The Alkaloids Volume 8

A Review of the Literature Published Between July 1976 and June 1977

By M. F. Grundon

The Royal Society of Chemistry

Copyright © 1978 The Chemical Society
All rights reserved.
ISBN: 978-0-85186-327-6

Contents

Chapter 1 Biosynthesis By R. B. Herbert, 1,
Chapter 2 Pvrrolidine, Piperidine, and Pyridine Alkaloids By A. R. Pinder, 37,
Chapter 3 Pyrrolizidine Alkaloids By D. J. Robins, 47,
Chapter 4 Indolizidine Alkaloids By J. A. Lamberton, 62,
Chapter 5 Quinolizidine Alkaloids By M. F. Grundon, 66,
Chapter 6 Quinoline, Quinazoline, and Acridone Alkaloids By M. F. Grundon, 77,
Chapter 7 β-Phenethylamines and the Isoquinoline Alkaloids By K. W. Bentley, 87,
Chapter 8 Aporphinoid Alkaloids By M. Shamma, 122,
Chapter 9 Amaryllidaceae and Related Alkaloids By J. N. Reed and V. A. Snieckus, 137,
Chapter 10 Erythrina and Related Alkaloids By S. 0. de Silva and V. A. Snieckus, 144,
Chapter 11 Indole Alkaloids By J. E. Saxton, 149,
Chapter 12 Lycopodium Alkaloids By W. A. Ayer, 216,
Chapter 13 Diterpenoid Alkaloids By S. W. Pelletier and S. W. Page, 219,
Chapter 14 Steroidal Alkaloids By D. M. Harrison, 246,
Author Index, 265,

CHAPTER 1

Biosynthesis

BY R. B. HERBERT

1 Introduction

In order to facilitate access to material discussed in previous Reports in this series, the practice of listing them as the first references is continued. This means that when new material is discussed the reference to material in, e.g., Vol. 3 appears most simply as ‘ref. 3’ (and will do so in all Reports beginning with Vol. 6),

2 Piperidine, Pyridine, and Pyrrolidine Alkaloids

Dioscorine. — Labelling of C-5, C-.10, and C-12 of dioscorine (2) by [1-14C]acetic acid indicates that C-5, C-6, C-9, C-10, C-11, C-12, and C-13 derive from acetate. This leaves a cyclic C5N unit unaccounted for, which, from the wealth of evidence on the biosynthesis of similar systems (see previous Reports), one expects will arise from the amino-acid lysine via Δ1-piperideine (1), an expectation not realized since neither compound is satisfactorily incorporated into dioscorine (2).

The alkaloid anatabine (6) is exceptional in that its dehydropiperidine ring derives from nicotinic acid (3), rather than from lysine which is the source of the similar fragment in anabasine (7) (see below). Most interestingly it has been shownthat the related heterocyclic fragment (C5N unit) of dioscorine also derives from nicotinic acid (3): [2-14C]- and [5,6-14C2,13 C2]-nicotinic acids were tested as precursors and in the latter case the dioscorine C labelling was shown by 13C n.m.r. to be of C-1 and C-7. 3,6-Dihydronicotinic acid (4) may be an important intermediate in the biosynthesis of nicotine and anabasine (7) and of anatabine (6). The evidence for dioscorine biosynthesis is consistent likewise with the intermediacy of (4) (see Scheme 1). The sequences of saturation and condensation leading to dioscorine, and also anatabine, cannot yet be specified, but are clearly well worth exploring. Of interest too, in the case of dioscorine, is the way in which the branched acetate chain originates (stepwise introduction of acetoacetate units in the course of the biosynthesis of the alkaloid?).

Anatabine. — A study of the biosynthesis of anatabine (6), which has been published in preliminary form and reviewed, is now available in a full paper. The incorporation of [2-14C]nicotinic acid equally into C-2 and C-2′ of anatabine (6) provides essentially the only new information. This result complements that obtained with [6-14C]nicotinic acid, i.e. labelling of C-6 and C-6′. Nicotinic acid was equally incorporated into both halves of (6) regardless of the length of the feeding experiment, which argues for alkaloid formation from two identical fragments [as (5)] rather than from (4) and (5), which might be formed from pools of material of different size and thus be dependent upon the length of the experiment.

In these experiments radioactive α,β-bipyridyl (8) was isolated and it appears from degradation studies that both rings of the alkaloid arise from nicotinic acid, so (8) can arise from anatabine (6) but not from anabasine (7), in which only one ring originates from nicotinic acid.

Lupin Alkaloids. — The C15 lupin alkaloids, e.g. sparteine (9) and lupanine (10), are biosynthesized from three C5 units derived (as shown in Scheme 2) from lysine and cadaverine, the two nitrogen atoms also arising from lysine. Although the later stages of biosynthesis of these and related alkaloids are fairly clear, the steps which lie between cadaverine and the alkaloids are obscure, the available hypotheses not being supported by experimental results. Recent results, however, support in a preliminary way an attractive new hypothesis, i.e. that these Calkaloids are modified trimers of Δ-piperideine (1) (which is an important intermediate in the biosynthesis of many alkaloids derived from lysine). One of these trimers is isotripiperidine (11), shown with the favoured all-trans stereo-chemistry. Modification of (11) as shown for (-)-sparteine (Scheme 3) leads to lupin alkaloids with the same relevant stereochemistry as (11), i.e. 6R, 7S, 9S; those with 6S, 7R, 9R-stereochemistry may be derived from the enantiomer of (11). (-)-β-Isosparteine is the only alkaloid with 6R, 1R, 9R-stereochemistry and it is suggested that it arises from another stereoisomer of (11).

The Δ1-piperideine trimer hypothesis is supported initially by the equal incorporation of lysine and cadaverine into all three alkaloid fragments but more significantly by the incorporation of three molecules of Δ1-piperideine (1) into lupanine (10) and by the manner of this incorporation: label from C-6 appeared at C-2, C-15, and, by inference, C-10, whereas C-2 label appeared at C-17, C-11, and, by inference, C-6, consistent with the hypothesis (see Scheme 3). Further, approximately a third of the label was located at each of the determined sites.

This trimer hypothesis may be extended to related alkaloids, e.g. matrine (12).Attractive though this is, it is argued against by the unequal utilization of Δ1-piperideine (1) units in the construction of this alkaloid skeleton.

Securinine. — Further details of one group’s study of the biosynthesis of securinine (13) have been published. The origins of this alkaloid are well defined, and information which adds to this definition is that tyrosine is incorporated without loss of tritium from the carbon atoms flanking the phenolic hydroxy-group.

Proferrorosamine A. — Further information on the biosynthesis of the bacterial metabolite proferrorosamine A (14), which is known to derive from picolinic acid, is that [1-14C]glycerol is more extensively incorporated into the pyrrolidine than the pyridine fragment of (14).

Nicotine. — [2-14C]Ornithine is incorporated into the pyrrolidine ring of nicotine (15) with equal labelling of C-2′ and C-5′. This requires passage of the amino-acid through a symmetrical intermediate, which combined evidence from experiments with labelled compounds and enzymes strongly indicates is putres-cine. The intermediacy of a symmetrical compound (putrescine) in biosynthesis is supported by the results of one set of experiments with 14CO2, but inconsistent with another. In this latter case the conclusion is that the pyrrolidine ring of nicotine derives partially via a symmetrical intermediate and partially without the intervention of such a compound. Re-examination of [2-14C]ornithine and 14CO2/13CO 2 incorporation, in association with a new degradation sequence, has led to a reaffirmation of the conclusion that both precursors are incorporated exclusively via a symmetrical intermediate. It must be noted, however, that the results from examining 13CO2 incorporation by n.m.r., in contrast to those obtained by degradation (14CO2), are less unequivocal.

It had been noted earlier that nornicotine (16) formed from [2-14 C]ornithine in an excised root culture of Nicotiana rustica was apparently unequally labelled at C-2′ and C-5′, arguing for partial incorporation of the amino-acid without intervention of a symmetrical intermediate. The validity of this conclusion is seriously questioned by the recent observation that, in N. glutinosa plants, ornithine was incorporated into nornicotine (16) in the same way as into nicotine (15), i.e. symmetrically.

Phenanthroindolizidine Alkaloids. — These alkaloids, which may be exemplified by tylophorine (17), have been shown to have their genesis in tyrosine, phenyl-alanine, and probably ornithine, as shown in Scheme 4; the phenylalanine is utilized by way of cinnamic acid. This latter observation is consistent with a normal pathway for phenylalanine incorporation in which benzoylacetic acid is an intermediate. Accordingly, incorporations of this acid (18), and also its mono-hydroxy-derivative (19), have been recorded; (25) was not incorporated. More importantly for the biosynthesis of phenanthroindolizidine alkaloids, the phenacylpyrrolidines [as (20)], which are analogous to many alkaloids with a pyrrolidine or piperidine ring, are defined as key intermediates. Thus the compounds (20), (21), and (22), doubly labelled, were incorporated intact into tylophorinine (24). The results obtained with these phenacylpyrrolidines and the keto-acids, when combined, lead to the early part of the pathway shown in Scheme 5; the incorporation of both (19) and (20) indicates that aromatic hydroxylation may occur before as well as after phenacylpyrrolidine formation.

Later steps of biosynthesis follow as the joining together of (22) with a tyrosine-derived molecule to give the amino-acid (23), analogues of which are involved in isoquinoline biosynthesis. Appropriate oxidative phenol coupling and subsequent steps as indicated (Scheme 5) could afford tylophorine (17) and tylophorinine (24).

Tropane Alkaloids. — It is known that tropine (26) is a precursor for meteloidine (27), and its close relative hyoscyamine (29) is a precursor for scopolamine (28). Experiments with samples of (26) labelled with β-tritium at C-6 and C-7 show that entry of the two β-hydroxy-groups in (27) must occur with normal retention of configuration since almost complete loss of tritium occurred. Tritium was again lost almost completely on formation of scopolamine (28). On the assumption that early conclusions on the sequential intermediacy of (30) and (31) in the biosynthesis of (28) are correct, formation of (30) involves normal retention of configuration (loss of half the tritium) and cis-dehydration then occurs to give (31) (loss of remaining tritium). [In these experiments the hyoscyamine (29) isolated showed appropriately no loss of tritium.]

3 Isoquinoline Alkaloids

The biosynthesis of isoquinoline alkaloids has been reviewed.

Protostephanine and Hasubanonine. — The unravelling of the biosynthesis of pro-tostephanine (36) and hasubanonine (37), both produced by Stephania japonica, has proved to be a long and difficult task. Painstaking experimentation has led to the conclusion that these alkaloids are both constructed from two C6-C2 units derived from tyrosine. The unit which is the source of ring C and the attached ethanamine side-chain in (36) and (37) combines as (32) with the other unit.

Sixteen possible benzylisoquinolines synthesized formally from (32) and the acids (38) have been examined as precursors for (36) and (37). Of these, the bases with O-methyl groups at C-8 and/or C-3′ [as (35)] were not incorporated, but (35) as well as (33) and (34) were, and specifically so where examined (see Scheme 6 for labelling); the results show that the timing of N-methylation is not critical. It follows then, and it is an important conclusion, that hasubanonine (37) and pro-tostephanine (36) are members of the large family of modified benzylisoquinolines. The pathway which the results indicate is illustrated in Scheme 6. It is to be noted from the results that phenol oxidative coupling must occur on (35) and not an O-methylated derivative of it — a so far unique example where two hydroxy-groups must be present on one ring for coupling to occur.

Also examined as precursors in these experiments were four possible bisphenethylamines derived from (32) as well as the acids (38) used to synthesize the benzylisoquinoline precursors. Consistent with biosynthesis of (36) and (37) along the pathway deduced above (Scheme 6) they were not incorporated.

Erythrina Alkaloids. — The biosynthesis of Erythrina alkaloids such as erythraline (42) has been proved to be from (S)-norprotosinomenine (39) by way of the dienone (40) and the dibenzazonine (41) (Scheme 7). The alkaloid isococculidine (46), which lacks a C-16 oxygen function, can be thought of as arising from (44), the reduction product of (40), by rearrangement to (45). Subsequent steps would parallel the route to the alkaloids such as erythraline, although cyclization of (45) could not take a course similar to that proposed for (41) (Scheme 7). Consequently a more acceptable route may be for loss of the oxygen atom in the biosynthesis of (46) to occur at a later stage by cyclization on (47), which is simply a lower-oxidation-level equivalent of (43).

Proof that the biosynthesis of isococculidine (46) parallels the normal Erythrina pathway has been obtained by showing that norprotosinomenine [as (39)] is an intact precursor for (46), is enormously more efficiently assimilated into the alkaloid than alternative isoquinoline precursors, nor-reticuline (48) and nororien-taline (49), and [by adding carrier (39) during isolation after a tyrosine feeding experiment] is a natural constituent of Cocculus laurifolius, the plants used in these experiments. It may be noted that as in the biosynthesis of other Erythrina bases (S)-norprotosinomenine (39) was much more efficiently utilized than its enan-tiomer. Further, incorporation of this isoquinoline does not involve loss of tritium label from C-1. Additional results are that norlaudanosoline (50), the undoubted precursor for (39), was efficiently incorporated into (46), but, as to be expected,(51) was not utilized in biosynthesis.

Cephalotaxus Alkaloids. — Preliminary results indicate that the homo-Erythrina alkaloid schelhammeridine (52) derives from phenylalanine and tyrosine by way of a phenethylisoquinoline precursor [as (53)]. Previous evidence for the biosynthesis of the related alkaloid cephalotaxine (54), obtained with tyrosine labelled in the side-chain, has indicated a different pathway which involves two molecules of this amino-acid. Recently, however, tyrosine labelled in the aromatic ring was examined as a cephalotaxine precursor and was found to label ring A of (54) almost exclusively, i.e. only one unit of tyrosine is used for biosynthesis. This is obviously inconsistent with the previous evidence and the early incorporations are now attributed in part to tyrosine catabolism in such a way that side-chain label but not ring label enters the pathway to cephalotaxine (54). By shortening the time of the feeding experiments it could be shown that DL-[2-14C]tyrosine primarily labels C-10, consistent with the ring-labelled-tyrosine results; in prolonged experiments label appeared elsewhere.

As expected, if rings C and D do not arise primarily from tyrosine, phenylalanine was incorporated, label from C-1 appearing largely at C-8. This establishes that the amino-acid provides a C6-C3 unit for cephalotaxine biosynthesis and is consistent with passage through a phenethylisoquinoline intermediate fas (53)]. Incorporation of phenylalanine would then be expected to be through cinnamic acid, but this compound was not incorporated into (54), although phenylalanine was assimilated with loss of one proton from C-3, consistent with passage through an intermediate such as cinnamic acid. The significance of the cinnamic acid incorporation must clearly rest on the results of further experiments.

Cephalotaxine (54) may occur naturally as various esters, e.g. deoxyharringtonine (55). The acid (58) is analogous to (61), which is an intermediate in the conversion of L-valine into L-leucine. This suggests a related pathway (Scheme 8) to (58) beginning with L-leucine. This is supported in a key way by the isolation from Cephalotaxus harringtonia of (56) specifically labelled as shown, after feeding [1-14C]leucine, and by the specific incorporation of (60) into (58), labelled as shown. On the basis of the proposed pathway (59), derivable simply from homoleucine (57), should be an immediate precursor for (58). A highly efficient and specific incorporation of (59) into (58) gives substance to this hypothesis. Attention may be drawn to the biosynthesis of similar esters of pyrrolizidine alkaloids, which also originate from branched-chain amino-acids.

Protoberberine Alkaloids. — In the course of the bioconversion of the protoberberine scoulerine (65) into chelidonine (62) and phthalide-isoquinolines, e.g. narcotine (63), C-13 becomes oxidized. Ophiocarpine (68), with a hydroxy-group at C-13, represents an intermediate stage in the modification of the protoberberine skeleton, and results of tracer experiments have shown that scoulerine (65) is also to be included in the biosynthesis of this alkaloid. Tetrahydro-protoberberine (67) is also a precursor, its incorporation indicating that C-13 hydroxylation is a terminal step. As for other protoberberine derivatives, nandinine (64) was not assimilated, and it follows then that (65) is probably converted into (67) by way of isocorypalmine (66).

Using chirally tritiated samples of the protoberberine (67) it has been established that hydroxylation of (67) to give (68) occurs with loss of the 13-pro-R hydrogen atom, i.e. normal retention of configuration, and does not involve an enamine intermediate since tritium is not lost from C-14 during the course of this biotransformation. It is to be noted that similar results, associated with C-13, have been observed for narcotine (63) and chelidonine (62) biosynthesis, except that here the 13-pro-S proton is removed.

(Continues…)Excerpted from The Alkaloids Volume 8 by M. F. Grundon. Copyright © 1978 The Chemical Society. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.