Lipid Peroxidation in Colorectal Carcinogenesis: Bad and Good News

and pro-apoptotic in colon cancer cells. These results, even obtained with very low doses of Colorectal cancer is a common disease, affecting millions worldwide and represents a global health problem. Effective therapeutic solutions and control measures for the disease will come from the collective research efforts of clinicians and scientists worldwide. This book presents the current status of the strides being made to understand the fundamental scientific basis of colorectal cancer. It provides contributions from scientists, clinicians and investigators from 20 different countries. The four sections of this volume examine the evidence and data in relation to genes and various polymorphisms, tumor microenvironment and infections associated with colorectal cancer. An increasingly better appreciation of the complex inter-connected basic biology of colorectal cancer will translate into effective measures for management and treatment of the disease. Research scientists and investigators as well as clinicians searching for a good understanding of the disease will find this book useful.


Mutagenicity
In this context, it would be important to know whether the HNE-dG adducts are mutagenic. Several authors suggest this possibility, since HNE has been shown to be mutagenic in mammalian cells (Cajelli et al., 1987). HNE was negative in bacterial mutagenicity tests, however its epoxidized form has been tested positive (Chung et al., 1993). HNE was found to be responsible for recombination, base substitutions and frameshift mutations in M13 phage transfected in E.coli (Kowalczyk et al., 2004). Moreover it has been reported that 50 μM HNE treatment in human cells induces a high frequency of G.C to T.A mutations at the third base of codon 249 (AGG*) of the p53 gene (Hussain et al., 2000), a mutational hot spot in human cancers, particularly in hepatocellular carcinoma (Hsu et al., 1991). Both eheno and propane type HNE-DNA-adduct at codon 249 can be responsible for such transitions (Feng et al., 2003). The stereochemistry of HNE-dG adducts seems to play an important role in determining mutations. Indeed, two of the HNE-dG adducts, (6R, 8S, 11R) and (6S, 8R, 11S), were significantly more mutagenic than (6R, 8S, 11S) and (6S, 8R, 11R) HNE-dG adducts. Only one of the HNE stereoisomers was able to form interstrand DNA-DNA cross-links. (Fernandes et al., 2003).

Genotoxicity
The genotoxic property of HNE was demonstrated in different cell types, such as on cultured human lymphocytes (Emerit et al., 1991), in primary hepatocytes  and cerebral microvascular endothelial cells (Eckl, 2003). In these cell lines an increase of micronuclei (a biomarker of chromosome breakage and/or whole chromosome loss), chromosomal aberrations and sister chromatid exchanges was observed after exposure to HNE at relatively low doses, ranging 0.1-10 μM. However these clastrogenic features of low doses of HNE failed to be confirmed in a recent multicentrum study on DNA of normal peripheral blood lymphocytes (Katic et al., 2010). Currently, the comet assay has been extensively used to measure DNA strand breaks, since it represents a sensitive and rapid assay to detect the mutagenic and genotoxicity of chemicals and xenobiotics (Tice, et al., 1991). Unfortunately, most HNE-induced DNA lesions are the stable 1,N2-propano adducts and they are not detected by this technique. By using this assay, the genotoxic property of 5-10 μM HNE in the K562 leukemic cell line has been shown; this feature was highly dependent on cellular GSH/GST/AR system (Yadav et al., 2008). The comet test was also used to demonstrate the genotoxicity of 200 μM HNE in human leukocytes (Park & Park, 2011).

Results in laboratory animals
To date, in contrast with several in vitro experimental results, tumor bioassays in laboratory animals failed to demonstrate the carcinogenic and mutagenic properties of HNE. HNE, in particular its epoxy derivate, has shown that to be a weak tumor-initiating agent, causing the development of renal preneoplastic tubule lesions in new-born mice (Chung et al., 1993). More interestingly, HNE lacks in vivo genotoxicity in lacI transgenic mice, a model for detecting mutagenicity in target organs, even when lethal doses are applied . The big gap between the in vitro and the in vivo data can be partially explained by carefully considering the elevated doses frequently used to demonstrated the carcinogenic properties of HNE in vitro. Indeed, several mutagenic assays with HNE have been performed with high doses of HNE (more than 100 µM). It seems rather unlikely that HNE or other aldehydes can reach overall concentrations in the range of 100 µM in cells and organs . It is conceivable that such levels may be built up locally, near or within peroxidizing membranes for a short time because of their high lipophilicity. It has been calculated, for example, that the concentration of HNE in the lipid bilayer of isolated peroxidizing microsomes is about 4.5 mM (Koster et al., 1986). Nevertheless, a convincing demonstration that this very high concentration can be reached into the cells has remained elusive. On the other hand, when HNE diffuses out from membranes, its concentration is reduced by the surrounding aqueous phase. Moreover, the cytosolic HNE-metabolizing enzymes destroy HNE produced in excess so that the steady-state HNE concentration into the cells, around 1 μM, is reached quickly Dianzani et al., 1999).

Cellular responses and signal transduction
As previously indicated, the adduct formation between HNE and DNA is only one of the several biological effects determined by this aldehyde. Indeed, HNE is considered as a signalling molecule influencing proliferation, differentiation and apoptosis of cancer cells (Dianzani, 2003;Leonarduzzi et al., 2004;Poli et al., 2008;Pizzimenti et al., 2010b). The majority of experimental evidence indicate an antiproliferative role of HNE, when added at low doses (1-10 μM) to cultured cells. The inhibition of proliferation has been observed in leukemic (HL-60, K562, U937, MEL, ML-1) (Barrera et al., 1991;Barrera et al., 1987, Rinaldi et al., 2000Pizzimenti et al., 2006), neuroblastoma (SK-N-BE) (Laurora et al. 2005), hepatoma (7777, J42) (Muzio et al., 2001;Canuto et al., 1999), osteosarcoma (SaOS2; HOS) (Calonghi et al., 2002;Sunjic et al., 2005), prostate cancer (PC3) (Pettazzoni et al., 2011) cells. This antiproliferative effect is sustained by the modulation of key genes involved in cell growth control, such as oncogenes (c-myc, c-myb, fos, AP1, cyclins) and anti-oncogenes (pRB, p53, SUFU-1, Mad-1) (Poli et al., 2008;Pizzimenti et al., 2009). Interestingly, the effect of HNE in normal cell proliferation is more variable if not opposite to that observed in tumor cells. For example, HNE has no effect on normal myeloid stem cells (Hassane et al., 2008) or on human peripheral blood lymphocytes (Semlitsch et al., 2002), while the respective tumour was sensitive to the anti-proliferative effect of aldehyde. On the contrary, in vascular smooth muscle cells 0.1 μM HNE stimulated cell proliferation (Kakishita et al., 2001). In several cell lines, the inhibition of proliferation was accompanied by apoptosis. The mechanisms of HNE-induced apoptosis through the extrinsic and intrinsic pathways, its self-regulatory role in this process and its interaction with Fas (CD95), p53, and Daxx has been recently reviewed (Awasti et al., 2008). HNE is also able to induce differentiation, as observed in HL-60, MEL, K562 and SaOS osteosarcoma cells (Barrera et al., 1991;Rinaldi et al., 2000;Calonghi et al., 2002;Cheng et al., 1999;Fazio et al., 1992). Moreover, HNE was shown to induce features of typical differentiated cells, such as chemotaxis (Curzio et al., 1988), phagocytosis and the ability to induce respiratory burst (Barrera et al., 1991) in myeloid cells. HNE also demonstrated the ability to regulate the replicative potential of cells, by inhibiting the telomerase activity. Indeed, in HNE-treated leukemic cells, the expression of the hTERT gene was downregulated by modulating the expression of transcription factors belonging to the Myc/Mad/Max network (Pizzimenti et al., 2006).

160
The anti-tumoral properties of HNE are also sustained by the demonstration of antiangiogenic properties. Stagos and collaborators demonstrated that 5 and 10 μM HNE were able to inhibit the tube formation of human bone marrow endothelial cells (HBMEC) (Stagos et al., 2009). However, conflicting results have been reported, since it has been demonstrated that 1 μM HNE induces an increase of VEGF expression in human retinal pigment epithelial cells (Ayalasomayajula & Kompella, 2002). The cellular responses to HNE are sustained by affecting cell signalling at multiple levels. Relevant findings in this area have been extensively reviewed (Poli et al., 2008;Leonarduzzi et al., 2004;Dianzani et al., 1999). In addition to the above cellular responses presented, HNE activates various cytoprotective, stress response pathways, promoting changes in gene expression that facilitate cell survival and recovery from stress (West & Marnett, 2005). For example, HNE activates the transcription factors Nrf2 (Nuclear factor erythroid-derived 2-like 2) and HSF1 (heat shock factor 1), which mediate the antioxidant and heat shock responses, respectively (Jacobs & Marnett, 2007). Nrf2 acts by binding Antioxidant Responsive Elements (ARE) sequences on promoters of certain genes promoting their expression (Thimmulappa et al., 2002). In regard to HNE metabolism, functional ARE sequences have been found on promoter of GST A4 ALDH and ADH (Reddy et al., 2007;Malhotra et al., 2010). Moreover, Nrf2 promotes de novo GSH synthesis by up-regulating expression of the GSH synthesis pathway (Harvey 2009). Nrf2 is controlled by both translational and post-translational mechanisms, in particular the protein Kelch-like ECH-associated protein 1 (KEAP1) mediates Nrf2 ubiquitinaltion followed by proteasomal destruction (Kaspar et al., 2010). In conditions of oxidative stress or in response to many chemicals KEAP1 undergo conformational changes responsible for loss of Nrf2 binding activity. As a consequence Nrf2 can accumulate, translocate in the nucleus and drive expression of the antioxidant program (Reddy et al., 2007). The heat shock response mediates the induction of a highly conserved set of heat shock proteins (Hsps) (Mosley, 1997). The inducible expression of Hsps is mediated by heat shock transcription factor 1 (HSF1), which translocates to the nucleus upon activation and enhances the expression of genes to form promoters containing heat shoch elements (HSE), such as Hsp70 (Sarge et al., 1993;Baler et al., 1993). A principal function of Hsps is to chaperone other proteins, binding to nascent polypeptide chains as well as to unfolded and damaged proteins. Their function as protein chaperones aids in the recovery of cells from thermal and chemical-induced damage (Hahn & 1982;Howard, 1993). In addition to acting as protein chaperones, Hsps inhibit cell death by directly inhibiting a variety of proapoptotic mediators, such as HNE (Jacobs et al., 2007). It is very likely that the majority of effects observed on cell signalling and cellular responses can be mediated by the reaction of HNE to proteins and peptides. Quantitatively, proteins and, among peptides, the GSH, represent the most important group of HNE-targeted biomolecules. It was estimated that 1-8% of the HNE formed in cells will modify proteins (Siems & Grune, 2003). Most of the identified targets are enzymes, carriers, receptors, ion channels, transport proteins, cytoskeletal, heat shock proteins and others. The biological significance of the HNE-protein adducts identified have been reviewed by several authors (Uchida, 2003;Poli et al., 2008). Some of the protein-adducts identified can explain the antitumoral effect exerced by this aldehyde. For example, it was demonstrated that the inhibition of cell proliferation in the human colorectal carcinoma cell line (RKO) and human lung carcinoma cell line (H1299) by HNE was mediated by the direct reaction of HNE with IκB kinase (IKK), the key enzyme regulating the NF-κB activation (Ji et al., 2001b). Moreover the HNE adducts with alpha-enolase, at the cellular surface of leukemic cells, suggest a new role for HNE in the control of tumour growth and invasion, since HNE causes a dose-and time-dependent reduction of the plasminogen binding to alpha-enolase. As a consequence, HNE reduces adhesion of HL-60 cells to HUVECs (human umbilical vein endothelial cells) (Gentile et al., 2009). New perspectives of HNE role in cancer-inducing signaling pathways have recently emerged, by recent findings on microRNA (miRNA) ), a class of conserved non-coding small RNAs, which regulate gene expression by translation repression of coding mRNAs (Bartel, 2004).

HNE content in human cancers
Several studies on human cancer tissues have analysed the HNE or HNE-protein adduct content, in order to find a possible correlation between this marker of lipid peroxidation and the progression of cancer. HNE content has been reported to increase along with the progression of breast cancer (Karihtala et al., 2011) and astrocytoma . In human renal cell carcinoma, immunohistochemistry for HNE-modified proteins showed positive staining in the cytoplasm of tumor cells, with respect to controls, without correlation to the clinical stage (Okamoto et al., 1994). However other reports have demonstrated the opposite: a low or undetectable lipid peroxidation, as well as HNE content, such as in hepatomas (Dianzani, 1993). Several studies have shown elevated lipid peroxidation markers in the sera, plasma or urine of breast carcinoma (Hung et al., 1999;Chandramathi et al., 2009), cervical intraepithelial neoplasia and carcinoma of the cervix (Looi et al., 2008), head and neck squamous cell carcinoma (Gupta et al., 2009) and prostate tumor (Kotrikadze et al., 2008), compared to healthy controls. However, these extratumoral measurement are likely, at least partly, to reflect generalized oxidative stress and /or inflammation in the whole body.

Sources and fate of HNE in colon
Colon cells can be exposed to HNE derived form different sources ( Figure 1). It is possible to find HNE directly in the food (Gasc et al., 2007), since it can derived from lipid peroxidation of PUFAs introduced with diet or from endogenous PUFAs presents in cellular membranes. A small amount of HNE can reach the colon also via bile. Following a single intravenous administration of [3H]-HNE, five metabolites were present in the bile, namely GSH-HNE, GSH-DHN, DHN, and HNA-lactone mercapturic acid conjugates (Laurent et al., 1999). Within 4 hr from injection of the radiolabel 3[H]-HNE, 19.5% of the injected radioactivity was found in the bile, whereas only 3% was found in the feces within 48 hr (Laurent et al., 1999). The existence of an enterohepatic circulation for HNE metabolites has been unequivocally demonstrated (Laurent et al., 1999) using a model linking donor rats (injected intravenously with [4-3H]HNE) and recipient rats (to which the bile from donor rats was delivered intraduodenaly). This enterohepatic circulation, approximately 8% of the total dose, may explain the low amount of radioactivity recovered from faeces when rat were dosed intravenously with [4-3H]HNE.

Fig. 1. Sources and fate of HNE in colon
Metabolic transformation of HNE starts in enterocites, where GSH-HNE is the main metabolite produced (Grune et al., 1991). The majority of HNE metabolites are found in the urine. Indeed, following the intravenous administration of [3H]-HNE in rats, 67%, 3%, 0.16%, and 6.5% of the injected radioactive dose was recovered from urine, faeces, liver and remaining tissues, respectively (Alary et al., 2003). The urinary HNE metabolites were separated by HPLC and the resolved peaks were identified as mercapturic acid conjugates of HNA, DHN, HNE and HNA-lactone, where DHN-MA, and to a lesser extent HNA lactone-MA, have been found to be the major urinary metabolites of HNE in rats (Boon et al., 1999). DHN-MA has been confirmed to be the major urinary HNE metabolite also in human urine (Alary et al., 1998). The microflora of the human intestine can also affect levels of lipid peroxidation, since the antioxidative effect of lactic acid bacteria has been demonstrated . In particular, the antioxidative activity of Bifidobacterium longum ATCC 15708 and Lactobacillus acidophilus ATCC 4356 was measured based on the inhibition of linoleic acid peroxidation. Both intact cells and intracellular cell-free extracts of B. longum and L. acidophilus demonstrated an antioxidative effect on inhibiting lipid peroxidation. The antioxidative activity ranged from 38 to 48% inhibition of linoleic acid peroxidation. This indicates that B. longum and L. acidophilus have a very strong antioxidative effect on inhibiting lipid peroxidation .
Low level of HNE and its metabolites can be found also in faecal water (Alary et al., 2003) and numerous studies have emphasized the lipid peroxidation products of faecal water in colon cancer as diet-related factors (Lapre et al., 1992;Glinghammar et al., 1997).

Pro-tumoral and anti-tumoral role of HNE in colon carcinogenesis
Several authors have reported evidence that sustains the pro-tumoral activity of HNE and other products of lipid peroxidation in colon carcinogenesis. These findings include in vitro (see table 1) and in vivo studies, which demonstrate the genotoxic properties of HNE on coloncarcinoma cell lines, the increase of HNE content along with the progression of colorectal cancer and the increase of HNE-DNA adducts in vivo. However other studies, seem to demonstrate the opposite (see table 1). Consistent with the hypothesis of an antitumoral role of HNE are the results showing the inhibition of cell growth, the induction of apoptosis in several colon cancer cell lines, as well as the demonstration that HNE content decreases in biopsies of colon-cancer tissues with respect to normal mucosa. A deeper discussion of these opposite results is here reported.

HNE-DNA adducts in colon and colon cancer
HNE-dG adducts, were found in normal human colon tissue, as well as DNA adducts with other lipid peroxidation products, such as acrolein and MDA (Chung et al., 2000). The levels of HNE-dG in tissue DNA examined so far are estimated to be in the range of 3-9 adducts per billion bases (3-9 nmol/mol guanine) (Chung et al,. 2000).
The etheno-DNA adducts, inter alia formed from epoxidized HNE, were found at increased level in colonic polyps of familial adenomatous polyposis (FAP) patients. Mean adduct levels in FAP polyps were 65 εdA/109 and 59 εdC/109 parent nucleotides, being 2 to 3 times higher than in unaffected colon tissue (Schmid et al., 2000). Interestingly, the level of etheno-DNA adducts in colon carcinoma tissues were found to be similar to unaffected colon (Schmid et al., 2000), suggesting a possible HNE role in the early events of colon carcinogenesis.
On the contrary, Obtułowicz and collaborators (2010) have found that, in colon cancer patients, the DNA-HNE adducts εdA and εdG, measured both in colon tissues and blood leukocytes, were lower in patients than in controls (Obtułowicz et al., 2010). These authors have measured the two corresponding metabolites also, 1,N6-Ethenoadenine (εAde) and 3,N4-ethenocytosine (εCyt), catalized by BER, the major pathway of etheno adduct elimination from DNA (Obtułowicz et al., 2010). Both excision activities were significantly higher in tumor than in normal colon tissues and this feature could be explained by the increased level of abasic site endonuclease (APE1), belonging to BER system, in coloncancer patients with respect to controls (Obtułowicz et al., 2010). A possible pro-cancinogenic role of etheno-DNA adducts is also sustained by the finding that in the colon of patients with inflammatory bowel disease εdC, but not εdA, are increased. In particular it has been demonstrated that εdC was 19-fold higher in colonic mucosa of Crohn's disease and 4-fold higher in the colonic mucosa of ulcerative colitis patients, when compared to normal tissues . Since patients with ulcerative colitis (UC), and Crohn's (CD) have an elevated risk for developing colon cancer (Konner et al., 2003), the authors suggest that the promutagenic etheno-DNA adducts, generated as a consequence of chronic inflammation, can act as a driving force to malignancy in cancerprone inflammatory diseases . HNE can also contribute to induce colon carcinogenesis, by inhibiting the DNA repair mechanism of such adducts. Indeed, Feng and collaborators (2004)

Genotoxicity and mutagenicity of HNE in colon cancer
Comet assay demonstrated that HNE, at concentration higher than 150 μM, displays a genotoxic effect in the colon carcinoma cell line HT-29 (Glei et al., 2006;Ebert et al., 2001;Knoll et al., 2005) and in HT29clone19A, a permanently differentiated sub-clone treated with sodium butyrate (Augeron & Laboisse, 1984). Moreover, such high doses of HNE were able to affect DNA integrity in primary human colon cells (Schaferhenrich et al., 2003;Glei et al., 2007) and in LT97, an established cell line derived from a differentiated microadenoma, representing a model of an early premalignant genotype, carrying adenomatous polyposis coli (APC) and Ki-ras mutated, but normal p53 (Richter et al., 2002), three well-characterized genes involved in coloncancer progression (Fearon et al., 1990).
Genotoxicity of HNE is higly dependent on cellular GSH level. Indeed, GSH depletion leads to and increase of HNE genotoxicity in the HT-29 colon carcinoma cell line (Knoll et al., 2005). Moreover, HNE displayed a higher genotoxicity in LT97 than in HT29clone19A and primary human colon cells. This result can be explained by the lower GST expression found in LT97 compared to HT29clone19A and primary human colon cells (Schaferhenrich et al., 2003).
Recently, by using a refined comet assay (Comet-FISH) , which combined the classical comet assay with the fluorescence in situ hybridisation, it has been demonstrated that HNE concentrations higher than 150 μM w e r e a b l e t o a f f e c t D N A integrity on the p53 (Schaferhenrich et al., 2003;Glei et al., 2007), Ki-Ras and APC genes (Glei et al., 2007), in primary human colon cells and the colon adenoma cell LT97. After cell incubation with HNE, the p53 gene, the crucial target gene for the progression of adenoma to carcinoma, migrated more efficiently into the comet tail than the global DNA, indicating a high susceptibility of the p53 gene to HNE (Glei et al., 2007). Moreover, the TP53 gene sensitivity to the DNA damage induced by HNE was significantly higher with respect to APC and KRAS genes. This particular sensitivity is especially apparent in LT97 cells (Glei et al., 2007). This may be due to the fact that LT97 cells normally carry damaged APC and KRAS, but undamaged TP53 (Richter et al., 2002). In normal colonocytes, APC and KRAS were also sensitive to damage (Glei et al., 2007). These findings are highly interesting when considering the sequence of mutational events that occur during human colon carcinogenesis (Vogelstein et al., 1988). APC and KRAS mutations transform normal epithelial (stem) cells into initiated, more rapidly proliferating cells to yield dysplasia and small adenoma. TP53 mutations in adenoma are then crucial alterations leading to further progression and to carcinoma. Based on studies of Glei and collaborators (2007), it is possible to conclude that HNE could potentially contribute to both cancer initiation and progression in the colon, if produced in sufficient amounts. However, as mentioned in the previously chapter, it is unlikely that HNE is reaching such high concentrations (150 μM) in colon in vivo. Moreover it still remains to be studied to what extent the observed genotoxicity of HNE is related to mutagenicity. Consistent with this hypothesis, as previously reported, it has been demonstrated that 50 μM HNE treatment in human TK-6 lymphoblastoid cell line induces a high frequency of G.C to T.A mutations at the third base of codon 249 (AGG*) of the p53 gene (Hussain et al., 2000), a mutational hot spot in human cancers, particularly in hepatocellular carcinoma (Hsu et al., 1991). The adduct of HNE to codon 249 of the p53 gene has been also found by Hu and collaborators (2002). These authors exposed DNA of exons 5, 7 and 8 of human p53 gene, where the large majority of p53 mutations occur, to a very high concentration of HNE (192 mM or more). They identified two main HNE adducts, the first already mentioned at codon 249 (exon 7) and the second at codon 174 (exon 5) (Hu et al., 2002). However, the possible contribution of HNE to p53 mutations, through the formation of DNA adducts remains to demonstrated, since codon 249 and codon 174 of p53 usually are not mutated in colorectal rectal. Indeed, mutations at codon 175, 245, 248, 273, and 282 account for approximately 43% of all p53 mutations in CRC (Soong et al., 2000;Soussi et al., 2000;Soussi & Beroud, 2003).

HNE role in controlling cell proliferation, apoptosis of colon cancer cell
Several findings have been collected through the years related to the anti-proliferative and pro-apoptotic in colon cancer cells. These results, even obtained with very low doses of HNE, easily reacheable in vivo, cast doubt on the pro-tumoral HNE role. 1 μM HNE is able to inhibit cell proliferation of Caco-2 and HT-29 colon cancer cells (Cerbone et al., 2007;Pizzimenti et al., 2010b;Vizio et al., 2005) and concentrations ranging form 1 to 100 μM are able to induce apoptosis in Caco-2, HT-29, RKO, HCT15 colon cancer cells. (see table I for references). A number of genes or cell signalling pathways have been found to be affected by HNE, and their modulation can explain the biological effects observed.
Results obtained in our laboratories demonstrated that the inhibition of proliferation in Caco-2 and HT-29 colon carcinoma cells by 1 μM HNE is sustained by the down-regulation of telomerase activity and hTERT expression, the catalytic subunit of telomerase (Pizzimenti et al., 2010b). The major mechanism of HNE action seems to be the modulation of expression and activity of transcription factors belonging to the Myc/Mad/Max network (Pizzimenti et al., 2010b). After HNE treatment, apoptosis of several colon cancer cell lines was investigated by different authors and different pathways were considered to be involved. In Caco-2 human colon adenocarcinoma cell line, 1 μM HNE caused an increase of bax expression (Cerbone et al., 2007) and the apoptosis induction is mediated by JNK activation. Indeed, the HNEmediated apoptotic cell death was significantly prevented by preincubating the cells with the selective JNK inhibitor SP600125 (Biasi et al., 2006). Ji and collaborators investigated the mechanism of HNE-induced cell death in human colorectal carcinoma cells and found that HNE-induced apoptosis depends on alteration of mitochondrial function, leading to the release of cytochrome c and subsequent activation of caspase cascade (Ji et al., 2001a). The authors have further demonstrated that HNE inhibited IκB kinase activity by direct interaction with IκB kinase and suggested that HNE is an endogenous inhibitor of NF-κB activation that acts by preventing IκB kinase activation and subsequent IκB degradation (Ji et al., 2001b). The molecular mechanism of HNE induced apoptosis was investigated in RKO colon cancer cells also. In this cell line, beside the pro-apoptotic stimuli, HNE activates the stress response pathways, that abrogate programmed cell death. Moreover, HNE elicits the nuclear translocation of HSF1 and promotes Hsp40 and Hsp72 expression (Jacobs & Marnett, 2007). The silencing of HSF1 sensitizes the colon cancer cells to HNE-induced apoptosis, through a mechanism involving the control of BCL-XL, BAG3 protein turnover (Jacobs & Marnett, 2007;Jacobs & Marnett, 2009)

HNE content in human colon cancers
Only a few studies have investigated the level of the lipid peroxidation products, in particular HNE, in human colon cancers and results are contradictory. It has been demonstrated that the levels of proteins modified by HNE and MDA in colorectal cancer tissues were significantly increased (Murawaki et al., 2008). By immunohystochemical analysis, Murawaki and collaborators (2008) have demonstrated that the proteins modified by HNE were stained diffusely in the cytoplasm of cancer cells, while they were weakly stained in normal tissues. Similar results have been obtained by Kondo and collaborators (1999). Immunostaining of HNE-histidine adducts was observed in the cytoplasm of colon cancer tissues. Immunoreactivity was also found in the cytosol of infiltrating inflammatory cells. Western blot analysis of HNE-histidine adducts confirmed the results, since larger amounts of modified proteins were detected in carcinomas than in nontumorous epithelial counterparts (Kondo et al., 1999). The authors also demonstrated that HNE content increased along with the progression of colorectal cancer, since tubular adenoma cells revealed a weaker staining, similar to the staining of non-tumorous epithelial cells (Kondo et al., 1999). An increase of HNE content in colon cancer tissues have been found also by Skrzydlewska and collaborators (2005). These authors analyzed the HNE content in homogenates of human colon cancer tissues, by measuring HNE as a fluorimetric derivative. These authors have demonstrated that the level of HNE was significantly increased (P<0.001) in cancer tissue compared to control group, with highest in G3-grade adenocarcinoma and mucinous adenocarcinoma and clinical IV stage of colorectal cancer.
In contrast with these results, other scientists demonstrated a decrease of HNE in colon cancer tissues. Indeed, it was demonstrated that HNE was significantly decreased in cancer specimens, with respect to normal tissues, by measuring the HNE content in tissue biopsies from patients with colon adenocarcinoma of different TNM and G stage (Biasi et al., 2002;Zanetti et al., 2003). This result was confirmed later by the same group (Biasi et al., 2006). Moreover, Chiarpotto and collaborators (1997) have demonstrated that the fluorescent adducts with plasma proteins and HNE were significantly lower in the plasma from cancer patients (all stage G3, pT3pN0) than in controls.

HNE metabolism in colon cancer
In colon cells, the enzymes of HNE metabolism are present. Staining with anti GST A4 specific antibodies revealed a significant expression of GST A4 in columnar and crypt epithelial cells of normal colon mucosae (Desmots et al., 2001), as well as in colon cancer cell lines (Scharmach et al., 2009;Knoll et al., 2005). Moreover, both the oxidative and reductive metabolisms of HNE are well represented in colon cells, since both ALDH or ADH have been found to be significantly expressed in colon mucosae (Seitz et al., 1996;Yin et al., 1994). The expression of AR is also enhanced in various forms of cancer, such as hepatoma (Zeindl-Eberhart et al., 1997) and melanoma cancer (Kawamura et al., 1999). By affecting HNE metabolism enzymes, it is possibly to modulate the HNE concentration inside cells. This could be critical for cancer growth regulation or DNA genotoxicity. Indeed, butyrate, produced during gut fermentation, has a chemoprotective role toward HNE injury, when added at high concentration, such as 100-200 μM in HT-29 colon cancer cells (Knoll et al., 2005). The chemoprotective effect of butyrate seems to be related to the increasing the expression of glutathione S-transferases GSTP1 (Ebert et al., 2001) and hGSTA4-4 (Knoll et al., 2005) able to catalyze the conjugation of HNE with glutathione. Similar results were obtained in HT-29 cells by using two wheat bran-derived arabinoxylans, fermented under anaerobic conditions in human feces. These two fermentation products inhibited growth and reduced the genotoxicity of HNE (100-200 μM) via up-regulation of the activity of GSTs, in absence of a GSTP1 or hGSTA4-4 increase (Glei et al., 2006). There is a growing interest in targeting aldose reductase (AR), as a novel therapeutic approach in preventing progression of colon cancer (Tammali et al., 2011). AR besides reducing aldo-sugars efficiently reduces toxic lipid aldehydes and their conjugates with glutathione (Tammali et al., 2006). Indeed, inhibition of AR by sorbinil or by antisense ablation, prevented FGF-induced and PDGF-induced proliferation of Caco-2 cells at Sphase (Tammali et al., 2006). Similar results were also obtained in other colon cancer cell lines, by Ramana and collaborators which show that the inhibition of AR prevents epidermal growth factor (EGF)-and basic fibroblast growth factor (bFGF)-induced HT29, and cell proliferation, by accumulating cells at the G1 phase of the cell cycle, through the AKT/Phosphoinositide 3-Kinase/E2F-1 pathway. Analogous results were obtained in SW480 and HCT-116 colon cancer cells (Ramana et al., 2010). More interestingly, in vivo studies showed that administration of aldose reductase-small interfering RNA (siRNA), or the AR inhibitor fidarest, to nude mice bearing SW480 human colon adenocarcinoma cells, led to a complete arrest of tumor progression. Such evidence suggests a key role for aldose reductase in growth factor-induced proliferation in colon cancer cells and it points to inhibition of aldose reductase as a novel therapeutic approach in preventing progression of colon cancer (Tammali et al., 2006;Ramana et al., 2010).
Recently, the ATP-depent transporter RLIP76 (Ral binding protein1) has been considered for its role in controlling HNE content inside the cells. Indeed, it has been demonstrated that this transporter with multi-specific transport activity towards glutathione-conjugates and chemotherapeutic agents, is also specific for GSH-HNE (Sharma et al., 2002). The expulsion of GS-HNE from cells represents another critical step in HNE detoxification since it avoids the accumulation of adducted GSH and permits the restoration of GSH/GSSG equilibrium. RLIP76 protein is frequently overexpressed in cancer lesions (Vatsyayan et al., 2010), included colon cancers (Singhal et al., 2007), thus there is a growing interest in considering this protein as target in cancer therapy (Vatsyayan et al., 2010). When RLIP76 is inhibited, a rapid increase in HNE-GSH is observed, both in vitro (Awasthi et al., 2003;Cheng et al., 2001;Yang et al., 2003) and in vivo (Vatsyayan et al., 2010). Recent studies show that the inhibition and/or depletion of RLIP76 by antibodies, siRNA, or antisense can lead to a drastic and sustained regression of lung, kidney, melanoma, prostate, and colon cancer xenografts with no observed recurrence of tumors (Vatsyayan et al., 2010). In particular, it has been shown that xenografts of SW480 human colon cancer cells in nude mice can be completely regressed by anti-RalBP1 immunoglubulin G or by suppression of RalBP1 expression using phosphorothioate antisense against it (Singhal et al., 2007). The super family of aldo-keto reductase (AKR) enzymes seems to be involved in tumor development, and growing evidence is accumulating, suggesting them as a new class of tumor marker. These enzymes are hydroxysteroid dehydrogenases with a broad substrate specificity for other carbonyl compounds including HNE. The isoform AKR1B10 seems to be particulary involved in the transformation of HNE to the oxidized counterpart 4-oxonon-2enal (4-ONE) (Martin et al., 2009). AKR1B10 is also up-regulated in many types of solid tumors (Fukumoto et al., 2005;Yoshitake et al., 2007;Breton et al., 2008;Satow et al., 2010), and its gene silencing results in growth inhibition of colorectal cancer cells (Yan et al., 2007), as well as in increasing HNE-elicited cell death (Matsunaga et al., 2011). Recently, some family members of AKR enzymes have been shown to be overexpressed and linked to resistance against anticancer drugs such as anthracyclines, cisplatin, and methotrexate (Veitch et al., 2009;Cheng al., 2008;Selga et al., 2008). As regarding colon cancer, experimental data suggest that the up-regulation of AKR1B10 was related with acquisition of resistance to the anticancer drug mitomycin-c (MMC) in HT-29 colon cancer cells (Matsunaga et al., 2011). The cytotoxic effects of MMC seems to be mediated by the formation of HNE. Thus, the biological significance of the increasing of AKR1B10 in MCC resistant cancer cells would be an ability to better detoxify cytotoxic aldehydes including HNE. (Matsunaga et al., 2011). In the resistant cells, treatment with an AKR1B10 inhibitor decreased their MMC tolerance (Matsunaga et al., 2011), suggesting its use as adjuvant therapy in drug resistant cells, in which AKR1B10 is over-expressed.
Many dietary cancer chemopreventive compounds, such as cruciferous vegetables, could activate the antioxidant responsive element (ARE), a critical regulatory element in the promoter sequence of genes encoding cellular Phase II detoxifying and antioxidant enzymes. Transcriptional activation of ARE is typically mediated by the transcription Nuclear factor-erythroid 2-related factor 2 (Nrf2). Thus, this transcription factor has emerged as a novel target for the prevention of colon cancer (Saw & Kong, 2011). However, stable RNAi-mediated knockdown of Nrf2 in human colon cancer cells suppressed tumor growth in mouse xenograft settings and colon tumor angiogenesis by inhibiting Hypoxia-Induced Activation of HIF-1a (Kim et al., 2011). Thus, the role of Nfr2 in colon carcinogenesis still has to be explored.

HNE and nutrition
It is well accepted that development and progression of colon cancer is generally associated with lifestyle-dependent risk factors, such as dietary choices (Pearson et al., 2009). HNE can be directly found in food (Gasc et al., 2007) or its production can be enhanced by the presence of some nutrients, i.e. ω-6 PUFAs, or some fermentation products of diet, i.e. butyrate, can modulate the metabolism of this aldehyde, thus modifying its concentration.
In this context, it is very interesting to explore the connection between HNE, nutrition and colon carcinogenesis. HNE has been founded in different foods, correlating with the amount of ω-6 (Surh et al., 2010). UsingGC-MS technology, scientists measured 4-hydroxy-alkenaks content in vegetable oils, fish and shellfish, calculating the HNE dietary intake of the Korean population (Surh et al., 2005). Korean daily exposure to 4-hydroxy-2-alkenals was found to be of 4.3 mg/day and HNE was found to be more represented (2.7 mg). There was an additional exposure to more than 11.8 mg/day 4-hydroxy-2-alkenal from fried foods. The combined exposure would be, therefore, 16.1 mg/day corresponding to 0.3 mg/kg body weight/day for a 60 kg Korean adult. Additionally, the screening of PUFA-fortified foods including infant formulas and baby foods commercially available on the Korean markets were screened, and it was estimated that 3-month to 1-year-old babies sticking exclusively to these products could be exposed to a maximum 20.2 μg/kg BW/ day of 4-hydroxy-2alkenals (Surh et al., 2007). However, in spite of the biological toxicity of 4-hydroxy-2alkenals, the risk for humans cannot be quantified due to the lack of a virtually safe dose of the compound (Surh et al., 2005). A diet high in red and processed meats can increase colon cancer risk by 12-20%. The mechanism of promotion by haem iron is not known, but may be linked to oxidative stress and subsequent events such as lipid pro-oxidation and HNE production (Sesink et al., 1999;Sawa et al., 1998). Indeed, the dietary haem, in the form of either haemoglobin or meat, promotes precancerous lesions, aberrant crypt foci (ACF) and mucin-depleted foci in the colon of rats (Pierre et al., 2003;Pierre et al., 2004). This haem-induced promotion was associated with increased lipid peroxidation in faecal water and strong cytotoxycity activity of faecal water on the cancerous colonic epithelial cell line (Pierre et al., 2003). Further, Pierre and collaborators (2007), have explored the effect of faecal water components of haem-fed rats, on normal APC +/+ or premalignant APC -/+ cells, demonstrating that the toxic effects observed correlated with the presence of HNE in the faeces. Moreover, the premalignat APC -/+ cells were more resistant to apoptosis with respect to normal APC +/+. The authors suggested, thus, that the premalignant mutation confer to cells the resistance to the inhibitory signal, allowing them to undergo further mutations and follow a tumoural pathway (Pierre et al., 2007). In a randomized human study, the urinary excretion of DHN-MA, the major metabolite of HNE detectable in urine was compared in volunteers consuming different levels of heme iron. The volunteers fed with a low red meat diet (60 g/day) showed a twofold increase of DHN-MA when supplemented with heme iron as blood sausage (70 g/day). Since colon preneoplastic lesions and DHN-MA excretion in the experimental animal were clearly associated with dietary heme iron, urinary DHN-MA was suggested as a promising biomarker of colon carcinogenesis (Pierre et al., 2006). The role of fat present in the diet in coloncarcinogenesis has been explored by several authors and comprehensive reviews have been published. In particular, diets rich in ω-6 PUFAs, contained in vegetable oils, seem to enhance the development of colon tumors, whereas ω-3 PUFA-containing diets, such as fish oil, reduce colon cancer incidence (Reddy, 2002;Kim & Milner, 2007). Thus, it is possible to suggest a putative HNE role in colon carcinogenesis, since HNE is derived from peroxidation of ω-6. However, the complexity of the issue forces us to be more cautious. Indeed, Eder and collaborators investigated the impact of different fatty-acids composition in the diet on cancer development, measuring the formation of the promutagenic HNE-dG in the mucosa of several organs, such as colon.
The correlation between adduct levels and the different fatty acids assumption was not uniform for all organs and they didn't find a clear relationship between fatty acids and adduct levels in the colon (Eder et al., 2008). Moreover, beside lipid peroxidation products it is necessary to consider the eicosanoids, also derived from PUFAs. Indeed, eicosanoids have different properties in cancer cell growth, invasion and angiogenesis when derived from ω-6 or ω-3 fatty acids (Berquim et al., 2008), thus suggesting a role in carcinogenesis. Epidemiological studies show a reduction in risk for individuals and populations consuming high amounts of vegetables. The protective effect of vegetables may be due to their content of complex carbohydrates such as dietary fiber and starch (Scheppach et al., 1999). A substantial amount of starch escapes digestion in the small intestine (Englyst et al., 1992) and this fraction is called enzyme-resistant starch (RS). Starch and dietary fiber together are the principal substrates controlling the pattern of fermentation in the colon and, thus, the metabolism of compounds, like bile acids, nitrate and enzyme activityes (bacterial and antioxidant enzymes), which have been implicated in carcinogenesis. The effect of enzyme-resistant starch (RS) on the development of colon cancer was reported to include both chemopreventive and tumorigenic activity in humans. Indeed, an inverse association between starch consumption and large bowel cancer incidence has been found in an international comparison in 12 populations worldwide (Cassidy et al., 1994). However, an increased cancer risk with high-starch intake has been also reported (Franceschi et al., 1998;Favero et al., 1999). Wacker and collaborators (2002) have studied the number of 1,N2propanodeoxyguanosine-30-monophosphate (HNE-dGp) adducts in the colonic mucosa of volunteers fed with starchy foods enriched with a highly resistant amylomaize starch (Hylon VII) and they found an increase of the HNE-dGp adduct, whereas there was no evidence for an increased cell proliferation in the upper crypt. Finally, as already mentioned, nutrients can modulate the HNE level in the colon, by affecting its metabolism. This is the case of fermented products of diet, such as butyrate (Knoll et al., 2005;Glei et al., 2006) and wheat bran-derived arabinoxylans, that can affect the HNE levels, by upregulating GSTs activities.

Conclusion
Lipid peroxidation is a physiological and pathological process that elicits a number of electrophilic compounds able to modulate several cellular processes. Among these, HNE is the most studied aldehyde, due to its high biological activity. Since HNE is a normal constituent of the diet or can be produced in the gut, colon cells can be exposed to this aldehyde. Low doses of HNE are able to inhibit cell proliferation and induce differentiation of colon cancer cells. Conversely, a high concentration of HNE exhibits genotoxic and mutagenic activity. We believe that the concentration of HNE and other lipid peroxidation products in the colon, represent a steady state level between production and catabolism. The alteration of this equilibrium elicits a stress condition for colon cells and, possibly, could be involved in colon carcinogenesis, although there is no scientific consensus in supporting its protumoral action.
Results on HNE content in human biopsies of coloncancer tissues are contradictory, and the positive correlation between HNE content and cancer progression doesn't allow an assumption whether the HNE increase during the progression of colon cancer may represent a cause or a consequence of this process. However, in colon cancer cells, HNE induces apoptosis and telomerase inhibition. Thus, we can hypothesize that HNE, produced during radiotherapy or chemotherapy, can participate to the control of tumor growth and tumor cell death.

References
Acharya