Limits...
Let-7 Represses Carcinogenesis and a Stem Cell Phenotype in the Intestine via Regulation of Hmga2.

Madison BB, Jeganathan AN, Mizuno R, Winslow MM, Castells A, Cuatrecasas M, Rustgi AK - PLoS Genet. (2015)

Bottom Line: Analysis of both mouse and human intestinal cancer specimens reveals that stem cell markers were significantly associated with loss of Let-7 miRNA expression, and that a number of Let-7 targets were elevated, including Hmga1 and Hmga2.In addition, inactivation of a single Hmga2 allele in the mouse intestine epithelium significantly represses tumorigenesis driven by Lin28b.In aggregate, we conclude that Let-7 depletion drives a stem cell phenotype and the development of intestinal cancer, primarily via Hmga2.

View Article: PubMed Central - PubMed

Affiliation: Division of Gastroenterology, Washington University School of Medicine, Saint Louis, Missouri, United States of America; Department of Medicine, Washington University School of Medicine, Saint Louis, Missouri, United States of America.

ABSTRACT
Let-7 miRNAs comprise one of the largest and most highly expressed family of miRNAs among vertebrates, and is critical for promoting differentiation, regulating metabolism, inhibiting cellular proliferation, and repressing carcinogenesis in a variety of tissues. The large size of the Let-7 family of miRNAs has complicated the development of mutant animal models. Here we describe the comprehensive repression of all Let-7 miRNAs in the intestinal epithelium via low-level tissue-specific expression of the Lin28b RNA-binding protein and a conditional knockout of the MirLet7c-2/Mirlet7b locus. This ablation of Let-7 triggers the development of intestinal adenocarcinomas concomitant with reduced survival. Analysis of both mouse and human intestinal cancer specimens reveals that stem cell markers were significantly associated with loss of Let-7 miRNA expression, and that a number of Let-7 targets were elevated, including Hmga1 and Hmga2. Functional studies in 3-D enteroids revealed that Hmga2 is necessary and sufficient to mediate many characteristics of Let-7 depletion, namely accelerating cell cycle progression and enhancing a stem cell phenotype. In addition, inactivation of a single Hmga2 allele in the mouse intestine epithelium significantly represses tumorigenesis driven by Lin28b. In aggregate, we conclude that Let-7 depletion drives a stem cell phenotype and the development of intestinal cancer, primarily via Hmga2.

No MeSH data available.


Related in: MedlinePlus

Hmga2 mediates Lin28b effects on stem cell colony formation and enteroid proliferation.Colony formation assay of small intestine enteroids from wild-type (WT) mice (A) and Vil-Lin28bMed mice (B). Two wells of a 24-well plate are pictured. C) Quantification of colony forming assay of WT mice and Vil-Lin28bMed mice. D) Schematic of lentiviral vector-mediated transduction of enteroids with a “tet-on” doxycycline (dox)-inducible vector. Inducible expression of a turbo RFP reporter in the unmodified pTRIPz vector is readily observed in enteroids (H) vs. untreated cells (G). Phase contrast images of untreated and doxycycline treated enteroids are pictured in (E) and (F), respectively. Immunostaining for Hmga2 in sectioned enteroids from pTRIPz-transduced (I), pTRIPz-Hmga2-transduced (J), and pTRIPz-Hmga2-transduced plus doxycycline (K). L) Comparison of colony forming potential of cells from dissociated enteroids transduced with pTRIPz (empty), Arid3a, Hif3a, or Hmga2 vectors. Phase contrast images of colony formation assays from enteroids transduced with pTRIPz (empty) (M) or Hmga2 (N) vectors. O) Colony forming assays of non-transgenic mice (NTG) or Vil-Lin28bMed mice with and without inactivation of one conditional Hmga2 allele using Vil-Cre. EdU incorporation of enteroids, as assayed by flow cytometry, transduced with pTRIPz-Hmga2 (P) or pTRIPz-Hif3a (Q) vectors. Cultures were treated with 100 ng/ml doxycycline in all experiments except in (F) and (H), which were treated with 500 ng/ml doxycycline. All experiments were performed at least in triplicate. Student’s two-tailed T-tests were performed to determine significance with * p-value < 0.05, ** p-value < 0.01, and *** p-value < 0.001, relative to pTRIPz vector controls, unless noted otherwise.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4526516&req=5

pgen.1005408.g006: Hmga2 mediates Lin28b effects on stem cell colony formation and enteroid proliferation.Colony formation assay of small intestine enteroids from wild-type (WT) mice (A) and Vil-Lin28bMed mice (B). Two wells of a 24-well plate are pictured. C) Quantification of colony forming assay of WT mice and Vil-Lin28bMed mice. D) Schematic of lentiviral vector-mediated transduction of enteroids with a “tet-on” doxycycline (dox)-inducible vector. Inducible expression of a turbo RFP reporter in the unmodified pTRIPz vector is readily observed in enteroids (H) vs. untreated cells (G). Phase contrast images of untreated and doxycycline treated enteroids are pictured in (E) and (F), respectively. Immunostaining for Hmga2 in sectioned enteroids from pTRIPz-transduced (I), pTRIPz-Hmga2-transduced (J), and pTRIPz-Hmga2-transduced plus doxycycline (K). L) Comparison of colony forming potential of cells from dissociated enteroids transduced with pTRIPz (empty), Arid3a, Hif3a, or Hmga2 vectors. Phase contrast images of colony formation assays from enteroids transduced with pTRIPz (empty) (M) or Hmga2 (N) vectors. O) Colony forming assays of non-transgenic mice (NTG) or Vil-Lin28bMed mice with and without inactivation of one conditional Hmga2 allele using Vil-Cre. EdU incorporation of enteroids, as assayed by flow cytometry, transduced with pTRIPz-Hmga2 (P) or pTRIPz-Hif3a (Q) vectors. Cultures were treated with 100 ng/ml doxycycline in all experiments except in (F) and (H), which were treated with 500 ng/ml doxycycline. All experiments were performed at least in triplicate. Student’s two-tailed T-tests were performed to determine significance with * p-value < 0.05, ** p-value < 0.01, and *** p-value < 0.001, relative to pTRIPz vector controls, unless noted otherwise.

Mentions: We next pursued 3-D culture and manipulation of intestinal organoids (enteroids) to explore the relationship between Let-7 targets and a stem cell phenotype. This method has facilitated the examination of stem cell phenotypes in the intestinal epithelium in multiple studies [34,38–44]. For these experiments we derived enteroids from Vil-Lin28bMed mice [18]. We have previously shown that crypt hyperplasia and Hmga2 expression is dependent on Let-7 depletion in crypts from Vil-Lin28bMed mice [18]. Enteroids derived from Vil-Lin28bMed mice exhibited enhanced colony forming potential of single cells (Fig 6A, 6B and 6C). This is unlikely to be a feature secondary to enhanced stem cell potential conferred by increased association with Paneth cells, as described previously [34], since this cell type is severely depleted following Let-7 repression [18]. To assay exogenous expression of Let-7 targets in enteroids, we used a lentivirus vector for transduction of wild-type mouse small intestine enteroids (Fig 6D–6G). This vector system yields low (Fig 6J) or high-level (Fig 6K) expression, in a doxycycline-dependent manner. We generated stable enteroid lines for inducible expression of mouse Hmga2, Igf2bp2, E2F5, Arid3a, or Hif3a and assayed colony forming potential and EdU incorporation. We focused on Hmga2, rather than Hmga1, as it is consistently up-regulated in non-malignant intestinal tissue from Vil-Lin28bMed and Lin28bLo/Let7IEC-KO and thus appears highly dependent on Let-7 [18]. For colony formation, only Hmga2 over-expression (O/E) exhibited a significant effect, with enhanced formation of new enteroids from singly plated cells (Fig 6L, 6M and 6N), whereas Igf2bp2, E2F5, Arid3a, and Hif3a had no apparent effect (Fig 6L and S4A Fig). Expression of Hmga2, Arid3a, Hif3a, or Igf2bp2 via lentiviral vectors did not induce any change in stem cell markers (S4B–S4K Fig). To determine if Hmga2 was necessary for the enhanced colony formation in Vil-Lin28b enteroids, we crossed Vil-Lin28bMed mice onto background in which one allele of Hmga2 is inactivated specifically in the intestine (Vil-Cre+/Hmga2CK/+) [45], and generated enteroids. We used Vil-Lin28bMed mice because their phenotype appears Let-7-dependent [18] and for simpler breeding. Effects on colony formation by Lin28b were greatly blunted by inactivation of a single Hmga2 allele (Fig 6O). Hmga2 could also trigger increased EdU incorporation in intestinal enteroids, whereas Hif3a repressed it, suggesting opposing effects of Hmga2 and Hif3a on cellular proliferation (Fig 6P and 6Q). Lentiviral-mediated expression and manipulation of the Hmga2 conditional allele were restricted to coding sequence only [45]. Perhaps consistent with its association with a stem cell phenotype, HMGA2 is also frequently co-expressed with the stem cell markers MSI1 and LGR5 in human CRC, and notably, more frequently than any of the other Let-7 targets evaluated here in this study (Fig 5L and S2 Table). Lastly, to evaluate the role of Hmga2 in intestinal tumorigenesis in the context of Let-7 depletion we examined tumor burden in Vil-Lin28bMed and Vil-Lin28bMed/Hmga2+/IEC-KO mice. As mentioned earlier, Vil-Lin28bMed mice have a lower penetrance of intestinal tumorigenesis compared to Lin28bLo/Let7IEC-KO mice, with about 50% of animals developing tumors by 9 months of age (S3 Table). Inactivation of one allele of Hmga2 in the intestinal epithelium significantly reduced disease penetrance and tumor burden in Vil-Lin28bMed/Hmga2+/IEC-KO mice (S3 Table).


Let-7 Represses Carcinogenesis and a Stem Cell Phenotype in the Intestine via Regulation of Hmga2.

Madison BB, Jeganathan AN, Mizuno R, Winslow MM, Castells A, Cuatrecasas M, Rustgi AK - PLoS Genet. (2015)

Hmga2 mediates Lin28b effects on stem cell colony formation and enteroid proliferation.Colony formation assay of small intestine enteroids from wild-type (WT) mice (A) and Vil-Lin28bMed mice (B). Two wells of a 24-well plate are pictured. C) Quantification of colony forming assay of WT mice and Vil-Lin28bMed mice. D) Schematic of lentiviral vector-mediated transduction of enteroids with a “tet-on” doxycycline (dox)-inducible vector. Inducible expression of a turbo RFP reporter in the unmodified pTRIPz vector is readily observed in enteroids (H) vs. untreated cells (G). Phase contrast images of untreated and doxycycline treated enteroids are pictured in (E) and (F), respectively. Immunostaining for Hmga2 in sectioned enteroids from pTRIPz-transduced (I), pTRIPz-Hmga2-transduced (J), and pTRIPz-Hmga2-transduced plus doxycycline (K). L) Comparison of colony forming potential of cells from dissociated enteroids transduced with pTRIPz (empty), Arid3a, Hif3a, or Hmga2 vectors. Phase contrast images of colony formation assays from enteroids transduced with pTRIPz (empty) (M) or Hmga2 (N) vectors. O) Colony forming assays of non-transgenic mice (NTG) or Vil-Lin28bMed mice with and without inactivation of one conditional Hmga2 allele using Vil-Cre. EdU incorporation of enteroids, as assayed by flow cytometry, transduced with pTRIPz-Hmga2 (P) or pTRIPz-Hif3a (Q) vectors. Cultures were treated with 100 ng/ml doxycycline in all experiments except in (F) and (H), which were treated with 500 ng/ml doxycycline. All experiments were performed at least in triplicate. Student’s two-tailed T-tests were performed to determine significance with * p-value < 0.05, ** p-value < 0.01, and *** p-value < 0.001, relative to pTRIPz vector controls, unless noted otherwise.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4526516&req=5

pgen.1005408.g006: Hmga2 mediates Lin28b effects on stem cell colony formation and enteroid proliferation.Colony formation assay of small intestine enteroids from wild-type (WT) mice (A) and Vil-Lin28bMed mice (B). Two wells of a 24-well plate are pictured. C) Quantification of colony forming assay of WT mice and Vil-Lin28bMed mice. D) Schematic of lentiviral vector-mediated transduction of enteroids with a “tet-on” doxycycline (dox)-inducible vector. Inducible expression of a turbo RFP reporter in the unmodified pTRIPz vector is readily observed in enteroids (H) vs. untreated cells (G). Phase contrast images of untreated and doxycycline treated enteroids are pictured in (E) and (F), respectively. Immunostaining for Hmga2 in sectioned enteroids from pTRIPz-transduced (I), pTRIPz-Hmga2-transduced (J), and pTRIPz-Hmga2-transduced plus doxycycline (K). L) Comparison of colony forming potential of cells from dissociated enteroids transduced with pTRIPz (empty), Arid3a, Hif3a, or Hmga2 vectors. Phase contrast images of colony formation assays from enteroids transduced with pTRIPz (empty) (M) or Hmga2 (N) vectors. O) Colony forming assays of non-transgenic mice (NTG) or Vil-Lin28bMed mice with and without inactivation of one conditional Hmga2 allele using Vil-Cre. EdU incorporation of enteroids, as assayed by flow cytometry, transduced with pTRIPz-Hmga2 (P) or pTRIPz-Hif3a (Q) vectors. Cultures were treated with 100 ng/ml doxycycline in all experiments except in (F) and (H), which were treated with 500 ng/ml doxycycline. All experiments were performed at least in triplicate. Student’s two-tailed T-tests were performed to determine significance with * p-value < 0.05, ** p-value < 0.01, and *** p-value < 0.001, relative to pTRIPz vector controls, unless noted otherwise.
Mentions: We next pursued 3-D culture and manipulation of intestinal organoids (enteroids) to explore the relationship between Let-7 targets and a stem cell phenotype. This method has facilitated the examination of stem cell phenotypes in the intestinal epithelium in multiple studies [34,38–44]. For these experiments we derived enteroids from Vil-Lin28bMed mice [18]. We have previously shown that crypt hyperplasia and Hmga2 expression is dependent on Let-7 depletion in crypts from Vil-Lin28bMed mice [18]. Enteroids derived from Vil-Lin28bMed mice exhibited enhanced colony forming potential of single cells (Fig 6A, 6B and 6C). This is unlikely to be a feature secondary to enhanced stem cell potential conferred by increased association with Paneth cells, as described previously [34], since this cell type is severely depleted following Let-7 repression [18]. To assay exogenous expression of Let-7 targets in enteroids, we used a lentivirus vector for transduction of wild-type mouse small intestine enteroids (Fig 6D–6G). This vector system yields low (Fig 6J) or high-level (Fig 6K) expression, in a doxycycline-dependent manner. We generated stable enteroid lines for inducible expression of mouse Hmga2, Igf2bp2, E2F5, Arid3a, or Hif3a and assayed colony forming potential and EdU incorporation. We focused on Hmga2, rather than Hmga1, as it is consistently up-regulated in non-malignant intestinal tissue from Vil-Lin28bMed and Lin28bLo/Let7IEC-KO and thus appears highly dependent on Let-7 [18]. For colony formation, only Hmga2 over-expression (O/E) exhibited a significant effect, with enhanced formation of new enteroids from singly plated cells (Fig 6L, 6M and 6N), whereas Igf2bp2, E2F5, Arid3a, and Hif3a had no apparent effect (Fig 6L and S4A Fig). Expression of Hmga2, Arid3a, Hif3a, or Igf2bp2 via lentiviral vectors did not induce any change in stem cell markers (S4B–S4K Fig). To determine if Hmga2 was necessary for the enhanced colony formation in Vil-Lin28b enteroids, we crossed Vil-Lin28bMed mice onto background in which one allele of Hmga2 is inactivated specifically in the intestine (Vil-Cre+/Hmga2CK/+) [45], and generated enteroids. We used Vil-Lin28bMed mice because their phenotype appears Let-7-dependent [18] and for simpler breeding. Effects on colony formation by Lin28b were greatly blunted by inactivation of a single Hmga2 allele (Fig 6O). Hmga2 could also trigger increased EdU incorporation in intestinal enteroids, whereas Hif3a repressed it, suggesting opposing effects of Hmga2 and Hif3a on cellular proliferation (Fig 6P and 6Q). Lentiviral-mediated expression and manipulation of the Hmga2 conditional allele were restricted to coding sequence only [45]. Perhaps consistent with its association with a stem cell phenotype, HMGA2 is also frequently co-expressed with the stem cell markers MSI1 and LGR5 in human CRC, and notably, more frequently than any of the other Let-7 targets evaluated here in this study (Fig 5L and S2 Table). Lastly, to evaluate the role of Hmga2 in intestinal tumorigenesis in the context of Let-7 depletion we examined tumor burden in Vil-Lin28bMed and Vil-Lin28bMed/Hmga2+/IEC-KO mice. As mentioned earlier, Vil-Lin28bMed mice have a lower penetrance of intestinal tumorigenesis compared to Lin28bLo/Let7IEC-KO mice, with about 50% of animals developing tumors by 9 months of age (S3 Table). Inactivation of one allele of Hmga2 in the intestinal epithelium significantly reduced disease penetrance and tumor burden in Vil-Lin28bMed/Hmga2+/IEC-KO mice (S3 Table).

Bottom Line: Analysis of both mouse and human intestinal cancer specimens reveals that stem cell markers were significantly associated with loss of Let-7 miRNA expression, and that a number of Let-7 targets were elevated, including Hmga1 and Hmga2.In addition, inactivation of a single Hmga2 allele in the mouse intestine epithelium significantly represses tumorigenesis driven by Lin28b.In aggregate, we conclude that Let-7 depletion drives a stem cell phenotype and the development of intestinal cancer, primarily via Hmga2.

View Article: PubMed Central - PubMed

Affiliation: Division of Gastroenterology, Washington University School of Medicine, Saint Louis, Missouri, United States of America; Department of Medicine, Washington University School of Medicine, Saint Louis, Missouri, United States of America.

ABSTRACT
Let-7 miRNAs comprise one of the largest and most highly expressed family of miRNAs among vertebrates, and is critical for promoting differentiation, regulating metabolism, inhibiting cellular proliferation, and repressing carcinogenesis in a variety of tissues. The large size of the Let-7 family of miRNAs has complicated the development of mutant animal models. Here we describe the comprehensive repression of all Let-7 miRNAs in the intestinal epithelium via low-level tissue-specific expression of the Lin28b RNA-binding protein and a conditional knockout of the MirLet7c-2/Mirlet7b locus. This ablation of Let-7 triggers the development of intestinal adenocarcinomas concomitant with reduced survival. Analysis of both mouse and human intestinal cancer specimens reveals that stem cell markers were significantly associated with loss of Let-7 miRNA expression, and that a number of Let-7 targets were elevated, including Hmga1 and Hmga2. Functional studies in 3-D enteroids revealed that Hmga2 is necessary and sufficient to mediate many characteristics of Let-7 depletion, namely accelerating cell cycle progression and enhancing a stem cell phenotype. In addition, inactivation of a single Hmga2 allele in the mouse intestine epithelium significantly represses tumorigenesis driven by Lin28b. In aggregate, we conclude that Let-7 depletion drives a stem cell phenotype and the development of intestinal cancer, primarily via Hmga2.

No MeSH data available.


Related in: MedlinePlus