Changes in villin synthesis and subcellular distribution during intestinal differentiation of HT29-18 clones.

Brush border in enterocytes is a cell surface specialization intimately associated with terminal differentiation of these cells. HT29-18, a clone derived from the HT-29 human colonic adenocarcinoma cell line, and HT29-18-C1, a subclone from HT29-18 described in the companion paper (Huet, C., C. Sahuquillo-Merino, E. Coudrier, and D. Louvard, 1987, J. Cell Biol., 105:345-357), undergo terminal differentiation with brush borders in the absence of glucose or upon replacement of glucose by galactose in the medium. Taking advantage of this clone and its subclone which can be manipulated in vitro, we have studied the synthesis and subcellular distribution of villin, one major protein in the microvillus core of the brush border. For this study, a monoclonal antibody against villin (BDID2C3) has been isolated and characterized in detail. In addition an ELISA has been set up to measure villin accurately in total cell extracts. Villin content in differentiated HT29-18 cells is close to that seen in normal human colonic cells but 10 times lower in undifferentiated HT29-18 cells. The rate of villin synthesis is dramatically increased in the course of enterocytic differentiation, while villin is remarkably stable after synthesis. We have recently shown, using a cDNA probe for villin, that this change is controlled either at the transcription level or by RNA stabilization (Pringault, E., M. Arpin, A. Garcia, J. Finidori, and D. Louvard, 1986, EMBO (Eur. Mol. Biol. Organ.) J., 5:3119-3124). As shown by immunofluorescence and immunogold labelings, villin is targeted to the brush border area of differentiated HT29-18 cells but remains diffusely distributed in undifferentiated ones.

In vivo, villin is required for Ca(2+)-dependent F-actin disruption in intestinal brush borders.

Villin is an actin-binding protein localized in intestinal and kidney brush borders. In vitro, villin has been demonstrated to bundle and sever F-actin in a Ca(2+)-dependent manner. We generated knockout mice to study the role of villin in vivo. In villin-null mice, no noticeable changes were observed in the ultrastructure of the microvilli or in the localization and expression of the actin-binding and membrane proteins of the intestine. Interestingly, the response to elevated intracellular Ca(2+) differed significantly between mutant and normal mice. In wild-type animals, isolated brush borders were disrupted by the addition of Ca(2+), whereas Ca(2+) had no effect in villin-null isolates. Moreover, increase in intracellular Ca(2+) by serosal carbachol or mucosal Ca(2+) ionophore A23187 application abolished the F-actin labeling only in the brush border of wild-type animals. This F-actin disruption was also observed in physiological fasting/refeeding experiments. Oral administration of dextran sulfate sodium, an agent that causes colonic epithelial injury, induced large mucosal lesions resulting in a higher death probability in mice lacking villin, 36 +/- 9.6%, compared with wild-type mice, 70 +/- 8.8%, at day 13. These results suggest that in vivo, villin is not necessary for the bundling of F-actin microfilaments, whereas it is necessary for the reorganization elicited by various signals. We postulate that this property might be involved in cellular plasticity related to cell injury.

I-SceI-induced gene replacement at a natural locus in embryonic stem cells.

Gene targeting is a very powerful tool for studying mammalian development and physiology and for creating models of human diseases. In many instances, however, it is desirable to study different modifications of a target gene, but this is limited by the generally low frequency of homologous recombination in mammalian cells. We have developed a novel gene-targeting strategy in mouse embryonic stem cells that is based on the induction of endogenous gap repair processes at a defined location within the genome by induction of a double-strand break (DSB) in the gene to be mutated. This strategy was used to knock in an NH2-ezrin mutant in the villin gene, which encodes an actin-binding protein expressed in the brush border of the intestine and the kidney. To induce the DSB, an I-SceI yeast meganuclease restriction site was first introduced by gene targeting to the villin gene, followed by transient expression of I-SceI. The repair of the ensuing DSB was achieved with high efficiency (6 x 10[-6]) by a repair shuttle vector sharing only a 2.8-kb region of homology with the villin gene and no negative selection marker. Compared to conventional gene-targeting experiments at the villin locus, this represents a 100-fold stimulation of gene-targeting frequency, notwithstanding a much lower length of homology. This strategy will be very helpful in facilitating the targeted introduction of several types of mutations within a gene of interest.

Conditional mouse models of cancer.

The development of inducible and conditional technologies allowed us to generate transgenic mouse models that faithfully recapitulate human tumorigenesis. It is possible to control, in time and space, the development of tumors in almost every mouse tissue. The result is that now we have available mouse models for all major human cancers. Novel noninvasive approaches to tumor imaging will enable us to follow tumor development and metastasis in vivo, as well as the effects of candidate therapeutic drugs. Such new generation tumor models, which accurately emulate the disease state in situ, should provide a useful platform with which to experimentally test drugs targeted to specific gene products, or combinations of genes that control rate-limiting steps of tumor development. In this review, we focus on the different mouse models for colon cancer.

Intestinal cancer progression by mutant p53 through the acquisition of invasiveness associated with complex glandular formation.

Tumor suppressor TP53 is frequently mutated in colorectal cancer (CRC), and most mutations are missense type. Although gain-of-functions by mutant p53 have been demonstrated experimentally, the precise mechanism for malignant progression in in vivo tumors remains unsolved. We generated ApcΔ716 Trp53LSL•R270H villin-CreER compound mice, in which mutant p53R270H was expressed in the intestinal epithelia upon tamoxifen treatment, and examined the intestinal tumor phenotypes and tumor-derived organoids. Mutant Trp53R270H, but not Trp53-null mutation accelerated submucosal invasion with generation of desmoplastic microenvironment. The nuclear accumulation of p53 was evident in ApcΔ716 Trp53R270H/R270H homozygous tumors like human CRC. Although p53 was distributed to the cytoplasm in ApcΔ716 Trp53+/R270H heterozygous tumors, it accumulated in the nuclei at the invasion front, suggesting a regulation mechanism for p53 localization by the microenvironment. Importantly, mutant p53 induced drastic morphological changes in the tumor organoids to complex glandular structures, which was associated with the acquisition of invasiveness. Consistently, the branching scores of human CRC that carry TP53 mutations at codon 273 significantly increased in comparison with those of TP53 wild-type tumors. Moreover, allografted ApcΔ716 Trp53R270H/R270H organoid tumors showed a malignant histology with an increased number of myofibroblasts in the stroma. These results indicate that nuclear-accumulated mutant p53R270H induces malignant progression of intestinal tumors through complex tumor gland formation and acquisition of invasiveness. Furthermore, RNA sequencing analyses revealed global gene upregulation by mutant p53R270H, which was associated with the activation of inflammatory and innate immune pathways. Accordingly, it is possible that mutant p53R270H induces CRC progression, not only by a cell intrinsic mechanism, but also by the generation or activation of the microenvironment, which may synergistically contribute to the acceleration of submucosal invasion. Therefore, the present study indicates that nuclear-accumulated mutant p53R270H is a potential therapeutic target for the treatment of advanced CRCs.

Presence of villin, a tissue-specific cytoskeletal protein, in sera of patients and an initial clinical evaluation of its value for the diagnosis and follow-up of colorectal cancers.

Villin is an actin-binding protein found in a few normal adult epithelia, namely epithelial cells in the digestive and urogenital tracts. Moreover, villin production is maintained in malignant cells. We assumed that cell lysis and necrosis of solid tumors producing villin might result in villin release into blood. We analyzed the villin content of sera from 788 patients and controls using an enzyme-linked immunosorbent assay. Patients and controls were classified into healthy donors, patients with benign diseases of the gastrointestinal tract, patients with colorectal cancers, and patients with malignant nondigestive diseases. In the panel of sera analyzed, the sensitivity of the assay for colorectal cancers was 50.5%, and its overall specificity for malignant digestive tumors was 94.5%. Results were statistically analyzed comparing each group of sera with each other. We conclude that the presence of villin is indicative of a pathological state in the gastrointestinal tract (P less than 0.001). Finally, we followed villin levels after tumor resections (60 patients). We found that the villin level in sera remains low in remissions but is raised in recurrences. We suggest that the villin assay may have clinical utility as a diagnostic adjunct for adenocarcinoma of the gastrointestinal tract. It may also have some value in monitoring patients with advancing colorectal carcinomas after resection of these tumors.

Ribosome biogenesis dysfunction leads to p53-mediated apoptosis and goblet cell differentiation of mouse intestinal stem/progenitor cells.

Ribosome biogenesis is an essential cellular process. Its impairment is associated with developmental defects and increased risk of cancer. The in vivo cellular responses to defective ribosome biogenesis and the underlying molecular mechanisms are still incompletely understood. In particular, the consequences of impaired ribosome biogenesis within the intestinal epithelium in mammals have not been investigated so far. Here we adopted a genetic approach to investigate the role of Notchless (NLE), an essential actor of ribosome biogenesis, in the adult mouse intestinal lineage. Nle deficiency led to defects in the synthesis of large ribosomal subunit in crypts cells and resulted in the rapid elimination of intestinal stem cells and progenitors through distinct types of cellular responses, including apoptosis, cell cycle arrest and biased differentiation toward the goblet cell lineage. Similar observations were made using the rRNA transcription inhibitor CX-5461 on intestinal organoids culture. Importantly, we found that p53 activation was responsible for most of the cellular responses observed, including differentiation toward the goblet cell lineage. Moreover, we identify the goblet cell-specific marker Muc2 as a direct transcriptional target of p53. Nle-deficient ISCs and progenitors disappearance persisted in the absence of p53, underlying the existence of p53-independent cellular responses following defective ribosome biogenesis. Our data indicate that NLE is a crucial factor for intestinal homeostasis and provide new insights into how perturbations of ribosome biogenesis impact on cell fate decisions within the intestinal epithelium.

Immunodetection of the microvillous cytoskeleton molecules villin and ezrin in the parasitophorous vacuole wall of Cryptosporidium parvum (Protozoa: Apicomplexa).

Microvilli - actin - villin - ezrin - Cryptosporidium parvum The sporozoites and merozoites of the Apicomplexan protozoan Cryptosporidium parvum (C. parvum) invade the apical side of enterocytes and induce the formation of a parasitophorous vacuole which stays in the brush border area and disturbs the distribution of microvilli. The vacuole is separated from the apical cytoplasm of the cell by an electron-dense layer of undetermined composition. In order to characterize the enterocyte cytoskeleton changes that occur during C. parvum invasion and development, we used both confocal immunofluorescence and immunoelectron microscopy to examine at the C.parvum-enterocyte interface the distribution of three components of the microvillous skeleton, actin, villin and ezrin. In infected cells, rhodamine-phalloidin and anti-villin and anti-ezrin antibodies recognized ring-like structures surrounding the developing parasites. By immunoelectron microscopy, both villin and ezrin were detected in the parasitophorous vacuole wall surrounding the luminal and lateral sides of the intracellular parasite. In contrast, anti-beta and anti-gamma actin antibodies showed no significant labelling of the vacuolar wall. These observations indicate that the parasitophorous vacuole wall contains at least two microvillus-derived components, villin and ezrin, as well as a low amount of F-actin. These data suggest that C.parvum infection induces a rearrangement of cytoskeleton molecules at the apical pole of the host cell that are used to build the parasitophorous vacuole.

Functional cystic fibrosis transmembrane conductance regulator tagged with an epitope of the vesicular stomatis virus glycoprotein can be addressed to the apical domain of polarized cells.

The cystic fibrosis transmembrane conductance regulator (CFTR) is a phosphorylation-activated chloride channel apically localized in epithelial cells. In cystic fibrosis patients, the gene encoding this N-linked glycoprotein is mutated. About 70% of CF patients express a mutated form of CFTR, deleted at the phenylalanine residue at position 508 (deltaF508). CFTR-deltaF508 fails to exit the endoplasmic reticulum; it remains incompletely glycosylated and is rapidly degraded. To optimize CFTR detection for membrane localization studies and biochemical studies, we tagged wild-type and deltaF508 CFTR with the VSV-G epitope at their carboxy-terminal ends. We have generated pig kidney epithelial cell clones (LLCPK1) expressing VSV-G-tagged human wild-type and deltaF508-CFTR. In CFTR-expressing cells, the transfected protein is maturated and transported to the apical membrane where it is concentrated. The cells exhibit a strong anion channel activity after stimulation by cAMP, as demonstrated by a halide sensitive fluorescent dye assay (6-methoxy-N-ethylquinominium, SPQ), and whole-cell patch-clamp approach. This activity of CFTR-VSV-G is indistinguishable from the wild-type CFTR. In contrast, in cells expressing tagged deltaF508-CFTR or in non-transfected cells, no anion channel activity could be detected after stimulation by cAMP. In deltaF508-CFTR-VSV-G-expressing cells, the mutated CFTR remained in the incompletely glycosylated form and was localized in the endoplasmic reticulum. These cell lines reproduce the cellular fate of wild-type and mutated CFTR-deltaF508. To our knowledge, they are the first differentiated epithelial cell lines stably expressing tagged CFTR and CFTR-deltaF508 in which cellular processing and functional activity of these two proteins are reproduced. Thus the addition of the VSV-G epitope does not impair the localization and function of CFTR, and these cell lines can be used to examine CFTR function in vitro.

Expression patterns of L-plastin isoform in normal and carcinomatous breast tissues.

Plastins are members of a family of actin-binding proteins which exhibit a tissue-specific expression pattern. L-plastin, which is specifically expressed in hematopoietic cell lineage, has been proposed to be involved in the control of cell adhesion and motility. This protein is also frequently expressed in cell lines derived from mammary solid tumors and therefore might be involved in cancer invasion and metastasis. We have analysed plastin expression in normal and carcinomatous breast tissues in vivo by immunohistochemistry and immunoblotting approaches using specific plastin isoform antibodies. L-plastin was not detected in normal epithelial cells of the mammary gland whereas a staining of myoepithelial cells was observed in 50% of the cases. In breast carcinomas, a significant immunostaining of malignant epithelial cells was observed in 4 of the 29 cases analysed (13.8%). No correlation between L-plastin expression and tumor size, histological grade or lymph node status was observed. In contrast, L-plastin was found expressed in 4 of the 11 estrogen and progesterone receptors negative tumors (p = 0.039). The potential role of plastin expression in the tumor process is discussed.

Illegitimate villin transcripts in normal bone marrow precludes detection of colon cancer micrometastases.

Villin is a specific marker for normal and tumoral colon tissue. We have developed a highly sensitive assay using reverse transcription (RT) and real-time PCR to detect villin transcripts. The sensitivity of detection is one colon cancer cell. However, high levels of illegitimate villin transcripts were observed in normal bone marrow, precluding the use of villin RT-PCR for routine detection of colon cancer cells in bone marrow of patients with colon cancer.

The Myb-p300-CREB axis modulates intestine homeostasis, radiosensitivity and tumorigenesis.

The gastrointestinal (GI) epithelium is constantly renewing, depending upon the intestinal stem cells (ISC) regulated by a spectrum of transcription factors (TFs), including Myb. We noted previously in mice with a p300 mutation (plt6) within the Myb-interaction-domain phenocopied Myb hypomorphic mutant mice with regard to thrombopoiesis, and here, changes in GI homeostasis. p300 is a transcriptional coactivator for many TFs, most prominently cyclic-AMP response element-binding protein (CREB), and also Myb. Studies have highlighted the importance of CREB in proliferation and radiosensitivity, but not in the GI. This prompted us to directly investigate the p300-Myb-CREB axis in the GI. Here, the role of CREB has been defined by generating GI-specific inducible creb knockout (KO) mice. KO mice show efficient and specific deletion of CREB, with no evident compensation by CREM and ATF1. Despite complete KO, only modest effects on proliferation, radiosensitivity and differentiation in the GI under homeostatic or stress conditions were evident, even though CREB target gene pcna (proliferating cell nuclear antigen) was downregulated. creb and p300 mutant lines show increased goblet cells, whereas a reduction in enteroendocrine cells was apparent only in the p300 line, further resembling the Myb hypomorphs. When propagated in vitro, crebKO ISC were defective in organoid formation, suggesting that the GI stroma compensates for CREB loss in vivo, unlike in MybKO studies. Thus, it appears that p300 regulates GI differentiation primarily through Myb, rather than CREB. Finally, active pCREB is elevated in colorectal cancer (CRC) cells and adenomas, and is required for the expression of drug transporter, MRP2, associated with resistance to Oxaliplatin as well as several chromatin cohesion protein that are relevant to CRC therapy. These data raise the prospect that CREB may have a role in GI malignancy as it does in other cancer types, but unlike Myb, is not critical for GI homeostasis.

ETV5 cooperates with LPP as a sensor of extracellular signals and promotes EMT in endometrial carcinomas.

Endometrial carcinoma (EC) is the most frequent among infiltrating tumors of the female genital tract, with myometrial invasion representing an increase in the rate of recurrences and a decrease in survival. We have previously described ETV5 transcription factor associated with myometrial infiltration in human ECs. In this work, we further investigated ETV5 orchestrating downstream effects to confer the tumor the invasive capabilities needed to disseminate in the early stages of EC dissemination. Molecular profiling evidenced ETV5 having a direct role on epithelial-to-mesenchymal transition (EMT). In particular, ETV5 modulated Zeb1 expression and E-Cadherin repression leading to a complete reorganization of cell-cell and cell-substrate contacts. ETV5-promoted EMT resulted in the acquisition of migratory and invasive capabilities in endometrial cell lines. Furthermore, we identified the lipoma-preferred partner protein as a regulatory partner of ETV5, acting as a sensor for extracellular signals promoting tumor invasion. All together, we propose ETV5-transcriptional regulation of the EMT process through a crosstalk with the tumor surrounding microenvironment, as a principal event initiating EC invasion.

The immune response to sporadic colorectal cancer in a novel mouse model.

Current mouse models do not reflect the sporadic nature of colon cancer and do not allow the analysis of antitumor immune response because of the lack of known tumor-specific antigens. Two transgenic mouse models with spontaneous tumor development were generated, directing the expression of SV40T antigen (Tag) either constitutively (Vil-Cre × LoxP-Tag-transgenic mice) or stochastically (Vil-Cre-ER(T2) × LoxP-Tag-transgenic mice) into the putative stem cell region of the crypt of Lieberkühn. Tumor development and antitumor immune response were monitored. Vil-Cre × LoxP-Tag mice developed multiple adenocarcinomas of the small intestine and colon at an average age of 6 months. During the tumor development, Tag-specific immunoglobulin G (IgG) antibodies were induced in half of the mice, although they had developed neonatal cytotoxic T lymphocyte (CTL) tolerance. This model shows similarity to hereditary colon cancer but not to the sporadic tumor development. Therefore, the conditional Vil-Cre-ER(T2) × LoxP-Tag mice were established, in which expression of the dormant Tag was induced by stochastic, tissue-specific activation of Cre recombinase. These mice spontaneously developed highly invasive, metastasizing colon carcinomas at an average age of 20 months. Colon carcinomas expressed epithelial and/or neuroendocrine markers depending on the grade of differentiation. Young Vil-Cre-ER(T2) × LoxP-Tag mice had retained CTL responses against epitope IV of Tag. The tumors induced strong anti-Tag IgG responses. We report, for the first time, a mouse model based on stochastic, tissue-specific activation of a dormant oncogene in the colon allowing the analysis of antitumor immune response against primary colorectal cancer.

[Mouse models of K-ras-initiated oncogenesis]. / Apport des modèles animaux pour l'étude du rôle de K-ras dans l'oncogenèse.

Activating mutations of the oncogene K-ras are found in one third of all human cancers. Much of our knowledge on K-ras signal transduction and its influence on tumor initiation and progression come from in vitro studies with cell lines. However, mouse models of human cancer allow a much more faithful recapitulation of the human disease, and the in vivo perspective is crucial for our understanding of neoplasia. In recent years, several new murine models for K-ras-induced tumorigenesis have been described. They allow new insights into the specific role that oncogenic K-ras proteins play in different solid tumors, and they permit the molecular dissection of the pathways that are initiated by somatic mutations in subsets of cells. Key advances have been made by the use of tissue-specific and inducible control of expression, which is achieved by the Cre/loxP technology or the tetracycline system. From these sophisticated models, a common picture emerges: the effects of K-ras on tumor initiation depend strongly on the cellular context, and different tissues vary in their susceptibility to K-ras transformation.

Cdx1, a dispensable homeobox gene for gut development with limited effect in intestinal cancer.

The homeobox gene Cdx1 is involved in anteroposterior patterning in embryos and its expression selectively persists in the intestinal epithelium throughout life. In human colon cancers, Cdx1 is overexpressed in few cases and lost in the majority of adenocarcinomas. We used mouse models of gain and loss-of-function to investigate the role of Cdx1 in intestinal development and cancers. Transgenic mice overexpressing Cdx1 and knockout mice exhibited a morphologically normal intestine. Cell proliferation, specification into the four differentiated lineages and migration along the crypt-villus axis were unchanged compared to wild-type mice. Changing Cdx1 caused an inverse and dose-dependent modification of the expression of the paralogous gene Cdx2, indicating that Cdx1 fine-tunes Cdx2 activity. Transgenenic and knockout mice failed to spontaneously develop tumours. Overexpressing Cdx1 was without incidence on the frequency of intestinal tumours induced chemically by azoxymethane treatment or genetically in Apc(Delta14/+) mice. However, it augmented the severity of the tumours in Apc(Delta14/+) mice. Inversely, the loss-of-function of Cdx1 in knockout mice was without incidence on the growth of tumours induced by azoxymethane. We conclude that Cdx1 is dispensable for intestinal development and that its overexpression could increase malignancy in early stages of tumourigenesis.

Structure of the human villin gene.

We have isolated and characterized the complete human villin gene. The villin gene is located on chromosome 2q35-36 in humans and on chromosome 1 in mice. Villin belongs to a family of calcium-regulated actin-binding proteins that share structural and functional homologies. The villin gene is expressed mainly in cells that develop a brush border, such as mucosal cells of the small and large intestine and epithelial cells of the kidney proximal tubules. Villin gene expression is strictly regulated during adult life and embryonic development in the digestive and urogenital tracts and, thus, may be used as a marker of the digestive and renal cell lineages. The human villin gene has one copy per haploid genome, encompasses about 25 kilobases, and contains 19 exons. Analysis of the structural organization of this gene shows that the two mRNAs that encode villin in humans arise by alternative choice of one of the two polyadenylylation signals located within the last exon. The overall organization of the exons reflects the gene duplication event from which this family of actin-binding proteins originated.

Epithelial cell growth and differentiation. IV. Controlled spatiotemporal expression of transgenes: new tools to study normal and pathological states.

The gut epithelium represents a dynamic, well-organized developmental system for examining self-renewal, differentiation, repair, and tumorigenesis. The apical pole of the enterocytes, the brush border, is composed of an array of well-organized actin microfilaments that support the plasma membrane. Villin, one actin-binding protein that contributes to the assembly and dynamics of the microvillus bundle, exhibits special features such as restricted tissue specificity and early expression in the immature crypt cells. The regulatory elements of the villin gene are suitable to control the expression of transgenes in intestinal cells. Engineering genetically modified animals by classic transgenesis using the villin promoter or by gene targeting in the villin locus will allow the establishment of animal models that may recapitulate human intestinal disorders.

Regulatory sequences on the human villin gene trigger the expression of a reporter gene in a differentiating HT29 intestinal cell line.

To develop a molecular tool for tissue-specific targeting of gene expression in immature and differentiated epithelial cells of the small and large intestinal mucosa, we have isolated the 2-kb 5'-flanking region of the human villin gene. This region contains numerous short sequences that are conserved among other tissue-specific promoters of genes expressed in differentiated enterocytes. This DNA fragment promotes the transcription and expression of the luciferase reporter gene in villin-positive intestinal, renal, and hepatoma cell lines but not in a villin-negative keratinocyte cell line. The pattern of expression corresponds that of the endogenous gene, indicating that this sequence can direct intestine-specific transcription. In the differentiating HT29 intestinal cell line, expression of the reporter gene is already detectable in undifferentiated cells, and dramatically increases when terminal differentiation is induced. Thus, as previously reported for the endogenous gene the isolated 5'-flanking region of the villin gene responds positively to conditions known to stimulate terminal differentiation of these cultured epithelial intestinal cells. The reported results indicate that this genomic fragment contains sufficient regulatory elements to recapitulate the expression pattern of the villin promoter during intestinal differentiation.