Epithelial argininosuccinate synthetase is dispensable for intestinal regeneration and tumorigenesis

Abstract

The epithelial signaling pathways involved in damage and regeneration, and neoplastic transformation are known to be similar. We noted upregulation of argininosuccinate synthetase (ASS1) in hyperproliferative intestinal epithelium. Since ASS1 leads to de novo synthesis of arginine, an important amino acid for the growth of intestinal epithelial cells, its upregulation can contribute to epithelial proliferation necessary to be sustained during oncogenic transformation and regeneration. Here we investigated the function of ASS1 in the gut epithelium during tissue regeneration and tumorigenesis, using intestinal epithelial conditional Ass1 knockout mice and organoids, and tissue specimens from colorectal cancer patients. We demonstrate that ASS1 is strongly expressed in the regenerating and Apc-mutated intestinal epithelium. Furthermore, we observe an arrest in amino acid flux of the urea cycle, which leads to an accumulation of intracellular arginine. However, loss of epithelial Ass1 does not lead to a reduction in proliferation or increase in apoptosis in vivo, also in mice fed an arginine-free diet. Epithelial loss of Ass1 seems to be compensated by altered arginine metabolism in other cell types and the liver.

Fig. 1. Ass1 expression increases during regeneration and upon loss of Apc.

A Schematic overview of the mouse model. B Heatmap showing the expression of genes in the KEGG geneset ‘Arginine biosynthesis’, from RNA isolated from the intestines of irradiated mice at t = 0, 24, 48, and 96 h (n = 10 mice per time point). C qRT-PCR for Ass1 on intestinal lysates from the same experiment (n = 5 mice per time point), Actb is used as a reference gene. D Immunohistochemistry for ASS1 in small intestine at t = 0 and t = 96 h after irradiation. Micrographs show representative images. Scale bar represents 25 µm. E qRT-PCR for Ass1 in murine adenoma-to-carcinoma sequence organoids relative to non-induced control organoid (relative to Actb/Ppia), A = Apc^−/−, K = Kras^G12D/+, S = Smad4 shRNA, P = Trp53^−/− (for single knockout) or Trp53 shRNA in combination with AKS. ^*p < 0.05, ^**p < 0.01 by student’s t-test. F Western blot for ASS1 and beta-actin in wild type (WT; from C57BL/6 mice), VillinCre^ERT2 Apc^fl/fl and VillinCre^ERT2 Apc^−/− organoids. G Immunohistochemistry for ASS1 in VillinCre^ERT2 Apc^fl/fl and VillinCre^ERT2 Apc^−/− organoids. Micrographs show representative images. Scale bar represents 50 µm. H Heatmap showing qRT-PCR results for various urea cycle and creatine synthesis, glutamate and proline metabolism, polyamine synthesis, and arginine transporter genes.

Fig. 2. ASS1 expression is upregulated and correlates with APC mutations in human colorectal adenomas and carcinomas.

AASS1 mRNA and ASS1 protein expression in 20 different colorectal cancer cell lines. Per cell line the mutational status of APC, KRAS, SMAD4, and TP53 is indicated below. B ASS1 mRNA expression in the same 20 colorectal cancer cell lines as in (A), separated by their mutational status. ^**p < 0.01 by Kruskal-Wallis test. C ASS1 expression in paired samples of normal mucosa and colorectal adenoma tissue (n = 32) from the publicly available expression dataset GDS2947. ^****p < 0.0001 by Wilcoxon signed rank test. D qRT-PCR for ASS1 in tissue from 60 adenoma and 8 carcinoma samples with paired controls. ^*p < 0.05, ^****p < 0.0001 by Wilcoxon signed rank test. E immunohistochemistry for ASS1 in healthy (n = 70), adenoma (n = 12), and carcinoma (n = 75) tissue. Micrographs show representative images. Scale bar indicates 100 µm. F Quantification of (E), which was scored in a blinded manner by two independent observers. ^**p < 0.01, ^***p < 0.001 by one-way ANOVA with Tukey’s multiple comparisons test.

Fig. 3. Loss of Apc leads to altered amino acid consumption and synthesis.

A Amino acid concentrations as measured by HPLC in supernatant of wild-type, VillinCre^ERT2 Apc^fl/fl, and VillinCre^ERT2 Apc^−/− organoids. B schematic overview of labeling. C LC-MS measurement of intracellular M4 citrulline, M4 arginine, and M4 argininosuccinate, in wild-type and VillinCre^ERT2 Apc^−/− organoids after incubation with labeled M4 citrulline (13C1;3,3,4-D3), and ratios between M3/M4-labeled citrulline and arginine. ^*p < 0.05 by student’s t-test.

Fig. 4. Ass1 knockout in the intestinal epithelium does not affect regeneration after damage by irradiation.

A Schematic overview of the mouse model. B qRT-PCR for Ass1 in small intestinal lysates of Ass1^wt/wt and Ass1^−/− mice 96 h after irradiation. Actb was used as a reference gene. ^***p < 0.001 by Student’s t-test. C Immunohistochemistry for ASS1 in small intestinal tissue of Ass1^wt/wt and Ass1^−/− mice 96 h after irradiation. Representative images shown. Scale bar indicates 100 µm. D Average villus length. E Amount of hyperproliferative crypts per mm intestine. n = 8 mice per group.

Fig. 5. Ass1 knockout in the intestinal epithelium does not affect tumorigenesis in vivo.

A Schematic overview of the Apc^fl/fl mouse model. B Relative weight compared to start tamoxifen injections. C Immunohistochemistry for ASS1 (upper panels) and BrdU (lower panels) of Ass1^wt/wt and Ass1^−/− mice 5 days after recombination. Representative images are shown. Scale bar represents 50 µm. D Amount of BrdU-positive cells per crypt in the small intestine. E Amount of cleaved caspase-3-positive cells per crypt in the small intestine. F Schematic overview of the Apc^wt/fl mouse model. G Relative weight compared to start tamoxifen injections. H Immunohistochemistry for ASS1 in adenomas in Ass1^wt/wt and Ass1^−/− mice. Scale bar represents 100 µm. I Amount of adenomas in both groups. J Average adenoma size.

Fig. 6. Growth of Ass1-deficient small intestinal organoids is impaired in arginine-low conditions only and arginine transporters could compensate for loss of Ass1.

A qRT-PCR for Ass1 in Ass1 and Apc knockout organoids. Actb and Ppia are used as reference genes. ^*p < 0.05 compared to Ass1^wt/wt Apc^wt/wt organoids, ^#p < 0.05 compared to Ass1^wt/wt Apc^−/− organoids by one-way ANOVA with Tukey’s multiple comparisons test. B Western blot for ASS1 and Vinculin in Ass1 and Apc knockout organoids. C Intracellular LC-MS measurements of M4 arginine in organoids cultured with M4 citrulline (U-13C;3,3,4-D3). D qRT-PCR for indicated urea-cycle genes. E qRT-PCR for indicated amino acid transporters. ^*p < 0.05, ^***p < 0.001, ^****p < 0.0001 by one-way ANOVA with Tukey’s multiple comparisons test. F Microscopic bright-field images of organoids cultured in regular and low-arginine medium. Representative images are shown. Scale bar represents 100 µm. G Quantification of organoid size cultured in regular and low-arginine medium (n = 3 independent experiments) ^*p < 0.05 student’s t-test.

Fig. 7. Ass1 knockout in the intestinal epithelium does not affect Apc-mediated hyperproliferation in mice fed an arginine-free diet.

A Schematic overview of the mouse model. B Relative weight compared to start tamoxifen injections. C Amount of BrdU-positive cells per crypt in the small intestine. D Amount of cleaved caspase-3-positive cells per crypt in the small intestine.

Fig. 8. Plasma glutamine levels and expression of urea-cycle enzymes in the liver, might compensate for intestinal loss of Ass1.

A qRT-PCRs for indicated genes in liver tissue from VillinCre^ERT2 Ass1^wt/wt and VillinCre^ERT2 Ass1^−/− mice 96 h after 14 Gy irradiation. Actb was used as a reference gene. ^**p < 0.01 by Student’s t-test. B Plasma levels of indicated amino acids measured by HPLC from VillinCre^ERT2 Ass1^wt/wt and VillinCre^ERT2 Ass1^−/− mice 96 h after 14 Gy irradiation. ^*p < 0.05 by Student’s t-test. C qRT-PCRs for indicated genes in liver tissue from VillinCre^ERT2 Ass1^wt/wt Apc^wt/wt, VillinCre^ERT2 Ass1^wt/wt Apc^−/−, and VillinCre^ERT2 Ass1^−/− Apc^−/− mice, 5 days after induction. Actb was used as a reference gene. ^*p < 0.05 by one-way ANOVA with Tukey’s multiple comparisons test.