Polyoma Middle T Research Articles

The application of a murine bone bioreactor as a model of tumor: bone interaction

A limited number of in vivo models that rapidly assess bone development or allow for the study of tumor progression in a closed in vivo environment exist. To address this, we have used bone tissue engineering techniques to generate a murine in vivo bone bioreactor. The bioreactor was created by implanting an osteoconductive hydroxyapatite scaffold pre-loaded with saline as a control or with bone morphogenetic protein-2 (BMP-2) to the murine femoral artery. Control and BMP-2 bioreactors were harvested and histologically assessed for vascularization and bone formation at 6 and 12 weeks post implantation. BMP-2 significantly enhanced the formation of osteoid within the bioreactor in comparison to the controls. To test the in vivo bone bioreactor as a model of tumor: bone interaction, FVB mice were implanted with control or BMP-2 treated bioreactors. After 6 weeks, an osteolytic inducing mammary tumor cell line tagged with luciferase (PyMT-Luc) derived from the polyoma virus middle T (PyMT) model of mammary tumorigenesis was delivered to the bioreactor via the femoral artery. Analysis of luciferase expression over time demonstrated that the presence of osteoid in the BMP-2 treated bioreactors significantly enhanced the growth rate of the PyMT-Luc cells in comparison to the control group. These data present a unique in vivo model of ectopic bone formation that can be manipulated to address molecular questions that pertain to bone development and tumor progression in a bone environment.

MEKK1 controls matrix degradation and tumor cell dissemination during metastasis of polyoma middle-T driven mammary cancer

Mammary tumor cells are required to degrade the surrounding matrix and disseminate in order to metastasize, and both of these processes are controlled by a tumor cell-signaling network that remains poorly defined. MEKK1 is a MAPKKK that regulates both the extracellular signal regulated kinase (ERK1/2) and the c-Jun amino terminal kinase (JNK) signaling pathways. MEKK1 signaling regulates migration through control of cell adhesion and is required for inducible expression of urokinase-type plasminogen activator (uPA). MEKK1-deficient mice with mammary gland-targeted expression of the polyoma middle T antigen (PyMT) transgene develop primary mammary tumors at a rate and frequency similar to wild-type littermates, indicating that MEKK1 deficiency does not affect PyMT-mediated transformation. However, MEKK1-/- mice display significantly delayed tumor cell dissemination and lung metastasis. Delayed MEKK1-dependent tumor dissemination is associated with markedly reduced tumor uPA expression, gelatinase activity, and prolonged tumor basement membrane integrity. siRNA-mediated MEKK1 knockdown inhibits uPA activity, cell migration and invasion in MDA-MB-231 human breast cancer cells. Thus MEKK1 controls tumor progression by regulating both the migration and proteolysis aspects of tumor cell invasiveness. To our knowledge, this is the first example of a MAPKKK that regulates metastasis through control of tumor invasiveness.

Oncogenic function for the Dlg1 mammalian homolog of the Drosophila discs-large tumor suppressor

The fact that several different human virus oncoproteins, including adenovirus type 9 E4-ORF1, evolved to target the Dlg1 mammalian homolog of the membrane-associated Drosophila discs-large tumor suppressor has implicated this cellular factor in human cancer. Despite a general belief that such interactions function solely to inactivate this suspected human tumor suppressor protein, we demonstrate here that E4-ORF1 specifically requires endogenous Dlg1 to provoke oncogenic activation of phosphatidylinositol 3-kinase (PI3K) in cells. Based on our results, we propose a model wherein E4-ORF1 binding to Dlg1 triggers the resulting complex to translocate to the plasma membrane and, at this site, to promote Ras-mediated PI3K activation. These findings establish the first known function for Dlg1 in virus-mediated cellular transformation and also surprisingly expose a previously unrecognized oncogenic activity encoded by this suspected cellular tumor suppressor gene.

Extracellular protease mRNAs are predominantly expressed in the stromal areas of microdissected mouse breast carcinomas

Solid tumors synthesize a number of extracellular matrix-degrading proteases that are important for tumor progression. Based on qualitative in situ hybridization studies in human cancer tissue, a range of components involved in proteolysis appear to be expressed by stromal cells rather than cancer cells. We have now used laser capture microdissection and real-time PCR to quantify the mRNA expression of components of matrix-degrading proteolytic systems in cancer and stromal areas of mouse mammary tumors genetically induced by the polyoma virus middle T (PyMT) antigen. We examined the mRNA levels of urokinase plasminogen activator, plasminogen activator inhibitor 1 and the matrix metalloproteases MMP-2, -3, -11, -13 and -14, and found that all these seven genes are predominantly expressed by stromal cells. Our results were qualitatively supported by in situ hybridization analysis of the expression of mRNAs for MMP-2, -3 and -13 in the PyMT tumors. Statistical analyses indicated that the quantitative expression patterns observed in cancer and stromal cells isolated from individual tumors from different PyMT mice are quite reproducible. The methodology described in this study provides excellent tools to study the possible interactions between cancer and stromal cells during the development of breast cancer, and the results suggest that stromal cells are involved in carcinogenesis and tumor progression, which may have important implications for the biology and therapy of cancer.

Haploid inactivation of the amplified-in-breast cancer 3 coactivator reduces the inhibitory effect of peroxisome proliferator-activated receptor gamma and retinoid X receptor on cell proliferation and accelerates polyoma middle-T antigen-induced mammary tumorigenesis in mice.

The amplified-in-breast cancer 3 (AIB3) is a nuclear receptor coactivator amplified and overexpressed in human breast cancers. AIB3(-/-) mice die during gestation, whereas AIB3(+/-) mice exhibit normal development. Here, we demonstrate that AIB3 protein is mainly located in the nuclei of mammary epithelial cells and tumor cells and its levels are elevated in mammary epithelial cells at middle pregnant stage and in mammary tumor cells. To examine whether AIB3 reduction affects mammary tumorigenesis, we generated wild-type mouse mammary tumor virus/polyoma middle-T (WT/PyMT) and AIB3(+/-)/PyMT mice. Mammary tumor development in AIB3(+/-)/PyMT female and male mice was substantially accelerated compared with that in WT/PyMT mice, because of increased cell proliferation in early tumorigenic lesions, including ductal hyperplasia and mammary intraepithelial neoplasia. Tumor formation in nude mice that received premalignant AIB3(+/-)/PyMT mammary tissue was much faster than in nude mice that received transplants of premalignant WT/PyMT mammary tissue, which indicated that the accelerated tumorigenesis in AIB3(+/-)/PyMT mammary glands is due to a mammary epithelial autonomous defect. Expression of PyMT, estrogen receptor alpha and estrogen receptor alpha-regulated genes was unaffected in AIB3(+/-)/PyMT mammary glands, which suggests that the acceleration of mammary tumor formation in AIB3(+/-)/PyMT mice was not a consequence of changes in PyMT expression or in estrogen receptor function. Importantly, the inhibitory effects of peroxisome proliferator-activated receptor gamma (PPARgamma) and retinoid-X receptor (RXR) ligands on AIB3(+/-)/PyMT cell proliferation and the transcriptional function of PPARgamma in AIB3(+/-)/PyMT cells were reduced. Thus, AIB3 haplodeficiency may facilitate PyMT-induced tumorigenesis through a partial impairment of PPARgamma and RXR function. These results suggest that AIB3 may be a tumor suppressor that is required for the inhibition of cell proliferation by PPARgamma and RXR.

Polyomavirus middle T–induced mammary intraepithelial neoplasia outgrowths: Single origin, divergent evolution, and multiple outcomes

The development of models to investigate the pathobiology of premalignant breast lesions is a critical prerequisite for development of breast cancer prevention and early intervention strategies. Using tissue transplantation techniques, we modified the widely used polyomavirus middle T (PyV-mT) transgenic mouse model of breast cancer to study the premalignant stages of tumorigenesis. Premalignant atypical lesions were isolated from PyV-mT transgenic mice and used to generate two sets of three mammary intraepithelial neoplasia (MIN) outgrowth lines. Investigation of these six unique lines, each of which fulfills the criteria for MIN, has provided new information regarding the biology of PyV-mT-induced neoplasia. Although expression of the PyV-mT transgene was the primary initiating event for all lines, they exhibited different tumor latencies, metastatic potentials, and morphologies. Six distinguishable morphologic patterns of differentiation were identified within the premalignant outgrowths that are likely to represent several tumorigenic pathways. Further, several tumor phenotypes developed from each line and the tumors developing from the six lines had different metastatic potentials. These observations are consistent with the hypothesis that distinct pathways of PyV-mT-initiated neoplastic progression lead to different outcomes with respect to latency and metastasis. The MIN outgrowth lines share several characteristics with precursors of human breast cancer including the observation that gene expression profiles of tumors are more similar to those of the MIN outgrowth line outgrowth from which they developed than to other tumors. These lines provide an opportunity to study the full range of events occurring secondary to PyV-mT expression in the mammary gland.

Immunotherapy of spontaneous mammary carcinoma with fusions of dendritic cells and mucin 1-positive carcinoma cells.

The tumour-associated antigen mucin 1 (MUC1) is a multifunctional protein involved in protection of mucous membranes, signal transduction, and modulation of the immune system. More than 70% of cancers overexpress MUC1, making MUC1 a potential target for immunotherapy. In the present study, MUC1 transgenic mice were crossed with syngeneic strains that express the polyomavirus middle-T oncogene (PyMT) driven by the mouse mammary tumour virus promoter long-terminal repeat (MMTV-LTR). The resultant breed (MMT mice) developed spontaneous MUC1-expressing mammary carcinomas with 100% penetrance at 8-15 weeks of age. As found in human breast cancer, the mammary carcinoma in MMT mice arose in multiple stages. Immunization with fusions of dendritic cells and MUC1-positive tumour cells (FC/MUC1) induced MUC1-specific immune responses that blocked or delayed the development of spontaneous breast carcinomas. In contrast, there was no delay of tumour development in MMT mice immunized with irradiated MC38/MUC1 tumour cells. The efficacy of fusion cells was closely correlated with the timing of initial immunization. Immunization with FC/MUC1 initiated in MMT mice at < 1, 1-2 and 2-3 months of age rendered 33, 5 and 0% of mice free of tumour, respectively, up to 6 months. Whereas mice immunized in the later stage of tumour development succumbed to their disease, immunization resulted in control of tumour progression and prolongation of life. These results indicate that immunization with FC/MUC1 can generate an anti-MUC1 response that is sufficient to delay the development of spontaneous mammary carcinomas and control tumour progression in MMT mice.

PyV-mT-induced parotid gland hyperplasia as detected by altered lectin reactivity is not modulated by inducible nitric oxide deficiency

The parotid gland was one of the first organs recognised to be sensitive to the transforming effects of polyomavirus. This study examines parotid gland pathology in mice expressing the polyomavirus middle T (PyV-mT) under the control of the mouse mammary tumour virus long terminal repeat (MMTV-LTR) to (1) demonstrate the utility of this model for studying premalignant disease; (2) identify early lesions by lectin staining and (3) determine effects of the inducible nitric oxide synthase (iNOS) in modulating tumorigenesis. The middle T oncogene is expressed in the parotid glands in addition to the mammary glands and results in the formation of parotid hyperplasias in 100% of transgenic female mice. These hyperplasias have the features of intraepithelial neoplasia including hypertrophic cells with prominent nucleoli and abnormal mitoses. Focal areas of parotid hyperplasia can be identified using peanut agglutinin (PNA), a lectin that recognises the tumour associated T antigen. In contrast to normal parotid gland, areas of hyperplasia do not bind PNA. Mice deficient in iNOS, an enzyme implicated in the promotion of tumorigenesis, were bred into the PyV-mT model. The loss of iNOS did not impact on the number or size of parotid gland hyperplasias, suggesting that NO produced by this enzyme is not a key regulator of PyV-mT-induced parotid gland hyperplasia. The consistent development of parotid hyperplasias in PyV-mT mice and clear identification of these lesions by loss of PNA lectin reactivity provides a useful model for studying early molecular changes in parotid tumorigenesis.

A bioinformatics-based strategy identifies c-Myc and Cdc25A as candidates for the Apmt mammary tumor latency modifiers.

The epistatically interacting modifier loci (Apmt1 and Apmt2) accelerate the polyoma Middle-T (PyVT)-induced mammary tumor. To identify potential candidate genes loci, a combined bioinformatics and genomics strategy was used. On the basis of the assumption that the loci were functioning in the same or intersecting pathways, a search of the literature databases was performed to identify molecular pathways containing genes from both candidate intervals. Among the genes identified by this method were the cell cycle-associated genes Cdc25A and c-Myc, both of which have been implicated in breast cancer. Genomic sequencing revealed noncoding polymorphism in both genes, in the promoter region of Cdc25A, and in the 3' UTR of c-Myc. Molecular and in vitro analysis showed that the polymorphisms were functionally significant. In vivo analysis was performed by generating compound PyVT/Myc double-transgenic animals to mimic the hypothetical model, and was found to recapitulate the age-of-onset phenotype. These data suggest that c-Myc and Cdc25A are Apmt1 and Apmt2, and suggest that, at least in certain instances, bioinformatics can be utilized to bypass congenic construction and subsequent mapping in conventional QTL studies.

Rac is Essential in the Transformation of Endothelial Cells by Polyoma Middle T

Expression of the Polyoma Middle T (PyMT) antigen in endothelial cells results in single-step transformation to hemangioma producing malignant cells. To study the mechanism of PyMT transformation, we used the PyMT induced mouse brain endothelial cell line, bEND.3, expressing constitutively active and dominant negative mutants of the small GTPase Rac. The bEND.3 cell phenotype of tumorigenesis, loss of normal growth control and formation of cysts rather than capillary tubes in fibrin gels was reversed by expression of dominant negative Rac. The mechanism of N17 Rac action in blocking the endothelial cell transformant, PyMT, did not involve effects of Rac on the actin cytoskeleton since this component of the bEND.3 cell phenotype was not affected. Furthermore, the PyMT induced activation of the plasminogen activator (PA)/plasmin system was not affected by Rac inhibition. Inhibition of the downstream effectors of Rac, phosphatidylinositol 3-kinase (PI3-K) and p70S6k, which are known to be constitutively activated by PyMT transformation, inhibited bEND.3 cell proliferation and cyst formation in fibrin gels even in cells expressing V12 constitutively active Rac, but they did not restore capillary tube formation. These results demonstrate that middle T antigen induced endothelial cell transformation requires signal transduction by Rac. The downstream Rac effectors, P13-K and p70S6k, mediate PyMT/Rac effects on cell proliferation and cyst formation, but other unknown effectors of PyMT are required for the cytoskeletal changes and activation of the PA/plasmin system.

Murine dendritic cells transduced with an adenoviral vector expressing a defined tumor antigen can overcome anti-adenovirus neutralizing immunity and induce effective tumor regression.

In this study transduced dendritic cells (DCs) were used to enhance immunogenecity of a specific tumor antigen. Using a polyoma middle T (PyMT) transgenic mammary carcinoma model we found that injections of DCs transduced with an adenoviral (Ad) vector expressing PyMT (DCAd-PymT) led to potent specific anti-tumor immunity. Efficacy was not affected by neutralizing Abs (high or low titers) and naive animals did not produce detectable anti-Ad Abs following two injections of transduced DCs. Repeated injections of transduced DCs significantly improved therapeutic efficacy in mice with established lung metastases. These data emphasize the ability of Ad-infected DCs to: i) minimize anti-Ad Ab production, ii) overcome pre-existing anti-Ad humoral immunity, and iii) improve vaccination efficacy when injected more than once.

P53 Is Not Inactivated in B6C3F1 Mouse Vascular Tumors Arising Spontaneously or Associated with Long-Term Administration of the Thiazolidinedione Troglitazone

Hemangiomas and hemangiosarcomas are uncommon in rodents and humans and, as such, the mechanisms giving rise to these tumors are poorly understood. Inactivating mutations in the p53 gene have been detected in sporadic and chemically induced human and rodent hemangiosarcomas. Additionally, experimental ablation of p53 function in mice by targeted gene disruption increases the incidence of both spontaneous and carcinogen-induced vascular tumors. These findings implicate p53 disruption in vascular tumor development. In this study, we characterized p53 inactivation immunocytochemically and by gene sequencing in a large number of vascular tumors that developed in B6C3F1 mice during a long-term (2-year) study of the thiazolidinedione troglitazone. For comparative purposes, a murine hemangiosarcoma induced by polyoma middle-T antigen, which transforms endothelial cells via a p53-independent mechanism, five spontaneous human hemangiosarcoma specimens, and species-specific positive control tissues were also evaluated by immunocytochemistry for p53 inactivation. While 20% of the human hemangiosarcomas and all positive control tissues expressed significant levels of nuclear p53, indicating functional inactivation of the protein, none of the 161 mouse vascular tumors studied expressed detectable p53 protein. The absence of inactivating mutations was confirmed in eight of the histologically most malignant mouse hemangiosarcomas by sequencing exons 5 to 8 of the p53 gene. These results demonstrate that p53 inactivation did not play a role in development of the vascular tumors seen in the long-term study of troglitazone, and they indicate that loss of p53 function is not essential for vascular tumor development in mice.

Adenoviral Vectors Expressing Lymphotactin and Interleukin 2 or Lymphotactin and Interleukin 12 Synergize to Facilitate Tumor Regression in Murine Breast Cancer Models

We have previously demonstrated that intratumoral injection with Ad vectors expressing IL-2 or IL-12 can induce regression in a murine breast cancer model. These IL-2- or IL-12-induced antitumor responses were mainly mediated by Ag-specific T cells. Lymphotactin is a novel lymphocyte chemokine that can cause local accumulation of CD4+, CD8+, and NK cells. We hypothesized that addition of lymphotactin may enhance the antitumor immune responses induced by locally produced IL-2 and IL-12 as we have previously shown. To this end we constructed two double-recombinant adenoviral vectors expressing lymphotactin along with either IL-2 (Ad5 Lym/IL-2) or IL-12 (Ad5 Lym/IL-12). Subcutaneous injection of polyoma middle T (PyMT) or Neu (8142) transgenically derived breast adenocarcinoma cells, in the hind flank of FVB/n mice, results in the formation of tumor nodules in 14-21 days. We show that these constructs elicit potent antitumor responses when administered intratumorally. The antitumor responses are long lasting as determined by rechallenge experiments and hence demonstrate a protective immunity. These observations indicate that by augmenting the antitumor response with adenoviral vectors expressing lymphotactin in combination with IL-2 or IL-12 is a novel way to enhance immunotherapeutic approaches.

Oncogene-induced up-regulation of Caco-2 cell proliferation involves IGF-II gene activation through a protein kinase C-mediated pathway.

We previously reported that ras and polyoma middle T (PyMT), a constitutive activator of the src protooncogene product, up-regulated Caco-2 cell proliferation along with protein kinase C (PKC) alpha expression and PKC activity. We aimed to investigate whether oncogene-induced up-regulation of Caco-2 cell proliferation involved stimulation of the autocrine IGF-II/IGF-I receptor (IGFIR) loop described in these cells and if so, to analyse the role of overexpressed and activated PKC. Compared with control vector transfected Caco-2 cells, ras- and PyMT-transfected cells exhibited increased expression of the 6.0 and 4.8 kb IGF-II transcripts. This was due to increased activity of the P3 and P4 promoters of the IGF-II gene which correlated with increased expression and DNA-binding activity of Sp1, a transcription factor interacting with several specific sites in P3 and P4 promoters. Oncogene-transfected cells displayed enhanced autocrine IGF-II production, which was fully responsible for the oncogene-induced increase in their proliferation since this increase was blunted by anti-human IGF-II and IGF1R (alphaIR3) antibodies. PKC mediated oncogene activation of the IGF-II gene presumably through action on Sp1 since (i) PKC activation by phorbol 12-myristate 13-acetate increased Sp1 expression, P3 and P4 activity and IGF-II mRNA in control but not in oncogene-transfected cells; and (ii) PKC inhibition by the PKC inhibitor Gö6976 reduced Sp1, P3 and P4 activity and IGF-II mRNA in all three cell lines. This is the first evidence that ras- and PyMT/src oncogenes up-regulate Caco-2 cell proliferation through a PKC-mediated pathway which stimulates IGF-II gene transcription and thereby increases autocrine IGF-II production. The mechanisms underlying IGF-II gene activation by PKC most probably involve action on Sp1.

At the onset of transformation polyomavirus middle-T recruits shc and src to a perinuclear compartment coincident with condensation of endosomes.

To examine the early cellular changes that accompany transformation, middle-T antigen (the transforming protein of polyomavirus), was expressed from an inducible promoter in Rat2 fibroblasts and in normal diploid human fibroblast (MRC5) cells using a replication defective adenovirus vector. The earliest detectable expression of middle-T antigen correlated closely with the initial changes in cellular morphology. At this time, expression of middle-T antigen resulted in the redistribution of shc and src to a brefeldin A resistant perinuclear compartment coincident with the formation of kinase active complexes containing middle-T antigen, shc and src. The first detectable changes directly related to altered morphology was reorganization of actin filaments, and the condensation of tubular endosomes in the perinuclear region of the cell. These results suggest a potential role for perinuclear localized mT molecules in some of the morphologic changes associated with the initial phase of cellular transformation.

Protein kinase B/Akt is activated by polyomavirus middle-T antigen via a phosphatidylinositol 3-kinase-dependent mechanism.

The middle tumor antigen (middle-T) of mouse polyomavirus is responsible for the transforming potential of this virus. Middle-T has been shown to interact with a variety of cellular proteins known to mediate mitogenic signaling, like phosphatase-2A, Src family kinases, phosphatidylinositol 3-kinase (PI 3-kinase), the adapter protein SHC, phospholipase Cgamma-1 and 14-3-3 family proteins. Association with SHC and PI 3-kinase, respectively, stimulates two independent signaling pathways that are indispensible for viral oncogenicity. SHC activates the Ras/MAPK pathway via Grb2/SOS resulting in changes in early gene expression. The downstream targets of PI 3-kinase are less well studied but seem to impinge on serum response factor (SRF) which is also involved in regulating early gene expression. Recently, the protein kinase B/Akt (PKB/Akt) has been identified as a target of PI 3-kinase in receptor tyrosine kinase signaling. Here we show that PKB/Akt is a target of wild type middle-T, but not of mutants unable to activate PI 3-kinase. These data were confirmed using inhibitors of PI 3-kinase as well as dominant-negative alleles of the catalytic subunit of this lipid kinase. In addition, mutants of PKB/Akt lacking a pleckstrin homology domain and therefore unable to bind to D3 phospatidylinositides were not activated by middle-T. Taken together these data suggest that middle-T activates PKB/Akt in a PI 3-kinase-dependent manner. Furthermore, direct association with D3 phosphatidylinositides seems to be essential for activation of PKB/Akt.

The Structural Basis for 14-3-3:Phosphopeptide Binding Specificity

The 14-3-3 family of proteins mediates signal transduction by binding to phosphoserine-containing proteins. Using phosphoserine-oriented peptide libraries to probe all mammalian and yeast 14-3-3s, we identified two different binding motifs, RSXpSXP and RXY/FXpSXP, present in nearly all known 14-3-3 binding proteins. The crystal structure of 14-3-3ζ complexed with the phosphoserine motif in polyoma middle-T was determined to 2.6 Å resolution. The bound peptide is in an extended conformation, with a tight turn created by the pS +2 Pro in a cis conformation. Sites of peptide–protein interaction in the complex rationalize the peptide library results. Finally, we show that the 14-3-3 dimer binds tightly to single molecules containing tandem repeats of phosphoserine motifs, implicating bidentate association as a signaling mechanism with molecules such as Raf, BAD, and Cbl.

Differential expression of the neu transgene in murine mammary tissues

The mammary glands of control FVB and mice with MTV-LTR promoted transgenes were stained using immunohistochemistry to detect neu expression. Neu expression in the terminal end buds of developing mammary glands and during early pregnancy in FVB mice was confirmed by in situ hybridization. Neu was expressed in all tumors from mice with the neu transgene but not in tumors expressing transforming growth factor alpha (TGF alpha) or polyoma virus middle T (PyV-MT). Neu was also expressed sporadically in non-neoplastic mammary cells of transgenic neu mice. However, most mammary cells expressing neu were dysplastic. The differential expression of the neu transgene has important implications for the interpretation of transgenic biology.

Microcirculation and metastasis in a new mouse mammary tumor model system

Two new metastatic mouse mammary tumor transplant lines have been established in nude mice. The Met-1 line, with the polyoma virus middle T (PyV-MT) transgene, metastasized with 100% efficiency. The Db-7 line, expressing a PyV-MT transgene mutated at positions 315 and 322, metastasized with 8.8% efficiency. Histology and computer-assisted intravital microscopy demonstrated that internal microcirculation in Met-1 was more complex than Db-7; Met-1 exhibited higher microvessel density and tortuosity (P < 0.0001). These indices of microvascular complexity correlated with the higher Met-1 metastatic rate (P < 0.0001). These two transplantable lines will be useful for investigating the complex relationship between angiogenesis and metastasis.

Evidence for multiple mechanisms for membrane binding and integration via carboxyl-terminal insertion sequences.

Subcellular localization of proteins with carboxyl-terminal insertion sequences requires the molecule be both targeted to and integrated into the correct membrane. The mechanism of membrane integration of cytochrome b5 has been shown to be promiscuous, spontaneous, nonsaturable, and independent of membrane proteins. Thus endoplasmic reticulum localization for cytochrome b5 depends primarily on accurate targeting to the appropriate membrane. Here direct comparison of this mechanism with that of three other proteins integrated into membranes via carboxyl-terminal insertion sequences [vesicle-associated membrane protein 1(Vamp1), polyomavirus middle-T antigen, and Bcl-2] revealed that, unlike cytochrome b5, membrane selectivity for these molecules is conferred at least in part by the mechanisms of membrane integration. Bcl-2 membrane integration was similar to that of cytochrome b5 except that insertion into lipid vesicles was inefficient. Unlike cytochrome b5 and Bcl-2, Vamp1 binding to canine pancreatic microsomes was saturable, ATP-dependent, and abolished by mild trypsin treatment of microsomes. Surprisingly, although the insertion sequence of polyomavirus middle-T antigen was sufficient to mediate electrostatic binding to membranes, binding did not lead to integration into the bilayer. Together these results demonstrate that there are at least two different mechanisms for correct membrane integration of proteins with insertion sequences, one mediated primarily by targeting and one relying on factors in the target membrane to mediate selective integration. Our results also demonstrate that, contrary to expectation, hydrophobicity is not sufficient for insertion sequence-mediated membrane integration. We suggest that the structure of the insertion sequence determines whether or not specific membrane-bound receptor proteins are required for membrane integration.