PIK3CA and APC Mutations Are Synergistic in the Development of Intestinal Cancers

ORIGINAL ARTICLE PIK3CA and APC mutations are synergistic in the development of intestinal cancers

DA Deming1, AA Leystra2, L Nettekoven3, C Sievers4, D Miller3, M Middlebrooks3, L Clipson2, D Albrecht3, J Bacher5, MK Washington6, J Weichert7 and RB Halberg3

Human colorectal cancers are known to possess multiple mutations, though how these mutations interact in tumor development and progression has not been fully investigated. We have previously described the FCPIK3ca* murine colon cancer model, which expresses a constitutively activated phosphoinositide-3 kinase (PI3K) in the intestinal epithelium. The expression of this dominantly active form of PI3K results in hyperplasia and invasive mucinous adenocarcinomas. These cancers form via a non-canonical mechanism of tumor initiation that is mediated through activation of PI3K and not through aberrations in WNT signaling. Since the Adenomatous Polyposis Coli (APC) gene is mutated in the majority of human colon cancers and often occurs simultaneously with PIK3CA mutations, we sought to better understand the interaction between APC and PIK3CA mutations in the mammalian intestine. In this study, we have generated mice in which the expression of a constitutively active PI3K and the loss of APC occur simultaneously in the distal small intestine and colon. Here, we demonstrate that expression of a dominant active PI3K synergizes with loss of APC activity resulting in a dramatic change in tumor multiplicity, size, morphology and invasiveness. Activation of the PI3K pathway is not able to directly activate WNT signaling through the nuclear localization of CTNNB1 (b-catenin) in the absence of aberrant WNT signaling. Alterations at the transcriptional level, including increased CCND1, may be the etiology of synergy between these activated pathways.

Oncogene (2014) 33, 2245–2254; doi:10.1038/onc.2013.167; published online 27 May 2013 Keywords: colon cancer; APC; PI3K; PIK3CA; mouse models

INTRODUCTION cancer types, including 20–30% of colorectal cancers.8 Mutations Colorectal adenocarcinoma continues to be a major etiology of of the PIK3CA gene encode for a dominant active form of the morbidity and mortality despite significant advances in the p110a catalytic subunit of PI3K and occur in three hotspots: E454K, understanding of tumor biology and treatment options.1 The E456K and H1047R.9 PIK3CA mutations have been investigated in profile of mutations has now been characterized in multiple numerous cancer cell lines; however, until recently the effect of a human cancers; however, the role of these mutations in dominant active PI3K in the mammalian intestine had not been tumorigenesis, progression and response to therapy has largely investigated. Our laboratory has developed a murine model, yet to be defined.2 FCPIK3ca*, that expresses a dominant active form of PI3K Adenomatous Polyposis Coli (APC) mutations have been well (p110*) in the distal small bowel and colon.10 This model characterized as important mediators of colorectal tumorigenesis.3 develops sessile large moderately differentiated invasive mucinous Germline mutations were discovered in patients with familial adenocarcinomas in the proximal colon through a non-canonical adenomatous polyposis, but somatic mutations are also present in mechanism and without a polypoid luminal component. 80–90% of sporadic colorectal polyps and cancers.4 These APC and PIK3CA mutations are commonly identified together in mutations lead to truncation of the APC protein resulting in the human colorectal cancers, yet the interaction between these loss of function of this tumor suppressor. Loss of APC function mutations remains to be investigated in the mammalian intes- results in dysregulation of CTNNB1 (b-catenin), leading to tine.11 The potential for cross-talk between these signaling increased WNT pathway signaling through MYC and CCND1 cascades has been an area of interest. These pathways converge (cyclin D1), among others.5 The ApcMin/ þ (Min) mouse carries a on glycogen synthase kinase 3 (GSK3) and potentially other germline mutation in Apc, which results in these mice developing mediators. Prior studies in cell culture models have had many adenomas throughout the intestinal tract.6 This model has contradictory findings when examining the potential for an been widely used to study the biology of intestinal tumors and as interaction between these signaling cascades.12–17 To examine a valuable tool for chemoprevention studies. what effect mutations in APC and PIK3CA have on tumorigenesis, Genetic alterations in the phosphoinositide-3 kinase (PI3K) we have crossed the Min mouse with the FCPIK3ca* (FC13K1) signaling pathway are the most common mutations found in mouse. This cross results in a murine model with the loss of one cancer.7 Oncogenic mutations of PIK3CA are important in multiple allele of APC throughout the intestine and the expression of a

1Division of Hematology and Oncology, Department of Medicine, University of Wisconsin, Madison, WI, USA; 2Department of Oncology, University of Wisconsin, Madison, WI, USA; 3Division of Gastroenterology and Hepatology, Department of Medicine, University of Wisconsin, Madison, WI, USA; 4Department of Pathology and Laboratory Medicine, University of Wisconsin, Madison, WI, USA; 5Promega Corporation, Madison, WI, USA; 6Department of Pathology and Vanderbilt-Ingram Cancer Center, Vanderbilt University School of Medicine, Nashville, TN, USA and 7Department of Radiology, University of Wisconsin, Madison, WI, USA. Correspondence: Professor RB Halberg, Department of Medicine, University of Wisconsin, Clinical Science Center, 600 Highland Avenue, K4/532, Madison, WI 53792, USA. E-mail: [email protected] Received 25 January 2013; revised 8 March 2013; accepted 14 March 2013; published online 27 May 2013 PIK3CA and APC mutations are synergistic DA Deming et al 2246

Figure 1. FC13K1ApcMin/ þ mice become moribund by 52 days of age on average due to obstructive enteropathy or anemia. At necropsy, large proximal colon cancers result in distention of the cecum and small intestine (a) compared with control (b). These tumors are associated with enlarged mesenteric adenopathy (c). In addition, an impressive vascular supply is noted supplying tumors in the colon (c) and distal small intestine (d). The colon was removed and split lengthwise demonstrating a large ﬂat proximal colon cancer and other smaller tumors in the colon (e). The distal small intestine was also resected and split lengthwise revealing multiple large tumors (f). The mesentery was also examined closely following removal of the colon and small intestine. Hyperplasia of the lymphatic tissue was seen in all mice (c, g). In some mice, tumor deposits were seen in the mesenteric adipose tissue (g, h) or within the lymphatic tissue (g, i). Size bars: (g) ¼ 2 mm; (h, i) ¼ 1 mm. (h, i) Enlargements of the boxes in (g).

Oncogene (2014) 2245 – 2254 & 2014 Macmillan Publishers Limited PIK3CA and APC mutations are synergistic DA Deming et al 2247 dominant active PI3K (3K1) in the distal small intestine and colon Tumor development and progression are mediated by the due to the expression of Cre under the control of the fatty acid expression of the dominant active form of PI3K binding protein-1 promoter (FC1). Here, we demonstrate increased Expression of the constitutively active PI3K in the tumors of FC13K1 tumor multiplicity, size and a more aggressive and poorly mice has previously been shown to activate the PI3K signaling differentiated phenotype as a consequence of synergy between cascade.10 In FC13K1ApcMin/ þ mice, all tumors in the distal small APC and PIK3CA mutations. and large intestine, regardless of morphology, demonstrate the presence of p110* and activation of the PI3K signaling pathway (Figures 2–4), including elevated phosphorylation of AKT compared with tumors from ApcMin/ þ mice (Figure 2c). These RESULTS observations confirm that the transgene was transcribed and Expression of a dominantly active PI3K in the setting of allelic translated in the colon of FC13K1ApcMin/ þ mice and that its loss of Apc results in increased tumor multiplicity and increased expression resulted in increased downstream activation of the tumor size PI3K/AKT/MTOR pathway. FC13K1ApcMin/ þ mice were dissected once moribund with age- matched control littermates (Figure 1; 1 denotes carrier and 0 þ denotes non-carrier for FC and 3K; denotes wild type for Apc). Differences in tumor morphology result from the expression of the The majority of mice developed massive proximal colon cancers dominant active form of PI3K in the setting of allelic loss of Apc in resulting in obstruction of the lumen (Figure 1a). These tumors the intestine were associated with an impressive vascular supply and hyper- Tumors with multiple distinct morphologies develop in the plastic lymphatic tissue (Figures 1c and d). Tumors are seen in the FC13K1ApcMin/ þ mice (Figure 3; Supplementary Figures S1 and colon and distal small intestine (Figures 1e and f). In some S2). The flat lesions, previously described in the FC13K1 mice, are instances, metastatic lesions were noted in the mesenteric present (Figure 3a).10 These flat lesions are moderately lymphatic and adipose tissue (Figures 1g–i). No evidence of liver differentiated invasive mucinous adenocarcinomas that arise in or lung metastases was identified. A total of 61 mice were evaluated for the presence of intestinal tumors, including 18 FC13K1ApcMin/ þ mice, 17 FC03K1ApcMin/ þ 1 0 Min/ þ 0 0 Min/ þ ab10 5 mice, 6 FC 3K Apc mice, 9 FC 3K Apc mice and 11 S1-S2 FC13K1Min 1 1 þ / þ 9 FC 3K Apc mice (Supplementary Table S1). The average age of S3-S4 Min control 4 FC13K1ApcMin/ þ mice at necropsy was 52 days old. Mice appeared 8 Cecum-colon well until they became moribund from obstructive enteropathy 7 relating to large obstructing colon cancers or secondary to anemia 6 3 from bleeding intestinal tumors. 5 Tumor multiplicity was significantly increased in each section of 4 2 the intestine where the activated PI3K was expressed in the 3 setting of allelic loss of Apc (Figure 2). Overall, an average of 8.7 1 1 1 Min/ þ 2 tumors were identified throughout the intestine in FC 3K Apc mean Number of tumors,

1 mean (mm) diameter, Tumor compared with 1.4 in FC13K1Apc þ / þ (Po0.001), and 1.4 in ApcMin/ þ 0 1 Min/ þ 1 0 Min/ þ 0 0 Min/ þ 0 0 controls (FC 3K Apc , FC 3K Apc or FC 3K Apc ; 1 ColonS4 1 and 2 1 Min 1 Min 0 Min 0 Min Po0.001). Specifically, segment 4 and colon tumors were much 1 3K Intestinal section more frequent in the FC13K1ApcMin/ þ mice. No increase in tumor 1 3K 0 3K 0 3K 1 3K FC multiplicity was noted in the proximal most intestinal sections, FC FC FC FC segments 1 and 2, which lack activated PI3K. This is useful as an c 1 1 FC13K1Min internal control and indicates that the increase in tumor multiplicity FC 3K Min is related to expression of constitutively active PI3K. Flat Polypoid The intestinal sections from FC13K1ApcMin/ þ mice and asso- p110* p110α ciated controls were also evaluated for tumor size. Fixed tissues were examined under the dissecting microscope. The maximum phospho-AKT diameter of each tumor was determined using a measuring GAPDH reticule. The flat FC13K1 proximal colon tumors, as previously described,10 were not measured as these tumors do not Figure 2. Activation of the PI3K pathway in the intestine results in have discrete borders, which are required for reproducible increased tumor number and size. In the FC13K1ApcMin/ þ , the measurements. median number of tumors seen was 8.7 compared with the control littermates which averaged 2.2 or less (a). This effect was seen An increase in tumor size was noted in segment 4 and the 1 1 Min/ þ colon (Figure 2b). In segment 4, a total of 66 tumors were prominently in the colon, where the FC 3K Apc mice have a 1 1 Min/ þ Min/ þ mean of 2.9 tumors per mouse and only 1 colon tumor was seen in measured (61 FC 3K Apc and 5 Apc controls). The Min/ þ 1 1 Min/ þ 32 Apc control littermates (a). In the proximal small intestine, FC 3K Apc segment 4 tumors averaged 2.8 mm (range 0.7–7; no difference was noted in tumor number which is consistent with P ¼ 0.004), whereas the control tumors averaged o0.8 mm (range the activated form of PI3K not being expressed in this area in the 0.4–2 mm) in diameter. These large tumors are forming by just 50 FC13K1ApcMin/ þ mice. An increase in tumor size was also noted in days of age or less. Tumors of this size would not be expected in FC13K1ApcMin/ þ (colon: mean 4.3 mm, median 3, range 2–10 mm; S4: Min mice until closer to 100 days of age or older. A total of 9 mean 2.8 mm, median 2.3 mm, range 0.7–7 mm) compared with Min/ þ FC13K1ApcMin/ þ colon tumors were measured and the mean Apc (colon: 1.6 mm, only one colon tumor was observed in 32 diameter was 4.3 mm (range 2–10 mm). Only one ApcMin/ þ mice; S4: mean 0.8 mm, median 0.5 mm, range 0.4–2 mm, P ¼ 0.004) control colon tumor was identified in 32 mice. This tumor (b). No change in tumor size was noted in the proximal 2 segments of the small intestine, which is consistent with the lack of expression measured 1.6 mm. A total of 14 tumors were identified in of the activated form of PI3K in this area. The increase in tumor segments 1 and 2 of the small intestine (5 FC13K1ApcMin/ þ and Min/ þ number and size is related to the expression of the activated form of 9 Apc controls). These tumors were not significantly different PI3K, p110*, in the FC13K1ApcMin/ þ mice, as this can be detected in in size owing to the lack of constitutively active PI3K in this area these tumors and is associated with increased phosphorylation of (1.2 and 1.1, respectively, P ¼ 0.85). AKT as compared with ApcMin/ þ control tumors (c).

& 2014 Macmillan Publishers Limited Oncogene (2014) 2245 – 2254 PIK3CA and APC mutations are synergistic DA Deming et al 2248

Figure 3. FC13K1ApcMin/ þ mice develop multiple tumors with distinct morphologies. Large ﬂat moderately differentiated mucinous adenocarcinomas develop in the proximal colon (a, d). This cancer penetrates through the serosal surface consistent with a T4 lesion. Polypoid intramucosal carcinomas arising in the setting of villous/serrated hyperplasia are seen also in the proximal colon (b, e). In the mid and distal colon, tumors range from small adenomas to large adenomas with high-grade dysplasia to adenocarcinomas. A large adenoma with solid growth pattern and a focal area of adenocarcinoma in the submucosa is pictured in (c, f). The ﬂat mucinous lesions do not possess nuclear CTNNB1 (g), while the polypoid tumors demonstrate nuclear CTNNB1 IHC staining (h, i). The MKI67 and phospho-AKT IHC staining of all of these tumors indicates that these tumors are highly proliferative and have activated PI3K/AKT signaling (j–o). Scale bars: (d) ¼ 1 mm; (e) ¼ 500 mm; (f) ¼ 1 mm. All experiments were completed at least in triplicate.

the setting of villous/serrated hyperplasia in the proximal 74 lesions were evaluated by a pathologist, and ranged from colon. Extension of these cancers through the serosal layer is small adenomas to larger adenomas with high-grade dysplasia usually seen. These lesions are associated with a dramatic to invasive adenocarcinomas. The polypoid tumors of the desmoplastic reaction and a high degree of tumor budding. No FC13K1ApcMin/ þ micearemorerepresentativeofApcMin/ þ appreciable differences in tumor morphology were noted in the tumors, but display a greater degree of architectural ﬂat FC13K1ApcMin/ þ versus the FC13K1Apc þ / þ invasive mucinous distortion. Compared with the ApcMin/ þ control colon tumor, adenocarcinomas of the proximal colon (Figure 3; Supplementary these are much larger and less differentiated. The polypoid Figure S3). In all cases, these tumors formed without signiﬁcant lesions of the FC13K1ApcMin/ þ mice demonstrate a high degree polypoid lesions extending into the lumen of the colon and of nuclear CTNNB1 (b-catenin), MKI67 and pAKT staining invaded through the serosal surface. (Figure 3; Supplementary Figure S1). A marked difference in In addition, polypoid tumors, which are not present in FC13K1 tumor morphology was also observed in the exophytic small mice,10 are seen in the colon and distal small intestine. A total of bowel lesions (Figure 4). Large invasive adenocarcinomas were

Oncogene (2014) 2245 – 2254 & 2014 Macmillan Publishers Limited PIK3CA and APC mutations are synergistic DA Deming et al 2249

Figure 4. Advanced small bowel tumors are seen in FC13K1ApcMin/ þ mice. Large, depressed adenocarcinomas are present in segments 3 and 4 of the small intestine (a, c). Some of these tumors even penetrate through the serosal surface. This is compared with the small adenomas that form less frequently in the ApcMin/ þ controls (b, d). CTNNB1 expression is nuclear in both FC13K1ApcMin/ þ tumors and controls (e, f). An increase in MKI67 and phospho-AKT IHC staining is noted in the FC13K1ApcMin/ þ tumors compared with control (g–j). Scale bars: (a, b) ¼ 1mm; (c–j) are 4 Â enlargements. seen in the small intestine of FC13K1ApcMin/ þ mice, whereas only exhibited uptake of a novel near-infrared phospholipid ether a limited number of small adenomas were seen in control analog, CLR1502, that selectively accumulates in advanced littermates. The lesions of both the small and large intestine tumors (Supplementary Figure S4).

& 2014 Macmillan Publishers Limited Oncogene (2014) 2245 – 2254 PIK3CA and APC mutations are synergistic DA Deming et al 2250 FC13K1ApcMin/ þ mice develop tumors with and without nuclear CTNNB1 were identified (Figure 5). The morphology of the cells in localization of CTNNB1 these regions also changed. This indicates that the cells with Immunohistochemistry (IHC) analysis of CTNNB1 has been used for nuclear CTNNB1 likely lost the other allele of Apc after tumor many years as a marker of activated WNT signaling. When initiation by activated PI3K. The change in morphology indicates functional APC is no longer expressed, CTNNB1 accumulates and that important phenotypic changes in the tumor might occur translocates to the nucleus. Here, we have examined CTNNB1 even if APC is lost after tumor initiation and not just as the initiator staining in tumor tissues of FC13K1ApcMin/ þ and their control of tumorigenesis. littermates. Similar to what was demonstrated previously, the flat proximal colon mucinous invasive adenocarcinomas of the FC13K1Apc þ / þ and FC13K1ApcMin/ þ mice largely did not demon- FC13K1ApcMin/ þ and FC13K1Apc þ / þ tumors are microsatellite strate nuclear CTNNB1 staining (Figure 5; Supplementary Figures S3 stable and S5). However, the polypoid lesions of the colon did Microsatellite instability (MSI) has been demonstrated to be an demonstrate a high degree of nuclear CTNNB1. Microscopic important marker for prognosis and response to treatment in adenomas are even able to be identified in the colon due humans.18 MSI analysis of intestinal tumors from mice with to their high degree of nuclear CTNNB1 staining (Supplementary FC13K1ApcMin/ þ and FC13K1Apc þ / þ mice was performed using Figure S5). mononucleotide repeats, which have been previously shown to be These data indicate that tumors in the FC13K1ApcMin/ þ mice are very sensitive for detection of MSI in mouse tumors developing via different mechanisms. In the flat proximal invasive (Supplementary Figure S6).19 The microsatellite allelic patterns of colon adenocarcinomas, the lack of nuclear CTNNB1 indicates that 18 out of 22 tumor samples and matching normal samples were these tumors are not initiated by aberrant WNT signaling. Their identical and therefore were considered as microsatellite stable. non-canonical mechanism of tumorigenesis needs to be further Four tumors contained a novel allele at the mBat-66 locus that was clarified. In the polypoid colonic lesions, where CTNNB1 is nuclear, not present in wild-type C57BL/6 mice. However, this novel allele these tumors are likely initiated by loss of APC and their more was the same in all four tumors and is, therefore, likely due to a aggressive phenotype driven by the presence of the dominantly polymorphism in these F1 mice. Thus, MSI does not appear to active PI3K. Interestingly, in two flat mucinous proximal colon have a role in intestinal tumorigenesis in FC13K1ApcMin/ þ and cancers, thought to be initiated by PI3K, small foci of nuclear FC13K1Apc þ / þ mice.

Figure 5. The tumors of the FC13K1ApcMin/ þ mice develop by both canonical and non-canonical pathways. Exophytic polypoid adenomas with high-grade dysplasia (a, left) and mucinous adenocarcinomas (a, right) develop in the proximal colon of FC13K1ApcMin/ þ mice. The exophytic polypoid lesions possess a high amount of nuclear CTNNB1 (tumor on the left in b, enlarged in c), while the flat mucinous adenocarcinomas do not possess significant levels of nuclear CTNNB1 (tumor on the right in b, enlarged in d). This indicates that the polypoid lesions are arising secondary to aberrant WNT signaling, while the flat mucinous lesions form via a non-canonical mechanism. Rarely, the large flat mucinous adenocarcinomas have islands of cells with nuclear CTNNB1 (e, H&E; f, CTNNB1). The cells with nuclear CTNNB1 also have a different morphology (g, h, enlargements of e and f, respectively), with a higher nucleus-to-cytoplasm ratio, nuclear enlargement and hyperchromasia, and a more basophilic cytoplasm. Scale bars: (a, b) ¼ 1 mm; (e, f) ¼ 1 mm; (c, d, g, h) ¼ 100 mm.

Oncogene (2014) 2245 – 2254 & 2014 Macmillan Publishers Limited PIK3CA and APC mutations are synergistic DA Deming et al 2251 Similar tumor multiplicity and phenotypes are seen when Apc is In addition, no change in total CTNNB1 was observed between the lost only in the epithelial cells of the intestine in the setting of a majority of FC13K1ApcMin/ þ and ApcMin/ þ controls. constitutively active PI3K The WNT and PI3K pathways also converge on important In the Min mouse, one allele of Apc is lost in all cells. Prior transcription factors including MYC and CCND1. At necropsy, investigations have indicated that the loss of APC in stromal tissue tumors were excised and prepared for histological analysis. may be important for tumorigenesis.20 To investigate this further, Nuclear MYC was identified in all FC13K1Apc þ / þ and FC13K1 we used a transgenic mouse with a Cre recombinase conditional ApcMin/ þ at similar levels to that seen for the ApcMin/ þ controls. knockout of Apc (Apcfl/ þ ).21 This mouse contains an allele of Apc Nuclear CCND1 was observed in the majority of malignant cells in that is floxed at exon 14. In the presence of Cre, exon 14 is FC13K1Apc þ / þ and FC13K1ApcMin/ þ , which is increased compared removed and the APC protein is truncated. These mice were used with the ApcMin/ þ controls. to generate genetically homogenous FC13K1Apcfl/ þ mice. Similarly to the FC13K1ApcMin/ þ mice, they also developed advanced small intestine and colon cancers (Figure 6). No appreciable significant DISCUSSION differences were identified between the FC13K1ApcMin/ þ and the 1 1 fl/ þ Human colorectal cancers are a heterogeneous group, with each FC 3K Apc mice, including tumor multiplicity and morphology. cancer individualized by its mutation profile. The presence of certain mutations may change with the biological context, such as the possible interaction between PIK3CA mutations and bowel FC13K1ApcMin/ þ tumors demonstrate downstream activation of inflammation. This is important clinically as patients with this the PI3K pathway, MYC expression, and increased CCND1 mutation are more likely to benefit from adjuvant aspirin Min/ þ compared with Apc controls therapy.22 In addition, certain mutations are often associated The expression of the dominant active PI3K in the FC13K1ApcMin/ þ with each other. Of colon tumors with PIK3CA mutations, 86% of tumors results in downstream activation of the PI3K pathway these cancers also possess mutant APC.11 In addition, PIK3CA (Figures 2c and 7a). RPS6 (ribosomal protein S6) and eukaryotic mutations have also been associated with KRAS and BRAF translation initiation factor 4E binding protein (EIF4EBP1) are mutations, CpG island methylator phenotype and MSI.23–25 phosphorylated in both flat and polypoid FC13K1ApcMin/ þ tumors. Concerted efforts to develop improved and more personalized Interestingly, a significant amount of phosphorylated RPS6 and treatment options for all types of cancer continue. Recent advances EIF4EBP1 is also seen in ApcMin/ þ . Only low levels of phosphory- include improved targeting of directed therapies against the lated ERK1/2 are seen in FC13K1ApcMin/ þ and ApcMin/ þ controls mutational activation of signaling pathways in cancer. This has led (Figure 7a). to the development of classes of exciting anti-neoplastic agents, such The PI3K and WNT signaling pathways are known to converge asEGFR,BRAF,MEK,EML4-ALK,MET,MTORandPI3Kinhibitors.To on GSK3; however, the role of GSK3 in cross-talk between these properly develop and utilize these new therapies, a better under- two pathways remains controversial. In the majority of flat standing of how the mutation profile of each cancer affects the FC13K1ApcMin/ þ cancers, phosphorylation of GSK3B is increased tumor biology is required. It is important to recognize the roles of compared with controls. Interestingly, in the polypoid FC13K1 these mutant proteins and how their function is altered by the ApcMin/ þ tumors, GSK3B appears to be phosphorylated less often. presence of mutations in other oncogenes or tumor suppressors. Two of these tumors that did not demonstrate phosphorylation of Here, we report the first in vivo description of the interaction GSK3B did show phosphorylation of GSK3A. Despite the between APC and PIK3CA mutations in mammalian intestinal tumors. phosphorylation of GSK3B in the flat FC13K1ApcMin/ þ cancers Previously, we have demonstrated that a dominant active PI3K nuclear localization of CTNNB1 was not observed (Figures 3 and 5). is able to initiate the development of flat mucinous invasive

Figure 6. In ApcMin/ þ mice, one functional allele of Apc is lost in all cells, including the epithelial and stromal cells. To examine if a similar phenotype exists when APC is lost only in the epithelial cells, mice possessing an allele of Apc with exon 14 floxed (Apcfl/ þ ) were utilized. In the FC13K1Apcfl/ þ mice, one allele of Apc was lost only in the epithelial cells of the distal small intestine and colon. No difference in tumor number was noted in FC13K1Apcfl/ þ and FC13K1ApcMin/ þ mice in the distal small intestine and colon (a). In addition, no difference in tumor morphology or CTNNB1 localization was noted (b–g). Scale bar (d–g) ¼ 1 mm.

& 2014 Macmillan Publishers Limited Oncogene (2014) 2245 – 2254 PIK3CA and APC mutations are synergistic DA Deming et al 2252

Figure 7. Several key mediators of PI3K and WNT signaling are altered in FC13K1ApcMin/ þ mice. (a, b) Western blots. Phosphorylation of RPS6 and EIF4EBP1 is observed in the FC13K1ApcMin/ þ tumors (a). Only low amounts of phospho-ERK1/2 are seen in these tumors (a). Increased pGSK3B is detected in the majority of ﬂat FC13K1ApcMin/ þ cancers with variable phosphorylation noted in the polypoid lesions (b). No change in the levels of total GSK3A or GSK3B was observed. In addition, in the majority of tumors no change in the levels of CTNNB1 was seen (b). Similar levels of MYC were seen in all tumors (c). Increased nuclear CCND1 levels were observed in the FC13K1Apc þ / þ and FC13K1ApcMin/ þ tumors compared with ApcMin/ þ tumors (c). Scale bars low mag 1 mm and high mag 100 mm. All experiments were completed at least in triplicate.

adenocarcinomas in the proximal colon via a non-canonical The mechanism for the synergy between WNT and PI3K mechanism of tumorigenesis.10 In addition to these lesions, we signaling pathways may occur due to a direct interaction between now demonstrate that the presence of a dominant active PI3K in these pathways. Previous studies have aimed to examine this the setting of allelic loss of APC results in increased tumor interaction; however, they have been contradictory. GSK3 is well multiplicity, size, and a more aggressive and less differentiated known to have an important role in both WNT and PI3K signaling, phenotype. We have demonstrated that this is related to and thus has been a target of multiple investigations examining increased activation of AKT and phosphorylation of downstream these pathways. GSK3 is a serine-threonine kinase that occurs targets including RPS6 and EIF4EBP1. in two isoforms GSK3A and GSK3B.26 These kinases are Interestingly, in the ﬂat proximal colonic mucinous invasive naturally constitutively active and are inhibited by N-terminal adenocarcinomas, these tumors carry germline mutations in one phosphorylation of Ser-21 for GSK3A and Ser-9 for GSK3B, which allele of Apc. However, in two tumors we observed the loss of APC results in the formation of a pseudosubstrate motif.27 In WNT activity as a likely later event. In this setting, cells that lost APC signaling, GSK3 has an important role in the degradation of after tumor initiation secondary to PI3K had a change in CTNNB1. GSK3 is bound by Axis inhibition protein (AXIN1) and morphology to a less differentiated state. This indicates that, in phosphorylates CTNNB1.28 This phosphorylation of CTNNB1 addition to tumor initiation, the loss of APC can be a late event in targets it for ubiquitination and degradation by proteosomes. If tumorigenesis, and could be responsible for clonal outgrowth in GSK3 is inhibited by WNT signaling, then this leads to CTNNB1 some instances, such as was observed here. accumulation and nuclear translocation, though the mechanism of

Oncogene (2014) 2245 – 2254 & 2014 Macmillan Publishers Limited PIK3CA and APC mutations are synergistic DA Deming et al 2253 WNT inhibition of GSK3 remains poorly understood. In PI3K then crossed with FC1 mice to generate FC13K1Apcfl/ þ mice. Mice were signaling, phosphorylated AKT in turn phosphorylates GSK3 on its genotyped for FC, 3K, Apcfl/ þ and Min as described previously.6,21,30,31 inhibitory serines.29 Tumors from ApcMin/ þ and (SWRxB6) F1 ApcMin/ þ mice treated with 4% Multiple studies have demonstrated that an interaction dextran sodium sulfate were used as controls. between these pathways might exist related to a single pool of 12–15 GSK3. Contrary to these findings, other groups have proposed Tumor fixation and counts that cross-talk between these two pathways does not exist.16,17 Ng 17 Mice were euthanized by CO2 asphyxiation. At necropsy, the entire et al. demonstrated that a pool of GSK3B is stably bound to intestine was excised. The small intestine was divided into four sections of AXIN1 and represents only about 3–5% of the total GSK3B. They equal length (numbered 1 through 4 from proximal to distal) and the colon also observed that the phosphorylation level of the AXIN1-bound and cecum were kept intact. Each section of the intestine was then split GSK3B was independent of PI3K pathway activation by insulin or lengthwise, splayed open, and rinsed. Tissues were fixed flat in 10% inhibition by wortmannin. buffered formalin for 24–48 h, then stored in 70% ethanol. Tumors in each Some evidence indicates that the context in which phospho- section of the intestine were counted under a dissecting microscope by a AKT signaling is taking place has an effect on its ability to inhibit single observer who was blinded to the genotypes. GSK3B that is bound to AXIN1 and thus acting in WNT signaling. Fukumoto et al.13 demonstrated that phospho-AKT is bound to Histology and immunohistochemistry the AXIN1:GSK3B complex in the presence of Dishevelled, Fixed tumors were isolated, embedded in paraffin, and cut into 5 mm promoting phosphorylation of GSK3B and accumulation of free sections. Every tenth section was stained with hematoxylin and eosin for CTNNB1. However, if the WNT cascade was not activated by histological review. IHC was carried out as described previously.10 The Dishevelled, then phospho-AKT was not found to be bound to the primary antibodies included rabbit anti-phospho-AKT (Ser473, 1:100, Cell AXIN1:GSK3B complex and free CTNNB1 levels were not elevated. Signaling Technology, Beverly, MA, USA), mouse anti-CTNNB1 (1:200, BD These data indicate that phospho-AKT is not sufficient, but can Biosciences—Clone 14, San Diego, CA, USA), rabbit anti-PCNA (Ki-67) enhance the activity of WNT signaling. These findings are (1:1000, Cell Signaling Technology), and CCND1 (1:25, Cell Signaling Technology). The IHC protocol was modified for the anti-MYC antibody consistent with those of Ng and colleagues as WNT was not N-262 (1:50, Santa Cruz Biotechnology, Dallas, TX, USA) as follows. Upon activated in their experiments. rehydration, slides were incubated in 0.05% sodium borohydride (Sigma- Here, we demonstrate synergy between the loss of the tumor Aldrich, St Louis, MO, USA) for 10 min followed by incubation in 0.5% Triton suppressor APC and the presence of a dominant active PI3K. X-100 (Sigma-Aldrich). Slides were blocked for 20 min at room temperature FC13K1ApcMin/ þ mice demonstrate increased tumor multiplicity, with a fresh blocking solution (2.5 ml 1 M Tris-HCl pH 8.0, 2 ml 5 M NaCl, size and aggressiveness compared with the tumors of control 2 ml 10% NP-40, 5 g non-fat dry milk, distilled deionized water to 100 ml). littermates. In the flat mucinous adenocarcinomas, an increase in pGSK3B is observed; however, this is without change in the total Western blot analysis levels or nuclear localization of CTNNB1. This proves that Tissue was processed and gels run as previously described.10 The gels were phospho-AKT is insufficient for activation of the WNT signaling probed with primary antibodies against p110a, phospho-AKT (Ser473), pathway in this context. In the polypoid tumors, variable RPS6 (Ser235/236), phosphor-EIF4EBP1 (Thr37/46), pMAPK (Thr202/Tyr phosphorylation of GSK3B was seen. Again, no change in total 204), pGSK3A, pGSK3B or Total GSK3 (Cell Signaling Technology) in bovine CTNNB1 levels was seen, but significant nuclear localization of serum albumin (Sigma-Aldrich) at a 1:1000 ratio for 16 h. Anti-CTNNB1 (BD CTNNB1 was noted similar to that seen in ApcMin/ þ controls. We Biosciences—Clone 14) was probed at a ratio of 1:5000. Anti-GAPDH cannot exclude the possibility that phospho-AKT signaling is able antibody (Cell Signaling Technology) was utilized as a loading control at a to enhance CTNNB1 signaling in the presence of an aberrant WNT ratio of 1:5000. pathway, such as mutant Apc. Significant levels of nuclear MYC were seen in all the tumors assayed and CCND1 was increased in MSI analysis 1 1 Min/ þ Min/ þ the FC 3K Apc tumors above the level seen in Apc Tumor DNAs from FC13K1Apc þ / þ and FC13K1ApcMin/ þ mice were tested for tumors, indicating a potential etiology for the synergy between MSI using a multiplex of six mononucleotide repeat loci designed these mutations at the transcriptional level. specifically for MSI testing of C57BL/6 mice, including: mBat-56 In summary, synergy exists between the loss of the tumor (GTGTGTATGCTGATTTTATATCCT, GCAAAAATATCATTCGGTATG), mBat-57 suppressor APC and the presence of a dominant active PI3K in this (TGAAATCCTTGATGTTCCTCTACTAGGT, GGTCATCCTGTTGTTTCTAAATGAT in vivo model of colon cancer. Further studies are required to TGT), mBat-58 (CCCCTAAAACTTTCCTTGCTATT, TTCTGAGTTCCAGGGCAG elucidate to what extent cross-talk between these pathways exists TCTG), mBat-59 (GGTCTTGCCCTGAGGCAAGTAAT, AACCATCCGTAACAAG and is responsible for this synergy. In addition, pharmacologic ATCTGACGT), mBat-64 (CAGCCCACACTCCTGAAAACAG, CACTTAGCCAGTT GCGTCACCT) and mBat-66 (TGGGGTGTCTAAAGACAGCTAAG, GCCCACTT studies using this and similar models will be vital to the CATGCGTAACAG). For MSI analysis, 42 ng of DNA was amplified with 1 Â development of novel combination therapies for the personalized C57BL/6 multiplex fluorescent primer mix, 1 Â Gold ST*R Buffer (Promega, treatment of patients with colorectal and other cancers. Madison, WI, USA) and 0.2 U GoTaqHot Start Polymerase (Promega) in a GeneAmp PCR System 9700 Thermocycler (Applied Biosystems, Foster City, CA, USA). The amplification conditions used were as follows: 95 1C (2 min); MATERIALS AND METHODS 10 cycles of 94 1C (30 s), 58 1C (30 s, 29% ramp), 65 1C (1 min, 23% ramp); 20 Mouse husbandry and genotyping cycles of 90 1C (30 s), 58 1C (30 s, 29% ramp), 65 1C (1 min, 23% ramp); and a final extension of 60 1C (30 min). PCR products were denatured in All animal studies were conducted under protocols approved by the deionized formamide with Internal Lane Standard 600 (Promega) for allele Institutional Animal Care and Use Committee at the University of sizing and analyzed on a 3130xl Genetic Analyzer using GeneMapper 4.0 Wisconsin-Madison, following the guidelines of the American Association 1 Software (Applied Biosystems). For MSI analysis, the microsatellite allelic for the Assessment and Accreditation of Laboratory Animal Care. FC mice patterns of tumor and matching normal samples were compared and (FVB/N-Tg(Fabp1-Cre)1Jig; NCI Mouse Repository; Strain number 01XD8), 1 tm7(Pik3ca*,EGFP)Rsky/ classified as microsatellite stable if no loci were unstable, MSI-Low if one 3K mice (C57BL/6-Gt(ROSA)26Sor J; The Jackson Labora- locus was unstable, and MSI-High if two or more loci were unstable.32 tory; Stock Number-012343) and ApcMin/ þ males (C57BL/6J ApcMin/J; The Jackson Laboratory; Stock number 002020) were maintained as previously described.6 ApcMin/ þ males were crossed with 3K1 females to generate 3K1ApcMin/ þ mice. Male 3K1ApcMin/ þ mice were then crossed with FC1 mice CONFLICT OF INTEREST to generate F1 FC13K1ApcMin/ þ mice and the associated controls. 3K1 mice Dr Jamey Weichert is the founder of Cellectar, Inc. (Madison, WI), which holds the were also crossed to Apcfl/fl mice (B6.Cg-Apctm2Rak; NCI Mouse Repository; licensing rights to the CLR1404 technology, and therefore has a financial interest in Strain number 01XAA) to generate 3K1Apcfl/ þ mice. These mice were this agent.

& 2014 Macmillan Publishers Limited Oncogene (2014) 2245 – 2254 PIK3CA and APC mutations are synergistic DA Deming et al 2254 ACKNOWLEDGEMENTS 14 Naito AT, Akazawa H, Takano H, Minamino T, Nagai T, Aburatani H et al. We thank Ella Ward and Jane Weeks in Experimental Pathology at the UW Carbone Phosphatidylinositol 3-kinase-Akt pathway plays a critical role in early Cancer Center for their technical assistance. This project was supported by the cardiomyogenesis by regulating canonical Wnt signaling. Circ Res 2005; 97:144–151. Conquer Cancer Foundation of the American Society of Clinical Oncology through A 15 Fang D, Hawke D, Zheng Y, Xia Y, Meisenhelder J, Nika H et al. Phosphorylation of Young Investigator Award (DAD); the National Cancer Institute of the US National beta-catenin by AKT promotes beta-catenin transcriptional activity. J Biol Chem Institutes of Health through T32 CA009614 (DAD), P50 CA095103 (Gastrointestinal 2007 11221–11229. Specialized Program of Research Excellence Grant, Vanderbilt Ingram Cancer Center), 16 Ding VW, Chen RH, McCormick F. Differential regulation of glycogen synthase R01 CA123438 (RBH), P30 CA014520 (Core Grant, University of Wisconsin Carbone kinase 3beta by insulin and Wnt signaling. J Biol Chem 2000; 275: 32475–32481. Cancer Center); and start-up funds (RBH) from the UW Division of Gastroenterology 17 Ng SS, Mahmoudi T, Danenberg E, Bejaoui I, de Lau W, Korswagen HC et al. and Hepatology, the UW Department of Medicine, and the UW School of Medicine Phosphatidylinositol 3-kinase signaling does not activate the Wnt cascade. J Biol and Public Health. Chem 2009; 285: 35308–35313. 18 Des Guetz G, Schischmanoff O, Nicolas P, Perret GY, Morere JF, Uzzan B. Does microsatellite instability predict the efﬁcacy of adjuvant chemotherapy in colorectal cancer? A systemic review with meta-analysis. Eur J Cancer 2009; 45: AUTHOR CONTRIBUTIONS 1890–1896. DAD, AAL and RBH designed, performed and analyzed experiments, and wrote 19 Bacher JW, Abdel Megid WM, Kent-First MG, Halberg RB. Use of mononucleotide repeat markers for detection of microsatellite instability in mouse tumors. the manuscript. LN, CS, MM, DA and JB performed and analyzed experiments. Mol Carcinog 2005; 44: 285–292. MKW, LC and JW analyzed experiments. All authors discussed results and edited 20 Tanwar PS, Zhang L, Roberts DJ, Teixeira JM. Stromal deletion of APC tumor the manuscript. suppressor in mice triggers development of endometrial cancers. Cancer Res 2011; 71: 1584–1596. 21 Hung KE, Maricevich MA, Richard LG, Chen WY, Richardson MP, Kunin A et al. Development of a mouse model for sporadic and metastatic colon tumors and its REFERENCES use in assessing drug treatment. Proc Natl Acad Sci USA 2010; 107: 1565–1570. 1 American Cancer Society. Cancer Facts & Figures 2012. American Cancer Society: 22 Liao X, Lochhead P, Nishihara R, Morikawa T, Kuchiba A, Yamauchi M et al. Aspirin Atlanta, 2012. use, tumor PIK3CA mutation, and colorectal-cancer survival. N Engl J Med 2012; 2 Cancer Genome Atlas Network. Comprehensive molecular characterization of 367: 1596–1651. human colon and rectal cancer. Nature 2012; 487: 330–337. 23 Whitehall VL, Rickman C, Bond CE, Ramsnes I, Greco SA, Umapathy A et al. 3 Kinzler KW, Vogelstein B. Colorectal tumors. In: Vogelstein B, Kinzler KWThe Oncogenic PIK3CA mutations in colorectal cancers and polyps. Int J Cancer 2012; Genetic Basis of Human Cancer. 2nd ed (McGraw-Hill: New York, 2002 583–612. 131: 813–820. 4 Goss KH, Groden J. Biology of the adenomatous polyposis coli tumor suppressor. 24 Velho S, Moutinho C, Cirnes L, Albuquerque C, Hamelin R, Schmitt F et al. BRAF, J Clin Oncol 2000; 18: 1967–1979. KRAS, and PIK3CA mutations in colorectal serrrated polyps and cancer: primary or 5 Clevers H, Nusse R. Wnt/b-catenin signaling and disease. Cell 2012; 149: secondary genetic events in colorectal carcinogenesis? BMC Cancer 2008; 8: 1192–1205. 255–260. 6 Su LK, Kinzler KW, Vogelstein B, Preisinger AC, Moser AR, Luongo C et al. Multiple 25 Armaghany T, Wilson JD, Chu Q, Mills G. Genetic alterations in colorectal cancer. intestinal neoplasia caused by a mutation in the murine homolog of the APC Gastrointest Cancer Res 2012; 5: 19–27. gene. Science 1992; 256: 668–670. 26 Doble BW, Woodgett JR. GSK-3: tricks of the trade for a multi-tasking kinase. 7 Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S et al. High frequency of J Cell Sci 2003; 116: 1175–1186. mutations of the PIK3CA gene in human cancers. Science 2004; 304: 554. 27 Frame S, Cohen P, Biondi RM. A common phosphate binding site explains the 8 Markowitz SD, Bertagnolli MM. Molecular basis of colorectal cancer. N Engl J Med unique substrate speciﬁcity of GSK3 and its inactivation by phosphorylation. 2009; 361: 2449–2460. Mol Cell 2001; 7: 1321–1327. 9 Zhao L, Vogt PK. Hot-spot mutations in p110a of phosphatidylinositol 3-kinase 28 Hinoi T, Yamamoto H, Kishida M, Takada S, Kishida S, Kikuchi A. Complex for- (PI3K):differential interactions with the regulatory subunit p85 and with RAS. Cell mation of adenomatous polyposis coli gene product and axin facilitates glycogen Cycle 2010; 9: 596–600. synthase kinase-3 beta-dependent phosphorylation of beta-catenin and down- 10 Leystra AA, Deming DA, Zahm CD, Farhoud M, Olson TJ, Hadac JN et al. Mice regulates beta-catenin. J Biol Chem 2000; 275: 34399–34406. expressing activated PI3K rapidly develop advanced colon cancer. Cancer Res 29 Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of gly- 2012; 72: 2931–2936. cogen synthase kinase-3 by insulin mediated by protein kinase B. Nature 1995; 11 Nosho K, Kawasaki T, Ohnishi M, Suemoto Y, Kirkner GJ, Zepf D et al. PIK3CA 378: 785–789. mutation in colorectal cancer: relationship with genetic and epigenetic altera- 30 Saam JR, Gordon JI. Inducible gene knockouts in the small intestinal and colonic tions. Neoplasia 2008; 10: 534–541. epithelium. J Biol Chem 1999; 274: 38071–38082. 12 Desbois-Mouthon C, Cadoret A, Blivet-Van Eggelpoel MJ, Bertrand F, Cherqui G, 31 Srinivasan L, Sasaki Y, Calado DP, Zhang B, Paik JH, DePinho RA et al. PI3 kinase Perret C. Insulin and IGF-1 stimulate the beta-catenin pathway through two signals BCR-dependent mature B cell survival. Cell 2009; 139: 573–586. signaling cascades involving GSK-3beta inhibition and Ras activation. Oncogene 32 Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW et al. A 2001; 20: 252–259. National Cancer Institute Workshop on Microsatellite Instability for cancer 13 Fukumoto S, Hsieh CM, Maemura K, Layne MD, Yet SF, Lee KH et al. Akt partici- detection and familial predisposition: development of international criteria for the pation in the Wnt signaling pathway through Dishevelled. J Biol Chem 2001; 276: determination of microsatellite instability in colorectal cancer. Cancer Res 1998; 17479–17483. 58: 5248–5257.

Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)

Oncogene (2014) 2245 – 2254 & 2014 Macmillan Publishers Limited