Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Feb 14;114(7):E1273-E1281.
doi: 10.1073/pnas.1621400114. Epub 2017 Jan 30.

Creatine Maintains Intestinal Homeostasis and Protects Against Colitis

Affiliations
Free PMC article

Creatine Maintains Intestinal Homeostasis and Protects Against Colitis

Emre Turer et al. Proc Natl Acad Sci U S A. .
Free PMC article

Abstract

Creatine, a nitrogenous organic acid, replenishes cytoplasmic ATP at the expense of mitochondrial ATP via the phosphocreatine shuttle. Creatine levels are maintained by diet and endogenous synthesis from arginine and glycine. Glycine amidinotransferase (GATM) catalyzes the rate-limiting step of creatine biosynthesis: the transfer of an amidino group from arginine to glycine to form ornithine and guanidinoacetate. We screened 36,530 third-generation germline mutant mice derived from N-ethyl-N-nitrosourea-mutagenized grandsires for intestinal homeostasis abnormalities after oral administration of dextran sodium sulfate (DSS). Among 27 colitis susceptibility phenotypes identified and mapped, one was strongly correlated with a missense mutation in Gatm in a recessive model of inheritance, and causation was confirmed by CRISPR/Cas9 gene targeting. Supplementation of homozygous Gatm mutants with exogenous creatine ameliorated the colitis phenotype. CRISPR/Cas9-targeted (Gatmc/c ) mice displayed a normal peripheral immune response and immune cell homeostasis. However, the intestinal epithelium of the Gatmc/c mice displayed increased cell death and decreased proliferation during DSS treatment. In addition, Gatmc/c colonocytes showed increased metabolic stress in response to DSS with higher levels of phospho-AMPK and lower levels of phosphorylation of mammalian target of rapamycin (phospho-mTOR). These findings establish an in vivo requirement for rapid replenishment of cytoplasmic ATP within colonic epithelial cells in the maintenance of the mucosal barrier after injury.

Keywords: N-ethyl-N-nitrosourea; glycine amidinotransferase; inflammatory bowel disease; mTOR.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. 1.
Fig. 1.
Mapping of the mrbig mutation in Gatm. (A) Weight loss analysis of the mrbig pedigree in response to 1.5% DSS. REF, Gatm+/+ (n = 8); HET, Gatmmb/+ (n = 7); VAR, Gatmmb/mb (n = 6). Data points represent individual mice; mean and SD are indicated. (B) Manhattan plot showing −log10 P values (y axis) plotted against the chromosome positions of 77 mutations (x axis) identified in the G1 male mouse of pedigree R0046. Linkage between the mrbig phenotype and the Gatm mutation using a recessive model of inheritance. Horizontal pink and red lines represent thresholds of P = 0.05, and the threshold for P = 0.05 after applying Bonferroni correction, respectively. (C) DNA chromatogram trace showing the Gatm point mutation. (D) Protein domain diagram and (E) crystal structure (PDB ID code 1JDW) of GATM with the mrbig mutation indicated. Amidinotransf, glycine amidinotransferase domain; TP, transit peptide. (F) Creatine concentration in Gatmmb/mb (n = 3) and littermate control (Gatm+/+, n = 4) colonic epithelial cells. ***P < 0.0001. (G) Creatine supplementation restored the body weights in male Gatmmb/mb mice (n = 3; ***P < 0.0001). (H) GATM protein levels in colonic tissue from Gatm+/+ and Gatmmb/mb mice. Blot is representative of four individual experiments. (A, F, and G) P values were calculated by Student’s t test. (F and G) Bars indicate mean ± SD.
Fig. 2.
Fig. 2.
Mrbig mice exhibit characteristic symptoms of colitis. (A) Kaplan–Meier survival plot showing that Gatmmb/mb mice (n = 8) exhibit increased mortality after DSS compared with littermate controls [Gatm+/+ (n = 8) and Gatmmb/+ (n = 14)]. (B) Colons of mice exposed to 10-d DSS treatment exhibited shortening (n = 3 per genotype). Representative colons (Upper; after DSS treatment) and colon length (Lower). Bars indicate mean ± SD; **P < 0.01. (C) Histologic examination of Gatmmb/mb colons after 7 d of DSS treatment (10× magnification). (D) Intestinal permeability was determined by serum concentration 4 h after FITC–dextran gavage. Gatm+/+ (n = 3) and Gatmmb/mb (n = 4) littermates were analyzed after 4 d of DSS treatment. Bars indicate mean ± SD; *P < 0.05. (E) Supplementation of Gatmmb/mb mice with exogenous creatine [1% (wt/vol) in drinking water] during DSS treatment rescues the DSS-induced weight loss. Data points indicate mean ± SD; *P < 0.05, **P < 0.005 (vs. Gatmmb/mb group). Gatm+/+ (n = 3), Gatmmb/mb (n = 5), and Gatmmb/mb plus creatine supplementation (n = 6). (F) Colon length was restored in Gatmmb/mb mice after creatine supplementation. Gatm+/+ (n = 3), Gatmmb/mb without creatine supplementation (n = 5), and Gatmmb/mb with creatine supplementation (n = 6) littermates were analyzed on day 10 of treatment. Data points represent individual mice; mean and SD are indicated. ns, not significant; **P < 0.01. (G) Supplementation of Gatm+/+ mice with a high dose [1% (wt/vol)] of creatine during high-dose (3%) DSS treatment does not rescue weight loss. Data from mice treated with water alone (n = 4) or 1% creatine (n = 4) are shown. Data points indicate mean ± SD. (B, D, E, and F) P values were calculated by Student’s t test.
Fig. 3.
Fig. 3.
Validation that a mutation in Gatm causes the mrbig phenotype. (A) Weight loss analysis of the Gatmmb/c mice in response to 1.5% DSS. Gatm+/+ (n = 3), Gatmmb/+ (n = 4), Gatmc/+ (n = 5), and Gatmmb/c (n = 5) mice were analyzed on day 8 of treatment. ***P < 0.001. (B) Colons of Gatmmb/c mice exposed to DSS on day 8 of treatment exhibited shortening (n = 5). Gatm+/+ (n = 3) mice were also analyzed. **P < 0.01. (C) Stool scores in Gatmmb/c mice 7 d after administration of 1.5% DSS. Gatm+/+ (n = 3) and Gatmmb/c (n = 5) mice were analyzed. **P < 0.01. (D) Relative mRNA expression levels of inflammatory cytokines Cxcl2, Il6, and Tnf in the distal colon of Gatm+/+ (n = 3) and Gatmc/c (n = 3) mice after 6 d of 1.5% DSS. Experiment was repeated three times; *P < 0.05. (A–D) P values were calculated by Student’s t test; bars indicate mean ± SD.
Fig. S1.
Fig. S1.
Creatine content and colitis induced shortening of Gatmc/c colons. (A) Creatine concentration in Gatmc/c (n = 4) and littermate control (Gatm+/+, n = 4) colonic epithelial cells. P < 0.005. (B) Colons of Gatmc/c mice exposed to 1.4% DSS for 8 d exhibited shortening. Representative colons are shown for each genotype.
Fig. S2.
Fig. S2.
Colons of trinitrobenzenesulfonic acid (TNBS)-treated Gatmc/c mice. Colons of Gatmc/c displayed shortening postintrarectal administration of TNBS. Five-month-old male Gatm+/+ (n = 4), Gatmc/+ (n = 5), and Gatmc/c (n = 5) mice were administered 3 mg of TNBS intrarectally, and colons were harvested and measured 4 d posttreatment. *P < 0.05; determined by Student's t test.
Fig. 4.
Fig. 4.
The Gatmc/c mice exhibit normal peripheral immune homeostasis and responses. (A) Flow cytometry analysis of peripheral blood B cells (B220+), CD4+ T cells (CD3+ CD4+), CD8+ T cells (CD3+CD8+), macrophages (F480+), neutrophils (CD11b+F480) and NK cells (NK1.1+CD3) in Gatm+/+ (n = 6), Gatmc/+ (n = 9), and Gatmc/c (n = 5) mice. (B and C) Serum NP-specific IgM (B) and β-gal–specific IgG (C) measured by ELISA in Gatm+/+ (n = 6), Gatmc/+ (n = 9), and Gatmc/c (n = 5) mice immunized 5 or 14 d before with NP-Ficoll and rSFV–β-gal, respectively. Bars indicate mean ± SD. Differences between genotypes were not found to be statistically significant (P > 0.05).
Fig. 5.
Fig. 5.
Colonic epithelium of Gatmc/c mice exhibit reduced proliferation and increased cell death during DSS treatment. (A) Images of TUNEL staining (green) of colon sections from Gatm+/+ and Gatmc/c mice on day 4 of 1.5% DSS treatment (20× magnification); images are representative of three independent experiments. Quantification was performed on a minimum of 50 crypts for Gatm+/+ (n = 3) and Gatmc/c (n = 3) colons. *P < 0.05. (B) Metabolic labeling of DNA in Gatm+/+ and Gatmc/c colons on day 4 of 1.5% DSS treatment. EdU-labeled DNA was stained with Click-iT EdU labeling Alexa Fluor 647 (20× magnification); images are representative of at least three independent experiments. Quantification was performed on a minimum of 50 crypts for Gatm+/+ (n = 5) and Gatmc/c (n = 5) colons. *P < 0.05. (C and D) Colonic epithelium extracts were assessed by Western blot on day 4 of 1.5% DSS treatment for phosphorylated AMPK (C) and phosphorylated mTOR (D). Results are representative of more than three independent experiments with at least three mice per group. (A and B) P values were calculated by Student’s t test; bars indicate mean ± SD.
Fig. S3.
Fig. S3.
TUNEL staining of Gatmmb/mb colons. Images of TUNEL staining (green) of colon sections from Gatm+/+ and Gatmmb/mb mice on day 6 of 1.5% DSS treatment (10× magnification); images are representative of three independent experiments.
Fig. S4.
Fig. S4.
Mitochondrial mass and functioning in Gatmc/c colonocytes. (A) Colonocytes from untreated and 3 d DSS-treated Gatmc/c (blue) vs. Gatm+/+ littermates (red) were harvested and stained with MitoTracker Green. Data are representative of three independent experiments. (B) Mitochondrial respiration of colonic crypts from mice of the indicated genotypes treated for 3 d with 1.5% DSS was tested using Seahorse Flux. Oligomycin, carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP), and a mix of rotenone and antimycin A were serially injected to measure ATP-linked respiration, maximal respiration, and nonmitochondrial respiration, respectively. Data are representative of three independent experiments.

Similar articles

See all similar articles

Cited by 13 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback