Preclinical Assays for Identifying Cancer Chemopreventive Phytochemicals

1. Introduction

Cancer mortality rates in the developed countries have increased throughout this century. It is already the leading cause of death in some Western countries [1, 2]. In Japan, the progressive introduction of Western dietary habits, especially increased fat intake and reduced carbohydrate and dietary fiber intake, has increased the incidence of colon cancer and related deaths [3]. Great advances have been made in the pharmacologic-based treatment of malignant epithelial neoplasms (cancers). In addition, there is a marked increase in the understanding of cell and molecular mechanisms underlying carcinogenic processes. However, therapy for advanced neoplastic disease remains limited. This may be due to the fact that advanced neoplasms contain a large number of genetic and molecular alterations that contribute to the maintenance of their neoplastic progression.

The chemopreventive approach against cancer development is highly attractive. Although highly attractive from a theoretical point of view, practical limitations may exist with respect to developing novel and effective chemopreventive agents. Most importantly, practical clinical endpoints are not definite. The efficacy of cancer chemoprevention can be determined by comparing the incidence of nondeveloped disease in a treated group to that of a control group. Such a clinical trial is labor intensive, very costly, and time-consuming. In addition, practically such trials cannot be conducted for the rapidly increasing number of chemopreventive agents being identified. These considerations suggest the use of certain intermediate biomarkers as indicators of clinical efficacy. Validated intermediate biomarkers thus may allow relatively small chemopreventive trials to be conducted within a short-term period. However, there are no universal approaches to determine intermediate biomarkers and their method of validation has still not been proven.

Some herbal and botanical products are likely to possess cancer preventive activities [4–7]. Many cancer patients use complementary and alternative medicines, including phytochemicals in addition to, or following the failure of standard cancer therapy [8]. The term phytochemical applies to any plant-based substance, but in the field of nutrition and cancer this term is usually applied to nutritive and nonnutritive chemicals that occur naturally in fruits and vegetables. A diet rich in fruits and vegetables has long been suggested to correlate with reduced risk of certain epithelial malignancies, including cancers in the lung, colon, prostate, oral cavity, and breast [4–7, 9–12]. Also, the cancer prevention potential of Mediterranean diets based mainly on olive tree products is known [13]. The major component of the leaves and unprocessed olive drupes of Olea europaea is oleuropein and the majority of polyphenols found in olive oil or table olives are derived from its hydrolysis. Oleuropein is a novel, naturally occurring antioxidant compound, which may possibly be used to prevent cancer [14–16] and cardiotoxicity induced by doxorubicin [17]. Searching for medicinal benefits from edible or inedible plants is not a new idea since numerous modern medicines have plant origins. Given that the ingestion of some plant foods results in reduced risk for cancer, researchers are delving into the identification of phytochemicals with cancer preventive ability in studies in vitro (cell culture), in vivo (model animals) and those in humans [18]. Phytochemicals can be roughly classified into four groups based on their mechanisms of chemopreventive action, as shown in Table 1. Preclinical studies focus on the identification of possible cancer preventive agents, short-term pharmacology, and assessment of toxicity. Agents that are within tolerable safety limits in humans then moved clinical trials to test their efficacy.

This review briefly summarizes the in vitro and in vivo assays used to discover possible new chemopreventive agents possessing novel mechanisms of action. Surely there are many more assay systems currently used in different laboratories to find candidate cancer chemopreventive agents and determine their efficacy. Because of limited space, only limited assays that might be useful for carcinogenesis and chemoprevention studies are herein introduced.

2. Anti-Inflammatory, Antioxidative, Antiangiogenic, and/or Apoptosis-Inducing Compounds in the Prevention of Cancer Development

One of the most exciting areas of cancer chemoprevention is the effect of anti-inflammatory compounds against CRC. Almost a dozen case-control and cohort studies have concluded that the daily or alternate day consumption of aspirin over extended periods results in halving one′s risk for colorectal cancer (CRC) [19]. Several prospective studies are underway to specifically assess whether the daily consumption of aspirin affects the recurrence rates of colorectal adenoma in high-risk subjects for CRC. These studies strongly suggest that eicosanoids, especially prostaglandins (PGs), are involved in colorectal oncogenesis [20]. Two genes encode the two major forms of cyclooxygenase (COX) that is responsible for the metabolic conversion of dietary arachidonate to PGs, thromboxanes, and leukotrienes. The constitutive form, COX-1, is present in different types of cells. COX-2, the isoform inducible by growth factors and tumor promoters, is rarely found in the normal intestinal epithelial cells but is overexpressed in colorectal neoplasms [21–23]. Since COX-1 is a housekeeping gene, drugs for COX-1 inhibition cause side effects, such as ulceration. The kidney function, platelet aggregation, and the ulceration in stomach are strongly affected by the overuse or overdose of aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) [24]. These limit the wide-scale application of NSAIDs in the general public for the chemoprevention of CRC. Although NSAIDs are excellent chemopreventive agents in animal models for CRC, dietary aspirin does not inhibit azoxymethane (AOM)-induced colon carcinogenesis in rats that received a nonhigh fat diet. Because of side effects of NSAIDs that affect both COX-1 and COX-2 expressions, specific COX-2 inhibitors (-coxibs) were introduced to obtain cancer chemopreventive ability without side effects [25, 26]. They are indeed effective for CRC development in animal studies [27]. However, certain COX-2 inhibitors have been reported to increase the risk of ischemic heart diseases [28].

Many phytochemicals and/or botanicals are routinely tested for anti-inflammatory properties in the hope of finding new sources of medicines for the treatment of chronic inflammation and for the cancer chemoprevention. The plant world is apparently rich in sources of COX and/or inducible nitric oxide (iNOS) inhibitors that are involved in inflammation-related carcinogenesis [29]. Several of these have been found to inhibit cancer development in certain animal models. Curcumin [30] is able to inhibit the COX activity [31] in the skin [32] and suppresses skin cancer development in mice [33], tongue [34] and CRC in rats [35]. Tea [36, 37], given to human volunteers in a clinical trial, could reduce PGE₂ activity in the rectal mucosa [38]. Resveratrol [39] is a potent inhibitor of COX [40] and is an inhibitor in the mouse skin and rat mammary carcinogenesis models [41]. Silymarin [42] a COX and iNOS inhibitor [43] was also found to inhibit skin [44, 45], colon [46], tongue [47], and prostatic [45, 48] cancer development. However, the cancer preventive ability of carnosol and ginger substances, both COX inhibitors [49, 50], has not fully been investigated [51, 52]. Numerous phytochemicals thus remain to be tested in animal models for the most common cancers such as colon, mammary, prostate, and lung. Nobiletin [53–56], zerumbone [57, 58], and garcinol [59–61] are potent anti-inflammatory compounds present in plants that inhibit carcinogenesis in different tissues. A recent study demonstrated the chemopreventive ability of zerumbone, which has been shown to possess anti-inflammatory effects in mouse lung and colon carcinogenesis [62]. These phytochemicals [63], for example, curcumin [31, 64], resveratrol [65, 66], silymarin [67], from plants have multifunctional effects, including COX-2 inhibition and anti-inflammatory function, on a variety of events that involved carcinogenesis [68–70].

Most phytochemicals with chemopreventive ability [71] have direct antioxidant activity but many can also induce oxidative stress within cells when applied at high doses [11, 72]. For example, a simple phenolic acid protocatechuic acid is a potential cancer chemopreventive agent in a variety of tissues [73], but it enhances tumor promotion and oxidative stress in female ICR mouse skin via enhancing a promoter-induced inflammatory responses and promotion by affecting tyrosinase-dependent oxidative metabolism of protocatechuic acid [74]. Therefore, care should be paid for applying strong antioxidants to clinical use as a chemopreventive agent. However, recently anti-inflammatory, antioxidative chemopreventive phytochemicals targeting signal transduction mediated NF-2 related factor-2 (Nrf2), nuclear factor-kappaB (NF-κB), and activator protein (AP)-1 that are redox-sensitive factors have been highlighted [72]. We recently have demonstrated that such a compound and melatonin effectively suppress inflammation-associated colon tumorigenesis [62, 75].

Angiogenesis the formation of new blood vessels from existing vasculature has been associated with neoplasms. Angiogenesis is critical to the transition of premalignant lesions in a hyperproliferative state to the malignant phenotype, which leads to tumor growth and metastasis. The intensity of angiogenesis as assessed by counting of microvessels in neoplastic tissue acts as a prognostic factor for many solid tumors, including CRC [76, 77]. Similarly, expression of angiogenic growth factors is associated with prognosis of a variety of cancers [76, 77]. Angiogenesis enables tumors to grow larger than 1-2 mm in diameter, invade surrounding tissue, and metastasise. Angiogenesis is already targeted by chemopreventive agents at various stages of drug development or in clinical practice [76, 77]. The biomarker measured in chemoprevention must have the potential for modification by therapeutic interventional agents. In this regard, angiogenesis is a particularly attractive biomarker. There are in vivo and in vitro assays using human endothelial cell neoplasms [78], umbilical vein endothelial cells [79] and for screening potential anti-angiogenic effects of candidate chemicals. Potentially nontoxic anti-angiogenic dietary compounds include green tea polyphenols, genistein, curcumin, resveratrol, linoleic acid, hesperidin, naringenin, and allyl disulfide [80].

The defect in apoptosis mechanism is recognized as an important cause of carcinogenesis [5, 6]. A dysregulation of proliferation alone is not sufficient for cancer development as suppression of apoptotic signaling is also required. Cancer cells acquire resistance to apoptosis by overexpression of anti-apoptotic proteins and/or by the downregulation or mutation of pro-apoptotic proteins. Various studies indicate that dietary constituents, particularly phytochemicals, can modulate the complex multistage process of carcinogenesis by several mechanism(s), including apoptosis-inducing effects [81]. They include (−)-epigallocatechin gallate, curcumin, genistein, indole-3-carbinol, resveratrol, isothiocyanates, luteolin, lycopene, caffeic acid, apigenin, sylymarin, gingerol, and capsaicin [81].

3. Mechanistic Screening Assays for Detecting Potential Chemopreventive Compounds

Potential chemopreventive agents are systematically screened in a battery of short-term assays which determine the inhibition or induction of biochemical and molecular processes involved in carcinogenic processes. As summarized in Table 2, these mechanistic-based assays are roughly divided into three major categories: (i) antimutagenesis assays which evaluate carcinogenesis blocking activities, (ii) antiproliferative and antiprogression screening assays, and (iii) assays assessing antioxidant and anti-inflammatory mechanisms.

In addition to these assays, studies are in progress to establish assays utilizing DNA microarray and proteomics to facilitate the discovery of new chemopreventive agents and novel molecular mechanisms of action [82, 83]. Such new emerging technologies allow the screening and monitoring of the expression levels of thousands of genes simultaneously [84]. More importantly, the technology to make customized gene chips with specific genes is also possible. In addition to monitoring the alterations in gene expression patterns in tissues undergoing carcinogenesis, these chips can be utilized to evaluate subjects at risk, such as those carrying specific germline mutations and genetic polymorphisms [85].

4. In Vitro Efficacy Model Systems

Besides the mechanistic assays mentioned above, the systematic evaluation of cancer preventive agents includes screening compounds in short-term in vitro screens which aim to select agents for subsequent whole animal testing with insight into potential mechanisms of action. Use of primary cultured cells without aneuploidy is ideal because they possess relatively intact drug metabolizing systems and normal gene numbers. Epithelial cells used are rat tracheal epithelial cells, human lung cells, hyperplastic alveolar nodules in mouse mammary gland organ cultures, JB6 epidermal cells, and human foreskin epithelial cells. In each assay, the substances are tested over a wide range of concentrations to determine EC₅₀ values. The assays include (i) a rat tracheal epithelial cell (RTE) assay which measures the ability of candidate chemopreventive agents to block the benzo[a] pyrene (B[a]P)-induced transformation of primary RTE cells [86]; (ii) an anchorage independence assay which is an effective method for detecting compounds that block carcinogenesis in the postinitiation stages and evaluates inhibition of anchorage independence in human lung tumor (A427) cells [87]; (iii) a mouse mammary gland organ cultures (MMOCs) assay which assesses the inhibitory activity of test chemicals on the development of carcinogen-induced hyperplastic alveolar nodules (HANs) in MMOC [88]. This assay is similar in appearance to the alveolar nodules produced in mouse mammary glands in vivo [89]; (iv) an in vitro assay for antipromoters or antiprogressors which is designed to identify chemopreventive agents effective in the promotion or progressive stages of carcinogenesis in JB6 epidermal cells [90]; and (v) a human foreskin epithelial assay which determines the inhibitory potential of chemopreventive agents in blocking cell growth stimulation induced by the carcinogen propane sulfone [91].

5. In Vivo Short-Term Screening Assays

This type of short-term assays identifies agents that might block or arrest carcinogenesis in the early stages. Two experimental models, which reflect major cancers in humans, are being used. They include the rat and mouse colorectal aberrant crypt foci (ACF) assay [11, 12, 92, 93] and a rat model of breast ductal carcinoma in situ model (DCIS) [94].

6. Animal Efficacy Assays

The use of animal efficacy models to establish organ specificity and to generate dose-response, toxicity, and other pharmacological data is a crucial component of the determinant process for chemoprevention agents. These assays with thetoxicity tests are used for decisions regarding recommendations for clinical use. Numerous animal models are used to study inhibition of chemical carcinogenesis in rodents. Important criteria considered in selecting an in vivo model for screening cancer chemoprevention agents include (i) short study duration and induction of carcinogenesis; (ii) target-specific experimental model evidenced by the induction of cancer in the target tissues comparable in such factors as histological type and hormone dependence to those found in humans; (iii) evaluation of in vitro mechanistic activities, efficacy profiles, and relevant published data prior to the selection of models for a given possible chemopreventive agent. Typically, test agents are administered in the diet unless problems with stability are encountered. During the course of chemoprevention studies a maximum tolerated dose (MTD), defined as the highest dose level that does not cause ≥10% reduction or gain in body weight over a six-week period, is determined. The treatment schedules include the administration of test agents either before, concurrently, or following exposure to the carcinogen. Efficacy is based upon the percent inhibition of tumor incidence and/or multiplicity, or increased tumor latency in comparison to carcinogen-treated controls. Representative carcinogenesis models are listed in Table 3.

7. Transgenic and Gene-Knockout Animal Models

Animal models that mimic the specific characteristics of human carcinogenesis may prove to be a valuable resource in both evaluating chemopreventive efficacy and identifying appropriate biomarkers for measuring the chemopreventive activity. Transgenic and gene knockout mice that carry well-characterized genetic lesions predisposing them to carcinogenesis are appropriate models for chemoprevention testing (Table 4). Some of the best developed models include the multiple intestinal neoplasia (Min) mouse [141] and other strains possessing lesions in the Apc gene [142]. The Min mouse carries an Apc mutation similar to that found in human familial adenomatous polyposis (FAP) patients. These mice are predisposed to develop predominantly small intestinal adenomas, but a few in the large intestine. By manipulating two or more carcinogenesis-associated genes, such as modifier genes, in a single animal, closer approximations of human carcinogenesis may be possible. Numerous colonic tumors develop in the large bowel in Min mice at 3 weeks after one-week-exposure of DSS [143], thus suggesting the importance of gene-environmental interaction in cancer development [68]. It might be feasible to knock out p53 in an animal that already carries another tumor suppressor defect such as Apc or p16. Recently, new transgenic animal models for mammary [144], tongue [145, 146], pancreas [147], and gall bladder [148] cancers have been reported. These models might be useful for discovering possible novel cancer chemopreventive agents.

8. Combination Treatment

One strategy for improving the efficacy and lessening toxicity is using combinations of agents [149, 150]. Synergistic or additive effects may be observed when two agents with different mechanisms of action are combined [149]. Such improved activity of inhibition may allow either or both of the agents to be given at lower doses, thereby reducing the toxic potential. Examples of chemical combinations producing positive results in experimental animal models include all-trans-N-(4-hydroxyphenyl)retinamide (4-HPR) and tamoxifen in the rat mammary gland [151] and 2-dinuoromethylorinithine (DFMO) and piroxicam in the rat colon [152, 153]. The identification and evaluation of other effective agent combinations is an interesting and important research effort for chemopreventive agent development [154, 155]. Since different treatment schedules of the two drugs can be compared and evaluated for synergism, additivity, or antagonism using a quantitative method based on the median-effect principle of Chou and Talalay [156–159] can identify the ideal combination treatment for obtaining an improved effect in comparison to a single exposure of the potential chemopreventive agent [160–162].

9. Intermediate Biomarkers

Carcinogenesis proceeds through a very long preclinical period. The collective hope is that multiple opportunities exist for chemoprevention to arrest or reverse progression towards malignancy. Intermediate biomarkers of malignant neoplasms are the phenotypic, genotypic, and molecular changes that occur during carcinogenesis [69]. An important component of chemopreventive agent development research is the identification and characterization of intermediate biomarkers that may serve as surrogate end points for cancer incidence reduction in chemoprevention clinical trials [163]. Such efforts are critical to the progress of chemoprevention and have the potential for cost-effective development of chemopreventive research [5, 6, 69]. Biomarkers include proliferation, apoptosis, growth factors and their receptors, genetic alterations, and so forth in the target tissues [162, 164, 165]. In the hope of faster progress with fewer subjects and lower total cost, much effort is being expended on the search for reliable biomarkers to predict the likelihood of developing cancer and/or to signal the effectiveness of chemopreventive therapy [166]. Considerable attention has been paid to identifying those markers that can act as surrogate markers for cancer development since favorable modulation of the surrogate end-point biomarker may demonstrate the effectiveness of a putative preventive treatment. However, the complexity of the biology challenges the ability to measure the effectiveness of attempts to arrest or reverse carcinogenesis, other than through costly and time-consuming prospective trials with the disease state as the endpoint. Despite much work, to date no prehistological biological or molecular intermediate marker has yet been validated for sporadic cancers.

For chemoprevention of prostate cancer [167] and other types of cancer, natural and synthetic agents that may suppress, reverse, or regress precancer and delay progression to invasive cancer can be used [77]. Epidemiological evidence suggests that environmental factors, such as diet, play a role in the development and progression of prostate cancer. The number of potential protective dietary compounds or whole dietary products that are indicated to have preventive effects is piling up and demands further evaluation. To face this scientific field, a strategy that combines prostate-specific antigen (PSA)-based clinical trials with experimental human xenograft studies to evaluate potential chemopreventive agents against prostatic cancer is proposed [168, 169]. PSA for prostate adenocarcinoma, even though has relatively poor specificity, is cheap and easily followed with minimally invasive procedures.

10. Future Direction

Epidemiologic evidence suggests that green tea polyphenols reduce the risk of some forms of cancer, while data for other commonly used phytochemicals is less convincing. Part of the problem lies in precise quantitation of foodborn phytochemicals where multiple dietary sources of a particular phytochemical are involved, as in recent studies of dietary quercetin on the prevention of heart disease. Future investigations may focus on specific foodborn phytochemicals and/or botanicals capable of intervening in critical pathways in carcinogenesis as indicated by the possible use of “natural” NSAIDs in the prevention of CRC and the soy-based phytoestrogens in the prevention of breast cancer. Much is to be learned by collaboration between cancer researchers and ethnopharmacologists for probing dietary prevention of cancer. In addition, the development and validation of new in vitro and in vivo assays of high predictive value for discovering agents with human chemopreventive potential is needed. Care, however, is needed when extrapolating in vitro data to in vivo models because it cannot be assumed that the effects seen when cells are exposed directly to active compounds that would be candidate chemopreventive agents will be seen when they are consumed in the diet. We should investigate whether they are capable of distribution throughout the body when they are absorbed after ingestion. The understanding of the molecular and cellular processes, such as gene and protein expression, apoptosis, angiogenesis, signal transduction, involved in carcinogenesis is rapidly increasing. Currently several translational studies using phytochemicals and bioactive compounds, such as indole-3-carbinol [170], ellagitannins [171], sulforaphane [172], lycopene [173], diallyl trisulfide [174], omega-3 fatty acids [175], proanthocyanidins [176], green tea polyphenols [177], genistein [178], silymarin [179], and curcumin [180, 181], from plants are underway.