Intestinal insulin/IGF1 signalling through FoxO1 regulates epithelial integrity and susceptibility to colon cancer.

Obesity promotes the development of insulin resistance and increases the incidence of colitis-associated cancer (CAC), but whether a blunted insulin action specifically in intestinal epithelial cells (IECs) affects CAC is unknown. Here, we show that obesity impairs insulin sensitivity in IECs and that mice with IEC-specific inactivation of the insulin and IGF1 receptors exhibit enhanced CAC development as a consequence of impaired restoration of gut barrier function. Blunted insulin signalling retains the transcription factor FOXO1 in the nucleus to inhibit expression of Dsc3, thereby impairing desmosome formation and epithelial integrity. Both IEC-specific nuclear FoxO1ADA expression and IEC-specific Dsc3 inactivation recapitulate the impaired intestinal integrity and increased CAC burden. Spontaneous colonic tumour formation and compromised intestinal integrity are also observed upon IEC-specific coexpression of FoxO1ADA and a stable Myc variant, thus suggesting a molecular mechanism through which impaired insulin action and nuclear FOXO1 in IECs promotes CAC.

ZEB1-repressed microRNAs inhibit autocrine signaling that promotes vascular mimicry of breast cancer cells.

During normal tumor growth and in response to some therapies, tumor cells experience acute or chronic deprivation of nutrients and oxygen and induce tumor vascularization. While this occurs predominately through sprouting angiogenesis, tumor cells have also been shown to directly contribute to vessel formation through vascular mimicry (VM) and/or endothelial transdifferentiation. The extrinsic and intrinsic mechanisms underlying tumor cell adoption of endothelial phenotypes, however, are not well understood. Here we show that serum withdrawal induces mesenchymal breast cancer cells to undergo VM and that knockdown of the epithelial-to-mesenchymal transition (EMT) regulator, Zinc finger E-box binding homeobox 1 (ZEB1), or overexpression of the ZEB1-repressed microRNAs (miRNAs), miR-200c, miR-183, miR-96 and miR-182 inhibits this process. We find that secreted proteins Fibronectin 1 (FN1) and serine protease inhibitor (serpin) family E member 2 (SERPINE2) are essential for VM in this system. These secreted factors are upregulated in mesenchymal cells in response to serum withdrawal, and overexpression of VM-inhibiting miRNAs abrogates this upregulation. Intriguingly, the receptors for these secreted proteins, low-density lipoprotein receptor-related protein 1 (LRP1) and Integrin beta 1 (ITGB1), are also targets of the VM-inhibiting miRNAs, suggesting that autocrine signaling stimulating VM is regulated by ZEB1-repressed miRNA clusters. Together, these data provide mechanistic insight into the regulation of VM and suggest that miRNAs repressed during EMT, in addition to suppressing migratory and stem-like properties of tumor cells, also inhibit endothelial phenotypes of breast cancer cells adopted in response to a nutrient-deficient microenvironment.

PP2A-B56α controls oncogene-induced senescence in normal and tumor human melanocytic cells.

Oncoprotein C-MYC is overexpressed in human metastatic melanomas and melanoma-derived cells where it is required for the suppression of oncogene-induced senescence (OIS). The genetic events that maintain high levels of C-MYC in melanoma cells and their role in OIS are unknown. Here we report that C-MYC in cells from several randomly chosen melanoma lines was upregulated at the protein level, and largely because of the increased protein stability. Of all known regulators of C-MYC stability, levels of B56α subunit of the PP2A tumor suppressor complex were substantially suppressed in all human melanoma cells compared with normal melanocytes. Accordingly, immunohistochemical analysis revealed that the lowest and the highest amounts of PP2A-B56α were predominantly detected in metastatic melanoma tissues and in primary melanomas from patients with good clinical outcome, respectively. Importantly, PP2A-B56α overexpression suppressed C-MYC in melanoma cells and induced OIS, whereas depletion of PP2A-B56α in normal human melanocytes upregulated C-MYC protein levels and suppressed BRAF(V600E)- and, less efficiently, NRAS(Q61R)-induced senescence. Our data reveal a mechanism of C-MYC overexpression in melanoma cells and identify a functional role for PP2A-B56α in OIS of melanocytic cells.

Negative regulation of Pim-1 protein kinase levels by the B56beta subunit of PP2A.

The Pim protein kinases are serine threonine protein kinases that regulate important cellular signaling pathway molecules, and enhance the ability of c-Myc to induce lymphomas. We demonstrate that a cascade of events controls the cellular levels of Pim. We find that overexpression of the protein phosphatase (PP) 2A catalytic subunit decreases the activity and protein levels of Pim-1. This effect is reversed by the application of okadaic acid, an inhibitor of PP2A, and is blocked by SV40 small T antigen that is known to disrupt B subunit binding to PP2A A and C subunits. Pim-1 can coimmunoprecipitate with the PP2A regulatory B subunit, B56beta, but not B56alpha, gamma, delta, epsilon or B55alpha. Using short hairpin RNA targeted at B56beta, we demonstrate that decreasing the level of B56beta increases the half-life of Pim-1 from 0.7 to 2.8 h, and decreases the ubiquitinylation level of Pim-1. We also find that Pin1, a prolyl-isomerase, is capable of binding Pim-1 and leads to a decrease in the protein level of Pim-1. On the basis of these observations, we hypothesize that phosphorylated Pim-1 binds Pin1 allowing the interaction of PP2A through B56beta. Dephosphorylation of Pim-1 then allows for ubiquitinylation and protein degradation of Pim-1.

Aberrant stabilization of c-Myc protein in some lymphoblastic leukemias.

Overexpression of the c-Myc oncoprotein is observed in a large number of hematopoietic malignancies, and transgenic animal models have revealed a potent role for c-Myc in the generation of leukemias and lymphomas. However, the reason for high c-Myc protein levels in most cases is unknown. We examined whether aberrant protein stabilization could be a mechanism of c-Myc overexpression in leukemia cell lines and in primary bone marrow samples from pediatric acute lymphoblastic leukemia (ALL) patients. We found that c-Myc protein half-life was prolonged in the majority of leukemia cell lines and bone marrow samples tested. There were no mutations in the c-myc gene in any of the leukemia cell lines that could account for increased c-Myc stability. However, abnormal phosphorylation at two conserved sites, Threonine 58 and Serine 62, was observed in leukemia cell lines with stabilized c-Myc. Moreover, stabilized c-Myc from the ALL cell lines showed decreased affinity for glycogen synthase kinase3beta, the kinase that phosphorylates c-Myc at Threonine 58 and facilitates its degradation. These findings reveal that deregulation of the c-Myc degradation pathway controlled by Serine 62 and Threonine 58 phosphorylation is a novel mechanism for increased expression of a potent oncoprotein known to be involved in hematopoietic malignancies.

The alternatively initiated c-Myc proteins differentially regulate transcription through a noncanonical DNA-binding site.

The myc proto-oncogene family has been implicated in multiple cellular processes, including proliferation, differentiation, and apoptosis. The Myc proteins, as heterodimers with Max protein, have been shown to function as activators of transcription through an E-box DNA-binding element, CACGTG. We have now found that the c-Myc proteins regulate transcription through another, noncanonical, DNA sequence. The non-AUG-initiated form of the c-Myc protein, c-Myc 1, strongly and specifically activates transcription of the C/EBP sequences within the EFII enhancer element of the Rous sarcoma virus long terminal repeat. In contrast, comparable amounts of the AUG-initiated form, c-Myc 2, fail to significantly affect enhancer activity. However, both c-Myc proteins trans-activate the CACGTG sequence comparably. In addition, Myc/Max heterodimers, but not Max homodimers, bind to the EFII enhancer sequence in vitro. Finally, c-Myc 1 overexpression, but not c-Myc 2 overexpression, significantly inhibits cell growth. These results reveal new transcriptional activities for the Myc proteins and demonstrate that the different forms of the Myc protein are functionally distinct. These results also suggest an interplay between two different growth regulatory transcription factor families.

Multiple forms of C/EBP beta bind the EFII enhancer sequence in the Rous sarcoma virus long terminal repeat.

In this report we demonstrate that C/EBP beta is a major component of three EFII DNA binding complexes, EFIIa, EFIIb, and EFIIc, which we have previously shown to specifically recognize a C/EBP consensus binding site found in the EFII enhancer sequence from the Rous sarcoma virus long terminal repeat (R. C. Sears and L. Sealy, J. Virol. 66:6338-6352, 1992). Three different forms of C/EBP beta, p42, p35, and p20, can bind the EFII DNA sequence as homodimers, and dimerization experiments show that EFIIa is a homodimer of p20 C/EBP beta, EFIIb is primarily composed of a p20/p35 heterodimer with minor amounts of p20/p42 heterodimer and p35 homodimer, and EFIIc is composed of p20 and/or p35 heterodimerized with a novel 60-kDa protein. p20 C/EBP beta is likely equivalent to the internally initiated translation product of C/EBP beta, LIP (liver inhibitor protein), described by P. Descombes and U. Schibler (Cell 67:569-579, 1991). In contrast to the low level of LIP expressed in liver, postulated to occur because of leaky ribosome scanning, we found high levels of expression of p20 C/EBP beta in fibroblasts and lymphocytes. In murine fibroblasts, p20 C/EBP beta has an extended half-life, four times longer than those of p42 and p35 C/EBP beta, which could contribute to its abundant accumulation in this cell type, even though its synthesis by leaky ribosome scanning might be inefficient. Interestingly, overexpression of either the long or short form of C/EBP beta represses EFII-mediated transcription, suggesting that another unidentified EFII transactivator(s) exists, which may be dominantly inhibited by C/EBP beta proteins, and/or that transactivation by C/EBP beta proteins requires posttranslational modifications that were lacking in the transient overexpression experiments.

Characterization of nuclear proteins that bind the EFII enhancer sequence in the Rous sarcoma virus long terminal repeat.

The EFII cis element is a 38-bp sequence at the 5' end of the Rous sarcoma virus long terminal repeat, extending from nucleotides -229 to -192 (with respect to the viral transcription start site), which is recognized by sequence-specific DNA-binding proteins in avian fibroblast nuclear extracts (L. Sealy and R. Chalkley, Mol. Cell. Biol. 7:787-798, 1987). We demonstrate that multiple copies of the EFII cis element strongly activate transcription of a reporter gene in vivo. We correlate the region of the EFII cis element which activates transcription in vivo with the in vitro binding site for three nuclear factors, EFIIa, EFIIb, and EFIIc. The sequence motif recognized by EFIIa, -b, and -c is also found in consensus binding sites for members of a rapidly growing family of transcription factors related to the CCAAT/enhancer-binding protein (C/EBP). EFIIa, -b, and -c are present in fibroblast and epithelial cell lines from various species but are much less abundant in differentiated rat liver and kidney cells. The EFIIa binding activity is particularly abundant in an avian B-cell lymphoma line. As judged from molecular weight analysis, cell type distribution, and sequence recognition properties, the EFII factors under study appear to differ from most of the previously described C/EBP-related factors and thus may expand the diversity of the C/EBP family.

Axial diffusion of retinol in isolated frog rod outer segments following substantial bleaches of visual pigment.

Partial bleaches of rhodopsin were made in either the proximal or distal halves of isolated Rana pipiens rod outer segments. The fluorescence of all-trans retinol was recorded 15, 60 and 120 min following 17% and 63% bleaches. Some of the retinol that formed remained immobilized in the bleached halves of the outer segments, while another portion was slowly mobilized and diffused along the cell axis.

Lengths of immunolabeled ciliary microtubules in frog photoreceptor outer segments.

Monoclonal anti-tubulin antibodies were used to label microtubules in the connecting cilia and outer segments of retinal photoreceptors isolated from Rana pipiens. In paraformaldehyde-fixed rods from frogs maintained on diurnal light cycles, the anti-tubulin labeling of ciliary microtubules (mean length = 27 micron) typically extends to slightly over half the length of the outer segments (mean length = 46 micron). Rod outer segments from frogs kept in constant darkness for 3-4 weeks are longer (mean length = 53 micron) than rod outer segments from frogs maintained in cyclic lighting. However, the distribution of fractional lengths of anti-tubulin labeling of ciliary microtubules is the same for both lighting regimens. Incubating retinas in 1.0 mM colchicine prior to outer-segment fixation has no effect on the length of immunolabeling of ciliary microtubules, suggesting that post-mortem elongation artifacts are not significant. Incubating retinas in 10 microM taxol prior to fixation significantly increases the length of stained ciliary microtubules, suggesting that taxol either promotes post-mortem assembly of microtubules, or that taxol reduces post-mortem disassembly. The mean position of the end of anti-tubulin-labeled ciliary microtubules does not correspond to the position of disk shedding for any of the experimental conditions employed.