Neurochem Res DOI 10.1007/s11064-014-1466-0

ORIGINAL PAPER

Non-competitive Inhibition of Nicotinic Acetylcholine Receptors by Ladybird Beetle Alkaloids Ron L. Leong • Hong Xing • Jean-Claude Braekman William R. Kem

•

Received: 17 May 2014 / Revised: 15 October 2014 / Accepted: 25 October 2014 Ó Springer Science+Business Media New York 2014

Abstract Ladybird beetles (Family Coccinellidae) secrete an alkaloid rich venom from their leg joints that protects them from predators. Coccinellines, the major venom constituents, are alkaloids composed of three fused piperidine rings that share a common nitrogen atom. Although many coccinellines have been isolated and chemically characterized, their pharmacological properties are essentially unknown. Using radioligand binding and functional assays we investigated the actions of several coccinellines on skeletal muscle and a7 nicotinic acetylcholine receptors (nAChRs). The alkaloids were shown to displace the specific binding of tritiated piperidyl-N-(1-(2-thienyl)cyclohexyl)3,4-piperidine ([3H]-TCP), which has been shown to bind deep within the ion channel of the electric fish (Torpedo) muscle nAChR. The stereoisomers precoccinelline and hippodamine (whose nitrogens are predicted to be ionized at physiological pH) and their respective analogs N-methylprecoccinelline and N-methyl-hippodamine (whose quaternary nitrogens are permanently charged) displayed similar IC50s for inhibition of [3H]-TCP binding. However, the corresponding precoccinelline and hippodamine N-oxides, coccinelline and convergine (which have an electronegative oxygen bonded to an electropositive nitrogen) displayed

significantly higher binding IC50s. Finally, exochomine, a dimeric coccinelline containing the hippodamine structure, displayed the highest IC50 (lowest affinity) for displacing specific [3H]-TCP binding. The presence of a desensitizing concentration (10-3 M) of carbachol (CCh) had little or no effect on the affinity of the Torpedo nAChR for the three coccinellines tested. High concentrations of the coccinellid alkaloids did not affect binding of [3H]-cytisine to Torpedo receptor ACh binding sites. Inhibition of the alpha7 nAChR with pre-equilibrated precoccinelline was insurmountable with respect to ACh concentration. We conclude that the coccinellines bind to one or more allosteric sites rather than to the ACh binding sites, and inhibit nAChR responses to ACh through a non-competitive mechanism. Future chemical and pharmacological investigations of other ladybird beetle alkaloids are likely to reveal other interesting alkaloids affecting ligand-gated receptors. Keywords Alpha7 nicotinic acetylcholine receptor  Torpedo muscle nicotinic receptor  Xenopus oocyte

Special Issue: In Honor of Lynn Wecker.

Abbreviations ACh Acetylcholine CCh Carbamylcholine nAChR Nicotinic acetylcholine receptor NCA Non-competitive antagonist [3H]-TCP [Piperidyl-3,4-3H(N)]-N-(1-(2thienyl)cyclohexyl)-3,4-piperidine

R. L. Leong  H. Xing  W. R. Kem (&) Department of Pharmacology and Therapeutics, University of Florida, Gainesville, FL 32610, USA e-mail: [email protected]

Introduction

J.-C. Braekman Laboratory of Bioorganic Chemistry, Faculty of Sciences, University of Brussels, 1050 Brussels, Belgium

Until recently the design of nAChR receptor targeted drugs has mainly focused on the development of ligands that bind

Ron L. Leong and Hong Xing have contributed equally to this study.

123

Neurochem Res

to the ACh (orthosteric) binding site and thereby activate, desensitize or inhibit the targeted receptor. However, other allosteric sites may also be useful targets for modulating nAChR sensitive processes, including nicotine dependence, cognition and inflammation. Allosteric modulators can enhance or diminish the amplitude of the normal pattern of cholinergic signaling, rather than producing chronic alterations in nAChR function such as maintained stimulation, desensitization or inhibition, as generated by orthosteric drugs. For instance, allosteric sites within the nAChR ion channel can be targeted to limit the action of drugs of abuse on the chemical reward pathway in the mid-brain. Mecamylamine, once used as an antihypertensive drug due to its preferential blockade of autonomic ganglionic nAChR channels, enhances the effectiveness of nicotine as a smoking cessation drug [1]. Memantine, a use-dependent non-competitive antagonist (NCA) of NMDA type glutamate receptor channels, is currently administered to Alzheimer’s disease patients [2]. The antidepressant bupropion seems to exert some of its smoking cessation actions as an NCA at high affinity brain nAChRs [3]. General anesthetics also exert some of their action as NCAs [4]. A variety of nAChR NCAs have been found in animal venoms and the pharmacological properties of some of these have been investigated. Some notable examples are the following toxins: solenopsins (2,6-disubstituted piperidines) isolated from fire ants [5], tetraponerines (tricyclic alkaloids containing two ring nitrogens) isolated from pseudomyrmicine ants [6–8], soft coral diterpenes [9, 10], and the amphibian histrionicotoxins [11] and decahydroquinolines [12]. Ladybird beetles (Family Coccinellidae, (Fig. 1), commonly referred to as ‘‘lady bugs’’ in the United States, produce a variety of compounds that serve as a chemical defence against predators [13–15]. Although only a few of the[6,000 described species have been examined so far, at least 50 compounds have been discovered [16, 17]. One group of alkaloids, the coccinellines, contains a 2-methylperhydro-9bazanaphalene ring system (Fig. 2). These compounds contain a nitrogen atom that is part of each of three fused piperidine rings. Two of the fused rings are co-planar and capable of coexisting in ‘‘boat’’ conformations, and the third ring is distinctly out of this plane and is predicted to be in a ‘‘chair’’ conformation (Fig. 3). Almost all of these compounds were isolated and chemically characterized in the 1970s by the Brussels group [13–15, 18]. A number of laboratory syntheses have been reported; the initial syntheses of racemic compounds [19, 20] were followed by several asymmetric syntheses [21–23]. The basic pathway of coccinelline biosynthesis was also elucidated [24]. In spite of these elegant chemical investigations, no pharmacological studies of coccinellid compounds have been reported. Here we examine the interaction of several

123

Fig. 1 European ladybird beetle (Coccinella punctata). Many ladybird beetles are brightly colored and have distinctive spot patterns to warn potential predators that they have a bitter taste and are poisonous

of these alkaloids with skeletal muscle and alpha7 nAChRs, using radioligand and two functional assay methods, and show that these compounds act as NCAs at these receptors.

Materials and Methods Materials Five of the seven coccinellid alkaloids (coccinelline, precoccinelline, hippodamine, convergine and exochomine) were isolated and purified from collections of several hundred whole beetles. Hippodamine and convergine were isolated from Hippodamia convergens, coccinelline and precoccinelline were isolated from Coccinella punctata, and exochomine was isolated from Exochomus pustulatus. N-methylprecoccinelline and N-methylhippodamine were prepared by treatment of precoccinelline and hippodamine, respectively, with methyliodide in acetonitrile. All seven compounds were shown to be homogeneous by thin-layer chromatographic, mass spectrometric and NMR spectroscopic determinations. The radioligands [piperidyl-3,4-3H(N)]-N-(1-(2-thienyl)cyclohexyl)-3,4-piperidine ([3H]-TCP, approximately 45 Ci/mmol) and [3H]-Cytisine (approximately 55 Ci/ mmol) were obtained from Perkin Elmer Life Sciences Products, Inc. (Boston, MA). and stored at -20 °C. Carbamylcholine chloride (CCh), acetylcholine chloride (ACh), polyethylenimine and proadifen were purchased from Sigma Chemical Co. (St. Louis, MO). Fetal bovine serum and penicillin/streptomycin were obtained from Media Tech, Inc. (Herndon, VA), Dulbecco’s modified Eagle’s medium was purchased from Fisher Scientific. The Blue Dye Kit sensing membrane potential was purchased from Molecular Devices (Sunnyvale, CA).

Neurochem Res Fig. 2 Two-dimensional structures of the coccinellid compounds whose nAChR blocking properties were measured in this study. The three compounds in the upper tier are stereoisomeric analogs of the three compounds below them. Exochomine is approximately twice as large as the six other coccinellines and contains a hippodamine moiety

Precoccinelline .HCI

N-Methyl-Precoccinelline iodide

Coccinelline

Hippodamine .HCI

N-Methyl-Hippodamine iodide

Convergine

H3C

N

N

CH3

Precoccinelline

Hippodamine

Fig. 3 Stereostructures of coccinelline and hippodamine, showing their similar three-dimensional structures

Frozen Pacific electric ray (Torpedo californica) muscle was obtained from Aquatic Research Consultants (San Pedro, CA). Human rhabdomyosarcoma (TE671) cells were obtained from the American Type Culture Collection (Manassas, VA). Mature female frogs (Xenopus laevis) containing oocytes were obtained from Nasco (Fort Atkinson, WI). Radioligand Binding Assays Displacement of specific binding of [3H]-TCP, a structural and functional analog of phencyclidine (PCP), was used to determine the relative potencies of the compounds for binding within the Torpedo neuromuscular nAChR ion channel [25]. Membranes were prepared according to Eldefrawi et al. [26]. Each concentration of coccinellid alkaloid to be tested was dissolved in binding saline (150 mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2 and 50 mM Tris Buffer, pH 7.4), containing 2 mg/ml bovine serum albumin to minimize non-specific binding. To each culture tube a final concentration of 50 lg of Torpedo membrane protein was added. Phencyclidine (final concentration 1 lM) was added to the [3H]-TCP (final concentration 5 nM) containing tubes used to

Exochomine

measure nonspecific binding. The order of addition was binding saline, coccinellid alkaloid, Torpedo membrane and finally the radioligand solution (all in saline, with final volume of 350 ll). Each tube was then vortexed and the entire rack of tubes was placed in an ice-containing pan and gently shaken at room temperature for 4 h. Quadruplicate measurements were made for each condition. Whatman GF/C filters were presoaked in 0.5 % polyethylenimine for 45 min prior to vacuum filtration of the membranes (Brandel Harvester) with ice-cold binding saline to quickly remove unbound radioligand. The recovered filters were placed into scintillation vials and then placed in a Beckman LS 7000 liquid scintillation system and counted for 5 min. Displacement curves and IC50 estimates were generated using Prism software. Since the reported Ki (196 nM, [27]) for [3H]-TCP binding to the Torpedo nAChR is 399 higher than the [3H]-TCP concentration employed in our experiments, our IC50 estimates for the compounds are also reasonable estimates of their Kis (Using the Cheng-Prusoff relation [28], Ki = 1.03 X IC50). [3H]-Cytisine displacement assays to assess coccinellid alkaloid binding to the ACh binding sites of the Torpedo muscle nAChR were carried out essentially as previously described for rat brain nAChRs [29], except that 3 nM [3H]-cytisine was used and the 3 h incubation was done at room temperature. Proadifen (200 lM final concentration) was added to favor the desensitized state of the Torpedo nAChR, which displays higher affinity for [3H]-cytisine (30).

Functional Analysis of Coccinellid Alkaloid Inhibition of Muscle nAChRs TE671 cells expressing the human fetal muscle nAChR subtype were cultured as previously described [25]. The TE671 were seeded at a density of approximately 5 9 104

123

Neurochem Res

cells/well into 96-well flat-bottom black-walled culture plates precoated with 50 ll per well of 50 lg/mL poly-Dlysine (70–150 kDa) and allowed to grow to near confluence overnight in 75 ll of culture medium. The kinetics of membrane depolarization after rapid addition of eleven different concentrations of ACh alone or in the presence of one of six concentrations of the coccinellid alkaloid were measured with a 96-well FlexStation robotic spectrofluorimeter (Molecular Devices) as previously described [25]. Membrane potential-sensing Blue dye (Molecular Devices) was used to measure changes in the membrane potential of The fluorescent dye solution was prepared by dissolving one vial from the Blue Dye Kit in 36 mL Hanks’ balanced salt solution (HBSS) supplemented with 20 mM Hepes (pH 7.4), hereafter referred to as HBSS/Hepes. The 96-well cell culture plates were equilibrated to room temperature for 10 min, then the cells were washed once with 100 ll HBSS/Hepes before adding 30 ll of the membrane potential dye solution to each well. The cells were incubated with the dye for 45 min before ACh addition and recording. Serial dilutions of the coccinellid compound dissolved in methanol for concentration–response analysis were prepared in 96-well V-bottom plates. After evaporation of the methanol, the compound in each well was redissolved in HBSS/Hepes [25]. Changes in membrane potential resulting from sudden addition of ACh or KCl were measured in the FlexStation. Readings was taken at 1.44 s intervals for 200 s, a total of 139 readings per well. The first 16 s were used as a basal reading. At 17 s 30 ll of a particular concentration of ACh was rapidly and simultaneously added to eight wells by the robotic instrument to measure agonist activity. A 30 ll pulse of isotonic KCl saline producing a final concentration of 40 mM KCl was added to identical wells with cells not exposed to ACh or coccinellid alkaloid, to serve as a depolarizing calibrant and to correct for differences in dye loading and cell counts between experiments. Peak depolarizing response (measured as fluorescence, F) was initially normalized to the KCl calibrant response: F = (FMax,Compound-FBasal)/(FMax,Calibrant-FBasal). These F values were then expressed relative to the maximum response generated by ACh alone. Normalized F values were then fitted to a four parameter logistic equation with variable slope using Prism software (GraphPad Software, San Diego, CA), for fitting sigmoidal concentration– response curves, to estimate the insect alkaloid concentration that produces 50 % inhibition of ACh evoked depolarization (IC50) and maximal degree of depolarization for each ACh concentration-activation curve. To avoid or at least minimize any desensitizing effect of either compound, we decided to test the effect of simultaneous addition of precoccinelline or coccinelline on the ACh responses. It is possible that the inhibitory potencies of

123

these compounds might be underestimated with this protocol, relative to the usual protocol where the putative antagonist is pre-equilibrated with the cell. Functional Analysis of Precoccinelline Inhibition of Alpha7 nAChRs To assess whether the coccinelline alkaloids have a similar actions on another nAChR subtype we tested the effects of three concentrations (10, 30 and 60 lM) of precoccinelline on ACh responses of human alpha7 receptors expressed in Xenopus oocytes using the two microelectrode voltage clamp method as previously described [29]. Initially the ability of each coccinelline concentration to stimulate alpha7 nAChRs was assessed, but no agonist actions were detected (results not shown). We then assessed the ability of precoccinelline to inhibit ACh (0.1, 0.3, 1, 5 and 15 mM concentrations) evoked currents. At this stage in our study, only small amounts of this alkaloid were still available and samples of the other compounds were depleted. During an experiment each of the three concentrations of precoccinelline was perfused on the oocyte for 3 min before it was then stimulated with one of two selected ACh concentrations also containing the same precoccinelline concentration, The current response to the alkaloid plus ACh combination was expressed relative to the mean current response obtained by administering the same concentration of ACh 5 min before and 5 min after the test response. The 5 min washing period was sufficiently long to permit full recovery from either the control (ACh only) or the combined alkaloid-ACh exposure. Then the next higher concentration of precoccinelline was tested and finally the highest concentration, each preceded and followed by the same ACh concentration administered alone. The same oocyte was tested at two ACh concentrations, starting with the lower concentration.

Results Inhibition of [3H]-TCP Binding to Torpedo Membrane Muscle nAChRs Each of the seven coccinellid alkaloids was tested several times for ability to displace the specific binding of [3H]TCP, an analog of PCP (Fig. 4). Both TCP and PCP are known to bind to the so-called chlorpromazine binding site within the nAChR ion channel. The stereoisomeric compounds hippodamine and precoccinelline displayed very similar low lM IC50s in this assay. The N-methylated analogs of these compounds, N-methylhippodamine and Nmethylpre-coccinelline, respectively, also displayed similar IC50s, that were not significantly different from their

Neurochem Res

% of Control

125 100

125

A

75

75

50

50

50

25

25

25

0

0

0

-25

-25

-25

-8

-7

-6

-5

-4

% of Control

125 100

-8

-7

-6

-5

-4

-3

125

100

75

75

75

50

50

50

25

25

25

0

0

0

-25

-25

-25

-6

-5

-4

-3

-8

Log [Hippodamine], M 125

-7

-7

-6

-5

-4

-6

-5

-4

Log [Coccinelline], M 125

E

100

-7

-8

Log [N-Methyl-Precoccinelline], M

D

-8

C

100

75

log [Precoccinelline], M

% of Contro

125

B

100

-3

Log [N-Methyl-Hippodamine], M

F

-7

-6

-5

-4

Log [Convergine], M

G

100 75 50 25 0 -25

-7

-6

-5

-4

-3

-2

Log [Exochomine], M

Fig. 4 [3H]-TCP displacement binding curves for the interaction of the seven coccinellid compounds with Torpedo muscle nAChRs. The compound IC50s, Hill slopes and number of experiments are found in Table 1, which is based on the same experiments. In each experiment, data for a given concentration was always obtained in quadruplicate.

Table 1 Relative potencies of the lady bird beetle alkaloids for inhibition of specific binding of [3H]-TCP in Torpedo membranes (from experimental data shown in Fig. 4)

a

The IC50 of hippodamine relative to its IC50 in the presence of carbachol (CCh) was statistically different (P = 0.03)

The average number of counts (cpms) for total specific binding of [3H]-TCP was approximately 300–400 and the non-specific binding was approximately 50–100 cpms (counting time per sample was 5 min)

Compound

IC50 (lM)

Hill slope

# Experiments

Precoccinelline

3.00 ± 1.0

3.3 ± 0.6

4

N-methyl-precoccinelline

4.80 ± 0.77

1.4 ± 0.1

3

N-methyl-precoccinelline ? CCh

3.38 ± 0.62

1.0 ± 0.2

3

Coccinelline

6.03 ± 0.73

1.5 ± 0.3

5

Hippodamine

1.30 ± 0.20

1.9 ± 0.4

5

Hippodamine ? CCh

2.18 ± 0.18a

1.2 ± 0.1

3

N-methyl-hippodamine

2.52 ± 0.78

1.8 ± 0.2

4

N-methyl-hippodamine ? CCh

4.36 ± 1.3

0.8 ± 0.2

2

Convergine

9.03 ± 0.58

0.9 ± 0.1

4

Exochomine

71.7 ± 8.4

2.1 ± 0.6

3

tertiary amine homologs (Table 1). In contrast, both Noxide compounds, convergine and coccinelline, displayed significantly greater IC50s (lower affinity for the TCP site) relative to the parent compounds. Finally, the dimeric alkaloid exochomine displayed a significantly lower

affinity for displacing [3H]-TCP binding to the Torpedo receptor compared with the other six compounds. [3H]-TCP displacement IC50s for three compounds were also determined in the presence of 1 mM CCh to assess whether the desensitized form of the receptor displays a

123

Neurochem Res Table 2 Inhibition of [3H]-cytisine binding at Torpedo muscle nAChRs Compound

% Contol specific binding

P value*

ACh

32.1 ± 1.9

\0.001

Precoccinelline

109 ± 14

[0.05

N-Methyl-precoccinelline

93.4 ± 7.0

[0.05

Coccinelline

124 ± 15

[0.05

Hippodamine

115 ± 12

[0.05

N-Methyl-hippodamine

97.8 ± 11

[0.05

Convergine Exochomine

126 ± 11 81.1 ± 11

[0.05 [0.05

None of the coccinelline compounds, at concentrations (30 lM) greatly exceeding their IC50s for inhibiting the binding of [3H]-TCP to the PCP site, significantly inhibited cytisine binding to the ACh binding sites. The average counts (cpms) for total binding of [3H]cytisine was approximately 5,000 and the non-specific binding was approximately 150 cpms (5 min count time per sample) * One-way ANOVA using Dunnett’s multiple comparison test with control specific binding

higher affinity for the coccinelline. The only statistically significant effect of CCh upon [3H]-TCP displacement IC50 was found for hippodamine (0.03 \ P); no significant differences were found for N-methylhippodamine (0.22 \ P) and N-methylprecoccinelline (0.22 \ P). [3H]-cytisine displacement assays were done to test whether any of the seven alkaloids was able to bind to the ACh binding sites on the Torpedo receptor. None of the compounds at 30 lM concentration affected specific [3H]cytisine binding (Table 2). Only the 100 lM ACh control was able to significantly reduce the cytisine binding. Its inhibition was partial, probably because no acetylcholinesterase inhibitor was included in the incubation solution (the Torpedo electric organ contains very large amounts of acetylcholinesterase). Inhibition of nAChR Function The ability of two of the most abundant compounds, precoccinelline and coccinelline, to inhibit the ACh- induced depolarization of TE671 cells is shown in Fig. 5. This transformed cell line expresses the fetal human nAChR subtype, which differs only from the adult neuromuscular receptor in possessing a c subunit in place of the e subunit present in the adult form. Neither compound depolarized the cells when given alone (results not shown). However, when different concentrations of the alkaloids were coadministered with different concentrations of ACh, the depolarizing responses to ACh were inhibited at all ACh concentrations to a similar extent, demonstrating an unsurmountable inhibition that is characteristic of noncompetitive antagonism. The IC50s for inhibition of the

123

ACh response were 14 and 50 lM, respectively, for precoccinelline and coccinelline. We also assessed whether one of the alkaloids, precoccinelline, inhibited the activation of the human a7 nAChR expressed in the Xenopus oocyte, using the two electrode voltage clamp method to record current responses [29]. The precoccinelline was preincubated with each oocyte for 3 min prior to measuring the response to ACh in the presence of the same concentration of precoccinelline. When administered alone, 1 mM ACh produced approximately 90 % of the maximal current response. Inhibition increased with increasing precoccinelline concentration to a similar extent at each ACh concentration (Fig. 6). Thus, the inhibitory effect of precoccinelline on alpha7 receptor function also is consistent with a non-competitive mechanism of action.

Discussion Our most extensive experiments measured the relative potencies of the various ladybird beetle alkaloids to inhibit the specific binding of [3H]–TCP to the Torpedo muscle type nAChR (Fig. 4; Table 1). The most potent inhibitors were hippodamine and precoccinelline. The pKas of the nitrogen atoms in these two compounds are estimated to be approximately 10, based on quinolizidine (pKa 10.1), whose nitrogen atom is in a very similar environment [31]. Therefore, these two compounds would be [99 % ionized at the physiological pH of our experiments. Since the permanently cationic N-methyl analogs of these compounds, N-methylhippodamine and N-methylprecoccinelline, also displayed very similar IC50s for TCP displacement as hippodamine and precoccinelline (Table 1), it can be inferred that the cationic forms of the coccinellines are capable of high affinity binding to the TCP binding site. On the other hand, the two N-oxides (convergine and coccinelline) that have N ? O- dipolar substituents displayed significantly higher IC50 values, indicating that the N-oxide group has only a small detrimental effect on interaction with the TCP site. The data for the N-oxide forms indicates that a formal net positive charge is not essential for non-competitive AChR inhibition. Since only one of the IC50s for three compounds (hippodamine, N-methylhippodamine and N-methylprecoccinelline) assayed in the presence of a 1 mM desensitizing concentration of carbamylcholine was statistically different from the IC50 measured in the absence of this agonist, it is anticipated that these compounds will not display a large use-dependent blockade. We assessed the possible binding of the coccinellid alkaloids to the ACh binding sites on the Torpedo muscle

Neurochem Res %Max Response 96.8±2.08

EC 50 (μM)

n

0.70±0.07

6

10μM Precoccinelline

70.4±2.84

0.59±0.11

4

20μM Precoccinelline

48.0±3.55

0.62±0.01

6

50μM Precoccinelline

37.3±3.87

0.58±0.16

4

100μM Precoccinelline 120

22.4±0.77

0.62±0.03

4

ACh

B

120 % of Max ACh Response

A

% of Control

100 80 60 40 20 0 -20 -10

-9

-8

-7

-6

-5

-4

100

60 40 20

log [ACh], M

Α Ch

120

EC50 (μM)

n

98.3±2.44 0.53±0.06

9

10μM Coccinelline 85.4±8.09 0.40±0.04

7

20μM Coccinelline 76.1±4.61 0.32±0.06

7

50μM Coccinelline 49.7±5.45 0.36±0.08

7

100μM Coccinelline 21.1±1.82 0.33±0.095 5

% of Control

100 80 60 40 20 0 -20 -10

-9

-8

-7

-6

-5

-4

-5.0 -4.5 -4.0 Log [Precoccinelline], M

-3.5

D

120 % of Max ACh Response

%Max Response

C

n 4

80

0 -5.5

-3

Hillslope IC50(μM) 2.58±0.55 14.3±0.81

-3

log [ACh], M

100

Hillslope IC50(μM) 1.86±0.35 49.6±5.87

n 4

80 60 40 20 0 -5.5

-5.0 -4.5 -4.0 Log [Coccinelline], M

-3.5

Fig. 5 Inhibition of TE671 cell muscle nAChR response to ACh. a and b, precoccinelline; c and d, coccinelline. Left figures a and c are ACh concentration–response curves; Right figures b and d display the

dependence of peak ACh response on alkaloid concentration (obtained from the left side figure), and were curve-fitted using GraphPad Prism software

nAChR, which is homologous with the fetal human muscle subtype expressed in the TE671 cell system. [3H]–cytisine avidly binds to this site of the Torpedo receptor, particularly in the presence of the desensitizing ligand proadifen [30]. Since relatively high concentrations of the seven alkaloids, which with the exception of exochomine were 5–109 higher than the IC50s for inhibiting [3H]-TCP binding (Table 1), were unable to significantly reduce specific binding of [3H]–cytisine, we can infer that the coccinellid alkaloids inhibit [3H]–TCP binding by interacting with another site than the ACh recognition site. Examination of the stereostructures of precoccinelline and hippodamine (Fig. 3) reveals that the ionizable nitrogen

group is embedded within a relatively rigid structure quite unlike the analogous nitrogens of ACh site ligands like ACh, nicotine, cytisine and other ligands that display high affinity binding to this site. Thus, it would seem unlikely on the basis of their structures, that the coccinellid alkaloids would fit within the aromatic box with significant affinity. The IC50s of the two compounds tested for inhibition of ACh responses in the TE671cells expressing the fetal human muscle receptor (Fig. 5) were significantly higher than their IC50s for inhibition of specific [3H]-TCP binding. Since the binding and TE671 functional experiments were done on fetal type muscle nAChRs from two different vertebrates (fish and human), these IC50 differences could

123

Neurochem Res ACh (mM)

10μM Precoccinelline 30μM Precoccinelline 60μM Precoccinelline n≥3

A

120

100

% of ACh Response

% of ACh Response

120

80 60 40 20 0 -20

-5

-4

-3

-2

-1

IC50 (μM)

0.1 0.3 1 5 15

B

100

36.9 46.9 36.5 26.4 42.6

n≥3

80 60 40 20 0 -20 -5.25

Log [ACh],M

-5.00

-4.75

-4.50

-4.25

-4.00

Log [Precoccinelline],M

Fig. 6 Precoccinelline inhibition of human a7 nAChRs expressed in Xenopus oocytes, measured by the two electrode voltage-clamp method. At all three alkaloid concentrations of alkaloid tested, the the oocyte was perfused with the compound for 3 min before the ACh solution, containing the same concentration of the compound, was

added. a Responses to ACh in the presence of the three different concentrations of precoccinelline; b precoccinelline inhibition of response at a particular ACh concentration. The curves and their computer fitted IC50s were obtained from the same data as in part a using GraphPad Prism software

be at least partly a species difference. Another factor that might have contributed to these differences in IC50 is that the radioligand binding experiments involved membrane fragments, while the functional experiments involved whole cells in which the receptors are still in a fully native state. A third possibility is that the TE671 cell functional IC50s were higher because the precoccinelline and coccinelline were added simultaneously with ACh and the maximum depolarizing response occurred within just a few seconds, whereas the radioligand binding experiments allowed full equilibration of the coccinelline and the [3H]TCP with the Torpedo receptor. To determine whether the coccinellines also act on other nAChRs in a similar manner we tested one of them on human alpha7 receptors expressed in the Xenopus oocyte expression system. As with the TE671 cells, precoccinelline failed to generate an agonist (in this case, inward currents) response when administered separately at 10, 30 and 60 lM concentrations. At this stage of our investigation we only had sufficient amounts of precoccinelline to fully assess whether their inhibition of the response to different concentrations of ACh was competitive or non-competitive. The data (Fig. 6) we obtained for precoccinelline are most consistent with a non-competitive mechanism, since the degree of inhibition by all three concentrations of precoccinelline was independent of the ACh concentration over a 150-fold range of ACh concentration. This initial study of just a few of the known coccinellines should be extended to other coccinellines and to other nAChR subtypes. There may be other novel potent defensive compounds that are produced by this family of insects. Certain nAChR subtypes may be more sensitive to some of

the compounds, as it is known that there are some significant differences in the ion channel lining, transmembrane segment M2 amino acid sequences between the various nAChRs [32, 33]. A few compounds from other sources have been reported to be rather selective NCAs of particular nAChRs [34, 35]. It is possible that the coccinellid alkaloids, besides affecting nAChRs, may also act as modulators of some other ion channels. Several nAChR NCAs have also been found to be NCAs at iontophoretic glutamate receptors, though their effective concentrations may differ considerably. Phencyclidine and MK801 are such compounds. Further pharmacological studies are desirable, as some NCAs may be of therapeutic value in limiting response to addictive drugs such as nicotine. The coccinellines are repellent to insect and vertebrate predators. It is not yet known if nAChRs are the main ecological targets for their actions. Nevertheless, due to the widespread occurrence of nAChRs in the animal kingdom and the importance of these receptors for central as well as peripheral nervous functions, including movement, feeding and other behaviours, it seems likely that at least some nAChRs are important ecological targets of the coccinellines.

123

References 1. Rose JE, Behm FM, Westman EC (2001) Acute effects of nicotine and mecamylamine on tobacco withdrawal symptoms, cigarette reward and ad lib smoking. Pharmacol Biochem Behav 68:187–197

Neurochem Res 2. Lipton SA (2006) Paradigm shift in neuroprotection by NMDA receptor blockade: memantine and beyond. Nat Rev Drug Discov 5:160–170 3. Arias HR (2009) Is the inhibition of nicotinic acetylcholine receptors by bupropion involved in its clinical actions? Intern J Biochem Cell Biol 41:2098–2108 4. Arias HR, Trudell JR, Kem WR, Blanton MP (2002) Unique general anesthetic binding sites within conformational states of the nicotinic acetylcholine receptor. Intern J Neurobiol 54:1–50 5. Aronstam RS, Edwards MW, Daly JW, Albuquerque EX (1988) Interactions of piperidine derivatives with the nicotinic cholinergic receptor complex from Torpedo electric organ. Neurochem Res 13:171–176 6. Braekman JC, Daloze D, Pasteels JM, Van Hecke P, Declercq JP, Sinnwell V, Francke W (1987) Tetraponerine-8, an alkaloidal contact poison in a Neoguinean pseudomyrmecine ant, Tetraponera sp. Z Naturforsch 42C:627–630 7. Kem WR, Wildeboer K, LeFrancois S, Raja M, Marzalec W, Braekman J-C (2004) Nicotinic receptor inhibition by Tetraponera ant alkaloids. Cell Mol Neurobiol 24:535–551 8. Rouchaud A, Braekman J-C (2009) Synthesis of new analogues of the tetraponerines. Eur J Org Chem 2009:2666–2674 9. Abramson SN, Culver P, Kline T, Li Y, Guest P, Gutman L, Taylor P (1988) Lophotoxin and related coral toxins covalently label the alpha subunit of the nicotinic acetylcholine receptor. J Biol Chem 263:18568–18573 10. Ferchmin PA, Pagan OR, Ulrich H, Szeto AC, Hann RM, Eterovic VA (2009) Actions of octacoral and tobacco cembranoids on nicotinic receptors. Toxicon 54:1174–1182 11. Aronstam RS, King CT, Albuquerque EX, Daly JW, Feigl DM (1985) Binding of [3H]perhydrohistrionicotoxin and [3H]phencyclidine to the nicotinc receptor-ion channel complex of Torpedo electroplax. Biochem Pharmacol 34:3037–3047 12. Daly JW, Nishizawa Y, Padgett WL, Tokuyama T, McCloskey PJ, Waykole L, Schultz AG, Aronstam RS (1991) Decahydroquinoline alkaloids: noncompetitive blockers for nicotinic acetylcholine receptor-channels in pheochromocytoma cells and Torpedo electroplax. Neurochem Res 16:1207–1212 13. Tursch B, Daloze D, DuPont M, Pasteels JM, Tricot M-C (1971) A defense alkaloid in a carnivorous beetle. Experientia 27:1381– 1382 14. Tursch B, Braekman JC, Daloze D (1976) Arthropod alkaloids. Experientia 32:401–407 15. Timmermans M, Braekman J-C, Daloze D, Pasteels JM, Merlin J, Declercq J-P (1992) Exochomine, a dimeric ladybird alkaloid, isolated from Exochomus quadripustulatus (Coleoptera: Coccinellidae). Tetrahedron Lett 33:1281–1284 16. King AG, Meinwald J (1996) Review of the defensive chemistry of coccinellids. Chem Rev 96:1105–1122 17. Giogi W, Vandenberg N (2014) Coccinellidae (lady beetles, ladybird beetles, ladybugs. Tree of life web project. http://tol. web.org 18. Lebrun B, Braekman JC, Daloze D (1999) Complete assignment of the 1H and 13C NMR spectra of the coccinellidae-defensive alkaloids myrrhine, precocccinelline and hippodamine, their N-oxides and the corresponding hydrochlorides. Magn Res Chem 37:60–64 19. Ayer WA, Furuichi K (1976) The total synthesis of coccinelline and precoccinelline. Can J Chem 54:1494–1495 20. Ayer WA, Dawe R, Eisner RA, Furuichi K (1976) A total synthesis of myrrhine, (±)-hippodamine, and (±)- convergine. Can J Chem 54:473–481 21. Stevens RV, Lee AWM (1979) On the stereochemistry of the Robinson-Schopf reaction. A stereospecific total synthesis of the

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

ladybug defense alkaloids precoccinelline and coccinelline. J Am Chem Soc 102:7032–7035 Mueller RH, Thompson ME, DiPardo RM (1984) Stereo- and regioselective total syntheses of the hydropyridyl[2,1,6-de]quinolizine ladybug defensive alkaloids. J Org Chem 49:2217–2231 Gerasyuto AI, Hsung RR (2006) Stereodivergent total syntheses of precoccinelline, hippodamine, coccinelline, and convergine. Org Lett 8:4899–4902 Laurent R, Braekman J-C, Daloze D, Pasteels JM (2002) In vitro production of adaline and coccinelline, two defensive alkaloids from ladybird beetles (coleoptera: coccinellidae). Insect Biochem Mol Biol 32:1017–1023 Arias HR, Xing H, MacDougall KMP, Blanton MP, Soti F, Kem WR (2009) Interaction of DMXBA (GTS-21) and its primary metabolite with agonist and allosteric sites on muscle nicotinic acetylcholine receptors. Brit J Pharmacol 157:320–330 [PMCID:2697798] Eldefrawi ME, Eldefrawi AT, Aronstam RS, Maleque MA, Warnick JE, Albuquerque EX (1980) [3H]Phencyclidine: a probe for the ionic channel of the nicotinic receptor. Proc Natl Acad Sci USA 77:7458–7462 Katz EJ, Cortes VI, Eldefrawi E, Eldefrawi AT (1997) Chorpyrifos, parathion, and their oxons bind to and desensitize a nicotinic acetylcholine receptor: relevance to their toxicities. Toxicol Appl Pharmacol 146:227–236 Cheng Y, Prusoff WH (1973) Relationship between the inhibition constant (Ki) and the concentration of inhibitor which causes 50 percent inhibition (IC50) of an enzymatic reaction. Biochem Pharmacol 22:3099–3108 Kem WR, Mahnir VM, Prokai L, Papke RM, Cao XF, LeFrancois S, Wildeboer K, Porter Papke J, Prokai-Tatrai K, Soti F (2004) Hydroxy metabolites of the Alzheimer’s drug candidate DMXBA (GTS-21): their interactions with brain nicotinic receptors, and brain penetration. Mol Pharmacol 65:56–67 Arias H, De Rosa MJ, Berge I, Feuerbach D, Bouzat C (2013) Differential pharmacological activity of JN403 between alpha7 and muscle nicotinic acetylcholine receptors. Biochemistry 52:8480–8488 Johnson CD, Jones RAY, Katritzky AR, Palmer CR, Schofield K, Wells RJ (1965) A re-examination of the stereochemistry of quinolizidine and the methylquinolizidines through measurement of their rate of quaternisation and those of the hexahydrojulolidines. J Chem Soc 1965:6797–6806 Zhorov BS, Brovtsyna NB, Gmiro VE, Lukomskaya NY, Serdyuk SE, Popapyeva NN, Magazanik LG, Kurenniy DE, Skok VI (1991) Dimensions of the ion channel in neuronal nicotinic acetylcholine receptor as estimated from analysis of conformation-activity relationships of open-channel blocking drugs. J Membr Biol 121:119–132 Sankararamakrishnan R, Adcock C, Sanson MSP (1996) The pore domain of the nicotinic acetylcholine receptor: molecular modeling, pore dimensions, and electrostatics. Biophys J 71:1659–1671 Papke RL, Burh JD, Francis MM, Choi KI, Thinschmidt JS, Horenstein NA (2005) The effects of subunit composition on the inhibition of nicotinic receptors by the amphipathic blocker 2,2,6,6-tetramethylpiperidin-4-yl heptanoate. Mol Pharmacol 67:1977–1990 Tsuneki H, You Y, Toyooka N, Kagawa S, Kobayashi S, Sasaoka T, Nemoto H, Kimura I, Dani JA (2004) Alkaloids indolizidine 235B0 , quinolizidine 1-epi-20171, and the tricyclic 205B are potent and selective noncompetitive inhibitors of nicotinic acetylcholine receptors. Mol Pharmacol 66:1061–1069l

123