Dietary Crocin Inhibits Colitis and Colitis-Associated Colorectal Carcinogenesis in Male ICR Mice

Abstract

A natural carotenoid crocin is contained in saffron and gardenia flowers (crocuses and gardenias) and is used as a food colorant. This study reports the potential inhibitory effects of crocin against inflammation-associated mouse colon carcinogenesis and chemically induced colitis in male ICR mice. In the first experiment, dietary crocin significantly inhibited the development of colonic adenocarcinomas induced by azoxymethane (AOM) and dextran sodium sulfate (DSS) in mice by week 18. Crocin feeding also suppressed the proliferation and immunohistochemical expression of nuclear factor- (NF-) κB but increased the NF-E2-related factor 2 (Nrf2) expression, in adenocarcinoma cells. In the second experiment, dietary feeding with crocin for 4 weeks was able to inhibit DSS-induced colitis and decrease the mRNA expression of tumor necrosis factor α, interleukin- (IL-) 1β, IL-6, interferon γ, NF-κB, cyclooxygenase-2, and inducible nitric oxide synthase in the colorectal mucosa and increased the Nrf2 mRNA expression. Our results suggest that dietary crocin suppresses chemically induced colitis and colitis-related colon carcinogenesis in mice, at least partly by inhibiting inflammation and the mRNA expression of certain proinflammatory cytokines and inducible inflammatory enzymes. Therefore, crocin is a candidate for the prevention of colitis and inflammation-associated colon carcinogenesis.

Figure 1

Structures of the principle constituents (crocetin, crocetin-diglycoside, crocetin-triglycoside, crocin, picrocrocin, and safranal) of saffron.

Figure 2

Experimental protocols for (a) Experiment 1 (18-week study) and (b) Experiment 2 (four-week study).

Figure 3

Representative histopathology of colonic proliferative lesions that developed in mice that received AOM and 1.5% DSS (Experiment 1). (a) Dysplastic crypts, high grade (bar, 30 μm); (b) tubular adenomas (bar, 60 μm); (c) tubular adenocarcinoma (bar, 200 μm). H&E stain.

Figure 4

(a) The incidence of severe colorectal inflammation, (b) the inflammation score of colorectum, (c) the incidence of high-grade dysplastic crypts (DYS), and (d) the multiplicity (no./colon) of high-grade DYS. *P < 0.05, **P < 0.01 versus the the AOM + 1.5% DSS group.

Figure 5

(a) The incidence of colorectal adenoma (AD), (b) multiplicity (no./colon) of colorectal AD, (c) incidence of colorectal adenocarcinoma (ADC), and (d) multiplicity (no./colon) of colorectal ADC. *P < 0.05, **P < 0.01, ***P < 0.001 versus the AOM + 1.5% DSS group.

Figure 6

Immunohistochemical staining for MCM2 in an adenocarcinoma that developed in a mouse from (a) group 1 (AOM + 1.5% DSS), (b) group 2 (AOM + 1.5% DSS + 50 ppm crocin), (c) group 3 (AOM + 1.5% DSS + 100 ppm crocin), and group 4 (AOM + 1.5% DSS + 200 ppm crocin). The insert in (a) is normal colonic mucosa. Bars, 100 μm. The graph summarizes the data on the MCM2-positive rates of adenocarcinomas from groups 1 through 4 (n = 5 each). *P < 0.001 versus the AOM + 1.5% DSS group.

Figure 7

Immunohistochemical expression of (a) NF-κB and (b) Nrf2 in adenocarcinoma cells (Experiment 1). Both proteins were expressed in the nuclei of cancer cells. The scores of immunohistochemical expression of both proteins were changed by crocin treatment: crocin feeding lowered the score for NF-κB (A) and increased it for Nrf2 (B). *P < 0.05, **P < 0.01 versus the AOM + 1.5% DSS group.

Figure 8

Representative histopathology of the colorectal mucosa (Experiment 2). When compared to (a) normal colorectal mucosa, 1.5% DSS treatment resulted in severe colitis with mucosal ulceration (b). In contrast, mucosal regeneration was observed in the colons of mice that were treated with crocin at (c) 100 ppm and (d) 200 ppm. Bars are 20 μm. H&E stain.

Figure 9

The inflammation scores in the colorectum of mice treated with DSS and or crocin (Experiment 2). Feeding with crocin at all three concentrations (50, 100, and 200 ppm) significantly decreased the inflammation score. *P < 0.05, **P < 0.01, ***P < 0.001 versus the AOM + 1.5% DSS group.

Figure 10

The mRNA expression levels of inducible inflammatory enzymes, (a) COX-2 and (b) iNOS, in the colorectum (Experiment 2) as determined by quantitative real-time RT-PCR. Crocin treatment significantly decreased the expression levels of COX-2 (50, 100, and 200 ppm) and iNOS (200 ppm), when compared with the AOM and DSS group. The expression was normalized to the β-actin mRNA expression. Samples were analyzed in triplicate. Data are the means ± SD from three independent assays (n = 5 from each group). The ordinates show the relative mRNA expression (/β-actin) versus the 1.5% DSS group. *P < 0.01, **P < 0.001 versus the 1.5% DSS group.

Figure 11

The mRNA expression levels of (a) IFN-γ, (b) TNF-α, (c) IL-1β, (d) IL-6, (e) NF-κB, and (f) Nrf2 in the colorectum (Experiment 2) as determined by quantitative real-time RT-PCR. Feeding with crocin significantly decreased the expression levels of IFN-γ (100 and 200 ppm), TNF-α (100 and 200 ppm), IL-1β (100 and 200 ppm), IL-6 (100 and 200 ppm), and NF-κB (100 and 200 ppm), compared with the AOM and DSS group. On the other hand, the mRNA expression of Nrf2 was significantly increased by the treatment with crocin (100 and 200 ppm). The expression was normalized to the β-actin mRNA expression. Samples were analyzed in triplicate. Data are the means ± SD from three independent assays (n = 5 from each treatment group). The ordinates are the relative mRNA expression levels (/β-actin) versus the 1.5% DSS group. *P < 0.05, **P < 0.01, ***P < 0.001 versus the 1.5% DSS group.