Colorectal cancer (CRC) is one of the main causes of death in Western countries. Many colon cancer treatment options are available, including surgery, chemotherapy and radiation, but chemoprevention is a fundamental approach to reduce cancer risk. Chemoprevention is the long-term use of different chemical agents, both synthetic and natural, by healthy individuals to prevent or delay the onset of disease. Since CRC has a significant environmental component, it is an ideal disease in which to evaluate the potential benefits of chemopreventive agents. Different CRC chemoprevention strategies are under investigation. They include prevention of radical formation and DNA hypomethylation, prevention or suppression of mutations, inhibition of cell proliferation and induction of tumor cell differentiation (1).

Among the several biochemical alterations found in cancer cells, one of the most noticeable is a change in the intracellular polyamine content. Polyamines are polycationic compounds that play a key role in almost all the steps of colorectal tumorigenesis. In CRC tissue, the polyamine content as well as the activities of two important enzymes in their biosynthesis such as ornithine decarboxylase (ODC) and S-adenosylmethionine decarboxylase (SAM-DC), are increased 3–4 fold over that found in the equivalent normal colonic tissue. Besides, polyamines have been considered as possible markers of neoplastic proliferation in the colon. Noticeably, it is not only polyamine synthesis, but also their uptake that is enhanced in rapidly proliferating neoplastic cells in the colon. In contrast to all other cell systems in the body, colon cancer cells are exposed to high concentration of polyamines. These derive from the cells of the gut mucosa, which release their polyamine content into the lumen during the process of cell extrusion and death, but also from the gut bacteria and food. Therefore, also the introduction of polyamines by exogenous environment with diet may be of importance in a strategy of cancer prevention (2).

This review summarizes data on the role of polyamine metabolism in neoplastic transformation of colorectal mucosa and as possible target for CRC chemoprevention. Attention will be focused on the influence of drugs and nutritional factors on polyamine metabolism, as well as the role played by dietary polyamines.

Polyamines (putrescine, spermidine and spermine) are organic cations present in every living cell. They have pleiotropic effects on cell physiology and play a relevant role in cell proliferation and differentiation (3). These molecules are positively charged at the primary and secondary amino groups at physiological pH. Thus, polyamines may act as ligands at multiple sites on DNA, RNA, proteins, phospholipids and nucleotide triphosphates (4). Accordingly, biological functions of polyamines are mainly involved in the regulation of gene expression by altering DNA structure and modulating signal transduction pathways (5,6).

The polyamine functions and their metabolic pathway have been extensively studied (7,8). Fig. 1 shows a schematic representation of the polyamine metabolic pathway. Briefly, biosynthesis is mediated by the key enzyme ODC which converts the amino acid ornithine into putrescine. This is then sequentially converted into spermidine and spermine, through the action of the enzyme SAM-DC and spermidine/spermine synthase. The central enzyme in the polyamine catabolic pathway is the spermidine-spermine-N1-acetyl transferase (SSAT), which adds acetyl groups to terminal amine groups in spermidine and spermine. These acetylated polyamines are then substrates for the enzyme polyamine oxidase (PAO) which retro-converts these acetylated derivatives into lower chain amines.

Intracellular levels of polyamines are tightly controlled and this occurs in addition to the multi-level control of synthesis and catabolism, also by uptake and efflux (9).

The protein ornithine decarboxylase antizyme (AZ) can effectively control polyamine levels not only by inactivating ODC and inducing its degradation, but also by increasing polyamine efflux and decreasing polyamine uptake. However, an antizyme inhibitor (AZI) has been characterized and the overexpression of this compound in certain forms of cancer has been reported (10).

Many biochemical alterations have been found in cancer cells, but one of the most consistent is a change in the intracellular polyamine content. Polyamine concentrations increase during carcinogenesis and an increase in ODC activity accompanies neoplastic transformation (11). As with other tumors, polyamine content of CRC is higher when compared to the adjacent mucosa and equivalent normal tissue (12,13). The increase is due to the loss in polyamine homeostasis occurring during the dysregulation of cell proliferation (2). This is also proven by evidence of an upregulation of polyamine biosynthesis (14), a decrease in their catabolism (15,16) and an increased uptake (17). Since polyamine metabolism is an integral component of the mechanism of the carcinogenesis in colorectal tissue, polyamine levels and ODC activity have been considered even as specific markers for neoplastic proliferation in the colon (18). Particularly, ODC has been shown to be critical in cell transformation and it has been suggested to be a proto-oncogene (19).

A possible role of polyamines in regulating oncogene expression and function through transcriptional and postranscriptional processes has also been suggested (20). On the contrary, two of the most commonly mutated genes in colon cancer, such as the adenomatous polyposis coli (APC) tumor suppressor gene and K-ras oncogene, have been shown to regulate the expression of several polyamine metabolic genes (e.g., ODC and SAT) (15,21). In this context, it has also been observed that polyamine biosynthesis is involved in human colorectal carcinogenesis in a manner that is K-ras-dependent and p53-independent; besides, K-ras mutation and polyamine biosynthesis seem to be preferentially associated with polypoid tumors rather than flat colorectal tumors of the colon (22,23). Given that polyamines and cancers seem to be tightly connected, modulations of the polyamine metabolic pathway such as polyamine uptake and efflux, have received much attention in cancer drug development for chemoprevention of human CRC (24).

ODC, being the rate limiting enzyme in polyamine biosynthesis, has represented the first target in the polyamine pathway for cancer therapy. Several studies have focused on the use of difluoromethylornithine (DFMO), an ODC inhibitor, as chemopreventive and chemotherapeutic agent (25,26). Although DFMO has been proven to inhibit the growth of tumor cells in vitro, it has not been so convincing in its anti-neoplastic properties, being cytostatic rather than cytotoxic, at least in vivo (27). Besides, prohibitively high doses of DFMO were usually required to inhibit malignant tumor growth in early chemotherapeutic trials (28,29).

The lack of efficacy by DFMO against established tumors has been put in relation to the availability of extracellular polyamines, derived from the diet, the retroconversion pathway as well as the polyamine content supplied by the gastrointestinal (GI) microbiota (30). However, it has also been observed that very low, non-toxic doses of DFMO were able to inhibit stimulation of proliferation by various carcinogens and this evidence has led to investigation of DFMO as a cancer chemopreventive agent. Probably, the greatest potential for DFMO action in chemoprevention may be exerted against CRC, where the decrease in ODC activity and polyamine content have been proven to limit significantly the formation of tumors (31).

A series of problem must be solved before conclusive evidence might be obtained from chemopreventive trials. Firstly, the putative chemopreventive agents have to be administered over long periods of time. Therefore, a major problem to be resolved concerns not only efficacy, but also the presence of minor toxicity and unpleasant side effects. Secondly, appropriate intermediate and endpoint biomarkers which allow the monitoring of the effectiveness of the drugs must be identified before the start of the research (32). In this connection, the early prevention clinical trials of DFMO have been conducted including pilot phase IIa and IIb studies aimed at evaluating the safety of treatment as well as the effective reduction in the colorectal tissue polyamine content (33).

It has been observed that an oral dose of 0.2 g/m²/day can be both effective in reducing colorectal mucosal polyamine contents and safe, given that no relevant signs of toxicity were observed in the treatment group, compared to the placebo group (34).

It is known that combining different molecules in a strategy of chemoprevention could offer the prospect of reduced toxicity by lowering doses of individual agents. Therefore, a variety of combined formulations in order to decrease intracellular polyamine levels have been proposed and studied for their efficacy in cancer prevention. As an example, one successful strategy is based on the administration of DFMO along with non-steroidal anti-inflammatory drugs (NSAIDs). As demonstrated by epidemiological reports, a regular use of NSAIDs is associated with a decreased risk for CRC (35), whereas experimental studies have shown that DFMO acts, at least additively, with a number of NSAIDs (36).

This drug combination proved to decrease polyamine synthesis by inhibiting ODC (action due to DFMO), while contemporary increasing catabolism and cellular export of polyamines by activating SAT (induced by NSAIDs) (37). This combination produced an additive reduction of colon carcinogenesis also in laboratory animals (38) and it was observed that sulindac and celecoxib are both potent stimulators of SAT and inhibitors of intestinal carcinogenesis in Apc Min/+ mouse (39).

More recently, a 3-year trial (40) strongly validated a significant effect of the combined administration of low doses of DFMO (500 mg/die) plus sulindac (150 mg/die) on the polyamine production in rectal mucosa. Although no relationship between changes in polyamine levels and response were observed, the baseline polyamine levels affected the DFMO plus sulindac effects for colorectal adenoma (CRA) prevention.

By combining DFMO with NSAIDs, or other proven chemopreventive agents, it is also expected that fewer side effects occur during treatment. In this connection, it has been reported that DFMO plus sulindac administration is safe and effective in chemoprevention of CRA in people with prior colon polyps (41). However, other phase III trials involving adenoma patients, showed that the same combination of DFMO plus sulindac is actually effective, but confers a modest risk of ototoxicity and the potential risk of cardiovascular toxicity (42).

The usual side-effect ototoxicity and differential treatment outcomes related to the DFMO plus sulindac treatment could be associated with the specific germline single nucleotide polymorphism (SNP) in the ODC-1 promoter region. SNP has been investigated as a marker for colorectal adenoma (43). In this connection, it has been observed that aspirin is able to decrease the risk of CRA recurrence especially in individuals homozygous for the ODC1 minor A allele compared with those with the major G allele (44). This has been confirmed and corroborated by other studies (45). These findings suggest that genetic features of the polyamine metabolism may be markers for both treatment benefit and toxicity.

Some data also indicate a potential link between obesity and polyamine inhibition in humans (46); besides, an association between obesity and risk of CRA has been reported (47). In spite of these observations, in a recent study it has been shown that obesity does not substantially modify the CRA risk reduction ascribed to DFMO plus sulindac versus placebo (48). This evidence strongly supports the need for chemopreventive clinical trials to refine the risk:benefit as well as the risk:risk profiles of the putative chemopreventive agents.

Based on the above, it is evident that the colorectal chemoprevention strategy with DFMO plus NSAIDs provides important proofs of principle that targeting polyamine metabolism can really be an effective strategy for reducing those risk factors closely associated with the development of colon cancer in humans.

Apart from ODC, the other polyamine biosynthesis enzymes, SAM-DC and spermidine/spermine synthase, have also received considerable attention as target for polyamine level reduction. Most SAM-DC inhibitors rapidly depleted spermidine and spermine concentrations in the cells, but their effects disappeared when in vivo models were tested. Probably, this happened because the accumulation of putrescine might compensate for spermidine in cells. Inhibitors of spermidine/spermine synthase have not been shown to be efficacious in either in vitro or in vivo system (49). PAO inhibitors have also been developed, since this enzyme converts spermine to spermidine, in order to overcome cell growth inhibition by polyamine depletion. These inhibitors are potent killers of cancer cells in vitro and showed promise when used with DFMO in carcinogenesis model in vivo (25,50).

As above mentioned, in addition to inhibition of polyamine biosynthesis, induction of polyamine catabolism has also been a major pharmacological goal. The enzyme SSAT can be induced by a variety of compounds like NSAIDs and polyamine analogues (51). These latter molecules need a special mention, due to their pleiotropic effects on cancer cells. Polyamine analogues are able to reduce the cell polyamine content by both upregulating catabolism (inducing SSAT), decreasing biosynthesis by negative feedback inhibition and by competing with exogenous polyamines for uptake. Besides, polyamine analogues can inhibit cell growth by acting like endogenous polyamines and bind to intracellular polyamine binding sites, thus rendering them ‘non-functional’ (52). In vitro studies have clearly demonstrated a role for polyamine analogues as multi-site inhibitors of the polyamine pathway, although the efficacy of each type of analogue in cancer chemoprevention and therapy remains to be fully characterized in vivo (53).

Other aspects of the polyamine metabolism pathway including polyamine uptake and efflux have also been targeted for cancer drug development. Uptake inhibitors have shown promise in vitro and have increased the efficacy of DFMO as an anticancer agent (17,54).

Aside from specific inhibitor/inducer drugs for the polyamine pathway and transport, also several nutritional components thought to be useful in colon chemoprevention have been shown to affect the polyamine metabolic pathway as well as to impair the tissue polyamine content in colorectal neoplastic tissue.

The GI tract is constantly connected with the external environment, therefore all the possible modifications in daily diet might significantly modify the exposure to different carcinogenetic factors. In this framework, the identification of those components in diet that may display at different degree some antitumor activity as well as the understanding of their mechanisms of action, may lead to significant advances in human cancer prevention. Several diet components have been demonstrated as having some anti-neoplastic activity. From the early 1970s, many studies following different designs (e.g., correlation studies, case-control, cohort studies) have hypothesized that dietary fibers may show protective effects against diverse neoplasms, including GI neoplasms (55). Their positive action in the GI environment has been essentially put in relation with the bacterial strains resident in the GI lumen (56).

The intestinal microbiota have been considered as a potential target for an active anticancer strategy. In view of its fundamental role in human health, manipulation of the intestinal microbiota by microrganisms shown to positively affect the GI tract (e.g., probiotics) or other compounds found in plants, has been considered as a logical approach to prevent or inhibit the neoplastic transformation of GI mucosa.

Probiotics are defined as ‘live microorganisms which, when administered in adequate amounts, confer a health benefit on the host’ (57). Together with these positive bacterial strains, other substances, mainly found in plants have been deeply investigated for their chemopreventive and chemotherapeutic properties against human cancers. These substances include flavonoids such as genistein (found in soy), quercetin (onions), apigenin (celery, parsley), green tea (polyphenols), etc. On these bases this chapter tries to review data on those nutritional components useful for CRC chemoprevention in relation to their potentiality in affecting the polyamine metabolism.

In addition to endogenous polyamines, dietary polyamines and their metabolites by intestinal microorganisms have been shown to be major determinants of the total body polyamine pool. The colonic lumen contains polyamines from both the diet or exported by enteric bacteria and these polyamines are transported via mechanisms not yet well described on the apical cell surface (110). As far as diet is concerned, poly amine content is high in several food products including fruits, cheeses and meat. Again, high concentrations of putrescine can be found in common diets, particularly in orange and grapefruit juice (111).

The early studies have shown that intestinal and dietary polyamines can enhance colonic tumorigenesis, and can minimize the effects of ODC inhibitors (112). In tumor-bearing animal models, it has been demonstrated that a polyamine deficient diet significantly enhances the antitumoral effect of DFMO plus neomycin (113). Overall, the use of polyamine transport inhibitors, alone or in combination with DFMO, provides a method to target cancers with high polyamine requirements (114).

It has been observed that the primary effect of dietary putrescine was to increase tumor grade; besides the effectiveness of sulindac to suppress intestinal carcinogenesis was partially abrogated by dietary putrescine in a murine FAP model (115).

Polyamine metabolism is also dependent on levels of the precursor amino acids arginine and ornithine. In a study by Yerushalmi et al (116), Apc Min/+ mice were fed with arginine concentrations corresponding to the higher range of arginine consumed by humans eating a western style diet. Authors reported that large amounts of arginine in diet increase colonic polyamine levels as well as carcinogenesis. This finding strongly supports the concept that dietary arginine could be a risk factor for colon carcinogenesis also in humans (117,118). In this connection, it has been found that patients with a family history of CRC and reporting meat consumption in the highest quartile had a statistically significantly decreased overall survival and increased risk of death, compared with those in the lower quartiles (119).

Recently, it has also been demonstrated that an association between high dietary polyamine intake and CRA risk in humans can be modified by sex and ODC genotype. Particularly for women, the risk of CRC development showed a positive trend with increasing quartiles of polyamine intake, and participants with higher polyamine intakes and the ODC GG genotype had significantly higher odds of CRC compared with those subjects showing the same genotype but lower polyamine intake (120). Therefore, a diet low in polyamines could represent an adjunctive strategy to therapeutic prevention using polyamine-inhibitory agents.

Raj et al (121) have observed a significant interaction between treatment with DFMO plus sulindac and dietary polyamine intake on the risk of recurrent adenomas in a CRA prevention phase III clinical trial. Patients in the highest quartile of dietary polyamine intake have been found to have no significant metachronous adenoma risk reduction after treatment with DFMO plus sulindac, in contrast to a significant 81% risk reduction observed for patients in the lowest quartile of dietary polyamine intake.

Collectively, the above evidence suggests that dietary polyamines may be involved in human colon carcinogenesis; therefore, approaches to limit dietary polyamines may represent additional strategies for CRC prevention. In this context, special diets low in polyamines have already been developed for other type of cancers and it has been observed that the polyamine-lowering regimen is associated with improved pain control and survival characteristics (122). However, due to the abundance of polyamines in the food supply and the strong preclinical data that relate polyamine exposure to tumor growth, these findings still support the need for additional investigation of dietary polyamines in human health.

Polyamines and their enzymes are strongly related to neoplastic proliferation in the GI tract. Therefore, all the strategies of effective chemotherapeutic and chemopreventive interventions targeting polyamines, will certainly require a combinatorial approach directed towards all the multiple features shown by their metabolic pathway.

Several agents in diet, thought to be useful in CRC chemoprevention, have been shown to affect the polyamine metabolic pathways. A combined chemopreventive intervention using protocols based on the use of these agents, along with polyamine inhibitors and/or inducers, would enhance their properties representing a suitable alternative option for the management of CRC patients. Moreover, also dietary polyamines may be involved in human colon carcinogenesis, conditioning the effects of drugs and nutritional components. It is, therefore, fundamental to acquire more data on this aspect that could represent an innovative and interesting approach to colorectal oncology. Future studies will determine whether strategies targeting the polyamine pathways will contribute to prevention of colon cancer.