The polyamines, spermidine and spermine, were first discovered in 1678 by Antonie van Leeuwenhoek. In the early part of the 20th century their structure was determined (Figure 1) and their pathway of biosynthesis established . The polyamines are essential elements of cells from all species. They are required for optimum cell growth, and cells where polyamine production has been prevented by mutation, or blocked by inhibitors, require exogenous provision of at least one polyamine for continued survival. Despite this critical function, the polyamines have not attracted as much attention as they deserve in the wider field of biochemistry and cell biology. They are rarely mentioned in standard textbooks, despite over 75000 research papers having been written on the subject since 1900, and more than half (54%) were published after 1990 (A.E. Pegg, personal communication). There have been a number of books dedicated to the polyamines published and “The Guide to the Polyamines” by Seymour Cohen  deserves mention as a work of outstanding scholarship describing “everything you ever wanted to know about the polyamines” in exquisite detail. The current volume of Essays in Biochemistry has a much humbler aim: to introduce the polyamines to interested researchers and students, and to describe how they are associated with, and might be utilized as a target for intervention in major diseases such as cancer.
One of the main drivers in the quest to understand the function of the polyamines in mammalian cells has been their role in the growth and maintenance of cancer cells. Cancer is the second most common cause of morbidity and mortality world-wide, with millions of new cases of the common cancers, such as breast, colorectal, prostate and lung, diagnosed every year. More critically, the success rate in terms of cancer treatment varies enormously depending on the tumour type. The 5-year survival, the measure of success of therapy, is 80–90% for breast cancer, 45–55% for colorectal cancer and only approx. 5–15% for lung cancer. So clearly there is still a long way to go, and new drugs and treatments are required, particularly as the population ages. The polyamine pathway, because of the established link between polyamine concentrations in cells and cancer cell growth, has been identified as a goal for the development of antiproliferative agents.
The polyamine metabolic pathways have been recognized as a target for therapeutic intervention for four main reasons: (i) polyamines are required for cell growth; (ii) polyamine concentrations are increased significantly in cancer cells and tissues; (iii) ODC (ornithine decarboxylase), the first enzyme in the pathway, is increased in cancers and is an oncogene; and (iv) preventing polyamine biosynthesis prevents the growth of cells.
Another huge advantage is that polyamine metabolism is essential for all cancers, thus there is potential for treatment of multiple forms of the disease. In addition, more recent studies have suggested that inhibition of polyamine production may well be a novel and relatively non-toxic strategy for chemoprevention of cancer (see Chapter 8).
Polyamine metabolism is intriguing with a number of fascinating individual regulatory features which, if not unique, are at least uncommon in biology. This includes the rapid turnover of ODC, of the order of 10 min in mammalian cells, the +1 frameshifting that leads to translation of antizyme which targets ODC for degradation and the ‘super-inducibility’ of SSAT (spermidine/spermine N1-acetyltransferase), all of which will be addressed in the present volume.
Spermidine, spermine and their diamine precursor, putrescine, are positively charged molecules at physiological pH (Figure 1). Spermidine and spermine interact electrostatically with DNA, but their charge is spread over the whole molecule rather than a point charge as found in mono- and di-valent cations making them subtly different in terms of charge. Both polyamines induce condensation and conformational changes in DNA and to some extent RNA . It has been suggested that the polyamines are simply ‘super’ cations, but the question then arises: would a seven enzyme intracellular pathway, as well as uptake and export transport systems, evolve simply to provide a tri- or tetra-valent cation? Logic would dictate that this is not the case and that the polyamines and, indeed, their metabolism, have a more extensive role within the cell than being only polycations.
Classically, biosynthesis begins with ornithine and methionine, but recent evidence indicates a role for arginase in mammalian cells . The major source of polyamines in the majority of mammalian cells is via de novo biosynthesis with diet playing a significant, but lesser, role. The smallest contribution to intracellular polyamine pools is made by the gut microflora (Figure 2).
Ornithine is decarboxylated by ODC to putrescine (Figure 3). Although strictly speaking not a polyamine, putrescine is usually considered along with spermidine and spermine as part of the polyamine family. Two aminopropyl groups are then added consecutively to putrescine to form spermidine, and to spermidine to form spermine. The aminopropyl groups are provided by decarboxylated S-adenosylmethionine which is itself produced from S-adenosylmethionine by the enzyme SAMDC (S-adenosylmethionine decarboxylase; also referred to as AdoMetDC in the present volume). The by-product of this reaction is 5′-methylthioadenosine which is recycled back to adenosine for further use. SAMDC is the subject of one of the chapters of this volume and will not be discussed further here (Chapter 3). The enzymes producing spermidine and spermine are synthases which are constitutive enzymes with little inducibility. This is in contrast with both ODC and SAMDC, which are readily induced by a range of agents. The ready inducibility of ODC by a variety of growth-promoting stimuli was one of the early indications that ODC was an important player in cell proliferation. ODC also has a fast turnover, of the order of 10 min, very fast for a mammalian enzyme. The regulation of ODC turnover by antizyme is also discussed later in this volume (Chapters 2 and 4).
Until quite recently polyamine biosynthesis was considered irreversible, a ‘one-way street’ from ornithine to spermine. Recycling of the polyamines does occur, but via a separate retroconversion pathway of acetylation catalysed by SSAT followed by oxidation catalysed by APAO (N1-acetylpolyamine oxidase) . The main reason for this two-step retroconversion seems to be to produce acetylated polyamine derivatives for export from the cell, providing a means of depleting intracellular polyamine concentrations . N1-Acetylspermidine and putrescine are the major polyamines exported from cells. In the early 2000s, Casero's group in Baltimore cloned another oxidase, SMO (spermine oxidase), which converts spermine directly into spermidine without the need for acetylation . Bearing in mind that the acetylpolyamines are for export, it is logical that SMO is the enzyme used for recycling the polyamines, whereas the SSAT and APAO pathway is used for polyamine depletion.
In addition to the metabolic pathway there are also polyamine transporters: an inward, uptake transporter and an outward exporter. The transporters in mammalian cells are the last big challenge in polyamine metabolism as these have not been isolated or cloned from these cells. Evidence from yeast and bacteria suggest that there will be multiple transporters  and the current thinking is that many cells have at least two separate transporters, one for putrescine (and possibly other diamines) and one for spermidine and spermine. While research continues in this area, work from my own laboratory using a competitive, polyamine-uptake inhibitor suggests that, in human cancer cells, the importer and exporter are separate systems (H. Wallace, unpublished work).
As a result of the positive correlation between polyamines and cell growth it is perhaps not surprising that there is also a strong link between the polyamine content of cancer cells and their rate of growth. Indeed human cancer cells contain significantly more polyamine than the equivalent normal cells or tissue. For example, breast cancer cells have 4–6 times the polyamine content of normal breast cells, and colon cancer cells have 3–4 times the polyamine content of normal colonic cells [6,7]. Several groups had observed that patients with a variety of different cancers all showed increased urinary excretion of polyamines [8,9]. ODC activity was also increased in cancer cells, and so all of this was suggestive of a causal relationship and led many groups to look for targets in polyamine metabolism that would prevent polyamine biosynthesis and then perhaps tumour growth. The most obvious targets were the decarboxylase enzymes, ODC and SAMDC (Figure 3). A number of single enzyme inhibitors were designed and synthesized [10,11] with the hope of finding a new and effective anticancer drug .
The most successful inhibitor of this class is without doubt the irreversible suicide inhibitor of ODC, DFMO (α-difluoromethylornithine) . This agent has stood the test of time well; having been synthesized in the late 1970s it has gone through several reincarnations. As a single-agent anticancer drug in man, DFMO was a disappointment. Rationally designed to inhibit a key pathway required for cancer cell growth it was hugely successful in vitro , but in vivo it failed to produce toxicity in cancer cells, and the compensatory increases in uptake of exogenous polyamines by cells treated with DFMO resulted in little effect on organ infiltration and overall survival . However, as a single agent in parasitic disease (marketed as elfornithine) it is a great success producing cures for several trypanosomal infections (for a review, see ). Unfortunately, success in treating diseases in the developing world attracts little interest from drug companies, so although used and useful, DFMO is not a first-line drug in treatment of trypanasomiasis. It has found a small niche market as a hair-removing cream in a commercial product that uses 11% DFMO (www.vanqia.com). Recently, DFMO has emerged again, not as a chemotherapeutic agent, but as a strong candidate as a chemopreventative agent, the opportunities for which are discussed in Chapter 8.
Although DFMO was not useful as a single agent against cancer, it did provide a useful ‘proof-of-concept’ that inhibiting polyamine biosynthesis can prevent cancer cell growth, albeit in the absence of an external polyamine source. DFMO also showed that more extensive polyamine depletion is required. Decreases in all three polyamines are needed to prevent cancer cell growth, not just loss of putrescine and spermidine as occurs with DFMO. Finally, this research indicated that compensatory pathways need to be blocked as well to prevent attenuation of the initial inhibitory effects of the agent. A specific polyamine-uptake inhibitor would make DFMO an attractive antiproliferative drug.
Armed with the knowledge provided by nearly 20 years experience with DFMO, several groups developed polyamine analogues based on the following: (i) analogues look like polyamines so they compete with the natural amines for transport (this may prevent the compensatory uptake seen with DFMO); (ii) analogues will result in negative-feedback inhibition of ODC and biosynthesis; and (iii) analogues are sufficiently different not to be able to mimic the function of the natural polyamines in cells (e.g. DNA condensation).
Three generations of polyamine analogues have now been synthesized from the simple symmetrically substituted bis(ethyl)spermidine and spermine compounds produced by Bergeron in Florida  through the unsymmetrically substituted mainly spermine analogues from Woster's group in Detroit , to the more complicated conformationally restricted and oligoamines synthesized originally by Frydmann and now produced by Progen . These multiple analogues have had mixed success therapeutically, but have provided great tools for scientific research. The most interesting feature of treatment of cells with analogues is that many of them superinduce SSAT, which results in a rapid polyamine depletion from cells via the export transporter. This was an unexpected result, but since the natural polyamines induce SSAT, then perhaps it is not so surprising. Chapters 6 and 7 will discuss the development of these and a class of newer agents in more detail.
Cancer cells are effectively cells without barriers or checkpoints regulating their growth, which generally means a loss of regulation over the cell cycle. One of the early observations in cell-cycle studies showed that polyamines showed a biphasic response during the cell cycle , with ODC increasing in the G1 phase and then again in G2 with the subsequent increases in polyamine content. More recent work has linked these changes to interaction with oncogenes, cyclins and CDKs (cyclin-dependent kinases) which drive the cell cycle, and this will be discussed in more detail in Chapter 5.
Many cancers have up-regulated oncogenes, for example k-ras in colorectal cancer, and these up-regulated genes contribute to the overall disease and the potential for treatment. One set of key observations in the polyamines and cancer field is that showing the link between c-myc and ODC. The ODC gene contains a conserved repeat of the c-myc-binding site (CACGTG; an ‘E-box’ motif) suggesting that c-myc can regulate ODC at the level of transcription . ODC per se has also been shown to be able to transform cells, suggesting that ODC is itself an oncogene .
Thus the weight of evidence linking polyamines and cancer looks very positive, with increased polyamine content or enzyme activity promoting the transformed phenotype. But what about the alternative: a link between polyamines and cell death? For many years it was hypothesized that any agent that caused cell death would by necessity have to induce loss of polyamines from cells because they were essentially growth promoting and therefore no longer required [1,10,23]. Early work showed that export of polyamines was the ‘terminal’ means of polyamine depletion, removing them from the cell and not recycling them . This export process involved both metabolism and transport. Metabolism here is acetylation and oxidation, which provides the substrates for the outward transporter. This being the case, induction of cell death and also growth arrest should induce SSAT and APAO and increase the transport activity. The effect of the analogues on these processes has already been discussed, but other cytotoxic agents also induce these changes, including etoposide . Thus induction, particularly of SSAT, may be considered a more general response to cellular stress leading to growth inhibition. The most common type of cell death induced by anticancer drugs is apoptosis, and polyamines have been found to both induce and inhibit apoptosis in mammalian cells. The bifunctional regulation of this key regulatory process remains to be fully understood. Another approach that has been taken in humans is to use a low-polyamine diet (a completely polyamine-free diet is not possible, practically). This work has been pioneered by the group of Moulinoux in Rennes where they have achieved some success in refractory cancer, particularly hormone refractory prostate cancer [26,27]. The original idea was to combine a low-polyamine diet with DFMO, an inhibitor of gut microflora metabolism (metronidazole) and an inhibitor of APAO (MDL 72,537) . For various reasons this protocol has been cut down over the years, and with simply a low-polyamine diet alone improvements have been observed in patients. Perhaps the most interesting observations from these studies are the analgesic effects of the low-polyamine diet . This may also be an exciting area for further development in the field.
An alternative approach to understanding polyamine and polyamine enzyme function in cells is to use transgenic technology to knock-out, knock-in or knock-down the key enzymes in the pathway. The group from Kuopio has led the way in this effort producing a range of transgenic animals with altered polyamine biochemistry (see Chapter 9).
The excitement in the field continues with new approaches being investigated all the time as technology advances. Currently, the use of DFMO as a chemopreventative agent against colorectal cancer is making huge strides forward with clinical trials ongoing. Similarly, the idea of using polyamines as vectors to introduce toxic agents more selectively to cancer cells is an area of growing interest with great potential for drug delivery .
In summary, the polyamines have a long history, are being actively studied at present and are set up for a bright future in chemoprevention and drug delivery.
- © The Authors Journal compilation © 2009 Biochemical Society