Absolute requirement for STAT3 function in small-intestine crypt stem cell survival - PubMed (original) (raw)

Absolute requirement for STAT3 function in small-intestine crypt stem cell survival

J R Matthews et al. Cell Death Differ. 2011 Dec.

Abstract

The transcription factor signal transducer and activator of transcription 3 (STAT3) is frequently activated in human cancers. Interestingly, STAT3 also maintains the pluripotency and self-renewal of murine embryonic stem cells, and several tissue stem cell types. To investigate whether STAT3 also maintains the small-intestine crypt stem cell, we conditionally inactivated a Floxed Stat3 allele (Stat3(fl)) in murine small-intestine crypt stem cells. Following Cre recombinase expression, apoptosis increased in Stat3(fl/-) experimental crypts relative to Stat3(wt/-) controls before declining. Control Stat3(wt/-) mice carrying a Flox-STOP LacZ reporter transgene stably expressed LacZ after Cre induction. In contrast, Stat3(fl/-) intestine LacZ expression initially increased modestly, before declining to background levels. Quantitative PCRs revealed a similar transient in recombined Stat3(fl) allele levels. Long-term bromodeoxyuridine labelling directly demonstrated that functional STAT3 is required for +4 to +6 region label-retaining small-intestine stem cell survival. Rapid clearance of recombined Stat3(fl/-) cells involves apoptosis potentially induced by elevated c-Myc in non-recombined cells and involves elevated p53 expression and caspase 3 activation. Intriguingly, Stat3(fl/-) intestine recombination triggered dramatically upregulated polycomb transcriptional repressor Bmi1 - potentially accelerating recombined crypt repopulation. In summary, STAT3 activity is absolutely required for small-intestine crypt stem cell survival at both the +4 to +6 label-retaining and crypt base columnar cell locations.

PubMed Disclaimer

Figures

Figure 1

Figure 1

Use of a Flox-STOP LacZ transgene as a surrogate marker of Stat3 fl recombination to visualise the transient appearance and disappearance of STAT3-null crypts. (a) LacZ enzyme activity in Stat3 fl/− and Stat3 wt/− small intestine at 1, 2 and 3 days after _β_-naphthoflavone injection/induction of Cre expression. (b) Cross-section of small intestine showing LacZ enzyme activity in Stat3 fl/− and Stat3 wt/− mice at 3 days after _β_-naphthoflavone injection/induction of Cre expression. (c) Crypt bases in Stat3 fl/− and Stat3 wt/− small intestine at 3 days following _β_-naphthoflavone injection/induction of Cre expression viewed at high power ( × 90 magnification) after removal of the smooth muscle layer. LacZ enzyme activity panels are representative of three Stat3 fl/− versus three Stat3 wt/− mice per time point. (d) Anti-STAT3 immunoreactivity (with the NEB/CST #9132 primary antibody recognising an epitope close to the STAT3 tyrosine 705 residue) in Stat3 fl/− versus Stat3 wt/− small intestine at 1.7 days after induction of Cre expression. Immunohistochemistry panels are representative of three Stat3 fl/− versus three Stat3 wt/− mice, the black bars represent 50 _μ_m. (e) Anti-STAT3 phosphotyrosine 705 immunoreactivity (using the NEB/CST #9131 primary antibody) in un-induced Stat3 wt/− control small intestine. The panel is representative of three Stat3 wt/− mice, the black bar represents 50 _μ_m. (f) Schematic of the small-intestine crypt base showing the thin CBC/Lgr5+ slowly proliferating stem cells interleaved by Paneth cells with the over-lying +4 to +6 region containing long-term, label-retaining, quiescent stem cells

Figure 2

Figure 2

Use of anti-STAT3 immunohistochemistry to trace migration of functionally STAT3-null small-intestine crypt epithelial cells after induction of Cre expression. (a) Average frequency of cells lacking any anti-STAT3 immunoreactivity for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 1.7 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. (b) Average frequency of cells lacking any anti-STAT3 immunoreactivity for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 2 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. (c) Average frequency of cells lacking any anti-STAT3 immunoreactivity for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 3 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. The error bars indicate the values of the standard error of the mean (S.E.M.), _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point

Figure 3

Figure 3

Changes in an average apoptotic and mitotic indices and average crypt length in Stat3 fl/− and wt/− small-intestine crypts in response to induction of Cre recombinase expression. (a) Apoptotic indices (average percentage of crypt cells undergoing apoptosis) in Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts between 1 and 4 days following _β_-naphthoflavone-driven induction of Cre recombinase expression. (b) Mitotic indices (average percentage of crypt cells undergoing mitosis) in Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts between 1 and 4 days following induction of Cre expression. (c) Average length of small-intestine crypts in Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) mice between 1 and 4 days following induction of Cre expression. The error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point

Figure 4

Figure 4

Changes in individual small-intestine crypt cell apoptotic indices in Stat3 fl/− and Stat3 wt/− crypts in response to induction of Cre recombinase expression. (a) Apoptotic indices for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 1.7 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. (b) Apoptotic indices for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 2 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. (c) Apoptotic indices for Stat3 fl/− (dashed line/triangles) and Stat3 wt/− (solid line/squares) crypts at individual crypt cell positions at 3 days after _β_-naphthoflavone injection/induction of Cre recombinase expression. The error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point

Figure 5

Figure 5

Correlation of elevated apoptotic indices in Stat3 fl/− crypts following induction of Cre recombinase expression with elevated activated (proteolytically cleaved) caspase 3 immunoreactivity. (a) Anti-activated caspase 3 immunoreactivity in Stat3 wt/− and Stat3 fl/− small-intestine crypts at 2 days after Cre induction, the black bars represent 50 _μ_m. (b) Frequency of activated caspase 3 immunoreactive cells at individual crypt cell positions in Stat3 fl/− (dashed line/triangles) versus Stat3 wt/− (solid line/squares) small intestine at 2 days after Cre induction. The error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point. (c) Overall frequency of activated caspase 3 immunoreactive cells in Stat3 wt/− (solid bar) versus Stat3 fl/− (open bar) small-intestine crypts at 2 days after Cre induction, error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice. (d) Representative apoptotic body (arrowed) in Stat3 fl/− small-intestine crypt at 2 days after induction of Cre expression, the black bar represents 20 _μ_m

Figure 6

Figure 6

Long-term BrdU labelling of Stat3 fl/− and Stat3 wt/− small-intestine crypts demonstrating the loss of BrdU-labelled +4 to +6 crypt cells in response to STAT3 inactivation. Stat3 fl/− and Stat3 wt/− mice were given BrdU in their drinking water for 3 weeks, followed by an 8-week water-only chase phase. This was followed by _β_-naphthoflavone injection and induction of Cre recombinase expression with a 7 day delay before harvesting. Small-intestine samples were then analysed by anti-BrdU immunohistochemistry and all BrdU-positive, non-Paneth, nuclei identified. The BrdU-positive cells were then divided up into three groups – crypt base positions +1 to +3 (base), the +4 to +6 positions (+4 to +6) and crypt positions from +7 to the crypt-villus junction (higher). (a) Anti-BrdU immunohistochemistry of a (Stat3 wt/−) crypt demonstrating a representative BrdU-positive nucleus (arrowed) in the +4 to +6 crypt position range, the black bar represents 20 _μ_m. (b) Frequency of crypts retaining a BrdU-positive nucleus at the crypt base, +4 to +6 and higher positions, Stat3 wt/− crypts are represented by hatched bars, whereas Stat3 fl/− crypts are represented by open bars. The error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice

Figure 7

Figure 7

Q-PCR demonstrating the transient appearance and disappearance of the recombined Stat3 fl allele. (a) Ct values (cycles) were plotted versus time after induction of Cre expression for the recombined Stat3 fl allele Q-PCR reaction using Stat3 fl/− small-intestine genomic DNA (solid line, solid squares) and for the recombined Stat3 fl allele Q-PCR reaction using Stat3 wt/− small-intestine genomic DNA (dashed line, solid triangles). Ct values were also plotted versus time after induction of Cre expression for the control Stat3 (wt/fl/− alleles) Q-PCR reaction using Stat3 fl/− small-intestine genomic DNA (solid line, open squares) and for the control Q-PCR reaction using Stat3 wt/− small-intestine genomic DNA (dashed line, open triangles). The error bars indicate S.E.M. values, N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point. (b) Q-PCR Ct values for the recombined Stat3 fl allele reaction versus time after induction of Cre expression were then converted into relative, linear-scaled, values for the level of the recombined Stat3 fl allele. This involved subtracting the average control Stat3 Q-PCR Ct value from the average recombined Stat3 fl Q-PCR Ct value to yield a difference value (difference). The un-induced (day 0) time point difference value from Stat3 wt/− intestine genomic DNA was then normalised to the Stat3 fl/− intestine genomic DNA difference value to give a normalised difference value (normalised difference). Then the Stat3 fl/− difference values were subtracted from the Stat3 wt/− normalised difference values to yield the difference of these difference values (difference of difference). Finally, the difference of difference values were raised to 2_x to yield a relative measure of the recombined Stat3 fl allele, which could be plotted directly on a linear scale (no error bars are provided). _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point

Figure 8

Figure 8

Upregulation of c-Myc expression in Stat3 fl/− relative to Stat3 wt/− crypts at day 0.5 after induction of Cre expression as a potential mechanism for non-recombined Stat3 fl/− cells to trigger p53-mediated apoptosis in Cre-recombined Stat3 fl/− cells. (a) Anti-c-Myc immunohistochemistry in Stat3 fl/− versus Stat3 wt/− crypts at day 0.5 after induction of Cre expression. Immunohistochemistry panels are representative of three Stat3 fl/− versus three Stat3 wt/− mice, the black bars represent 20 _μ_m. (b) Statistically significant increase in numbers of c-Myc-positive nuclei in Stat3 fl/− (open bar) versus Stat3 wt/− (hatched bar) crypts at day 0.5 after induction of Cre expression (_P_=0.0404), the error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice. (c) Quantitation of anti-p53 immunohistochemistry demonstrating a statistically significant increase in the frequency of p53-positive cell nuclei in Stat3 fl/− (open bars) relative to Stat3 wt/− (hatched bars) crypts at 2 days (_P_=0.0259, _N_=4 Stat3 fl/− versus 3 Stat3 wt/− mice) and 3 days (_P_=0.0259 or 0.0249 adjusted for ties), _N_=4 Stat3 fl/− versus 3 Stat3 wt/− mice) after induction of Cre expression

Figure 9

Figure 9

Confocal fluorescence immunohistochemistry demonstrating activation of caspase 3 cleavage in a Stat3 fl/− crypt at 2 days after induction of Cre expression and its correlation with loss of STAT3 immunoreactivity in corresponding cells identified (arrowed) on the subsequent serial tissue section. The white scale bars represent 20 _μ_m

Figure 10

Figure 10

Quantitation of a transient peak in Bmi1 expression in Stat3 fl/− crypts following _β_-naphthoflavone injection and induction of Cre expression. (a) Stat3 fl/− and Stat3 wt/− small intestines were subjected to anti-Bmi1 immunohistochemistry. Strongly anti-Bmi1 immunoreactive crypt nuclei were counted and the average percentage of strongly positive nuclei in Stat3 wt/− (hatched bars) and Stat3 fl/− (open bars) crypts charted. The error bars indicate S.E.M. values, _N_=3 Stat3 fl/− versus 3 Stat3 wt/− mice per time point. (b) Anti-Bmi1 immunohistochemistry from a Stat3 wt/− small intestine at 3 days after induction of Cre expression. The immunohistochemistry panel is representative of three Stat3 wt/− mice, the black bar represents 20 _μ_m. (c) Anti-Bmi1 immunohistochemistry from a Stat3 fl/− small intestine at 3 days after induction of Cre expression. The immunohistochemistry panel is representative of three Stat3 fl/− mice, the black bar represents 20 _μ_m

References

    1. Schindler C, Darnell JE. Transcriptional responses to polypeptide ligands – the JAK-STAT pathway. Ann Rev Biochem. 1995;64:621–651. - PubMed
    1. Zhong Z, Wen ZL, Darnell JE. STAT3 – a STAT family member activated by tyrosine phosphorylation in response to epidermal growth factor and interleukin-6. Science. 1994;264:95–98. - PubMed
    1. Yu H, Jove R. The stats of cancer – new molecular targets come of age. Nat Rev Cancer. 2004;4:97–105. - PubMed
    1. Haura EB, Turkson J, Jove R. Mechanisms of disease: insights into the emerging role of signal transducers and activators of transcription in cancer. Nat Clin Pract Oncol. 2005;2:315–324. - PubMed
    1. Niwa H, Burdon T, Chambers I, Smith A. Self-renewal of pluripotent embryonic stem cells is mediated via activation of STAT3. Genes Dev. 1998;12:2048–2060. - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources