Original Article

Oncogene (2012) 31, 2423–2437; doi:10.1038/onc.2011.434; published online 26 September 2011

A knock-in mouse model reveals roles for nuclear Apc in cell proliferation, Wnt signal inhibition and tumor suppression

M Zeineldin1, J Cunningham1, W McGuinness1, P Alltizer1, B Cowley2, B Blanchat1, W Xu3, D Pinson4 and K L Neufeld1

  1. 1Department of Molecular Biosciences, University of Kansas, Lawrence, KS, USA
  2. 2Eccles Institute of Human Genetics, University of Utah, Salt Lake City, UT, USA
  3. 3Department of Microbiology, University of Virginia, Charlottesville, VA, USA
  4. 4Department of Pathology and Laboratory Medicine, University of Kansas Medical Center, Kansas City, KS, USA

Correspondence: Dr KL Neufeld, Department of Molecular Biosciences, University of Kansas, 7049 Haworth Hall, 1200 Sunnyside Ave., Lawrence, KS 66045, USA. E-mail: klneuf@ku.edu

Received 5 December 2010; Revised 23 August 2011; Accepted 24 August 2011
Advance online publication 26 September 2011



Mutation of the tumor suppressor adenomatous polyposis coli (APC) is considered an initiating step in the genesis of the vast majority of colorectal cancers. APC inhibits the Wnt-signaling pathway by targeting the proto-oncogene β-catenin for destruction by cytoplasmic proteasomes. In the presence of a Wnt signal, or in the absence of functional APC, β-catenin can serve as a transcription cofactor for genes required for cell proliferation such as cyclin-D1 and c-Myc. In cultured cells, APC shuttles between the nucleus and the cytoplasm, with nuclear APC implicated in the inhibition of Wnt target gene expression. Adopting a genetic approach to evaluate the functions of nuclear APC in the context of a whole organism, we generated a mouse model with mutations that inactivate the nuclear localization signals (NLSs) of Apc (ApcmNLS). ApcmNLS/mNLS mice are viable and fractionation of mouse embryonic fibroblasts (MEFs) isolated from these mice revealed a significant reduction in nuclear Apc as compared with Apc+/+ MEFs. The levels of Apc and β-catenin protein were not significantly altered in small intestinal epithelia from ApcmNLS/mNLS mice. Compared with Apc+/+ mice, ApcmNLS/mNLS mice showed increased proliferation in epithelial cells from the jejunum, ileum and colon. These same tissues from ApcmNLS/mNLS mice showed more mRNA from three genes upregulated in response to canonical Wnt signal, c-Myc, axin-2 and cyclin-D1, and less mRNA from Hath-1, which is downregulated in response to Wnt. These observations suggest a role for nuclear Apc in the inhibition of canonical Wnt signaling and the control of epithelial proliferation in intestinal tissue. Furthermore, we found ApcMin/+ mice, which harbor a mutation that truncates Apc, to have an increased polyp size and multiplicity if they also carry the ApcmNLS allele. Taken together, this analysis of the novel ApcmNLS mouse model supports a role for nuclear Apc in the control of Wnt target genes, intestinal epithelial cell proliferation and polyp formation.


adenomatous polyposis coli; polyp; Wnt; nuclear; mouse model; NLS



The tumor-suppressor protein adenomatous polyposis coli (APC) is large, with multiple subcellular localizations and functions. Mutation of the APC gene is considered the initiating event in the formation of most intestinal polyps, the precursors to colorectal cancer (Miyoshi et al., 1992a, 1992b; Smith et al., 1993). As such, an intense investigation of potential APC functions involved in tumor suppression has led to the identification of APC as a Wnt-signaling pathway antagonist (Giles et al., 2003). In this capacity, APC is part of a cytoplasmic protein complex that targets the proto-oncoprotein β-catenin for proteasome-mediated destruction (Munemitsu et al., 1995). APC has also been observed in the nuclei of cultured colon cells and in colonic epithelia from human tissue by conventional and confocal immunofluorescence microscopy and immunoelectron microscopy (Neufeld and White, 1997; Anderson et al., 2002; Sena et al., 2006). Nuclear APC has three proposed roles in regulating Wnt signaling. First, nuclear APC binds to nuclear β-catenin and likely competes with the transcription factor T-cell factor/ lymphocyte enhancer factor (TCF)/(LEF) for β-catenin binding (Neufeld et al., 2000). Second, nuclear APC has been implicated in the nuclear export of β-catenin (Henderson, 2000; Neufeld et al., 2000; Rosin-Arbfeld et al., 2000). Finally, the interaction of nuclear APC with the transcriptional co-repressor C-terminal binding protein further contributes to the modulation of Wnt signaling (Sierra et al., 2006). In addition to its role as a repressor of β-catenin, nuclear APC has been implicated in DNA synthesis and repair (Jaiswal and Narayan, 2008; Qian et al., 2008).

Because APC is a large protein (~310kDa), it is unable to simply diffuse into the nucleus but must instead be actively transported through nuclear pores (Peters, 1986; Mattaj and Englmeier, 1998). Several domains of APC have been proposed to mediate this nuclear entry, including two monopartite nuclear localization signals (NLSs) in the central part of the protein (Zhang et al., 2000), as well as an armadillo repeat region closer to the N-terminus (Galea et al., 2001). In addition, phosphorylation of APC regulates its localization (Zhang et al., 2000) and relative levels of nuclear and cytoplasmic APC appear to correlate with the proliferative status of a cell (Zhang et al., 2001).

Although many mouse models have been generated with alterations in Apc, they either produce no Apc protein (Cheung et al., 2009), full-length Apc at a reduced level (Ishikawa et al., 2003) or a truncated Apc (Moser et al., 1993; Fodde et al., 1994; Oshima et al., 1995; Kan et al., 1997; Smits et al., 1999; Sasai et al., 2000; Colnot et al., 2004). Elimination of the C-terminal portion of Apc leads to at least partial loss of function and mice homozygous for mutations that truncate Apc typically do not develop beyond embryonic day 8 (Taketo, 2006). To date, no mouse model with targeted disruption of the two Apc NLSs has been generated.

To assess the functions of nuclear Apc in the context of a whole organism, we introduced germline mutations into the mouse Apc gene that result in the inactivation of the two characterized Apc NLSs (ApcmNLS). Inactivation of the two APC NLSs was previously shown to attenuate the ability of the full-length exogenous human APC to enter the nucleus (Zhang et al., 2000). In the present study, mice heterozygous (ApcmNLS/+) or homozygous (ApcmNLS/mNLS) for mutant Apc NLS were viable. Intestinal epithelial cells in ApcmNLS/mNLS mice showed increased levels of Wnt target gene expression and more proliferation compared with their wild-type littermates. For the past two decades, the ApcMin/+ mouse has been a popular model organism to study Apc mutation-driven tumorigenesis (Moser et al., 1990). ApcMin/+ mice have a missense mutation in Apc that results in the truncation of the C-terminal 2/3 of the protein. ApcMin/+ mice develop 20–60 intestinal polyps at the time of their natural death, ~120 days of age. We found that ApcMin/mNLS mice showed significantly more intestinal polyps than ApcMin/+ mice and these polyps were larger. Together, these results suggest that nuclear Apc participates in the regulation of Wnt signaling, proliferation and tumor suppression.



Mutation of two Apc NLSs in ES cells and generation of mutant mice

To introduce specific mutations that inactivate both Apc NLSs in mouse embryonic stem (ES) cells and ultimately generate ApcmNLS/mNLS mice, we constructed a gene-targeting vector (Figure 1a). The targeted mutations alter a total of six amino acids in Apc, four in NLS1 and two in NLS2 (Figure 1b). These mutations substitute a neutral alanine for the basic lysine residues in the monopartite NLSs and have been shown to inhibit the nuclear import of exogenously expressed APC (Zhang et al., 2000). Each NLS is adjacent to a SAMP repeat region (so named because they each contain a central Ser–Ala–Met–Pro), which is involved in the binding of APC to axin. However, mutation of both NLSs does not appear to interfere with axin binding as full-length APC with mutations in both NLSs still interacts with axin (Supplementary Figure S1). These mutations also establish two novel restriction sites, SstII in NLS1 and EagI, in NLS2, which were used for screening purposes (Figure 1b). The targeting vector contains an 11.5-kb genomic NheI/NotI fragment encompassing exons 14 and 15 of Apc as well as the surrounding introns. The entire Apc region of the vector was validated by sequencing. The positive selection marker contained in a germline-induced self-excision cassette was placed in the intron after Apc exon-15 (Bunting et al., 1999). In mice, the testes-specific murine angiotensin-converting enzyme (tACE) promoter initiates the transcription of Cre-recombinase during spermatogenesis, inducing Cre-mediated self-excision of the selectable Neor marker in the germline of the transgenic animals. Following self-excision, only one 34-bp minimal LoxP element remains in the last Apc intron. The targeting construct also contains a copy of the herpes simplex virus (HSV) thymidine kinase gene (TkHSV) to allow negative selection of homologous recombinant ES clones.

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Generation of ApcmNLS/+ mouse ES cell lines. (a) A schematic representation of Apc NLS targeting vector. Drawn roughly to scale, the vector includes Apc exons 14 and 15 (dark blue), with surrounding introns (light blue). A neomycin resistance gene (Neor purple) inserted in the final Apc intron is flanked by LoxP sites (green) that will facilitate its excision when expressed in the testes where the tACE promoter (murine angiotensin-converting enzyme) drives the expression of Cre recombinase. The HSV thymidine kinase gene (TkHSV) allows the negative selection of homologous recombinant ES cell clones. (b) Mutations in Apc NLS1 and NLS2 result in amino-acid substitutions inactivating the NLSs and also introduce unique restriction enzyme sites for screening purposes. (c) Analysis of ES cell lines by PCR using primers specific for the wild-type (wt) or mutant (m) Apc NLS allowed identification of 107 lines with both Apc NLS1 and NLS2 mutated. (d) Samples 2 and 4 in the gel image shown each have both Apc NLSs mutated. PCR without template DNA (neg). (e) Excision of the Ace-Cre-Neor selection cassette was verified in N2 ApcmNLS/+ mice by PCR using primers specific for Cre. Genomic DNA isolated from wild-type (WT) and ApcmNLS/+ ES cell lines was used as negative and positive control, respectively, for the Cre primer PCR (Cr). PCR using primers specific for wild-type sequence (wt) and mutant sequence (m) confirmed that the mouse analyzed was ApcmNLS/+ and demonstrated genomic DNA quality sufficient for PCR analysis. (f) Scheme to determine correct integration of the 5′- and 3′-ends of the targeting vector. (g) Correct integration of the 5′-end of the targeting vector was established by PCR analysis using a primer specific for mutant Apc NLS2, with a second primer that recognizes genomic sequence outside the targeting construct and upstream from a correctly integrated vector. Products of this PCR were digested with SstII to confirm integration of mutant Apc NLS1. (h) PCR analysis using a primer unique to the targeting construct, with a second primer that recognizes genomic sequence outside the targeting construct and downstream from a properly integrated vector, allowed identification of eight recombinant ES cell lines that had correctly integrated the 3′-end of the targeting vector (samples 1 and 2). TC, targeting vector control product is ~1800bp if targeting vector is integrated into genomic DNA; AI, Apc integrated product is ~2500bp if targeting vector is properly inserted into the Apc gene.

Full figure and legend (249K)

Homologous recombinant ES clones were identified by PCR using different primer pairs inside and outside the targeting construct. Of the 161 ES cell clones that survived positive and negative selection, 107 contained Apc with both mutant NLSs (Figures 1c and d). Eight of these ES cell clones were identified as homologous recombinants by PCR to verify correct insertion of the 3′-region of the vector (Figures 1f and h). Two were further analyzed by PCR and then subjected to restriction digestion in order to verify correct insertion of the 5′-region of the vector and to confirm that both mutant NLSs were integrated (Figures 1f and g). One ES line with normal karyotype in 75% of the 50 cells analyzed was used for the initial blastocyst injections. From these injections, 11 chimeric mice, 4 females and 7 males were obtained; three males showed evidence of germline transmission when bred with wild-type C57BL/6J mice.

Elimination of the Neor/Cre selection cassette in ApcmNLS/+ mice

In the Apc1638N mouse model, a Neor selection gene inserted in exon-15 of Apc resulted in a dramatic decrease in the expression of the mutant Apc (Fodde et al., 1994; Kielman et al., 2002). Thus it was critical to ensure that the Neor gene was excised from the ApcmNLS/+ mice. Excision of the Neor cassette in animals was verified by PCR analysis using oligos within the Neor/Cre selection cassette to prime the DNA amplification of the genomic DNA isolated from ApcmNLS/+ mice in the N1 and N2 generations (Figure 1e).

ApcmNLS/mNLS mice are viable

Many of the previously generated Apc mouse models show early embryonic lethality as homozygous mutants. When ApcmNLS/+ mice were interbred, both ApcmNLS/+ and ApcmNLS/mNLS mice were obtained. Thus, the two characterized monopartite NLSs of Apc are not essential for viability. ApcmNLS/mNLS mice showed no obvious growth defects compared with their wild-type littermates in generations N1–N13. ApcmNLS/mNLS mice are fertile, giving birth to litters of typical size. A long-term survival analysis of 55 progeny from first-generation ApcmNLS/+ females bred to first-generation ApcmNLS/+ males revealed no significant difference in mouse survival between ApcmNLS/mNLS and Apc+/+ mice (data not shown). Thus, it appears that mutation of Apc NLS sequence has limited impact on the lifespan of the mice.

Reduced nuclear Apc in embryonic fibroblasts isolated from ApcmNLS/mNLS mice

Because of complications associated with fractionation of cells obtained from intestinal tissue, we used mouse embryonic fibroblasts (MEFs) isolated from congenic (generation N11) mice to determine the effects of ApcmNLS on subcellular Apc localization (Figure 2a). In Apc+/+ MEFs, more than one-third of the total Apc was associated with the nuclear fraction (Figure 2b). ApcmNLS/mNLS MEFs were significantly compromised for nuclear Apc distribution, with 11% of the total Apc associated with the nuclear fraction. ApcmNLS/+ MEFs showed an intermediate phenotype, but nuclear Apc was not significantly different from that in Apc+/+ MEFs. β-Catenin distribution was similar in all MEF lines (Figure 2c). Moreover, β-catenin appeared predominantly at cell–cell junctions in intestinal tissue from Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS mice as assessed by immunohistochemistry (Supplementary Figure S2a). A few crypts showed limited nuclear β-catenin staining in cells near their base, but this was consistent for all mouse genotypes (Supplementary Figure S2b).

Figure 2.
Figure 2 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Reduced nuclear Apc levels in MEFs from ApcmNLS/mNLS mice. (a) MEFs isolated from congenic Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS mice were subjected to fractionation followed by immunoblotting. A single blot, probed for Apc, β-catenin, tubulin (cytoplasmic marker) and fibrillarin (nuclear marker) is shown. (b) Apc band intensities were determined and results from five independent fractionation experiments performed on three different isolations of MEF cells are presented as the fraction of the total Apc protein in the nucleus±s.e.m. ApcmNLS/mNLS MEFs showed significantly less nuclear Apc than Apc+/+ MEFs (P=0.012) as indicated by asterisk. (c) β-Catenin distribution, determined as in panel b, was similar in all MEF lines.

Full figure and legend (78K)

Apc and β-catenin levels in intestinal tissue

To determine whether the single LoxP site remaining in the last Apc intron impacts Apc level in ApcmNLS mice, we measured the relative amount of Apc in whole-cell protein lysates prepared from epithelial cells isolated from mouse jejunum, ileum and colon (Figure 3). The Apc level did not vary significantly in epithelia from jejunum or ileum tissue isolated from ApcmNLS/mNLS ApcmNLS/+ and Apc+/+ mice. Unexpectedly, colon epithelia from ApcmNLS/mNLS mice had higher levels of Apc than colon epithelia from Apc+/+ mice. Because Apc targets the proto-oncoprotein β-catenin for destruction, higher Apc levels would be expected to result in reduced β-catenin levels. Unexpectedly, colon tissue from ApcmNLS/mNLS mice showed higher levels of β-catenin than colon tissue from Apc+/+ mice. In the small intestine, β-catenin levels were not altered in ApcmNLS/mNLS mice as compared with Apc+/+ mice.

Figure 3.
Figure 3 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Apc and β-catenin levels in intestinal epithelia from ApcmNLS/+ and ApcmNLS/mNLS mice. Epithelial cells were isolated from three intestinal segments (jejunum, ileum and colon) of congenic Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS mice. Proteins from whole-cell lysates were resolved by sodium dodecyl sulfate–PAGE. Immunoblots were probed for Apc, β-catenin and β-actin, which served as a loading control (top panels). Band intensities were determined for four samples from each genotype and are presented as average band intensity±s.e.m. relative to β-actin and normalized to the Apc+/+ sample, which was set to 1 (middle and bottom panels). Apc and β-catenin levels appeared comparable in epithelial cells isolated from the jejunum and ileum of mice of each genotype. Using Mann–Whitney non-parametric test, colon tissue from ApcmNLS/mNLS mice showed significantly higher levels of both Apc (P=0.017) and β-catenin (P=0.01) than colon tissue from Apc+/+ mice. Significant values indicated by asterisks.

Full figure and legend (140K)

Increased expression of Wnt targets in intestinal epithelia of ApcmNLS/mNLS mice

Based on previous studies performed in cultured cells, we proposed that nuclear APC binds to nuclear β-catenin and sequesters it from transcription factor TCF/LEF. Consequently, Wnt target genes activated by the β-catenin/TCF/LEF complex are downregulated by nuclear APC (Neufeld et al., 2000). Thus we predicted that cells defective in nuclear Apc would be less able to dampen Wnt target gene expression. To test this prediction, mRNA isolated from intestinal epithelial cells was evaluated for relative expression of various Wnt-regulated genes (Figure 4). Three genes typically upregulated in response to a canonical Wnt signal, c-Myc, axin-2 and cyclin-D1, showed higher expression throughout the intestinal epithelia of ApcmNLS/mNLS mice as compared with Apc+/+ mice. In these same tissues, Hath-1, a gene downregulated in response to a canonical Wnt signal, showed lower expression throughout the intestinal epithelia of ApcmNLS/mNLS mice as compared with Apc+/+ mice. BTEB2, a gene reportedly upregulated in response to non-canonical Wnt signal, showed no expression alterations in jejunum or ileum from ApcmNLS/mNLS mice as compared with Apc+/+ mice. Unexpectedly, in colon tissue, BTEB2 expression was higher in both ApcmNLS/mNLS and ApcmNLS/+ mice as compared with Apc+/+ mice.

Figure 4.
Figure 4 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Elevation of Wnt target gene expression in ApcmNLS/mNLS mice. Epithelial cells were isolated from three intestinal segments (jejunum, ileum and colon) of congenic Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS mice. For each sample, the mRNA levels of three genes upregulated by canonical Wnt signaling (c-Myc, axin-2 and cyclin-D1), one gene, which is downregulated by canonical Wnt signaling (Hath-1), and one gene, which is not regulated by canonical Wnt signaling (BTEB2), each normalized to HGPRT (housekeeping gene, control), were determined by real-time quantitative reverse transcription–PCR. Results from 4–6 mice are presented as average mRNA level relative to that found in the Apc+/+ sample±s.e.m. P-values <0.05, as calculated by Mann–Whitney non-parametric test, are indicated by asterisk. The levels of Hath-1 mRNA in jejunum samples from ApcmNLS/mNLS mice were around the lower limit of detection, precluding calculation of ΔΔC(t) for some samples. Significant values indicated by asterisks.

Full figure and legend (132K)

Increased proliferation in intestinal epithelia of ApcmNLS/mNLS mice

Given the increased expression of c-Myc and cyclin-D1 in the intestines of ApcmNLS/mNLS mice, we expected to find an accompanying increase in proliferation. To evaluate epithelial cell proliferation in intestinal tissue, mice were analyzed 4h after injection with the thymidine analogue ethynyl deoxyuridine (EdU). Crypts from jejunum, ileum and colon each showed significantly more proliferating cells in ApcmNLS/mNLS mice than in ApcmNLS/+ and Apc+/+ mice (Figure 5).

Figure 5.
Figure 5 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Increased epithelial cell proliferation in intestines of ApcmNLS/mNLS mice. Representative images of crypts from EdU-labeled intestinal tissue isolated from Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS mice, with the white line indicating crypt border (left panels). 4′,6-Diamidino-2-phenylindole (Dapi) staining allowed identification of cell nuclei. Scale bar: 10μm. Right panels: The average number of EdU-positive cells per crypt cross-section normalized to the total crypt cell number is presented, with error bars indicating s.e.m. Samples were collected from three mice of each genotype. Top panels: Jejunum tissues showed a significant increase in proliferation in ApcmNLS/mNLS mice as compared with Apc+/+ mice (P<0.0001). Middle panels: Ileum tissues showed a significant increase in proliferation in ApcmNLS/mNLS mice as compared with Apc+/+ mice (P<0.05). Bottom panels: Colon tissues showed a significant increase in proliferation in ApcmNLS/mNLS mice as compared with Apc+/+ mice (P<0.0001). Significant values indicated by asterisks.

Full figure and legend (276K)

Enhanced polyp formation in intestines of ApcMin/mNLS mice

Mice with germline truncating mutations in Apc occasionally show a few adenomatous polyps in the colon, with the great majority of polyps found in the small intestine (Taketo, 2006). Therefore, we targeted both the colon and the small intestine for our initial phenotypic analysis. No adenomatous polyps of the small intestine were identified in any of the ApcmNLS/+ mice from generations N1–N4 analyzed for intestinal lesions (n=33). One dysplastic colon polyp was found in an ApcmNLS/mNLS mouse (Figures 6a–d). A polyp was also found in the stomach of another ApcmNLS/mNLS mouse (generation N11).

Figure 6.
Figure 6 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Polyps in ApcmNLS/mNLS mice. (a) Low-power magnification of a lesion in the colon of ApcmNLS/mNLS mouse. Scale bar: 1mm. (b) The same lesion, with a higher magnification view of the region outlined by the white box in panel a showing epithelial atypia. Nuclei are enlarged with loss of polarization. Scale bar: 100μm. (c) Higher power magnification of the region with cell atypia shown in panel b, showing enlarged nuclei. Scale bar: 25μm. (d) A progression of atypia is shown in the same tissue (approximate area shown in black box in panel a) from lower right, with more polarized nuclei, to upper left, with enlarged, non-polarized nuclei. A representative dysplastic crypt is outlined in white. Scale bar: 50μm.

Full figure and legend (473K)

The few intestinal lesions observed in the ApcmNLS/mNLS mice left open the possibility that nuclear Apc participates in polyp suppression. To examine this more efficiently, congenic ApcmNLS/+ mice were bred with ApcMin/+ mice. Polyp number, size and distribution were compared in 14-week-old progeny ApcMin/+ and ApcMin/mNLS mice. Significantly more polyps were found in the jejunum and ileum of ApcMin/mNLS mice than in ApcMin/+ mice (Figure 7A and Supplementary Figure S3). There were only a few polyps observed in any colon tissue with no significant variation between ApcMin/+ and ApcMin/mNLS mice (Figure 7B and Supplementary Figure S3). Even fewer polyps were observed in the stomach and duodenum tissues, but the average polyp number was slightly larger in ApcMin/mNLS mice than in ApcMin/+ mice (Figure 7B and Supplementary Figure S3). Combining results from the entire gastrointestinal tissue, we observed nearly twice as many polyps in ApcMin/mNLS mice than in ApcMin/+ mice (Figure 7C). These results suggest that the ApcmNLS enhances polyp initiation or development in the ApcMin mouse model. Furthermore, the average polyp size was significantly larger in the jejunum and ileum tissue of ApcMin/mNLS mice as compared with ApcMin/+ mice (Figure 7D). Collectively, the results from this study suggest that nuclear Apc contributes to a tumor-suppressor phenotype.

Figure 7.
Figure 7 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

ApcMin/mNLS mice have more and larger polyps than ApcMin/+ mice. Polyps were identified and measured in 14-week-old ApcMin/+ (n=12) and ApcMin/mNLS (n=8) mice. (A) Results presented as average polyp number per mouse±s.e.m. indicate significantly more polyps in the jejunum (P=0.035) and ileum (P=0.0002) of ApcMin/mNLS mice as compared with ApcMin/+ mice. (B) Polyps were also identified in the colon, stomach and duodenum, but showed no statistically significant differences between ApcMin/mNLS and ApcMin/+ mice. (C) Results from all gastrointestinal tissues combined (jejunum, ileum, colon, stomach and duodenum) indicate significantly more polyps in ApcMin/mNLS as compared with ApcMin/+ mice (P=0.0021). (D) Diameters of all polyps are shown as a box-and-whisker plot. Polyps in both the jejunum and ileum were significantly larger in ApcMin/mNLS mice than in ApcMin/+ mice (P<0.0001). (E) Significantly more proliferation, as determined by Ki-67 expression, was observed in polyps isolated from the ileums of ApcMin/mNLS mice than from ApcMin/+ mice (P=0.0003). (F) A tissue section of a polyp from ApcMin/+ or ApcMin/mNLS mice stained for β-catenin (a–d), with a subsequent section stained for Ki-67 (e–h). The scale bar in panel a measures 200μm and also corresponds to (c), (e) and (g). Panels (b), (d), (f) and (h) show higher power magnifications of the regions indicated by black boxes in the previous panels, with a 50-μm scale bar shown in panel b. Significant values indicated by asterisks.

Full figure and legend (591K)

Increased proliferation in intestinal polyps from ApcMin/mNLS mice

To begin to explore the mechanism of enhanced polyp formation in ApcMin/mNLS mice, we examined cellular proliferation and β-catenin distribution in polyps from ApcMin/mNLS and ApcMin/+ mice. Strong nuclear β-catenin staining was observed in polyps from both ApcMin/mNLS and ApcMin/+ mice (Figures 7F, a–d). By contrast, β-catenin appeared predominantly at the cell–cell junctions of normal intestinal cells from both ApcMin/mNLS and ApcMin/+ mice. Although nuclear β-catenin was prevalent in both ApcMin/mNLS and ApcMin/+ polyps, not every cell with nuclear β-catenin was positive for the proliferation marker Ki-67 (Figures 7F, e–h). Closer examination of Ki-67-positive cells revealed that polyps from ApcMin/mNLS mice had significantly more proliferating cells than polyps from ApcMin/+ mice (Figure 7E). Combined with the observation that there were more polyps and these polyps were larger in the intestines of the ApcMin/mNLS mice than in the ApcMin/+ mice (Figures 7A–D), it is likely that contributions of nuclear Apc to regulation of cellular proliferation impact tumor suppression in this newly developed model.



Nuclear APC has been observed in cultured cells, in cells of various model organisms and in human tissues (Neufeld, 2009). Roles for nuclear APC in DNA repair and replication, as well as Wnt signal regulation, have been proposed (Henderson, 2000; Neufeld et al., 2000; Rosin-Arbesfeld et al., 2003; Sierra et al., 2006; Jaiswal and Narayan, 2008; Qian et al., 2008). However, to date, experimental manipulation of nuclear APC has been confined to cultured cells. In this report, we describe a new mouse model with mutations ‘knocked-in’ the Apc gene, which inactivate the two canonical Apc NLSs (Zhang et al., 2000). To our knowledge, the ApcmNLS/mNLS model is the first mouse generated to specifically inhibit the nuclear import of any protein by knock-in mutations that target and inactivate an NLS. ApcmNLS/mNLS mice were viable and showed increased proliferation of epithelia throughout the small intestine and the colon. Epithelia from jejunum, ileum and colon also showed increased canonical Wnt signaling as evidenced by increased expression of genes upregulated by canonical Wnt signal (cyclin-D1, c-myc and axin-2) and decreased expression of genes downregulated by canonical Wnt signal (Hath-1). This increased Wnt signaling could explain the enhanced proliferation observed in these same tissues. Although only occasionally observed in ApcmNLS/mNLS mice, polyps in jejunum and ileum were significantly more abundant and larger, with increased proliferative index in ApcMin/mNLS mice as compared with ApcMin/+ mice, implicating nuclear Apc in suppression of polyp formation.

The best characterized role for nuclear APC involves negative regulation of Wnt signaling (Neufeld et al., 2000; Sierra et al., 2006). Based on studies performed in cultured cells, it was proposed that nuclear APC binds to nuclear β-catenin, sequestering it from a complex with activated TCF/LEF (Neufeld et al., 2000) and mediating β-catenin export to the cytoplasm (Henderson, 2000; Neufeld et al., 2000; Rosin-Arbfeld et al., 2000). There is also evidence that APC can interact with the transcriptional co-repressor C-terminal binding protein (Sierra et al., 2006), resulting in the inhibition of TCF/LEF-mediated transcription and dampening of a Wnt signal (Neufeld, 2009). Our data indicate that Wnt signaling is upregulated in ApcmNLS/mNLS mice, supporting a role for nuclear Apc as a negative regulator of Wnt signaling. Of note, in this study, the alterations in Wnt gene expression occurred without increased β-catenin levels. This finding supports an alternate mechanism that does not require destruction of β-catenin for Apc to elicit the dampening of a Wnt signal. We also found that cells from ApcmNLS/mNLS mice had no increase in nuclear β-catenin, consistent with a role for nuclear Apc in the transport of transcription repressors to the β-catenin/TCF/LEF complex and/or sequestration of nuclear β-catenin. However, as cells from ApcmNLS/mNLS mice showed some nuclear Apc, we cannot fully exclude the possibility that Apc also has a role in the nuclear export of β-catenin.

Some Wnt target genes such as c-Myc and cyclin-D1 are associated with cell-cycle progression (He et al., 1998; Tetsu and McCormick, 1999). If nuclear Apc downregulates Wnt signaling, then reduced nuclear Apc should lead to enhanced TCF/LEF-mediated transcription, increased expression of c-Myc and cyclin-D1, and, consequently, more proliferation. Our finding that the intestinal epithelia of ApcmNLS/mNLS mice show increased Wnt signaling, including upregulation of c-Myc and cyclin-D1, and increased proliferation further supports a role for nuclear Apc in the control of cell-cycle progression.

The mutations introduced into Apc were minimal, and we expected that the phenotype of ApcmNLS/mNLS mice would be more subtle than that of previous Apc mouse models, which express truncated Apc proteins. The combined six amino-acid substitutions in Apc NLS1 and NLS2 do not appear to impact the binding of Apc to β-catenin or axin (Supplementary Figure S1). The Apc and β-catenin levels in epithelia isolated from either jejunum or ileum were similar in ApcmNLS/mNLS and Apc+/+ mice (Figure 3). Therefore, the alteration in the expression of canonical Wnt targets seen in these tissues did not likely result from compromised β-catenin degradation and it is likely that the observed phenotypes of ApcmNLS/mNLS mice result from compromised nuclear import of Apc.

Nuclear Apc entry was severely compromised, but not entirely eliminated in MEFs from ApcmNLS/mNLS mice (Figure 2). The two Apc NLSs that were inactivated in our mouse model are each classic monopartite NLSs, which bind to importin-α in the cytoplasm to target protein transport through the nuclear pore (Dingwall et al., 1982; Gorlich et al., 1994). The continued presence of some Apc in the nuclei of MEFs isolated from ApcmNLS/mNLS mice supports the existence of alternative processes by which Apc may gain nuclear access. In the mouse model described here, the two canonical NLSs are eliminated, but the Armadillo repeat domain, which has been implicated previously in the nuclear import of APC, remains intact (Galea et al., 2001). In human polyp tissue and colon cancer cell lines, the Armadillo repeat domain has been proposed to mediate the nuclear entry of truncated forms of APC that lack the canonical NLSs (Anderson et al., 2002; Fagman et al., 2003). Moreover, the 15-amino-acid repeat region of APC (aa 959–1338) can facilitate the nuclear import of a fused green fluorescent protein (Wang et al., 2008). It is possible that one or both of these auxiliary Apc domains binds to the nuclear import machinery directly, or may allow binding of Apc to other proteins that are able to facilitate nuclear import or retention of Apc.

The most surprising result from this study was that ApcMin/mNLS mice had nearly twice as many polyps as ApcMin/+ mice (Figure 7C). Furthermore, polyps from ApcMin/mNLS mice were larger on average than those from ApcMin/+ mice (Figure 7D). Together, these observations are consistent with a role for nuclear Apc in the suppression of polyp initiation or progression. ApcmNLS/mNLS mice showed increased proliferation and Wnt signaling, either of which would be expected to enhance polyp formation. A recent report showed an association of Wnt target gene upregulation with increased intestinal adenomas in Apc1322T/+ mice (Pollard et al., 2009). It is also possible that the observed increase in polyp number and size might result from a role for nuclear Apc in suppressing polyp progression, once the polyp is initiated. In this scenario, polyps would initiate at the same rate in ApcMin/mNLS and ApcMin/+ mice, but would more rapidly grow large enough for detection in ApcMin/mNLS mice. Our finding of increased proliferation in polyps from ApcMin/mNLS mice as compared with ApcMin/+ mice does not distinguish between these two scenarios, but does implicate nuclear Apc regulation of cell proliferation as a contributing factor to tumor suppression.

In the current study, three separate parameters each showed a response in the colon distinct from that seen in the jejunum or ileum. Epithelia from jejunum or ileum did not show elevated levels of either Apc or β-catenin (Figure 3). By contrast, colon epithelia showed Apc and β-catenin protein levels that were significantly higher in ApcmNLS/mNLS mice as compared with Apc+/+ mice. This finding confirms that high levels of Apc do not always result in constitutive β-catenin degradation. Changes in the expression of Wnt target genes were consistent with elevated canonical Wnt signaling in the jejunum, ileum and colon tissues from ApcmNLS/mNLS mice as compared with Apc+/+ mice (Figure 4). However, BTEB2, a gene not regulated by canonical Wnt signaling, was unchanged in the jejunum and ileum, but was elevated in colonic epithelial cells from ApcmNLS/mNLS mice as compared with Apc+/+ mice. Finally, unlike small intestinal tissue, colon tissue from ApcMin/mNLS mice did not show elevated polyp size or multiplicity compared with ApcMin/+ mice (Figures 7B and D). Why do the colon and small intestine tissues show distinct responses to manipulation of Apc NLSs? We speculate that this variability might result from potential alternate feedback responses (chromosomal instability, apoptosis, survival) dependent on tissue type. Moreover, colon tissue lacks villi, performs a distinct function and maintains a different flora than the small intestine. Although not completely understood, humans and rats with germline APC mutations are prone to colon polyps, whereas mice with similar Apc mutations have polyps predominantly in the small intestine. Any of these factors might contribute to the differences seen in the colon and small intestinal tissue from ApcmNLS mice. Future studies using the novel ApcmNLS/mNLS model might offer clues to some of these longstanding puzzles in Apc biology.

In summary, the novel ApcmNLS/mNLS mouse model contains a subtle alteration that inactivates both Apc NLSs. Increased expression of Wnt targets and increased proliferation in the intestines of adult ApcmNLS/mNLS mice implicate nuclear Apc in the control of cellular proliferation and Wnt signaling. The finding of increased numbers and size of small intestinal polyps in ApcMin/mNLS mice as compared with ApcMin/+ mice implicates nuclear Apc in tumor suppression. Future studies of this novel mouse model will elucidate nuclear Apc contributions to other cellular events critical for tissue homeostasis, such as DNA repair, transcription regulation and cell-cycle progression.


Materials and methods

Creating the gene replacement vector

The homologous sequences of the DNA used for the targeting construct were obtained from a lambda phage library of genomic DNA isolated from 129 mouse ES cells (provided by Kirk Thomas and Mario Capecchi, University of Utah). This library was screened for fragments of Apc containing the two primary Apc NLSs. After identification of an Apc NLS-containing plaque by hybridization to a radioactive probe, a 14-kb stretch of mouse genomic DNA (Apc exons 14, 15 and surrounding introns) was isolated from the phage and inserted into the pBluescript KSII+ vector (Stratagene, Santa Clara, CA, USA). An EcoRI fragment containing both Apc NLSs was subcloned into the pUC19 vector. Mutations that inactivated each NLS and introduced novel restriction sites were inserted by PCR mutagenesis (Figure 1b). A second pUC19 vector was modified to destroy its EcoRI restriction site and introduce NheI and NotI sites. An 11537-bp region of Apc (NheI/NotI) was subcloned into this modified vector. The EcoRI fragment of Apc containing wild-type NLSs was replaced by the same fragment with mutant NLSs. Using restriction sites for KpnI and AflII in the non-coding region 3′ to Apc exon-15, a tACE-Cre-Neor cassette, flanked by two LoxP sites, was introduced into the pUC19 vector containing the Apc gene fragment with mutant Apc NLSs. This tACE-Cre-Neor cassette included a neomycin resistance gene (Neor) driven by the promoter for RNA polymerase-II and was linked to a gene encoding Cre recombinase under the control of the tACE promoter (Bunting et al., 1999). The HSV thymidine kinase gene controlled by the phosphoglycerate kinase promoter was inserted upstream from the Apc homology sequences in the pUC19 vector to create the 19947-bp targeting vector (Figure 1a).

Electroporation into ES cells

The targeting construct DNA was cleaved with NotI restriction endonuclease to linearize at a unique site 3′ of the region homologous to Apc. Digestion with NotI was followed by two phenol/chloroform extractions before the linearized gene replacement vector was electroporated into R1 mouse ES cells (Nagy et al., 1993) at the Transgenic and Gene-Targeting Institutional Facility at The University of Kansas Medical Center. Cells were selected for growth in 300μg/ml G418 (Mediatech Inc., Manassas, VA, USA) and 2μM ganciclovir (Roche, Indianapolis, IN, USA). After 8–10 days of culture in selective media, each surviving colony was picked and expanded into two separate wells. DNA was isolated from one well for genotype analysis and the cells in the remaining well were frozen in 96-well plates for injections.

Screening and verifying targeted ES cell lines

ES cell lines containing mutations in both Apc NLSs were identified by using a two-step PCR screen and the following primer sets: wild-type Apc NLS1 forward and NLS2 reverse: wNLS1fw (5′-CTAAGAAAAAGAAGCCTACTTCAC) and wNLS2rv (5′-GGCCTTTTCTTTTTTGGCATGGC); mutant Apc NLS1 forward and NLS2 reverse: mNLS1fw (5′-GCAGCCGCGGCACCTACT) and mNLS2rv (5′-TTGAAGGCCTTTTTGCGGCC) (Figure 1c). To identify homologous cassette insertion at the 3′-end, a primer annealing to DNA in the LoxP/tACE promoter region of the tACE-Cre-Neor cassette (5′-CCTGGCCCATGGAGATCCAT) was used with a primer annealing to genomic Apc 3′ of the targeting construct (5′-CATACCACCCACCATCCCTA) or with a primer annealing to the 3′-end of the targeting construct (5′-TCTCCCATTGCTTATGGCAAC) as a positive control (Figure 1h). Cell lines were also screened for correct incorporation of the 5′-end of the targeting construct by PCR. The reverse primers wNLS2rv and mNLS2rv were used in conjunction with the following primer that anneals to a region of genomic DNA upstream from the 5′-end of the targeting construct: (5′-AAATTGAACTCAGGACCTTCTC) (Figure 1f). PCR products digested with SstII produced a 6100-bp DNA product if mutant Apc NLS1 was present (Figure 1g). Each cell line with correct homologous recombination was further evaluated by determining the karyotype of 40–50 cells.

Generation of chimeric mice

For generation of chimeric mice, ApcmNLS/+ ES cells were injected into C57BL/6J blastocysts at the Gene Targeting and Transgenic Facility at the University of Virginia. Male chimeric mice that carried the mutant copy of Apc were bred to C57BL/6J females originally obtained from The Jackson Labs (Bar Harbor, ME, USA). Black progeny were culled and agouti-colored progeny were genotyped by using tail DNA isolated by following a protocol from The Jackson Laboratory (http://www.jax.org/imr/tail_nonorg.html) and PCR to screen for the presence of the mutant nuclear localization sites. Selection cassette loss in the ApcmNLS/+ progeny was verified by PCR using the following primers: forward primer, 5′-TCGGCCATTGAACAAGATGGA-3′ and reverse primer, 5′-ATTCGCCGCCAAGCTCTTCA-3′ (Figure 1e).

Mouse husbandry

Mice were maintained at the Animal Care Unit at the University of Kansas according to animal use statement number 137-01. The research complied with all relevant federal guidelines and institutional policies. The chimeric mice were bred with C57BL/6J mice from The Jackson Labs. ApcmNLS/+ progeny from the original mating of male chimeric mice with female C57BL/6J mice were repeatedly backcrossed to the inbred C57BL/6J mice 10–18 times to generate a line that was considered to be congenic (N10–N18). Detailed records, including date of birth, lineage, coat color, sex, genotype and date of killing, were maintained for each individual mouse. For the survival curve, N1 male and female ApcmNLS/+ mice were bred and F1 progeny were housed with same-sex siblings for the duration of the study (n=18 for Apc+/+; 19 for ApcmNLS/+ and 18 for ApcmNLS/mNLS mice). Cause of death for mice in the survival study was not determined because mice were allowed to die naturally and thus most were not collected for many hours after their death. Mice were fed ad libitum with Purina Lab Diet 5001 and were housed in cages in adjoining animal rooms. Congenic mice weighed at different time points between ages 6 and 14 weeks showed no significant differences when comparing Apc+/+ and ApcmNLS/mNLS mice. ApcMin/+ mice were purchased from The Jackson Labs and were maintained by breeding males with C57BL/6J females. ApcMin/mNLS mice were generated by breeding male ApcMin/+ mice (from Jackson Labs) with congenic (N10–N15) female ApcmNLS/+ mice.

Mouse genotyping

After weaning, mouse pups were tagged with a metal ear tag or by injection of an implantable electronic transponder (Bio Medic Data Systems Inc., Seaford, DE, USA) into the subcutaneous space above the shoulders. The Apc genotype of each mouse pup was determined by using isolated tail DNA and PCR to screen for the presence of the wild-type and mutant NLS coding sequences by using the following primers: forward, 5′-TAGT;GATGCGGTGAGTCCAA-3′ and reverse, 5′-ACCAAGTCCAACAAGCATCC-3′. Reaction conditions were as follows: 94°C for 5min, 35 cycles of 94°C for 1min, 54°C for 1min, 68°C for 1min and final 3min at 68°C. The PCR products (295bp) were cut with the SacII restriction enzyme. The restriction enzyme does not cut the wild-type allele but cuts the mutant allele into two fragments of 235 and 60bp (Supplementary Figure S4). The ApcMin allele was detected by using the standard PCR protocol published by The Jackson Laboratories (http://jaxmice.jax.org).

Analysis of gross and microscopic pathology, polyp measurement

The gross and cellular histology of intestinal tissues were examined in the N1–N13 generations of ApcmNLS mice and in 14-week old ApcMin/+ mice and ApcMin/mNLS mice. For each mouse, gastrointestinal tract from the stomach to the anal canal was dissected, opened longitudinally and fixed in 10% buffered formalin. Using a dissecting microscope, an investigator blind to the animal's genotype examined the intestinal luminal surface for any irregularities and polyps. Regions of tissue with abnormalities were recorded, removed from the surrounding tissue and stored in 10% buffered formalin. Tissue that appeared grossly abnormal was sent for pathologic evaluation. Intestinal polyps were located and diameter was measured with the aid of a dissection microscope (MZ8; Leica, Richmond, IL, USA) equipped with an eyepiece graticule and calibrated to a 50-mm-scale stage micrometer with 0.1 and 0.01-mm graduation. Fourteen-week-old ApcMin/mNLS and ApcMin/+ mice were analyzed for this study.

Evaluation of proliferation in intestinal epithelia

For proliferation analysis, congenic mice were injected with EdU and intestinal tissue prepared and stained as described by Nathke et al. (1996). Briefly, at the time of killing, mouse small and large intestines were removed and opened lengthwise. The colon, and proximal, middle and distal regions of the small intestine were individually rolled as described by Magnus (1937) to form multiple ‘Swiss Rolls’. The sections of rolled colon and small intestine were incubated in fixative (4% paraformaldehyde, 0.1% Triton X-100, in phosphate-buffered saline (PBS)) on ice for 1h, rinsed in PBS and incubated in 2.5M sucrose at 4°C overnight. After PBS rinse, rolled tissue sections were frozen in OCT tissue freezing media (VWR, West Chester, PA, USA) at −20°C. Sections were sliced immediately or stored at −80°C. Tissue cryosections (7μm) were made using a Leica CM1900 cryotome and adhered to glass slides coated with histomount (Invitrogen, Carlsbad, CA, USA) for immunohistochemistry. Slides were air-dried for approximately 1min, then tissue slices were permeabilized in 70% methanol before staining for EdU as described by Salic and Mitchison (2008). For each tissue in each genotype, 50–100 crypts were analyzed for proliferation by scoring the number of EdU-positive cells and the total cell number by 4′,6-diamidino-2-phenylindole (Dapi) staining in a crypt cross-section. Samples were coded so that analysis could be performed by a trained observer blind to the study parameters. Data from three mice of each genotype were collected to determine proliferation in the various intestinal tissues. Only crypts with 40 or more cells in a cross-section were scored.

Immunohistochemistry of intestinal epithelia and polyps

Tissues were fixed in 10% saline-buffered formalin for 16–20h then stored in 70% ethanol. Tissues were embedded in paraffin and sectioned at 6μm. Immunohistochemistry for β-catenin and Ki-67 was performed by using the Histomouse kit (Invitrogen, cat. no. 95–9541) according to the manufacturer's protocol, and with mouse monoclonal anti-β-catenin antibodies (1:100; BD Biosciences, Palo Alto, CA, USA, cat. no. 610153) and rat anti-mouse Ki-67 antibodies (1:20; Dako, Carpintera, CA, USA, cat. no. M7249). Tissues were coded, allowing the analysis to be performed by an observer blind to the tissue genotype. Normal-appearing jejunum, ileum and colon tissues from at least three mice of each genotype (Apc+/+, ApcmNLS/+ and ApcmNLS/mNLS) were examined for β-catenin localization. In addition, 6 polyps from 4 ApcMin/+ mice and 7 polyps from 3 ApcMin/mNLS mice were examined for β-catenin localization. After surveying the entire length of tissue for β-catenin localization, the average distribution was recorded and representative images were captured by using × 20 and × 40 objectives. The polyps were also scored for Ki-67 by counting Ki-67-positive cells per field at × 40 magnification. Three fields were photographed and scored for each polyp. The P-values for all proliferation studies were calculated by using unpaired, two-tailed t-tests and the GraphPad Prism software (GraphPad, La Jolla, CA, USA).

Isolation of mouse intestinal epithelial cells

Intestinal epithelial cells were isolated with modifications to a previous protocol (Whitehead et al., 1999). Immediately after killing, the small and large intestines were removed from congenic mice, opened lengthwise and rinsed with cold PBS. Tissue was incubated in 0.04% sodium hypochlorite for 15min on ice and then rinsed in cold PBS. The colon and small intestine were then incubated on ice for 15min in individual 15-ml conical tubes containing an EDTA/DTT solution (1.5–3mm EDTA and 0.5mM dithiothreitol (DTT) in PBS). The EDTA/DTT solution was poured off and replaced with cold PBS. The tubes were shaken forcefully for 10s to release the epithelial cells from the underlying tissue. The intestinal tissue was removed and placed in a fresh 15-ml conical tube containing the EDTA/DTT solution and the process was repeated two additional times. The released epithelial cells were collected by centrifugation at 700g for 5min at room temperature. Pellets of epithelia from all three rounds of extraction were resuspended in PBS with protease inhibitors and combined into one sample. The small intestinal epithelial cells from the second round of extraction were used in the experiments described here. Cell pellets were lysed in Reporter Lysis Buffer (Promega, Madison, WI, USA) with protease inhibitors (aprotinin, leupeptin and pepstatin each at 10μg/ml and 1mM phenylmethylsulfonyl fluoride) and briefly sonicated. Samples were boiled after addition of Sample Buffer (3 × Sample Buffer: 6% w/v sodium dodecyl sulfate, 30% glycerol, 150mM Tris (pH 6.8), ~0.2mg/ml bromophenol blue) and resolved by sodium dodecyl sulfate–PAGE before transfer to nitrocellulose membranes.

Immunoblot analysis of Apc and β-catenin in intestinal tissue and MEF fractionation

Immunoblots were processed as described by Wang et al. (2008). Membranes were probed with the following primary antibodies diluted as indicated in 5% non-fat dry milk/TBST: rabbit anti-APC M2 1:3000 (Wang et al., 2009), mouse anti-β-catenin (1:2000; BD Biosciences, cat. no. 610153), mouse anti-α-tubulin (1:50; Developmental Studies Hybridoma Bank, Iowa City, IA, USA), rabbit anti-fibrillarin (1:1000, ab5821; Abcam, Cambridge, MA, USA) and β-actin (1:2000; Sigma, St Louis, MO, USA). The following secondary antibodies were diluted as indicated: horseradish peroxidase goat anti-mouse (1:25000; Invitrogen) and horseradish peroxidase goat anti-rabbit (1:25000; Bio-Rad, Hercules, CA, USA). Blots were developed by using Western Lightning Chemiluminescence Reagent Plus (Perkin Elmer, Boston, MA, USA) or the SuperSignal West Femto Maximum Sensitivity Substrate (Pierce, Rockford, IL, USA) and a Kodak Image Station 4000R. Band analysis was conducted by using the Kodak ID Image Analysis Software, with protein levels first normalized to β-actin and then values for the Apc+/+ samples set to 1. The P-values were calculated by Mann–Whitney non-parametric test. MEFs of each genotype were isolated from congenic (N10–N18) mice on three independent occasions following the protocol described by Nagy et al. (2003). Independently isolated early-passage MEFs were subjected to fractionation in five independent experiments as described by Neufeld and White (1997). Nuclear Apc and β-catenin were calculated based on the band intensity in the nuclear fraction divided by the intensity of the nuclear plus the cytoplasmic bands. The P-values were calculated by using unpaired, two-tailed t-tests and the GraphPad Prism software.

Analysis of mRNA by real-time reverse transcription–PCR

For preparation of RNA, 200μl of suspended epithelial cells were added to 1ml of Trizol (Invitrogen) and tubes were stored at −80°C until use. RNA extraction was performed by using Trizol according to the manufacturer's protocol. For preparing cDNA, 1μg total RNA was incubated for 1h at 42°C with the cDNA reaction mix containing 1mM dNTPs, 1μg random hexamer primers (NEB, Ipswich, MA, USA), 1 × M-MLuV enzyme buffer (NEB) and 200U of M-MLuV reverse transcriptase enzyme (NEB). The reverse transcriptase enzyme was then inactivated by heating at 95°C for 5min. Quantitative PCR was performed for c-Myc, axin-2, cyclin-D1, BTEB2 and Hath-1 cDNA from mouse intestinal epithelial cells. The cDNA of the housekeeping gene HGPRT was used as an internal control. Table 1 shows the primers used.

PCRs were performed in a DNA engine Opticon 2 instrument (MJ Research, Waltham, MA, USA), using the SYBR green detection system. The total reaction volume was 25μl, containing 1 × DyNAmo HS SYBR Green qPCR kit (Finnzymes, Waltham, MA, USA), 15pmol of each primer and 3μl of 1:2.5 diluted cDNA. Each reaction was performed in triplicate and repeated three times. The reaction condition was initial denaturation at 95°C for 15min, followed by 40 cycles of denaturation at 94°C for 20s, annealing at 54°C for 30s and extension at 72°C for 30s. Fluorescence was measured at the end of every cycle and a melting curve was analyzed between 40 and 95°C, with 0.2°C increment. Samples were included in the analysis only if the melting curve was a single peak at the expected temperature. Average ΔC(t) was calculated for different genotypes relative to the housekeeping gene transcript, whereas the ΔΔC(t) of Wnt target cDNA for ApcmNLS/+ and ApcmNLS/mNLS was calculated relative to the Apc+/+ mice. P-values were calculated by using Mann–Whitney non-parametric test and the GraphPad Prism software.


Conflict of interest

The authors declare no conflict of interest.



  1. Anderson CB, Neufeld KL, White RL. (2002). Subcellular distribution of Wnt pathway proteins in normal and neoplastic colon. Proc Natl Acad Sci USA 99: 8683–8688. | Article | PubMed | CAS |
  2. Bunting M, Bernstein KE, Greer JM, Capecchi MR, Thomas KR. (1999). Targeting genes for self-excision in the germ line. Genes Dev 13: 1524–1528. | Article | PubMed | ISI | CAS |
  3. Cheung AF, Carter AM, Kostova KK, Woodruff JF, Crowley D, Bronson RT et al. (2009). Complete deletion of Apc results in severe polyposis in mice. Oncogene 29: 1857–1864. | Article | PubMed | ISI |
  4. Colnot S, Niwa-Kawakita M, Hamard G, Godard C, Le Plenier S, Houbron C et al. (2004). Colorectal cancers in a new mouse model of familial adenomatous polyposis: influence of genetic and environmental modifiers. Lab Invest 84: 1619–1630. | Article | PubMed | ISI | CAS |
  5. Dingwall C, Sharnick SV, Laskey RA. (1982). A polypeptide domain that specifies migration of nucleoplasmin into the nucleus. Cell 30: 449–458. | Article | PubMed | ISI | CAS |
  6. Fagman H, Larsson F, Arvidsson Y, Meuller J, Nordling M, Martinsson T et al. (2003). Nuclear accumulation of full-length and truncated adenomatous polyposis coli protein in tumor cells depends on proliferation. Oncogene 22: 6013–6022. | Article | PubMed | ISI | CAS |
  7. Fodde R, Edelmann W, Yang K, van Leeuwen C, Carlson C, Renault B et al. (1994). A targeted chain-termination mutation in the mouse Apc gene results in multiple intestinal tumors. Proc Natl Acad Sci USA 91: 8969–8973. | Article | PubMed | CAS |
  8. Fodde R, Kuipers J, Rosenberg C, Smits R, Kielman M, Gaspar C et al. (2001). Mutations in the APC tumour suppressor gene cause chromosomal instability. Nat Cell Biol 3: 433–438. | Article | PubMed | ISI | CAS |
  9. Galea MA, Eleftheriou A, Henderson BR. (2001). ARM domain-dependent nuclear import of adenomatous polyposis coli protein is stimulated by the B56 alpha subunit of protein phosphatase 2A. J Biol Chem 276: 45833–45839. | Article | PubMed | ISI | CAS |
  10. Giles RH, van Es JH, Clevers H. (2003). Caught up in a Wnt storm: Wnt signaling in cancer. Biochim Biophys Acta 1653: 1–24. | PubMed | ISI | CAS |
  11. Gorlich D, Prehn S, Laskey RA, Hartmann E. (1994). Isolation of a protein that is essential for the first step of nuclear protein import. Cell 79: 767–778. | Article | PubMed | ISI | CAS |
  12. He TC, Sparks AB, Rago C, Hermeking H, Zawel L, da Costa LT et al. (1998). Identification of c-MYC as a target of the APC pathway. Science 281: 1509–1512. | Article | PubMed | ISI | CAS |
  13. Henderson B. (2000). Nuclear–cytoplasmic shuttling of APC regulates β-catenin subcellular localization and turnover. Nat Cell Biol 2: 653–660. | Article | PubMed | ISI | CAS |
  14. Ishikawa T-O, Tamai Y, Li Q, Oshima M, Taketo MM. (2003). Requirement for tumor suppressor Apc in the morphogenesis of anterior and ventral mouse embryo. Dev Biol 253: 230–246. | Article | PubMed | ISI | CAS |
  15. Jaiswal AS, Narayan S. (2008). A novel function of adenomatous polyposis coli (APC) in regulating DNA repair. Cancer Lett 271: 272–280. | Article | PubMed | ISI | CAS |
  16. Kan Y, Winfried E, Kunhua F, Kirkland L, Venkateswara RK, Riccardo F et al. (1997). A mouse model of human familial adenomatous polyposis. J Exp Zool 277: 245–254. | Article | PubMed | ISI | CAS |
  17. Kaplan KB, Burds AA, Swedlow JR, Bekir SS, Sorger PK, Nathke IS. (2001). A role for the adenomatous polyposis coli protein in chromosome segregation. Nat Cell Biol 3: 429–432. | Article | PubMed | ISI | CAS |
  18. Kielman MF, Rindapaa M, Gaspar C, van Poppel N, Breukel C, van Leeuwen S et al. (2002). Apc modulates embryonic stem-cell differentiation by controlling the dosage of beta-catenin signaling. Nat Genet 32: 594–605. | Article | PubMed | ISI | CAS |
  19. Magnus HA. (1937). Observations on the presence of intestinal epithelium in the gastric mucosa. J Pathol Baceriol 44: 389–398. | Article |
  20. Mattaj IW, Englmeier L. (1998). Nucleocytoplasmic transport: the soluble phase. Annu Rev Biochem 67: 265–306. | Article | PubMed | ISI | CAS |
  21. Miyoshi Y, Ando H, Nagase H, Nishisho I, Horii A, Miki Y et al. (1992a). Germ-line mutations of the APC gene in 53 familial adenomatous polyposis patients. Proc Natl Acad Sci USA 89: 4452–4456. | Article | PubMed | CAS |
  22. Miyoshi Y, Nagase H, Ando H, Horii A, Ichii S, Nakatsuru S et al. (1992b). Somatic mutations of the APC gene in colorectal tumors: mutation cluster region in the APC gene. Hum Mol Genet 1: 229–233. | Article | PubMed | CAS |
  23. Moser A, Pitot H, Dove W. (1990). A dominant mutation that predisposes to multiple intestinal neoplasia in the mouse. Science 247: 322–324. | Article | PubMed | ISI | CAS |
  24. Moser AR, Mattes EM, Dove WF, Lindstrom MJ, Haag JD, Gould MN. (1993). ApcMin, a mutation in the murine Apc gene, predisposes to mammary carcinomas and focal alveolar hyperplasias. Proc Natl Acad Sci USA 90: 8977–8981. | Article | PubMed | CAS |
  25. Munemitsu S, Albert I, Souza B, Rubinfeld B, Polakis P. (1995). Regulation of intracellular beta-catenin levels by the adenomatous polyposis coli (APC) tumor-suppressor protein. Proc Natl Acad Sci USA 92: 3046–3050. | Article | PubMed | CAS |
  26. Nagy A, Gertsenstein M, Vintersten K, Behringer R. (2003). Manipulating the Mouse Embryo: a Laboratory Manual 3rd edn. Cold Spring Harbor Press: New York.
  27. Nagy A, Rossant J, Nagy R, Abramow-Newerly W, Roder JC. (1993). Derivation of completely cell culture-derived mice from early-passage embryonic stem cells. Proc Natl Acad Sci USA 90: 8424–8428. | Article | PubMed | CAS |
  28. Nathke IS, Adams CL, Polakis P, Sellin JH, Nelson J. (1996). The adenomatous polyposis coli tumor suppressor protein localizes to plasma membrane sites involved in active cell migration. J Cell Biol 134: 165–179. | Article | PubMed | ISI | CAS |
  29. Neufeld KL. (2009). Nuclear functions of APC. Adv Exp Med Biol 656: 13–29. | PubMed |
  30. Neufeld KL, White RL. (1997). Nuclear and cytoplasmic localizations of the adenomatous polyposis coli protein. Proc Natl Acad Sci USA 94: 3034–3039. | Article | PubMed | CAS |
  31. Neufeld KL, Zhang F, Cullen BR, White RL. (2000). APC-mediated downregulation of β-catenin activity involves nuclear sequestration, nuclear export. EMBO Rep 1: 519–523. | Article | PubMed | ISI | CAS |
  32. Neufeld KL, Zhang F, Cullen BR, White RL. (2000). APC-mediated downregulation of beta-catenin activity involves nuclear sequestration, nuclear export. EMBO Rep 1: 519–523. | Article | PubMed | ISI | CAS |
  33. Oshima M, Oshima H, Kitagawa K, Kobayashi M, Itakura C, Taketo M. (1995). Loss of Apc heterozygosity, abnormal tissue building in nascent intestinal polyps in mice carrying a truncated Apc gene. Proc Natl Acad Sci USA 92: 4482–4486. | Article | PubMed | CAS |
  34. Peters R. (1986). Fluorescence microphotolysis to measure nucleocytoplasmic transport, intracellular mobility. Biochim Biophys Acta 864: 305–359. | PubMed | ISI |
  35. Pollard P, Deheragoda M, Segditsas S, Lewis A, Rowan A, Howarth K et al. (2009). The Apc1322 T mouse develops severe polyposis associated with submaximal nuclear β-catenin expression. Gastroenterology 136: 2204–2213. | Article | PubMed | ISI | CAS |
  36. Qian J, Sarnaik AA, Bonney TM, Keirsey J, Combs KA, Steigerwald K et al. (2008). The APC tumor suppressor inhibits DNA replication by directly binding to DNA via its carboxyl terminus. Gastroenterology 135: 152–162. | Article | PubMed | ISI |
  37. Rosin-Arbesfeld R, Cliffe A, Brabletz T, Bienz M. (2003). Nuclear export of the APC tumour suppressor controls beta-catenin function in transcription. EMBO J 22: 1101–1113. | Article | PubMed | ISI | CAS |
  38. Rosin-Arbfeld R, Townsley F, Bienz M. (2000). The APC tumour suppressor has a nuclear export function. Nature 406: 1009–1012. | Article | PubMed | ISI | CAS |
  39. Salic A, Mitchison TJ. (2008). A chemical method for fast and sensitive detection of DNA synthesis in vivo. Proc Natl Acad Sci USA 105: 2415–2420. | Article | PubMed |
  40. Sasai H, Masaki M, Wakitani K. (2000). Suppression of polypogenesis in a new mouse strain with a truncated Apc(Delta474) by a novel COX-2 inhibitor, JTE-522. Carcinogenesis 21: 953–958. | Article | PubMed | ISI |
  41. Sena P, Saviano M, Monni S, Losi L, Roncucci L, Marzona L et al. (2006). Subcellular localization of beta-catenin and APC proteins in colorectal preneoplastic and neoplastic lesions. Cancer Lett 241: 203–212. | Article | PubMed | ISI |
  42. Sierra J, Yoshida T, Joazeiro CA, Jones KA. (2006). The APC tumor suppressor counteracts beta-catenin activation and H3K4 methylation at Wnt target genes. Genes Dev 20: 586–600. | Article | PubMed | ISI | CAS |
  43. Smith KJ, Johnson KA, Bryan TM, Hill DE, Markowitz S, Willson JK et al. (1993). The APC gene product in normal and tumor cells. Proc Natl Acad Sci USA 90: 2846–2850. | Article | PubMed | CAS |
  44. Smits R, Kielman MF, Breukel C, Zurcher C, Neufeld K, Jagmohan-Changur S et al. (1999). Apc1638T: a mouse model delineating critical domains of the adenomatous polyposis coli protein involved in tumorigenesis and development. Genes Dev 13: 1309–1321. | Article | PubMed | ISI | CAS |
  45. Taketo MM. (2006). Mouse models of gastrointestinal tumors. Cancer Sci 97: 355–361. | Article | PubMed | ISI | CAS |
  46. Tetsu O, McCormick F. (1999). Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature 398: 422–426. | Article | PubMed | ISI | CAS |
  47. Wang Y, Azuma Y, Friedman DB, Coffey R, Neufeld KL. (2009). Novel association of APC with intermediate filaments identified using a new versatile APC antibody. BMC Cell Biol 10: 75–88. | Article | PubMed | ISI |
  48. Wang Y, Azuma Y, Moore D, Osheroff N, Neufeld KL. (2008). Interaction between tumor suppressor APC and topoisomerase IIα: implication for the G2/M transition. Mol Biol Cell 19: 4076–4085. | Article | PubMed | ISI | CAS |
  49. Whitehead RH, Demmler K, Rockman SP, Watson NK. (1999). Clonogenic growth of epithelial cells from normal colonic mucosa from both mice and humans. Gastroenterology 117: 858–865. | Article | PubMed | ISI | CAS |
  50. Zhang F, White R, Neufeld K. (2000). Phosphorylation near nuclear localization signal regulates nuclear import of APC protein. Proc Natl Acad Sci USA 97: 12577–12582. | Article | PubMed | CAS |
  51. Zhang F, White RL, Neufeld KL. (2001). Cell density and phosphorylation control the subcellular localization of adenomatous polyposis coli protein. Mol Cell Biol 21: 8143–8156. | Article | PubMed | ISI | CAS |


This work was supported by RO1 CA10922 from the National Cancer Institute and 5P20 RR15563. We extend our gratitude to Alan Godwin, Kirk Thomas and Mario Capecchi for advice on the generation of the knock-in mouse and for the lambda phage library, and the tACE-Cre-Neor- and TkHSV-containing constructs. We thank Marc Roth, Areli Monarrez, Ashrita Abraham and Travis Friesen for technical assistance; the staff at the University of Kansas Animal Care Unit for excellent mouse husbandry and David Davido for critical reading of the manuscript. We also thank Reka Nagy and Drs Andras Nagy, Janet Rossant and Wanda Abramow-Newerly (Mount Sinai Hospital) for the mouse R1 ES cells.

Supplementary Information accompanies the paper on the Oncogene website