunits were added in the following proportions: α1:β21:1, α1:β2:γ2L1:1:3, α1:β2:δ1:1:4 (μg). The day after transfection, the cells were detached and replated on coverslips (Carl Roth) coated with poly-D-lysine. CD4 magnetic binding beads (Dynabeads CD4, Invitrogen Dynal) were used for the detection of transfected cells. Electrophysiological experiments were performed 48−72 h after transfection. Electrophysiological Recording. Currents were recorded using an Axopatch 200B amplifier (Molecular Devices) in the whole-cell configuration of the patch-clamp technique at a holding potential of −40 mV. Typically, stable recordings were available for 30−40 min. Signals were low-pass filtered at 3 kHz and sampled at 10 kHz with a Digidata 1440 (Molecular Devices) acquisition card. For acquisition and signal analysis pClamp 10.2 software (Molecular Devices) was used. The intrapipette solution contained (in mM): 137 CsCl, 1 CaCl2, 2 MgCl2, 11 EGTA, 2 ATP, and 10 HEPES, pH 7.2 with CsOH. The external solution contained (in mM): 137 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 20 glucose, and 10 HEPES, pH 7.2 with NaOH. For GABA applications, the multibarrel perfusion system (Bio-Logic RSC-200) was used. Pipettes with resistance of 2−3.5 MΩ were used for the whole-cell recordings. Cells for which rundown of measured currents was greater than 25% during the entire recording period, or access resistance was larger than 10 MΩ were rejected. All experiments were performed at room temperature (22−24 °C) and the external solution was persistently perfused at the rate of ca. 1 mL/min to prevent excessive accumulation of toxin. Data Analysis. For data analysis, pClamp 10.2 software was used. The onset kinetics of current responses was fitted with an
be excluded that upon prolonged washing, beyond the time of a typical whole-cell recording, a reversal of 1a activity might have been observed. Despite the structural similarity between falcarinol (1a) and falcarindiol (1b), the effect of 1a on GABAergic current was significantly larger. Indeed, at 0.3 μM, 1b exerted no effect on GABAergic current responses in hippocampal neurons,16 while 1a at 0.1 μM significantly enhanced responses mediated by α1β2γ2L receptors. Higher concentrations of 1b (1 μM) significantly enhanced the current responses, but the effect was still weaker than that observed for 1a at the same concentrations.16 In addition, the impact of 1a on the time course of GABA-evoked currents (except for fading) was larger than in the case of 1b. Substantial changes in the modulatory actions of 1a and 1b on GABAARs exist, underscoring the specificity of their interactions with these receptors. These findings mirror the differences observed between the closely related isomeric monoterpene ketones α-thujone and dihydroumbellulone, which, despite their close similarity, showed different potencies for GABAARs.31 Given the heterogeneity of neuronal GABAARs, it is difficult to predict the overall physiological effect of falcarinol (1a), for which the activity is basically hormesic. At low concentrations, enhancement of synaptic currents (mediated predominantly by α1β2γ2L GABAARs) would be promoted, but in the case of intense barrages of synaptic activity, this effect would be attenuated by the use-dependent block, which would become predominant at higher concentrations. Tonic conductance mediated by α1β2 receptors would be reduced, but tonic inhibition associated with δ subunit-containing receptors (expected to be by far predominant over that related to α1β2 receptors) would be markedly increased. Thus, although the present data cover only a few representative types of GABAARs in the CNS, it does not seem unreasonable to assume that falcarinol (1a), overall, upregulates GABAergic activity, a conclusion backed up by the observation that in animal experiments a sedative and not a convulsant activity was reported for falcarinol (1a).32 This type of activity might be one associated with a high human dietary intake of the compound. Moreover, since GABAergic interneurons typically show much larger tonic currents with respect to the principal cells,33 our data related to the receptors containing the δ subunit suggest that tonic inhibition in interneurons is affected by falcarinol to a larger extent than that in the pyramidal neurons. In addition, GABAARs are important targets of neuroactive pesticides,34 and the potent insecticidal action of falcarinol (1a)35 might be related to the GABAergic block associated with the higher intake expected in herbivorous insects.
■
EXPERIMENTAL SECTION General Experimental Procedures. Silica gel 60 (70−230 mesh) used for gravity column chromatography was purchased from Macherey-Nagel. Flash chromatography was carried out on a Biotage SP-1 apparatus. Optical rotations (CHCl3) were measured at 589 nm on a JASCO P2000 polarimeter, and IR spectra were determined with a FT-IR Thermo Nicolet equipment. 1H (500 MHz) NMR spectra were measured on a Varian INOVA spectrometer, and high-resolution ESIMS were obtained on a LTQ OrbitrapXL (Thermo Scientific) mass spectrometer. Purities were established on a Knauer HPLC apparatus, equipped with a Luna 5 μm C18(2), 150 mm × 2.0 mm (Phenomenex, Torrance, CA) column, protected with a C18−Security Guard cartridge, 4 mm × 2.0 mm (Phenomenex), F
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
exponential function: y(t) = A(1 − exp(-t/τ)).The mean rise time 10−90% was calculated as trise= τ·ln9. Current fading was assessed as the difference between the maximal current (peak current) amplitude and current value at the end of agonist application, divided by peak current: fading = (Imax − Iend)/Imax. The impact of falcarinol (1a) was determined from the comparison between control and test recordings obtained from the same cell. Data are expressed as means ± SEM. For data comparison, the paired Student’s t test was used. Differences were considered significant when p < 0.05.
■
(17) Christensen, L. P.; Brandt, K. J. Pharm. Biomed. Anal. 2006, 41, 683−689. (18) Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. Nat. Rev. Drug Discovery 2011, 10, 307−317. (19) Olsen, R. W.; Sieghart, W. Neuropharmacology 2009, 56, 141− 148. (20) Sieghart, W.; Sperk, G. Curr. Top. Med. Chem. 2002, 2, 795− 816. (21) Siegel, E.; Steinmann, M. E. J. Biol. Chem. 2012, 287, 40224− 40231. (22) Nusser, Z.; Sieghart, W.; Somogyi, P. J. Neurosci. 1998, 18, 1693−1703. (23) Farrant, M.; Nusser, Z. Nat. Rev. Neurosci. 2005, 6, 215−229. (24) Martirolo, O. Phytoalimurgia Pedemontana; Blu Edizioni, Peveragno (CN), Italy, 2001; pp 65−69. (25) Mortensen, M.; Smart, T. G. J. Physiol. 2006, 577, 841−856. (26) Mody, I.; Pearce, R. A. Trends Neurosci. 2004, 27, 569−575. (27) Mozrzymas, J. W. J. Physiol. 2010, 588, 1381−1382. (28) Bianchi, M. T.; Macdonald, R. L. Neuropharmacology 2001, 41, 737−744. (29) Bianchi, M. T.; Macdonald, R. L. J. Neurosci. 2003, 23, 10934− 10943. (30) Szczot, M.; Kisiel, M.; Czyzewska, M. M.; Mozrzymas, J. W. J. Neurosci. 2014, 34, 3193−3209. (31) Szczot, M.; Czyzewska, M. M.; Appendino, G.; Mozrzymas, J. W. J. Nat. Prod. 2012, 75, 622−629. (32) Duan, X.-C.; Wang, Y.-Z; Ju, J.; Zhang, J.-R.; Xia, L.-Z. Anhui Yiyao 2008, 12, 1−3. (33) Semyanov, A.; Walker, M. C.; Kullmann, D. M. Nat. Neurosci. 2003, 6, 484−90. (34) Casida, J. E.; Durkin, K. A. Annu. Rev. Entomol. 2013, 58, 99− 117. (35) Eckenbach, U.; Lampman, R. L.; Seigler, D. S.; Ebinger, J.; Novak, R. J. J. Chem. Ecol. 1999, 25, 1885−1893. (36) Christensen, L. P.; Kreutzmann, S. J. Sep. Sci. 2007, 30, 483− 490.
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Tel: +48 71 7841550. Fax: +48 71 7841399. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by Polish National Science Centre grant “Preludium” No UMO-2013/11/N/NZ3/00972. We are grateful to Dado Luciano and Renzo Salvo (Gruppo Micologico Cebano, Ceva, Italy) for the collection and identification of B. bulbocastanum and for the plant photograph used in the graphical Table of Contents.
■
REFERENCES
(1) Christensen, L. P. Recent Pat. Food Nutr. Agric. 2011, 3, 64−67. (2) Holzinger, F.; Chenot, J. F. Evid.-Based Complement. Alternat. Med.20112011, Article ID 382789, 9 pages. (3) Christensen, L. P.; Brandt, K. In Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet; Crozier, A., Clifford, M. N., Ashihara, H., Eds.; Blackwell, Oxford, U.K., 2009; Chapter 5, pp 147−173. (4) Tanaka, S.; Ikeshiro, Y.; Tabata, M.; Knoshima, M. Arzneimittelforschung 1977, 27, 2039−2045. (5) Matsunaga, H.; Katano, M.; Yamamoto, H.; Fujito, H.; Mori, M.; Takata, K. Chem. Pharm. Bull. 1990, 38, 3480−3482. (6) Purup, S.; Larsen, E.; Christensen, L. P. J. Agric. Food Chem. 2009, 57, 8290−8296. (7) Leonti, M.; Casu, L.; Raduner, S.; Cottiglia, F.; Floris, C.; Altmann, K. H.; Gertsch, J. Biochem. Pharmacol. 2010, 79, 1815−1826. (8) Ohnuma, T.; Nakayama, S.; Anan, E.; Nishiyama, T.; Ogura, K.; Hiratsuka, A. Toxicol. Appl. Pharmacol. 2010, 244, 27−36. (9) Hausen, B. M.; Bröhan, J.; König, W. A.; Faasch, H.; Hahn, H.; Bruhn, G. Contact Dermatitis 1987, 17, 1−9. (10) Paulsen, E.; Christensen, L. P.; Andersen, K. E. Contact Dermatitis 2010, 62, 201−209. (11) Due to an easier purification procedure, it has been found more convenient to run the thia-Michael reaction with the odorless and lipophilic thioldodecanethiol rather than with cysteamine, the thiol used in ref 7. A comprehensive structure−reactivity study on the reaction of various classes of polyacetylenic compounds will be reported in due course. (12) Wang, Z. J.; Nie, B. M.; Chen, H. Z.; Lu, Y. Chem. Biol. Interact. 2006, 159, 58−64. (13) Appendino, G.; Pollastro, F.; Verotta, L.; Ballero, M.; Romano, A.; Wyrembek, P.; Szczuraszek, K.; Mozrzymas, J. W.; TaglialatelaScafati, O. J. Nat. Prod. 2009, 72, 962−965. (14) Wyrembek, P.; Lebida, K.; Mercik, K.; Szczuraszek, K.; Szczot, M.; Pollastro, F.; Appendino, G.; Mozrzymas, J. W. Br. J. Pharmacol. 2010, 160, 1302−1315. (15) Wyrembek, P.; Negri, R.; Appendino, G.; Mozrzymas, J. W. Eur. J. Pharmacol. 2012, 683, 35−42. (16) Wyrembek, P.; Negri, R.; Kaczor, P.; Czyżewska, M.; Appendino, G.; Mozrzymas, J. W. J. Nat. Prod. 2012, 75, 610−616. G
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Dietary Acetylenic Oxylipin Falcarinol Differentially Modulates GABAA Receptors Marta Magdalena Czyzewska,*,† Lukasz Chrobok,† Alan Kania,† Magdalena Jatczak,† Federica Pollastro,‡ Giovanni Appendino,‡ and Jerzy Wladyslaw Mozrzymas†,§ †
Laboratory of Neuroscience, Department of Biophysics, Wroclaw Medical University, ul. Chalubinskiego 3, 50-368 Wroclaw, Poland Dipartimento di Scienze del Farmaco, Università del Piemonte Orientale, Via Bovio 6, 28100 Novara, Italy § Department of Animal Molecular Physiology, Institute of Zoology, Wroclaw University, Cybulskiego 30, 50-205 Wroclaw, Poland ‡
ABSTRACT: The dietary oxylipins falcarinol (1a) and falcarindiol (1b) trap thiols by direct nucleophilic addition to their diyne system, but despite this, only falcarinol (1a) is a reversible agonist of cannabinoid receptors, providing a rationale for comparing their activity also on other neuronal targets. Because GABAA receptors (GABAARs) are exquisitely sensitive to polyacetylenic oxylipins in terms of either potentiation (falcarindiol, 1b) or inhibition (oenanthotoxin, 2a), the activity of 1a was investigated on synaptic (α1β2γ2L) and extrasynaptic (α1β2δ and α1β2) subtypes of GABAARs. Falcarinol (1a) significantly enhanced the amplitude of currents mediated by α1β2γ2L receptors, but this effect was associated with a use-dependent block. Conversely, α1β2 receptors were inhibited without any sign of use-dependent block for the entire range of concentrations tested (1−10 μM). Interestingly, responses mediated by α1β2δ receptors, showing no or very little macroscopic desensitization, were strongly potentiated by 1a, exhibiting a fading reminiscent of macroscopic desensitization. When compared to the activity of falcarindiol (1b), falcarinol (1a) showed a higher affinity for GABAARs and, overall, a substantially different profile of pharmacological action. Taken together, the present data support the view that modulation of GABAARs might underlie the insecticidal and sedative activity of falcarinol (1a).
F
alcarinol (1a) is a major dietary polyacetylenic oxylipin. It is common in food plants from the family Apiaceae [carrot (Daucus carota L.), celery (Apium graveolens subsp. dulce (Mill.) Schübl. & G. Martens), parsnip (Pastinaca sativa L.), coriander (Coriandrum sativum L.) and parsley (Petroselinum crispum (Mill.) Fuss)], and also occurs in medicinal plants from the family Araliaceae, like ginseng (Panax ginseng L.) and ivy (Hedera helix L.).1,2 Falcarinol, a bitter compound, was long considered an undesirable dietary toxin, as cogently documented by its alternative name “carotatoxin”.3 However, over the past three decades, this view has been revised substantially in the wake of the discovery of the antinociceptive4 and potential anticancer5 properties of this compound. Current interest in falcarinol (1a) was triggered by distinct lines of research that exemplify the pleiotropic profile of bioactivity of this polyacetylene. Thus, while carotenoid-rich plants like carrots have been related epidemiologically to a preventive activity on cancer, clinical trials of β-carotene as a chemopreventive agent have failed, and it has been suggested that the epidemiological correlations that spurred these studies would better fit the biological profile of falcarinol (1a), an inhibitor of cancer resistance proteins as well as a selective cytotoxin,6 rather than that of β-carotene.3 On the other hand, falcarinol (1a) was also identified as an inverse agonist of the cannabinoid receptor CB1, capable of selectively alkylating the anandamide binding site, increasing the expression of chemokines in the skin, and overall inducing pro-allergic effects in the © XXXX American Chemical Society and American Society of Pharmacognosy
skin.7 As an electrophilic agent, falcarinol (1a) can trap thiol groups in a covalent way8 and is considered the causative agent of contact dermatitis from ivy,9 a growing problem in several European countries as well as in the U.S.10 The reaction of falcarinol (1a) and falcarindiol (1b) with thiols proceeds by direct addition to the triple bond and leads to a mixture of diastereomeric vinyl sulfides (Scheme 1).8,11 This electrophilic behavior underlies the CB1 reverse-agonist activity of 1a7 as well as its pro-allergenic properties, and might well represent Received: August 1, 2014
A
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
nature, it might cross the blood-brain barrier and enter the brain, where, apart from neuritogenic activity12 and the interaction with CB1,7 it could also interact with GABAA receptors (GABAARs), which are exquisitely sensitive to polyacetylenic compounds.13−15 Thus, oenanthotoxin (2a) and dihydrooenanthotoxin (2b), the poisonous acetylenic constituents of Oenanthe species, inhibit GABAergic responses in a dose-dependent manner (EC50 value in the low micromolar range) by a complex mechanism comprising allosteric modulation of agonist binding and desensensitization, with some contribution of the open channel block.13,14 Minor structural changes greatly modify the modulatory effects on GABAARs, with allosteric effects and open channel block being strongly dependent on polarity,15 as shown by the observation that the related polyacetylene falcarindiol (1b) enhances rather than inhibits current responses to low GABA concentrations (Scheme 1).16 Falcarindiol (1b) and falcarinol (1a) react in the same way with thiols,8,11 but show different biological profiles, since
Scheme 1. Reaction of Falcarinol (1a, R = H) and Falcarindiol (1b, R = OH) with Thiols
the hallmark of its biological profile. In humans, falcarinol is rapidly absorbed after oral administration.3 Due to its lipophilic
Figure 1. Current responses mediated by α1β2γ2L receptors are affected by falcarinol (1a) by a complex mechanism comprising initial upregulation and progressive accumulation of the use-dependent inhibition. (A) Typical superimposed current traces evoked by 3 μM GABA under control conditions (black line) and in the presence of 1 μM 1a (gray line, first response to GABA in the presence of 1a). (B) Superimposed traces of currents evoked by 3 μM GABA in control conditions (black line) and in the presence of 3 μM 1a (gray line, additionally sweep numbering is provided). Note the use-dependent current inhibition. (C) Statistics of relative values of current amplitude for considered range of 1a concentrations. Gray bins present the saturation of the 1a effect (observed upon recordings of several current responses), while white bins present the relative value of amplitude for the first sweep. (D) Normalized and superimposed initial phases of currents evoked by 3 μM GABA in control conditions 1a (black line) or in the presence of 1 μM 1a (dark gray line) and 10 μM 1a (gray line). (E) Statistics of relative values of current rise time for all tested 1a concentrations. (F) Normalized and superimposed currents evoked by 3 μM GABA under control conditions (black line) or in the presence of 3 μM 1a. Deepening of current fading in the presence of 1a is shown by comparison of the 1st, 5th, and 6th sweeps (note the saturation of this effect in 5th and 6th sweeps). (G) Statistics of the extent of fading recorded under control conditions (black column) and in the presence of 1a within the considered concentration range. Gray bins present mean values of the saturated 1a effect, whereas white bins present fading of the first sweep. Asterisks indicate statistical significance (p < 0.05). B
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 2. Falcarinol (1a) downregulates the amplitude of currents mediated by α1β2 receptors and affects the receptor kinetics. (A) Superimposed current responses under control conditions (black line) and in the presence of 1 μM of 1a. (B) Statistics of relative values of current amplitudes measured in the presence of the range considered of 1a concentrations. (C) Normalized and superimposed traces of the initial phases of current responses under control conditions (black line) and in the presence of 0.1 μM 1a (gray line) and 5 μM 1a (dark gray line). (D) Statistics of relative rise time of currents in the presence of all 1a concentrations tested. (E) Normalized and superimposed current traces recorded under control conditions (black line) and in the presence of 5 μM 1a (dark gray line). (F) Statistics of fading calculated as (Imax − Iend)/Imax. The black column presents the mean value for fading under control conditions. Asterisks indicate statistical significance (p < 0.05).
−40 mV. The mean value of current amplitude mediated by α1β2γ2L receptors was −643.82 ± 8.77 pA (n = 58). Low concentration of 1a (0.1−1 μM) significantly increased the amplitude of evoked currentsthe relative values were the following: 1.18 ± 0.05 (p < 0.05), 1.29 ± 0.06 (p < 0.05), and 1.41 ± 0.09 (p < 0 0.05) for 0.1, 0.3, and 1 μM 1a, respectively (Figure 1A,C). The first GABA application (first sweep) in the presence of 3 μM 1a had a markedly larger amplitude than in control conditions (Figure 1B,C) but at 5 μM a reversal of current enhancement (for the first sweep) was observed (Figure 1C). Interestingly, in addition to this biphasicity of falcarinol (1a) concentration-dependence, an apparently usedependent effect of this compound was revealed, because in consecutive sweeps, a profound and progressive reduction of current amplitude was seen (Figure 1B,C). The saturation of this use-dependent effect was achieved after 5 to 6 sweeps. Notably, as shown in Figure 1F, after the first pulse in the presence of 1a, subsequent current responses were characterized by increased current fading (see also below), further suggesting a use-dependent block. The relative amplitude values determined for the first sweep were 1.76 ± 0.25, n = 6 (p < 0.05) and 1.17 ± 0.21, n = 8 (p > 0.05) for 3 and 5 μM 1a, respectively. After saturation of the effect of 1a on the current amplitude, for responses mediated by α1β2γ2L, the relative values were 0.68 ± 0.12, n = 8 (p < 0.05) and 0.40 ± 0.05, n = 8 (p < 0.05) for 3 and 5 μM 1a, respectively. In the presence of a high concentration of 1a (10 μM), the current amplitude showed a tendency to decrease already at the first sweep (Figure 1C), whereas at subsequent ones, a profound current reduction was observed, reaching saturation at the second response (for the first sweep, the relative amplitude value was
falcarindiol is completely inactive in assays of CB1 binding7 and also displays lower cytotoxic activity than falcarinol.17 These differences highlight the role of the molecular milieu for covalent ligands 18 and provide a rationale for investigating the action of falcarinol on GABAA receptors. GABAARs are responsible for inhibition in the adult mammalian central nervous system and have a pentameric structure, with as many as 19 subunits identified so far, namely, α1 − α6, β1 − β3, γ1 − γ3, ε, θ, ρ1 − ρ3, δ, and π.19 Subunit composition has been shown to determine their functional and pharmacological properties.20,21 In particular, two major forms of GABAergic inhibition have been described, referred to as phasic (synaptic) and tonic (extrasynaptic) and mediated by different receptor types, for instance, α1β2γ2L and α1β2δ (with some contribution from α1β2), respectively.22,23 In the present report, the differential modulation of these two forms of inhibition by falcarinol (1a) is described.
■
RESULTS AND DISCUSSION Falcarinol (1a) was isolated in a very large yield (ca. 1% on dry weight basis) from the edible tubers of earth chestnut (Bunium bulbocastanum L., Apiaceae), a plant popular in Europe before the introduction of potatoes from the New World,24 with the seeds used as a spice (black cumin) in Asia. To assess the impact of 1a on GABAergic currents, responses mediated by receptors implicated in tonic and phasic inhibition were investigated. First, the effect was examined on the most abundant GABAA receptor in the central nervous system (CNS), α1β2γ2L, which plays a crucial role in phasic inhibition. For this purpose, whole-cell responses to 8 s application of 3 μM GABA were recorded at the holding membrane voltage of C
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
Figure 3. Falcarinol (1a) upregulates the amplitude of currents mediated by α1β2δ receptors and affects their kinetics. (A) Superimposed traces of currents evoked by 3 μM GABA alone (black line) and in the presence of 3 μM of 1a. (B) Statistics of relative values of current amplitude. (C) Normalized and superimposed traces of current onset measured in control conditions (black line) in the presence of 0.1 μM of 1a (gray line) and in the presence of 5 μM of 1a (dark gray line). (D) Statistics of relative values of current onset (rise time). (E) Normalized and superimposed of current traces evoked by 3 μM GABA alone (black line) or in the presence of 5 μM 1a (gray line). (F) Statistics of fading calculated as (Imax − Iend)/ Imax. The black column presents the mean value for fading under control conditions. Asterisks indicate the statistical significance (p < 0.05).
conductance with respect to α1β2γ2L receptors, and it has been proposed that these receptors may play a role in mediating the tonic GABAergic conductance.25 Thus, the impact of 1a on this type of GABAARs was also investigated. The mean amplitude of currents mediated by α1β2 receptors (evoked by 3 μM GABA) was −576.40 ± 10.9 pA (n = 5). Interestingly, in contrast to α1β2γ2L receptors, 1a, at all tested concentrations, reduced current amplitude in a dose-dependent manner (relative values: 0.86 ± 0.05, n = 7, p < 0.05 for 0.1 μM; 0.81 ± 0.03, n = 8, p < 0.05 for 0.3 μM; 0.50 ± 0.03, n = 14, p < 0.05 for 1 μM; 0.20 ± 0.01, n = 8, p = 0.05 for 3 μM; 0.30 ± 0.05, n = 10, p < 0.05 for 5 μM; 0.06 ± 0.01, n = 6, p < 0.05 for 10 μM), and this effect was not associated with any apparent use-dependence (Figure 2A,B). Under control conditions, the time course of currents mediated by α1β2 receptors was characterized by a faster rise time (121.89 ± 9.66 ms, n = 45) with respect to α1β2γ2L receptors, and this difference was significant (p < 0.05). However, α1β2 receptor-mediated currents were characterized by a profound fading (Figure 2A,E), reflected by a large value of the fading coefficient 0.67 ± 0.03 (n = 34), which was significantly larger (p < 0.05) in comparison to that determined for α1β2γ2L receptors (0.35 ± 0.02, n = 40). Falcarinol (1a) showed a concentration-dependent effect on the current onset, accelerating it at low (relative rise time 0.70 ± 0.05, n = 7, p < 0.05 for 0.1 μM 1a and 0.82 ± 0.05, n = 8, p < 0.05 for 0.3 μM 1a, Figure 2C,D) and slowing it down at high concentrations (relative rise time 2.16 ± 0.38, n = 10, p < 0.05 for 5 μM 1a; 1.52 ± 0.12, n = 4, p < 0.05 for 10 μM 1a, Figure 2C,D). At 1 and 3 μM, the effect on the current onset was not significant, although at 3 μM, a clear trend to slow it down was seen (1.02 ± 0.15, n = 11, p > 0.05 for 1 μM 1a; 1.43 ± 0.36, n = 5, p > 0.05 for 3 μM 1a, Figure 2D). The effect of falcarinol on
0.52 ± 0.18, n = 6, p > 0.05; for the second sweep, the mean relative value was 0.08 ± 0.03, n = 6, p < 0.05; the third sweep had an amplitude undistinguishable from the second one). The effect of 1a on the amplitude of currents mediated by the α1β2γ2L receptor was reversible, ruling out a dock-and-lock covalent mechanism of activity mediated by the alkylation of cysteine groups. The time course of the α1β2γ2L receptor-mediated currents was described by determination of the current rise time and fading parameters (see Experimental Section). Under control conditions, the mean values for the rise time and fading coefficient were 164.99 ± 9.57 ms (n = 50) and 0.35 ± 0.02 (n = 40), respectively. Falcarinol (1a), at low and medium concentrations (0.1−5 μM), had no effect or showed a slightly prolonged rise time of evoked currents (the relative values for 0.1, 0.3, 1, 3, 5 μM of 1a respectively: 0.90 ± 0.05, n = 8, p > 0.05; 1.25 ± 0.10, n = 7, p < 0.05; 1.21 ± 0.12, n = 18, p > 0.05; 0.99 ± 0.16, n = 6, p > 0.05; 0.79 ± 0.11, n = 8, p > 0.05). Falcarinol (1a) at high concentration (10 μM) clearly reduced the rise time of current responses (relative value: 0.19 ± 0.07, n = 3, p < 0.05) (Figure 1D,E). Acceleration of current onset was accompanied by increased fading except for the lowest concentrations of 1a (under control conditions 0.35 ± 0.02, n = 40; mean values for effect at its saturation: 0.21 ± 0.06, n = 9, p < 0.05 for 0.1 μM 1a; 0.32 ± 0.08, n = 8, p > 0.05 for 0.3 μM; 0.50 ± 0.04, n = 19, p < 0.05 for 1 μM; 0.85 ± 0.02, n = 8, p < 0.05 for 3 μM; 0.85 ± 0.03, n = 8, p < 0.05 for 5 μM; 0.97 ± 0.03, n = 6, p < 0.05 for 10 μM 1a; mean values for the first sweep: 0.78 ± 0.02, n = 7, p < 0.05 for 3 μM 1a, 0.59 ± 0.05, n = 8, p < 0.05 for 5 μM 1a, 0.86 ± 0.04, n = 5, p < 0.05 for 10 μM 1a, Figure 1F,G). It is known that the α1 and β2 subunits may coassemble and form α1β2 receptors characterized by higher affinity and lower D
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
strongly dependent on the identity of the subunit completing the pentamer comprising two α1 and two β2 subunits. For the α1β2γ2L receptors, which play a crucial role in phasic inhibition and in being also the most abundant GABAAR in the CNS, low falcarinol (1a) concentrations enhanced current amplitude elicited by nonsaturating GABA, similar to previous observation for falcarindiol (1a) in primary hippocampal neurons.16 This effect may suggest an upregulation of agonist binding rate, but such a mechanism would predict also acceleration of the current onset that has not been observed in the present experiments. Most likely, 1a may affect not only binding, but also gating (e.g., the opening/closing rate), although this possibility would require a detailed receptor kinetic study beyond the scope of this current investigation. Interestingly, high concentrations of 1a induced a use-dependent inhibition of α1β2γ2L receptors, strongly suggesting the open channel block. Observation of cumulative amplitude reduction, shortening of current rise time correlated with amplitude decrease and increased current fading were all compatible with the open channel block mechanism. Interestingly, the use-dependent blockade observed for compound 1a closely resembled the effect described for oenanthotoxin (2a) on current responses elicited by low GABA.14 In addition, it was found that oenanthotoxin analogues characterized with a low hydrophobicity also induced a use-dependent block.15 Thus, the open channel block might be a peculiar feature of several polyacetylenes acting on GABAA receptors containing a γ2L subunit. Taken altogether, the present data indicated that in the case of α1β2γ2L receptors, falcarinol (1a) may exert an allosteric modulation affecting probably both binding and gating properties and such actions are accompanied by an open channel block. In contrast to α1β2γ2L GABAARs, α1β2 receptors were inhibited by 1a with no sign of an open channel block. A relatively potent amplitude inhibition is associated with a biphasic effect on the rise time and a weak impact on current fading (Figure 2). Again, such effects appear incompatible with a pure modulation of receptor binding properties but the precise mechanism of action of falcarinol (1a) on these receptors requires further studies. In contrast to α 1 β 2 GABAARs, α1β2δ receptors, which are believed to be more representative for mediating tonic currents, were potentiated strongly by medium and high doses of 1a. No signs of usedependent inhibition were observed, although the trend of growth of relative increase in current amplitude appeared to reverse at high concentrations of 1a (Figure 3). Interestingly, amplitude increase in the presence of falcarinol is accompanied by appearance of marked current fading, which was absent under control conditions. This finding is intriguing as some δ subunit-containing receptors have been described as “nondesensitizing”. However, the lack of macroscopic desensitization does not provide sufficient evidence for the absence of microscopic desensitization (i.e., that the desensitized conformation is not available for the receptor).27,28 Indeed, in the presence of 3α,21-dihydroxy-5α-pregnan-20-one (THDOC), responses of δ subunit-containing receptors greatly increased and pronounced fading appeared revealing the receptor ability to desensitize.29 Similar observations lacking macroscopic desensitization could be revealed by a high potency agonist has been described for α1β1γ2L receptors with mutation at the agonist binding site (α1F64).30 In contrast to α1β2γ2L and α1β2, the effect of falcarinol (1a) on α1β2δ receptors appeared irreversible. The nature of this interaction is not clear. It cannot
current fading was to moderately reduce the fading coefficient in a dose-dependent manner while at low concentrations of 1a (0.1−1 μM) there was no effect (0.65 ± 0.06, n = 7, p > 0.05 for 0.1 μM 1a; 0.60 ± 0.03, n = 8, p > 0.05 for 0.3 μM 1a; 0.64 ± 0.02, n = 12, p > 0.05 for 1 μM 1a). At higher doses (3,10 μM), however, this effect became significant (0.42 ± 0.04, n = 6, p < 0.05 for 3 μM; 0.54 ± 0.08, n = 10, p > 0.05 for 5 μM 1a; 0.45 ± 0.02, n = 5, p < 0.05 for 10 μM 1a, Figure 2E,F). As already mentioned, α1β2 receptors have been shown to be putatively extrasynaptic.25 However, it is widely accepted that δ subunit-containing receptors play a crucial role in mediating tonic inhibition.23,26 Therefore, it was decided to extend the current studies to an extrasynaptic receptor containing δ subunit. To minimize the number of changes in the subunit composition, α1β2δ receptors were chosen as a representative extrasynaptic GABAAR. Under control conditions, application of 3 μM GABA elicited responses with an average amplitude of −174.34 ± 29.51 pA (n = 49). The 10−90% rise time for these currents in control conditions was 141.19 ± 4.47 ms (n = 36). As expected, currents mediated by α1β2δ receptors showed minimal fading, consistent with weak macroscopic desensitization. Indeed, the fading coefficient was close to zero (Figure 3E) as differences between Imax and Iend values (methods) were at the level of current noise. Low concentrations of falcarinol (1a) (0.1−1 μM) either had no effect or slightly increased the amplitude of evoked currents (relative values: 1.04 ± 0.05, n = 7, p > 0.05 for 0.1 μM 1a; 1.46 ± 0.07, n = 15, p < 0.05 for 0.3 μM 1a; 1.26 ± 0.08, n = 7, p > 0.05 for 1 μM, Figure 3B). Interestingly, at medium and high concentrations, falcarinol profoundly increased the amplitude of current responses (relative values: 3.20 ± 0.58, n = 9, p < 0.05 for 3 μM 1a; 8.31 ± 1.74, n = 7, p < 0.05 for 5 μM 1a; 5.68 ± 1.97, n = 4, p < 0.05 for 10 μM 1a, Figure 3A,B). In contrast to observations made for α1β2γ2L and α1β2 receptors, this effect was irreversible in the time scale of the recordings made (after tens of minutes of rinsing, only a slight washout of effect of 1a was observed). Falcarinol, at low concentration, except for 1 μM, slowed down the current onset and had no effect at medium and high concentrations (relative values for 10−90% rise time: 1.69 ± 0.27, n = 6, p < 0.05 for 0.1 μM 1a; 1.87 ± 0.18, n = 8, p < 0.05 for 0.3 μM 1a; 1.05 ± 0.05, n = 8, p > 0.05 for 1 μM 1a; 1.50 ± 0.15, n = 5, p > 0.05 for 3 μM 1a; 1.78 ± 0.33, n = 5, p > 0.05 for 5 μM 1a; 1.39 ± 0.13, n = 4, p > 0.05 for 10 μM 1a, Figure 3C,D). At low concentrations (0.1−1 μM), 1a had no effect on current fading, which remained at the detection level as under control conditions (Figure 3F). However, surprisingly, at higher concentrations, a strong upregulation of current amplitude (Figure 3E,F) was accompanied by the appearance of a marked fading (0.17 ± 0.05, n = 9, p < 0.05 for 3 μM 1a; 0.37 ± 0.07, n = 7, p < 0.05 for 5 μM 1a; 0.47 ± 0.04, n = 4, p < 0.05 for 10 μM 1a, Figure 3F). The major finding of this study was that falcarinol (1a) exerted differential effects on GABAA receptor subtypes involved in phasic and tonic forms of GABAergic inhibition. Strikingly, the effect on each GABAAR subtype investigated was characterized by a substantially different mechanism. Indeed, responses mediated by the α1β2γ2L receptors were potentiated with a concomitant use-dependent block, with the α1β2 receptors inhibited and the α1β2δ receptors irreversibly potentiated (within a recording period up to 40 min), with no sign of the use-dependent block. The present data thus have demonstrated that the effect of 1a on GABAA receptors is E
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
using the conditions developed by Christensen and Kreutzmann.36 Plant Material. Bulbs of Bunium bulbocastanum were collected around Ceva (Italy) in August 2010. The plant material was identified by Dado Luciano and Renzo Salvo, and a voucher specimen (number 03/10), is kept at the Dipartimento di Scienze del Farmaco, Università del Piemonte Orientale, Novara, Italy. Extraction and Isolation. Powdered dried tubers of Bunium bulbocastanum (27.6 g) were extracted with acetone (2 × 250 mL) to afford, after removal of the solvent, 600 mg of a brownish oil, that was purified by gravity-column chromatography on silica gel (25 mL) using petroleum ether-EtOAc (9:1) as eluant, to afford 280 mg (1%) of 1a as a colorless oil, identified by comparison with an authentic sample available from previous studies (optical rotation, IR, and 1H NMR).12 Before biological evaluation, 1a was further purified by flash chromatography on silica gel (hexane-EtOAc 9:1 as eluant). The product so obtained (purity 96% by HPLC) was dissolved in DMSO and the solution was stored frozen until assaying.13 Cell Culture and Transfection. Human embryonic kidney 293 (HEK 293) cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; Life Technologies) supplemented with 10% fetal bovine serum (Life Technologies), 50 U/mL streptomycin and 50 U/mL penicillin (Life Technologies) at 37 °C in a humidified incubator with 5% CO2 in Nunc flasks. Before transfection, the cells were detached from the flasks and replated on Petri dishes (Becton, Dickinson Co.) coated with 1 μg/mL poly-D-lysine. A standard calcium phosphate precipitation method was used for cell transfection. Cells were cotransfected with rat GABAAR subunits cloned in pCMV vectors with plasmid that encoded human CD4. DNA samples of rat GABAA receptor subunits were added in the following proportions: α1:β21:1, α1:β2:γ2L1:1:3, α1:β2:δ1:1:4 (μg). The day after transfection, the cells were detached and replated on coverslips (Carl Roth) coated with poly-D-lysine. CD4 magnetic binding beads (Dynabeads CD4, Invitrogen Dynal) were used for the detection of transfected cells. Electrophysiological experiments were performed 48−72 h after transfection. Electrophysiological Recording. Currents were recorded using an Axopatch 200B amplifier (Molecular Devices) in the whole-cell configuration of the patch-clamp technique at a holding potential of −40 mV. Typically, stable recordings were available for 30−40 min. Signals were low-pass filtered at 3 kHz and sampled at 10 kHz with a Digidata 1440 (Molecular Devices) acquisition card. For acquisition and signal analysis pClamp 10.2 software (Molecular Devices) was used. The intrapipette solution contained (in mM): 137 CsCl, 1 CaCl2, 2 MgCl2, 11 EGTA, 2 ATP, and 10 HEPES, pH 7.2 with CsOH. The external solution contained (in mM): 137 NaCl, 5 KCl, 2 CaCl2, 1 MgCl2, 20 glucose, and 10 HEPES, pH 7.2 with NaOH. For GABA applications, the multibarrel perfusion system (Bio-Logic RSC-200) was used. Pipettes with resistance of 2−3.5 MΩ were used for the whole-cell recordings. Cells for which rundown of measured currents was greater than 25% during the entire recording period, or access resistance was larger than 10 MΩ were rejected. All experiments were performed at room temperature (22−24 °C) and the external solution was persistently perfused at the rate of ca. 1 mL/min to prevent excessive accumulation of toxin. Data Analysis. For data analysis, pClamp 10.2 software was used. The onset kinetics of current responses was fitted with an
be excluded that upon prolonged washing, beyond the time of a typical whole-cell recording, a reversal of 1a activity might have been observed. Despite the structural similarity between falcarinol (1a) and falcarindiol (1b), the effect of 1a on GABAergic current was significantly larger. Indeed, at 0.3 μM, 1b exerted no effect on GABAergic current responses in hippocampal neurons,16 while 1a at 0.1 μM significantly enhanced responses mediated by α1β2γ2L receptors. Higher concentrations of 1b (1 μM) significantly enhanced the current responses, but the effect was still weaker than that observed for 1a at the same concentrations.16 In addition, the impact of 1a on the time course of GABA-evoked currents (except for fading) was larger than in the case of 1b. Substantial changes in the modulatory actions of 1a and 1b on GABAARs exist, underscoring the specificity of their interactions with these receptors. These findings mirror the differences observed between the closely related isomeric monoterpene ketones α-thujone and dihydroumbellulone, which, despite their close similarity, showed different potencies for GABAARs.31 Given the heterogeneity of neuronal GABAARs, it is difficult to predict the overall physiological effect of falcarinol (1a), for which the activity is basically hormesic. At low concentrations, enhancement of synaptic currents (mediated predominantly by α1β2γ2L GABAARs) would be promoted, but in the case of intense barrages of synaptic activity, this effect would be attenuated by the use-dependent block, which would become predominant at higher concentrations. Tonic conductance mediated by α1β2 receptors would be reduced, but tonic inhibition associated with δ subunit-containing receptors (expected to be by far predominant over that related to α1β2 receptors) would be markedly increased. Thus, although the present data cover only a few representative types of GABAARs in the CNS, it does not seem unreasonable to assume that falcarinol (1a), overall, upregulates GABAergic activity, a conclusion backed up by the observation that in animal experiments a sedative and not a convulsant activity was reported for falcarinol (1a).32 This type of activity might be one associated with a high human dietary intake of the compound. Moreover, since GABAergic interneurons typically show much larger tonic currents with respect to the principal cells,33 our data related to the receptors containing the δ subunit suggest that tonic inhibition in interneurons is affected by falcarinol to a larger extent than that in the pyramidal neurons. In addition, GABAARs are important targets of neuroactive pesticides,34 and the potent insecticidal action of falcarinol (1a)35 might be related to the GABAergic block associated with the higher intake expected in herbivorous insects.
■
EXPERIMENTAL SECTION General Experimental Procedures. Silica gel 60 (70−230 mesh) used for gravity column chromatography was purchased from Macherey-Nagel. Flash chromatography was carried out on a Biotage SP-1 apparatus. Optical rotations (CHCl3) were measured at 589 nm on a JASCO P2000 polarimeter, and IR spectra were determined with a FT-IR Thermo Nicolet equipment. 1H (500 MHz) NMR spectra were measured on a Varian INOVA spectrometer, and high-resolution ESIMS were obtained on a LTQ OrbitrapXL (Thermo Scientific) mass spectrometer. Purities were established on a Knauer HPLC apparatus, equipped with a Luna 5 μm C18(2), 150 mm × 2.0 mm (Phenomenex, Torrance, CA) column, protected with a C18−Security Guard cartridge, 4 mm × 2.0 mm (Phenomenex), F
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
Journal of Natural Products
Article
exponential function: y(t) = A(1 − exp(-t/τ)).The mean rise time 10−90% was calculated as trise= τ·ln9. Current fading was assessed as the difference between the maximal current (peak current) amplitude and current value at the end of agonist application, divided by peak current: fading = (Imax − Iend)/Imax. The impact of falcarinol (1a) was determined from the comparison between control and test recordings obtained from the same cell. Data are expressed as means ± SEM. For data comparison, the paired Student’s t test was used. Differences were considered significant when p < 0.05.
■
(17) Christensen, L. P.; Brandt, K. J. Pharm. Biomed. Anal. 2006, 41, 683−689. (18) Singh, J.; Petter, R. C.; Baillie, T. A.; Whitty, A. Nat. Rev. Drug Discovery 2011, 10, 307−317. (19) Olsen, R. W.; Sieghart, W. Neuropharmacology 2009, 56, 141− 148. (20) Sieghart, W.; Sperk, G. Curr. Top. Med. Chem. 2002, 2, 795− 816. (21) Siegel, E.; Steinmann, M. E. J. Biol. Chem. 2012, 287, 40224− 40231. (22) Nusser, Z.; Sieghart, W.; Somogyi, P. J. Neurosci. 1998, 18, 1693−1703. (23) Farrant, M.; Nusser, Z. Nat. Rev. Neurosci. 2005, 6, 215−229. (24) Martirolo, O. Phytoalimurgia Pedemontana; Blu Edizioni, Peveragno (CN), Italy, 2001; pp 65−69. (25) Mortensen, M.; Smart, T. G. J. Physiol. 2006, 577, 841−856. (26) Mody, I.; Pearce, R. A. Trends Neurosci. 2004, 27, 569−575. (27) Mozrzymas, J. W. J. Physiol. 2010, 588, 1381−1382. (28) Bianchi, M. T.; Macdonald, R. L. Neuropharmacology 2001, 41, 737−744. (29) Bianchi, M. T.; Macdonald, R. L. J. Neurosci. 2003, 23, 10934− 10943. (30) Szczot, M.; Kisiel, M.; Czyzewska, M. M.; Mozrzymas, J. W. J. Neurosci. 2014, 34, 3193−3209. (31) Szczot, M.; Czyzewska, M. M.; Appendino, G.; Mozrzymas, J. W. J. Nat. Prod. 2012, 75, 622−629. (32) Duan, X.-C.; Wang, Y.-Z; Ju, J.; Zhang, J.-R.; Xia, L.-Z. Anhui Yiyao 2008, 12, 1−3. (33) Semyanov, A.; Walker, M. C.; Kullmann, D. M. Nat. Neurosci. 2003, 6, 484−90. (34) Casida, J. E.; Durkin, K. A. Annu. Rev. Entomol. 2013, 58, 99− 117. (35) Eckenbach, U.; Lampman, R. L.; Seigler, D. S.; Ebinger, J.; Novak, R. J. J. Chem. Ecol. 1999, 25, 1885−1893. (36) Christensen, L. P.; Kreutzmann, S. J. Sep. Sci. 2007, 30, 483− 490.
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Tel: +48 71 7841550. Fax: +48 71 7841399. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by Polish National Science Centre grant “Preludium” No UMO-2013/11/N/NZ3/00972. We are grateful to Dado Luciano and Renzo Salvo (Gruppo Micologico Cebano, Ceva, Italy) for the collection and identification of B. bulbocastanum and for the plant photograph used in the graphical Table of Contents.
■
REFERENCES
(1) Christensen, L. P. Recent Pat. Food Nutr. Agric. 2011, 3, 64−67. (2) Holzinger, F.; Chenot, J. F. Evid.-Based Complement. Alternat. Med.20112011, Article ID 382789, 9 pages. (3) Christensen, L. P.; Brandt, K. In Plant Secondary Metabolites: Occurrence, Structure and Role in the Human Diet; Crozier, A., Clifford, M. N., Ashihara, H., Eds.; Blackwell, Oxford, U.K., 2009; Chapter 5, pp 147−173. (4) Tanaka, S.; Ikeshiro, Y.; Tabata, M.; Knoshima, M. Arzneimittelforschung 1977, 27, 2039−2045. (5) Matsunaga, H.; Katano, M.; Yamamoto, H.; Fujito, H.; Mori, M.; Takata, K. Chem. Pharm. Bull. 1990, 38, 3480−3482. (6) Purup, S.; Larsen, E.; Christensen, L. P. J. Agric. Food Chem. 2009, 57, 8290−8296. (7) Leonti, M.; Casu, L.; Raduner, S.; Cottiglia, F.; Floris, C.; Altmann, K. H.; Gertsch, J. Biochem. Pharmacol. 2010, 79, 1815−1826. (8) Ohnuma, T.; Nakayama, S.; Anan, E.; Nishiyama, T.; Ogura, K.; Hiratsuka, A. Toxicol. Appl. Pharmacol. 2010, 244, 27−36. (9) Hausen, B. M.; Bröhan, J.; König, W. A.; Faasch, H.; Hahn, H.; Bruhn, G. Contact Dermatitis 1987, 17, 1−9. (10) Paulsen, E.; Christensen, L. P.; Andersen, K. E. Contact Dermatitis 2010, 62, 201−209. (11) Due to an easier purification procedure, it has been found more convenient to run the thia-Michael reaction with the odorless and lipophilic thioldodecanethiol rather than with cysteamine, the thiol used in ref 7. A comprehensive structure−reactivity study on the reaction of various classes of polyacetylenic compounds will be reported in due course. (12) Wang, Z. J.; Nie, B. M.; Chen, H. Z.; Lu, Y. Chem. Biol. Interact. 2006, 159, 58−64. (13) Appendino, G.; Pollastro, F.; Verotta, L.; Ballero, M.; Romano, A.; Wyrembek, P.; Szczuraszek, K.; Mozrzymas, J. W.; TaglialatelaScafati, O. J. Nat. Prod. 2009, 72, 962−965. (14) Wyrembek, P.; Lebida, K.; Mercik, K.; Szczuraszek, K.; Szczot, M.; Pollastro, F.; Appendino, G.; Mozrzymas, J. W. Br. J. Pharmacol. 2010, 160, 1302−1315. (15) Wyrembek, P.; Negri, R.; Appendino, G.; Mozrzymas, J. W. Eur. J. Pharmacol. 2012, 683, 35−42. (16) Wyrembek, P.; Negri, R.; Kaczor, P.; Czyżewska, M.; Appendino, G.; Mozrzymas, J. W. J. Nat. Prod. 2012, 75, 610−616. G
dx.doi.org/10.1021/np500615j | J. Nat. Prod. XXXX, XXX, XXX−XXX
