Polyamine Pools and Cancer Prevention Eugene W Gerner and David E Stringer

1. Introduction

Polyamine contents increase during epithelial carcinogenesis in both rodents and humans. The activity of ornithine decarboxylase (ODC), the first enzyme in polyamine synthesis in mammals, is increased in epithelial cancers, and ODC has become a target for cancer chemoprevention. Difluoromethylornithine (DFMO) is a specific inhibitor of ODC and is under evaluation in clinical trials for prevention of epithelial cancers, including those of the colon, prostate, and skin. Tissue, serum, and other bodily fluid endpoints have been used to monitor effects of DFMO in human clinical chemoprevention trials. Cell and tissue polyamine contents are regulated by mechanisms influencing synthesis, catabolism, uptake, and efflux. These regulatory mechanisms include both transcriptional and posttranscriptional processes. Tissue measurements of ODC enzyme activity have often failed to be reliable measures of either carcinogenesis or response to DFMO. Levels of target tissue polyamines are less subject to measurement variability than are measures of ODC enzyme activity, and are closely associated with administered DFMO dose in randomized clinical chemoprevention trials in humans. Target tissue polyamine contents are validated biomarkers for the biochemical efficacy of DFMO in clinical cancer chemo-prevention trials. There is a need for new predictive biomarkers of agents that affect polyamine metabolism that could be measured in sources other than target tissues.

2. Regulation of Polyamine Metabolic Genes in Normal Development

Polyamine metabolic genes are targets of several signaling pathways regulating normal development in mammals. ODC, the first enzyme in polyamine synthesis in animals, is a transcriptional target of the c-MYC gene (1). The c-MYC protein is a transcriptional regulator of survival and apoptotic responses (2), and is regulated in part by the adenomatous polyposis coli (APC) tumor suppressor gene (3). APC regulates ODC transcription in intestinal and colonic tissues of mice (4). APC regulation of ODC transcription occurs via a c-MYC-dependent mechanism in human colon tumor cells (5). Wnt and Hedgehog are two important developmental regulatory pathways in

From: Polyamine Cell Signaling: Physiology Pharmacology and Cancer Research Edited by: J.-Y. Wang and R. A. Casero, Jr. © Humana Press Inc., Totowa, NJ

eukaryotes that have also been associated with carcinogenesis (6). Polyamine metabolic genes are targets of these pathways. The Wnt pathway member, APC, regulates ODC, whereas the ODC inhibitor protein antizyme has been reported to be a target of the Hedgehog pathway (7). ODC is essential for cell survival during embryonic development in mice (8). ODC expression is downregulated via a posttranscriptional mechanism in senescing human fibroblasts (9), and putrescine and spermidine pools are lowest in slowly turning over adult epithelial tissues (10). Reactivation of polyamine synthesis is crucial for repair processes in certain normal tissues (11).

3. Regulation of Polyamine Metabolic Genes in Neoplasia

In humans suffering from familial adenomatous polyposis (FAP), an inherited form of colon cancer caused by loss of wild-type alleles encoding the APC gene, c-MYC, and subsequently ODC are aberrantly expressed (12,13). Studies in the multiple intestinal neoplasia (ApcMin/+) mouse model of FAP, and in genetically modified human cells, indicate that mutant APC is associated with an increase in ODC and a decrease in ODC inhibitor protein antizyme (4). These changes in polyamine metabolic genes are associated with increases in intestinal mucosal polyamine contents in the ApcMin/+ mice compared with normal littermates. Treatment of the ApcMin/+ mice with the selective ODC inhibitor DFMO suppresses these elevated intestinal mucosal polyamine contents and reduces the number of intestinal adenomas. Other genes mutated or deleted in cancer also regulate polyamine metabolic genes. Activating mutations in the K-RAS oncogene, a signaling molecule activated by guanosine 5'-triphosphate, are found in about half of colon tumors induced by treatment of mice with the chemical carcinogen azoxymethane (14). K-RAS activation is associated with the transcriptional repression of the spermidine/spermine A^-acetyltransferase (SSAT), a regulatory enzyme in polyamine catabolism (15). This transcriptional repression is mediated by the K-RAS-dependent suppression of expression of the peroxisomal proliferator-activated receptor gamma (PPARy), which promotes SSAT transcription via a response element near the SSAT transcription start site (16). Consequently, loss of function of the APC tumor suppressor and activation of the K-RAS oncogene, which are common events in human colon cancers, causes increases in intestinal polyamine contents because of increased synthesis and decreased catabolism of the polyamines. Evidence suggests that polyamine metabolism is altered in other epithelial cancers and neoplasia. Chemically induced skin carcinogenesis is associated with increases in ODC (17), and ODC inhibitors are potent inhibitors of this process (18). Polyamine metabolic gene expression is altered in human prostate cancer (19). Recent studies indicate aberrant expression of c-MYC, the transcriptional activator of ODC in prostate epithelium, is sufficient to induce invasive prostate cancer in mice (26). Treatment of the TRAMP mouse, another model of prostate cancer, with DFMO reduces prostate neoplasia (21). These studies provide evidence for causative roles of polyamine metabolic gene alterations in several types of epithelial cancer and neoplasia.

4. General Factors Controlling Polyamine Pool Sizes

The polyamines spermidine and spermine, and their diamine precursor, putrescine, derive from either intracellular synthesis, via decarboxylation of the amino acid

Fig. 1. In humans, increases in polyamine pools occur as a consequence of synthesis from ornithine or mechanisms transporting polyamines from extracellular sites into cells and tissues. Decreases in polyamine contents occur as a consequence of catabolism and export of intracellu-lar pools to extracellular sites.

Fig. 1. In humans, increases in polyamine pools occur as a consequence of synthesis from ornithine or mechanisms transporting polyamines from extracellular sites into cells and tissues. Decreases in polyamine contents occur as a consequence of catabolism and export of intracellu-lar pools to extracellular sites.

ornithine, or transport from extracellular sources in mammals. Figure 1 depicts these biosynthetic and uptake processes, along with catabolic and efflux mechanisms, all of which act to modulate intracellular polyamine pools.

These processes are highly regulated. Polyamines negatively regulate their synthesis and uptake from extracellular sources (22), and positively regulate their catabolism and export (23,24). Davis et al. (25) have noted a lack of evidence for polyamines allo-sterically regulating their own metabolism. They have argued that the apparent stability of polyamine pool sizes in unstressed cells is owing to sequestration of the polyamines, rather than as a consequence of allosteric feedback regulation. They present a compelling argument for the conclusion that only a small fraction of cellular polyamines are unbound, and consequently metabolically active. They point to observations that addition of exogenous polyamines to cultured mammalian cells can cause dramatic changes in intracellular processes involving polyamines, such as induction of the ODC antizyme, without causing detectable changes in intracellular polyamine pools. On the other hand, treatment of cultured mammalian cells with inhibitors of polyamine synthesis can deplete intracellular pools of some of the polyamines without dramatically affecting cell growth. These observations suggest that polyamine pools may be useful biomarkers to predict effects of agents that inhibit or activate polyamine metabolic enzymes, but may not be useful as markers of normal or disease processes.

5. Assessment of Errors in Measures of Polyamine Parameters in Human Tissues

Early in our cancer chemoprevention studies, it became clear that we needed to identify surrogate biomarkers as endpoints for our studies. We wanted to treat people who had not yet developed, but were at risk of developing, cancer. Development of colon and other epithelial cancers from noninvasive intraepithelial neoplasia has been estimated to take 15-30 yr. Thus, we determined sources of errors associated with several measures of polyamine parameters to determine which parameters were most reliably measured in human tissues. We conducted our most extensive study of errors in polyamine parameters in colonic mucosa of normal volunteers (26). We chose this particular tissue for study because it was relatively easily accessible and highly relevant to our planned studies of colon cancer chemoprevention. Our endpoints included measures of ODC and SSAT enzyme activities, ODC RNA expression, and polyamine contents. Enzyme activities were measured by documenting metabolism of radiolabeled substrates in tissue extracts, RNA expression was assessed by Northern blot analysis, and tissue polyamine contents were determined by high-performance liquid chromatography (HPLC) methods. Our measurement methods first used HPLC to separate individual amines, which were then derivatized with fluorescent markers. The fluorescently labeled amines were quantified. We evaluated measures of these endpoints in normal volunteers. First, we collected colonic tissues using several bowel preparation methods. We found that intrain-dividual measures of colorectal polyamine contents were less affected by bowel preparation methods compared with measures of enzyme activities. All measures of polyamine parameters in colorectal mucosa were highly variable between individuals. However, we found that differences in measures of polyamine contents, and especially spermidine:spermine ratios, were greatest between individuals. Differences in poly-amine contents were less variable when compared in colorectal biopsies taken from similar locations in the same individual compared with those taken from different individuals. We evaluated variability in our measures of polyamine parameters in biopsies of various sizes taken at various locations in the colons of these normal volunteers. Finally, we evaluated measures of polyamine endpoints in biopsies obtained sequentially during insertion and extraction of the colonoscope to evaluate the influence of bowel location and irritation on these parameters. Measures of polyamine parameters were least variable when comparing values within individuals compared with those between individuals. Of all parameters measured, polyamine contents were least variable, especially spermidine:spermine ratios. We reasoned that the low intraindividual variability in spermidine:spermine ratios was a consequence of this parameter having only one source of technical measurement error. Both spermidine and spermine values are obtained from a single HPLC chromatogram, so that only technical measurement is required. Individual polyamine contents are determined as a ratio of two separate measurements, including the HPLC measurement, which is then normalized to tissue protein content. We concluded that measures of polyamine contents were the most reliable measures of polyamine parameters in colonic tissue of normal human volunteers. We reached the same conclusions regarding measurements of errors in polyamine parameters in other tissues in the intestinal tract (10) and prostate (27), although our analysis of errors in these latter tissues was not as extensive as our initial studies in the colon.

6. Polyamine Parameters as Prognostic Markers in Cancer

Twenty yr ago, there was great interest in measures of polyamine metabolism as potential prognostic markers for epithelial cancers. Pioneering studies by Luk and Baylin found that ODC enzyme activity was elevated in the apparently normal colorectal mucosa of individuals with FAP compared with people not afflicted with FAP (28). ODC enzyme activity and polyamine contents were found to be elevated in several examples of nonheritable intraepithelial neoplasia, including sporadic colon polyps

(29,30) and Barrett's esophagus (31), compared with adjacent, normal tissues. Most of these studies found that average values of polyamine parameters were elevated in neo-plasia compared with those in apparently normal tissue, but that substantial overlap existed between values in normal and neoplastic tissues. Further, measures of polyamine parameters in apparently normal colorectal mucosa were not prognostic for, or did not foretell the risk of, colon cancer development (32-24). Measurements of polyamine parameters in human serum or urine were also found to be poor prognostic markers for cancer (35), although recent technical developments, to be discussed at the end of this chapter, might provide new avenues for investigation. From these results, we concluded that ODC enzyme activity and polyamine contents were generally elevated in neoplastic compared with normal, intestinal tissue, but that measurements of these endpoints were not useful prognostic factors for people with either Barrett's esophagus or colon polyps.

7. Polyamine Parameters as Predictive Markers in Cancer Chemoprevention

Undeterred by these negative findings, we focused our attention on determining if polyamine parameters could be used as surrogate endpoint biomarkers in clinical cancer chemoprevention trials. The goal of cancer chemoprevention is to prevent the occurrence of invasive cancer. The target populations for chemoprevention trials are individuals at risk for, but who do not yet have, cancer. Because many epithelial cancers develop over one or more decades, clinical chemoprevention trials have used surrogate endpoints in trials lasting weeks to a few years to assess potential utility of agents for use in longer term trials (36). When we initiated our cancer chemoprevention studies, DFMO had already been shown to be a potent inhibitor of carcinogenesis in several animal models (37). Although a number of other inhibitors and activators of polyamine metabolism are now known, we have focused on DFMO because of its apparent safety when administered to humans (37,38). Our choice of DFMO dose schedule merits discussion. Pharmacological studies of DFMO had shown that serum half-life of DFMO was on the order of hours in rodents, dogs, and humans (39). These findings caused clinical investigators to employ multiple daily dosing or continuous intravenous infusion techniques to maintain maximal serum levels of DFMO in clinical trials. For cancer chemopreven-tion trials, we preferred simple methods of agent delivery. Cell culture studies with DFMO had shown that ODC enzyme activity continued to be suppressed, even after removal of DFMO from culture medium (40). The persistent anti-ODC enzyme activity of DFMO was both DFMO concentration- and treatment time-dependent. We reasoned that DFMO was cleared rapidly from the serum, but that the ODC-inhibiting activity of the drug was longer lasting in cells and tissues. As a consequence of these published data and unpublished results, we chose an administration schedule of single daily oral doses of DFMO. Our early studies of DFMO involved drug administered orally in a liquid. We have recently changed to oral DFMO administered in tablet formulation.

8. Validation of Polyamine Contents as Biomarkers for DFMO Effect

Several polyamine parameters have been evaluated as potential biomarkers in chemoprevention trials of DFMO. Studies in rodent models showed that topical application of 12-O-tetradecanoyl-13-acetate (TPA)-induced ODC enzyme activity in skin (17). Uninduced ODC enzyme activity is difficult to detect in extracts of human skin punch biopsies. However, the University of Wisconsin group used these observations to evaluate inhibition of TPA-induced ODC activity in skin punch biopsies as a marker of DFMO effect in a chemoprevention trial (41). They evaluated a range of doses to determine which oral DFMO doses, administered once daily, were able to inhibit ODC enzyme activity (induced by ex vivo TPA treatment) of punch biopsies by at least 50%. Their results indicated that doses of 0.5 g/m2 per day could achieve this endpoint. Higher doses were found to cause reversible hearing loss, whereas lower doses failed to inhibit TPA-induced ODC activity by at least 50% in their ex vivo assay. Based on our evaluation of sources of errors in measurements of polyamine parameters, we chose to evaluate measures of polyamine contents as potential biomarkers of DFMO effect. In one study, we documented polyamine parameters in 111 men and women who had colon polyps. After removal of their polyps, we assessed polyamine contents, including spermidine:spermine ratios, in apparently normal rectal mucosa before beginning any drug treatment. In many experimental models, increased ODC synthesis is associated with increased putrescine and spermidine pool sizes, whereas spermine is often not affected. Thus, increases in spermidine:spermine ratios are general indicators of increased polyamine synthesis, whereas decreases in this parameter generally reflect decreases in polyamine synthesis (40). As shown in Fig. 2, we found that spermidine:spermine ratios in rectal mucosa were negatively associated with age in a statistically significant manner.

This result was interesting because it was consistent with results discussed earlier, which showed that polyamine synthesis is downregulated in adult tissues and during senescence in experimental models. We conducted short-term trials of DFMO administration in humans with colon polyps. We wanted to identify the safest dose of DFMO to use in our chemoprevention trials because potential participants in these trials would be basically healthy individuals identified by risk factors for cancer. In collaboration with our long-time collaborator Frank Meyskens and his colleagues at the University of California at Irvine, we designed and conducted a unique dose-de-escalation trial of DFMO, with polyamine contents as our primary endpoint (42). We found that DFMO taken as an oral liquid once per day for 1 mo reduced colorectal polyamine contents in a dose-dependent manner. In this nonrandomized trial, we found that DFMO dose was associated with decreased colorectal tissue contents of putrescine and spermidine, but not spermine. Spermidine:spermine ratios were also reduced in colorectal tissue in a statistically significant manner. Serum levels of DFMO were elevated in individuals taking this drug. Serum levels of DFMO were dependent on administered doses, but were highly variable among participants in the trial. Polyamine parameters, such as spermidine:spermine ratios, were reduced by administered dose, but not in a manner that was dependent on DFMO in the 0.1 to 3.0 g/m2 range. As shown in Table 1, we found that single daily oral doses as low as 0.10 mg/m2 or higher reduced spermi-dine:spermine ratios in the majority of patients at each dose group.

Patient responses measured by rectal spermidine:spermine ratios to oral DFMO were dependent on age, as seen in Fig. 3. These response data are consistent with the results

Fig. 2. Influence of patient age on spermidine:spermine ratios of colorectal mucosa. Biopsy specimens were obtained, with appropriate consents, from individuals who had a colon polyp, and were about to enroll in a chemoprevention trial of DFMO for colon polyp recurrence. Values shown are spermidine:spermine ratios before patients received DFMO. (Reprinted from ref. 42.)

Fig. 2. Influence of patient age on spermidine:spermine ratios of colorectal mucosa. Biopsy specimens were obtained, with appropriate consents, from individuals who had a colon polyp, and were about to enroll in a chemoprevention trial of DFMO for colon polyp recurrence. Values shown are spermidine:spermine ratios before patients received DFMO. (Reprinted from ref. 42.)

Table 1

Change in Colorectal Mucosal Spermidine:Spermine Ratios by DFMO Dose Group

Table 1

Change in Colorectal Mucosal Spermidine:Spermine Ratios by DFMO Dose Group

DFMO dose

Total number

Number with

Number with

Was this article helpful?

0 0
Healthy Chemistry For Optimal Health

Healthy Chemistry For Optimal Health

Thousands Have Used Chemicals To Improve Their Medical Condition. This Book Is one Of The Most Valuable Resources In The World When It Comes To Chemicals. Not All Chemicals Are Harmful For Your Body – Find Out Those That Helps To Maintain Your Health.

Get My Free Ebook

Post a comment