1

APC differentially regulates β-catenin phosphorylation and ubiquitination in colon cancer cells

Jun Yang1, Wen Zhang1, Paul M. Evans1, Xi Chen1, Xi He2 and Chunming Liu1, 3

1Sealy Center for Cancer Cell Biology and Department of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, TX 77555

2Division of Neuroscience, Children’s Hospital, Harvard Medical School, Boston, MA 02115 Running Title: APC regulates β-catenin ubiquitination

3Correspondence: Chunming Liu, Sealy Center for Cancer Cell Biology, University of Texas Medical Branch, Galveston, TX 77555-1048. Tel: (409)747-1909; Fax: (409)747-1938; e-mail: chliu@utmb.edu

Most colorectal cancers have mutations of the APC (adenomatous polyposis coli) gene or the β-catenin gene which stabilize β-catenin and activate β-catenin target genes, leading ultimately to cancer. The molecular mechanisms of APC function in β-catenin degradation are not completely known. APC binds β-catenin and is involved in the Axin complex, suggesting that APC regulates β-catenin phosphorylation. Some evidence also suggests that APC regulates β-catenin nuclear export. Here, we examine the effects of APC mutations on β-catenin phosphorylation, ubiquitination and degradation in colon cancer cell lines SW480, DLD-1 and HT29, each of which contains a different APC truncation. Although the current models suggest that β-catenin phosphorylation should be inhibited by APC mutations, we detected significant β-catenin phosphorylation in these cells. However, β-catenin ubiquitination and degradation were inhibited in SW480, but not in DLD-1 and HT29 cells. The ubiquitination of β-catenin in SW480 cells can be rescued by exogenous expression of APC. The APC domains required for β-catenin ubiquitination were analyzed. Our results suggest that APC regulates β-catenin phosphorylation and ubiquitination by distinct domains and by separate molecular mechanisms. Colorectal cancer is the second leading cause of cancer-related death in the United States and the incidence of colorectal cancer is rising in developing countries (1). APC is an important tumor suppressor for colorectal cancer. APC mutations are responsible for FAP (familial adenomatous polyposis) and up to 85% of sporadic colorectal cancers (2). APC is an essential component of the Wnt/β-catenin

signaling pathway, which plays a key role in intestinal homeostasis and colorectal cancers (3,4). β-catenin is a multifunctional protein. It binds E-cadherin on the cell membrane, regulating cell adhesion. It also binds T-cell factor (TCF) in the nucleus and controls gene expression (3-6). In normal cells, β-catenin degradation is initiated by serine/threonine (S/T) phosphorylation at its amino terminal region (7,8). Without Wnt signaling, β-catenin forms a complex with Casein Kinase Iα (CKIα), Glycogen Synthase Kinase-3 (GSK-3) and a scaffold protein, Axin, resulting in CKIα mediated phosphorylation of β-catenin at S45. S45 phosphorylation provides a recognition site for GSK-3 and allows GSK-3 to phosphorylate β-catenin at positions T41, S37 and S33 (9-11). Phosphorylated β-catenin is recognized by β-Trcp and targeted for ubiquitination and degradation (12-15). Wnt signalling inhibits GSK-3 phosphorylation of β-catenin but not CKIα phosphorylation of β-catenin (10). However, it is still unknown whether APC regulates β-catenin phosphorylation, or if APC regulates GSK-3 or CKIα phosphorylation of β-catenin. Furthermore, it is unclear how β-catenin shifts from a complex with Axin to a separate complex with β-Trcp. APC has two types of β-catenin binding domains, three 15aa repeat domains and seven 20aa repeat domains (3-6). APC also has three Axin binding domains, which are called SAMP motifs (16). Most colorectal cancers have an APC truncation in one allele and loss-of-heterozygosity in the other allele (2,17). APC mutations associated with cancer are consistently truncated before amino acid 1638 and lack the Axin binding domains (18). Since APC has both β-catenin and Axin binding domains, it has been suggested that APC is an essential component in the Axin

http://www.jbc.org/cgi/doi/10.1074/jbc.M600831200The latest version is at JBC Papers in Press. Published on May 3, 2006 as Manuscript M600831200

Copyright 2006 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

2

complex (16,19,20). It would be expected that, in these colon cancer cells, APC mutations prevent the assembly of Axin complex and thus prevent β-catenin phosphorylation. Recently, crystal structures of β-catenin/APC and β-catenin/Axin complex suggest that APC may compete with Axin for β-catenin binding, thus releasing phosphorylated β-catenin from the Axin complex for degradation (21-23). APC also has a nuclear export signal and can shuttle between the cytoplasm and nucleus (24-26). It has been suggested that APC may transport β-catenin out of the nucleus. According to this model, APC mutations should prevent the nuclear export of β-catenin and thus prevent β-catenin phosphorylation and degradation in colon cancer cells. To understand APC function in vivo, and to understand why APC mutations inhibit β-catenin degradation in colon cancer cells, we analyzed β-catenin phosphorylation in colon cancer cell lines containing different APC truncations. We found that these APC truncations did not prevent β-catenin phosphorylation, suggesting that truncated APC retains at least partial activity for β-catenin phosphorylation. However, β-catenin ubiquitination and degradation was inhibited in SW480 cells. We demonstrated that the defect in β-catenin ubiquitination and degradation in SW480 cells is due to an APC truncation. The truncated APC only inhibits β-catenin ubiquitination but not β-catenin phosphorylation. These results indicate that the mechanisms of APC function in β-catenin regulation are more complicated than originally thought.

EXPERIMENTAL PROCEDURES Cell lines−Colon cancer cell lines, HEK293T, HCT116, SW480, DLD-1 and HT29 were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% of Fetal Bovine Serum (FBS), at 37oC, 5% CO2. HEK293T cells were transfected using the calcium phosphate method. SW480 cells were transfected with Lipofectamine 2000 (Invitrogen, Carlsbad, CA) or infected with adenovirus. For proteasome inhibition, cells were treated with MG132 (2.5µM) for 6h.

Constructs−Truncated APC fragments were amplified by PCR and cloned into the pCS2MT vector. DNA sequences were confirmed by automatic sequencing. Adenovirus Ad-CRB was a gift from Dr. B. Vogelstein (27). Adenoviruses that express different APC fragments were constructed using the Ad-Easy Kit from ATCC. GST-APC2,3 was a gift from Dr. W. Xu. The putative APC phosphorylation sites S1385, T1388, S1389, S1391, S1501, S1504, S1505, S1507 and S1510 (according to Ha et al. (21)) were changed to alanine in GST-APC2,3 to make an S→A mutant. This site-directed mutagenesis was designed using the computer program SiteFind (28). CKI kinase was purchased from New England Biolabs (Beverly, MA). Western blot−Cells were lysed in 1% Triton lysis buffer containing 1% Triton 100, 50 mM HEPES (pH 7.4), 1.5 mM EDTA, 150 mM NaCl, 10% glycerol, 10 mM NaF, 1 mM Na3VO4, 0.5 mM DTT, and a cocktail of protease inhibitors. GST pull down assays and Western blots were performed as previously described (13). Cell fractions were isolated using a Cell Fractionation Kit from Sigma (St. Louis, MI). Phospho-specific antibodies for β-catenin were described previously (10). Mouse anti-β-Trcp antibody was purchased from Zymed Lab (San Francisco, CA). Immunohistochemistry−Cells grown on coverslips were fixed for 10 min with 4% paraformaldehyde. The cells were permeabilized with PBS containing 0.2% (w/v) Triton X-100 for 20 min and then blocked by serum free protein blocking buffer (DakoCytomation, Denmark) for 1 h. Anti-β-catenin antibody (1:100, Upstate, Charlottesville, VA) was diluted in antibody diluent (DakoCytomation) and incubated with cells overnight. The cells were washed 3 times with TBST, incubated with Alexa-594-labeled anti-mouse IgG (1:800) and then diluted in TBST for 1 h. The coverslips were washed, mounted on glass slides, and photographed with a Zeiss LSM510 confocal microscope (Germany). Enzymatic assay−Luciferase activity was analyzed by the Dual-luciferase Reporter Assay System (Promega, Madison WI). In vitro kinase assays were performed as previously described (13). The alkaline phosphatase assay was performed by incubating the immunoprecipitated

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

3

β-catenin with Calf-Intestinal alkaline Phosphatase (CIP) at 37oC for 30 min.

RESULTS β-catenin phosphorylation was not inhibited in colon cancer cells containing APC mutations −The majority of colon cancer cells have APC mutations; therefore, it was expected that these cells would not have phosphorylated β-catenin. We analyzed β-catenin phosphorylation in APC mutant colon cancer cell lines SW480, DLD-1 and HT29 using phospho-specific antibodies against β-catenin. APC is deleted at the C-terminus at residue 1338 in SW480 cells, residue 1427 in DLD-1 cells and residue 1555 in HT29 cells (29-31) (Fig. 1A). The colon cancer cell line HCT116, which has wild type APC was used as a control. As described in our previous study, antibody Ab45 only recognizes β-catenin phosphorylation at residue Ser45, and Ab(33+37)/41 only recognizes S33/S37/T41 phosphorylation of β-catenin (10). Unexpectedly, we found that both S33/S37/T41 and S45 were phosphorylated in these cells, suggesting that APC truncations in these cell lines do not inhibit GSK-3 or CKIα phosphorylation of β-catenin (Fig. 1B). To further confirm that these antibodies only recognize phosphorylated β-catenin, β-catenin protein was immunoprecipitated with a β-catenin antibody from HT29 and HCT116 cell lysates and treated with calf-intestinal alkaline phosphatase (CIP). The levels of β-catenin protein were remained the same as the control samples without CIP treatment. However, β-catenin phosphorylation could not be detected in the CIP treated samples (Fig. 1C). These data confirmed the specificity of the phospho-specific antibodies, further suggesting that, in human colon cancer cell lines SW480, DLD-1 and HT29, APC mutations do not prevent β-catenin phosphorylation. HCT116 cells have one allele of wild type β-catenin and one allele of mutated β-catenin that has a deletion at S45 (32). Since the S45 deletion mutant is stabilized relative to the wild type protein, it comprises most of the β-catenin within the cell. We observed a lesser degree of S45 phosphorylation in these cells. APC mutation inhibits β-catenin ubiquitination and degradation in the SW480

colon cancer cells−To further study how APC regulates β-catenin degradation, several colon cancer cell lines were treated with MG132, a proteasome inhibitor, and analyzed with β-catenin antibodies. HCT116 cells were used as a control. It is known that proteasome inhibitors can stabilize ubiquitinated β-catenin. After MG132 treatment, two major bands were detected by Western blot. The lower band is β-catenin without ubiquitination; the upper band is β-catenin with mono-ubiquitination (30,33). As expected, after MG132 treatment, ubiquitinated β-catenin was detected in HCT116 cell lysates (Fig. 2A). The ubiquitinated bands of β-catenin were also detected in DLD-1 and HT29 cells. In HCT116, DLD-1 and HT29 cells treated with MG132, β-catenin levels increased, suggesting that β-catenin degradation was not completely inhibited in these cells. In contrast, SW480 cells demonstrated strong β-catenin phosphorylation, almost undetectable ubiquitination, and no increase in β-catenin levels after MG132 treatment. This suggests that β-catenin ubiquitination and degradation were inhibited in SW480 cells (Fig. 2A). Similar results have been reported by Sadot et al. (34). However, it is not known whether the defect of β-catenin ubiquitination in SW480 cells was caused by an APC mutation or by other mutations. To confirm that β-catenin phosphorylation is required for ubiquitination, we transfected wild type β-catenin and mutant β-catenin into HEK293T cells. The cells were treated with or without MG132. We found that wild type β-catenin, but not mutant β-catenin, was modified by ubiquitination (Fig. 2B). We infected SW480 cells with Ad-CBR, which carries a functional APC fragment (27). This fragment contains all of the 15aa repeats and 20aa repeats, in addition to the Axin binding domains. We found that Ad-CBR can rescue β-catenin ubiquitination and degradation in SW480 cells, indicating that APC is required for the ubiquitination and degradation of phosphorylated β-catenin (Fig. 2C). β-catenin phosphorylation was stronger in SW480 cells, most likely because phosphorylated β-catenin is degraded more slowly and accumulates in SW480 cells. MG132 treatment also stabilized phosphorylated β-catenin. Ad-CBR increased β-catenin phosphorylation,

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

4

indicating that APC is required for the efficient phosphorylation of β-catenin. APC domains required for β-catenin ubiquitination−Notably, the APC mutations in these colon cancer cells are truncated mutations and not null mutations. These APC proteins still have the 15aa repeat β-catenin binding domains and at least one 20aa β-catenin binding domain. β-catenin phosphorylation may be regulated by the β-catenin binding domains that remain in the APC mutants. The CBR fragment of APC can rescue β-catenin ubiquitination in SW480 cells. To determine which APC domains are required for β-catenin ubiquitination, we cloned truncated CBR fragments into the CS2MT vector as well as an adenovirus vector (Fig. 3A). Except for the APC fragment containing the first two 20aa repeats (APC-I2), the expression of these APC fragments can be detected by Western blot using anti-Myc antibody (Fig. 3B). We have made several constructs for APC-I2, but were unable to detect this protein. The effects of these APC fragments on a TCF reporter were analyzed in SW480 cells. APC fragments with SAMP domains strongly inhibited the TCF reporter. An APC fragment with 1 or 3 20aa repeats modestly inhibited TCF reporter activity (Fig. 3C). SW480 cells were infected with several adenoviruses that express each of these APC fragments and then treated with MG132. We found that the APC fragment with the first three 20aa repeats (APC-I3) can sufficiently rescue β-catenin ubiquitination (Fig. 3D). Control virus (GFP) or a virus containing the first 20aa repeat (APC-I1) did not rescue β-catenin ubiquitination in SW480 cells. These data demonstrate that the second and third 20aa repeats are required for β-catenin ubiquitination, whereas the SAMP domains are not required for β-catenin ubiquitination. It is also possible that any combination of two 20aa repeats may be sufficient for β-catenin ubiquitination. β-catenin localization in SW480 cells−Since APC has a nuclear export signal, APC may regulate β-catenin subcellular localization. It has been suggested that a mutation in APC might inhibit the nuclear export of β-catenin and thus inhibit β-catenin degradation, resulting in β-catenin accumulation in the nucleus of colon cancer cells. To test whether β-catenin

ubiquitination is regulated by β-catenin localization, the SW480 cells were infected with APC adenoviruses, which co-express the GFP protein, and then stained with anti-β-catenin antibody. Control virus (GFP) or an APC virus with the first 20aa repeat (APC-I1) had no effect on the cytoplasmic levels of β-catenin (Fig. 4). β-catenin was detected in both the nucleus and cytoplasm, with or without infection. APC viruses with 3-5 20aa repeats (APC-I3, APC-I4 and APC-I5) significantly reduced the cytoplasmic β-catenin levels (Fig. 4). These data confirm that the second and third 20aa repeats of APC are required for β-catenin degradation. We didn’t find a clear correlation between β-catenin localization, degradation and APC status. Since β-catenin can be detected in both nucleus and cytoplasm of SW480 cells, the inhibition of β-catenin ubiquitination is not due to the effects on β-catenin localization. Although the SAMP domains are not absolutely required for β-catenin degradation, APC-I4 and APC-I5 are more efficient in β-catenin degradation. The function of 15aa repeats of APC−The 15aa repeats also bind β-catenin. To determine whether the 15aa repeats are required for β-catenin ubiquitination and degradation, we deleted the 15aa repeats from the CBR fragment (Fig. 5A). We found that the CBR fragment, without 15aa repeats (APC-II5), can sufficiently inhibit the TCF reporter and rescue β-catenin ubiquitination in SW480 cells (Fig. 5B, C). This construct also reduced cytoplasmic β-catenin levels, suggesting that the 15aa repeats are not required for β-catenin ubiquitination (Fig. 5D). Whether these 15aa repeats are required for β-catenin phosphorylation is not known. Mechanisms for APC mediated β-catenin ubiquitination−To confirm our results from immunostaining, nuclear and cytoplasmic proteins were isolated from SW480 and HT29 cells (Fig. 6A). Strong β-catenin signals were detected in both nuclear and cytoplasmic fractions. These data further confirm that β-catenin ubiquitination but not phosphorylation or nuclear export is inhibited in SW480 cells. Since β-catenin ubiquitination was inhibited in SW480 but not DLD-1 and HT29 cells, and β-Trcp couples β-catenin phosphorylation and ubiquitination, we examined

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

5

the interaction between β-catenin and β-Trcp in these cells. The interaction was detected in all three cell lines, and the interaction in SW480 cells was not dependent on functional APC (Fig. 6B). Similar result has been reported using a Myc-tagged β-Trcp (12). β-catenin ubiquitination in SW480 cells was rescued by APC fragments containing three 20aa repeats but not an APC fragment containing the first 20aa repeat, suggesting that the second and third 20aa repeats of APC may play a critical role in β-catenin ubiquitination. The 20aa repeats of APC have several putative phosphorylation sites, which are similar to β-catenin phosphorylation sites (Fig. 6C). X-ray crystallographic studies suggest that phosphorylated 20aa repeats of APC can compete with Axin for β-catenin binding and release phosphorylated β-catenin from the Axin complex for degradation (21-23). We previously identified a dual-kinase mechanism for β-catenin phosphorylation (10). To examine whether APC can be phosphorylated by a similar mechanism, we used a GST fusion protein containing the second and third 20aa repeats of APC (GST-APC2,3). GST-APC2,3 was phosphorylated in vitro by purified CKI, and then incubated with HCT116 cell lysate. GST-APC2,3 was pulled down using GST-beads and bound β-catenin was analyzed by Western blot. CKI induced a band shift of GST-APC2,3, as detected by an anti-GST antibody. Phosphorylation of APC2,3 indeed increased β-catenin binding affinity (Fig. 6D). As a control, mutations in the putative phosphorylation sites of the second and third 20aa repeats inhibited CKI phosphorylation and abolished binding with β-catenin (Fig. 6D). Several quantitative studies have demonstrated the ability of APC phosphorylation to enhance of its binding affinity with β-catenin (21,23). These data further support our hypothesis that APC regulates β-catenin phosphorylation and ubiquitination by distinct molecular mechanisms.

DISCUSSION

As a tumor suppressor for colorectal cancers, the major function of APC is to regulate β-catenin (3-6). It is generally believed that APC regulates β-catenin degradation by regulating β-catenin phosphorylation and localization. We found that

APC also regulates β-catenin ubiquitination. The APC mutation in SW480 cells inhibits β-catenin ubiquitination regardless of the status of β-catenin phosphorylation and localization. Our data suggest that APC regulates β-catenin degradation by distinct domains and separate molecular mechanisms. β-catenin phosphorylation is carried out by the Axin complex containing CKIα, GSK-3 and APC (16,19,20). APC binds both β-catenin and Axin, and deleting the APC binding domain from Axin blocks β-catenin/Axin binding (20). A simple model to explain these findings is that APC regulates the assembly of the Axin complex, thus regulating β-catenin phosphorylation. Colon cancer cell lines HT29, DLD-1 and SW480 all contain a truncated APC protein lacking the Axin binding domains. Despite the fact that β-catenin degradation was inhibited, we found that β-catenin can still be phosphorylated in these cells. We hypothesize that APC is required for β-catenin phosphorylation in normal cells. In tumor cells with high levels of β-catenin, β-catenin can be phosphorylated independent of the presence of full length APC. The rate of both β-catenin protein synthesis and its degradation are tightly controlled in normal cells, in order to maintain a low level of β-catenin in both the nuclear and cytoplasmic compartments. In APC mutant cells, however, the rate of β-catenin phosphorylation and degradation is lower than the rate of β-catenin synthesis, resulting in an accumulation of β-catenin in the nucleus, contributing to tumorigenesis. High levels of β-catenin phosphorylation in SW480 cells were also observed by Sadot et al., who proposed that the difference between β-catenin phosphorylation in different APC truncated cells may be the result of a loss in a phosphatase binding site (34). We observed β-catenin ubiquitination in DLD-1 and HT29 cells, which have the first two or first three 20aa repeats, but not in the SW480 cell line, which only has the first 20aa repeat, suggesting that the second and third 20aa repeats are required for β-catenin ubiquitination. It appears that the first 20aa repeat of APC in SW480 cells is not sufficient for regulating β-catenin ubiquitination. Given that APC in DLD-1 cells is truncated between the second and third 20aa repeats, and the fact that β-catenin

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

6

ubiquitination can be detected in DLD-1 cells, the first two 20aa repeats should be sufficient for β-catenin ubiquitination. Whether each of these seven 20aa repeats has similar a function is not known. It is possible that any combination of two 20aa repeats is sufficient for β-catenin ubiquitination. In our TCF reporter assay, APC fragments with one 20aa repeat, which cannot rescue β-catenin ubiquitination, also inhibited the reporter activity. This may be because the APC fragment binds β-catenin and inhibits the activity of the β-catenin/TCF complex. It is also possible that the constructs with multiple 20aa repeats sequestered more β-catenin from the reporter. Although the Axin binding domains are not required for β-catenin ubiquitination, APC fragments containing at least one SAMP repeat are indeed more efficient at promoting β-catenin degradation. APC protein in both HT29 and DLD-1 cells lacks SAMP domains. Although these cells demonstrate limited degree of β-catenin ubiquitination and degradation, β-catenin levels are dramatically increased compared with normal cells. Again, this may due to the fact that loss of SAMP domain decreases the efficiency of APC-mediated degradation, disrupting the balance of β-catenin synthesis and degradation. After phosphorylation, β-catenin must be released from the Axin complex for ubiquitination and degradation. β-catenin phosphorylation, but not ubiquitination, is noted in SW480 cells, suggesting that mutations in APC prevent the proper assembly of the ubiquitin ligase complex. A possible explanation for this phenomenon is that truncated APC cannot release phosphorylated β-catenin from the Axin complex, thus preventing the proper assembly of β-catenin/β-Trcp ubiquitin ligase complex. The interaction between β-catenin and β-Trcp was detected in SW480 cells, suggesting other steps of β-catenin ubiquitination were inhibited by APC truncation. However, we cannot rule out the possibility that β-Trcp doesn’t bind β-catenin in vivo in SW480 cells, but interacts with β-catenin during immunoprecipitation. The second and third 20aa repeats are involved in β-catenin ubiquitination. We have confirmed that the 20aa repeats can be phosphorylated in vitro and that phosphorylated APC binds β-catenin more strongly. Our data

support the model proposed from crystallographic studies. However, it is still unknown which kinase phosphorylates which sites of APC in vivo. To address these questions, we are currently generating phospho-specific antibodies against these putative phosphorylation sites. It has been reported that the localization of β-catenin is regulated by the cell density in SW480 cells, and that several antibodies used for APC immunostaining detect none-specific bands in the nucleus, suggesting that the nuclear function of APC should be interpreted with caution (29,35). Since APC protein in these colon cancer cell lines still has β-catenin binding domains, these domains may be sufficient β-catenin nuclear export. However, in at least the colon cancer cell line SW480, APC truncation did not completely block β-catenin nuclear export and phosphorylation, but did inhibit β-catenin ubiquitination and degradation. Recently, Krieghoff et al. reported that APC regulates β-catenin subcellular localization by retention, rather than by active transport (36). Both 15aa and 20aa repeats bind to the conserved β-catenin surface made by Armadillo repeats 5-9. The 20aa repeats have additional interactions when they are phosphorylated (21). We have shown that the 15aa repeats are not required for β-catenin ubiquitination. In the colon cancer cell lines used in this study, truncated APC contains three 15aa repeats but β-catenin can still be phosphorylated. If APC is required for β-catenin phosphorylation, it is possible that the 15aa repeats are required for β-catenin phosphorylation and the 20aa repeats are required for β-catenin ubiquitination. APC-mediated regulation of β-catenin degradation by different domains and separate steps has important clinical implications. For example, different APC truncations correspond to different levels of tumor progression (4,37). Truncations that still retain the three 15aa repeats and one or two of the 20aa repeats provide the strongest selective advantage and demonstrate a more aggressive tumor phenotype. Truncations that have lost all of the β-catenin regulating domains or retain both β-catenin and Axin binding domains produce more attenuated tumors (37). Our results suggest that different APC truncations have different molecular consequences on β-

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

7

catenin regulation, thus providing novel insights into the correlation between APC truncations and colon cancer phenotypes. Acknowledgment−We are grateful to Dr. Bert Vogelstein for the Ad-CBR adenovirus, to Dr.

Wenqing Xu for GST-APC2,3 construct and to Dr. B. Mark Evers for advice and critical reading of the manuscript. C.L. is supported by a John Sealy Memorial Fund Recruitment Award and by R21 CA112007 from the NCI.

REFERENCES 1. Cancer Facts & Figures 2005, American Cancer Society 2. Kinzler, K. W., and Vogelstein, B. (1996) Cell 87(2), 159-170 3. Giles, R. H., van Es, J. H., and Clevers, H. (2003) Biochim Biophys Acta 1653(1), 1-24 4. Polakis, P. (2000) Genes Dev 14(15), 1837-1851 5. Moon, R. T., Brown, J. D., and Torres, M. (1997) Trends Genet 13(4), 157-162 6. Wodarz, A., and Nusse, R. (1998) Annu Rev Cell Dev Biol 14, 59-88 7. Aberle, H., Bauer, A., Stappert, J., Kispert, A., and Kemler, R. (1997) Embo J 16(13), 3797-3804 8. Orford, K., Crockett, C., Jensen, J. P., Weissman, A. M., and Byers, S. W. (1997) J Biol Chem

272(40), 24735-24738 9. Amit, S., Hatzubai, A., Birman, Y., Andersen, J. S., Ben-Shushan, E., Mann, M., Ben-Neriah, Y.,

and Alkalay, I. (2002) Genes Dev 16(9), 1066-1076 10. Liu, C., Li, Y., Semenov, M., Han, C., Baeg, G. H., Tan, Y., Zhang, Z., Lin, X., and He, X.

(2002) Cell 108(6), 837-847 11. Yanagawa, S., Matsuda, Y., Lee, J. S., Matsubayashi, H., Sese, S., Kadowaki, T., and Ishimoto,

A. (2002) Embo J 21(7), 1733-1742 12. Hart, M., Concordet, J. P., Lassot, I., Albert, I., del los Santos, R., Durand, H., Perret, C.,

Rubinfeld, B., Margottin, F., Benarous, R., and Polakis, P. (1999) Curr Biol 9(4), 207-210 13. Liu, C., Kato, Y., Zhang, Z., Do, V. M., Yankner, B. A., and He, X. (1999) Proc Natl Acad Sci U

S A 96(11), 6273-6278 14. Spencer, E., Jiang, J., and Chen, Z. J. (1999) Genes Dev 13(3), 284-294 15. Winston, J. T., Strack, P., Beer-Romero, P., Chu, C. Y., Elledge, S. J., and Harper, J. W. (1999)

Genes Dev 13(3), 270-283 16. Behrens, J., Jerchow, B. A., Wurtele, M., Grimm, J., Asbrand, C., Wirtz, R., Kuhl, M., Wedlich,

D., and Birchmeier, W. (1998) Science 280(5363), 596-599 17. Rowan, A. J., Lamlum, H., Ilyas, M., Wheeler, J., Straub, J., Papadopoulou, A., Bicknell, D.,

Bodmer, W. F., and Tomlinson, I. P. (2000) Proc Natl Acad Sci U S A 97(7), 3352-3357 18. Smits, R., Kielman, M. F., Breukel, C., Zurcher, C., Neufeld, K., Jagmohan-Changur, S.,

Hofland, N., van Dijk, J., White, R., Edelmann, W., Kucherlapati, R., Khan, P. M., and Fodde, R. (1999) Genes Dev 13(10), 1309-1321

19. Ikeda, S., Kishida, S., Yamamoto, H., Murai, H., Koyama, S., and Kikuchi, A. (1998) Embo J 17(5), 1371-1384

20. Itoh, K., Krupnik, V. E., and Sokol, S. Y. (1998) Curr Biol 8(10), 591-594 21. Ha, N. C., Tonozuka, T., Stamos, J. L., Choi, H. J., and Weis, W. I. (2004) Mol Cell 15(4), 511-

521 22. Xing, Y., Clements, W. K., Kimelman, D., and Xu, W. (2003) Genes Dev 17(22), 2753-2764 23. Xing, Y., Clements, W. K., Le Trong, I., Hinds, T. R., Stenkamp, R., Kimelman, D., and Xu, W.

(2004) Mol Cell 15(4), 523-533 24. Henderson, B. R. (2000) Nat Cell Biol 2(9), 653-660 25. Henderson, B. R., and Fagotto, F. (2002) EMBO Rep 3(9), 834-839 26. Rosin-Arbesfeld, R., Townsley, F., and Bienz, M. (2000) Nature 406(6799), 1009-1012

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

8

27. Shih, I. M., Yu, J., He, T. C., Vogelstein, B., and Kinzler, K. W. (2000) Cancer Res 60(6), 1671-1676

28. Evans, P. M., and Liu, C. (2005) BMC Mol Biol 6, 22 29. Brocardo, M., Nathke, I. S., and Henderson, B. R. (2005) EMBO Rep 6(2), 184-190 30. Hinoi, T., Yamamoto, H., Kishida, M., Takada, S., Kishida, S., and Kikuchi, A. (2000) J Biol

Chem 275(44), 34399-34406 31. Nishisho, I., Nakamura, Y., Miyoshi, Y., Miki, Y., Ando, H., Horii, A., Koyama, K.,

Utsunomiya, J., Baba, S., and Hedge, P. (1991) Science 253(5020), 665-669 32. Wang, Z., Vogelstein, B., and Kinzler, K. W. (2003) Cancer Res 63(17), 5234-5235 33. Kitagawa, M., Hatakeyama, S., Shirane, M., Matsumoto, M., Ishida, N., Hattori, K., Nakamichi,

I., Kikuchi, A., Nakayama, K., and Nakayama, K. (1999) Embo J 18(9), 2401-2410 34. Sadot, E., Conacci-Sorrell, M., Zhurinsky, J., Shnizer, D., Lando, Z., Zharhary, D., Kam, Z., Ben-

Ze'ev, A., and Geiger, B. (2002) J Cell Sci 115(Pt 13), 2771-2780 35. Davies, M. L., Roberts, G. T., Spiller, D. G., and Wakeman, J. A. (2004) Oncogene 23(7), 1412-

1419 36. Krieghoff, E., Behrens, J., and Mayr, B. (2006) J Cell Sci 119(Pt 7), 1453-1463 37. Matsumoto, T., Iida, M., Kobori, Y., Mizuno, M., Nakamura, S., Hizawa, K., and Yao, T. (2002)

Gut 50(3), 402-404

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

9

FIGURE LEGENDS Fig. 1. β-catenin phosphorylation in colon cancer cells. A, Linear representation of different APC truncations in the colon cancer cell lines, SW480, DLD-1 and HT29. Colon cancer cell line, HCT116, contains wild type APC. B, The cytoplasmic lysates from SW480, DLD-1, HT29 and HCT116 cells were analyzed by Western blot using antibodies against β-catenin and phosphorylated β-catenin. Ab45 recognizes S45 phosphorylation and Ab(33+37)/41 recognizes S33/S37/T41 phosphorylation of β-catenin. C, The β-catenin immunoprecipitation products from HT29 and HCT116 cell lysates were treated with or without calf-intestine alkaline phosphatase (CIP) and analyzed with antibodies against β-catenin and phosphorylated β-catenin. Fig. 2. β-catenin ubiquitination/degradation in colon cancer cells. A, SW480, DLD-1, HT29 and HCT116 cells were treated with MG132 and analyzed with antibodies against β-catenin and phosphorylated β-catenin. The upper bands are ubiquitinated β-catenin (28, 32). B, HEK293T cells were transfected by Flag-tagged wild type β-catenin (WT) and mutant β-catenin, S33A (Serine 33 changed to Alanine ) and S→A (S33, S37, T41 and S45 were changed to Alanine), and then treated with MG132. Transfected β-catenin was detected by anti-Flag antibody. C, SW480 cells were infected with the APC adenovirus Ad-CBR, or the control adenovirus, Ad-GFP, and then treated with MG132. Cytoplasmic fractions were analyzed with antibodies against β-catenin and phosphorylated β-catenin. Fig. 3. APC domains required for β-catenin ubiquitination. A, Schematic representation of various truncated APC fragments incorporated into CS2MT as well as the Ad-Easy adenovirus vectors. The 15-amino acid repeats (A-C) and 20-amino acid repeats (1-7) are β-catenin binding domains and the SAMP repeats (SAMP1-3) are Axin binding domains. B, The expression of APC fragments in HEK293T cells were analyzed by Western blot using anti-Myc antibody. C, The APC fragments were transfected into SW480 cells with both the TopFlash reporter and the Renilla reporter. The luciferase reporter was analyzed and normalized with Renilla reporter. D, SW480 cells were infected with various APC Adenoviruses and then treated with MG132. β-catenin phosphorylation and ubiquitination were analyzed by Western blot using β-catenin antibody and phospho-specific antibodies against β-catenin. Fig. 4. β-catenin localization in various APC adenovirus infected SW480 cells. SW480 cells were infected with various APC adenoviruses or a control adenovirus. Cells were immunostained with a β-catenin antibody (red fluorescence) and the nucleus was counterstained with DAPI (blue fluorescence). Green fluorescence indicates the Adenovirus-infected cells. Arrows indicate β-catenin staining in the adenovirus-infected cells. Fig. 5. The function of 15aa repeats of APC. A, Schematic representation of APC fragments without 15aa repeats (APC-II5). B, Top: Luciferase reporter activity was analyzed in SW480 cells transfected with APC-II5. Bottom: The expression of APC-II5 detected by an anti-Myc antibody. C, SW480 cells were infected with the APC-II5 adenovirus and then treated with MG132. Cells were analyzed with antibodies against both β-catenin and phosphorylated β-catenin. D, Immunofluorescent staining of β-catenin in SW480 cells infected with APC-II5 Adenovirus. Arrows indicate β-catenin staining in the adenovirus-infected cells. Fig. 6. Mechanisms for APC mediated β-catenin ubiquitination. A, The nuclear and cytoplasmic fractions were isolated from SW480 and HT29 cells treated with or without MG132. β-catenin was analyzed by Western blot. Lamin B and HSP 90 were used as markers for the nuclear and cytoplasmic fractions, respectively. B, The interaction between β-Trcp and β-catenin was detected in HT29, DLD-1 and SW480 cells. C, The putative phosphorylation sites in the second and third 20aa repeats of APC were aligned with β-catenin phosphorylation sites T41 and S45. D, Top: GST-APC2,3, a GST fusion protein

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

10

containing the second and third 20aa repeats of APC, was in vitro phosphorylated by purified CKI. The band shift of phosphorylated GST-APC2,3 was detected by anti-GST antibody (WT). When the putative phosphorylation sites in the second and third 20aa repeats (according to Ha et al. (21)) were mutated to alanine (S→A), the band shift was greatly reduced. Bottom: Phosphorylated GST-APC2,3 (WT) and its mutant (S→A) were incubated with HCT116 cell lysates and pulled down using GST-beads, and β-catenin was analyzed by Western blot. CKI phosphorylation significantly enhanced the binding between β-catenin and APC2,3 (WT). S→A mutations of APC inhibited its ability to interact with β-catenin.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

11

Fig. 1.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

12

Fig. 2.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

13

Fig. 3.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

14

Fig. 4.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

15

Fig. 5.

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

16

Fig. 6

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from

Jun Yang, Wen Zhang, Paul M. Evans, Xi Chen, Xi He and Chunming Liucancer cells

-catenin phosphorylation and ubiquitination in colonβAPC differentially regulates

published online April 25, 2006J. Biol. Chem. 

  10.1074/jbc.M600831200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on Decem

ber 10, 2020http://w

ww

.jbc.org/D

ownloaded from