The National Cancer Institute

The National Cancer Institute (NCI) was established in the USA in 1937 ‘to provide for, foster and aid in coordinating research related to cancer’. This organization has been involved in many of the great discoveries in fundamental research of cancer and particularly in the characterization of cytotoxic natural products. During the 1950s it was realized that there was a need for screening of natural extracts in a discovery programme and the Cancer Chemotherapy National Service Center was set up (Cragg et al 1993). The objectives of this organization were to procure compounds and to screen and evaluate them in a preclinical and, finally, full clinical setting. This service developed into the modern organization known as the Developmental Therapeutics Program (DTP).

Apart from endeavours to find anticancer agents from natural and synthetic sources, the DTP has also initiated a programme to discover and preclinically develop anti-AIDS agents. Being a US government-funded agency, the NCI accepts and screens compounds from many sources, including academics in universities and pharmaceutical companies. This philanthropic activity increases the chances of finding lead compounds with activity against the many types of cancer and against HIV. Much of the early NCI screening focused on natural products produced by fermentation of microbes, such as filamentous bacteria, and, prior to 1960, only a small numbers of plants (1500) were investigated.

Early successes to exploit plants as a source of anticancer agents included the characterization of podophyllotoxin-type lignans from Podophyllum peltatum, and their semi-synthetic derivatives, and vincristine and vinblastine from Catharanthus roseus (Fig. 8.1). These discoveries drove further interest in plants at the NCI, and collections of plant material were expanded to include 60 countries. This collection strategy was quite extensive and, between 1960 and 1982, 114,000 extracts from 35,000 plants species were screened against a number of tumour types (Cragg et al 1993).

Fig. 8.1

Methods to assess antitumour activity are being continually improved and, after 1975, the P388 mouse leukaemia cell line was used to assay for the presence of compounds that are cytotoxic. Following bioassay-guided isolation of the active agents, a secondary screen known as the human xenograft mouse model was used to assess agents with significant activity. In this model, human tumour cells are introduced beneath the skin of mice lacking an immune response. The cells rapidly grow to form a tumour; compounds may then be administered to the mouse and any effects on the reduction in tumour size measured. This mouse model is still used today and cells from different tumour types may be introduced. Any compounds with broad-spectrum activity were prioritized for preclinical development and eventually clinical trials.

Bioassay-guided fractionation of active extracts led to the characterization of a large number of agents from many different natural product classes. Arguably the most significant discovery from the NCI screening programme was named taxol, from Taxus brevifolia (Taxaceae) (Fig. 8.2). The compound had been previously isolated in 1971 by Wall and co-workers and its activity against a melanoma cell line and in the human xenograft model led to its selection for preclinical development in 1977. Initially, there were problems in acquiring large amounts of the compound, but solutions to these problems and the report of its unique mode of action by promotion of tubulin polymerization and stabilization of microtubules against depolymerization increased interest. Taxol (now known as paclitaxel; see below) has excellent activity against ovarian and breast cancers and was approved in the USA by the FDA in 1993.

Fig. 8.2

Another agent of note, again from the group of Wall, is camptothecin from the Chinese ornamental tree, Camptotheca acuminata (Nyssaceae) (Fig. 8.2). Chinese clinical trials were conducted on this agent, which showed promise against a number of different cancer types, including gastric, liver, head, neck and bladder cancers. However, US clinical trials of the sodium salt were terminated due to a limited number of responses. There has been much work on this class of natural product, with two companies (SmithKline Beecham and Daiichi) having developed and released two products: topotecan and irinotecan.

Between 1985 and 1990 the NCI developed a new in vitro screen that uses 60 human cell lines from nine cancer types representing leukaemia, lung, colon, central nervous system (CNS), melanoma, ovarian, renal, prostate and breast cancers. Synthetic or natural products are made up at concentrations ranging from 10^–4 M to 10^–8 M and then tested against the 60 different tumour cell lines. The data that are acquired from this test system are evaluated to decide if further investigation is warranted (Alley et al 1988, Grever et al 1992).

The results from the screening are reported in a graph format which calculates the potency of the compound against all 60 cell lines in terms of three indicators:

The importance of these three factors is that a fingerprint of a compound’s anticancer activity is generated which can be correlated, using a mathematical model, with test results for known anticancer drugs (Paull et al 1989). This can even give an idea of whether a compound works by a similar mechanism of action as a known anticancer agent such as vincristine. Promising compounds with interesting selective profiles of activity (e.g. activity against one particular cell line) are then retested to confirm the reproducibility of these results. In 1995 a three cell line prescreen was introduced to increase the throughput of compounds whilst not reducing the number of compounds with activity in the 60 cell line assay. These cell lines were from lung, CNS and breast tumour types.

A new acquisition programme was initiated in 1986 with contracts for the cultivation and extraction of fungi and bacteria and the collection of marine organisms and terrestrial plants (Cragg et al 1997). Marine organisms were collected from many diverse locations, including the Caribbean, Australasia, Central and South Pacific, Indian Ocean and Polynesia. Plant material was collected in over 25 countries in the tropical and subtropical regions by specialist organizations with botanical expertise such as the Missouri Botanical Garden, the University of Chicago and the New York Botanical Garden.

Collaborations with the NCI rely on working closely with qualified organizations (e.g. universities, research institutes, botanical gardens) in the source country that has access to the biological diversity. Botanists and biologists from source countries collaborate in collections, which are invaluable to the NCI, and in many cases training is provided and workshops are organized for local personnel so that there is a transfer of expertise in certain areas. The source countries send biomass back to the National Products Repository in Frederick, Maryland. In the past, invitations have been made to local scientists of source countries to visit the NCI for collaborative research in natural products (Cragg et al 1997). This fosters openness and true collaboration between the NCI and source country.

For biomass collection, dried plant material (usually in the dry weight range of 0.3–1.0 kg) and frozen marine organism samples (1 kg) are shipped to Frederick and stored at –20°C, lessening the possibility of sample degradation. Biomass is extracted with methanol/dichloromethane (1:1) and then water to give organic and aqueous extracts which are given NCI numbers and stored at –20°C until required for screening (Cragg et al 1997). The extracts are then tested in the in vitro human cancer cell line screen and active extracts undergo bioassay-guided fractionation to isolate pure natural products. Structure elucidation is carried out using NMR and mass spectrometry. Compounds demonstrating significant antitumour activity are selected for secondary testing in in vivo systems and those with good activity advance to preclinical and, finally, clinical development. During this process there is the opportunity to license compounds out to companies for development; this is the preferred route due to the large costs of conducting the development process.

Importantly, agreements ensure that any royalties that may arise are shared with the country of origin and, as previously mentioned with the Convention on Biodiversity, it is a legal requirement that the collection of organisms is only possible with the ‘prior informed consent’ of the authorities of the source country (see Chapter 5, p. 55).

A full review of antitumour agents is beyond the scope of this text but we will briefly cover the three main sources of anticancer agents (marine sources, plants and microbes), giving selected examples of agents and their sources. The reader is urged to consult the excellent reviews by Cragg and co-workers on this subject (Cragg & Newman 1999, 2000, Cragg et al 1993, 1997).

Camptotheca acuminata (Nyssaceae)

In 1957, a small number of ethanolic plant extracts (1000) were screened at the Cancer Chemotherapy National Center (USA) and extracts of Camptotheca acuminata were shown to possess high activity. The late Dr Monroe Wall, who will be remembered as a leading expert in plant antitumour agents and a wonderful enthusiast for natural product research, established a natural product research group at Research Triangle Institute which conducted some research that was funded by the NCI. Dr Wall, who was later joined by Dr Mansukh Wani, became interested in the constituents of Camptotheca, particularly due to the high activity of these extracts. Wood and bark (20 kg) were collected for extraction and the extracts were shown to be active against a mouse leukaemia life prolongation assay in which it was unusual to find activity (Wall & Wani 1995). During the 1960s, the bioassay-guided isolation process was slow and it could take as long as 3 months to get bioassay results from the mouse life-prolongation assay back to the chemists working on the extracts (Wall & Wani 1995).

When compared with cellular assays today, in which it may only take a few days to decide if extracts and fractions are active, it is worth noting how diligent, persistent and thorough these researchers were. Extraction of the plant material was carried out first using a non-polar solvent (hot heptane) to remove very non-polar natural products. The plant material was then extracted with ethanol and this extract was treated with chloroform and aqueous ethanol to give two phases, the lipophilic chloroform phase being highly active. Additional extraction of the aqueous phase with chloroform gave a further active chloroform extract. High activity was associated with all of the chloroform extracts of the ethanolic portion, indicating that the active natural product had a degree of lipophilicity. Column chromatography was initially attempted on these active extracts using alumina as the sorbent, but what was not known at the time was that the active component binds very tightly to alumina and could not be eluted from the column. Extensive partitioning of the extract was tried in separating funnels using the quaternary system (four solvents) chloroform/carbon tetrachloride/methanol/water and each of the fractions were assayed in the mouse life-prolongation assay and in vitro cellular cytotoxicity assay. This partitioning system gave excellent separation and active fractions were combined to yield a yellow precipitate, which was subjected to chromatography on silica to which the active agent did not irreversibly bind. Further crystallization of active fractions led to the isolation of the active agent, camptothecin (Fig. 8.6). The structure elucidation of this natural product was carried out by conversion to its chloroacetate, which was then further converted to the iodoacetate salt which gave crystals of sufficient quality for X-ray structural analysis.

Fig. 8.6

At the time, camptothecin had a unique structure, possessing an α-hydroxylactone, and being a highly unsaturated alkaloid (nitrogen-containing) natural product of the quinoline alkaloid group. When treated with sodium hydroxide, the lactone ring opened with the formation of a sodium salt that was much more water-soluble than camptothecin. When this salt was acidified, the lactone was again regenerated and the water solubility decreased.

Camptothecin was shown to be extremely active in the life-prolongation assay of mice treated with leukaemia cells and in solid tumour inhibition. These activities encouraged the NCI to initiate clinical trials with the water-soluble sodium salt. Trials were conducted in a small number of patients (18) and there were five partial responses against gastrointestinal tumours of short duration (Gottlieb & Luce 1972). There were, however, toxic side effects. A phase II study was conducted but unfortunately there were only two responses to the treatment (Moertel et al 1972). The sodium salt was also studied in a much larger clinical trial in China using 1000 patients with better results against head, neck, gastric, intestinal and bladder carcinomas. These results were more promising than those of the US trial possibly due to the fact that the US patients had already been treated with other cytotoxic drugs and their tumours may have become multidrug-resistant. It is also now known that the lactone (which is absent in the lactone ring-opened sodium salt) of the compound is important for activity.

Wall and Wani continued to isolate and evaluate Camptotheca metabolites and found that one such product in particular, 10-hydroxycamptothecin (Fig. 8.6), was more active than camptothecin, and this may have stimulated a number of companies to prepare further water-soluble analogues of 10-hydroxycamptothecin.

Interest in this natural product and its analogues was low until 1985 when it was discovered that camptothecin works by inhibition of topoisomerase I, which is an enzyme involved in many important cellular processes by interacting with DNA. This finding fuelled further interest in this class of compound as the mechanism by which it functioned as an inhibitor of tumour growth was understood. Much work was conducted by pharmaceutical companies on this template and two products became available as a result of this research, irinotecan and topotecan (Fig. 8.7). The structural importance of the parent natural product, 10-hydroxycamptothecin, as a template can be seen in both of these products.

Fig. 8.7

Irinotecan was approved in 1994 in the USA. It is less toxic than camptothecin and is used to treat metastatic colorectal cancer; it is also effective against lung cancer and leukaemias. Irinotecan has much greater water solubility than camptothecin. It is a prodrug, being metabolized in vivo by hydrolysis of the carbamate ester to give the phenolic topoisomerase I inhibitor, which is 1000 times more potent than the parent compound.

Topotecan (Fig. 8.7) was approved for use in the USA in 1996. It is active against ovarian cancer.

Taxus brevifolia (Taxaceae)

In the early 1960s a number of plants were collected from the USA and assessed for antitumour activity under an NCI-supported programme. One of these specimens was Taxus brevifolia. This plant, commonly known as the Pacific yew, is a member of the Taxaceae family and is related to the English yew (Taxus baccata), which is common in churchyards in the UK. The Pacific yew is a slow-growing tree common to the western coast of the USA. Active extracts were tested at the Research Triangle Institute in 1964 (Wall & Wani 1995).

The extraction of the plant material was conducted using a partitioning protocol similar to that employed in the isolation of camptothecin. The isolation of the active component employed a highly extensive partitioning procedure utilizing many steps between aqueous and organic solvents. The biological activity of the fractions was monitored at all stages and 0.5 g of the active compound was isolated. This was an exceptionally low yield (0.004%) from 12 kg of plant material and the name taxol was assigned to the active natural product. Taxol was shown to be active against solid tumours and highly active against leukaemia models in the mouse life-prolongation assay. The compound was also active against a melanoma cell line. As with camptothecin, structure elucidation was not a trivial process (it can still be difficult today!) and there were only a handful of techniques that could be used to determine the structure, including UV, IR, elemental analysis (determination of C, H and N composition) and mass spectrometry; there were not many NMR techniques available in the early 1960s, which made structure elucidation an extremely difficult task. The structure elucidation of this compound is an example of a masterful piece of natural product research and the reader is urged to consult the superb review by Wall & Wani (1995) and the book by Goodman & Walsh (2001).

¹H NMR spectra suggested that taxol had a number of ester groups attached to it and was possibly a diterpene natural product of the taxane group. Elemental analysis and mass spectrometry gave a molecular formula of C₄₇H₅₁NO₁₄ for taxol, but it was not possible to obtain crystals of the compound of sufficient quality for X-ray structural analysis. At this stage chemical degradation techniques were used to help elucidate the structure. Base-catalysed methanolysis yielded compounds that could be crystallized and submitted for X-ray structural analysis for structure elucidation. Elegant work, which included the use of further chemical degradation plus ¹H NMR and high-resolution mass spectrometry, was conducted to piece these fragments together to give the final structure of taxol (Fig. 8.8) (Wall & Wani 1995).

Fig. 8.8

This was the first report of a taxane diterpene with cytotoxic activity. Taxol is a highly functional molecule possessing esters, epoxides, hydroxyls, amide, ketone groups and unsaturation. It has a large number of chiral centres (11) and is very difficult to synthesize; this was achieved in 1994 after 26 steps and is obviously not a feasible option for production on a large scale.

For the antitumour activity it is essential that the ester group at position C₁₃ is present as the hydrolysed product is inactive. Work on taxol at RTI ended in 1971. Although Dr Wall and co-workers tried to get the NCI to assess taxol further, it was thought that the concentration in the plant was too low, the extraction and isolation were too difficult and the tree supply too limited. There were two developments which re-ignited interest in this compound, the first being its activity in a melanoma cell line and the discovery of its unique mode of action. Initial work suggested that taxol had similar activity to other agents which were ‘spindle poisons’ such as vincristine and colchicine. The work of Susan Horowitz’s group demonstrated that, whilst taxol inhibited mitosis, it actually stabilized microtubules and inhibited their depolymerization back to tubulin, which is the exact opposite of other agents that bind to soluble tubulin and inhibit the polymerization of tubulin into microtubules.

The fact that taxol worked by a new mechanism and that it was a highly unusual and novel structure encouraged further research into this agent, resulting in clinical trials and the development of analogues such as taxotere (Fig. 8.8).

The supply issues concerning taxol were overcome with its semi-synthesis by the conversion of metabolites present in larger amounts (e.g. 10-deacetylbaccatin III; Fig. 8.8) in the needles of the related English yew (Taxus baccata). As the needles are a renewable resource, there is no need to destroy trees by the removal of bark. Taxol is now commercially produced by plant cell culture by large-scale fermentation of the plant cells.

Taxol, now renamed paclitaxel, was approved in 1993 and marketed under the trade name Taxol by Bristol Myers Squibb for ovarian cancer and the secondary treatment for breast and non-small-cell lung cancers. Docetaxel is marketed as Taxotere by Rhone-Poulenc Rorer and, like Taxol, prevents the mitotic spindle from being broken down by stabilizing microtubule bundles. Docetaxel, approved for use in the USA in 1995, is slightly more water-soluble than Taxol and is also administered by the intravenous route. It is used for the treatment of breast and ovarian cancers.

Betula alba (Betulaceae)

Betulinic acid (Fig. 8.9) is a pentacyclic triterpene that occurs in several plants. It can be chemically derived from betulin, a natural product found in abundance in the outer bark of white birch trees (Betula alba).

Fig. 8.9

There is currently interest in betulinic acid as it selectively kills human melanoma cells leaving healthy cells alive. The incidence of melanoma has been increasing at a higher rate than any other type of cancer and, therefore, the need for new anticancer agents that selectively target melanoma is great. The effectiveness of betulinic acid against melanoma cancer cells has also been evaluated in a xenograft assay with athymic (nude) mice. When mice were injected with melanoma cells, the tumour size was observed for 40 days following injections of betulinic acid. Betulinic acid effectively inhibited tumour growth in the mice, with little drug toxicity and side effects such as weight loss. This natural product has been shown to be an inducer of apoptosis (programmed cell death) in cancer cells with a high degree of specificity for melanoma cells. This unique specificity is unusual when compared to other cytotoxic agents such as camptothecin or paclitaxel, which exhibit a far broader spectrum of activity.

This natural product appears to have limited toxicity, is inexpensive and abundantly available from the bark of white birch trees in the form of betulin (the hydroxyl derivative) which can be readily converted to betulinic acid through a simple oxidation reaction. The compound is currently undergoing preclinical development.

Podophyllum peltatum (Berberidaceae)

Podophyllum peltatum has a number of common names, including the mayapple, Devil’s apple and American mandrake, and is a perennial plant found in the woodlands of Canada and eastern USA.

The plants reach 10–45 cm in height and have long, thin rhizomes which are the underground stem from which the roots grow. The rhizomes are known to be poisonous and are the most important part of the plant, containing high concentrations of podophyllotoxin and α- and β-peltatin, all of which are cytotoxic.

The closely related Asian species, P. emodi (syn. P. hexandrum), which is known as Indian podophyllum, contains these active lignan ingredients at a lower concentration. Podophyllotoxin and related lignans are also found in the rhizomes of another species P. pleianthum, which in Japan and China is used to make a preparation to treat snakebites and genital tumours.

The rhizomes of P. peltatum have a long history as a medicine among native North American tribes (Penobscot Indians of Maine). They are gathered in autumn, dried and ground to a powder, and the material is eaten or drunk as an infusion of the powder as a laxative or to get rid of intestinal worms. The powder was also used as a poultice to treat warts and skin growths. Currently, extracts of the plant are used in topical medications for genital warts and some skin cancers. However, the mayapple rhizome powder has a strong purgative action and the compounds in it are too toxic for self-medication.

Ethanolic extracts of the rhizomes are known as podophyllin, which is included in many pharmacopoeias for the topical treatment of warts and condylomata acuminata, which are benign tumours. Podophyllin resin is highly irritant and unpleasant and cannot be used systemically. The main natural product is the podophyllotoxin lignan class (see Chapter 6) of which podophyllotoxin was the first to be isolated in 1880 and the structure proposed in 1932. There are many components in podophyllin; the most important from the antitumour perspective are the lignans. Podofilox is a purified form of podophyllin which acts as a cell poison against cells undergoing mitosis, and, although this extract is not a systemic chemotherapeutic agent, it is used topically in creams as a treatment for genital warts.

Chemists working at the Sandoz pharmaceutical company developed the hypothesis that podophyllin resin may contain anticancer lignan glycosides that are more water-soluble and less toxic than podophyllotoxin, and several of these natural products were isolated. The compounds did, indeed, possess greater water solubility but unfortunately had less antitumour activity. Much work was done on these glycosides to make them resistant to enzymic hydrolysis, maintain their water solubility and improve their cellular uptake. This was a difficult task as, paradoxically, improving the water solubility decreased the cellular uptake (and therefore activity).

By a serendipitous event, a crude fraction of podophyllin was treated with benzaldehyde, resulting in a mixture of products that were mainly benzyl derivatives of lignan glycosides (Stähelin & von Wartburg 1991). The crude reaction mixture was highly cytotoxic and active in the mouse life-prolongation assay against a leukaemia cell line. Some of the components of this mixture were isolated but were not as active as the whole mixture. Condensation of benzaldehyde with the reaction mixture generated a series of acetals that were not hydrolysed by glucosidase enzymes and were less water-soluble. Additionally, the crude mixture worked by a different mechanism from the purified products by stopping the tumour cells from undergoing mitosis. This preparation was marketed for cancer treatment under the name of Proresid. Work on Proresid with bioassay-guided isolation against a mouse life-prolongation assay with a leukaemia cell line indicated that there was a highly active agent present. As with paclitaxel and camptothecin, in the 1960s it was very difficult to isolate and identify minor components and the process took several years to complete. The most abundant component of Proresid is podophyllotoxin benzylidene glucoside (1) (Fig. 8.10); a minor component is 4′-demethylepipodophyllotoxin benzylidene glucoside (2), which was found to be the most active agent (Fig. 8.10).

Fig. 8.10 Structures of podophyllotoxin benzylidene glucoside (1) and 4′-demethylepipodophyllotoxin benzylidene glucoside (2).

There was much synthetic work conducted to produce analogues that retain the same structural features of compound (2), which is epimeric at position 1 and lacking a methyl at position 4′ with respect to compound (1). The two most important analogues synthesized so far are etoposide and teniposide (Fig. 8.11) which have much more potency than the parent compound.

Fig. 8.11

Etoposide is marketed as Vepesid for small cell lung cancer, testicular cancer and lymphomas; teniposide is also used in the treatment of brain tumours. Podophyllotoxin binds to tubulin and is a member of the ‘spindle poison’ group of agents and functions by preventing microtubule formation. Etoposide and teniposide work via a different mechanism by inhibiting the enzyme topoisomerase II preventing DNA synthesis and replication. The difference in mechanism is attributable to the small adjustment in structure with etoposide and teniposide being 4′-demethyl compounds and having different stereochemistry at position C₁.

Catharanthus roseus (Apocynaceae)

The Madagascar periwinkle (Catharanthus roseus, syn. Vinca rosea) was originally native to Madagascar. It has been widely cultivated for hundreds of years and can now be found growing wild in most countries, including the UK. The natural wild plants are a pale pink with a purple eye in the centre but many colours have been developed by horticulturalists, ranging from white to pink and purple. The plant has had a long history of treating a wide assortment of diseases and was used as a folk remedy for diabetes in Europe for centuries. In China, the plant has been used for its astringent and diuretic properties and as a cough remedy, and in the Caribbean it is used to treat eye infections and for diabetes. Historically, the periwinkle has had a reputation as a magic plant: Europeans thought it could ward off evil spirits, and the French referred to it as ‘the violet of the sorcerers’.

Jamaicans have traditionally used a tea from C. roseus to treat diabetes although it has not been possible to find a basis for this use. Over 150 alkaloids have been characterized in the plant, a number of which are indole alkaloids and include dimeric or bis-indole alkaloids. These components are broadly referred to as vinca alkaloids after the name of a synonym for this plant, Vinca rosea.

The discovery of the vinca alkaloids from the Madagascan periwinkle is a classic example of drug discovery and, although extracts of the species had a reputation as being useful in the treatment of diabetes, a screening programme at the pharmaceutical company Eli Lilly revealed that extracts inhibited the growth of certain types of cancer cells. Bioassay-guided isolation of extracts of the plant led to the characterization of the active alkaloidal compounds vincristine and vinblastine (Fig. 8.12).

Fig. 8.12

These natural products are structurally highly complex and are dimeric indole alkaloids that can be synthesized but are too expensive to produce in this way. The natural yield of the drugs in the plant is exceptionally low (0.0002% for vincristine), which makes them very expensive antitumour agents. They exert their anticancer effects by inhibiting mitosis by binding to tubulin, thus preventing the cell from making the spindles it needs to be able to move its chromosomes around as it divides. Vinblastine is marketed as Velbe by Eli Lilly and is useful for treating Hodgkin’s disease, lymphomas, advanced testicular cancer, advanced breast cancer and Kaposi’s sarcoma. This drug has a number of side effects, including hair loss, nausea, lowered blood cell counts, constipation and mouth sores. Vincristine is marketed as Oncovin by Eli Lilly and is used to treat acute leukaemia, Hodgkin’s disease and other lymphomas. Semi-synthetic vinca alkaloids of note include vindesine (marketed under the name Eldisine), used to treat leukaemia and lung cancers, and vinorelbine (marketed as Navelbine by GlaxoSmithKline), which is used as a treatment for ovarian cancer. Vinorelbine has a wider range of antitumour activity than the other vinca alkaloids and is used, in combination with cisplatin, in treating patients with non-small-cell lung cancers.

The anthracyclines

This group is a large complex family of antibiotics and many were investigated before a useful antitumour agent was isolated. They are structurally and biosynthetically related to the tetracyclines (Chapter 6), being derived from the polyketide pathway. One of the first agents to be described in this class was daunorubicin from Streptomyces peucetius and Streptomyces caeruleorubidis (Fig. 8.13).

Fig. 8.13

This agent is active against leukaemias. One of the most widely used related antitumour agents is doxorubicin (Adriamycin) from Streptomyces peucetius var. caesius, discovered in the late 1960s, which is highly active against a broad spectrum of both solid and liquid tumours. A number of semi-synthetic analogues have been produced of which idarubicin has enhanced antitumour potency and is less cardiotoxic than doxorubicin. These compounds possess a tetracyclic linear ring system (based on anthracene, hence the term anthracycline antitumour antibiotics) to which an amino sugar is attached. These agents bind to DNA and inhibit DNA and RNA synthesis. The main antitumour action of this group is by inhibition of topoisomerase II.

The bleomycins

This group consists of a closely related mixture of glycopeptide antibiotics from the filamentous bacterium Streptomyces verticillus. They were discovered in 1966 by screening of culture filtrates against tumours. The bleomycin mixture is partly resolved (purified) prior to formulation for clinical use under the name blenoxane, which consists of a mixture of bleomycin A₂ (55–70%) and bleomycin B₂ (30%) (Fig. 8.14). These natural products occur as blue copper chelates.

Fig. 8.14

The different analogues are distinct from each other only in the terminal amine functional group and a number of unusual functional groups are present in the bleomycin skeleton, including amino acids, pyrimidine rings and sugars. The bleomycins are DNA-cleaving drugs and cause single- and double-strand breaks in DNA. The dithiazole groups are essential for activity and are believed to be important in the binding of the bleomycins to DNA. This class of drugs has use in the treatment of lymphomas, head and neck tumours and testicular cancer with very little bone-marrow toxicity.

The actinomycins

Actinomycin D (dactinomycin) is an antibiotic from Streptomyces parvullus first isolated in 1940. It is an antimicrobial compound which is toxic and has limited use as an antitumour agent. Structurally, it consists of a planar phenoxazinone dicarboxylic acid attached to two identical pentapeptides (Fig. 8.15). A number of analogues of different peptide composition are known.

Fig. 8.15

The planar group of this agent intercalates with double-stranded DNA, inhibits topoisomerase II and RNA synthesis, and can also cause single-strand DNA breaks. The principal use of this group is in paediatric tumours, including kidney (Wilms) tumours, but the use is limited due to unpleasant side effects.