«MODULATION OF POLYAMINE METABOLISM AS A CHEMOPREVENTIVE STRATEGY OF PHYTOCHEMICALS IN A CELL CULTURE MODEL OF COLORECTAL CANCERS Dissertation zur ...»
After detecting potent cell growth inhibitory properties of ursolic acid we started to measure the expression status of several cell cycle regulating proteins, whereby the most prominent effects could be observed in the upregulation of cell cycle inhibitors p21WAF1/Cip1 and p27Kip1. The observed induction of 21WAF1/Cip1 is in accordance with results found in other malignant cell lines which could show the same effects ursolic acid [83;84;164], but we are the first to describe regulatory effects on p27Kip1. Contrary to expectations incubation with ursolic acid also leads to a conspicuous increase of cell cycle progressor cyclin E, which is however consistent with earlier findings by Wolter et al.  and Schneider et al. , which could show the same effects after treatment with the polyphenol resveratrol and take this as a result of a cell cycle arrest in the SDISCUSSION phase. However, cell cycle analysis of ursolic acid treated cells presents a predominant arrest in the G1 phase [83;165]. These controversial results will deserve further investigations. Moreover, there are several lines of evidence, that the induction of cyclin E by genotoxic stress, such as ionizing radiation  or chemotherapeutic agents  could play a functional role in the initiation phase of apoptosis in malignant cell lines, in addition to its reported key regulatory role on the control of the G1 to S-phase transition and the initiation of DNA replication .
Another interesting observation is the simultaneous accumulation in both p27Kip1 and cyclin E levels, which normally correlate negatively as there exist interdependence regulatory mechanisms. These simultaneous events are known to be characteristic phenotypes in cells derived from mice lacking Sphase kinase associated protein 2 (Skp2)  suggesting a possible involvement of protooncogene Skp2 in the regulation of p27Kip1 and cyclin E, which might provide another target of ursolic acid mediated actions.
3.2.2 Induction of apoptosis
Apoptosis, a form of programmed cell death, plays a fundamental role in the maintenance of tissues and organ systems by providing a controlled cell deletion to balanced cell proliferation. It is now apparent that many dietary chemopreventive agents with promise to human consumption can also preferentially inhibit tumor cell growth by targeting one or more signalling intermediates leading to induction of apoptosis. Two major apoptosis pathways have been identified, the death receptor or extrinsic pathway and the mitochondrial or intrinsic pathway. The mitochondrial pathway is regulated by members of the bcl-2 protein family, which can be divided in pro- and antiDISCUSSION apoptotic groups . Because the activation of mitochondria has been considered as the “point of no return” in the apoptotic process, the manipulation of mitochondrial activation with proapoptotic intentions has been envisaged as a potential therapeutic target. While ursolic acid leads to an upregulation of proapoptotic BAX protein levels, the expression of the antiapoptotic Bcl-2 was diminished after 24h of incubation. This results in a significant increase of the BAX/Bcl-2 protein ratio which is generally known to trigger apoptosis [172;173].
The extrinsic pathway of apoptosis is activated at the cell-surface when a specific ligand binds to its corresponding death receptor. Death receptors, like tumor necrosis factor (TNF) receptor, TRAIL receptor and Fas belong to the TNF receptor superfamily which consists of more than 20 proteins with a broad range of biological functions . The TRAIL ligand and its receptors are of special interest for cancer therapy, since TRAIL has been shown to predominantly kill cancer cells, while sparing most other cells . In certain tumor cell lines, TRAIL protein expression could be induced by chemopreventive agents resulting in TRAIL-mediated apoptosis in an autocrine or paracrine manner [176-178]. This suggests that endogenously expressed TRAIL, which we could also observe in our cells after ursolic acid treatment, may be at least partly responsible for the observed chemopreventive effects.
Taken together ursolic acid seems to lead to activation both of extrinsic and intrinsic signaling pathways both resulting in caspase-3 activation, followed by DNA fragmentation and programmed cell death.
3.2.3 Modulation of polyamine metabolism
Based on the findings in Manuscript I and II we were interested whether the modulatory effects on the polyamine metabolism are specific for polyphenolic resveratrol or if reduction of intracellular polyamines is a common chemopreventive strategy used also by other phytochemicals like triterpene ursolic acid. A correlation between polyamines and cancer have been extensively studied for decades, pointing out the inhibition of polyamine biosynthetic enzymes ODC and SAMDC or activation of catabolic enzyme SSAT as a potential chemopreventive strategy [179;180]. An association between SSAT induction followed by catabolism of the ubiquitous intracellular DISCUSSION polyamines and subsequent apoptotic responses was first reported by Ha et al . Furthermore, Chen et al.  could demonstrate that selective interference of polyamine-analogue induced SSAT prevents apoptotic signaling and apoptosis in human melanoma cells. Much on the focus on SSAT has been on the functional level, but the regulation of SSAT gene expression has also been a subject of recent investigations. As already mentioned, the transcription factor PPARγ was shown to be involved in this regulatory mechanisms  [MANUSCRIPT I]. We could show that also ursolic acid seems to activate SSAT in a PPARγ-dependent manner. But unlike resveratrol [MANUSCRIPT I], ursolic acid could not simultaneously inhibit the activity of biosynthetic ODC (unpublished data).
3.2.4 Summary and conclusion
Summarizing our results, the observed reduction of cell growth of colon cancer cell lines after treatment with ursolic acid presumably results from a large increase in the number of apoptotic cells. The modulation of the polyamine metabolism, especially the induction of the catabolic SSAT via PPARγdependent mechanisms thereby seems to present the major molecular target in the induction of programmed cell death.
Hence, providing potent chemopreventive activities in vitro, ursolic acid could in theory possibly serve as alternatives to chemically designed antineoplastic agents, as constituents of therapeutic drug combinations in advanced disease, or as adjuvant treatments.
3.3 Bioavailability of Phytochemicals
The bioavailability of a nutrient is defined by its degree to which it becomes available to the target tissue after administration. One important cause of failure in cancer therapies it due to a defect of drug accumulation in cancer cells.
Indeed, the action of chemopreventive or chemotherapeutic agents can be nullified by a failure of their absorption, distribution, metabolism or an increase
in their excretion. Hence, the knowledge of absorption and metabolism of a compound in vivo is the precondition to determine its bioavailability.
The absorption and transport of resveratrol have been studied in several models: isolated rat intestine [182;183], rats and mice after oral administration [184-188], human colon carcinoma Caco-2 cell line , hepatocytes  and healthy human subjects [187;191;192].
These experiments showed, that jejunum and, to lesser extend, illeum are involved in the absorption of resveratrol. Using radiolabelled resveratrol administered orally, an appreciable fraction, 50-70% of the dose, was absorbed in rats , and radioactivity could be recovered from the stomach, liver, kidney, intestine, bile and urine in mice . Intragastric administration of 2mg/kg, 20mg/kg and 50mg/kg resveratrol to rats generated peak values of 2µM, 1.2µM and 6.6µM in plasma. After oral administration to humans (25mg/70kg) the amount of free resveratrol in plasma and serum reached ~37nM (less than 2% of total resveratrol) . The appearance of a new resveratrol peak 6h after consumption suggests enteric recirculation of conjugated metabolites by reabsorption after intestinal hydrolysis .
In contrast, most mechanistic studies in vitro suggest that carcinogenesismodulating effects of resveratrol require the sustained presence of 5-100µM, which, according to the mentioned studies, seems hardly to be realizable in vivo.
Pharmacokinetic studies in mice and rats suggest consistently that resveratrol is well absorbed and rapidly glucuronidated and sulphated both in the liver and intestinal epithelial cells [183;187;188;194]. In humans, following its absorption, resveratrol is rapidly metabolized in the liver by phase-2 drug-metabolizing enzymes to water-soluble trans-resveratrol-3-o-glucuronide and transresveratrol-3-o-sulfate, accounting for its predominant urine excretion.
Compared to resveratrol, which has a plasma half-life of 8-14min, these metabolites have a plasma half-life of about 9.2h . Although modifications such as glucuronidation and sulphation typically reduce the cell permeability of drugs and aid in their excretion, the undeniable in vivo efficiacy of resveratrol (see next chapter), despite its low bioavailability, has led to the speculation that its metabolites could retain some activity. Research into the actions of metabolites has been hampered by the lack of commercial sources, but should
proceed more readily now that synthetic molecules have been established by several groups [192;193;195].
Most bioavailability studies on resveratrol concentrate on absorption rates and plasma levels, but in the case of colorectal cancers, it might also be interesting, which amount of resveratrol is not absorbed and reaches the colon.
Unfortunately, there is a lack of data measuring fecal concentrations after resveratrol consumption. To our knowledge solely Walle et. al examined fecal recovery of [14C]-Resveratrol after oral (25mg) and i.v. (0.2mg) doses in six volunteers which was highly variable (0.3-38%) . Unfortunately, the number of participants in this study was very small, which is particularly critical due to the interindividual variability in xenobiotic metabolisms. The relevance of fecal concentrations becomes apparent in a recent study, when Karlsson et al.
could show that fecal water from volunteers consuming a vegetarian diet potently reduced COX-2 protein levels and PGE2 production in colorectal cancer cells . This is in line with earlier publications demonstrating that dietary alterations can influence the biochemical composition of fecal water [197;198].
Hence, intraluminal accumulation of resveratrol might be another hypothesis to explain the discrepancy between low bioavailability and effectiveness in vivo which requires further investigations.
While the bioavailability of resveratrol is comparatively well characterized by now, little work was done on pharmacokinetic studies of ursolic acid. In one study plasma samples taken from rats that had received Lu-Ying extract, whose major effective constituent is ursolic acid, orally were analysed using a rapid and sensitive LC-MS method. These measurements demonstrated that ursolic acid exhibits a high binding activity in organs and a low blood distribution. This low bioavailability was due to a poor absorption rate and to rapid metabolization in entero- and hepatocytes . But even though these results doubt the effectiveness of ursolic acid in vivo, a very recent study, examining the effect of orally administered ursolic acid on the formation of aberrant crypt foci (ACF) and intestinal SMase activity in azoxymethane (AOM)-treated rats, indicates that ursolic acid provides chemopreventive effects in the initiation phase of colon cancer associated with changes in SM metabolism, as the incidence of ACF could be significantly reduced after ursolic acid consumption when
compared to control animals . These auspicious results raise hope for a clinical relevance of ursolic acid in therapeutic approaches of the future.
3.4 Red wine consumption as a chemopreventive strategy?