return to L2W Research - Science Writing Portfolio

About this report...

A Review of the Hypothetical Biogenesis and Regulation of
Hypericin synthesis via the Polyketide Pathway in Hypericum perforatum and
Experimental Methods Proposed to Evaluate the Hypothesis

By: Loren W. Walker

Portland State University

May, 1999

    flower and hypericin glands


     I.    Overview of hypericin

     II.   Multiple pathways to quinones complicate the   hypothetical elucidation of hypericin biosynthesis

     III.  A hypothetical polyketide pathway to hypericin is formulated

     IV.  Possible hypericin metabolism regulatory mechanisms

     V.   Experiments proposed to evaluate the validity of the hypothetical pathway and regulatory mechanisms

     VI.  Conclusion

     Works Cited


    dark glands contain the biologically active compound hypericin in
    Hypericum perforatum L. 

I. Overview of hypericin
Polyketides are defined as a class of molecules produced through the successive condensation of small carboxylic acids (Katz and Donodio, 1993). Because of its structure and, perhaps more importantly, because of its hypothetical mode of biogenesis hypericin is considered a polyketide derived quinone. It is described as a highly condensed polycyclic quinone (C30H16O8) and given the descriptive name naphthodianthrone, alluding to the configuration of the compound constructed from two tricyclic anthrones (reduced anthraquinones) called emodinanthrone (Figure 1). This pigment, colorless in its reduced form within the cell, is known to have a wide range of biological activities. Like many of the hundreds of other polyketides hypericin exhibits antifungal, antitumor, and anthelmintic properties (Hobbs, 1996 Katz and Donodio, 1993). The history of use of Hypericum perforatum (containing hypericin) preparations as a component of galenicals dates back to before the turn of the turn of the millennium
Return to Contents
Figure 1
The structure of hypericin (Robinson)
The legacy of St. John's Wort the medicine has made Hypericum species prime candidates for natural products research. Hypericin was first identified from Hypericum species and serves as a chemotaxonomic marker of the genera. Since its initial identification and characterization (Brockmannin Thomson, 1957) it has been isolated from coccid insects and recognized in fossil crinoids (Thomson, 1987).

The biological activity of hypericin has ecological ramifications as well. The pigment is known to be toxic to most animals because of its photooxidative properties. The putative defense that hypericin affords Hypericum species against insectivory and pathogen attack is not its only ecological consequence. Several species of beetles (Coleoptera) in the genus Chrysolina have co-evolved with the hypericin producing plants and consequently have adapted the ability to metabolize the pigment safely. In fact, it has been demonstrated that the Chrysolina beetle uses the presence of hypericin in Hypericum plants as a way of identifying their food source (Rodriguez, 1975).

II.    Multiple pathways to quinones complicate the hypothetical elucidation of hypericin synthesis

Multiple pathways to the biosynthesis of aromatic rings -

The three ring structures that are the intermediates of hypericin (a naphthodianthrone) are called anthrones (in their reduced form) or anthraquinones (in their oxidized form). Both of these structures can be isolated in small quantities from H. perforatum (Thompson, 1957). Ascertaining the biochemical origin of these compounds is complicated because anthraquinones with similar structures can be derived from at least two different metabolic pathways.
Return to Contents
Figure 2.
Similar structures with different biosynthetic origins (Vickery and Vickery)

Emodin, which is oxidized from emodinanthrone, the direct precursor to protohypericin, is an anthraquinone that is formed from a C15-16 polyketide chain. The anthraquinone alizarin, which bears a strong structural resemblance to emodin and its derivatives (Figure 2), are biosynthesized via the shikimic acid pathway (Vickery and Vickery, 1981). Hypothetical biosynthetic pathways for these compounds can be formulated. However, because such similar structures (alizarin and emodin) can have different origins, the actual pathway can be confirmed only through careful experimentation (Section IV).

The Polyketide Hypothesis -
Aromatic rings are known to be synthesized via the shikimate/phenylpropanoid, acetate/malonate, and isoprenoid pathways (Dennis, 1997). The acetate/malonate, or polyketide, pathway to aromatic ring synthesis was first described by Birch (1967). He explained the mechanism by which a polyketide chain was formed from an acetyl CoA starter unit by the sequential addition of malonyl CoA units. He further demonstrated that cyclization of these chains could be accomplished by either Claisen or Aldol condensation (Figure 3); the former process is similar to that which builds the chain itself. This hypothesis for the formation of aromatic rings from acetate and malonate precursors, now widely accepted, was not unanimously embraced from its inception. In fact, as reported by Birch, referees of "some of the best journals" rejected earlier forms of the manuscript that ultimately appeared in Science.
Figure 3.
Condensation of polyketide chain to a ring structure (Dey and Harbourne)

At the time of Birch's publication many details regarding the polyketide pathway remained to be elucidated especially details of the enzymes and the processes they control. Recently, there has been much progress made toward the goal of characterizing the enzymes that catalyze the chain elongation and cyclization reactions leading towards aromatic polyketides. These so-called polyketide synthase complexes will be addressed (Section III).

Polyketides and the Fatty Acid Pathway -
Polyketide synthesis is believed to be analogous to fatty acid synthesis. Therefore, much of the dogma that describes polyketide synthesis and characterizes the enzymes purportedly involved has been extrapolated from the well characterized polypeptides and mechanisms of long chain fatty acid synthesis (Katz and Donadio). Polyketides are distinguished from fatty acids and their derivatives by their mode of biosynthesis. This involves a hypothetical chain of alternating keto and methylene groups.

-CO-CH2-CO-CH2-CO-CH2-CO-CH2- - Polyketide chain

Whereas fatty acid precursors contain only methylene groups, the malonyl keto group being reduced after each condensation (Vickery and Vickery).


Chemical principles applied to biological systems -
The combination of reduction and oxidation reactions that have been hypothesized since as early as the 1950's to lead to polyketide chains and their derived aromatic rings are based on well-established chemical principles. The implicit, but unstated (by Collie, 1907) assumption that chemical mechanisms should be recognizable in biosynthetic processes, even though enzymes and coenzymes are clearly important, is a reoccurring theme in the hypothetical mechanism of polyketide biosynthesis. That these early hypothetical reaction sequences are being confirmed by evermore scrutinizing analyses would seem to demonstrate that Collie's early assumption was valid.

III.   A Hypothetical polyketide pathway to hypericin is proposed

The hypothetical biosynthesis of hypericin via the polyketide pathway:

Shortly after the isolation and characterization of hypericin as extracted from H. perforatum a hypothetical biosynthetic pathway was proposed (Brockmann, 1950; Thomson, 1957). By isolating intermediate compounds workers formulated a theoretical pathway based largely on the paradigm that chemical principles should apply to biosynthetic processes. The hypothetical metabolism of hypericin in H. perforatum reported hereafter is formulated from a review of the original treatments of the polyketide pathway (Birch) and hypericin synthesis (Thomson) in combination with information gleaned from more recent publications.
Return to Contents
Generation of the precursors -
The polyketide pathway, also known as the acetate/malonate pathway, derives the latter descriptive name from compounds that serve as the precursors to the pathway. Acetate is generated as a product of glycolysis, as the result of fatty acid metabolism by b -oxidation, or by a number of other pathways yet to be confirmed. A pool of acetyl CoA is present in the plastids. The pool is relatively "shallow" - only 30 to 50 m M - but the level of the pool remains fairly constant even when demand is high (Ohlrogge and Browse, 1995). The system by which acetyl CoA is made readily available in the plastid is not fully understood, though there are a number of hypotheses. Plastidal pyruvate dehydrogenase may act on pyruvate derived from glycolysis or from a side reaction of ribulose bisphosphate carboxylase to form acetyl CoA. Also, chloroplasts contain "an extremely active acetyl CoA synthetase," and free acetate has been found superior to pyruvate and other substances as a precursor of fatty acid synthesis by isolated chloroplasts (Ohlrogge and Browse). Also proposed is the synthesis of acetyl CoA in the mitochondria followed by transport into the chloroplast (Ohlrogge and Browse). All of these probable pathways make deciphering how carbon moves from photosynthesis to acetyl CoA exceedingly complicated. Acetyl CoA is a seminal compound for a variety of metabolic pathways. Because of the central role that acetyl CoA plays in many metabolic pathways it is likely that more than one route of biosynthesis may contribute to maintaining the acetyl CoA pool. Furthermore, the active pathway at any given time may vary with tissue, developmental stage, light/dark conditions and species. (Ohlrogge and Browse) Malonyl CoA can be derived from acetyl CoA by the
addition of CO2 as catalyzed by acetyl CoA carboxylase - ACCase (equation 1). Additionally, malonate is derived from oxaloacetate (OAA) from the tricarboxilic acid (TCA) cycle. Both compounds are activated by addition of coenzyme A (CoA).
Figure 4.
Polyketide synthase (PKS) complex (Katz and Donodio)
Equation 1
CO2 + CH3.CO.S.CoA + ATP + H2O  ------ (ACCase)------->  HOOC.CH2.CO.S.CoA + ADP + Pi

Chain Elongation (condensation) -

Polyketide chains and subsequent aromatic compounds are built mainly from acetate and malonate units through a reiterative process. This is the "polyketide hypothesis" of aromatic ring formation established in the Birch Science publication. Since that report it has been further hypothesized that an enzyme complex - the polyketide synthase (PKS), catalyzes the reactions leading to a polyketide chain that will form an aromatic ring compound. PKSs bears a strong structural and operational resemblance to the type II fatty acid synthases (FAS II) also referred to as the prokaryotic pathway to fatty acid synthesis (Ohlrogge and Browse, Katz and Donodio). In very few instances have particular PKSs been characterized biochemically. Therefore, much of the hypothetical model for aromatic polyketide synthesis is based upon what is known about the formation of long chain fatty acids (Katz and Donodio). This model supposes that a number of discrete polypeptides work cooperatively to catalyze the reactions required for chain elongation and subsequent cyclization (Figure 4). Each polypeptide carries a distinct activity in the loosely associated complex called PKS. Like fatty acid synthesis, these reactions are purported to take place in the chloroplast stroma and the cytosol.

Figure 5.
Probable final stages of 6-methylsalicylic acid biosynthesis (Vickery and Vickery)


The metabolism of hypericin precursors, like the polyketide metabolites in general, is presumed to follow a pathway analogous to fatty acid synthesis. A series of Claisen condensations between 2 carbon units (malonate) yields polyketomethylene chains (Figure 3), which lead by reduction to fatty acids and by further cyclization to many classes of aromatic compounds (Bruneton). Acetyl CoA serves as a starter unit for polyketide chain elongation just as in fatty acid synthesis. Malonyl CoA serves as the extendor unit of the chain. A concomitant decarboxylation occurs during the attack on the carbonyl group of acetyl-S.CoA. This reaction leaves the malonyl moiety as a stronger nucleophile so it becomes the chain extendor unit (Bruneton). The malonate extendor unit is transferred from CoA the pantotheine arm on an acyl carrier protein (ACP) by acetyl transferase (AT) as shown in figure 4(Katz and Donadio). The nascent chain, while attached to the active-site cysteine residue of the b -ketoacyl ACP synthase (KS) condensing enyme, is joined with the ACP-bound malonate by decarboxilative condensation (Katz and Donodio). Once the acetyl CoA starter unit is in place the chain is elongated by subsequent additions of malonyl ACP (2C) extendor units. The successive processing steps of b -ketoreduction, dehydration and enoylreduction require catalysis by b -ketoreductase (KR), dehydratase (DH), and enoylreductase (ER) respectively (figure 4). In fatty acids the elongated chain would then by transacylated to KS thereby initiating a new cycle. The analogous enzymes that are proposed to catalyze the biosynthesis of poyketides are now collectively referred to as the polyketide synthases (PKSs). They PKS appears to have a genetic organization similar to the enzymes of the fatty acid synthase complex (Katz and Donodio). Because this enzymatically-catalyzed pathway to polyketide chains and ultimately to aromatic rings is not well characterized it is not known what sort of reducing power or other cofactors are required for each reaction. Most of the experimental information in this area comes from what is known about the biosynthesis of 6-methylsalicylic acid. The PKS that catalyzes this reaction is the only one extensively characterized to date (Figure 5). It has been shown that NADPH is required for the reductase catalyzed reactions in 6-methylsalicylic acid. NADPH may also be required for functioning of the putative cyclase/dehydratase enzyme that catalyzes anthrone ring formation.

Exactly how the correct starter unit (acetyl CoA in the case of hypericin) is selected, how the ACP is charged with malonate, and which mechanisms determine folding and chain release are not currently known. Additionally, the mechanism that accomplishes the programming that governs the number of condensations, the proper folding of the completed acyl chain, the cyclization into the correct number of rings, and the release of the resulting structure is not well understood. However, this programming clearly must reside in the PKS components themselves (Katz and Donodio).

Cyclization and the formation of emodinanthrone -
The hypothetical cyclization of an acetate unit derived chain and subsequent addition of those rings to form hypericin was originally proposed (Thomson) at the same time that Birch was elucidating his polyketide hypothesis. The hypericin metabolism hypothesis was strongly corroborated by Birch's findings. The direct precursor to protohypericin is believed to be emodinanthrone. Emodinanthrone is the first cyclization product (Figure 6) of the chain formed by the condensation of one acetyl CoA molecule and seven malonyl CoA molecules (Chen et al, 1995). A cyclase/dehydratase is believed to be involved here. The bifunctional cyclase/dehydratase is also part of the loosely associated PKS complex (Katz and Donodio).

Figure 6.
Cyclization of a polyketide chain to form emodinanthrone (Robinson, 1991)

Oxidation of emodinanthrone to the naphthodianthrone protohypericin and irradiation to hypericin -

It is postulated that a dianthrone would arise from emodinanthrone, probably by oxidative coupling of the anthranol, and would lead by further oxidation of its enol form to a dehydrodianthrone and then from a helianthrone derivative (protohypericin) to hypericin (Thomson). Protohypericin is readily converted (by oxidation) to hypericin upon irradiation. In fact, protohypericin can be isolated from H. perforatum extracts only in the dark. The knowledge that anthrone - dianthrone interconversion takes place readily coupled with a basic understanding of how anthracene nuclei are cross-linked by oxidation contributed to the original scheme for the biosynthesis of hypericin from emodinanthrone as put forth by Brockman and co-workers in 1942 and 1950 (in Thomson, 1957). Emodinanthrone can be detected in Hypericum plant extracts (Thomson). Additionally, this compound can be converted in vitro into the hypericin pigments.

IV. Possible regulatory mechanisms of hypericin
biosynthesis via the polyketide pathway

Induction of polyketide pigment biosynthesis by the excess of reduced pyridine nucleotides -

While anthropocentric reports of biologically active secondary compounds from plants are widespread the activity and function of these compounds in vivo is often more difficult to ascertain. Polyketides are known to have anti-pathogenic qualities; hence their use as antibiotics (i.e. erythromycin). Therefore one hypothetical role would be for defense against pathogens. Hypericin specifically is toxic to most animals in addition to being an effective anti-microbial. Again, a hypothetical defense role (against insectivory) can be postulated. However, secondary compound production as a response to regulation of the intracellular environment should be considered as well.
Return to Contents
It has been observed, even in the course of my own research into the production of secondary compounds by Hypericum perforatum, that the concentration at which these low MW compounds occur in plants is usually subject to variation over the course of the growing season. One hypothesis asserts that increased accumulation of secondary metabolites is the natural consequence of conditions under which excess precursors accumulate. This phenomenon has been referred to as overflow metabolism (Matsuki, 1996). Overflow metabolism is presumed to take place during times of slow growth when photosynthates, which continue to accumulate in the presence of light, are abundant while other precursors necessary to sustain vegetative growth are limiting or during periods of seasonal growth depression of a normally fast-growing plant. It is notable that the accumulation of the hypericin pigment in H. perforatum is most pronounced during the flowering season at which time the plant has produced substantial leaf area for photosynthesis and vegetative growth is minimal. The effect of resource stress on hypericin accumulation in the species has not yet been reported.

It has been proposed that an increase in "defense compounds" during times of slow growth is simply a mechanism to protect the plant against pathogens during a time when damaged cells cannot be rapidly replaced. However, another hypothesis put forth nearly a decade ago by a pair of scientists working in the former Soviet Union (Trutko and Akimenko, 1991) takes into consideration the influence of intracellular conditions on the accumulation of secondary compounds in plants. Photons impinging on the light harvesting complex of photosystem 1 (PS1) provide the energy for a series of reactions that ultimately reduce the pyridine nucleotide pool in the chloroplast stroma. The NAD(P)H produced as a result of the activity in PS1 is used a reducing agent in myriad anabolic and catabolic events throughout the plant cell. A large portion of the NAD(P)H produced in this way is used in the reductive pentose phosphate (Calvin) cycle for the reduction of carbon. The reduced carbon is shuttled from the chloroplast as triose phosphate units and distributed to a number of anabolic pathways. When there is a decrease in the amount of NAD(P)H needed for primary metabolism the amount of reduced pyridine nucleotides in the cytosolic pool increases. As observed in microbes, an excessive amount reducing equivalents can inhibit the essential oxidative activity of cells (Akimenko, 1985 in Trutko and Akimenko). It has been shown that biosynthesis of pigments in microbes (Pseudomonas and Serratia) is a means of eliminating excess reducing equivalents (Trutko and Akimenko, 1991, 1988). Furthermore, the authors report that the induction of polyketide biosynthesis in microbes is coupled with cessation of growth. The results of experiments designed to test the effect of excess NAD(P)H on the synthesis of polyketide pigments in Streptomyces cultures suggest that the rate of reduced pyridine nucleotide production is determined by the activity of the enzyme systems generating and oxidizing NAD(P)H. In microorganisms they observed that inhibition of growth decreases the oxidizing activity of the cells leading to an increase in the reduction of the pyridine nucleotide pool. Upon the addition of synthetic electron acceptors to the experimental cultures the authors observed an inversely proportional relationship between the concentration of acceptor solution added and the level of reduction in the pyridine nucleotide pool. Furthermore, the degree of inhibition of biosynthesis of pigments was observed to be proportional to the concentration of the shunting agent. The authors assert, with reference to this observation, that enzymes which regulate the production of (polyketide) pigments are regulated by NAD(P)H cofactors.

In response to the widely known fact that many pigments are efficient electron donor-acceptors (i.e. plastoquinones) the authors performed experiments to test whether the pigments that were being accumulated in their experimental cultures were influencing the redox state of the reduced pyridine nucleotide pool. They found that none of the accumulated pigments could serve as a natural acceptor of reducing equivalents in the cultures investigated. Therefore they conclude that an excess of NAD(P)H, not utilized for the growth and functioning of the cell, induces the biosynthesis of these pigments.

Jasmonates -
It is well known that hypericin is a potent anti-microbial and that it is phototoxic to most animals (Upton, 1997). The pigment which is purportedly sequestered in a number of "dark glands" located on the leaves, stems, calyx, corolla and anthers of Hypericum perforatum is presumed to protect to the plant from insectivory and pathogen attack. The defensive role that hypericin is purported to play in the species gives cause to consider whether or not the action of wounding, as by insectivory, may stimulate increased accumulation of the pigment. While there are no published reports confirming this phenomenon I have made limited observations that support the hypothesis. I have observed that in areas where evidence of insectivory pressure is abundant (i.e. defoliated leaves or the observation of swarms of grasshoppers) the concentration of hypericin in the leaves is greater (in some cases) than the concentration in the flowers of the same plant. This is contrary to the expected concentration stratification between these two tissue types on the same plant (Upton).

It has been reported that an ubiquitous plant hormone, jasmonic acid, may serve as a signaling mechanism within plants that stimulates the production and accumulation of low MW weight compounds in response to wounding or pathogen attack. To date most of the compounds accumulated in response to jasmonate signaling are derived from the shikimic acid/phenylpropanoid pathway. Researchers who have worked with jasmonates report that their activity does not seem to be taxonomically limited, nor do they believe it will be found to be limited to action on the shikimate/phenylpropanoid pathway. There are no published reports on the effect of jasmonates on hypericin accumulation in H. perforatum, but such experiments would seem to have salient value.


V.    Experiments proposed to evaluate the validity of the hypothetical
pathway and the possible regulatory mechanisms described

The hypothetical biogenesis of polyketides and hypericin specifcally in H. perforatum plants is well established in the literature, but there is still much to be confirmed experimentally. While the proposed metabolic pathway - from the polyketide chain formed by an acetyl CoA starter unit and elongated by 2-Carbon malonyl CoA extendors to the tricyclic anthrones that combine to form the naphthodianthrone precursor to hypericin - makes sense in terms of chemical principles it needs to be confirmed experimentally within a natural system. Additionally, there are still unanswered questions about the regulation of hypericin synthesis and accumulation. Some regulatory actions that remain unexplained include: that which stimulates increased hypericin production during the summer months when the plant flowers; the putative effect of insectivory pressure on hypericin production; and the effect of stress/slow growth on hypericin synthesis. Additionally, the enzymes that make up the putative PKS complex which catalyzes chain elongation and ring cyclization await characterization.
Return to Contents

Figure 7.
Use of radiolabeled acetate onle falsely confirms that 6-methylsalicylic acid is formed by four moles of acetate (Mann)
The hypothetical acetate/malonate pathway to hypericin synthesis could be confirmed by careful administration of radiolobeled precursors to plant cell cultures. Treating cell cultures with 14C-acetate precursors and then looking for radioactive hypericin and intermediates (by scintillation counting or autoradiography) at intervals after the addition of the radiolabel is one well-established way of elucidating a metabolic pathway. However, early experiments designed to elucidate the biosynthesis of polyketide rings using 14C-acetate confirmed, incorrectly, that 6-methylsalicylic acid was derived from four moles of acetate (Figure 7). The addition of 14C-labeled malonate to precursors and the detection of the radiolabel in hypericin would strongly suggest that the pigment is derived via the acetate/malonate pathway. 14C isotopes of both acetate and malonate are available commercially (Sigma-Aldrich). Additionally, it has been reported that the use of 13C labeled acetate is becoming increasingly common in the study of polyketide biosynthesis (Bruneton). The reason for this label's popularity is that it is readily detected, without preliminary degradation, by nuclear magnetic resonance (NMR) spectroscopy (Bruneton).

Northern blotting experiments using cloned DNA from 6-methylsalicylic acid synthase, the only well-characterized PKS to date (Katz and Donodio), could be employed to detect DNA segments in Hypericum cultures that would hybridize with the cloned fractions indicating the presence of PKS-like DNA in Hypericum. Initial experiments that would investigate the regulation of these enzymes might involve protocols by which the known regulators of the similar fatty acid synthase complex would be added to cell cultures of Hypericum species. Quantification of reaction products would allow conclusions to be drawn with regard to how the regulators affect the reactions supposed to be governed by some sort of PKS complex.

The stage of growth effects the intracellular environment. The effect of intracellular conditions on polyketide pigment synthesis such as the level of reduction in the pyridine nucleotide pool could be analyzed toward ascertaining the influence of slow or fast vegetative growth. The regulation of light regimes and/or the addition of synthetic reducing equivalent acceptors (as reported in Trutko and Akimenko) could be used to control the amount of available NAD(P)H. In this way experiments could be performed to investigate the hypothesis that the synthesis of polyketide pigments is regulated by the level of reduced pyridine nucleotides available.

How hypericin pigment biosynthesis is affected by insectivory pressure is another area that needs to be evaluated. First, it must be confirmed that there is a correlation between insectivory damage to foliage and increased hypericin accumulation. The defoliation of plants by hand is one way that insectivory could be simulated. Assuming that a positive relationship between defoliation (insectivory) and increased hypericin production does exist then the question of a mechanism that stimulates increased production must be addressed.

Endogenous signaling mechanisms propagated in response to wounding have been identified for a variety of plant species. The relatively recent discovery of jasmonate signaling activity in a wide variety of species gives cause to look for its presence in any species that reacts to wounding and/or pathogen attack with increased synthesis of defense compounds. An assay for the presence of jasmonates and/or other putative endogenous signalers in Hypericum species would be valuable.


VI. Conclusion

It has been over 50 years since hypericin was first isolated from Hypericum species. The principles of chemistry applied to that early research resulted in a strong biosynthetic hypothesis. In that time a lot of progress has been made toward an increased understanding of polyketide biosynthesis. The information on the mechanisms of polyketide synthesis can be applied generally to hypericin synthesis. From this hypothetical foundation, a number of experiments using well-established protocols could be conducted to elucidate with certainty the metabolic pathway that leads to hypericin.
Return to Contents
The strong chemical defense system that protects Hypericum species from pathogens and insectivory is the same system from which we derive valuable medicinals. Furthermore, it is in part because of these chemical defenses that Hypericum is considered a noxious weed across much of the globe. A more complete understanding of the mechanisms that regulate the synthesis of defense compounds like hypericin would provide insights that have intrinsic scientific value as well as practical agricultural and pharmaceutical value.

Works Cited:

Return to Contents
Akimenko, V.K. 1985. Environmental regulation of microbial metabolism. Eds: Kulalev, I.S., et al. New York : Academic. 127.

Birch, A.J. 1967. "Biosynthesis of polyketides and related compounds." Science. 156: 202-206.

Brockmann, H., et al. 1950. Naturwissenschaften. 540.

Bruneton, J. 1995. Pharmacognosy, Phytochemistry, Medicinal Plants. Trans. Caroline K. Hatton. Andover: Intercept.

Chen, Z., et al. 1995. "Purification and characterization of emodinanthrone oxygenase from Aspergillus terreus." Phytochemistry. 38(2): 299-305.

Collie, J.N. 1907. Journal of the American Chemical Society. 91: 1806.

Dennis, D.T., et al, eds. 1990. Plant Metabolism. Essex: Addison Wesley Longman.

Dey, P.M. and J.B. Harbourne. eds. 1997. Plant Biochemistry. New York: Academic.

Hobbs, C. 1996. "St. John's Wort (Hypericum perforatum L.): A review." HerbalGram.

Katz, L and S. Donadio.  1993.  "Polyketide synthesis: prospects for hybrid antibiotics."  Ann. Rev. Microbiol.  47:875-912. Mann, J. 1994. Chemical Aspects of Biosynthesis. Oxford: Oxford University.

Matsuki, M. 1996. "Regulation of Plant Phenolic Synthesis: from Biochemistry to Ecology and Evolution." Australian Journal of Botany. 44: 613-634.

Ohlrogge, J. and J. Browse. 1995. "Lipid Biosynthesis." The Plant Cell. 7:957-970.

Robinson, T. 1991. The Organic Constituents of Higher Plants: Their Chemistry and Interrelationships. North Amherst: Cordus.

Rodriguez, E. and D.A. Levin. "Biochemical parallelisms of repellants and attractants in higher plants and arthropods." In Wallace, J.W. and R.L. Mansell. eds. 1975. Biochemical Interaction Between Plants and Insects. New York: Plenum.

Thomson, R.H. 1957. Naturally Occurring Quinones. London: Butterworths Scientific.

Trutko, S.M. and V.K. Akimenko. 1988. "Overproduction of Microbial Products." Second International Symposium. Prague. 247.

Trutko, S.M. and V.K. Akimenko. 1991. "Biosynthesis of pigments via the polyketide pathway - a means of regulating the oxidative metabolism of Streptomyces cultures producing these pigments." Microbiology. 59(5):650-654.

Upton, R. ed.  1997.  "American Herball Pharmacopoeia Monograph: St. John's Wort, Hypericum perforatum."  HerbalGram. 40: Supplement.

Vickery, M.L. and B. Vickery. 1981. Secondary Plant Metabolism. Baltimore: University Park.



Return to Contents
Anderson, J.A. 1988. "Deoxygenation of phenolic natural products. Enzymatic conversion of emodin to chrysophanol." Journal of the American Chemical Society. 110: 1623-1624.

Banks, H.J., et al. 1976. "Chemistry of the coccoideae. II condensed polycyclic pigments from two australian pseudococcids (hemiptera)." Australian Journal of Chemistry. 29: 1509-1521.

                    Birch, A.J. 1951. Journal of the American Chemical Society. 3026.

                    Brockmann, H., et al. 1939. Leibigs Annals. 553:1.

Cameron, D.W. and W.D. Raverty. 1976. "Pseudohypericin and other phenanthroperlyne quinones." Australian Journal of Chemistry. 29: 1523-1533.

Cameron, D.W., et al. 1976. "Oxidation of emodin anthrone and stereochemistry of emodin bianthrone." Australian Journal of Chemistry. 29: 1535-1548.

Durand, R. and M.H. Zenk. 1974. "The homogentisate ring-cleavage pathway in the biosynthesis of acetate-derived naphthoquinones of the droseraceae." Phytochemistry. 13: 1483-1492.

Fairbairn, J.W. and F.J. Muhtadi. 1972. "The biosynthesis and metabolism of anthraquinones in Rumex obtusifolius." Phytochemistry. 11: 215-219.

Falk, H. and G. Schoppel. 1991. "A synthesis of emodin anthrone." Monatshefte fur Chemie. 122: 739-744.

Falk, H. and W. Schmitzberger. 1992. "On the nature of 'soluble' hypericin in Hypericum species." Monatshefte fur Chemie. 123: 731-739.

Falk, H. and G. Schoppel. 1992. "On the synthesis of hypericin by oxidative trimethyl-emodin anthrone and emodin anthrone dimerization: isohypericin." Monatshefte fur Chemie. 123: 931-938.

Goodwin, T.W. and E.I. Mercer. 1983. Introduction to Plant Biochemistry. 2nd ed. Oxford: Pergamon.

Inoue, K., et al. 1984. "Biosynthesis of naphthoquinones and anthraquinones in Streptocarpus dunnii cell cultures." Phytochemistry. 23(2): 313-318.

Kartnig, Th. and I. Gobel. 1992. "Determination of hypericin and pseudohypericin by thin-layer chromatography-densitometry. Journal of Chromatography. 609: 423-426.

Khouri, H. and R. Ibrahim. 1987. "Purification and some properties of five anthraquinone-specific glucosyltransferases from Cinchona succirubra cell suspension culture." Phytochemistry. 26(9): 2531-2535.

Knox, J.P. and A.D. Dodge. 1985. "Isolation and activity of the photodynamic pigment hypericin." Plant, Cell and Environment. 8: 19-25.

Leistner, E. 1973. "Biosynthesis of morindone and alizarin in intact plants and cell suspension cultures of Morinda citrifolia." Phytochemistry. 12: 1669-1674.

Rawlings, B.J. 1997. "Biosynthesis of polyketides." Natural Products Reports. 14(5): 523-556.

Simantiras, M. and E. Leistner. 1989. "Formation of o-succinylbenzoic acid from iso-chorismic acid in protein extracts from anthraquinone-producing plant cell suspension cultures." Phytochemistry. 28(5): 1381-1382.

Tropf, S., et al. 1995. "Reaction mechanisms of homodimeric plant polyketide synthases (stilbene and chalcone synthase)." The Journal of Biological Chemistry. 270(14): 7922-7928.

Stumpf, P.K. and E.E. Conn, eds. 1981. The Biochemistry of Plants: A Comprehensive Treatise. New York: Academic.

Trutko, S.M. and V.K. Akimenko. 1987. "Biosynthesis of secondary metabolites in : Reports of All-Union conference." Puschino. 55.

Return to top of page


© Loren W. Walker 1999-2004

Return to L2W Research Home Page