Purpose: Skp2 is a subunit of the SCF ubiquitin protein ligase, which plays a vital role in the control of tumorigenesis via its regulation of G1-S transition. Deregulation of Skp2 in various types of cancers correlates with aggressive clinical behavior and poor prognosis. Recent studies suggest that cell cycle–dependent fluctuation of Skp2 is governed by APCCdh1, another important E3 ligase, thereby preventing premature entry into S phase. To assess the potential role of APCCdh1 in tumorigenesis through proteolysis of Skp2, we have dissected the APCCdh1-Skp2 cascade.

Experimental Design: We manipulated the APCCdh1-Skp2 cascade and examined its cellular behavior using both breast cancer and normal breast epithelial cells. Furthermore, applying immunohistochemistry, we analyzed the clinicopathologic significance of these molecules in patients with breast cancer.

Results: Analysis of tissue arrays indicated that the percentage of samples positive for Cdh1 in breast cancer was significantly lower compared with normal breast tissues (P = 0.004). Conversely, the percentage of samples scored as positive for Skp2 in cancer was significantly higher than in normal tissues (P < 0.001). Moreover, prognostic studies revealed that relatively high levels of Cdh1 are associated with survivability in patients with breast cancer. In addition, depletion of Cdh1 by small interfering RNA in normal breast cells resulted in increased cellular proliferation, whereas knockdown of Skp2 significantly suppressed growth in breast cancer cells.

Conclusions: This study shows a correlation between Skp2 and APCCdh1 in breast cancer. Thus, Cdh1 may act as an important component in tumor suppression and could be considered as a novel biomarker in breast cancer.

Skp2 (S phase kinase–associated protein 2) has been suggested to be a promising biological marker for several types of cancer, which could also be used as a prognostic indicator in certain types of malignancy (1, 2). Skp2 in association with SCF complex (Skp1, Cullin, and F-box) targets substrate proteins for degradation, resulting in the regulation of cell cycle progression and other cellular processes (3). The major role of Skp2 in the control of the cell cycle is to ubiquitylate p27 for proteolysis, regulating the onset of S phase (4, 5). Both Skp2 transcripts and protein oscillate in a cell cycle–dependent manner (610). Recent studies have revealed that APC (anaphase-promoting complex), a pivotal E3 ligase, degrades Skp2 and thus, prevents abnormal entry into S phase (8, 9). Nevertheless, the correlation between APCCdh1 and Skp2 has not yet been validated in tumorigenesis.

APC is a multifunctional E3 ligase, regulating several critical cellular events including mitotic progression, DNA replication, cellular differentiation, genomic integrity, and signal transduction (1118). Activation of APC is controlled by two WD40 family proteins (e.g., Cdh1 and Cdc20). Cdh1, in association with APC, regulates APC function in G1 and postmitotic processes, whereas the interaction of Cdc20 with APC controls chromatid separation during mitosis (12, 1922). Recognition of the substrate by the substrate-specific activator (e.g., Cdh1 and Cdc20) is facilitated by several well-characterized degrons/recognition domains including destruction box (RXXL), KEN box and, A box present in the substrate (21). Recent evidence has drawn our attention to the connection between APC function and human diseases. Pathologic and epigenetic studies have shown that dysfunction in several components of the APC pathway including APC6, Cdc16, Cdc23, and Cdh1 or Cdc20, is correlated with different types of cancer such as colon cancer, B lymphoma, gastric, and lung cancer (2326). However, the underlying mechanism by which APC is involved in the aforementioned types of carcinogenesis remains largely unknown. Dissection of the APC pathway in human cancer will facilitate our understanding of APC in tumor progression.

Overexpression of Skp2 is often correlated with malignancy. Understanding the mechanism by which Skp2 protein levels is down-regulated could provide strategies for manipulating the status of tumor cells. The notion that Cdh1 targets Skp2 for degradation suggests a potential role for Cdh1 in the suppression of tumor growth. To validate the connection between Cdh1 and Skp2 in tumor formation, we have carried out a human breast cancer tissue array and prognostic analyses with the components of the APCCdh1-Skp2 cascade. In addition, using RNA interference technology, we have further dissected the function of Cdh1 and Skp2 in suppressing or enhancing cell growth in both normal and breast cancer cells. Our study shows that Cdh1 expression is inversely correlated with the expression of Skp2 in cancer and in normal condition. Suppressing Skp2 protein levels by enhancing Cdh1 function results in the inhibition of tumor cell growth. The present results suggest that Cdh1 may function as a critical component in the suppression of breast tumor.

Plasmid preparation and construction of small interfering RNA–stable cell lines. pCS2-Myc-Cdh1 and pCS2-HA-Skp2 plasmids have been engineered and reported previously (11). Stable knockdown using small interfering RNA (siRNA) for Cdh1 has also been previously reported (16). Construction of the Skp2-siRNA stable cell line 1 Skp2 NH2-terminal (5′-GAGGAGCCCGACAGTGAGA-3′) was engineered using pSUPER system (OligoEngine). Transfection was carried out using LipofectAMINE 2000 (Invitrogen) according to the manufacturer's protocol. Thereafter, positive clones were selected in the presence of puromycin-containing medium (Promega).

Antibodies and reagent. Antibodies were Cdh1 (Calbiochem), Skp2 (Santa Cruz), p27(Santa Cruz), tubulin (Calbiochem), and horseradish peroxidase–conjugated goat anti-mouse and anti-rabbit secondary antibody (Promega). 17β-Estradiol was from Sigma. Western blot analysis was done using an enhanced chemiluminescence detection kit (Amersham). Semiquantification of data was done using an image densimeter.

Cell cycle analysis. DNA fragmentation was measured by propidium iodide staining and fluorescence-activated cell sorting analysis. Following treatments, cells were harvested and pelleted by centrifugation, and were resuspended and fixed in ethanol. Cells were incubated in the propidium iodide solution (Sigma) with 5 μg/mL of RNase (Sigma). Flow cytometric analysis of stained cells was done using a FACScan (Becton Dickinson).

Colony formation by soft agar assay. Twenty-four hours after transfection, viable cells were counted until a concentration of 2.0 × 105/mL was achieved. Cells were seeded into soft agar as described previously (27), with slight modification (Dr. Flemington). Briefly, 1% agarose solution was made with sterile water, and 5 mL of agarose was added in a six-well plate until the plate was covered completely. Excess agarose was then removed with a pipette, leaving a thin film of agarose on the bottom and sides of each well and plate (2.0 × 105 cells/well). Colony formation was assessed by microscopic inspection (×10) and counting 7 days after cell seeding. Because the aggregates of the untreated MCF10A cells did not grow further throughout the experiment period, they were not considered as colonies. Each experiment was repeated at least thrice. The values given are the results of mean (± SD) value scores.

Bromodeoxyuridine labeling. The proliferative rate of cells grown was measured by assaying 5-bromo-2-deoxyuracil bromodeoxyuridine (BrdUrd) incorporation with commercially available labeling and detection kits (Roche Diagnosis). Briefly, 24 h after transfection, cells were spread and labeled nuclei were detected, according to the manufacturer's instructions. BrdUrd-labeled indices were determined by visually scoring nuclei stained with 4′,6-diamidino-2-phenylindole (Vector Laboratories) from 50 to 100 cells in 10 independent visual fields, and thereafter scoring BrdUrd-positive cells as a percentage of the total cell number (28). Each experiment was repeated at least thrice. Values given are the results of mean (± SD) value score.

Immunofluorescence. Immunofluorescence analysis was done using the following concentration of first antibodies: Cdh1 (rabbit anti-rat 1:500), Skp2 (rabbit anti-mouse 1:100). Second antibodies used were Cy2 (anti-mouse 1:500; Jackson ImmunoResearch), Texas red (anti-rat 1:100; Jackson ImmunoResearch). Semiquantification of data was done using Scion Image imaging software.

Immunohistochemical staining and prognostic analysis. Samples were deparaffinized in xylene and rehydrated in a series of graded alcohols, and the antigen was retrieved in 0.01 mol/L of sodium citrate buffer; thereafter, sections were treated with 0.6% hydrogen (29). Samples were incubated using rabbit anti-human APC2 antibody (1:150), rat anti-human Cdh1 antibody (1:150), rabbit anti-human Skp2 antibody (1:100), and rabbit anti-human p27 antibody (1:150). Sections were thereafter treated with biotinylated mouse anti-rat immunoglobulin (Jackson ImmunoResearch) and donkey anti-rabbit antibody (Vector Laboratories) followed by incubations with avidin-biotin peroxidase complex solution (DAKO Cytomation) and 3-amino-9-ethylcarbazole solution (DAKO Cytomation). The counterstaining was carried out using Mayer's hematoxylin (Sigma). Tissue arrays were purchased from U.S. Biomax. The expression of each molecule was tested in a breast cancer tissue array, which contained breast cancer tissues as well as the matched normal adjacent breast tissues of each patient. For patient survival analysis, human breast tissue samples were provided from Breast Tissue Bank in the Department of Cancer and Thoracic Surgery, Okayama University, Okayama, Japan. To verify the specificity and optimal concentration of the antibody, each antibody and its concentration was verified using the test tissue array slides (BR241t, BR804t).

Scoring of immunohistochemical staining. APC2, Cdh1, Skp2, and p27 immunohistochemical staining were examined under the microscope (Olympus), and staining intensity and subcellular localization were evaluated twice in a blinded manner based on the pre-agreed staining scoring standard from specialized pathologist (Dr. Roodman, Dr. Cheng). Staining intensity was scored separately using the following scoring criteria: (a) 0 to 1, negative or low staining intensity in >50% of tumor cells or moderate to high in <50% of the cells (hereafter referred to as low); and (b) 2 to 3, moderate to high staining intensity in >50% of tumor cells (hereafter referred to as high; ref. 30).

Statistical analysis. In the quantification of data, each value represents at least three independent experiments. Levels of statistical significance were evaluated with data from at least three independent experiments by using two-tailed Student's t test and χ2 test. Fisher's exact test and Spearman correlation test were used for statistical analysis of immunostaining results and analysis of clinicopathologic data. P < 0.05 was considered statistically significant. All data were analyzed with SPSS 14.0 for Windows.

APCCdh1 pathway is altered in cancer cell lines. Skp2 and p27 are thought to be important players involved in cancer formation, wherein Skp2 acts as an oncogenic protein usually enhancing carcinogenesis, whereas p27 often functions as a tumor suppressor inhibiting tumorigenesis (1). Both Skp2 and p27 are high-turnover proteins governed by the ubiquitin proteosome pathway with deregulation resulting in aberrant protein function and therefore leading to cancer (1). Current studies have implicated APCCdh1 in having a pivotal role in orchestrating the Skp2-p27 axis, thereby preventing precocious entry into S phase (8, 9, 11). To correlate the regulatory axis of APCCdh1-Skp2-p27 with malignant tumor status, we determined the protein levels of Cdh1, APC2, Skp2, and p27 in breast cancer and colorectal cancer cells relative to the respective normal cells. As shown in Fig. 1A and B, the protein expression levels of Cdh1 and p27 were lower in the breast tumor cell line (MCF7) compared with the normal breast epithelial cell line (MCF10A), whereas Skp2 protein levels were higher in breast cancer cells but lower in the normal breast cells. No significant difference was seen in the levels of APC2 between cancer cells and normal breast epithelial cells (Fig. 1A). Similar expression patterns for Cdh1, APC2, Skp2, and p27 were detected in colon cancer cells (HCT116) relative to normal colon cells (CCD18Co; Fig. 1A).

Fig. 1.

Expression of Cdh1, APC, Skp2, and p27 in normal and cancer cells. A, a, Cdh1 and p27 are down-regulated, whereas Skp2 is up-regulated in breast cancer and colorectal cancer cells. b, quantification of protein expression in normal and cancer cells. Results were analyzed from the data independently carried out thrice. B, immunocytochemical analysis of Cdh1 and Skp2 in normal breast epithelial and breast cancer cells. Both Cdh1 (red) and Skp2 (green) are expressed and colocalized in the nucleus. C, estimation of fluorescent signals reflecting the expression of Cdh1 and Skp2 in normal and cancer cells using surface plotting.

Fig. 1.

Expression of Cdh1, APC, Skp2, and p27 in normal and cancer cells. A, a, Cdh1 and p27 are down-regulated, whereas Skp2 is up-regulated in breast cancer and colorectal cancer cells. b, quantification of protein expression in normal and cancer cells. Results were analyzed from the data independently carried out thrice. B, immunocytochemical analysis of Cdh1 and Skp2 in normal breast epithelial and breast cancer cells. Both Cdh1 (red) and Skp2 (green) are expressed and colocalized in the nucleus. C, estimation of fluorescent signals reflecting the expression of Cdh1 and Skp2 in normal and cancer cells using surface plotting.

Close modal

To confirm the results obtained using immunoblotting, we studied MCF7 as well as MCF10A cells and performed immunocytochemical analyses. As shown in Fig. 1B, both Cdh1 and Skp2 are nuclear proteins and they mainly colocalize in the nucleus in both MCF7 and MCF10A cells. Quantification analyses showed that the expression of Cdh1 in normal breast cells was ∼40% higher than in breast cancer cells, whereas Skp2 was 55% lower in the normal breast cells than in the breast cancer cells (Fig. 1B and C). The results based on immunocytochemical analyses are consistent with the results from immunoblotting. Taken together, these results correlate the function of Cdh1, Skp2, and p27 with their protein expression status in tumorigenesis and further suggest that Cdh1 could potentially suppress tumor formation.

Depletion of Cdh1 or overexpression of Skp2 enhances cell proliferation in normal breast cell line. To confirm the role of Skp2 in promoting cellular growth and function of Cdh1 in suppressing proliferation (1, 8, 9, 31), we overexpressed Skp2 and depleted Cdh1 using RNA interference in normal breast cells. Upon the manipulation of the protein levels of Skp2 and Cdh1, we then measured the expression of downstream proteins and determined their effects on cellular oncogenesis and proliferation. As shown in Fig. 2A, overexpression of Skp2 in MCF10A cells significantly reduced p27 levels. Similarly, depletion of Cdh1 using RNA interference significantly elevated Skp2 levels, which then leads to a drop in p27 expression levels for MCF10A cells (Fig. 2B).

Fig. 2.

Effect of overexpression of Skp2 or depletion of Cdh1 in normal breast epithelial cells. A, a, overexpression of Skp2 results in the down-regulation of p27 in normal breast epithelial cells. b, quantification of expression of Skp2 and p27. Summary of the three independent experiments. B, a, depletion of Cdh1 by siRNA. b, knockdown of Cdh1 leads to up-regulation of Skp2 and down-regulation of p27. c, quantification of a Cdh1 knockdown experiment. Results were analyzed from data independently carried out thrice. C, a, depletion of Cdh1 or overexpression of Skp2 promotes anchorage-independent growth in normal breast epithelial cells. b, quantification of anchorage-independent growth analysis from three independent experiments (columns, mean; bars, SD). D, a, effect of Cdh1 knockdown and overexpression of Skp2 on the cell cycle profile of normal breast epithelial cells. b, effect of Cdh1 knockdown or overexpression of Skp2 on cellular proliferation of normal breast epithelial cells measured by BrdUrd analysis. c, quantification of BrdUrd analysis from three independent experiments (columns, mean; bars, SD).

Fig. 2.

Effect of overexpression of Skp2 or depletion of Cdh1 in normal breast epithelial cells. A, a, overexpression of Skp2 results in the down-regulation of p27 in normal breast epithelial cells. b, quantification of expression of Skp2 and p27. Summary of the three independent experiments. B, a, depletion of Cdh1 by siRNA. b, knockdown of Cdh1 leads to up-regulation of Skp2 and down-regulation of p27. c, quantification of a Cdh1 knockdown experiment. Results were analyzed from data independently carried out thrice. C, a, depletion of Cdh1 or overexpression of Skp2 promotes anchorage-independent growth in normal breast epithelial cells. b, quantification of anchorage-independent growth analysis from three independent experiments (columns, mean; bars, SD). D, a, effect of Cdh1 knockdown and overexpression of Skp2 on the cell cycle profile of normal breast epithelial cells. b, effect of Cdh1 knockdown or overexpression of Skp2 on cellular proliferation of normal breast epithelial cells measured by BrdUrd analysis. c, quantification of BrdUrd analysis from three independent experiments (columns, mean; bars, SD).

Close modal

To examine the ability for oncogenic colony formation and the capacity for accelerating cell growth in response to alteration of Cdh1 and Skp2, we conducted an anchorage-independent growth assay and further measured cell cycle profiles as well as BrdUrd-positive cells using MCF10A cells. We engineered a Cdh1 siRNA-stable cell line as indicated in Fig. 2B (11, 16). As shown in Fig. 2C, either depletion of Cdh1 or overexpression of Skp2 in MCF10A cells promoted the increase in number and size of the colony in soft agar, whereas nonappreciable colonies were found for control cells.

Results from the fluorescence-activated cell sorting analyses showed that either depletion of Cdh1 or overexpression of Skp2 in MCF10A cells significantly elevated the fraction of cells in S phase (Fig. 2D, a). Analysis based on BrdUrd staining showed that either depletion of Cdh1 or overexpression of Skp2 resulted in an increase of BrdUrd-positive staining cells (Fig. 2D, b and c). In summary, the results based on the above analyses suggest that an increase in Skp2 protein levels or loss of Cdh1 could lead to an aberrant cell cycle, which in turn, induces oncogenesis in normal breast cells.

Depletion of Skp2 or overexpression of Cdh1 slow down cellular growth in breast cancer cell line. To confirm the above results in normal breast cells, we also depleted Skp2 or overexpressed Cdh1 in breast cancer cells and subsequently examined its effects on colony formation as well as proliferation. We established a Skp2 siRNA-stable cell line (Fig. 3B; refs. 11, 16). Predictably, depletion of Skp2 resulted in a significant increase in p27 levels (Fig. 3B; Supplementary Fig. S1A and B). Moreover, overexpression of Cdh1 in MCF7 cells largely reduced Skp2 protein levels leading to elevation of p27 abundance (Fig. 3A; Supplementary Fig. S1A and B).

Fig. 3.

Effect of overexpression of Cdh1 or depletion of Skp2 in breast cancer cells. A, a, overexpression of Cdh1 results in the down-regulation of Skp2 in breast cancer cells. b, summary of three independent experiments. B, a, engineering of Skp2 siRNA clones. b, knockdown of Skp2 results in the elevation of p27. c, quantification of a Skp2 knockdown experiment. Results were analyzed from data independently carried out thrice. C, a, depletion of Skp2 or overexpression of Cdh1 suppresses anchorage-independent growth in breast cancer cells. b, quantification of anchorage-independent growth analysis from three independent experiments (columns, mean; bars, SD). D, a, effect of Skp2 knockdown and overexpression of Cdh1 on the cell cycle profile of breast cancer cells. b, effect of Skp2 knockdown or overexpression of Cdh1 on cellular proliferation of breast cancer cells measured by BrdUrd analysis. c, quantification from three independent experiments (columns, mean; bars, SD).

Fig. 3.

Effect of overexpression of Cdh1 or depletion of Skp2 in breast cancer cells. A, a, overexpression of Cdh1 results in the down-regulation of Skp2 in breast cancer cells. b, summary of three independent experiments. B, a, engineering of Skp2 siRNA clones. b, knockdown of Skp2 results in the elevation of p27. c, quantification of a Skp2 knockdown experiment. Results were analyzed from data independently carried out thrice. C, a, depletion of Skp2 or overexpression of Cdh1 suppresses anchorage-independent growth in breast cancer cells. b, quantification of anchorage-independent growth analysis from three independent experiments (columns, mean; bars, SD). D, a, effect of Skp2 knockdown and overexpression of Cdh1 on the cell cycle profile of breast cancer cells. b, effect of Skp2 knockdown or overexpression of Cdh1 on cellular proliferation of breast cancer cells measured by BrdUrd analysis. c, quantification from three independent experiments (columns, mean; bars, SD).

Close modal

We next analyzed oncogenic colony formation and the capacity of accelerating cell growth using MCF7 with either Cdh1 overexpressed or Skp2 depleted. As shown in Fig. 3C, MCF7 cells have a great capacity for colony formation in soft agar, although both the number and the size of the colony were significantly decreased when Cdh1 was overexpressed or Skp2 was depleted in MCF7 cells, confirming that the major effects of Cdh1 in p27 regulation are mediated via Skp2 proteolysis (Supplementary Fig. S1A-C).

Consistent with the results based on the anchorage-independent growth assay, fluorescence-activated cell sorting analysis showed that the fraction of cells in S phase was significantly reduced in Cdh1-overexpressed or Skp2-depleted MCF7 cells (Fig. 3D, a). In addition, the results based on BrdUrd staining showed that either overexpression of Cdh1 or depletion of Skp2 led to a significant reduction of BrdUrd-positive staining in MCF7 cells (Fig. 3D, b and c). Taken together, the above results further suggest that Skp2 promotes oncogenic proliferation, whereas Cdh1 could potentially orchestrate Skp2, thereby suppressing the acceleration of tumor growth.

Tissue array analysis of human breast tissue in the APCCdh1 cascade. Our results, based on the loss of function analyses of APCCdh1-Skp2-p27 using RNA interference in combination with overexpression analyses in both normal and breast cancer cells, showed that Cdh1 plays a pivotal role in dictating Skp2-p27 function in cellular proliferation for breast tumor cells. These results suggest that deregulation of Cdh1 could contribute to aberration in Skp2/p27 function in breast cancer.

To determine the importance of the molecular pathway of APCCdh1-Skp2-p27 in controlling G1-S transition in the pathology of breast cancer, we have done human tissue arrays using 325 breast tumors (U.S. Biomax). As shown in Fig. 4A, no significant difference in APC2 expression was observed between normal breast tissue and breast cancer, whereas a significantly higher frequency of positive Cdh1 expression was detected in normal breast tissue with prominent nuclei localization compared with breast cancer tissue (Fig. 4A). Meanwhile, a higher frequency of positive Skp2 expressions were observed in breast cancer tissue, whereas a lower frequency of p27 expressions were measured in the cancer area (Fig. 4A). Furthermore, statistical analyses have shown that no difference existed in the expression of APC2 between cancer and normal tissues (P = 0.483). However, significant differences were seen in Cdh1 (P = 0.004), Skp2 (P < 0.001), and p27 (P = 0.017) between breast cancer and normal breast tissue (Table 1A). The above results show the inverse correlation between the expression levels of Cdh1 and Skp2, in which breast cancer cells have a lower expression of Cdh1 and p27 but higher expression of Skp2. Overall, the results based on the human breast tissue array are consistent with the results from the cell culture studies (8, 9).

Fig. 4.

Human tissue array analysis of proteins in the APCCdh1-Skp2-p27 cascade. A, a significant reduction of Cdh1 protein levels was measured in breast cancer tissue, whereas abundant Cdh1 proteins were detected in normal breast tissue. Expression of APC2, a key subunit of the APC, showed similar levels between normal and breast cancer tissue. The accumulation of Skp2 was observed in breast cancer tissue, whereas moderate Skp2 protein levels were found in normal breast tissue. Lower levels of p27 protein were examined in breast cancer tissue, whereas abundant p27 protein was tested in normal breast tissue. B, distribution of Cdh1-positive breast cancer in each histologic grade, showing the correlation between higher population of positive Cdh1 cancer with early histologic grade of breast tumor.

Fig. 4.

Human tissue array analysis of proteins in the APCCdh1-Skp2-p27 cascade. A, a significant reduction of Cdh1 protein levels was measured in breast cancer tissue, whereas abundant Cdh1 proteins were detected in normal breast tissue. Expression of APC2, a key subunit of the APC, showed similar levels between normal and breast cancer tissue. The accumulation of Skp2 was observed in breast cancer tissue, whereas moderate Skp2 protein levels were found in normal breast tissue. Lower levels of p27 protein were examined in breast cancer tissue, whereas abundant p27 protein was tested in normal breast tissue. B, distribution of Cdh1-positive breast cancer in each histologic grade, showing the correlation between higher population of positive Cdh1 cancer with early histologic grade of breast tumor.

Close modal
Table 1.

Analysis of tissue arrays from breast cancer and normal breast epithelial tissue

(A)
Normal breast tissue
Breast cancer
P
+1+
APC2 67.7% 32.3% 61.1% 38.9% N.S. (0.483) 
Cdh1 40.7% 59.3% 72.2% 27.8% 0.004 
Skp2 85.1% 14.9% 42.6% 57.4% <0.001 
p27 45.9% 54.1% 72.5% 27.5% 0.017 
    
(B)
 
   
Variables
 
Cdh1 (−)
 
Cdh1 (+)
 
P
 
Age   N.S. (0.468) 
    <50 38.9% 11.1%  
    ≤50 35.6% 14.4%  
Tumor size   N.S. (0.878) 
    T1, T2 44.4% 11.1%  
    T3 14.4% 4.5%  
    T4 13.3% 5.6%  
    No data 2.2% 4.4%  
Lymph node metastasis   N.S. (0.791) 
    N (−) 26.7% 7.8%  
    N (+) 45.6% 13.3%  
    No data 2.2% 4.4%  
Distant metastasis   N.S. (0.122) 
    M (−) 60.0% 13.3%  
    M (+) 12.2% 7.8%  
    No data 2.2% 4.4%  
Histologic grade   0.027 
    G1 2.2% 4.4%  
    G2, G3 67.8% 17.8%  
    No data 4.4% 3.3%  
Stage   N.S. (0.619) 
    I, II 38.9% 7.8%  
    III 21.1% 8.9%  
    IV 12.2% 4.4%  
    No data 2.2% 4.4%  
(A)
Normal breast tissue
Breast cancer
P
+1+
APC2 67.7% 32.3% 61.1% 38.9% N.S. (0.483) 
Cdh1 40.7% 59.3% 72.2% 27.8% 0.004 
Skp2 85.1% 14.9% 42.6% 57.4% <0.001 
p27 45.9% 54.1% 72.5% 27.5% 0.017 
    
(B)
 
   
Variables
 
Cdh1 (−)
 
Cdh1 (+)
 
P
 
Age   N.S. (0.468) 
    <50 38.9% 11.1%  
    ≤50 35.6% 14.4%  
Tumor size   N.S. (0.878) 
    T1, T2 44.4% 11.1%  
    T3 14.4% 4.5%  
    T4 13.3% 5.6%  
    No data 2.2% 4.4%  
Lymph node metastasis   N.S. (0.791) 
    N (−) 26.7% 7.8%  
    N (+) 45.6% 13.3%  
    No data 2.2% 4.4%  
Distant metastasis   N.S. (0.122) 
    M (−) 60.0% 13.3%  
    M (+) 12.2% 7.8%  
    No data 2.2% 4.4%  
Histologic grade   0.027 
    G1 2.2% 4.4%  
    G2, G3 67.8% 17.8%  
    No data 4.4% 3.3%  
Stage   N.S. (0.619) 
    I, II 38.9% 7.8%  
    III 21.1% 8.9%  
    IV 12.2% 4.4%  
    No data 2.2% 4.4%  

Abbreviations: +, positive staining; −, negative staining; N.S., not significant.

Additional pathologic analyses showed that no significant differences were observed in the tumor size, lymph node metastasis, distant metastasis, or stages between Cdh1-positive and -negative patients (Table 1B). Interestingly, positive Cdh1 is more frequently observed in high-grade tumors with statistically significant difference seen between grade 1 tumor (Cdh1+; 66.6%) and grade 2 and 3 tumors (Cdh1+; 20.7%; P = 0.027, χ2 test; Fig. 4B). Results from the clinicopathologic analysis indicate that abundant Cdh1 is correlated with low histologic grade tumor, and therefore, suggest that loss of Cdh1 could be associated with aggressive cellular behavior and potentially poor prognosis for patients with breast cancer.

Prognostic implication of APCCdh1 in patients with breast cancer. The results from the tissue array suggest a potential function for Cdh1 in suppressing breast tumor progression. To correlate such results based on the molecular dissection to prognostic relevance, we have carried out a prognostic analysis for the APCCdh1-Skp2-p27 axis for patients with breast cancer. We analyzed disease-free survival (DFS) for patients with breast cancer. A different set of 105 breast cancer samples comprising 54 stage I cases, 30 stage II cases, 12 stage III cases, 6 stage IV cases, and 3 unknown, with at least 1 year (mean, 613.1 days) of follow-up, was independently analyzed immunohistochemically.

Among the 105 samples, 31 (29.5%) of the cases detected were Cdh1 positive, which is consistent with observations in tissue arrays (Table 1A). There were no significant differences in patient's backgrounds, including treatment strategy, between these groups. DFS in these patients are shown in Fig. 5, with the time of failure being 755 days (25%) and 1,764 days (75%), respectively, and the median time of DFS being 1,190 days (Fig. 5A). Meanwhile, analyses of additional breast cancer markers, including estrogen receptor, human epidermal growth factor receptor type 2, as well as the relevant disease stage were included in the DFS assay for Cdh1 (Fig. 5A, Supplementary Table S1). As to the status of Cdh1 and stage, the results from the Kaplan-Meier analysis showed significant differences in DFS of the stratified patients (log-rank P = 0.028; hazard ratio, 0.47; 95% confidence interval, 0.187-0.923 according to Cdh1; log-rank P < 0.001; hazard ratio, 0.20; 95% confidence interval, 0.076-0.452 according to stage; Fig. 5). In the Cdh1-positive population, the median time of failure was 1,390 days, whereas 968 days is the median time of failure in Cdh1 negative patients.

Fig. 5.

Summary of DFS analysis of patients according to Cdh1 levels as well as stage of disease. A, expression profile of APCCdh1-Skp2-p27 cascade in breast cancer tisseus and their adjacent normal breast epithelial tissues. B, correlation of Cdh1 and clinicopathologic variables in patients with breast cancer. Cdh1 status was significantly correlated with DFS duration (B and D).

Fig. 5.

Summary of DFS analysis of patients according to Cdh1 levels as well as stage of disease. A, expression profile of APCCdh1-Skp2-p27 cascade in breast cancer tisseus and their adjacent normal breast epithelial tissues. B, correlation of Cdh1 and clinicopathologic variables in patients with breast cancer. Cdh1 status was significantly correlated with DFS duration (B and D).

Close modal

To further address the benefit of Cdh1 in the same stage of disease, we only analyzed the DFS for the early stage (stage I and II) populations. As shown in Fig. 5D, the Cdh1-positive group showed a favorable outcome (log-rank P = 0.0699; hazard ratio, 0.41; 95% confidence interval, 0.095-1.003), which means that DFS in the Cdh1-positive group was 3,712 days, whereas 968 days was measured in the Cdh1-negative group. Taken together, these results indicate that Cdh1 has an appreciable function in suppressing breast tumorigenesis and could be a potential prognostic molecule in patients with breast cancer.

Regulation of G1-S transition during the cell cycle through Skp2-p27-cyclin E/CDK2 is thought to be an important mechanism governing tumorigenesis (32). Loss of control for this regulatory cascade has been implicated in multiple types of cancers (1, 31). Previous studies have revealed that aberration in components of the above pathway tightly correlate with different types of breast cancers (1). Recent demonstration that Cdh1 is a critical factor of the E3 ligase in orchestrating the Skp2-p27 axis sheds a light on its potential clinical relevance (8, 9). The present work validates the role of Cdh1 in the modulation of cell proliferation in normal and transformed breast cells, and further examined its expression in breast cancer tissue as well as its prognostic effect. Our results suggest that Cdh1, in association with APC, may contribute to the deregulation of the Skp2-p27 axis in breast cancer.

Proteolytic control of Skp2-p27 cascade in breast cancer. Aberrant protein levels of Skp2 and p27 are thought to be an important molecular basis for several types of carcinogenesis because dysregulated Skp2 and p27 often lead to abnormal cell cycle progression (8, 9). Identification of APCCdh1 as the upstream ubiquitin protein ligase governing the turnover of Skp2 during G1-S transition unveils the molecular mechanism for orchestrating Skp2 function via proteolytic regulation. This leads to the question as to whether Cdh1 could be a potential suppressing component coordinating Skp2-p27 for the control of cyclin E/CDK2 in breast cancer. To assess the involvement of Cdh1 in modulating Skp2-p27 in breast cancer, we conducted loss of function analyses of Cdh1 as well as Skp2 in either normal or breast cancer cells accompanied with the overexpression of Cdh1 and Skp2 in such cells. Data strongly suggest that in both normal and breast cancer cells, the alteration of Cdh1 leads to the correlatively opposite protein expression pattern for Skp2, and correlatively similar protein expression for p27. Manipulation of Cdh1 by siRNA or overexpression induces a significant change in the number of BrdUrd-positive cells as well as the property of oncogenic growth on soft agar in both normal and breast cancer cells which is consistent with its known effects in the regulation of Skp2. 17β-Estradiol is also reported to be involved in the regulation of Skp2 and subsequent p27 level in MCF7 cells (33), however, we could not find a clear connection between Cdh1 and 17β-estradiol in this study (Supplementary Fig. S2A-D). Therefore, different molecular pathways may be involved in the regulation of Skp2, in which regulation of Skp2 via estrogen receptor is believed to be at the transcriptional level, whereas alteration of Skp2 by APCCdh1 is via the ubiquitin-proteasomal pathway.

Prognostic effect of Cdh1 in breast cancer. The results of examination of Cdh1, Skp2, p27, and related breast cancer markers such as human epidermal growth factor receptor type 2 and estrogen receptor based on >300 human breast tumors using immunohistochemistry provide the notion that Cdh1 could antagonize tumorigenesis via down-regulation of Skp2, thereby leading to up-regulation of p27 (1, 31). Prognostic analysis based on information from >100 patients unveils that patients with positive Cdh1 have a significant increase in survival time, which is consistent with the findings that the APCCdh1-Skp2-p27 cascade is critical to preventing immature entry into S phase and coordinating appropriate cell cycle progression (8, 9). The present assessment suggests that the expression levels of Cdh1 could be a potential prognostic indicator in patients with breast cancer. A previous report which suggests that lower levels of SnoN, an APCCdh1 substrate in the transforming growth factor-β pathway, as a prognostic marker of estrogen receptor–positive breast cancer supports this present finding (30). The current implication that APCCdh1 could stabilize p27 through the degradation of Skp2 in response to transforming growth factor-β signaling further explained the prognostic value of Cdh1 protein levels in patients with breast cancer (11).

Potential of the APCCdh1 pathway involved in breast cancer. The function of APC has been initially characterized in faithfully ensuring the separation of duplicated daughter genomes during mitosis in which dysfunction of APC could often result in aneuploidy, one of the hallmarks of cancer (21, 22). The results from epigenetic studies have further implicated the correlation of APC with tumorigenesis in which a deregulation in the components of the APC pathway including APC6, Cdc16, Cdc23, and Cdh1 are found in different types of cancer such as colon cancer, B lymphoma, and gastric and lung cancers (2326). Current pathologic analysis has shown that aberrant APC expression are present in multiple types of malignant tumors (34). The finding that APC mediates transforming growth factor-β signaling targeting SnoN as well as Skp2 unveiled the potential role of APCCdh1 in tumor inhibition (11, 17). Previous evidence from several lines has sketched a framework for APC in tumor progression. The present results based on human breast cancer tissue array and prognostic dissection of the APCCdh1-Skp2-p27 cascade have confirmed results obtained from studies based on cultured cells with pathologic relevance, which further suggests the role of the APC pathway in tumorigenesis.

Integration of the present finding with current paradigm. Ubiquitin-dependent proteolysis facilitates normal cell cycle progression. Malfunction of the ubiquitin-proteasome pathway could result in carcinogenesis by disrupting the balance between oncoproteins and tumor suppressor proteins (3537). Mitotic regulation and G1-S transition are key regulatory sites during the cell cycle, in which their aberration usually leads to genomic instability and uncontrolled growth. APC and SCF complexes are critical E3 ligases, dictating chromatid separation during mitosis and orchestrating cyclin E/CDK2 during G1-S transition. Loss of control for these two major ubiquitin-mediated pathways has been correlated with a variety of malignancies (38). This work shows the importance of an ubiquitylation-regulatory cascade in tumorigenesis showing how one E3 ligase (APC) could regulate another E3 ligase (SCF) complex resulting in scheduled G1-S progression. Analyses based on human specimens validated the molecular paradigm of APCCdh1-Skp2-p27 in breast cancer formation. The expression pattern of APCCdh1-Skp2-p27 in breast cancer samples supports the mechanism of this regulatory axis in the control of cellular proliferation with its dysfunction leading to carcinogenesis.

For further dissection of the role of Cdh1 in governing Skp2-p27 in breast tumorigenesis, a xenograft human breast cancer model is necessary. Indeed, to clarify the potential tumor suppressor role of Cdh1, a breast cancer mouse xenograft study is currently under way (data not shown). Combinatorial studies based on biochemistry, mouse dissection, and pathologic analysis will advance our understanding of APCCdh1 in breast cancer.

Grant support: NIH grants CA115943. Y. Wan is a scholar of the American Cancer Society and V Cancer Research Foundation.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

We thank Wan laboratory members for critical discussion and reading of the manuscript; the Yong-Tae Kwon, Shiyuan Cheng, and David Roodman laboratories for assisting us in tissue array analysis; and Richard D. Wood for critical discussion and reading of the manuscript.

1
Signoretti S, Di Marcotullio L, Richardson A, et al. Oncogenic role of the ubiquitin ligase subunit Skp2 in human breast cancer.
J Clin Invest
2002
;
110
:
633
–41.
2
Oliveira AM, Okuno SH, Nascimento AG, Lloyd RV. Skp2 protein expression in soft tissue sarcomas.
J Clin Oncol
2003
;
21
:
722
–7.
3
Pagano M, Tam SW, Theodoras AM, et al. Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27.
Science
1995
;
269
:
682
–5.
4
Carrano AC, Eytan E, Hershko A, Pagano M. SKP2 is required for ubiquitin-mediated degradation of the CDK inhibitor p27.
Nat Cell Biol
1999
;
1
:
193
–9.
5
Sutterluty H, Chatelain E, Marti A, et al. p45SKP2 promotes p27Kip1 degradation and induces S phase in quiescent cells.
Nat Cell Biol
1999
;
1
:
207
–14.
6
Zhang H, Kobayashi R, Galaktionov K, Beach D. p19Skp1 and p45Skp2 are essential elements of the cyclin A-CDK2 S phase kinase.
Cell
1995
;
82
:
915
–25.
7
Wirbelauer C, Sutterluty H, Blondel M, et al. The F-box protein Skp2 is a ubiquitylation target of a Cul1-based core ubiquitin ligase complex: evidence for a role of Cul1 in the suppression of Skp2 expression in quiescent fibroblasts.
EMBO J
2000
;
19
:
5362
–75.
8
Bashir T, Dorrello NV, Amador V, Guardavaccaro D, Pagano M. Control of the SCF (Skp2-1) ubiquitin ligase by the APC/C(Cdh1) ubiquitin ligase.
Nature
2004
;
428
:
190
–3.
9
Wei W, Ayad NG, Wan Y, Zhang GJ, Kirschner MW, Kaelin WG, Jr. Degradation of the SCF component Skp2 in cell-cycle phase G1 by the anaphase-promoting complex.
Nature
2004
;
428
:
194
–8.
10
Wang W, Ungermannova D, Jin J, Harper JW, Liu X. Negative regulation of SCFSkp2 ubiquitin ligase by TGF-β signaling.
Oncogene
2004
;
23
:
1064
–75.
11
Liu W, Wu G, Li W, Lobur D, Wan Y. Cdh1-anaphase-promoting complex targets Skp2 for destruction in transforming growth factor β-induced growth inhibition.
Mol Cell Biol
2007
;
27
:
2967
–79.
12
Jackson PK. Linking tumor suppression, DNA damage and the anaphase-promoting complex.
Trends Cell Biol
2004
;
14
:
331
–4.
13
Pagano M. Cell cycle regulation by the ubiquitin pathway.
FASEB J
1997
;
11
:
1067
–75.
14
Stewart S, Fang G. Anaphase-promoting complex/cyclosome controls the stability of TPX2 during mitotic exit.
Mol Cell Biol
2005
;
25
:
10516
–27.
15
Ayad NG, Rankin S, Murakami M, Jebanathirajah J, Gygi S, Kirschner MW. Tome-1, a Trigger of mitotic entry, is degraded during G1 via the APC.
Cell
2003
;
113
:
101
–13.
16
Wu G, Glickstein S, Liu W, et al. The anaphase-promoting complex coordinates initiation of lens differentiation.
Mol Biol Cell
2007
;
18
:
1018
–29.
17
Wan Y, Liu X, Kirschner MW. The anaphase-promoting complex mediates TGF-β signaling by targeting SnoN for destruction.
Mol Cell
2001
;
8
:
1027
–39.
18
Bonni S, Wang HR, Causing CG, et al. TGF-β induces assembly of a Smad2-2 ubiquitin ligase complex that targets SnoN for degradation.
Nat Cell Biol
2001
;
3
:
587
–95.
19
Peters JM. The anaphase-promoting complex: proteolysis in mitosis and beyond.
Mol Cell
2002
;
9
:
931
–43.
20
Fang G, Yu H, Kirschner MW. Direct binding of CDC20 protein family members activates the anaphase-promoting complex in mitosis and G1.
Mol Cell
1998
;
2
:
163
–71.
21
Harper JW, Burton JL, Solomon MJ. The anaphase-promoting complex: it's not just for mitosis any more.
Genes Dev
2002
;
16
:
2179
–206.
22
Yu H. Regulation of APC-Cdc20 by the spindle checkpoint.
Curr Opin Cell Biol
2002
;
14
:
706
–14.
23
Wang Q, Moyret-Lalle C, Couzon F, et al. Alterations of anaphase-promoting complex genes in human colon cancer cells.
Oncogene
2003
;
22
:
1486
–90.
24
Wang CX, Fisk BC, Wadehra M, Su H, Braun J. Overexpression of murine fizzy-related (fzr) increases natural killer cell-mediated cell death and suppresses tumor growth.
Blood
2000
;
96
:
259
–63.
25
Singhal S, Amin KM, Kruklitis R, et al. Alterations in cell cycle genes in early stage lung adenocarcinoma identified by expression profiling.
Cancer Biol Ther
2003
;
2
:
291
–8.
26
Kim JM, Sohn HY, Yoon SY, et al. Identification of gastric cancer-related genes using a cDNA microarray containing novel expressed sequence tags expressed in gastric cancer cells.
Clin Cancer Res
2005
;
11
:
473
–82.
27
Kolligs FT, Hu G, Dang CV, Fearon ER. Neoplastic transformation of RK3E by mutant β-catenin requires deregulation of Tcf/Lef transcription but not activation of c-myc expression.
Mol Cell Biol
1999
;
19
:
5696
–706.
28
Wang XJ, Liefer KM, Tsai S, O'Malley BW, Roop DR. Development of gene-switch transgenic mice that inducibly express transforming growth factor β1 in the epidermis.
Proc Natl Acad Sci U S A
1999
;
96
:
8483
–8.
29
Guo P, Imanishi Y, Cackowski FC, et al. Up-regulation of angiopoietin-2, matrix metalloprotease-2, membrane type 1 metalloprotease, and laminin 5γ2 correlates with the invasiveness of human glioma.
Am J Pathol
2005
;
166
:
877
–90.
30
Zhang F, Lundin M, Ristimaki A, et al. Ski-related novel protein N (SnoN), a negative controller of transforming growth factor-β signaling, is a prognostic marker in estrogen receptor-positive breast carcinomas.
Cancer Res
2003
;
63
:
5005
–10.
31
Masuda TA, Inoue H, Sonoda H, et al. Clinical and biological significance of S-phase kinase-associated protein 2 (Skp2) gene expression in gastric carcinoma: modulation of malignant phenotype by Skp2 overexpression, possibly via p27 proteolysis.
Cancer Res
2002
;
62
:
3819
–25.
32
Pei XH, Xiong Y. Biochemical and cellular mechanisms of mammalian CDK inhibitors: a few unresolved issues.
Oncogene
2005
;
24
:
2787
–95.
33
Foster JS, Fernando RI, Ishida N, Nakayama KI, Wimalasena J. Estrogens down-regulate p27Kip1 in breast cancer cells through Skp2 and through nuclear export mediated by the ERK pathway.
J Biol Chem
2003
;
278
:
41355
–66.
34
Lehman NL, Tibshirani R, Hsu JY, et al. Oncogenic regulators and substrates of the anaphase promoting complex/cyclosome are frequently overexpressed in malignant tumors.
Am J Pathol
2007
;
170
:
1793
–805.
35
Weissman AM. Themes and variations on ubiquitylation.
Nat Rev Mol Cell Biol
2001
;
2
:
169
–78.
36
Hoeller D, Hecker CM, Dikic I. Ubiquitin and ubiquitin-like proteins in cancer pathogenesis.
Nat Rev Cancer
2006
;
6
:
776
–88.
37
Datto M, Wang XF. Ubiquitin-mediated degradation: a mechanism for fine-tuning TGF-β signaling.
Cell
2005
;
121
:
2
–4.
38
Nakayama KI, Nakayama K. Ubiquitin ligases: cell-cycle control and cancer.
Nat Rev Cancer
2006
;
6
:
369
–81.