Chemistry of southeast asian plants 2

Chapter The Hepaticae 3.1 Introduction Bryophytes are taxonomically placed between pteridophytes and algae. They are the simplest terrestrial plants and the most primitive higher plants. They are divided morphologically into three classes: Hepaticae (liverworts, 6000 species), Musci (mosses, 14000 species) and Anthocerotae (hornworts, 3000 species). Hepaticae is further divided into two subclasses: the Jungermanniidae and the Marchantiidae. These two subclasses together contain 61 families with about 6000 species found worldwide.111 Chemical investigations of liverworts (and the bryophytes in general) have been neglected for a long time, for two main reasons. First, they are tiny plants and tend to grow intermingled with other liverwort or moss species; therefore, it is difficult to collect them in adequate amounts as pure samples for chemical investigations. Second, their identification is difficult and they are considered to have no nutritional value for human. Started from 1980s, the development of superconducting magnets for NMR spectroscopy and the advances in computing technology have helped scientist to overcome the difficulties, by reducing the minimum amount of pure sample needed for analysis. Among the bryophytes, the phytochemistry of Hepaticae has been studied in greater detail because they possess oil bodies, while the other two classes lack such features. Up to 2001, more than 700 terpenoids and 220 aromatic compounds (excluding flavonoids) have been isolated from or detected in the Hepaticae, while the number of terpenoids found in Musci is signficantly lower.112 79 Monoterpenoids are responsible for the distinctive smell that many liverworts produce when crushed. Among approximately 60 monoterpenoids found in liverworts, α-pinene (164), β-pinene (165), and limonene (160) are the most abundant compounds.113 Structures of representative compounds are shown in Fig. 3-1 OH OH OH (156) Myrcene (157) Linalool (158) Nerol (159) Geraniol O OH (160) Limonene (161) Pulegone (162) β-Sabinene (163) α-Terpineol OH OH (164) α-Pinene (165) β-Pinene (166) Myrtenol (167) Borneol O (168) (+)-Camphor (169) Camphene Fig. 3-1. Monoterpenoids found in liverworts Compared to monoterpenoids, sesquiterpenoids are a lot more abundant in liverworts, and in greater diversity. More than 500 sesquiterpenoids, which belong to around sixty 80 types of carbon skeleton, have been isolated or detected in the Hepaticae so far.112-113 80% of these sesquiterpenoids are enantiomeric to those found in higher plants.114 The most widespread sesquiterpene in liverworts is ent-bicyclogermacrene (170), although the number of bicyclogermacranes isolated from liverworts is rather small. It is considered to be the biosynthetic precursor of many type of sesquiterpenoids, including aromadendranes, maalianes, aristolanes, and vitranes. The most common type of sesquiterpenoids in liverworts is the eudesmanes, of which ent-α-selinene (171) and ent-β-selinene (172) are the two most widespread compounds. The family of Frullaniaceae, which comprises more than 500 species, is a rich source of eudesmanolides, which contribute to about three-quarters of eudesmanes found in liverworts. Among them, (-)-frullanolide (173) and dihydrofrullanolide (174) are the most frequently encountered compounds.113 H (170) H (171) O (172) O O (173) O (174) The Panamanian liverwort Plagiochila moritziana produces four rare C-35 sesquiterpene lactones (sesquiterpene + diterpene), plagiospirolides A-D (175-178), and a rare C-30 sesquiterpene lactone (sesquiterpene + sesquiterpene), plagiospirolide 81 E (179). The C-35 sesquiterpene lactones could be the products of a Diels-Alder cycloaddition reaction between the dienophiles ent-diplophyllolide (180) and entdiplophylline (181) (both were also found in P. moritziana) and either the diterpene diene (182) (in case of plagiospirolides A & B) or (183) (in case of plagiospirolides C & D). For the C-30 lactone dimer, plagiospirolide E, the two suggested sesquiterpene moieties are (181) and an aromadendrene derivative (184).115-116 H H O O O H O H H R R H H (175) R = H (177) R = OH (176) R = H (178) R = OH O O O O R H H (180) (182) R = H (183) R = OH (181) H O O H O O + H H (179) (181) (184) 82 Aromadendranes comprise another widespread class of liverwort sesquiterpenoids. Ent-aromadendranes are present in many species of the Jungermanniales, while secoaromadendranes are found largely in Plagiochila species. Among the aromadendranes, anastreptene (185) and ent-spathulenol (186) are the most regularly isolated compounds, especially the latter which has been found in more than 80 species. Mylia taylorii was found to be a rich source of aromadendranes and secoaromadendranes.114 This species also produces dimeric sesquiterpenoids, bitaylorione (187), myltayloriones A (188) and B (189). The latter two are probably products of Diels-Alder reaction of mylione (190) and (191), an isomer of taylorione (192).117 O H H HO (185) O H (186) (187) O O O O (188) O (189) O O (188), (189) (192) (191) (190) 83 The biosynthetically interesting pinguisane sesquiterpenoids are found in liverworts only, and they are also restricted to a number of families, namely Lejeuneaceae, Trichocoleaceae, Ptilidiaceae, Porellaceae and Riccardiaceae.112 The carbon skeleton of pinguisanes does not conform to the biogenetic isoprene rule. Acutifolones A (193) and B (194), together with three dimeric pinguisane sesquiterpenoids, bisacutifolones A (195), B (196) and C (197) have been isolated from Porella acutifolia subsp. Tosana.118-119 Further modifications of the pinguisane skeleton lead to the norpinguisanes (198, 199)247 and pinguisanolides (200, 201),248 isolated from the liverworts Porella recurva (Taylor) Kuhnemann and Trocholejeunea sandvicensis, respectively. R O H O O MeOOC MeOOC O COOMe O O (193) O O OAc (198) R = H2 (199) R = O CHO CHO R O (195) R = β-OH (196) R = α-OH (197) R = H (194) O O O OH O O (200) (201) The biosynthesis pathway for pinguisone (203) from trans,cis-farnesol (202) has been proposed, based on labeling experiment in the cultured liverwort Aneura pinguis, involving a series of hydride and methyl shifts.120 84 OPP OPP OPP H+ H H (202) PPO O O O HO H (203) Fig. 3-2. Proposed biosynthetic pathway of pinguisone The main constituent of Saccogyna viticulosa is the alcohol saccogynol (204) with the zierane carbon skeleton.121 This type of sesquiterpene is very rare in nature, and zierone (205) is the only other known member of this group. H H HO H O (204) (205) 85 Liverworts are also rich sources of diterpenoids. More than 200 diterpenoids, representing about 20 carbon skeletons, have been found in the Hepaticae. Among them, clerodanes, labdanes, and kauranes are the most widepread types. Clerodanes are distributed in both higher plants and liverworts. They have a vast stereochemical and functional variability. More than half of the clerodanes found in liverworts were isolated from Jungermannia and Schistochila species. The liverwort Anastrepta orcadensis produced two bitter clerodanes anastreptin (206) and orcadensin (207).122 The latter possess a structure similar to that of gymnocolin (208) from Gymnocolea inflata. 1H-NMR data suggested that gymnocolin was a trans-clerodane, but X-ray crystallographic study revealed that (208) was actually a cis- clerodane.123 These cis-clerodane lactones with a β-substituted furan ring exhibit intense bitterness.114 O O O OAc O O O H H O O O O O O (206) O O (207) (208) Kaurane diterpenoids encountered in liverworts belong to the ent-series, and all of them were isolated from the Jungermanniales. ent-Kaurene (209) is the most common compound of this series. ent-(16S)-11α-Hydroxykauran-15-one (210) was isolated from Jungermannia infusca, together with a number of known ent-kauranes, including ent-15α-hydroxykaurene (211), ent-kauren-15-one (212), and ent-(16R)-kauran-15-one 86 (213).124 These known ent-kauranes were also found in Nardia species,125 another rich source of this type of diterpenoids. R R' H R O H H (210) R = OH, R' = α-Me, β-H (213) R = H, R' = β-Me, α-H (209) R = H2 (211) R = α-H, β-OH (212) R = O Jungermannia infusca also produced a number of infuscasides, including infuscasides A (214) and B (215), the first terpene glucosides found in bryophytes.126-127 HO OH O OH OH O H OH OAc O AcO O OH OH H OH OH O (214) (215) HO HO HO O O O H H (216) O OH OH (217) OH (218) 87 The majority of labdane diterpenoids were isolated from the Jungermanniales. Scapania undulata and Ptychanthus striatus are rich sources of these compounds: 22 labdane diterpenoids have been found in these species prior to 2001, including labda12E,14-dien-8α,11ζ-diol (216), scapanins A (217) and B (218), and ptychantins E-G (219-221).112,114,122,128,129 AcO HO AcO AcO O OR OAc OH AcO H H OAc OAc (219) (220) R = H (221) R = Ac A few rare classes of diterpenoids have been found in liverworts, e.g. pleuroziol (222) and levierol (223), which belong to the chettaphanins (rearranged labdane),129-130 and the cembrane diterpenoid, setiformenol (224).131-132 OH H OH H O HO OH (222) (223) (224) The bis(bibenzyls) are a new class of natural products that is restricted to the liverworts. Since the first bis(bibenzyl), marchantin A (225) was isolated from Marchantia polymorpha,133 more than 60 compounds of this class have been reported. 88 COOH COOH HO O NH2 HO HO (279) COOH HO NH2 HO HO + COOH HO NH (281) (283) NH2 HO (282) (280) MeO MeO NCH3 HO OH HO NH NH HO HO H H H HO MeO HO (286) HO (285) (284) MeO MeO MeO MeO HO HO HO HO NCH3 NCH3 MeO MeO OH MeO O (287) NCH3 NCH3 MeO HO (288) H H MeO MeO MeO MeO O O O O NCH3 NCH3 HO O (277) O O (291) O NCH3 O (3) (278) HO NCH3 HO NCH3 MeO (292) HO (290) NCH3 O HO OAc (289) NCH3 MeO (294) (293) Scheme 4-1. Proposed biosynthetic pathway of morphine in Papaver somniferum. 179 The last steps of morphine biosynthesis in Papaver somniferum can proceed by two different routes: thebaine (278) å neopinone (291) å codeinone (292) å codeine (45) å morphine (3); and thebaine (278) å oripavine (293) å morphinone (294) å morphine (3).187-190 In an attempt to find endogenous opioids in mammalian tissues, many groups have employed immunoassays which can detect morphine immunoreactivity in brain tissues. Their efforts led to the detection of morphine and other natural opiates in mammalian tissues.191-195 However, it was still not clear whether these compounds were of endogenous or exogenous in nature, possibly derived from dietary sources. The picture became clearer when Weitz et al. discovered that rat liver can synthesize salutaridine from [3H]reticuline.196 This transformation from reticuline to salutaridine is the critical step in morphine biosynthesis, as it creates the rigid and chiral four-ring system of the morphinans, via intramolecular coupling. Later, this finding was strongly supported in a series of experiments by Amann and Zenk, showing that a highly regio- and stereoselective cytochrome P-450 enzyme from pig liver can convert (R)-[N14CH3]reticuline to salutaridine.197 These results showed that the opiates are of endogenous origin, that they are actually biosynthesized from a precursor, and even more, that the biosynthetic pathway of morphine in mammalian tissues and poppy share the same key step. The total synthesis of morphine was first reported by Gates198-199 and Ginsburg.200 The poor overall yield of this synthesis has prompted many other groups to explore alternative approaches and in 1985, a completed total synthesis of (+) and (-) morphine, as well as (+) and (-) codeine, with high overall yield is reported by Rice et al., and their approach is shown in Scheme 4-2,187 in which the practical conversion from dihydrothebainone (304) to codeine (45) is achieved by Weller and Rapoport.201 180 Although morphine is a powerful analgesic, it is also famous for its potential for abuse. This lead to countless attempts by medicinal chemists to find a “safer” analgesic, which can maintain analgesia efficacy and reduce abuse potential as well as other negative side effects (such as respiratory depression). Heroin (311), synthesized in 1898, was claimed as such compound, but has been proved to have just as much potential for abuse as morphine. MeO O NH2 (295) + O MeO MeO NCHO NH HO HO MeO MeO HO COOH HO NCHO MeO MeO (296) (297) H N (298) NCHO Br (299) O O Br NCHO O NCHO HO HO MeO OH O MeO OH O MeO Br MeO (303) (302) NCH3 (301) Br (300) NCH3 NCH3 NCH3 OMe MeO OH O MeO O (304) O MeO (305) NCH3 O MeO MeO (306) NCH3 O OMe (307) NCH3 NCH3 Br OMe OMe RO O OH (277) R = CH3 (3) R = H MeO O (310) O MeO O MeO (309) MeO O MeO (308) Scheme 4-2. Total synthesis of morphine. 181 The pharmacological properties of morphine suggested that there are specific sites in the nervous system that can recognize and bind to this alkaloid. As early as 1965, based on structure-activity relationship analysis studies, Portoghese proposed that there are more than one opioid receptor type exists, since the multiple modes of opioid receptor-ligand interactions could not be used to properly explain certain experimental data.202 Extensive studies lead to the demonstration of opioid binding sites in mammalian brain and peripheral tissues.203-205 Soon after, pharmacological evidence for the multiple opioid receptor concept became available, and three types of opioid receptors, namely µ, κ, and σ, were proposed based on different pharmacological profiles of the three agonists morphine, ketazocine and SKF 10 047 (N-allylnormetazocine).206-207 Assays have demonstrated, at a cellular level, the existence of multiple opioid receptor types, as well as another opioid receptor type in mouse vas deferens, which was named the δ-receptor.208 Subsequent studies showed that the σ-receptor is actually not an opioid receptor, since its physiological effects can not be reversed by opioid antagonists, even at very high concentrations. Of the three opioid receptors, the δ-receptor is the first to have been cloned,209-210 followed closely by the µ-211-214 and the κ-receptor.215-218 All these cloned opioid receptors are members of the seven transmembrane G-protein coupled receptors family. In both animals and humans, these opioid receptors are synthesized in the dorsal root ganglia and transported within the nerves to the central and peripheral nerve endings.219 The three major types of opioid receptors are further divided into subtypes, based on pharmacological studies and binding assays. The two subtypes of µ receptors, µ1 and µ2, showed different affinities for opioid peptides: while µ1 binds with equally high affinity to both morphine and opioid peptides, µ2 has markedly 182 lower affinity towards opioid peptides compare to morphine.220 Recently, another subtype of µ receptor, µ3, which is selective for opioid alkaloids but insensitive to opioid peptides, was reported by Stefano and coworkers.221 Pharmacological studies of δ receptors also revealed two subtypes δ1 and δ2, with δ1 receptors activate analgesia in the brain, and δ2 receptors exert their effects at the spinal cord level. The subtypes of κ receptors are κ1, κ2 and κ3. While the κ2 subtype has been identified by binding assays only, and its pharmacology remains unexplored; the other two subtypes are well studied, in terms of pharmacology and radioligand binding experiments, especially with the development of effective ligands.222 However, all attempts to identify subtypes of opioid receptors at molecular level have not been successful: cloning of opioid receptor subtypes always yielded cDNAs corresponding to the original receptor types. The endogenous ligands of opioid receptors were discovered in the mid-1970s. Hughes et al. reported the isolation and identification of two pentapeptides Leuenkephalin and Met-enkephalin,223 which have pharmacological properties similar to those of morphine. These three opioid receptor families are now usually referred to as traditional/classical opioid receptors, due to the discovery of a novel receptor in 1994, which has been cloned from rat,224-229 human,230-232 and mouse.230,233 The novel receptor, initially named Opioid Receptor Like-1 (ORL1 or nociceptin receptor), was an unexpected result from the screening of genomic and cDNA libraries, in attempts to identify putative subtypes of the three traditional opioid receptors. ORL1 receptor is also a member of the family of G-protein coupled receptor with seven transmembrane domains, and its structure displayed a significant degree of homology with the other opioid receptors (68% with µ, 67% with δ, 66% with κ receptors, and up to 80% 183 homologies in transmembrane regions). Despite this similarity, its pharmacological properties are quite distinct: all ligands of the classical opioid receptors, including non-selective ligands that bind µ, κ and δ receptors with equally high affinity, showed very low affinity for the ORL1 receptor. This receptor was initially considered “orphan” opioid receptor, due to the aforementioned low affinity towards opioid ligands and the lack of its own endogenous ligand. In 1995, the ORL1 receptor was de-orphaned when its endogenous ligand, a heptadecapeptide, has been isolated independently at the same time by two research groups.234-235 The first group named it nociceptin, due to its supposed pronociceptive activity. The name orphanin FQ, used by the second group, has two parts: orphanin refers to the fact that this peptide is the endogenous ligand of the “orphan” opioid receptor, and FQ are abbreviations of the first and last amino acid, phenylalanine and glutamine. Most groups in this field use both names together, hence the name is orphanin FQ / nociceptin, or OFQ/N. This peptide, which is broadly distributed throughout the central nervous system, now has been accredited with a wide range of neuronal functions, including modulation of motivational and emotional behavior, but most significantly, pain modulation. With this new receptor system, nomenclature of opioid receptors has become a controversial issue. In 1996, the International Union of Pharmacology Receptor Nomenclature Committee (NC-IUPHAR) proposed a new naming scheme for the opioid receptor family, in the order they were cloned: OP1 (δ), OP2 (κ), OP3 (µ), hence ORL1 would be named OP4. However, many pharmacologists ignored this proposal and continued to use the widely accepted Greek symbol names, in order to avoid disintegration of the literature. Recently, NC-IUPHAR recommended a temporary naming scheme,236 which will be used hereafter. 184 4.1.7 Structure of OFQ/N: Sequence of OFQ/N: Phe-Gly-Gly-Phe-Thr-Gly-Ala-Arg-Lys-Ser-Ala-Arg-Lys-LeuAla-Asn-Gln. OFQ/N is a heptadecapeptide with two internal pairs of basic amino acids (at positions 7-8 and 11-12), raising the possibility of further processing. This basic structure can be modified and even truncated at its C-terminal without major loss of activity, but the initial motif FGGF must remain intact. Of these, the phenylalanine in position is very important in binding selectivity, since the analog with tyrosine at position has greater affinity for traditional opioid receptors (and lower binding affinity at NOP receptors). Like most neuropeptides, OFQ/N has a larger precursor peptide, named preproOFQ/N (ppOFQ/N). This precursor has been cloned from rat, mouse, and human, with high interspecies homology (80% identity). 4.1.8 Receptor binding The first report of OFQ/N binding to NOP employed an iodinated analog, 125 I- [Tyr14]OFQ/N.234 In 1996, another analog, 3H-OFQ/N, was reported to yield results virtually identical to those of the iodinated peptide.237 Although most research groups found that OFQ/N has high binding affinity to NOP, the reported values fall within a moderately wide range, which is likely due to differences in assay techniques, buffers, and transfected cell lines, as well as radioligand stability. OFQ/N can block analgesia from various exogenous and endogenous opioid ligands. For example, OFQ/N can block morphine analgesia and this anti-opioid effect can last up to to hours. 185 Since OFQ/N does not interact with any of the traditional opioid receptors, it must act through its own neural circuits as a functional antagonist. NOP binding is also found to be sensitive to sodium and divalent cations. Studies of ligand effects on NOP using cAMP as a functional endpoint are not possible in binding assay with membrane preparations, since NOP regulation of cAMP is undetectable. Therefore, these studies have to be performed in a cellularbased assay, which makes comparisons between ligand-receptor binding and ligandcAMP regulation studies inaccurate, because OFQ/N binding to NOP is sensitive to the ionic composition of the incubation buffer. The use of recombinant receptor systems (NOP expressed in CHO cells) to address the coupling between G-protein and secondary messenger has its limitations. The ideal situation would be to express opioid receptors in neuronal tissues at levels found in nature. At higher level of expression, receptors can act differently. One example is the effects of [F/G]Nociceptin(1-13)NH2 at NOP: depending on the condition, this peptide can act as an antagonist, partial agonist, or full agonist. Initial experiments showed that OFQ/N was hyperalgesic when administered by intracerebroventricular injection (icv). Later, results from many other groups indicated that OFQ/N – NOP system has a rather complex pharmacological profile.238 For example, the hyperalgesic effect seems region and assay specific, since OFQ/N was reported to be analgesic and induce allodynia when injected into the spinal cord.239-240 At present, the pharmacology of OFQ/N – NOP system is difficult to characterize, due to the absence of a selective antagonist. It is very important for OFQ/N, since many groups have demonstrated that this peptide possesses both nociceptive and anti- 186 nociceptive activities. Furthermore, opiate antagonists (at high concentration and in vitro) can block certain affects caused by OFQ/N but show no effect against others.238 4.1.9 OFQ/N agonists and antagonists: There are more than twenty compounds, both peptide and non-peptide, identified as NOP-selective ligands. The non-peptide ligands generally have poor selectivity and specificity, but are more stable compared to those of a peptide nature. Based on their activities, the ligands are divided into agonist, antagonist, and partial agonist groups. Recently, Calo and coworkers reported [Nphe1,Arg14,Lys15]Nociceptin-NH2 (UFP101), a novel potent and selective NOP antagonist with more than a 3000-fold selectivity over classical opioid receptors.241-242 Another artificially designed peptide, ZP120, was characterized in vitro and in vivo by Rizzi et al. as an NOP partial agonist.243 A major issue that needs to be resolved regarding the NOP receptors is the binding site heterogeneity, with evidence come from both pharmacological and binding studies. However, without more definitive evidence, OFQ/N binding site heterogeneity remains to be established. The regional distribution of OFQ/N and NOP has been well documented. Literature showed that the localization of OFQ/N corresponds relatively well with that of the NOP. Its particularly wide distribution patterns suggested that the NOP receptor system is involved in nociception, as well as reinforcement and reward, motor and balance control, autonomic control of physiological process, stress response, sexual behavior and aggression. Among these proposed functional roles of OFQ/N, the ability of this peptide in the mediation and modulation of pain in the supraspinal, 187 spinal and peripheral levels of the nervous system is receiving the most attention from research scientists. 4.1.10 OFQ/N metabolism: Metabolic fragmentation of bioactive peptides is important since it may lead to products that are less potent, inactive peptides, or even generate molecules with another type of activity. That knowledge will allow scientists to design enzyme inhibitors that may be pharmacologically relevant. In the case of OFQ/N, its metabolism has been studied in various types of tissue and conditions (Table 4-2). The results clearly showed that OFQ/N metabolism is tissue-dependent. However, our knowledge about the natural occurrence of those OFQ/N fragments is very limited so far. Species Tissue Condition Major fragments Mouse Brain Incubation with cortical slices N1-7, N1-11 Rat Brain Microinjection into hippocampus N1-13, N1-9 Brain cells in culture Incubation with cell extract N1-13, N1-9 Spinal cord Incubation with spinal cord tissue N1-11, N1-6 Tumor cells in culture Incubation with cell extract N1-13, N1-9 Plasma Incubation with plasma N2-17, N3-17 Human Table 4-2. Metabolism fate of OFQ/N in different tissues.244 4.1.11 Methods for study pain mechanism: The methods used to elucidate the pain mechanism and to evaluate analgesics for exploring effective treatment can be classified into two groups: in vitro and in vivo. The in vitro methods are performed at a molecular level, while the in vivo methods involve various animal models, using mostly those based on the mouse and rat. These animal models also have been employed for behavioral studies of normal and transfected mice. The advantage of using mouses is that they are inexpensive, easy to 188 breed and have a wider variety of genotypes, compare to the rat. Recently, the transgenic “knockout” technology, which was developed for the mouse, has made this species an even more popular animal model in biomedical studies.245 However, in our investigation, we will focus on the in vitro methods only. The in vitro methods for NOP-OFQ/N system can be further divided into two types: ligand-receptor binding and ligand-cAMP regulation. The former measures the binding affinity between the potential ligand and NOP (compare to the endogenous ligand OFQ/N), while the latter tells us whether the ligand is an agonist or an antagonist. The first type can be used for fast screening of a large number of candidates, while the second type is more suitable for further study of a confirmed NOP ligand. Therefore in our study, since we want to search for potential NOP ligand(s) in the isolated natural products, we chose the ligand-receptor binding method. This method is based on the competitive binding principle: [3H]-labeled OFQ/N was allowed to bind with NOP, then the test compound was added to displace the bound [3H]-labeled OFQ/N. 189 4.2 Results & discussion The crude extracts of Salacia chinensis, Singaporean Pallavicinia cf. lyellii, and selected first level fractions of the Vietnamese Pallavicinia lyellii were screened for potential NOP ligand. Only four fractions from the crude acetone extract of Vietnamese P. lyellii showed higher binding affinity to NOP than the endogenous ligand OFQ/N (i.e. displaced [leucyl-3H]OFQ/N from NOP). The results are showed in table 4-3. Fractions of P. lyelli VN E F G H P P. lyellii SG 27109 14001 15122 21901 42217 56233 Sample Total binding (cpm) Salacia Control 45644 47073 Table 4-3. The total binding of the control sample (see Experimental) represents maximum binding without competition. Therefore, a sample can be considered as potential NOP ligand if it has the total binding significantly lower than that of the control sample. Fraction G was therefore selected for bioassay-guided purification to identify the active component(s). Results from binding assays performed on the subsequent fractions confirmed that the crude acetone extract of the Vietnamese P. lyellii contained potential NOP ligand(s). Bioassay results from the last separation step is shown in table 4-4. Fractions Total binding (cpm) G4-19-14 G4-19-18 G4-19-19 G4-19-21 G4-19-22 Control 9455 5242 12869 20853 35163 76401 Table 4-4. 190 NMR spectra of the most active fraction (after four separation steps), showed that it may be a mixture of two compounds which did not correspond to any of the isolated compounds. The 1H-NMR spectrum contained signals due to one methoxy group, two tertiary methyls and two secondary methyls. The 13C-NMR spectra showed fifty two signals, including nine methyls, thirteen methylenes, twelve methines, and eight quaternary carbons. There was one carbonyl, one aldehyde, one methoxy, and at least five oxygenated carbons. Attempts to further purify this fraction by HPLC (using RP18 and RP-8) resulted in unresolved peaks. HPLC purifications using silicagel and DIOL failed due to possible interconversion between the major compound and the two minor compounds in the mixture. The interconversion was also observed when neighboring fractions (G4-19-14 and G4-19-18) were purified by preparative TLC and flash chromatography. The 1D and 2D-NMR data (see Appendix, page 241-244) only provide enough information for partial structure elucidation. Below are the two possible fragments that can be built from the NMR data: CH2 Me H Me Me H O H H CHO H 191 Many research groups are focusing on the NOP – OFQ/N system, led to the identification of a number of potential ligands. However, according to the IUPHAR Receptor Database, only compound J113397 (312) is currently recognized as an antagonist of NOP, and there is no agonist available for this receptor.246 The lack of selective agonists hinders the progress in this field. Further investigation of the Vietnamese P. lyelli is needed to isolate and identify this potential ligand for NOP, as well as establish its pharmacological profile. N N N O OH (312) 192 4.3 Experimental Membranes prepared from Chinese hamster ovary cells expressing the recombinant human NOP (CHOhNOP) were used for screening. CHOhNOP cells were maintained in RPMI 1640 medium, supplemented with 10% foetal bovine serum (v/v), and 1% antibiotic (v/v) at 37oC in 5% carbon dioxide humidified air. Cells were sub-culture every 5-6 days. Membranes prepared from the cells were then resuspend in 50 mM Tris buffer pH 7.4 containing 15 µg/ml protease inhibitor (leupeptin) and 100 µg/ml bovine serum albumin (BSA). [leucyl-3H]OFQ/N (specific activity, 152 Ci/mmol) was purchased from Amersham Biosciences UK. Unlabeled OFQ/N (human) was purchased from Peptide Institute, Inc., Japan. For binding assay, pM of [leucyl-3H]OFQ/N was added to the membranes (75 µg in 100 µl Tris buffer) minutes prior to the addition of test compound (in acetone) to the final volume of 205 µl. For each binding assay batch, there was one control sample, in which the test compound solution was replaced by the same volume of Tris buffer. The mixture was then incubated at room temperature for 60 minutes. The cell membranes were then harvested using multi-channel harvester by filter the binding reactions through GF/F glass fiber filter paper (pretreated with 0.5% polyethyleneimine). The filters were washed twice on the harvester with ml ice-cold 50 mM Tris buffer pH 7.4. After drying the glass fiber filter paper for 24 hours in the oven, filter discs were punched out into scintillation vials and incubated overnight with ml of scintillation fluid, on a rotating shaker. Counting was performed on a βcounter, minutes for each vial. The selected fraction was subjected to gel permeation chromatography on Sephadex LH-20, followed by VLC on silicagel, then flash chromatography (DIOL, CHCl3- 193 hexane). For each step of the bioassay-guided purification, 7-12 fractions were selected for binding experiment, based on TLC and 1H-NMR. Fractions that showed positive results in binding experiment were combined for the next separation step. Mixture G4-19-19 (4.1mg): 1H-NMR: 11.08 (1H, s), 10.30 (1H, s), 9.57 (1H, s), 8.53 (1H, s), 7.98 (1H, dd, J = 17.9, 11.5 Hz), 6.37 (1H, dd, J = 17.8, 1.0 Hz), 3.91 (1H, s), 3.66 (1H, s), 3.38 (1H, s), 0.85 (1H, s); 13 C-NMR (including DEPT-90 and DEPT- 135): 190.6, 188.2 (CH), 176.1, 173.9, 170.8, 165.8, 160.0, 151.7, 148.1, 144.5, 143.2, 139.1, 138.2, 137.7, 134.0, 133.6, 133.4, 131.2, 130.1 (CH), 124.1 (CH2), 120.0 (CH), 108.3 (CH), 107.2, 101.7 (CH), 95.3 (CH), 66.1 (CH), 62.2 (CH2), 53.8 (CH3), 53.1 (CH), 51.6 (CH), 40.9 (CH2), 40.7 (CH2), 38.6 (CH2), 38.6 (CH2), 38.5 (CH2), 37.8 (CH2), 34.0 (CH), 33.9 (CH), 32.5 (CH2), 29.3 (CH), 26.2 (CH2), 26.0 (CH2), 25.6 (CH2), 24.1 (CH3), 23.6 (CH3), 23.5 (CH3), 20.6 (CH3), 20.1 (CH3), 19.4 (CH2), 16.9 (CH3), 12.8 (CH3), 12.6 (CH3). 194 [...]... 5.97 (dt, 17.0, 10 .2 Hz) 4, 5 136.7 7E 5 .24 (dd, 17.0, 2. 2 Hz) 5, 6 117.8 7Z 5.13 (dd, 10 .2, 2. 7 Hz) 8 118.3 9 2. 15 (d, 6.9 Hz) 11, 12, 20 1, 5, 11, 17, 20 61.5 10 51 .2 11 4.81 (dd, 6.6, 3.8 Hz) 7, 9 8, 12 83.3 12 4.85 (dd, 3.8, 1.0 Hz) 9, 15 11, 14, 16 76.3 13 127 .5 14 7.00 (qd, 7 .2, 1.0 Hz) 15 12, 13, 15, 16 141.8 15 2. 02 (d, 7 .2 Hz) 12, 14 13, 14 15.9 16 169 .2 17 1.60 (s) 9, 12 8, 9 26 .0 18 0.90 (s)... Hz) 2, 3, 9, 12, 20 78.4 9, 3 , 15 1.75 (m) 1, 3, 4, 10 23 .8 2 2 1.56 (m) a 1. 32 (dt, J = 3.9, 13.5 Hz) 1, 2, 4 39.3 3 a 3 1.56 (m) 4 34.1 a 5 1.56 (m) 60.1 6 5.64 (dt, J = 17.1, 10 .2 Hz) 5, 10 135.0 7E 5.00 (dd, J = 17.1, 2. 0 Hz) 5, 6 120 .1 7Z 5.11 (dd, J = 10 .2, 2. 0 Hz) 8 21 0.8 9 2. 32 (d, J = 10.0 Hz) 1, 14 5, 8, 10, 11, 20 63.4 10 42. 6 11 4.68 (dd, J = 10.0, 7.8 Hz) 12, 20 8, 12, 13, 16 75.6 12 5 .26 ... 8, 9 26 .1 18 0.95 (s) 3, 4, 5, 19 32. 4 19 0.97 (s) 3, 4, 5, 18 23 .2 20 1 .26 (s) 11, 12, 17, 19 1, 5, 9, 10 12. 9 21 169.8 22 2. 07 (s) 21 21 .6 a ( ): Approximate position of overlapping multiplets assigned using HMQC correlations 109 Pallavicinin E Pallavicinin E (25 2), was obtained as a white powder, mp 20 4 -20 5 oC, [ ]D +95.5 (c 0 .28 , CHCl3) It was determined to have the molecular formula C22H32O6 by... Hz) 8, 10, 11, 17, 20 66.7 10 42. 5 11 5.19 (m) a 16 80.0 a 12 5 .21 (m) 14 72. 9 13 128 .1 14 7.09 (dq, J = 1.7, 7.1 Hz) 15 12, 15, 16 144.9 15 2. 04 (dd, J = 7.1, 0.5 Hz) 3, 4 16.0 16 168.4 17 1.37 (s) 7, 8, 9 22 .7 18 0.93 (s) 1, 6 3, 4, 5, 19 32. 4 19 1.01 (s) 20 3, 4, 5, 18 22 .6 20 1 .29 (s) 11, 17, 19 1, 5, 9, 10 12. 7 21 169.6 22 2. 08 (s) 21 21 .7 a ( ): Approximate position of overlapping multiplets assigned... isolation of ten compounds, seven of which are novel: pallavicinins B-F (24 9 -25 3), 8-hydroxymarchantin C (26 7), and 7-oxo-riccardin D (26 8) 14 15 15 H 17 O O 20 H 9 O 16 11 10 H O 7 18 19 (24 9) (25 0) O 14 13 O 14 13 15 16 H 16 H 12 O O 12 11 H 20 H 1 O O 11 H 20 17 H 1 9 10 17 9 10 OH 5 OH 5 3 7 3 7 H H 19 8 5 3 18 15 17 9 10 7 19 H 11 20 1 H 5 3 O O O 1 16 12 13 12 8 O 13 18 O 21 O (25 1) 19 18 O 21 O (25 2)... 7.8, 2. 8 Hz) 11 1, 9, 13, 14 72. 9 13 124 .5 14 7.10 (qd, J = 7.3, 2. 8 Hz) 15 12, 15, 16 1 42. 0 15 2. 07 (dd, J = 7.3, 2. 5 Hz) 1, 14 13, 14 13.6 16 168 .2 17 2. 20 (s) 8 36.8 18 0.80 (s) 3, 4, 5, 19 32. 8 3 , 7E 19 0.86 (s) 6, 20 3, 4, 5, 18 22 .5 20 1 .26 (s) 6, 11, 19 1, 5, 9, 10 10.4 a ( ): Approximate position of overlapping multiplets assigned using HMQC correlations 104 The biosynthesis of compound 25 0... (22 5) R = H, R' = OH, R" = H2 (22 6) R = R' = OH, R" = H2 (22 7) R = R' = H, R" = H2 (23 1) R = H, R' = OH, R" = O R (22 8) R = R' = H (22 9) R = H, R' = Me (23 0) R = OH, R' = H It is noteworthy that marchantin C (22 7) is not only isolated from the Marchantiales, but also from the Jungermanniales, including Plagiochila sciophila and Schistochila glaucescens 137-138 Riccardin A (23 2) is representative of. .. 2, 9, 20 78.0 , 5, 9, 12 2 1. 82 (dq, J = 13.7, 3.9 Hz) 1, 4, 10 26 .6 2 1.67 (dq, J = 13.7, 3.9 Hz) 1, 2, 5, 19 42. 1 3 1.41 (dt, J = 13.5, 3.4 Hz) 4 33 .2 5 0.98 (d, J = 2. 9 Hz) 1, 4, 6, 9, 10, 19, 20 55.4 6 5.57 (dt, J = 3.9, 2. 8 Hz) 5, 8, 10, 21 70.4 5, 18, 22 , 7 7 1.74 (dd, J = 13.9, 4 .2 Hz) 43.9 7 2. 20 (dd, J = 13.9, 2. 7 Hz) 5, 6, 8, 9, 17 6, 17, 7 8 81.9 9 1.51 (d, J = 7.8 Hz) 8, 10, 11, 17, 20 66.7... enhancements of (24 9) Table 3-1 1H-NMR (500 M Hz), 13C-NMR ( 125 M Hz), HMBC and NOEDIFF spectral data of compound (24 9) in CDCl3 NOEDIFF HMBC No C H 1 3.78 (dd, 12. 2, 3.7 Hz) 5, 2 , 3 3, 5, 9, 20 85 .2 2 1.77 (dq, 12. 2, 3.9 Hz) 1, 3, 4, 10 22 .5 2 1.68 (dq, 3.4, 12. 5 Hz) 1.37 (dt, 4 .2, 13.6 Hz) 1, 2, 4, 5, 19 39.7 1.57 (dt, 13.9, 3.4 Hz) 4 34.0 5 3.01 (d, 10.3 Hz) 1, 4, 6, 7, 9, 10, 18, 19, 20 51.5 6 5.97... 9, 10, 19, 20 52. 9 6 5.51 (dt, J = 2. 4, 3.1 Hz) 8, 10 69.4 5, 7 , 18 a 1.73 (m) 45.5 7 5, 6, 8, 9, 17 7 2. 15 (dd, J = 15 .2, 3.1 Hz) 8 70 .2 a 9 1.57 (m) 8, 11, 17, 20 58.5 10 38.8 11 4.97 (dd, J = 10.8, 7.8 Hz) 12, 17, 20 8, 12, 13, 16 73.8 12 5.09 (dt, J = 7.8, 2. 7 Hz) 11 1, 11 73.4 13 123 .4 14 6.48 (qd, J = 7.5, 2. 9 Hz) 12, 16 141.4 15 2. 26 (dd, J = 7.5, 2. 9 Hz) 1, 14 13, 14 14.0 16 166 .2 17 1.50 (s) . to 20 01, including labda- 12E,14-dien-8α,11ζ-diol (21 6), scapanins A (21 7) and B (21 8), and ptychantins E-G (21 9 -22 1). 1 12, 114, 122 , 128 , 129 OAc OR AcO AcO AcO H O A c O H O A c OH HO AcO (21 9). (22 6- 23 1). 134-136 OH R' OH O O R OH OH OH O O OR' R R" (22 5) R = H, R' = OH, R" = H 2 (22 8) R = R' = H (22 6) R = R' = OH, R" = H 2 (22 9) R = H, R' = Me (22 7) R = R' = H, R" = H 2 (23 0) R = OH,. chettaphanins (rearranged labdane), 129 -130 and the cembrane diterpenoid, setiformenol (22 4). 131-1 32 H OH OH H OH O HO (22 2) (22 3) (22 4) The bis(bibenzyls) are a new class of natural products that