secondary metabolism: the building blocks and construction...

Medicinal Natural Products. Paul M DewickCopyright 2002 John Wiley & Sons, Ltd

ISBNs: 0471496405 (Hardback); 0471496413 (paperback); ; 0470846275 (Electronic)

2SECONDARY METABOLISM: THE

BUILDING BLOCKS ANDCONSTRUCTION MECHANISMS

Distinctions between primary and secondary are defined, and the basic building blocks used inthe biosynthesis of secondary natural products are introduced. The chemistry underlying how thesebuilding blocks are assembled in nature is described, subdivided according to chemical mechanism,including alkylation reactions, Wagner–Meerwein rearrangements, aldol and Claisen reactions, Schiffbase formation and Mannich reactions, transaminations, decarboxylations, oxidation and reductionreactions, phenolic oxidative coupling, and glycosylations.

PRIMARY AND SECONDARYMETABOLISM

All organisms need to transform and interconvert avast number of organic compounds to enable themto live, grow, and reproduce. They need to pro-vide themselves with energy in the form of ATP,and a supply of building blocks to construct theirown tissues. An integrated network of enzyme-mediated and carefully regulated chemical reac-tions is used for this purpose, collectively referredto as intermediary metabolism, and the pathwaysinvolved are termed metabolic pathways. Some ofthe crucially important molecules of life are car-bohydrates, proteins, fats, and nucleic acids. Apartfrom fats, these are polymeric materials. Carbohy-drates are composed of sugar units, whilst proteinsare made up from amino acids, and nucleic acidsare based on nucleotides. Organisms vary widelyin their capacity to synthesize and transform chem-icals. For instance, plants are very efficient at syn-thesizing organic compounds via photosynthesisfrom inorganic materials found in the environ-ment, whilst other organisms such as animals andmicroorganisms rely on obtaining their raw mate-rials in their diet, e.g. by consuming plants. Thus,many of the metabolic pathways are concernedwith degrading materials taken in as food, whilst

others are then required to synthesize specializedmolecules from the basic compounds so obtained.

Despite the extremely varied characteristics ofliving organisms, the pathways for generally mod-ifying and synthesizing carbohydrates, proteins,fats, and nucleic acids are found to be essentiallythe same in all organisms, apart from minorvariations. These processes demonstrate the fun-damental unity of all living matter, and are col-lectively described as primary metabolism, withthe compounds involved in the pathways beingtermed primary metabolites. Thus degradationof carbohydrates and sugars generally proceedsvia the well characterized pathways known asglycolysis and the Krebs/citric acid/tricarboxylicacid cycle, which release energy from the organiccompounds by oxidative reactions. Oxidation offatty acids from fats by the sequence called β-oxidation also provides energy. Aerobic organismsare able to optimize these processes by adding ona further process, oxidative phosphorylation. Thisimproves the efficiency of oxidation by incorpo-rating a more general process applicable to theoxidation of a wide variety of substrates rather thanhaving to provide specific processes for each indi-vidual substrate. Proteins taken in via the diet pro-vide amino acids, but the proportions of each willalmost certainly vary from the organism’s require-ments. Metabolic pathways are thus available to

8 SECONDARY METABOLISM

interconvert amino acids, or degrade those notrequired and thus provide a further source ofenergy. Most organisms can synthesize only a pro-portion of the amino acids they actually require forprotein synthesis. Those structures not synthesized,so-called essential amino acids, must be obtainedfrom external sources.

In contrast to these primary metabolic pathways,which synthesize, degrade, and generally inter-convert compounds commonly encountered in allorganisms, there also exists an area of metabolismconcerned with compounds which have a muchmore limited distribution in nature. Such com-pounds, called secondary metabolites, are foundin only specific organisms, or groups of organ-isms, and are an expression of the individualityof species. Secondary metabolites are not neces-sarily produced under all conditions, and in thevast majority of cases the function of these com-pounds and their benefit to the organism is not yetknown. Some are undoubtedly produced for easilyappreciated reasons, e.g. as toxic materials provid-ing defence against predators, as volatile attractantstowards the same or other species, or as colouringagents to attract or warn other species, but it islogical to assume that all do play some vital rolefor the well-being of the producer. It is this areaof secondary metabolism that provides most ofthe pharmacologically active natural products. It isthus fairly obvious that the human diet could beboth unpalatable and remarkably dangerous if allplants, animals, and fungi produced the same rangeof compounds.

The above generalizations distinguishing pri-mary and secondary metabolites unfortunatelyleave a ‘grey area’ at the boundary, so that somegroups of natural products could be assigned toeither division. Fatty acids and sugars provide goodexamples, in that most are best described as pri-mary metabolites, whilst some representatives areextremely rare and found only in a handful ofspecies. Likewise, steroid biosynthesis produces arange of widely distributed fundamental structures,yet some steroids, many of them with pronouncedpharmacological activity, are restricted to certainorganisms. Hopefully, the blurring of the bound-aries will not cause confusion; the subdivisioninto primary metabolism (≡ biochemistry) or sec-ondary metabolism (≡ natural products chemistry)

is merely a convenience and there is considerableoverlap.

THE BUILDING BLOCKS

The building blocks for secondary metabolites arederived from primary metabolism as indicated inFigure 2.1. This scheme outlines how metabolitesfrom the fundamental processes of photosynthe-sis, glycolysis, and the Krebs cycle are tappedoff from energy-generating processes to providebiosynthetic intermediates. The number of build-ing blocks needed is surprisingly few, and as withany child’s construction set a vast array of objectscan be built up from a limited number of basicbuilding blocks. By far the most important buildingblocks employed in the biosynthesis of secondarymetabolites are derived from the intermediatesacetyl coenzyme A (acetyl-CoA), shikimic acid,mevalonic acid, and 1-deoxyxylulose 5-phosphate.These are utilized respectively in the acetate,shikimate, mevalonate, and deoxyxylulose phos-phate pathways, which form the basis of succeed-ing chapters. Acetyl-CoA is formed by oxidativedecarboxylation of the glycolytic pathway productpyruvic acid. It is also produced by the β-oxidationof fatty acids, effectively reversing the processby which fatty acids are themselves synthesizedfrom acetyl-CoA. Important secondary metabolitesformed from the acetate pathway include phenols,prostaglandins, and macrolide antibiotics, togetherwith various fatty acids and derivatives at theprimary/secondary metabolism interface. Shikimicacid is produced from a combination of phos-phoenolpyruvate, a glycolytic pathway intermedi-ate, and erythrose 4-phosphate from the pentosephosphate pathway. The reactions of the pentosephosphate cycle may be employed for the degra-dation of glucose, but they also feature in the syn-thesis of sugars by photosynthesis. The shikimatepathway leads to a variety of phenols, cinnamicacid derivatives, lignans, and alkaloids. Meval-onic acid is itself formed from three molecules ofacetyl-CoA, but the mevalonate pathway channelsacetate into a different series of compounds thandoes the acetate pathway. Deoxyxylulose phos-phate arises from a combination of two glycolyticpathway intermediates, namely pyruvic acid andglyceraldehyde 3-phosphate. The mevalonate anddeoxyxylulose phosphate pathways are together

THE BUILDING BLOCKS 9

CO2H

NH2

POO

OH

OH

CO2H

NH2HO

CO2H

HOOH

OH

O

OHOH

OHOH

PO

HO2C

HO

OH

NH

CO2H

NH2

HO2C CO2H

NH2

SHIKIMIC ACID

OHC

PO

OH

OHO

OH

OP

H2NCO2H

NH2

HO2C

PO

OH

O

OHOH

OHOH

HO

HO2C OP

CO2H

NH2

HO2C CO2H

O

HO2C O

NH

CO2H

NH2

H2N

NH

HOCO2H

NH2

CoAS O

CO2H

NH2

HO2CCO2H

O

CO2H

NH2

HSCO2H

NH2

CO2H

NH2

HO2CCO2H

NH2

H2N CO2H

NH2

CO2H

NH2

S CO2H

NH2

erythrose 4-Pglucose 6-P

3-phosphoglyceric acid

phosphoenolpyruvate

pyruvic acid

L-phenylalanine

oxaloacetic acid

ACETYL-CoA

PHOTOSYNTHESIS

L-tryptophan

MEVALONIC ACID

L-tyrosine

L-alanine

L-valine

2-oxoglutaric acid

PENTOSE PHOSPHATE CYCLE

L-serine

L-leucine

glycine

L-cysteine

KREBS CYCLE

GLYCOLYSIS

L-isoleucine

L-methionine L-lysine L-ornithine

L-aspartic acid L-glutamic acid

L-arginine

glyceraldehyde 3-P

D-glucose

DEOXYXYLULOSE 5-P

Figure 2.1

responsible for the biosynthesis of a vast array ofterpenoid and steroid metabolites.

In addition to acetyl-CoA, shikimic acid, meval-onic acid, and deoxyxylulose phosphate, otherbuilding blocks based on amino acids are fre-quently employed in natural product synthesis.Peptides, proteins, alkaloids, and many antibioticsare derived from amino acids, and the originsof the most important amino acid components ofthese are briefly indicated in Figure 2.1. Interme-diates from the glycolytic pathway and the Krebscycle are used in constructing many of them, butthe aromatic amino acids phenylalanine, tyrosine,

and tryptophan are themselves products fromthe shikimate pathway. Ornithine, a non-proteinamino acid, along with its homologue lysine, areimportant alkaloid precursors having their originsin Krebs cycle intermediates.

Of special significance is the appreciation thatsecondary metabolites can be synthesized by com-bining several building blocks of the same type,or by using a mixture of different building blocks.This expands structural diversity, and consequentlymakes subdivisions based entirely on biosyntheticpathways rather more difficult. A typical naturalproduct might be produced by combining elements


from the acetate, shikimate, and deoxyxylulosephosphate pathways. Many secondary metabolitesalso contain one or more sugar units in theirstructure, either simple primary metabolites suchas glucose or ribose, or alternatively substantiallymodified and unusual sugars. To appreciate howa natural product is elaborated, it is of valueto be able to dissect its structure into the basicbuilding blocks from which it is made up, andto propose how these are mechanistically joinedtogether. With a little experience and practice, thisbecomes a relatively simple process, and it allowsthe molecule to be rationalized, thus exposing logi-cal relationships between apparently quite differentstructures. In this way, similarities become muchmore meaningful than differences, and an under-standing of biosynthetic pathways allows rationalconnecting links to be established. This forms thebasic approach in this book.

Relatively few building blocks are routinelyemployed, and the following list, though not com-prehensive, includes those most frequently encoun-tered in producing the carbon and nitrogen skeletonof a natural product. As we shall see, oxygenatoms can be introduced and removed by a vari-ety of processes, and so are not considered in theinitial analysis, except as a pointer to an acetate(see page 62) or shikimate (see page 123) origin.The structural features of these building blocks areshown in Figure 2.2.

• C1: The simplest of the building blocks is com-posed of a single carbon atom, usually in theform of a methyl group, and most frequently itis attached to oxygen or nitrogen, but occasion-ally to carbon. It is derived from the S -methylof L-methionine. The methylenedioxy group(OCH2O) is also an example of a C1 unit.

• C2: A two-carbon unit may be supplied byacetyl-CoA. This could be a simple acetylgroup, as in an ester, but more frequently itforms part of a long alkyl chain (as in a fattyacid) or may be part of an aromatic system (e.g.phenols). Of particular relevance is that in thelatter examples, acetyl-CoA is first convertedinto the more reactive malonyl-CoA before itsincorporation.

• C5: The branched-chain C5 ‘isoprene’ unit is afeature of compounds formed from mevalonateor deoxyxylulose phosphate. Mevalonate itself

is the product from three acetyl-CoA molecules,but only five of mevalonate’s six carbons areused, the carboxyl group being lost. The alter-native precursor deoxyxylulose phosphate, astraight-chain sugar derivative, undergoes askeletal rearrangement to form the branched-chain isoprene unit.

• C6C3: This refers to a phenylpropyl unit andis obtained from the carbon skeleton of eitherL-phenylalanine or L-tyrosine, two of theshikimate-derived aromatic amino acids. This,of course, requires loss of the amino group. TheC3 side-chain may be saturated or unsaturated,and may be oxygenated. Sometimes the side-chain is cleaved, removing one or two carbons.Thus, C6C2 and C6C1 units represent modifiedshortened forms of the C6C3 system.

• C6C2N: Again, this building block is formedfrom either L-phenylalanine or L-tyrosine, L-tyrosine being by far the more common. In theelaboration of this unit, the carboxyl carbon ofthe amino acid is removed.

• indole.C2N: The third of the aromatic aminoacids is L-tryptophan. This indole-containingsystem can undergo decarboxylation in a sim-ilar way to L-phenylalanine and L-tyrosine soproviding the remainder of the skeleton as anindole.C2N unit.

• C4N: The C4N unit is usually found as a hetero-cyclic pyrrolidine system and is produced fromthe non-protein amino acid L-ornithine. Inmarked contrast to the C6C2N and indole.C2Nunits described above, ornithine supplies not itsα-amino nitrogen, but the δ-amino nitrogen. Thecarboxylic acid function and the α-amino nitro-gen are both lost.

• C5N: This is produced in exactly the same wayas the C4N unit, but using L-lysine as precursor.The ε-amino nitrogen is retained, and the unittends to be found as a piperidine ring system.

These eight building blocks will form the basis ofmany of the natural product structures discussed inthe following chapters. Simple examples of howcompounds can be visualized as a combination ofbuilding blocks are shown in Figure 2.3. At thisstage, it is inappropriate to justify why a particularcombination of units is used, but this aspectshould become clear as the pathways are described.

THE BUILDING BLOCKS 11

CO2H

NH2

O

OP

OH

OH

SCoA

O

SCoA

O

S CO2H

NH2

H3C

OH

HO

OP

OH

mevalonicacid

HOHO2C

OH

CO2H

NH2HO

CO2H

NH2

CO2H

NH2HO

NNH

CO2H

NH2

H2NCO2H

NH2

CO2H

NH2

H2N

N

N

N

N

C6C3

C6C2N

C5N

C4N

SCoA

O

CO2H

C6C1

C6C2

N

N

X CH3

C1

C5

C2

C C

The building blocks

malonyl-CoA

acetyl-CoA

acetyl-CoA

(X = O, N, C)

3 x

L-Phe

isoprene unit

L-Tyr

L-Met

L-Tyr

L-Phe

L-Trp

L-Orn

L-Lys

indole.C2N

deoxyxylulosephosphate

methylerythritolphosphate

Figure 2.2


C6C3 C6C3

N Me O

CO2Me

OO

OH

CO2H

N

MeO

MeO

MeO

MeO NH

NHO2C Me

O

OO

OH

O

MeO

OMe

OMe

OO

O

OO

OH O

OH

Glucose

RhamnoseCO2H

HO

OH

tetrahydrocannabinolic acid

6 x C2 + 2 x C5

orsellinic acid

4 x C2

podophyllotoxin

2 x C6C3 + 4 x C1

naringin

C6C3 + 3 x C2 + sugars

parthenolide

3 x C5

papaverine

C6C2N + (C6C2) + 4 x C1

cocaine

C4N + 2 x C2 + (C6C1) + 2 x C1

lysergic acid

indole.C2N + C5 + C1

Figure 2.3

A word of warning is also necessary. Some naturalproducts have been produced by processes inwhich a fundamental rearrangement of the carbonskeleton has occurred. This is especially commonwith structures derived from isoprene units, and itobviously disguises some of the original buildingblocks from immediate recognition. The same istrue if one or more carbon atoms are removed byoxidation reactions.

THE CONSTRUCTION MECHANISMS

Natural product molecules are biosynthesized bya sequence of reactions which, with very fewexceptions, are catalysed by enzymes. Enzymes areprotein molecules which facilitate chemical mod-ification of substrates by virtue of their specificbinding properties conferred by the particular com-bination of functional groups in the constituentamino acids. In many cases, a suitable cofactor, e.g.NAD+, PLP, HSCoA (see below), as well as thesubstrate, may also be bound to participate in thetransformation. Although enzymes catalyse somefairly elaborate and sometimes unexpected changes,it is generally possible to account for the reactionsusing sound chemical principles and mechanisms.As we explore the pathways to a wide variety ofnatural products, the reactions will generally be

discussed in terms of chemical analogies. Enzymeshave the power to effect these transformations moreefficiently and more rapidly than the chemical anal-ogy, and also under very much milder conditions.Where relevant, they also carry out reactions in astereospecific manner. Some of the important reac-tions frequently encountered are now described.

Alkylation Reactions: NucleophilicSubstitution

The C1 methyl building unit is supplied from L-methionine and is introduced by a nucleophilicsubstitution reaction. In nature, the leaving groupis enhanced by converting L-methionine into S -adenosylmethionine (SAM) [Figure 2.4(a)]. Thisgives a positively charged sulphur and facilitatesthe nucleophilic substitution (SN2) type mech-anism [Figure 2.4(b)]. Thus, O-methyl and N -methyl linkages may be obtained using hydroxyland amino functions as nucleophiles. The genera-tion of C -methyl linkages requires the participationof nucleophilic carbon. Positions ortho or para toa phenol group, or positions adjacent to one ormore carbonyl groups, are thus candidates for C -methylation [Figure 2.4(c)].

A C5 isoprene unit in the form of dimethylallyldiphosphate (DMAPP) may also act as an

THE CONSTRUCTION MECHANISMS 13

OH

S

CO2HH2N

H3CATP

S CO2HH3C

NH2

Ad

S

CO2HH2N

H3C

N

NN

NH2

O

HO OH

NCH2

NH2

OH

S CO2HH3C

NH2

Ad

PPO

OH

PPO

SAM

O CH3

H

O CH3

NH CH3

OH

S CO2H

NH2

Ad

O

O

H

O

OH

S CO2HH3C

NH2

Ad

neutral molecule is good leaving group

S-adenosylmethionine(SAM)

carbonyl groups increase acidity and allow formation of enolate anion

or

dimethylallyl diphosphate(DMAPP)

P, PP

SN2 reaction

– H

ortho (and para) positions are activated by OH

L-Met

or

SN2 reaction – H

SN1 reaction

or

resonance stabilized allylic carbocation

diphosphate is good leaving group

S-adenosylhomocysteine

(c) C-alkylation using SAM

(b) O- and N-alkylation using SAM

(d) O-alkylation using DMAPP

Alkylation reactions: nucleophilic substitution

(a) formation of SAM

Figure 2.4

alkylating agent, and a similar SN2 nucleophilicdisplacement can be proposed, the diphosphatemaking a good leaving group [Figure 2.4(d)]. Insome cases, there is evidence that DMAPP mayionize first to the resonance-stabilized allylic car-bocation and thus an SN1 process operates instead.C -Alkylation at activated positions using DMAPPis analogous to the C -methylation process above.

Alkylation Reactions: ElectrophilicAddition

As indicated above, the C5 isoprene unit in theform of dimethylallyl diphosphate (DMAPP)can be used to alkylate a nucleophile. In theelaboration of terpenoids and steroids, two or moreC5 units are joined together, and the reactions


are rationalized in terms of carbocation chem-istry, including electrophilic addition of carbo-cations on to alkenes. DMAPP may ionize togenerate a resonance-stabilized allylic carboca-tion as shown in Figure 2.4(d), and this can thenreact with an alkene [e.g. isopentenyl diphos-phate (IPP)] as depicted in Figure 2.5(a). Theresultant carbocation may then lose a protonto give the uncharged product geranyl diphos-phate (GPP). Where the alkene and carbocationfunctions reside in the same molecule, this type

of mechanism can be responsible for cyclizationreactions [Figure 2.5(a)].

The initial carbocation may be generated bya number of mechanisms, important examplesbeing loss of a leaving group, especially diphos-phate (i.e. SN1 type ionization), protonation ofan alkene, and protonation/ring opening of epox-ides [Figure 2.5(b)]. S -Adenosylmethionine mayalso alkylate alkenes by an electrophilic additionmechanism, adding a C1 unit, and generating anintermediate carbocation.

L

OPP

H

SAM

S CO2HH3C

NH2

Ad

OPP

H

HH

H

H

H

CH3

H2O

H

O

H

OPP

OH

H

HO

Alkylation reactions: electrophilic addition

– H

electrophilic addition of cation on to alkene

isopentenyldiphosphate

(IPP)

geranyl diphosphate(GPP)

intramolecular addition: cyclization

(b) generation of carbocation

loss of leaving group protonation of alkene protonation and ring opening of epoxide

methylation of alkene via SAM

(c) discharge of carbocation

loss of proton cyclization / loss of proton quenching with nucleophile (water)

(a) inter- and intra-molecular additions

Figure 2.5


The final carbocation may be discharged byloss of a proton (giving an alkene or sometimes acyclopropane ring) or by quenching with a suitablenucleophile, especially water [Figure 2.5(c)].

Wagner–Meerwein Rearrangements

A wide range of structures encountered in nat-ural terpenoid and steroid derivatives can onlybe rationalized as originating from C5 isopreneunits if some fundamental rearrangement processhas occurred during biosynthesis. These rearrange-ments have, in many cases, been confirmed experi-mentally, and are almost always consistent with theparticipation of carbocation intermediates. Rear-rangements in chemical reactions involving car-bocation intermediates, e.g. SN1 and E1 reactions,are not uncommon, and typically consist of 1,2-shifts of hydride, methyl, or alkyl groups. Occa-sionally, 1,3- or longer shifts are encountered.These shifts, termed Wagner–Meerwein rear-rangements, are readily rationalized in terms ofgenerating a more stable carbocation, or relax-ing ring strain (Figure 2.6). Thus, tertiary carbo-cations are favoured over secondary carbocations,

and the usual objective in these rearrangementsis to achieve tertiary status at the positive cen-tre. However, a tertiary to secondary transitionmight be favoured if the rearrangement allowsa significant release of ring strain. These gen-eral concepts are occasionally ignored by nature,but it must be remembered that the reactions areenzyme-mediated and carbocations may not existas discrete species in the transformations. An inter-esting feature of some biosynthetic pathways, e.g.that leading to steroids, is a remarkable series ofconcerted 1,2-migrations rationalized via carboca-tion chemistry, but entirely a consequence of theenzyme’s participation (Figure 2.6).

Aldol and Claisen Reactions

The aldol and Claisen reactions both achievecarbon–carbon bond formation and in typical base-catalysed chemical reactions depend on the gener-ation of a resonance-stabilized enolate anion froma suitable carbonyl system (Figure 2.7). Whetheran aldol-type or Claisen-type product is formeddepends on the nature of X and its potential asa leaving group. Thus, chemically, two molecules

HH HH

HH

H

H

H

H

HH H H

1,3-hydrideshift

1,2-alkyl shift

secondary carbocation

1,2-methyl shift

secondary carbocation

tertiary carbocation, but strained 4-membered ring

tertiary carbocation


secondary carbocation, but reduced ring strain in 5-membered ring

a series of concerted 1,2 hydride and methyl shifts

1,2-hydride shift

Wagner–Meerwein rearrangements


resonance-stabilized allylic cation

Figure 2.6


CH CO

XH

R CH CO

XR CH C

O

XR

CH CO

XR

CH2 CO

XR CH2 CR

O

X

CH CRO

X

CH2 CR

OH

X

CH CRO

X

CH2 CO

CH

R

CO

XR

R = H, X = OEt, ethyl acetoacetate

aldol-type product

Claisen-type product

nucleophilic addition on to carbonyl

if no suitable leaving group, protonation occurs

resonance-stabilized enolate anion

loss of leaving group

Aldol and Claisen reactions

R = X = H, acetaldehydeR = H, X = OEt, ethyl acetate

R = X = H, aldol

H

B

+

−

Figure 2.7

H3C SCoA

O

H3CC

OR

O

H

H3CC

OR

O

H2C SCoA

O

H3CC

SR

O

N

NN

CH2

NH2

O

PO OH

NOPOPO

O O

OH OH

NH

NH

OH

OO

HS

H3CC

SR

O

+

Coenzyme AHSCoA

acetyl-CoA

thioesterester

resonance decreases acidity of α-hydrogens

resonance of this type is less favourable in the sulphur ester

cysteamine

pantotheine

pantothenic acid

Figure 2.8

of acetaldehyde yield aldol, whilst two moleculesof ethyl acetate can give ethyl acetoacetate. Theseprocesses are vitally important in biochemistry forthe elaboration of both secondary and primarymetabolites, but the enzyme catalysis obviates theneed for strong bases, and probably means the eno-late anion has little more than transitory existence.Nevertheless, the reactions do appear to parallelenolate anion chemistry, and are frequently respon-sible for joining together of C2 acetate groups.

In most cases, the biological reactions involvecoenzyme A esters, e.g. acetyl-CoA (Figure 2.8).This is a thioester of acetic acid, and it has sig-nificant advantages over oxygen esters, e.g. ethylacetate, in that the α-methylene hydrogens are nowmore acidic, comparable in fact to those in theequivalent ketone, thus increasing the likelihood ofgenerating the enolate anion. This is explained interms of electron delocalization in the ester func-tion (Figure 2.8). This type of delocalization is


more prominent in the oxygen ester than in thesulphur ester, due to oxygen’s smaller size and thuscloser proximity of the lone pair for overlap withcarbon’s orbitals. Furthermore, the thioester has amuch more favourable leaving group than the oxy-gen ester, and the combined effect is to increasethe reactivity for both the aldol and Claisen-typereactions.

Claisen reactions involving acetyl-CoA aremade even more favourable by first convertingacetyl-CoA into malonyl-CoA by a carboxylationreaction with CO2 using ATP and the coenzymebiotin (Figure 2.9). ATP and CO2 (as bicarbonate,HCO3

−) form the mixed anhydride, which car-boxylates the coenzyme in a biotin–enzyme com-plex. Fixation of carbon dioxide by biotin–enzymecomplexes is not unique to acetyl-CoA, andanother important example occurs in the gener-ation of oxaloacetate from pyruvate in the syn-thesis of glucose from non-carbohydrate sources

(gluconeogenesis). The conversion of acetyl-CoAinto malonyl-CoA means the α-hydrogens arenow flanked by two carbonyl groups, and haveincreased acidity. Thus, a more favourable nucle-ophile is provided for the Claisen reaction. No acy-lated malonic acid derivatives are produced, andthe carboxyl group introduced into malonyl-CoA issimultaneously lost by a decarboxylation reactionduring the Claisen condensation (Figure 2.9). Analternative rationalization is that decarboxylationof the malonyl ester is used to generate the acetylenolate anion without any requirement for a strongbase. Thus, the product formed from acetyl-CoAas electrophile and malonyl-CoA as nucleophileis acetoacetyl-CoA, which is actually the same asin the condensation of two molecules of acetyl-CoA. Accordingly, the role of the carboxylationstep is clear cut: the carboxyl activates the α-carbon to facilitate the Claisen condensation, and itis immediately removed on completion of this task.

loss of CoAS

as leaving group

ADP

ATP HCO3

OHC

O

OP

OH

O

HONHHN

S CO

H H

O

Enz

CH3 C

O

SCoA

H2C C

O

SCoA

CO O H

CH3 C

O

SCoA

CH2 C

O

SCoACO2

CoAS C

O

CH2

CoAS C

O

CH2

CO2H

CH3 C

O

CH2 C

O

SCoA

NHN

S CO

H H

O

Enz

CO

HO

nucleophilic attack on carbonyl but with simultaneous loss of CO2

biotin-enzyme N1-carboxybiotin−enzyme

acetyl-CoA(enolate)

nucleophilic attack on to mixed anhydride

malonyl-CoA

+

nucleophilic attack on to carbonyl; loss of biotin−enzyme as leaving group

acetoacetyl-CoA

biotin−enzyme

mixed anhydride

Figure 2.9


CH3 C

O

OEt CH2 CO2EtNaOEt

CH3 C

O

OEt CH CO2Et

CO2EtNaOEt

EtOH

EtOHCH3 C

O

CH2 CO2Et

CH3 C

O

CH CO2Et

CO2Et

H

HCH3 C

O

CH

C O

HOC

O

HO

CH3 C

O

CH2 CO2H

thermal decarboxylation

∆– CO2

enolate anion from ethyl acetate

enolate anion from diethyl malonate

ethyl acetoacetate

acylated diethyl malonate

acetoacetic acid

gem-diacid

Figure 2.10

R SCoA

OFAD FADH2

R SCoA

OH2O

SR SCoA

OHHONAD+ NADH

R

O

SCoA

R SCoA

OO

HSCoA

CH3 SCoA

O

acetyl-CoA

E

dehydrogenation; hydrogen atoms passed to FAD

stereospecifichydration of double bond

dehydrogenation; hydrogen atoms passed to NAD+

reverse Claisen reaction

β-Oxidation of fatty acids

fatty acyl-CoA(chain length C2n)

fatty acyl-CoA(chain length C2n−2)

Figure 2.11

By analogy, the chemical Claisen condensationusing the enolate anion from diethyl malonate inFigure 2.10 proceeds much more favourably thanthat using the enolate from ethyl acetate. The sameacetoacetic acid product can be formed in the mal-onate condensation by hydrolysis of the acylatedmalonate intermediate and decarboxylation of thegem-diacid.

Both the reverse aldol and reverse Claisenreactions may be encountered in the modificationof natural product molecules. Such reactionsremove fragments from the basic skeleton alreadygenerated, but may extend the diversity of struc-tures. The reverse Claisen reaction is a promi-nent feature of the β-oxidation sequence for thecatabolic degradation of fatty acids (Figure 2.11),

in which a C2 unit as acetyl-CoA is cleaved offfrom a fatty acid chain, leaving it two carbonsshorter in length.

Schiff Base Formation and theMannich Reaction

Formation of C−N bonds is frequently achieved bycondensation reactions between amines and alde-hydes or ketones. A typical nucleophilic additionis followed by elimination of water to give animine or Schiff base [Figure 2.12(a)]. Of almostequal importance is the reversal of this process,i.e. the hydrolysis of imines to amines and alde-hydes/ketones [Figure 2.12(b)]. The imine so pro-duced, or more likely its protonated form the


R NH2 C O R N

H

H

C OH R N

H

C OH2 R N C

R N C R N C

H

OH

R NH2 C

OH

N CH

R

C N C C

R

H

N C

R1

RNH C O R1 N

R

H

C OH

R

R1

R1 N

R

C OH2

– H , + H

loss of amine leaving group; formation of carbonyl

imine or Schiff base

nucleophilic attack on to carbonyl

equilibrium between protonated species; proton may be on N or O

elimination of water

iminium ion

nucleophilic attack on to imine or protonated imine

carbanion-type nucleophile,

e.g. enolate anion

(a) Schiff base formation

(c) Mannich reaction

nucleophilic addition on to iminium ion

secondaryamine quaternary Schiff base

primaryamine

(b) Schiff base hydrolysis

HH2O

H

H

H

H

Figure 2.12

iminium ion, can then act as an electrophile ina Mannich reaction [Figure 2.12(c)]. The nucle-ophile might be provided by an enolate anion, orin many examples by a suitably activated centrein an aromatic ring system. The Mannich reactionis encountered throughout alkaloid biosynthesis,and in its most general form involves combinationof an amine (primary or secondary), an aldehydeor ketone, and a nucleophilic carbon. Secondaryamines will react with the carbonyl compoundto give an iminium ion (quaternary Schiff base)directly, and the additional protonation step is thusnot necessary.

It should be appreciated that the Mannich-likeaddition reaction in Figure 2.12(c) is little different

from nucleophilic addition to a carbonyl group.Indeed, the imine/iminium ion is merely actingas the nitrogen analogue of a carbonyl/protonatedcarbonyl. To take this analogy further, protonson carbon adjacent to an imine group will beacidic, as are those α to a carbonyl group, and theisomerization to the enamine shown in Figure 2.13is analogous to keto–enol tautomerism. Just astwo carbonyl compounds can react via an aldolreaction, so can two imine systems, and this isindicated in Figure 2.13. Often aldehyde/ketonesubstrates in enzymic reactions become covalentlybonded to the enzyme through imine linkages; inso doing they lose none of the carbonyl activationas a consequence of the new form of bonding.


R

N CCH

H

R

N CCH2 H

R

NC

H

R

N CCH

H

H

N C

R

H

CH C NHR

R

N CCH

Henamineimine

imine−enamine tautomerism

enamine protonated imine

aldol-type addition product

aldol-type reaction between two imine systems behaving as enamine−iminium ion pair

Figure 2.13

Transamination

Transamination is the exchange of the aminogroup from an amino acid to a keto acid, andprovides the most common process for the intro-duction of nitrogen into amino acids, and forthe removal of nitrogen from them. The cou-ple glutamic acid/2-oxoglutaric acid are theusual donor/acceptor molecules for the aminogroup. Reductive amination of the Krebs cycleintermediate 2-oxoglutaric acid to glutamic acid(Figure 2.14) is responsible for the initial incorpo-ration of nitrogen, a reaction which involves imineformation and subsequent reduction. Transamina-tion then allows the amino group to be transferredfrom glutamic acid to a suitable keto acid, or in thereverse mode from an amino acid to 2-oxoglutaricacid. This reaction is dependent on the coenzymepyridoxal phosphate (PLP) and features a Schiffbase/imine intermediate (aldimine) with the alde-hyde group of PLP (Figure 2.14). The α-hydrogenof the original amino acid is now made consider-ably more acidic and is removed, leading to theketimine by a reprotonation process which alsorestores the aromaticity in the pyridine ring. Theketo acid is then liberated by hydrolysis of theSchiff base function, which generates pyridoxam-ine phosphate. The remainder of the sequence isnow a reversal of this process, and transfers theamine function from pyridoxamine phosphate toanother keto acid.

Decarboxylation Reactions

Many pathways to natural products involve stepswhich remove portions of the carbon skeleton.Although two or more carbon atoms may be

cleaved off via the reverse aldol or reverse Claisenreactions mentioned above, by far the most com-mon degradative modification is loss of one carbonatom by a decarboxylation reaction. Decarboxy-lation is a particular feature of the biosyntheticutilization of amino acids, and it has alreadybeen indicated that several of the basic buildingblocks, e.g. C6C2N, indole.C2N, are derived froman amino acid via loss of the carboxyl group. Thisdecarboxylation of amino acids is also a pyridoxalphosphate-dependent reaction (compare transam-ination) and is represented as in Figure 2.15(a).This similarly depends on Schiff base formationand shares features of the transamination sequenceof Figure 2.14. Decarboxylation is facilitated in thesame way as loss of the α-hydrogen was facilitatedfor the transamination sequence. After protonationof the original α-carbon, the amine is released fromthe coenzyme by hydrolysis of the Schiff basefunction.

β-Keto acids are thermally labile and rapidlydecarboxylated in vitro via a cyclic mechanismwhich proceeds through the enol form of thefinal ketone [Figure 2.15(b)]. Similar reactions arefound in nature, though whether cyclic processesare necessary is not clear. ortho-Phenolic acids alsodecarboxylate readily in vitro and in vivo, and it isagain possible to invoke a cyclic β-keto acid tau-tomer of the substrate. The corresponding decar-boxylation of para-phenolic acids cannot have acyclic transition state, but the carbonyl group inthe proposed keto tautomer activates the system fordecarboxylation. The acetate pathway frequentlyyields structures containing phenol and carboxylicacid functions, and decarboxylation reactions maythus feature as further modifications. Although thecarboxyl group may originate by hydrolysis of the


HO2C

HO2C

ONAD(P)H

NH3 H2O

N

CHO

OH

CH3

PO

R CO2H

NH2H

N

OH

CH3

PO

N

CO2HR

H

HO2C

HO2C

NH2 R

HO2CO

H

HO2C

HO2C

O

H

HO2C

HO2C

NH2

N

OH

CH3

PO

N

CO2HR

H

R

HO2CNH2

N

OH

CH3

PO

NH2

N

OH

CH3

PO

N

CO2HR

R CO2H

O

2-oxoglutaricacid

formation of imine from aldehyde and amino acid

α-hydrogen is now acidic

restoring aromaticity

hydrolysis of imine to keto acid and amine

pyridoxal P(PLP)

aldimine ketimine

pyridoxamine P

NAD(P)+

2-oxoglutaric acid

glutamatedehydrogenase

transaminase

glutamic acidL-Glu

Transamination

reductive amination

keto acid amino acidglutamic acid

Figure 2.14

thioester portion of the acetyl-CoA precursor, thereare also occasions when a methyl group can besequentially oxidized to a carboxyl, which thensubsequently suffers decarboxylation.

Decarboxylation of α-keto acids is a fea-ture of primary metabolism, e.g. pyruvic acid→ acetaldehyde in glycolysis, and pyruvic acid→ acetyl-CoA, an example of overall oxidativedecarboxylation prior to entry of acetyl-CoA

into the Krebs cycle. Both types of reactiondepend upon thiamine diphosphate (TPP). TPPis a coenzyme containing a thiazole ring, whichhas an acidic hydrogen and is thus capable ofyielding the carbanion. This acts as a nucle-ophile towards carbonyl groups. Decarboxylationof pyruvic acid to acetaldehyde is depicted asin Figure 2.15(c), which process also regeneratesthe carbanion. In the oxidation step of oxidative


N

CHO

OH

CH3

PO

N

OH

CH3

PO

N

R

H

O

OH

R CO2H

NH2H

N

OH

CH3

PO

N

HR

H

H

RC

CH2

CO

O OH

RC

CH2

OH

OH

CO2H

O

CO

H

OH

CO2H

HO

CO

O

O

HH

H

H

CO2

HO

OH

RC

CH3

O

N

CHO

OH

CH3

PO

N

OH

CH3

PO

N

HR H

R

NH2

CO2

CO2

....

pyridoxal P(PLP)

pyridoxal P(PLP)

hydrolysis of imine to aldehyde and amine

restoring aromaticitydecarboxylation

formation of imine

Decarboxylation reactions

6-membered H-bonded system

keto tautomer

phenolic acid (i.e. enol tautomer)

intermediate enol

keto tautomer≡ β-keto acid

β-keto acid

6-membered H-bonded system

(a) amino acids

(b) β-keto acids

....keto-enoltautomerism

Figure 2.15


OHSS

O

R1

N S

R2

CH3C OH

N

N

N S

OPP

NH2

R1

N S

R2

CH3C

O R3

SHS

H

H

R3

SS

B

H3CC

H

O

R1

N S

R2

R1

N S

R2

H

TPP

R3

SHSC

H3C

O

R1

N S

R2

CH3C

OH

H

R1

N S

R2

HSCoA

H3CC

CO2H

OH

R1

N S

R2

SCoAC

H3C

O

R3

HSHS

R3

SS

R1

N S

R2

CH3C OH

R1

N S

R2

CCH3C

HO

OH

O

FAD

H

acidic hydrogen

nucleophilic attack of carbanion on to carbonyl: aldol-type reaction

TPP anion

decarboxylation of β-iminium acid

lipoic acid

enamine

thiamine diphosphate(TPP)

TPP anion regenerated

enzyme-bound lipoic acid

enzyme-bound lipoic acidenamine enamine attacks S of

lipoic acid fragment with S–S bond fission

regeneration of TPP carbanion leaves acetyl group attached to dihydrolipoic acid

acetyl group displaced by coenzyme A

original lipoic acid fragment has become reduced to dithiol; oxidation regenerates the enzyme-bound lipoic acid

iminium ion

reverse aldol-type reaction

enamine−imine tautomerism

acetyl-CoA

(c) α-keto acids

Figure 2.15 (continued )


decarboxylation, the enzyme-bound disulphide-containing coenzyme lipoic acid is also involved.The intermediate enamine in Figure 2.15(c), in-stead of accepting a proton, is used to attack asulphur in the lipoic acid moiety with subsequentS−S bond fission, thereby effectively reducing thelipoic acid fragment. This allows regeneration ofthe TPP carbanion, and the acetyl group is boundto the dihydrolipoic acid. This acetyl group is thenreleased as acetyl-CoA by displacement with thethiol coenzyme A. The bound dihydrolipoic acidfragment is then reoxidized to restore its function.An exactly equivalent reaction is encountered inthe Krebs cycle in the conversion of 2-oxoglutaricacid into succinyl-CoA.

Oxidation and Reduction Reactions

Changes to the oxidation state of a molecule arefrequently carried out as a secondary metaboliteis synthesized or modified. The processes are notalways completely understood, but the followinggeneral features are recognized. The processes maybe classified according to the type of enzymeinvolved and their mechanism of action.

Dehydrogenases

Dehydrogenases remove two hydrogen atomsfrom the substrate, passing them to a suitable

coenzyme acceptor. The coenzyme system in-volved can generally be related to the functionalgroup being oxidized in the substrate. Thus if theoxidation process is

CH OH C O

then a pyridine nucleotide, nicotinamide adeninedinucleotide (NAD+) or nicotinamide adeninedinucleotide phosphate (NADP+), tends to beutilized as hydrogen acceptor. One hydrogen fromthe substrate (that bonded to carbon) is transferredas hydride to the coenzyme, and the other, asa proton, is passed to the medium (Figure 2.16).NAD(P)+ may also be used in the oxidations

HC O

CH NH2 C NH

CO2H

The reverse reaction, i.e. reduction, is alsoindicated in Figure 2.16, and may be comparedwith the chemical reduction process using complexmetal hydrides, e.g. LiAlH4 or NaBH4, namely nu-cleophilic addition of hydride and subsequent pro-tonation. The reduced forms NADH and NADPHare conveniently regarded as hydride-donating

O

HO OH

N

CONH2

N

N N

NH2

RO OHO

N P

O

OH

O P

O

OH

O CH2OCH2

HC

OH

N

CONH2

R

NAD+

NADP+

P

NADHNADPH

P

CO

N

CONH2

R

HH

P

H

≡

adenine

ribose ribose

nicotinamide

R = H, NAD+

R = P, NADP+

Dehydrogenases: NAD+ and NADP+

Figure 2.16


FAD

N

N N

NH2

HO OHO

N P

O

OH

O P

O

OH

O

N

N

N

NHH3C

H3C

O

O

OH OH

OH

OCH2

FMN

CC

H

H

N

N

N

NHH3C

H3C

O

O

R HFADFMN

CC

H

P

N

N

N

NHH3C

H3C

O

O

R

H

H

FADH2FMNH2

P

Dehydrogenases: FAD and FMN

adenine

ribose ribitol

flavin

≡

Figure 2.17

reducing agents. In practice, NADPH is generallyemployed in reductive processes, whilst NAD+ isused in oxidations.

Should the oxidative process be the conversion

CH2 CH2 CH CH

the coenzyme used as acceptor is usually a flavinnucleotide, flavin adenine dinucleotide (FAD) orflavin mononucleotide (FMN). These entities arebound to the enzyme in the form of a flavoprotein,and take up two hydrogen atoms, representedin Figure 2.17 as being derived by addition ofhydride from the substrate and a proton from themedium. Alternative mechanisms have also beenproposed, however. Reductive sequences involvingflavoproteins may be represented as the reversereaction in Figure 2.17. NADPH may also beemployed as a coenzyme in the reduction of acarbon–carbon double bond.

These oxidation reactions employing pyridinenucleotides and flavoproteins are especially impor-tant in primary metabolism in liberating energyfrom fuel molecules in the form of ATP. Thereduced coenzymes formed in the process are nor-mally reoxidized via the electron transport chain

of oxidative phosphorylation, so that the hydrogenatoms eventually pass to oxygen giving water.

Oxidases

Oxidases also remove hydrogen from a substrate,but pass these atoms to molecular oxygen or tohydrogen peroxide, in both cases forming water.Oxidases using hydrogen peroxide are termed per-oxidases. Mechanisms of action vary and neednot be considered here. Important transforma-tions in secondary metabolism include the oxi-dation of ortho- and para-quinols to quinones(Figure 2.18), and the peroxidase-induced phenolicoxidative coupling processes (see page 28).

Mono-oxygenases

Oxygenases catalyse the direct addition of oxygenfrom molecular oxygen to the substrate. They aresubdivided into mono- and di-oxygenases accord-ing to whether just one or both of the oxygen atomsare introduced into the substrate. With mono-oxygenases, the second oxygen atom from O2

is reduced to water by an appropriate hydrogen


OH

OH O

O

OH

OH

O

O

Oxidases

ortho-quinol

+ H2O

+ H2O+ 1/2 O2

+ 1/2 O2

ortho-quinone

para-quinol para-quinone

Figure 2.18

donor, e.g. NADH, NADPH, or ascorbic acid(vitamin C). In this respect they may also beconsidered to behave as oxidases, and the term‘mixed-function oxidase’ is also used for theseenzymes. Especially important examples of theseenzymes are the cytochrome P-450-dependentmono-oxygenases. These are frequently involvedin biological hydroxylations, either in biosyn-thesis, or in the mammalian detoxification andmetabolism of foreign compounds such as drugs,and such enzymes are thus termed ‘hydroxylases’.Cytochrome P-450 is named after its intenseabsorption band at 450 nm when exposed to CO,which is a powerful inhibitor of these enzymes.It contains an iron–porphyrin complex (haem),which is bound to the enzyme, and a redox changeinvolving the Fe atom allows binding and the

cleavage of an oxygen atom. Many such sys-tems have been identified, capable of hydroxy-lating aliphatic or aromatic systems, as well asproducing epoxides from alkenes (Figure 2.19).In most cases, NADPH features as hydrogendonor.

Aromatic hydroxylation catalysed by mono-oxygenases (including cytochrome P-450 systems)probably involves arene oxide (epoxide) interme-diates (Figure 2.20). An interesting consequenceof this mechanism is that when the epoxide opensup, the hydrogen atom originally attached to theposition which becomes hydroxylated can migrateto the adjacent carbon on the ring. A high pro-portion of these hydrogen atoms is subsequentlyretained in the product, even though enolizationallows some loss of this hydrogen. This migrationis known as the NIH shift, having been origi-nally observed at the National Institute of Health,Bethesda, MD, USA.

C HO2

H

NADPH

O2

NADPH

O2

NADPH

O

C OH

OH

Mono-oxygenases

Figure 2.19

R

H

H

R

H

H

O

R

HH

O

R

R R

O

H

H H

OHOH

H

retention of labelled hydrogen

arene oxide

hydride migrationNIH shift

enolization– H – H

NIH shift

loss of labelled hydrogen

Figure 2.20


The oxidative cyclization of an ortho-hydroxy-methoxy-substituted aromatic system giving amethylenedioxy group is also known to involvea cytochrome P-450-dependent mono-oxygenase.This enzyme hydroxylates the methyl to yielda formaldehyde hemiacetal intermediate, whichcan cyclize to the methylenedioxy bridge (theacetal of formaldehyde) by an ionic mechanism(Figure 2.21).

Dioxygenases

Dioxygenases introduce both atoms from molecu-lar oxygen into the substrate, and are frequentlyinvolved in the cleavage of bonds, including

aromatic rings. Cyclic peroxides (dioxetanes) arelikely to be intermediates (Figure 2.22). Oxidativecleavage of aromatic rings typically employs cat-echol (1,2-dihydroxy) or quinol (1,4-dihydroxy)substrates, and in the case of catechols, cleavagemay be between or adjacent to the two hydrox-yls, giving products containing aldehyde and/orcarboxylic acid functionalities (Figure 2.22).

Some dioxygenases utilize two acceptor sub-strates and incorporate one oxygen atom intoeach. Thus, 2-oxoglutarate-dependent dioxyge-nases hydroxylate one substrate, whilst also trans-forming 2-oxoglutarate into succinate with therelease of CO2 (Figure 2.23). 2-Oxoglutarate-dependent dioxygenases also require as cofactors

OCH3

OH

O2

NADPH

OCH2

OH

OH

H

OCH2

O

H

O

O

formaldehydehemiacetal

nucleophilic attack on to carbonyl equivalent

methylenedioxyderivative

ortho-hydroxy-methoxy derivative

Methylenedioxy groups

Figure 2.21

O

O

O

O

O

O

CH

CH O

O

dioxetane

+ O2

+ O2

OH

OH

O2

O

O

OH

OH

OH

OH

O

O

OHOH

O2

CO2H

CO2H

CO2H

CHO

OH

cleavage between hydroxyls

cleavage adjacent to hydroxyls

Dioxygenases

Figure 2.22


HO2C

HO2C

OR H

O2 CO2CO2H

HO2C

R OH

2-oxoglutaric acid

2-Oxoglutarate-dependent dioxygenases

succinic acid

2-oxoglutarate-dependent

dioxygenase

Figure 2.23

Fe2+ to generate an enzyme-bound iron–oxygencomplex, and ascorbic acid (vitamin C) to subse-quently reduce this complex.

Amine Oxidases

In addition to the oxidizing enzymes outlinedabove, those which transform an amine into analdehyde, the amine oxidases, are frequentlyinvolved in metabolic pathways. These includemonoamine oxidases and diamine oxidases.Monoamine oxidases utilize a flavin nucleotide,typically FAD, and molecular oxygen, and involveinitial dehydrogenation to an imine, followedby hydrolysis to the aldehyde and ammonia(Figure 2.24). Diamine oxidases require a diaminesubstrate, and oxidize at one amino group usingmolecular oxygen to give the correspondingaldehyde. Hydrogen peroxide and ammonia arethe other products formed. The aminoaldehyde soformed then has the potential to be transformedinto a cyclic imine via Schiff base formation.

H2N NH2

RCH2NH2

NH3

RCHO

H2NCHO

Amine oxidases

FAD, O2

monoamineoxidase

n n

O2, H2O

diamine oxidase

NH3, H2O2

Figure 2.24

Baeyer–Villiger Oxidations

The chemical oxidation of ketones by peracids, theBaeyer–Villiger oxidation, yields an ester, andthe process is known to involve migration of analkyl group from the ketone (Figure 2.25). Forcomparable ketone → ester conversions knownto occur in biochemistry, cytochrome-P-450- orFAD-dependent enzymes requiring NADPH andO2 appear to be involved. This leads to formationof a peroxy–enzyme complex and a mechanismsimilar to that for the chemical Baeyer–Villigeroxidation may thus operate. The oxygen atomintroduced thus originates from O2.

Phenolic Oxidative Coupling

Many natural products are produced by thecoupling of two or more phenolic systems, ina process readily rationalized by means of freeradical reactions. The reactions can be brought

O

R2R1 OO

O

OO

O

HOR2

R1R1 O

R2

O O

HO

O

R2R1

OEnzOHO

R2R1

R1 OR2

O

Baeyer−Villigeroxidation

ester

ketone

nucleophilic attack of peracid on to carbonyl

carbonyl reforms: alkyl group migrates from carbon to adjacent oxygen

Baeyer−Villiger oxidations

O2, NADPH

(HOO–Enz)

peracid

HH

Figure 2.25


OH O O O

O

O

H

O

OH

O

H

OH

OHOH

H

OH

OH

O

OH

O

H

H

OH

O

HO

x 2

– e

x 2

phenol

resonance-stabilizedfree radical

bis-dienonebis-dienone

ether linkage

keto tautomers

enol tautomers

enolization

coupling of two radicals

Phenolic oxidative coupling

ortho–ortho coupling

bis-dienone

– H

ortho–para coupling para–para coupling

Figure 2.26

about by oxidase enzymes, including peroxidaseand laccase systems, known to be radicalgenerators. Other enzymes catalysing phenolicoxidative coupling have been characterized ascytochrome P-450-dependent proteins, requiringNADPH and O2 cofactors, but no oxygen isincorporated into the substrate. A one-electronoxidation of a phenol gives the free radical, andthe unpaired electron can then be delocalizedvia resonance forms in which the free electronis dispersed to positions ortho and para to theoriginal oxygen function (Figure 2.26). Couplingof two of these mesomeric structures gives a rangeof dimeric systems as exemplified in Figure 2.26.The final products indicated are then derived byenolization, which restores aromaticity to the rings.Thus, carbon–carbon bonds involving positionsortho or para to the original phenols, or etherlinkages, may be formed. The reactive dienonesystems formed as intermediates may in somecases be attacked by other nucleophilic groupings,extending the range of structures ultimatelyderived from this basic reaction sequence.

Glycosylation Reactions

The widespread occurrence of glycosides andpolysaccharides requires processes for attaching

sugar units to a suitable atom of an aglyconeto give a glycoside, or to another sugar givinga polysaccharide. Linkages tend to be throughoxygen, although they are not restricted tooxygen, since S -, N -, and C -glycosides arewell known. The agent for glycosylation isa uridine diphosphosugar, e.g. UDPglucose.This is synthesized from glucose 1-phosphateand UTP, and then the glucosylation processcan be envisaged as a simple SN2 nucleophilicdisplacement reaction [Figure 2.27(a)]. SinceUDPglucose has its leaving group in the α-configuration, the product has the β-configuration,as is most commonly found in natural glucosides.Note, however, that many important carbohydrates,e.g. sucrose and starch, possess α-linkages,and these appear to originate via double SN2processes (see page 470). Other UDPsugars, e.g.UDPgalactose or UDPxylose, are utilized inthe synthesis of glycosides containing differentsugar units.

The hydrolysis of glycosides is achieved by spe-cific hydrolytic enzymes, e.g. β-glucosidase for β-glucosides and β-galactosidase for β-galactosides.These enzymes mimic the readily achieved acid-catalysed processes [Figure 2.27(b)]. Under acidicconditions, the α- and β-anomeric hemiacetalforms can also equilibrate via the open chain


ROH

O

HO OH

CH2N

O

OP

O

OH

OP

O

OH

O

OHOHO

HOH

O

OH

OHOHO

HOH

OH

OP

OHOHO

HOOR

OH

H

H OH2OHO

HOHO

OH

H

OHOHO

HOOR

OH

H

OHOHO

HOOR

OHH

H

OHOHO

HOH

OH

OHOH2

HOHO

HOHO

OH

OH

H

OHHOHO

HOOH

OH

H

HO

OH

OHOHO

HOH

OH

OPPU

OHOHO

HO

OH

H OH

HO

b

+ UTP

UDPglucose

+ UDP

UDPglucose

glucose 1-P

C-β-D-glucoside

O-β-D-glucoside

SN2 reaction

O-β-D-glucoside

α-D-glucose

β-D-glucose

(a) O-glucosylation

(b) hydrolysis of O-glucosides

(c) C-glucosylation

Glycosylation reactions

a

b

a

Figure 2.27

sugar. Of particular importance is that althoughO-, N -, and S -glycosides may be hydrolysedby acid, C -glycosides are stable to acid. C -Glycosides are produced in a similar manner tothe C -alkylation process described above, where

a suitable nucleophilic carbon is available, e.g.aromatic systems activated by phenol groups[Figure 2.27(c)]. The resultant C -glycoside thuscontains a new carbon–carbon linkage, andcleavage would require oxidation, not hydrolysis.

SOME VITAMINS ASSOCIATED WITH THE CONSTRUCTION MECHANISMSVitamin B1

Vitamin B1 (thiamine) (Figure 2.28) is a water-soluble vitamin with a pyrimidinylmethylthia-zolium structure. It is widely available in the diet, with cereals, beans, nuts, eggs, yeast,and vegetables providing sources. Wheat germ and yeast have very high levels. Dietarydeficiency leads to beriberi, characterized by neurological disorders, loss of appetite, fatigue,

(Continues )


(Continued )

HN NH

S CO2H

O

HH

N

N

NS

OH

NH2

N

N

N

NH

O

O

OH

OH OH

OH

HO2CNH

OH

O

OH

N

HO

OH

OH

N

CHO

HOOH

N

HOOH

NH2

pyridoxine(pyridoxol) biotin

(vitamin H)

thiamine(vitamin B1) riboflavin

(vitamin B2)

pantothenic acid(vitamin B5)

(vitamin B6)

pyridoxal pyridoxamine

5′

Figure 2.28

and muscular weakness. Thiamine is produced synthetically, and foods such as cerealsare often enriched. The vitamin is stable in acid solution, but decomposes above pH 5, andis also partially decomposed during normal cooking. As thiamine diphosphate, vitamin B1

is a coenzyme for pyruvate dehydrogenase which catalyses the oxidative decarboxylationof pyruvate to acetyl-CoA (see page 21), and also for transketolase which transfers a two-carbon fragment between carbohydrates in the pentose phosphate pathway (see page 446).Accordingly, this is a very important component in carbohydrate metabolism.

Vitamin B2

Vitamin B2 (riboflavin) (Figure 2.28) is a water-soluble vitamin having an isoalloxazine ringlinked to D-ribitol. It is widely available in foods, including liver, kidney, dairy products, eggs,meat, and fresh vegetables. Yeast is a particularly rich source. It is stable in acid solution,not decomposed during cooking, but is sensitive to light. Riboflavin may be producedsynthetically, or by fermentation using the yeastlike fungi Eremothecium ashbyii and Ashbyagossypii. Dietary deficiency is uncommon, but manifests itself by skin problems and eyedisturbances. Riboflavin is a component of FMN (flavin mononucleotide) and FAD (flavinadenine dinucleotide), coenzymes which play a major role in oxidation–reduction reactions(see page 25). Many key enzymes containing riboflavin (flavoproteins) are involved in metabolicpathways. Since riboflavin contains ribitol and not ribose in its structure, FAD and FMN arenot strictly nucleotides, though this nomenclature is commonly accepted and used.

Vitamin B5

Vitamin B5 (pantothenic acid) (Figure 2.28) is a very widely distributed water-soluble vitamin,though yeast, liver, and cereals provide rich sources. Even though animals must obtainthe vitamin through the diet, pantothenic acid deficiency is rare, since most foods provide

(Continues )


(Continued )

adequate quantities. Its importance in metabolism is as part of the structure of coenzyme A(see page 16), the carrier molecule essential for carbohydrate, fat, and protein metabolism.Pantothenic acid is specifically implicated in enzymes responsible for the biosynthesis of fattyacids (see page 36), polyketides (page 62) and some peptides (page 421).

Vitamin B6

Vitamin B6 covers the three pyridine derivatives pyridoxine (pyridoxol), pyridoxal, andpyridoxamine, and also their 5′-phosphates (Figure 2.28). These are water-soluble vitamins,pyridoxine predominating in plant materials, whilst pyridoxal and pyridoxamine are themain forms in animal tissues. Meat, salmon, nuts, potatoes, bananas, and cereals aregood sources. A high proportion of the vitamin activity can be lost during cooking, buta normal diet provides an adequate supply. Vitamin B6 deficiency is usually the result ofmalabsorption, or may be induced by some drug treatments where the drug may actas an antagonist or increase its renal excretion as a side-effect. Symptoms of deficiencyare similar to those of niacin (vitamin B3) and riboflavin (vitamin B2) deficiencies, andinclude eye, mouth, and nose lesions, and neurological changes. Synthetic pyridoxineis used for supplementation. Pyridoxal 5′-phosphate is a coenzyme for a large numberof enzymes, particularly those involved in amino acid metabolism, e.g. in transamination(see page 20) and decarboxylation (see page 20). The production of the neurotransmitterγ-aminobutyric acid (GABA) from glutamic acid is an important pyridoxal-dependent reaction.

Vitamin B12

Vitamin B12 (cobalamins) (Figure 2.29) are extremely complex structures based on a corrinring, which, although similar to the porphyrin ring found in haem, chlorophyll, and cytochromes,

CONH2

CONH2

Co+

N

N N

N

H2NOC

H2NOC

P O

OH

O

HH2NOC

R

CONH2

HOO

O

N

N

HO

NH

O O

HO OH

N

NN

N

NH2

N

NH N

N

HN

NNH

N

R = CN, cyanocobalamin (vitamin B12)R = OH, hydroxocobalamin (vitamin B12a)R = H2O, aquocobalamin (vitamin B12b)R = NO2, nitritocobalamin (vitamin B12c)R = Me, methylcobalamin (methyl vitamin B12)

R =

5′-deoxyadenosylcobalamin

(coenzyme B12)

corrin ring system porphyrin ring system

Figure 2.29

(Continues )

FURTHER READING 33

(Continued )

has two of the pyrrole rings directly bonded. The central metal atom is cobalt; haem andcytochromes have iron, whilst chlorophyll has magnesium. Four of the six coordinations areprovided by the corrin ring nitrogens, and a fifth by a dimethylbenzimidazole moiety. The sixthis variable, being cyano in cyanocobalamin (vitamin B12), hydroxyl in hydroxocobalamin(vitamin B12a), or other anions may feature. Cyanocobalamin is actually an artefact formed as aresult of the use of cyanide in the purification procedures. The physiologically active coenzymeform of the vitamin is 5′-deoxyadenosylcobalamin (coenzyme B12). Vitamin B12 appears tobe entirely of microbial origin, with intestinal flora contributing towards human dietary needs.The vitamin is then stored in the liver, and animal liver extract has been a traditional source.Commercial supplies are currently obtained by semi-synthesis from the total cobalaminextract of Streptomyces griseus, Propionibacterium species, or other bacterial cultures. Thismaterial can be converted into cyanocobalamin or hydroxocobalamin. The cobalamins arestable when protected against light. Foods with a high vitamin B12 content include liver,kidney, meat, and seafood. Vegetables are a poor dietary source, and strict vegetarians maytherefore risk deficiencies. Insufficient vitamin B12 leads to pernicious anaemia, a disease thatresults in nervous disturbances and low production of red blood cells, though this is mostlydue to lack of the gastric glycoprotein (intrinsic factor) which complexes with the vitamin tofacilitate its absorption. Traditionally, daily consumption of raw liver was used to counteractthe problem. Cyanocobalamin, or preferably hydroxocobalamin which has a longer lifetime inthe body, may be administered orally or by injection to counteract deficiencies. Both agentsare converted into coenzyme B12 in the body. Coenzyme B12 is a cofactor for a number ofmetabolic rearrangements, such as the conversion of methylmalonyl-CoA into succinyl-CoAin the oxidation of fatty acids with an odd number of carbon atoms, and for methylations,such as in the biosynthesis of methionine.

Vitamin HVitamin H (biotin) (Figure 2.28) is a water-soluble vitamin found in eggs, liver, kidney, yeast,cereals, and milk, and is also produced by intestinal microflora so that dietary deficiency israre. Deficiency can be triggered by a diet rich in raw egg white, in which a protein, avidin,binds biotin so tightly so that it is effectively unavailable for metabolic use. This affinitydisappears by cooking and hence denaturing the avidin. Biotin deficiency leads to dermatitisand hair loss. The vitamin functions as a carboxyl carrier, binding CO2 via a carbamate link,then donating this in carboxylase reactions, e.g. carboxylation of acetyl-CoA to malonyl-CoA(see page 17), of propionyl-CoA to methylmalonyl-CoA (see page 92), and of pyruvate tooxaloacetate during gluconeogenesis.

FURTHER READING

Natural Products, Biosynthesis

Mann J (1994) Chemical Aspects of Biosynthesis . Ox-ford Chemistry Primers, Oxford.

Mann J, Davidson RS, Hobbs JB, Banthorpe DV andHarborne JB (1994) Natural Products: Their Chem-istry and Biological Significance. Longman, Harlow.

Torssell KBG (1997) Natural Product Chemistry. AMechanistic, Biosynthetic and Ecological Approach.

Apotekarsocieteten, Stockholm.

Vitamins

Battersby AR (2000) Tetrapyrroles: The pigments oflife. Nat Prod Rep 17, 507–526.

Burdick D (1998) Vitamins [pyridoxine (B6)]. Kirk–Othmer Encyclopedia of Chemical Technology , 4thedn, Vol 25. Wiley, New York, pp 116–132.

Burdick D (1998) Vitamins [thiamine (B1)]. Kirk–Othmer Encyclopedia of Chemical Technology , 4thedn, Vol 25. Wiley, New York, pp 152–171.

Kingston R (1999) Supplementary benefits? Chem Brit35 (7), 29–32.


Outten RA (1998) Vitamins (biotin). Kirk–Othmer En-cyclopedia of Chemical Technology , 4th edn, Vol 25.Wiley, New York, pp 48–64.

Rawalpally TR (1998) Vitamins (pantothenic acid).Kirk–Othmer Encyclopedia of Chemical Technology ,4th edn, Vol 25. Wiley, New York, pp 99–116.

Scott JW (1998) Vitamins (vitamin B12). Kirk–OthmerEncyclopedia of Chemical Technology , 4th edn, Vol25. Wiley, New York, pp 192–217.

Yoneda F (1998) Vitamins [riboflavin (B2)]. Kirk–Othmer Encyclopedia of Chemical Technology , 4thedn, Vol 25. Wiley, New York, pp 132–152.

secondary metabolism: the building blocks and construction...

Documents