Different Susceptibilities between Apoe- and Ldlr-Deficient Mice to Inflammation-Associated Colorectal Carcinogenesis

Hypercholesterolemia resulting in atherosclerosis is associated with an increased risk of ischemic heart disease and colorectal cancer (CRC). However, the roles of apoliprotein (Apo) E (Apoe) and low-density lipoprotein (Ldl) receptor (Ldlr) in colorectal carcinogenesis have not yet been investigated. In this study, we examined the susceptibility of Apoe-deficient and Ldlr-deficient mice, which are genetic animal models of atherosclerosis to azoxymethane (AOM)/dextran sodium sulfate (DSS)-induced colorectal carcinogenesis. In Experiment 1, male Apoe-deficient (n = 20) and wild type (WT) mice (C57BL/6J, n = 21) were treated with a single intraperitoneal (i.p.) injection of AOM (10 mg/kg body weight) and then given 1.5% DSS in drinking water for seven days. They were maintained up to week 20 and sacrificed for the histopathological examination of colorectal tumors. The mRNA expression of cyclooxygenase (Cox)-2, inducible nitric oxide synthase (Nos2), tumor necrosis factor (Tnf)-α interleukin (Il)-1β, and Il-6 was assayed in the colorectal mucosa. In Experiment 2, male Ldlr-deficient (n = 14) and WT mice (C57BL/6J, n = 10) were given a single i.p. injection of AOM (10 mg/kg body weight) and then given 2% DSS in drinking water for seven days. They were sacrificed at week 20 to evaluate their colorectum histopathologically. In Experiment 1, the multiplicity of CRCs was significantly higher in the Apoe-deficient mice (2.75 ± 1.48) than in the WT mice (0.62 ± 0.67). The serum lipoprotein levels in the Apoe-deficient mice were also significantly higher than in the WT mice. In Experiment 2, the incidence (29%) and multiplicity (0.50 ± 0.94) of CRCs in the Ldlr mice were significantly lower than in the WT mice (80% incidence and 3.10 ± 2.38 multiplicity). The mRNA expression of two inducible enzymes and certain pro-inflammatory cytokines in the colorectum of each genotype was greater than in the respective WT mice. The values in the Apoe-deficient mice were much greater than in the Ldlr mice. These findings suggest that Apoe-deficient mice showed increased susceptibility to inflammation-associated colorectal carcinogenesis due to their high reactivity to inflammatory stimuli.

Certain cancers, including CRC and human cancer cell lines, have increased levels in Ldlr protein [47,48]. A loss of feedback regulation of Ldlr in CRCs was also reported [49]. In addition, Cox-2 was up-regulated in colorectal neoplasms that over-expressed Ldlr mRNA compared with normal colorectal mucosa. These findings may suggest that Ldlr is abnormally regulated in tumors and may play a certain role in the up-regulation of Cox-2 in neoplasms.
Alterations in the plasma lipid profiles and in intracellular cholesterol homoeostasis were reported in various malignancies, including CRCs. However, the significance of these alterations, if any, in colorectal carcinogenesis and cancer biology is not clear. In the current study, we examined whether or not Apoe and Ldlr were involved in colorectal carcinogenesis in mice. For this, we used an inflammation-associated colorectal carcinogenesis model of Apoeor Ldlr-deficient mice developed with azoxymethane (AOM) and promoted by dextran sodium sulfate (DSS), where many colorectal neoplasms develop within a short period of time [50].

AOM/DSS-Induced Colorectal Carcinogenesis in the Apoe-Deficient Mice (Experiment 1)
Both the Apoe-deficient mice and WT mice tolerated treatment with AOM and 1.5% DSS well and survived to week 20 ( Figure 1a). As listed in Table 1, at sacrifice, the body (p < 0.05) and liver weights (p < 0.001) of the Apoe-deficient mice were significantly greater than those of the WT mice. The mean relative liver weights of both groups were comparable. When given AOM and DSS, the colon length of the Apoe-deficient mice was slightly shorter than that of the wild type of mice (Table 1). Treatment with AOM followed by 1.5% DSS resulted in the development of colorectal tumors in both genotypes (Figure 2a). The incidences and multiplicities of several colorectal lesions, such as mucosal ulcer, adenoma (AD), and AD + adenocarcinoma (ADC), were larger in the Apoe-deficient mice than in the WT mice, but the differences between the two genotypes were not significant ( Table 2). In addition, the multiplicities of dysplastic lesions (DYS) and ADC in the Apoe-deficient mice were also larger than in the WT mice, and these differences were statistically significant (dysplastic lesions, p < 0.02; and ADC, p < 0.005, Table 2). The mean volume (1150.2 ± 396.7 mm 3 ) of colorectal tumors in the Apoe-deficient mice was significantly greater than that (597.9 ± 234.6 mm 3 ) in the WT mice (p < 0.05), as shown in Figure 3a. Histopathologically, three types (well, moderately, and poorly) of differentiation were observed in ADCs (Figure 4a). Poorly differentiated adenocarcinomas developed in a few Apoe-deficient mice that received AOM and DSS (Figure 4b). The incidence and multiplicity of mucosal ulcer in the Apoe-deficient mice were higher than in the WT mice, but the differences were not statistically significant (Table 2). 234.6 mm ) in the WT mice (p < 0.05), as shown in Figure 3a. Histopathologically, three types (well, moderately, and poorly) of differentiation were observed in ADCs (Figure 4a). Poorly differentiated adenocarcinomas developed in a few Apoe-deficient mice that received AOM and DSS (Figure 4b).
The incidence and multiplicity of mucosal ulcer in the Apoe-deficient mice were higher than in the WT mice, but the differences were not statistically significant (Table 2).           WT, wild type. a Mean ± standard deviation (SD); b−e Significantly different from the respective WT mice ( b p < 0.02, c p < 0.001, d p < 0.005, and e p < 0.05).   With regard to the serum biochemistry (Table 3), the cholesterol (p < 0.001), very-low-density lipoprotein (VLDL, p < 0.001), and low-density lipoprotein (LDL, p < 0.001) levels in the Apoe-deficient mice were significantly higher than in the WT mice. However, the serum TG, high-density lipoprotein (HDL), glucose, and adiponectin levels were comparable between the genotypes.  With regard to the serum biochemistry (Table 3), the cholesterol (p < 0.001), very-low-density lipoprotein (VLDL, p < 0.001), and low-density lipoprotein (LDL, p < 0.001) levels in the Apoe-deficient mice were significantly higher than in the WT mice. However, the serum TG, high-density lipoprotein (HDL), glucose, and adiponectin levels were comparable between the genotypes.   The expression of all five molecules was significantly higher in the Apoe-deficient mice than in the WT mice (p < 0.001). The expression of the molecules was slightly but not significantly higher in the Ldlr-deficient mice than in the WT mice. The expression was normalized to the β-actin mRNA expression level. The data are represented as the means ± SD from three independent assays (n = 5 from each group). Y-axis shows expression of the mRNA relative to the "Standard condition" and normalized to β-actin. The expression of all five molecules was significantly higher in the Apoe-deficient mice than in the WT mice (p < 0.001). The expression of the molecules was slightly but not significantly higher in the Ldlr-deficient mice than in the WT mice. The expression was normalized to the β-actin mRNA expression level. The data are represented as the means ± SD from three independent assays (n = 5 from each group). Y-axis shows expression of the mRNA relative to the "Standard condition" and normalized to β-actin.

AOM/DSS-Induced Colorectal Carcinogenesis in the Ldlr-Deficient Mice (Experiment 2)
The Ldlr-deficient mice and WT mice tolerated treatment with AOM and 2% DSS well and survived to week 20 (Figure 1b). At sacrifice, the body, liver, and relative liver weights of the Ldlr-deficient mice and WT mice were almost comparable ( Table 1). The colon length of the Ldlr-deficient mice was slightly smaller than that of the wild type of mice (Table 1). Treatment with AOM followed by 2% DSS resulted in the development of colorectal tumors in both genotypes (Figure 2b). The incidences (p < 0.02 or p < 0.05) and multiplicities (p < 0.001) of several proliferative colorectal lesions, including DYS, AD, ADC, and AD + ADC, in the Ldlr-deficient mice were significantly lower than in the WT mice ( Table 2). The mean volume (569.8 ± 417.6 mm 3 ) of tumors in the Ldlr-deficient mice was smaller than that in the WT mice (966.2 ± 1362.8 mm 3 ), but the difference was not statistically significant (Figure 3b). Well-and moderately differentiated ADCs (Figure 4a) developed in the colorectum, but there were no poorly differentiated types (Figure 4c). The incidence and multiplicity of mucosal ulcer in the Ldlr-deficient mice were lower than in the WT mice, but the differences were not statistically significant ( Table 2).
The serum levels of cholesterol (p < 0.001), TG (p < 0.001), LDL (p < 0.001), and adiponectin (p < 0.01) in the Ldlr-deficient mice were significantly higher than in the WT mice (Table 3). However, the serum VLDL, HDL, and glucose levels were comparable between the genotypes.
As illustrated in Figure 5, the mRNA expression of Cox-2, Nos2, Tnf-α, Il-1β, and Il-6 in the non-lesional colorectal mucosa of the Ldlr-deficient mice was slightly elevated when compared to the WT mice.

Discussion
This is the first study, to our knowledge, to examine the susceptibility of inflammation-associated colorectal carcinogenesis in Apoeand Ldlr-deficient male mice in comparison with their background genotype using our AOM/DSS model [50]. Our findings here suggest that Apoe and Ldlr inversely affect inflammation-associated colorectal carcinogenesis, irrespective of their serum lipoprotein profiles. Surprisingly, the Apoe-deficient mice were much more susceptible to AOM/DSS-induced colorectal carcinogenesis than the WT mice. The mRNA expression levels of two inducible enzymes, Cox-2 and Nos2, and three pro-inflammatory cytokines, Tnf-α, Il-1β, and Il-6, in the colorectum of Apoe-deficient mice were much higher than in the Ldlr-deficient mice.
When given AOM and DSS, the colon length of the Apoe-deficient mice or Ldlr-deficient mice was shorter than their respective wild mice (Table 1) without statistical significance. This may be related to the findings of severe colitis and enhancement of colorectal carcinogenesis in the Apoe-deficient mice or Ldlr-deficient mice. Histopathologic investigation revealed that poorly differentiated ADCs were developed only in the Apoe-deficient mice that received AOM and DSS, but not in the Ldlr-deficient mice and their respective wild mice (Figure 4b,c). Severe inflammation and increased levels of Cox-2, Nos2, Tnf-α, Il-1β, and Il-6 observed in the colorectum of Apoe-deficient mice treated with AOM and DSS may be related to these differences of histopathological findings of ADCs.
The Apoeand Ldlr-deficient mice, which have elevated serum levels of cholesterol, TG, VLDL, LDL, and/or HDL, are frequently used for research of atherosclerosis and in developing new drugs against atherosclerosis [51,52]. In contrast to the Ldlr-deficient mice, the Apoe-deficient mice develop atherosclerosis spontaneously without an atherogenic diet. The lipid profiles of these mice differ slightly. For example, hypercholesterolemia in the Apoe-deficient mice is more severe than in the Ldlr-deficient mice. Compared to the WT mice, the serum VLDL level of the Apoe-deficient mice is markedly increased (five-fold that of the WT mice [51]), but the elevation in the VLDL level of the Ldlr-deficient mice is moderate. While the serum HDL levels of the Apoe-deficient mice are decreased (45% of that in the WT mice [51]), the levels in the Ldlr-deficient mice are modestly increased. The TG level in the Apoe-deficient mice was found to be 68% higher than that in the WT mice [51]. The different susceptibilities of the two genotypes to AOM/DSS-induced colorectal carcinogenesis may be due to their differing lipid profiles. However, the precise reason for the differences in susceptibility observed in this study is unclear.
Apoe is a major modulator of lipoprotein metabolism, and the allele-specific effects of Apoe on lipoprotein metabolism were reported [53]. Apoe also has other crucial functions, and aberration of these functions may lead to carcinogenesis [54][55][56][57]. In vitro studies suggest that treatment of the colon cancer cell line, HT29, with Apoe increased the cell polarity by translocating β-catenin from the cytoplasm to cell-cell adhesion sites [56]. Apoe is able to inhibit cell proliferation and de novo DNA biosynthesis [57]. Functional experiments on Apoe isoforms showed that Apoe4, but not wild-type Apoe, inhibits glycogen synthase kinase (GSK)-3β and increases the amount of active protein kinase B (PKB), which further inactivates GSK-3β, leading to enhancement of β-catenin translocation into nuclei [54]. Nuclear β-catenin can promote transcription of genes involved in cell survival and division [58][59][60]. Since patients with UC show impaired lipoprotein metabolism [61], the effect of Apoe polymorphism on the risk of UC development should be investigated.
In the present study, we assayed the mRNA expression of Cox-2, Nos2, TNF-α, Il-1β, and Il-6 in the colorectal mucosa of the Apoe-deficient, Ldlr-deficient, and their respective WT mice. The values of the Apoe-deficient mice treated with AOM and DSS were significantly and much greater than the WT mice that received AOM and DSS, suggesting that the Apoe-deficient mice have hyper-inflammation status. Sensitivity to inflammatory stimuli was reported to be greater in Apoe-deficient mice [62] than in Ldlr-deficient mice [63]. This was related to a decrease in the production of certain pro-inflammatory cytokines in the Ldlr-deficient mice, although macrophages maintained an elevated cytokine production capacity [62,63]. Apoe involved in cholesterol and lipid metabolism has also altered both innate and adaptive immune responses [64]. Importantly, mice lacking Apoe exhibit increased inflammatory responses and higher mortality following lipopolysaccharide challenge [65]. This may suggest that Apoe has anti-inflammatory effects. Genetic factors have been reported to contribute to the pathophysiology of IBD [66]. Apoe inhibits the production of T lymphocytes and regulates immune reactions by interacting with several cytokines [67,68]. Apoe thus plays a key role in regulating the immune response in various autoimmune diseases [69].
Pro-inflammatory cytokines are central mediators of the chronic inflammatory process in several tissues. IL-6 is part of a central pathway in the pathogenesis of chronic inflammation diseases, such as IBD [70], and inflammation-associated colorectal cancer [71][72][73]. IL-6 trans-signaling is also an independent risk factor for coronary artery disease and is involved in inflammatory processes in vessels [74]. Other cytokines (Tnf-α, IL-1β, Interferon-γ), inflammatory enzymes (Cox-2, Nos2), nuclear factor (NF)-κB, and signal transducer and activator of transcription 3 (Stat3) are also involved in inflammation-associated colorectal carcinogenesis [72]. Uncontrolled activation of Nf-κB, Stat3, and Wnt-β-catenin signaling pathways enhances the aberrant proliferation of crypal cells in the sustained inflammatory microenvironment and promotes CRC development [75]. Although we did not assay Stat3 and Nf-κB in the present study, the mRNA expression of Tnf-α, Il-1β, Il-6, Cox-2, and Nos2 was assayed in the colorectum of Apoe-deficient, Ldlr-deficient, and their respective WT mice treated with AOM and DSS. The expression of all molecules was greater in the Apoe-deficient mice than that of the WT mice. The expression of Nos2, Tnf-α, and Il-1β in the Ldlr-deficient mice was slightly higher than that of the WT mice. The Apoe-deficient mice showed even more elevation of these molecules than the Ldlr-deficient mice. Thus, the Apoe-deficient mice were in a higher-inflammation status compared to the Ldlr-deficient mice. Further studies should investigate colorectal carcinogenesis in Apoe/Ldlr double-knockout mice.

Animals, Chemicals, and Diets
Twenty male five-week-old Apoe −/− mice (C57BL/6J background, Jackson Laboratories, Bar Harbor, ME, USA), 14 male five-week-old Ldlr −/− mice (C57BL/6J background, Jackson Laboratories), and 31 male C57BL/6J mice were used in this study. Two experiments were conducted. The experiment for AOM/DSS-induced colorectal carcinogenesis in the Apoe-deficient mice (Experiment 1) was conducted using Apoe −/− (n = 20) and C57BL/6J mice (n = 21), and contained two experimental groups that were treated with AOM and 1.5% DSS in their drinking water. Similarly, the experiment for AOM/DSS-induced colorectal carcinogenesis in the Ldlr-deficient mice (Experiment 2) was conducted using Ldlr −/− (n = 14) and C57BL/6J mice (n = 10), and contained two experimental groups that were given AOM and 2% DSS in their drinking water. A colorectal carcinogen AOM was obtained from Sigma-Aldrich Chemical (St. Louis, MO, USA), and DSS with a molecular weight of 36,000-50,000 Da (Lot No. 6046H) was purchased from MP Biomedicals (Aurora, OH, USA). AOM was diluted in physiological saline just before injection. DSS for colitis induction was dissolved in water at 1.5% (w/v) or 2% (w/v). A pelleted Charles River Formula (CRF)-1 diet (Oriental Yeast, Tokyo, Japan) was used as the basal diet throughout the study. All of the mice were maintained in the animal facility of the University, according to the Institutional Animal Care Guidelines. The animals were housed in plastic cages (three to five mice/cage) with free access to tap water and CRF-1 (Oriental Yeast Co., Ltd., Tokyo, Japan), under controlled conditions of humidity (50% ± 10%), light (12/12 h high/dark cycle), and temperature (23 ± 2 • C). Study designs were approved by the University and animal handling and procedures were performed in accordance with the Institutional Animal Care Guidelines.

Study Design
In the AOM/DSS-induced colorectal carcinogenesis in the Apoe-deficient mice (Experiment 1), twenty five-week old Apoe-deficient male mice and twenty-one, five-week old C57BL/6J male mice were given a single intraperitoneal (i.p.) injection of AOM (10 mg/kg body weight), as shown in Figure 6a. One week after the injection, they received 1.5% DSS in their drinking water for seven days and then were maintained on a basal diet and tap water for 18 weeks. In the AOM/DSS-induced colorectal carcinogenesis in the Ldlr-deficient mice (Experiment 2), fourteen five-week-old Ldlr-deficient male mice and ten five-week-old C57BL/6J male mice were given a single i.p. injection of AOM (10 mg/kg body weight), as shown in Figure 6b. Similar to Experiment 1, they received 2% DSS in their drinking water for seven days and then were maintained on a basal diet and tap water for 18 weeks.
In both experiments, all of the animals were killed by an overdose of ether at week 20 to evaluate the colorectal lesions histopathologically. for AOM/DSS-induced colorectal carcinogenesis in the Ldlr-deficient mice (Experiment 2) was conducted using Ldlr −/− (n = 14) and C57BL/6J mice (n = 10), and contained two experimental groups that were given AOM and 2% DSS in their drinking water. A colorectal carcinogen AOM was obtained from Sigma-Aldrich Chemical (St. Louis, MO, USA), and DSS with a molecular weight of 36,000-50,000 Da (Lot No. 6046H) was purchased from MP Biomedicals (Aurora, OH, USA). AOM was diluted in physiological saline just before injection. DSS for colitis induction was dissolved in water at 1.5% (w/v) or 2% (w/v). A pelleted Charles River Formula (CRF)-1 diet (Oriental Yeast, Tokyo, Japan) was used as the basal diet throughout the study. All of the mice were maintained in the animal facility of the University, according to the Institutional Animal Care Guidelines. The animals were housed in plastic cages (three to five mice/cage) with free access to tap water and CRF-1 (Oriental Yeast Co., Ltd., Tokyo, Japan), under controlled conditions of humidity (50% ± 10%), light (12/12 h high/dark cycle), and temperature (23 ± 2 °C). Study designs were approved by the University and animal handling and procedures were performed in accordance with the Institutional Animal Care Guidelines.

Study Design
In the AOM/DSS-induced colorectal carcinogenesis in the Apoe-deficient mice (Experiment 1), twenty five-week old Apoe-deficient male mice and twenty-one, five-week old C57BL/6J male mice were given a single intraperitoneal (i.p.) injection of AOM (10 mg/kg body weight), as shown in Figure 6a. One week after the injection, they received 1.5% DSS in their drinking water for seven days and then were maintained on a basal diet and tap water for 18 weeks. In the AOM/DSS-induced colorectal carcinogenesis in the Ldlr-deficient mice (Experiment 2), fourteen five-week-old Ldlr-deficient male mice and ten five-week-old C57BL/6J male mice were given a single i.p. injection of AOM (10 mg/kg body weight), as shown in Figure 6b. Similar to Experiment 1, they received 2% DSS in their drinking water for seven days and then were maintained on a basal diet and tap water for 18 weeks. In both experiments, all of the animals were killed by an overdose of ether at week 20 to evaluate the colorectal lesions histopathologically. At sacrifice, complete necropsy was done on all mice. The body and liver weights were measured and processed for a histopathological examination by conventional methods. The colon was flushed with normal saline and then removed. After measuring the length, the colon was cut open longitudinally along the main axis and gently washed with normal saline. The colonrectum was macroscopically inspected for the presence of lesions and fixed in 10% buffered formalin for at least 24 h. A histopathological examination was conducted on paraffin-embedded sections after hematoxylin and eosin (H & E) staining.

Real-Time Quantitative Polymerase Chain Reaction (RT-PCR)
We determined the mRNA expression in the non-lesional colorectal mucosa from five mice of each genotype, Apoe-deficient mice, Ldlr-deficient mice, or their respective wild type of mice.
RNA was extracted from the olorectum and stored at −80 • C using TRIzol reagent (Thermo Fisher Scientific, K.K., Yokohama, Japan) according to the manufacturer's protocol. RNA concentration and quality were verified, and reversely transcribed to produce cDNA. Quantitative RT-PCR analyses of Cox-2, Nos2, Tnf-α, Il-1β, and Il-6 were performed with ABI Prism 7500 (Applied Biosystems Japan Ltd., Tokyo, Japan) using TaqMan Gene Expression Assays (Applied Biosystems Japan Ltd., Tokyo, Japan): Cox-2 (Ptgs2), Mm00478374-ml; Nos2 (Mm00440485-ml); Tnf-α, Mm00443258-m1; Il-1β, Mm00434228-m1; and IL-6, Mm00446190-ml. β-Actin (Mm00607939-sl) was used to normalize the expression level of the mRNA genes. The cycling protocol of RT-PCR was conducted at a DNA denaturation temperature of 95 • C for 5 min and followed by 40 cycles of 95 • C for 15 s, 60 • C for 20 s, and an elongation temperature 72 • C for 40 s. Each experiment was performed in triplicate, and data were calculated by ∆∆C t methods.

Clinical Chemistry (Serum Lipid Profiles)
At sacrifice, blood samples were collected to measure the serum concentrations of total cholesterol, TG, very-low-density lipoprotein (VLDL), LDL, high-density lipoprotein (HDL), glucose, and adiponectin after overnight fasting from 10 or 14 mice in each group. Whole blood anti-coagulated with heparin lithium was taken from the inferior vena cava with a sterile syringe (Terumo, Tokyo, Japan). The serum was obtained by centrifugation (3000 rpm for 10 min), and stored at −80 • C until measurement. The serum TG and total cholesterol levels were determined using commercial enzymatic assay kits (TG, L-Type WAKO-TG·H; and total cholesterol, L-Type WAKO-CHO·H), obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). The serum levels of HDL, LDL, and VLDL were determined using an HDL and LDL/VLDL Cholesterol Quantitation Kit (BioVision, Inc., Milpitas, CA, USA). The serum glucose level was determined using commercial enzymatic assay kit (Glucose CII-test WAKO, Wako Pure Chemical Industries). These measurements were expressed as mg/dL. The serum adiponectin level (µg/mL) was determined with Mouse/Rat Adiponectin ELISA kits (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan).

Statistical Analysis
The incidences of colonic lesions between the groups were compared using the chi-square test or Fisher's exact probability test (GraphPad Instat version 3.05; GraphPad Software, San Diego, CA, USA). Other measures expressing mean ± standard deviation (SD) were statistically analyzed using one-way analysis of variance (ANOVA), followed by the Bonferroni or Tukey-Kramer multiple comparison post-test (GraphPad Instat version 3.05; GraphPad Software). mRNA expression was statistically analyzed by the Kruskal-Wallis test. Differences were considered statistically significant at p < 0.05.

Conclusions
Our findings indicate that Apoe and Ldlr are inversely involved in inflammation-associated colorectal carcinogenesis induced by AOM/DSS, irrespective of their serum lipoprotein profiles: the Apoe-deficient mice were much more susceptible to inflammation-associated colorectal carcinogenesis than the WT mice. The mRNA expression levels of two inducible enzymes and certain pro-inflammatory cytokines in the colorectum of Apoe-deficient mice were much more elevated than in the Ldlr-deficient mice.