Protective effects of dietary curcumin in mouse model of chemically induced colitis are strain dependent

Abstract

Background: Curcumin (diferulolylmethane) has been shown to have a protective role in mouse models of inflammatory bowel diseases (IBD) and to reduce the relapse rate in human ulcerative colitis (UC), thus making it a potentially viable supportive treatment option. Trinitrobenzene sulfonic acid (TNBS) colitis in NKT-deficient SJL/J mice has been described as Th1-mediated inflammation, whereas BALB/c mice are believed to exhibit a mixed Th1/Th2 response.

Methods: We therefore investigated the effect of dietary curcumin in colitis induced in these 2 strains.

Results: In the BALB/c mice, curcumin significantly increased survival, prevented weight loss, and normalized disease activity. In the SJL/J mice, curcumin demonstrated no protective effects. Genomewide microarray analysis of colonic gene expression was employed to define the differential effect of curcumin in these 2 strains. This analysis not only confirmed the disparate responses of the 2 strains to curcumin but also indicated different responses to TNBS. Curcumin inhibited proliferation of splenocytes from naive BALB/c mice but not SJL/J mice when nonspecifically stimulated in vitro with concanavalin A (ConA). Proliferation of CD4(+) splenocytes was inhibited in both strains, albeit with about a 2-fold higher IC(50) in SJL/J mice. Secretion of IL-4 and IL-5 by CD4(+) lymphocytes of BALB/c mice but not SJL/J mice was significantly augmented by ConA and reduced to control levels by curcumin.

Conclusions: The efficacy of dietary curcumin in TNBS colitis varies in BALB/c and SJL/J mouse strains. Although the exact mechanism underlying these differences is unclear, the results suggest that the therapeutic value of dietary curcumin may differ depending on the nature of immune dysregulation in IBD.

FIGURE 1

Purity and stability of curcumin used in the study. A: HPLC analysis of curcumin was performed as described in the Materials and Methods section. No contaminating curcuminoids (demetoxycurcumin, or bis-demetoxycurcumin), commonly detected as contaminants in commercial curcumin, were detectable in the preparation. B: Mass spectrometry (MS) analysis of curcumin. Its melting point, elemental analysis, molecular weight by MS, and H NMR were in accord with the structure. C: Chromatographic analysis of curcumin stability in the mouse diet did not indicate any degradation over a 24-hour period. A slight increase in the amount of extracted curcumin was observed between times 0 and 9 hours, possibly because of partitioning of hydrophobic curcumin into the lipid phase of the diet, which was more accessible to the organic solvents used in the extraction procedure.

FIGURE 2

A: Survival of TNBS-treated BALB/c and SJL/J mice fed control diet (dotted line) and 2% curcumin–supplemented diet (solid line). Dietary curcumin significantly improved the survival of colitic BALB/c mice but not SJL/J mice. B: Body weight loss and recovery in control and TNBS-treated BALB/c and SJL/J mice fed the control or the 2% curcumin–supplemented diet. The curcumin diet prevented body weight loss in BALB/c mice but not SJL/J mice.

FIGURE 3

Histological analysis (H&E staining) of colonic morphology in control and TNBS-treated BALB/c and SJL/J mice fed the control or the 2% curcumin–supplemented diet. A–C: BALB/c mice; D–F: SJL/J mice. G: Summary of histological damage scoring expressed as the sum of the scores from the proximal and distal segments of the mouse colon. Consistent with the effects on mortality and body weight loss, curcumin treatment resulted in statistically significant improvement of the histology scores of BALB/c mice but not SJL/J mice.

FIGURE 4

Venn diagram depicting a difference in the colonic gene expression pattern in BALB/c and SJL/J mice in response to TNBS. The number of genes/probe sets up- or down-regulated at least 1.5-fold (P < 0.05, Student t test) in the BALB/c mice (blue and gray overlap) and the SJL/J mice (green and gray overlap) indicated a significantly different response to the same haptenizing agent, with a relatively small number of genes dysregulated in both strains.

FIGURE 5

A: Genes/probe sets dysregulated by TNBS colitis (≥1.5-fold, P < 0.05) in the BALB/c mice (965 genes) or the SJL/J mice (800 genes) were subjected to 1-way ANOVA followed by the Student-Newman-Keuls (SNK) post hoc test. Four hundred and sixty-nine genes/probe sets in the BALB/c mice and 211 in the SJL/J mice passed ANOVA. Of these, 428 genes were normalized by curcumin treatment of the BALB/c mice (SNK, P < 0.05, TNBS versus TNBS/curcumin), but only 41 by curcumin treatment of the SJL/J mice. To demonstrate dramatic difference in the effect of curcumin in the 2 mouse strains, 428 genes from the SNK analysis of the BALB/c mice were plotted with their respective expression values from the BALB/c mice (B) and the SJL/J mice (C).

FIGURE 6

Gene ontology analysis using the DAVID Functional Annotation Tool (http://david.abcc.ncifcrf.gov/) of the 116 gene/probe sets (97 nonredundant well-characterized genes) whose expression was normalized by curcumin in BALB/c mice but not SJL/J mice based on post hoc analyses. Genes categorized based on biological process were grouped and ranked (threshold of 4, P < 0.05). Categories were sorted according to the EASE score, a modified Fisher exact P value.

FIGURE 7

Real-time RT-PCR analysis of colonic expression of TNFα mRNA in BALB/c and SJL/J mice. The results were analyzed with TATA box–binding protein as an internal control using the ΔΔCt method, with control BALB/c mice (open bar) used as a calibrator. Statistical analysis was performed with ANOVA (P < 0.0007) followed by the Fisher PLSD post hoc test. Different letters next to bars indicate statistical differences (Fisher PLSD test, P < 0.05).

FIGURE 8

Effects of low, noncytotoxic concentrations of curcumin on ConA-induced proliferation of splenocytes (A,B) or magnetically selected CD4⁺ lymphocytes (C,D) obtained from naive BALB/c (A,C) or SJL/J mice (B,D). Dose-dependent inhibition of splenocyte proliferation was only observed in BALB/c mice, whereas the calculated IC₅₀ of curcumin for the inhibition of CD4+ cell proliferation was significantly higher in SJL/J mice (P < 0.05; Student t test). All data were derived from 5 independent experiments with a minimum of 4 repetitions per treatment (^*statistical difference between control and ConA-treated cells; ^#statistical difference between ConA and ConA+Curcumin treatment groups).