Cancer Research Cancer Epigenetics EMT and Cancer Progression and Treatment
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 68, 759, February 1, 2008. doi: 10.1158/0008-5472.CAN-07-5014
© 2008 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Contreras, C. M.
Right arrow Articles by Castrillon, D. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Contreras, C. M.
Right arrow Articles by Castrillon, D. H.

Cell, Tumor, and Stem Cell Biology

Loss of Lkb1 Provokes Highly Invasive Endometrial Adenocarcinomas

Cristina M. Contreras1,2, Sushma Gurumurthy5, J. Marshall Haynie1,2, Lane J. Shirley1,2, Esra A. Akbay1,2, Shana N. Wingo3, John O. Schorge2,3,4, Russell R. Broaddus7, Kwok-Kin Wong6, Nabeel Bardeesy5 and Diego H. Castrillon1,2

1 Department of Pathology, 2 Simmons Comprehensive Cancer Center, 3 Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, and 4 Hamon Center for Therapeutic Oncology Research, University of Texas Southwestern Medical Center, Dallas, Texas; 5 Department of Medicine, Massachusetts General Hospital Cancer Center and Harvard Medical School; 6 Department of Medical Oncology, Dana-Farber Cancer Institute and Harvard Medical School, Boston, Massachusetts; and 7 Department of Pathology, M. D. Anderson Cancer Center, Houston, Texas

Requests for reprints: Diego H. Castrillon, University of Texas Southwestern Medical Center, Department of Pathology, 6000 Harry Hines Boulevard, Dallas, TX 75390-9072. Phone: 214-648-4032; Fax: 214-648-7355; E-mail: diego.castrillon{at}utsouthwestern.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Mutations in the LKB1 tumor suppressor gene result in the Peutz-Jeghers syndrome, an autosomal dominant condition characterized by hamartomatous polyps of the gastrointestinal tract and a dramatically increased risk of epithelial malignancies at other sites, including the female reproductive tract. Here we show that female mice heterozygous for a null Lkb1 allele spontaneously develop highly invasive endometrial adenocarcinomas. To prove that these lesions were indeed due to Lkb1 inactivation, we introduced an adenoviral Cre vector into the uterine lumen of mice harboring a conditional allele of Lkb1. This endometrial-specific deletion of the Lkb1 gene provoked highly invasive and sometimes metastatic endometrial adenocarcinomas closely resembling those observed in Lkb1 heterozygotes. Tumors were extremely well differentiated and histopathologically distinctive and exhibited alterations in AMP-dependent kinase signaling. Although Lkb1 has been implicated in the establishment of cell polarity, and loss of polarity defines most endometrial cancers, Lkb1-driven endometrial cancers paradoxically exhibit (given their highly invasive phenotype) normal cell polarity and apical differentiation. In human endometrial cancers, Lkb1 expression was inversely correlated with tumor grade and stage, arguing that Lkb1 inactivation or down-regulation also contributes to endometrial cancer progression in women. This study shows that Lkb1 plays an important role in the malignant transformation of endometrium and that Lkb1 loss promotes a highly invasive phenotype. [Cancer Res 2008;68(3):759–66]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Inherited germ line mutations in the LKB1 gene result in the Peutz-Jeghers syndrome, an autosomal dominant condition characterized by hamartomatous polyps of the gastrointestinal tract (1, 2). Peutz-Jeghers syndrome is also associated with a significantly elevated (>15x) general cancer risk, suggesting that LKB1 plays a broad tumor suppressor role (3). Consistent with this idea, somatic inactivation of the LKB1 gene by point mutation (or other mechanisms such as promoter hypermethylation) has more recently been shown to be a common event driving carcinomas of the gall bladder, pancreas, and lung (4–7).

Peutz-Jeghers syndrome is strongly associated with clinically and histopathologically distinctive neoplasms of the female reproductive tract, including the ovary and uterine cervix. In the endocervix, Peutz-Jeghers syndrome is associated with an extremely well-differentiated variant of endocervical adenocarcinoma (minimal deviation adenocarcinoma of the endocervix, also known as adenoma malignum). The histopathologic diagnosis of this tumor (i.e., in biopsies) can be difficult because the malignant glands so closely resemble normal endocervical glands. Their benign histologic appearance is deceptive, however, because these tumors are invasive, often large, and clinically aggressive (8).

The Lkb1 protein participates in at least two distinct but potentially overlapping biological pathways that seem to synergize in tumorigenesis. In the mouse, Lkb1 establishes cell polarity in cortical neurons (9) and colonic epithelium (2), and it has been suggested that most of the consequences of Lkb1 deficiency in tumors are indirect consequences of cell polarization defects (10). Loss of cell polarity strongly correlates with aggressive and invasive malignant cell growth (11) and is a general feature of adenocarcinomas, including those of endometrial origin (12). Second, Lkb1 directly phosphorylates and regulates AMP-dependent kinase (AMPK), a sensor of intracellular ATP levels. AMPK controls protein synthesis by regulating the mammalian target of rapamycin (mTOR) through phosphorylation of TSC2 (tuberin; ref. 13). Through this pathway, loss of Lkb1 activity in tumors may derepress protein synthesis, thereby promoting cell growth, proliferation, and tumorigenesis. Intriguingly, the regulation of cell polarity by Lkb1 seems to be at least partially mediated by AMPK, which, in addition to its roles in the mTOR pathway, also regulates the assembly of epithelial tight junctions and thus directly contributes to the maintenance of epithelial cell polarity (14–16). Although these findings suggest that AMPK is of central importance as a downstream effector of Lkb1 inactivation, Lkb1 is also the master upstream kinase of at least 13 AMPK-related kinases (17), suggesting that Lkb1 may contribute to tumorigenesis through other as yet unknown mechanisms.

A mouse Lkb1 knockout model of Peutz-Jeghers syndrome closely mimics the human condition. Mice heterozygous for a null Lkb1 mutation are viable and externally normal, but develop intestinal polyps identical to those seen in Peutz-Jeghers syndrome patients (Lkb1–/– embryos die early in gestation; ref. 18). Here we report that Lkb1–/+ mice also spontaneously develop recurring distinctive neoplasms within the uterine corpus. These tumors are extremely well differentiated yet highly invasive, similar to the uterine endocervical neoplasms in women with Peutz-Jeghers syndrome, a notable finding given that endocervical and endometrial epithelia share a common embryologic origin as Müllerian duct derivatives and are directly contiguous within the uterus. Inactivation of a floxed Lkb1 allele at 6 weeks of age proved that these endometrial cancers were the result of Lkb1 inactivation and arise secondarily following normal uterine development. Further analyses showed that tumor cells with biallelic Lkb1 inactivation do not have gross defects in cell polarity, suggesting that Lkb1 restrains endometrial tumor cell growth and invasion via biological pathways separable from the establishment of cell polarity per se. Lastly, we show that low Lkb1 expression in human endometrial cancers strongly correlates with invasiveness. Our results thus show that Lkb1 serves an important and general role in constraining the malignant transformation of the endometrium and suggest that Lkb1 loss may have clinical utility as a prognostic marker.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Colony generation and Ad-Cre administration. This study was approved by an institutional animal care and use committee. The mouse null (–) and floxed (L) alleles of Lkb1 used in this study have previously been reported. Both were backcrossed for six generations to strain FVB, and genotyping was done as described (18). Experimental Lkb1–/+ animals were generated by outcrossing to FVB mice; Lkb1L/L mice were propagated as homozygotes. Intrauterine Ad-Cre injections of Lkb1L/L mice were done as described (19) with modifications: Ad5CMVCre adenovirus (University of Iowa) was stored in 3% sucrose/PBS at 6 x 10–10 plaque-forming units (pfu)/µL in 20-µL aliquots, thawed immediately on ice before use, diluted in MEM (Sigma) to a final concentration of 2.4 x 10–9 pfu/µL, and incubated at room temperature x 30 min. Female mice (6–8 weeks old) at diestrus or postestrus were anesthetized, and Genemate 57-mm mulitFlex Tips (ISCBioExpress) connected to 1-mL syringes were used to deliver 50 µL of adenovirus/MEM through the cervical os and into each uterine horn.

Tissue microarray generation and analysis, histologic techniques, immunohistochemistry, and RNA in situ hybridization. The tissue microarray contained 2 x 0.4-mm cores of paraffin blocks from each of 190 randomly selected primary endometrial cancers representing a broad range of histologic subtypes. Grade and stage were determined at the time of surgery. Grade 1 tumors were classified as low grade and grade 2 + 3 tumors as high. For immunohistochemistry, 5-µm sections were deparaffinized and hydrated in an ethanol series. Slides for all antibodies except lectin were subjected to antigen retrieval by boiling in 10 mmol/L Na citrate and cooled at room temperature x 20 min; slides for lectin immunohistochemistry were not subjected to antigen retrieval. Antibodies used were lectin, Ulex Europaeus (Sigma) at 1:100 dilution, smooth muscle actin 1:1,000 (Sigma), phosphorylated (p-) eukaryotic translation initiation factor 4E –binding protein 1 (4EBP1; Thr70) (Cell Signaling Technologies), p-Akt(Ser473) (Cell Signaling Technologies), p-AMPK (Cell Signaling Technologies), phosphorylated acetylcoenzymeA carboxylase (Cell Signaling Technologies), phosphorylated S6 kinase (p-S6K; Cell Signaling Technologies), and Lkb1 1:1,000 (18). The detection system was Immpress (Vector). For RNA in situ hybridization, sense and antisense probes of a full-length Lkb1 cDNA were made using the DIG RNA Labeling Mix (Roche) and in situ hybridization was done according to standard protocols (20). For apoptotic and mitotic counts in primary tumors, apoptotic bodies and mitotic figures were counted per standard morphologic criteria on routine H&E sections. Counts were obtained from normal endometrial and tumor epithelium from the same animal.

Western blot analysis. Normal and tumorous uterine tissues were homogenized in lysis buffer [50 mmol/L HEPES, 250 mmol/L NaCl, 2 mmol/L EDTA, 25% Glycerol, 1% NP40, 0.1% SDS, protease inhibitors (Roche), and phosphatase inhibitors (Calbiochem)] on ice. Retroviruses expressing Lkb1 or Lkb-KD were generated by transfecting 293T cells with pBabe-Lkb1 or Lkb-KD plasmids and collecting supernatant media at 36 h posttransfection. Lkb1-deficient squamous cell carcinoma cell lines were infected with retroviruses expressing Lkb1 or Lkb1-KD,8 selected with puromycin and harvested in lysis buffer 48 h postinfection, and then subjected to SDS-PAGE and immunoblotted with indicated antibodies. Cell lines were derived from squamous cell carcinoma as described (21).

Electron microscopy. Tissues were cut into 1-mm cubes, fixed in 2% glutaraldehyde, 0.1 mol/L sodium cacodylate, and postfixed in 1% OsO4/2% uranyl acetate. Samples were dehydrated, rinsed with propylene oxide, embedded in Epon, polymerized overnight at 60°C, sectioned (50–70 nm) using a Reichert ultracut E microtome, and mounted on formvar-carbon–coated copper grids. Images were obtained on a JEOL 1200EX transmission electron microscope.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Female Lkb1–/+ mice spontaneously develop uterine tumors (endometrial adenocarcinomas). The well-documented association between Peutz-Jeghers syndrome and tumors of the female reproductive tract prompted an investigation of the female reproductive tract in Lkb1–/+ mice. Cohorts of Lkb1–/+ and sibling wild-type mice were generated and aged. As previously reported, by 43 weeks of age, 40 of 59 Lkb1–/+ mice (versus 0 of 65 wild-type control mice; P = 2.5 x 10–18, Fisher's exact test) developed gastrointestinal polyps resulting in symptoms of gastrointestinal obstruction. Of surviving females up to 55 weeks of age, 4 of 15 (26.7%) also had large uterine tumors readily detectable on routine gross examination. Increased tumor incidence was not observed in anatomic sites other than the uterus and gastrointestinal tract. Serial sectioning and microscopic analysis of uteri from all 15 females revealed the presence of smaller but histologically identical tumors in an additional 4 of 15 (26.7%) of females. Thus, in total, 8 of 15 (53%) of surviving Lkb1–/+ females spontaneously developed uterine neoplasms by 55 weeks of age (versus 0 of 54 wild-type control mice; P = 7.7 x 10–7, Fisher's exact test).

Histologic analysis revealed that these uterine neoplasms were histologically distinctive, well-differentiated endometrial adenocarcinomas. The uterus consists of three layers: endometrium (epithelium plus stroma), myometrium, and serosa (the outer single-cell layer of mesothelium); the normal endometrial-myometrial interface is shown by a dashed line in Fig. 1A . The Lkb1–/+ uterine tumors consisted of endometrial-type glands invading into myometrium (Fig. 1A), where endometrial glands are not normally present. Higher magnification revealed that the tumors consisted of very well-differentiated glands diffusely infiltrating through surrounding tissue. Despite their well-differentiated appearance, the tumors frequently invaded through the entire myometrial thickness and reached the serosa (Fig. 1B). To better define the uterine layers and confirm myometrial invasion in Lkb1–/+ mice, immunohistochemistry was done against smooth muscle actin, revealing that in tumor areas, myometrium was normally developed yet extensively invaded by well-differentiated glands of adenocarcinoma (Fig. 1C). Furthermore, tumor glands were not associated with endometrial stroma, arguing against adenomyosis (maldevelopment of normal endometrium within the myometrium) and consistent with true invasion (Fig. 1C, inset).


Figure 1
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Extremely well-differentiated endometrial carcinomas arise spontaneously in Lkb1–/+ female mice. A, example of tumor invading through myometrium, H&E-stained section. Dashed line demarcates endometrial-myometrial interface. Asterisks, tumor in myometrium. Bar, 1 mm. B, higher magnification of boxed area in A. Tumors consist of well-differentiated tumor epithelium forming glandular structures separated by abundant intervening stroma. The tumor invades through the full thickness of the myometrium, with some glands reaching the serosal surface (arrow). Bar, 100 µm. C, low-magnification view of tissue section of another tumor. Immunohistochemistry against smooth muscle actin (SMA) highlights invasive growth and extensive infiltration into myometrium. Inset, high magnification showing tumor cells in direct contact with smooth muscle actin–positive myometrium and absence of intervening endometrial stroma. D, in situ hybridization of normal endometrium (wild-type mouse) using antisense probe reveals high Lkb1 expression in endometrial glands and surface epithelium. Inset, higher magnification of area in dashed line showing that all cells uniformly express Lkb1. A sense control probe gave no detectable signal (not shown). L, lumen. Bar, 100 µm.

 
Northern blot analysis showed that Lkb1 is broadly expressed in many mouse tissues, including the uterus (data not shown). To further define the cellular compartments where Lkb1 is expressed within the uterus, RNA in situ hybridization was done. Lkb1 was detectable throughout the uterus but most highly expressed in the endometrial epithelium, consistent with a cell-autonomous role in the regulation of endometrial cell growth (Fig. 1D). No significant variation in Lkb1 expression was observed on different days of the estrus cycle, suggesting that Lkb1 expression is not steroid hormone dependent (data not shown).

Postnatal endometrial-specific gene inactivation shows that Lkb1 is a bona fide endometrial tumor suppressor. The above findings were most consistent with neoplastic growth but did not formally exclude a developmental abnormality such as adenomyosis. To exclude this possibility and to determine if Lkb1 functions in a cell-autonomous manner, we injected adenovirus expressing the Cre recombinase (Ad-Cre) into the uterine lumen of female mice homozygous for a floxed Lkb1 allele (L). As previously shown, this approach results in efficient and transient transduction of endometrial epithelium but, because of the presence of tight junctions, not of endometrial stromal or other cell types within the uterus (19).

Cohorts of 17 Lkb1L/L homozygous floxed and 30 control wild-type female mice were injected with Ad-Cre at 6 weeks of age and euthanized at 9 months posttreatment. PCR confirmed Cre-mediated recombination and the presence of the null allele in uterine DNA, but not in control tail DNA (data not shown). Of the 17 Lkb1L/L mice, 11 (65%) developed uterine tumors, versus 0 of 30 of controls mice similarly treated with Ad-Cre (P = 7.1 x 10–7, Fisher's exact test). The majority of these tumors were confined to the uterus, but one was diffusely metastatic within the peritoneum (Fig. 2A ). No extrauterine tumors were observed. Histologically, the tumors in Ad-Cre–treated Lkb1L/L females were identical to those in Lkb1–/+ females. The tumors in Lkb1L/L females similarly involved the myometrium (Fig. 2B) and consisted of diffusely infiltrating well-differentiated glandular tumor epithelium that frequently invaded through the entire thickness of the myometrium, often reaching the serosal surface. Distant metastatic tumor glands were indistinguishable from those in primary tumors (Fig. 2C, peripancreatic tumor deposit). Mitotic and apoptotic counts were ascertained for uterine tumors relative to adjacent normal endometrial epithelium within the same mouse (n = 8). Data were plotted for individual mice because in randomly cycling females, endometrial cell growth/death vary through the estrus cycle, potentially obscuring aggregate analysis. In some tumors, mitotic activity and apoptotic index were significantly altered (Fig. 2D), consistent with the notion that Lkb1–/+ endometrial tumor cells, despite their well-differentiated appearance, do exhibit abnormal responses to exogenous growth and survival stimuli.


Figure 2
View larger version (57K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Endometrial-specific deletion of Lkb1 provokes well-differentiated endometrial adenocarcinomas. A, left, small tumor (arrow) protruding through uterine horn; the ovary and oviduct are near top of the picture. Right, large tumor with widespread peritoneal metastases (arrows). Ad-Cre was introduced into the uterine lumen of Lkb1L/L females at 6 wk of age and euthanized 9 mo posttreatment. B, left, low-magnification view of full-thickness tumor invasion through myometrium, H&E-stained section. Dashed line demarcates endometrial-myometrial interface. Asterisks, tumor in myometrium. Right, higher magnification of boxed area in B. Tumor is well differentiated and histologically identical to tumors observed in Lkb1–/+ females. Tumor glands invade through myometrium (asterisks) with some glands reaching the serosa (S). Some tumor epithelium also lines the serosal surface (arrows). Bar, 1 mm (A and B). C, metastatic tumor far from the primary (peripancreatic location) retains well-differentiated appearance and tumor glands are histologically identical to those primary tumors, being characterized by diffusely infiltrating scant glands (asterisk). P, pancreas. Bar, 100 µm. D, apoptotic and mitotic counts in primary tumors. Y axis, absolute number of mitotic figures or apoptotic bodies per 100 glands. Each data line shows counts obtained from normal endometrium and tumor epithelium from the same animal.

 
Loss of Lkb1 protein in endometrial tumor cells results in misregulation of AMPK signaling downstream of Lkb1. PCR analysis of DNA from various tissues in Ad-Cre–treated animals showed Cre-mediated recombination and the presence of the null allele in uterus but not other tissues (data not shown). Immunohistochemistry confirmed that Lkb1 protein was indeed undetectable in tumor epithelial cells, whereas it remained detectable in adjacent stromal cells (Fig. 3A ). To gain insights into the molecular mechanisms by which Lkb1 loss contributes to endometrial carcinogenesis, we analyzed the activation/phosphorylation status of selected components of the kinase cascade downstream of Lkb1. These in situ studies permitted analysis of pathway status in Lkb1 endometrial tumors although they are composed predominantly of stromal cells.


Figure 3
View larger version (65K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Misregulation of AMPK signaling in endometrium following Lkb1 deletion. A, immunohistochemical studies. For p-S6K, most areas of tumor showed few positive cells similar to normal control endometrium (arrows), but small foci (comprising <20% of tumor epithelium) contained somewhat increased numbers of positive cells (inset). All images are of same magnification except the top right image, which is shown at 2.5x higher magnification to highlight abundant Lkb1 expression in stroma (S). Slides were immunostained, developed in diaminobenzidene, and counterstained with hematoxylin. B, Western blot analysis of another tumor also shows decreased levels of p-AMPK. Lkb1-deficient squamous carcinoma cells (SCC) with enforced expression of wild-type or kinase-dead (KD) forms of Lkb1 served as additional controls.

 
Lkb1-mediated repression of mTOR signaling through the AMPK-Tsc pathway is thought to be important for tumor suppression (2). Endometrial tumor cells showed significantly decreased p-AMPK(Thr172) levels, consistent with inactivation of the Lkb1 kinase. Correspondingly, we also observed decreased phosphorylation of the AMPK target, acetylcoenzymeA carboxylase, indicating that Lkb1 deletion was associated with defective AMPK signaling in these tumors (Fig. 3A). Surprisingly, however, inactivation of Lkb1 in endometrial epithelium did not globally affect the phosphorylation status of ribosomal protein S6K (Fig. 3A), an important target of mTOR, and, indirectly, AMPK (2). Previous studies (using an antibody from the same source) have documented uniformly increased p-S6K levels in the epithelium of gastrointestinal polyps from Lkb1 mice (22). Although we did observe small foci of p-S6K–positive cells (Fig. 3B, inset), this seems to be secondary to focal increases in mitotic rate because S6K is phosphorylated during mitosis (23) and p-S6K is also a well-documented marker of mitotic cells (24). We also did not observe significant alterations in the patterns or levels of phosphorylation of other putative mTOR substrates including p-4EBP1(Thr70) or p-Akt(Ser473) (data not shown). Next, Western blot analyses were done. Although the presence of stroma may have limited the sensitivity of this approach, significant decreases in p-AMPK were observed, whereas levels of total AMPK were unaffected. Furthermore, analysis of other mTOR pathway components was consistent with the in situ studies, including increased phosphorylation of S6K (Fig. 3A and B). Thus, although misregulated mTOR signaling has been reported in Lkb1–/– associated gastrointestinal polyps (22), it seems that this is not a prime pathway by which Lkb1 loss promotes endometrial carcinogenesis.

Lkb1 tumors are extremely well differentiated without overt defects in cell polarity. At high magnification, Lkb1 tumors seemed to be extremely well differentiated, so much so that isolated tumor glands were virtually indistinguishable from normal endometrial glands (Fig. 4A ). Although loss of polarity is characteristic of adenocarcinomas, Lkb1 tumor cells exhibited normal polarity in H&E-stained tissue sections and consisted of tall columnar cells with basally located nuclei and vacuolated secretory-type apical cytoplasm. Nuclear morphology was also not significantly altered, suggesting that these tumors are not highly aneuploid (Fig. 4A). These findings were surprising because Lkb1 is required for the establishment of cell polarity in a variety of other cell types including gastrointestinal epithelium (2, 9). To further analyze potential defects in cell polarity, we used an antibody against {alpha}-lectin (Ulex Europaeus; ref. 25). Expression in tumors was confined to the apical surface, further arguing against gross polarity defects (Fig. 4B). Lastly, ultrastructural analysis also failed to reveal cell polarity defects. Tumor cells expressed numerous microvilli on apical, but not basolateral, surfaces, and microvillous morphology and density were unaltered. Furthermore, no defects in other polarized structures such as tight junctions were observed in tumor cells (Fig. 4C).


Figure 4
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. Normal cell polarity in Lkb1 tumors from Ad-Cre–treated Lkb1L/L females. A, high magnification showing portion of a single gland, H&E-stained sections. Left, normal endometrium (em); right, tumor (tu). Tumor glands are indistinguishable from normal glands. Tall columnar tumor cells exhibit normal polarity with basal nuclei and apical secretory-type cytoplasm. Bar, 10 µm. B, lectin immunohistochemistry shows normal differentiation and expression on apical membrane in both normal endometrium and tumor (arrowheads; a single positive secretory vacuole is apparent in the tumor gland). Bar, 10 µm. C, ultrastructural analysis reveals normal apical differentiation including presence of abundant microvilli on apical cell surface. Morphology, distribution, and number of microvilli in tumor are similar to those of normal endometrium. Normal apical tight junctional complexes are also observed in tumor epithelial cells (white arrows). Images are representative of large areas scanned by electron microscope. Bar, 0.4 µm (both images at same magnification).

 
Low LKB1 expression in human endometrial cancers correlates with invasiveness. Sequencing of LKB1 coding exons in 40 human endometrial tumors revealed a non–germ line mutation leading to a single conservative amino acid change (p.Arg331His) in only 1 of 40 human endometrial tumors.9 To determine if LKB1 down-regulation by other mechanisms might contribute to endometrial cancer progression, expression was assessed by immunohistochemistry of 190 endometrial cancers in a single tissue microarray. The antibody was first validated for LKB1 detection following formalin fixation/paraffin embedding of 293T cells with enforced LKB1 expression (Fig. 5A ). Expression in primary tumor cells was classified as high, intermediate, or low (Fig. 5B). In high-grade versus low-grade tumors, a statistically significant increase in cases with low LKB1 expression was observed (P = 0.035, Fisher's exact test; Fig. 5C). We then categorized expression by clinical stage. Stage 1A tumors are defined by lack of myometrial invasion, whereas 1B tumors invade through <50% of the myometrium and 1C tumors invade through >50% of the myometrium. LKB1 expression strongly correlated inversely with stage, with stage 1C tumors being much more likely to express low levels of Lkb1 than 1A tumors (P = 0.0053, Fisher's exact test; Fig. 5D). These data suggest that LKB1 constrains invasion in vivo, consistent with our mouse studies. Furthermore, these findings suggest that down-regulation of LKB1, which may occur through a variety of mechanisms, is an important molecular event driving endometrial cancer progression. Our results also suggest that detection of LKB1 in primary tumors, as assessed by immunohistochemistry or other techniques, may aid in the identification of tumors more likely to be invasive and hence associated with a worse prognosis.


Figure 5
View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Low expression levels of Lkb1 in endometrial cancers correlate with invasiveness. A, validation of anti-Lkb1 antibodies in samples subjected to formalin fixation and paraffin embedding. 293T cells were transfected with a full-length human Lkb1 cDNA, trypsinized 24 h later, resuspended in a soft agar plug, and the solid plug was then fixed in 10% formalin overnight. Left, control untransfected cells; right, Lkb1-transfected cells. Antibody detects endogenous Lkb1 in untransfected control 293T cells by Western blot (not shown) and endogenous Lkb1 is also detectable as weak signal by immunohistochemistry (left). In transfected cells, a proportion of cells shows much higher signal (transfection is <100% efficient). Slides were counterstained with hematoxylin. Bar, 20 µm. B, representative Lkb1 expression levels categorized as high, intermediate, and low. Only staining in tumor cells (not stroma) was scored. A and B, bar, 20 µm; slides were counterstained with hematoxylin. C, Lkb1 expression in low-grade and high-grade endometrial cancers (n = 190). D, Lkb1 expression in stage 1A, 1B, and 1C endometrial cancers (n = 121). FIGO, International Federation of Gynecology and Obstetrics.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
It is striking that germ line or somatic inactivation of a single gene, Lkb1, provokes endometrial cancers in >50% of female mice. This high tumor incidence suggests that Lkb1 is, unexpectedly, a very potent endometrial tumor suppressor, and our finding that decreased expression correlates with invasiveness in human tumors is the first evidence that Lkb1 similarly drives human endometrial carcinogenesis. Another significant finding in our studies relates to the role of mTOR as a downstream target of Lkb1. The Lkb1-mediated repression of mTOR through the AMPK-Tsc pathway is thought to be important for tumor suppression, and the common regulation of mTOR by Tsc, Pten, and Lkb1 provides a unifying molecular explanation for hamartoma syndromes. The Lkb1-mutant endometrial tumors in our studies did not show overt mTOR dysregulation despite having defects in AMPK signaling. Hence, the Lkb1 tumor suppressor program in endometrium may not involve mTOR repression, or alternatively, there may be different mechanisms for malignant and benign tumor suppression. Along these lines, it is notable that Lkb1 deficiency promotes a highly invasive phenotype, whereas Pten inactivation promotes multifocal endometrial hyperplasias that become invasive only infrequently and with relatively long latency (26–31).

That only a subset of mice developed uterine tumors and the focal nature of these neoplasms suggest that cooperating genetic events are required for Lkb1-driven neoplasia. Loss of the second Lkb1 allele may not be requisite because some gastrointestinal polyps in Peutz-Jeghers syndrome patients or Lkb1–/+ mice do not exhibit loss of heterozygosity (LOH) or loss of Lkb1 expression (2, 18). We similarly did not find consistent evidence of LOH in Lkb1–/+ tumors, although the highly stromal makeup of these tumors may have obscured this analysis.10 We also note that many tumors invaded through the entire myometrium, yet remained confined to the uterus with no evidence of metastasis, suggesting that additional genetic events are required for extrauterine spread. The nature of cooperating genetic events in Lkb1-driven neoplasia and their precise order in tumor initiation and progression remain important unresolved issues, but our mouse model should prove a useful tool to address these questions.

We show that Lkb1 functions as an endometrial cancer invasion gene, consistent with a recently described mouse model of lung cancer where Lkb1 deficiency conferred significantly increased mortality (32). Cells in most cancers exhibit multiple morphologic abnormalities (dedifferentiation, nuclear atypia, etc.) that are manifestations of growth signal self-sufficiency, invasion, and genomic instability (33). It is remarkable that Lkb1 endometrial cancers exhibit a relatively pure invasive phenotype with minimal dedifferentiation. The lack of gross polarity defects further suggests that Lkb1 inactivation can promote cancer through diverse, highly context-dependent mechanisms, and that the functions of Lkb1 in regulating invasion and cell polarity are separate and distinct. Along these lines, the absence of widespread S6K hyperphosphorylation in Lkb1 endometrial cancers (in contrast to gastrointestinal polyps in Lkb1 mice; ref. 22) lends further credence to the notion that Lkb1 promotes tumor growth via context-dependent mechanisms, with different downstream effectors varying in their significance in different tumor types.

We speculate that a variety of mechanisms may serve to down-regulate Lkb1 in human cancers; these may include promoter hypermethylation (5) or proteasomal degradation (34). Lastly, important parallels can be drawn between the endometrial tumors we describe in mice and the endocervical neoplasms observed in women with Peutz-Jeghers syndrome. The striking similarities of these tumors (extremely well-differentiated morphology, yet highly invasive) reveal fundamental similarities in Lkb1-driven carcinogenesis in the female reproductive tract in mice and women.


   Acknowledgments
 
Grant support: Sidney Kimmel scholarship (D.H. Castrillon), Sidney Kimmel Translational Science Award, Young Investigator Award from the American Cancer Society/University of Texas Southwestern Simmons Comprehensive Cancer Center, K26 mouse pathobiologist award from the NIH (K26RR024196), NIH NRSA F31 fellowship (C.M. Contreras), Joan Scarangello Foundation to Conquer Lung Cancer (K-K. Wong), the Flight Attendant Medical Research Institute (K-K. Wong), NIH Specialized Program of Research Excellence in Uterine Cancer grant 1P50CA098258-01 (R.R. Broaddus), the Waxman Foundation (N. Bardeesy), the Harvard Stem Cell Institute (N. Bardeesy), and the Linda Verville Foundation (N. Bardeesy).

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.

We thank the M. D. Anderson Uterine Cancer Spore Pathology Core Facility for technical assistance and Tom Januszewski and George Lawton of the University of Texas Southwestern Molecular and Cellular Imaging Core Facility for assistance in electron microscopy.


   Footnotes
 
8 In preparation. Back

9 Unpublished data. Back

10 Unpublished data. Back

Received 8/23/07. Revised 11/ 5/07. Accepted 11/19/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Hemminki A, Markie D, Tomlinson I, et al. A serine/threonine kinase gene defective in Peutz-Jeghers syndrome. Nature 1998;391:184–7.[CrossRef][Medline]
  2. Alessi DR, Sakamoto K, Bayascas JR. Lkb1-dependent signaling pathways. Annu Rev Biochem 2006;75:137–63.[CrossRef][Medline]
  3. Hearle N, Schumacher V, Menko FH, et al. Frequency and spectrum of cancers in the Peutz-Jeghers syndrome. Clin Cancer Res 2006;12:3209–15.[Abstract/Free Full Text]
  4. Ghaffar H, Sahin F, Sanchez-Cepedes M, et al. LKB1 protein expression in the evolution of glandular neoplasia of the lung. Clin Cancer Res 2003;9:2998–3003.[Abstract/Free Full Text]
  5. Esteller M, Avizienyte E, Corn PG, et al. Epigenetic inactivation of LKB1 in primary tumors associated with the Peutz-Jeghers syndrome. Oncogene 2000;19:164–8.[CrossRef][Medline]
  6. Sato N, Rosty C, Jansen M, et al. STK11/LKB1 Peutz-Jeghers gene inactivation in intraductal papillary-mucinous neoplasms of the pancreas. Am J Pathol 2001;159:2017–22.[Abstract/Free Full Text]
  7. Sanchez-Cespedes M, Parrella P, Esteller M, et al. Inactivation of LKB1/STK11 is a common event in adenocarcinomas of the lung. Cancer Res 2002;62:3659–62.[Abstract/Free Full Text]
  8. Young RH, Clement PB. Endocervical adenocarcinoma and its variants: their morphology and differential diagnosis. Histopathology 2002;41:185–207.[CrossRef][Medline]
  9. Barnes AP, Lilley BN, Pan YA, et al. LKB1 and SAD kinases define a pathway required for the polarization of cortical neurons. Cell 2007;129:549–63.[CrossRef][Medline]
  10. Baas AF, Kuipers J, van der Wel NN, et al. Complete polarization of single intestinal epithelial cells upon activation of LKB1 by STRAD. Cell 2004;116:457–66.[CrossRef][Medline]
  11. Spicer J, Ashworth A. LKB1 kinase: master and commander of metabolism and polarity. Curr Biol 2004;14:R383–5.[CrossRef][Medline]
  12. Clement PB, Young RH. Endometrioid carcinoma of the uterine corpus: a review of its pathology with emphasis on recent advances and problematic aspects. Adv Anat Pathol 2002;9:145–84.[CrossRef][Medline]
  13. Inoki K, Zhu T, Guan KL. TSC2 mediates cellular energy response to control cell growth and survival. Cell 2003;115:577–90.[CrossRef][Medline]
  14. Mirouse V, Swick LL, Kazgan N, St Johnston D, Brenman JE. LKB1 and AMPK maintain epithelial cell polarity under energetic stress. J Cell Biol 2007;177:387–92.[Abstract/Free Full Text]
  15. Zhang L, Li J, Young LH, Caplan MJ. AMP-activated protein kinase regulates the assembly of epithelial tight junctions. Proc Natl Acad Sci U S A 2006;103:17272–7.[Abstract/Free Full Text]
  16. Zheng B, Cantley LC. Regulation of epithelial tight junction assembly and disassembly by AMP-activated protein kinase. Proc Natl Acad Sci U S A 2007;104:819–22.[Abstract/Free Full Text]
  17. Lizcano JM, Goransson O, Toth R, et al. LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR-1. EMBO J 2004;23:833–43.[CrossRef][Medline]
  18. Bardeesy N, Sinha M, Hezel AF, et al. Loss of the Lkb1 tumour suppressor provokes intestinal polyposis but resistance to transformation. Nature 2002;419:162–7.[CrossRef][Medline]
  19. Beauparlant SL, Read PW, Di Cristofano A. In vivo adenovirus-mediated gene transduction into mouse endometrial glands: a novel tool to model endometrial cancer in the mouse. Gynecol Oncol 2004;94:713–8.[CrossRef][Medline]
  20. Gallardo TD, John GB, Shirley L, et al. Genomewide discovery and classification of candidate ovarian fertility genes in the mouse. Genetics 2007;177:179–94.[Abstract/Free Full Text]
  21. Pera MF, Gorman PA. In vitro analysis of multistage epidermal carcinogenesis: development of indefinite renewal capacity and reduced growth factor requirements in colony forming keratinocytes precedes malignant transformation. Carcinogenesis 1984;5:671–82.[Abstract/Free Full Text]
  22. Shaw RJ, Bardeesy N, Manning BD, et al. The LKB1 tumor suppressor negatively regulates mTOR signaling. Cancer Cell 2004;6:91–9.[CrossRef][Medline]
  23. Shah OJ, Ghosh S, Hunter T. Mitotic regulation of ribosomal S6 kinase 1 involves Ser/Thr, Pro phosphorylation of consensus and non-consensus sites by Cdc2. J Biol Chem 2003;278:16433–42.[Abstract/Free Full Text]
  24. Schmidt T, Wahl P, Wuthrich RP, et al. Immunolocalization of phospho-S6 kinases: a new way to detect mitosis in tissue sections and in cell culture. Histochem Cell Biol 2007;127:123–9.[CrossRef][Medline]
  25. Munson L, Wilkinson JE, Schlafer DH. Effects of substrata on the polarization of bovine endometrial epithelial cells in vitro. Cell Tissue Res 1990;261:155–61.[CrossRef][Medline]
  26. Stambolic V, Tsao MS, Macpherson D, Suzuki A, Chapman WB, Mak TW. High incidence of breast and endometrial neoplasia resembling human Cowden syndrome in pten+/– mice. Cancer Res 2000;60:3605–11.[Abstract/Free Full Text]
  27. Wang H, Douglas W, Lia M, et al. DNA mismatch repair deficiency accelerates endometrial tumorigenesis in Pten heterozygous mice. Am J Pathol 2002;160:1481–6.[Abstract/Free Full Text]
  28. You MJ, Castrillon DH, Bastian BC, et al. Genetic analysis of Pten and Ink4a/Arf interactions in the suppression of tumorigenesis in mice. Proc Natl Acad Sci U S A 2002;99:1455–60.[Abstract/Free Full Text]
  29. Podsypanina K, Ellenson LH, Nemes A, et al. Mutation of Pten/Mmac1 in mice causes neoplasia in multiple organ systems. Proc Natl Acad Sci U S A 1999;96:1563–8.[Abstract/Free Full Text]
  30. Di Cristofano A, Pesce B, Cordon-Cardo C, Pandolfi PP. Pten is essential for embryonic development and tumour suppression. Nat Genet 1998;19:348–55.[CrossRef][Medline]
  31. Vilgelm A, Lian Z, Wang H, et al. Akt-mediated phosphorylation and activation of estrogen receptor {alpha} is required for endometrial neoplastic transformation in Pten+/– mice. Cancer Res 2006;66:3375–80.[Abstract/Free Full Text]
  32. Ji H, Ramsey MR, Hayes DN, et al. LKB1 modulates lung cancer differentiation and metastasis. Nature 2007;448:807–10.[CrossRef][Medline]
  33. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:57–70.[CrossRef][Medline]
  34. Boudeau J, Deak M, Lawlor MA, Morrice NA, Alessi DR. Heat-shock protein 90 and Cdc37 interact with LKB1 and regulate its stability. Biochem J 2003;370:849–57.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Sci SignalHome page
H. Cheng, P. Liu, Z. C. Wang, L. Zou, S. Santiago, V. Garbitt, O. V. Gjoerup, J. D. Iglehart, A. Miron, A. L. Richardson, et al.
SIK1 Couples LKB1 to p53-Dependent Anoikis and Suppresses Metastasis
Sci. Signal., July 21, 2009; 2(80): ra35 - ra35.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. Rev.Home page
M. Jansen, J. P. ten Klooster, G. J. Offerhaus, and H. Clevers
LKB1 and AMPK Family Signaling: The Intimate Link Between Cell Polarity and Energy Metabolism
Physiol Rev, July 1, 2009; 89(3): 777 - 798.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. Rev.Home page
G. R. Steinberg and B. E. Kemp
AMPK in Health and Disease
Physiol Rev, July 1, 2009; 89(3): 1025 - 1078.
[Abstract] [Full Text] [PDF]

Home page
 CarcinogenesisHome page
S. Rodriguez-Nieto and M. Sanchez-Cespedes
BRG1 and LKB1: tales of two tumor suppressor genes on chromosome 19p and lung cancer
Carcinogenesis, April 1, 2009; 30(4): 547 - 554.
[Abstract] [Full Text] [PDF]

Home page
 Genes Dev.Home page
R. G. Jones and C. B. Thompson
Tumor suppressors and cell metabolism: a recipe for cancer growth
Genes & Dev., March 1, 2009; 23(5): 537 - 548.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Contreras, C. M.
Right arrow Articles by Castrillon, D. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Contreras, C. M.
Right arrow Articles by Castrillon, D. H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online