Effects of forced expression of an NH2-terminal truncated beta-Catenin on mouse intestinal epithelial homeostasis

Abstract

beta-Catenin functions as a downstream component of the Wnt/Wingless signal transduction pathway and as an effector of cell-cell adhesion through its association with cadherins. To explore the in vivo effects of beta-catenin on proliferation, cell fate specification, adhesion, and migration in a mammalian epithelium, a human NH2-terminal truncation mutant (DeltaN89 beta-catenin) was expressed in the 129/Sv embryonic stem cell-derived component of the small intestine of adult C57Bl/6-ROSA26 left and right arrow 129/Sv chimeric mice. DeltaN89 beta-Catenin was chosen because mutants of this type are more stable than the wild-type protein, and phenocopy activation of the Wnt/Wingless signaling pathway in Xenopus and Drosophila. DeltaN89 beta-Catenin had several effects. Cell division was stimulated fourfold in undifferentiated cells located in the proliferative compartment of the intestine (crypts of Lieberkühn). The proliferative response was not associated with any discernible changes in cell fate specification but was accompanied by a three- to fourfold increase in crypt apoptosis. There was a marked augmentation of E-cadherin at the adherens junctions and basolateral surfaces of 129/Sv (DeltaN89 beta-catenin) intestinal epithelial cells and an accompanying slowing of cellular migration along crypt-villus units. 1-2% of 129/Sv (DeltaN89 beta-catenin) villi exhibited an abnormal branched architecture. Forced expression of DeltaN89 beta-catenin expression did not perturb the level or intracellular distribution of the tumor suppressor adenomatous polyposis coli (APC). The ability of DeltaN89 beta-catenin to interact with normal cellular pools of APC and/or augmented pools of E-cadherin may have helped prevent the 129/Sv gut epithelium from undergoing neoplastic transformation during the 10-mo period that animals were studied. Together, these in vivo studies emphasize the importance of beta-catenin in regulating normal adhesive and signaling functions within this epithelium.

Figure 1

Jejunum from normal B6-ROSA26↔ 129/Sv chimeric mice. (A) The jejunum has been opened, stained with X-Gal, and then photographed looking down on the tips of villi. The B6-ROSA26–derived epithelium expresses β-gal and appears blue after incubation with X-Gal. The 129/Sv component lacks β-gal (white). Closed arrow, wholly 129/Sv villus that receives cells from several monoclonal 129/Sv crypts positioned around its base. Open arrow, polyclonal villus. It appears striped because it receives columns of β-gal–negative (white) cells from monoclonal 129/Sv crypts and β-gal–positive cells from monoclonal B6-ROSA26 crypts. (B) Paraffin-embedded section prepared from the same wholemount and stained subsequently with nuclear fast red. The villus on the left is polyclonal: it is supplied by a monoclonal B6-ROSA26 crypt that contains an entirely β-gal–positive population of blue cells and by a monoclonal 129/ Sv crypt that only contains β-gal–negative cells. The villus on the right is only supplied by 129/Sv crypts. Bar, 25 μm.

Figure 1

Jejunum from normal B6-ROSA26↔ 129/Sv chimeric mice. (A) The jejunum has been opened, stained with X-Gal, and then photographed looking down on the tips of villi. The B6-ROSA26–derived epithelium expresses β-gal and appears blue after incubation with X-Gal. The 129/Sv component lacks β-gal (white). Closed arrow, wholly 129/Sv villus that receives cells from several monoclonal 129/Sv crypts positioned around its base. Open arrow, polyclonal villus. It appears striped because it receives columns of β-gal–negative (white) cells from monoclonal 129/Sv crypts and β-gal–positive cells from monoclonal B6-ROSA26 crypts. (B) Paraffin-embedded section prepared from the same wholemount and stained subsequently with nuclear fast red. The villus on the left is polyclonal: it is supplied by a monoclonal B6-ROSA26 crypt that contains an entirely β-gal–positive population of blue cells and by a monoclonal 129/ Sv crypt that only contains β-gal–negative cells. The villus on the right is only supplied by 129/Sv crypts. Bar, 25 μm.

Figure 2

Evidence for expression of the Fabpl–ΔN89β-catenin transgene. (A) RT-PCR analysis of RNAs isolated from the jejunum and skeletal muscle of a 6-wk-old-chimeric-transgenic mouse (ΔN89β-catenin) and from the jejunum of a 6-wk-old chimera generated using nontransfected ES cells (nl chimera). The arrow points to the expected size of the product generated from ΔN89β-catenin mRNA. (B) Duplicate immunoblots are shown each containing total cellular proteins isolated from the jejunum of a 6-wk-old normal control B6-ROSA26↔ 129/Sv chimeric mouse and a 6-wk-old B6-ROSA26↔ 129/Sv(ΔN89β-catenin) chimeric-transgenic animal (both with 30–40% 129/Sv contribution based on coat color). The blot on the left of the panel was probed with antibodies that recognize the myc epitope tag present at the NH₂ terminus of the 80-kD ΔN89β-catenin mutant (arrow). The blot on the right of the panel was probed with antibodies raised against the COOH terminus of β-catenin (molecular mass of wild-type β-catenin = 91 kD). (C) PLP-fixed frozen section of a polyclonal villus from a B6-ROSA26↔ 129/Sv(ΔN89β-catenin) mouse. The section was incubated with affinity-purified rabbit antibodies to the myc tag, Cy3-tagged donkey anti–rabbit Ig, and the nuclear stain bis-benzimide (dark blue). Myc-tagged ΔN89β-catenin (magenta) is prominently represented in 129/Sv but not B6-ROSA26 villus epithelial cells. (Genotyping was accomplished by staining an adjacent serial section with antibodies to β-gal.) (D) Section of a polyclonal villus from a normal chimeric mouse stained with the same reagents as in C. 129/Sv epithelial cells lack the transgene, and therefore do not contain any myc-tagged protein. Bars: (C and D) 25 μm.

Figure 2

Evidence for expression of the Fabpl–ΔN89β-catenin transgene. (A) RT-PCR analysis of RNAs isolated from the jejunum and skeletal muscle of a 6-wk-old-chimeric-transgenic mouse (ΔN89β-catenin) and from the jejunum of a 6-wk-old chimera generated using nontransfected ES cells (nl chimera). The arrow points to the expected size of the product generated from ΔN89β-catenin mRNA. (B) Duplicate immunoblots are shown each containing total cellular proteins isolated from the jejunum of a 6-wk-old normal control B6-ROSA26↔ 129/Sv chimeric mouse and a 6-wk-old B6-ROSA26↔ 129/Sv(ΔN89β-catenin) chimeric-transgenic animal (both with 30–40% 129/Sv contribution based on coat color). The blot on the left of the panel was probed with antibodies that recognize the myc epitope tag present at the NH₂ terminus of the 80-kD ΔN89β-catenin mutant (arrow). The blot on the right of the panel was probed with antibodies raised against the COOH terminus of β-catenin (molecular mass of wild-type β-catenin = 91 kD). (C) PLP-fixed frozen section of a polyclonal villus from a B6-ROSA26↔ 129/Sv(ΔN89β-catenin) mouse. The section was incubated with affinity-purified rabbit antibodies to the myc tag, Cy3-tagged donkey anti–rabbit Ig, and the nuclear stain bis-benzimide (dark blue). Myc-tagged ΔN89β-catenin (magenta) is prominently represented in 129/Sv but not B6-ROSA26 villus epithelial cells. (Genotyping was accomplished by staining an adjacent serial section with antibodies to β-gal.) (D) Section of a polyclonal villus from a normal chimeric mouse stained with the same reagents as in C. 129/Sv epithelial cells lack the transgene, and therefore do not contain any myc-tagged protein. Bars: (C and D) 25 μm.

Figure 2

Evidence for expression of the Fabpl–ΔN89β-catenin transgene. (A) RT-PCR analysis of RNAs isolated from the jejunum and skeletal muscle of a 6-wk-old-chimeric-transgenic mouse (ΔN89β-catenin) and from the jejunum of a 6-wk-old chimera generated using nontransfected ES cells (nl chimera). The arrow points to the expected size of the product generated from ΔN89β-catenin mRNA. (B) Duplicate immunoblots are shown each containing total cellular proteins isolated from the jejunum of a 6-wk-old normal control B6-ROSA26↔ 129/Sv chimeric mouse and a 6-wk-old B6-ROSA26↔ 129/Sv(ΔN89β-catenin) chimeric-transgenic animal (both with 30–40% 129/Sv contribution based on coat color). The blot on the left of the panel was probed with antibodies that recognize the myc epitope tag present at the NH₂ terminus of the 80-kD ΔN89β-catenin mutant (arrow). The blot on the right of the panel was probed with antibodies raised against the COOH terminus of β-catenin (molecular mass of wild-type β-catenin = 91 kD). (C) PLP-fixed frozen section of a polyclonal villus from a B6-ROSA26↔ 129/Sv(ΔN89β-catenin) mouse. The section was incubated with affinity-purified rabbit antibodies to the myc tag, Cy3-tagged donkey anti–rabbit Ig, and the nuclear stain bis-benzimide (dark blue). Myc-tagged ΔN89β-catenin (magenta) is prominently represented in 129/Sv but not B6-ROSA26 villus epithelial cells. (Genotyping was accomplished by staining an adjacent serial section with antibodies to β-gal.) (D) Section of a polyclonal villus from a normal chimeric mouse stained with the same reagents as in C. 129/Sv epithelial cells lack the transgene, and therefore do not contain any myc-tagged protein. Bars: (C and D) 25 μm.

Figure 3

Villus branching in 129/Sv(ΔN89β-catenin) epithelium. (A) Wholemount preparation of the jejunum from a 6-mo-old chimeric-transgenic mouse. Arrow, branched 129/Sv villus. (B) Section from the same wholemount stained with nuclear fast red. Each limb of the branched villus is equivalent in height. There are no obvious morphologic abnormalities in the crypts associated with this villus (arrows). (C) A branched polyclonal villus. Open arrow, a stripe of β-gal–positive (blue) B6-ROSA26 epithelium in one of the branches. The other branch is wholly 129/Sv (closed arrow; white). Inset, section of the branched villus incubated with rabbit antibodies to β-gal and Cy3-conjugated donkey anti–rabbit Ig. The stripe of B6-ROSA26 epithelium in one of the branches appears red. Closed arrow, the other, β-gal–negative, 129/Sv branch. Bars: (B and C) 25 μm.

Figure 4

Quantitation of villus branching. Branching was defined and quantitated as described in Materials and Methods. Branched villi were scored in the B6-ROSA26 and 129/Sv jejunal components of 6-mo-old normal chimeras (B6-ROSA26↔ 129/ Sv) and chimeric-transgenic (B6-ROSA26↔ 129/Sv[NΔ89β-catenin]) mice. Each line of chimeric-transgenic mice was derived from an independent clone of ES cells stably transfected with Fabpl–NΔ89β-catenin DNA. Mean values ±1 SD are plotted. Asterisk, the frequency of branched villi was <0.01%.

Figure 5

Forced expression of ΔN89β-catenin has no discernible effects on epithelial cell differentiation. Frozen sections were prepared from PLP-fixed jejunums of 6-mo-old chimeric-transgenic animals. (A) A polyclonal villus stained with biotin-conjugated Dolichos biflorus agglutinin (DBA), Cy3-conjugated avidin, rabbit anti–β-gal, and FITC-conjugated donkey anti–rabbit Ig. Glycoconjugates containing GalNAcα3GalNAc and GalNAcα3Gal recognized by DBA appear yellow-orange; β-gal appears green. The polarity and differentiation of enterocytes appears to be unaffected, as judged by the distribution of these glycoconjugates in apical membranes and supranuclear Golgi apparatus (closed arrow). Similarly, based on their reaction with DBA, the number and differentiation of goblet cells (open arrows) is equivalent in the 129/Sv and B6-ROSA26 components of the polyclonal villus. (B) The base of a polyclonal villus with its crypt-villus junction indicated by closed arrows. Three crypts are seen (open arrows at their base): the one on the left is supplying cells to another villus. The section was incubated with rat anti-β₄ integrin subunit, Cy3 donkey anti–rat Ig, rabbit anti–β-gal, FITC donkey anti–rabbit Ig, and bis-benzimide. Nuclei (blue); the β₄ integrin subunit (orange); β-gal (green-brown). The location of β₄ integrin at the base of epithelial cells and its distribution along the crypt-villus unit are unaffected by ΔN89β-catenin. (C) Villi sectioned perpendicular to their crypt-villus axis. The tight junction protein ZO-1 (orange) was detected with rat anti-ZO-1 and Cy3 donkey anti–rat Ig. β-Gal (green) was visualized with the same reagents used in the preceding sections. The levels and location of ZO-1 in the 129/Sv(ΔN89β-catenin) and B6-ROSA26 components of polyclonal villi are similar (e.g., open arrows). (D) Villi sectioned perpendicular to their crypt-villus axis as in C. The section was incubated with rat anti-β₇ integrin, Cy3-donkey anti–rat Ig, rabbit anti-laminin, rabbit anti–β-gal, FITC donkey anti–rabbit Ig, and bis-benzimide. B6-ROSA26 cells exhibit diffuse staining of their cytoplasm due to the presence of β-gal (green). β₇ integrin is confined to intraepithelial lymphocytes (orange). Comparable numbers of these cells are seen in the B6-ROSA26 and 129/ Sv(ΔN89β-catenin) components of polyclonal villi and in wholly 129/Sv(ΔN89β-catenin) villi. Laminin appears as linear green immunoreactivity underlying 129/Sv and B6-ROSA26 epithelium (e.g., closed arrows). The intensity of staining is similar under cells of both genotypes. Bars, 25 μm.

Figure 6

ΔN89β-catenin stimulates proliferation and apoptosis in 129/Sv crypts. (A) The number of M-phase cells in juxtaposed B6-ROSA26 and 129/Sv jejunal crypt sections were scored in 6-mo-old normal chimeras, in two lines of chimeric-transgenic mice generated using separate ES cell clones stably transfected with Fabpl–ΔN89β-catenin DNA, and in two lines of chimeric-transgenic mice produced from independent ES cells clones stably transfected with Fabpl-mouse E-cadherin DNA (Hermiston et al., 1996). Each mouse served as its own control: the average number of M-phase cells in their 129/Sv crypts were divided by the average number of M-phase cells in their B6-ROSA26 crypts. Mean values ±1 SD obtained from animals in each group are plotted except for line 2 of B6↔ 129/Sv (E-cadherin) chimeras where only two animals were examined. Note that the stimulation of crypt proliferation is similar in mice belonging to a given line of B6-ROSA26↔ 129/Sv(ΔN89β-catenin) chimeras. Differences between lines produced from different Fabpl–ΔN89β-catenin ES cell clones were attributed to differences in the levels of expression of this transgene. (B) Apoptotic cells were scored in jejunal sections prepared from the same animals used to define the mitotic index in A.

Figure 7

Forced expression of ΔN89β-catenin is associated with a slowing of epithelial migration. A 6-mo-old chimeric-transgenic mouse was pulse labeled with BrdU 60 h before killing. Left, a section stained with rabbit anti–β-gal and Cy3 donkey anti–rabbit Ig. The villus encompassed by the box is polyclonal. Right, an adjacent section of the polyclonal villus, stained with goat anti-BrdU, Cy3 donkey anti–goat Ig, and bis-benzimide. The leading edge of the column of BrdU–positive 129/Sv(ΔN89β-catenin) epithelial cells (magenta-colored nuclei) has only reached the middle portion of the villus (arrow), whereas the leading edge of the BrdU–positive B6-ROSA26 column is already near the villus tip. Bars, 25 μm.

Figure 8

ΔN89β-Catenin expression results in augmented cellular pools of E-cadherin. Frozen sections were prepared from PLP-fixed jejunum recovered from a 6-mo-old B6-ROSA26↔ 129/ Sv(ΔN89β-catenin) chimeric-transgenic animal (A–D) and a 6-mo-old normal B6-ROSA26↔ 129/Sv chimera (E and F). (A) Polyclonal villus incubated with rabbit anti–β-catenin, Cy3 donkey anti–rabbit Ig, and bis-benzimide. Prominent β-catenin staining (red) is evident at the adherens junctions and basolateral surfaces of epithelial cells. No immunoreactive protein is detectable in 129/Sv or B6-ROSA26 nuclei. (B) Polyclonal villus incubated with rabbit anti-APC, Cy3 donkey anti–rabbit Ig, and bis-benzimide. The levels and intracellular distribution of APC (red) are similar in B6-ROSA26 and 129/Sv(ΔN89β-catenin) epithelial cells. (C and D) Villi from the chimeric-transgenic mouse that have been sectioned perpendicular their crypt-villus axis. (C) Section stained with rat anti–E-cadherin and Cy3-donkey anti– rat Ig. (D) Dual exposure of the same section after incubation with rabbit anti–β-gal and FITC donkey anti–rabbit Ig. Steady-state levels of E-cadherin are markedly increased in the 129/ Sv(ΔN89β-catenin) component of polyclonal villi. (E and F) Villi from a control normal chimera processed as in (C and D). E-Cadherin levels are comparable in the B6-ROSA26 and 129/Sv villus epithelium. Bars, 25 μm.