Insight into the Tumor Suppression Mechanism from the Structure of Human Polypyrimidine Splicing Factor (PSF/SFPQ) Complexed with a 30mer RNA from Murine Virus-like 30S Transcript-1.

Human polypyrimidine-binding splicing factor (PSF/SFPQ) is a tumor suppressor protein that regulates the gene expression of several proto-oncogenes and binds to the 5'-polyuridine negative-sense template (5'-PUN) of some RNA viruses. The activity of PSF is negatively regulated by long-noncoding RNAs, human metastasis associated in lung adenocarcinoma transcript-1 and murine virus-like 30S transcript-1 (VL30-1). PSF is a 707-amino acid protein that has a DNA-binding domain and two RNA recognition motifs (RRMs). Although the structure of the apo-truncated PSF is known, how PSF recognizes RNA remains elusive. Here, we report the 2.8 Å and 3.5 Å resolution crystal structures of a biologically active truncated construct of PSF (sPSF, consisting of residues 214-598) alone and in a complex with a 30mer fragment of VL30-1 RNA, respectively. The structure of the complex reveals how the 30mer RNA is recognized at two U-specific induced-fit binding pockets, located at the previously unrecognized domain-swapped, inter-subunit RRM1 (of the first subunit)-RRM2 (of the second subunit) interfaces that do not exist in the apo structure. Thus, the sPSF dimer appears to have two conformations in solution: one in a low-affinity state for RNA binding, as seen in the apo-structure, and the other in a high-affinity state for RNA binding, as seen in the sPSF-RNA complex. PSF undergoes an all or nothing transition between having two or no RNA-binding pockets. We predict that the RNA binds with a high degree of positive cooperativity. These structures provide an insight into a new regulatory mechanism that is likely involved in promoting malignancies and other human diseases.

From a retrovirus infection of mice to a long noncoding RNA that induces proto-oncogene transcription and oncogenesis via an epigenetic transcription switch.

Here I review the properties of the mouse retroelement VL30-1, which apparently derived from retrotranspostions of a founder VL30 retrovirus that infected the mouse germline after the mouse-human speciation. The VL30-1 gene is transcribed as a long noncoding RNA (lncRNA) with an essential host function in an epigenetic transcription switch (ETS) that regulates transcription of multiple genes, including proto-oncogenes that control cell proliferation and oncogenesis. The ETS involves the tumor suppressor protein PSF that has a DNA-binding domain (DBD) and two RNA-binding domains (RBDs). The DBD binds to promoters that have a DBD-binding sequence and switches off transcription, and the RBDs bind lncRNAs that have a RBD-binding sequence, releasing PSF and switching on transcription. VL30-1 lncRNA has two RBD-binding sequences, apparently acquired by mutations during retrotranspositions of the founder retrovirus, which drive proto-oncogene transcription and oncogenesis via the ETS. VL30-1 lncRNA is a seminal example of the key role of endogenous retroviruses (ERVs) and their retroelements in the evolution of transcription regulatory systems.

The immunoconjugate "icon" targets aberrantly expressed endothelial tissue factor causing regression of endometriosis.

Endometriosis is a major cause of chronic pain, infertility, medical and surgical interventions, and health care expenditures. Tissue factor (TF), the primary initiator of coagulation and a modulator of angiogenesis, is not normally expressed by the endothelium; however, prior studies have demonstrated that both blood vessels in solid tumors and choroidal tissue in macular degeneration express endothelial TF. The present study describes the anomalous expression of TF by endothelial cells in endometriotic lesions. The immunoconjugate molecule (Icon), which binds with high affinity and specificity to this aberrant endothelial TF, has been shown to induce a cytolytic immune response that eradicates tumor and choroidal blood vessels. Using an athymic mouse model of endometriosis, we now report that Icon largely destroys endometriotic implants by vascular disruption without apparent toxicity, reduced fertility, or subsequent teratogenic effects. Unlike antiangiogenic treatments that can only target developing angiogenesis, Icon eliminates pre-existing pathological vessels. Thus, Icon could serve as a novel, nontoxic, fertility-preserving, and effective treatment for endometriosis.

Regulation of proto-oncogene transcription, cell proliferation, and tumorigenesis in mice by PSF protein and a VL30 noncoding RNA.

We describe the role of PSF protein and VL30-1 RNA, a mouse retroelement noncoding RNA, in the reversible regulation of proto-oncogene transcription, cell proliferation, and tumorigenesis in mice. The experiments involved increasing expression of PSF or VL30-1 RNA in NIH/3T3 fibroblast cells and B16F10 melanoma cells by transfecting the respective coding genes under control of a strong promoter or decreasing expression by transfecting a shRNA construct that causes degradation of PSF mRNA or VL30-1 RNA. The results are as follows: (i) PSF binds to the proto-oncogene Rab23, repressing transcription, and VL30-1 RNA binds and releases PSF from Rab23, activating transcription; (ii) increasing expression of PSF or decreasing expression of VL30-1 RNA suppresses cell proliferation in culture and tumorigenesis in mice; and (iii) decreasing expression of PSF or increasing expression of VL30-1 RNA promotes cell proliferation in culture and tumorigenesis in mice. These results indicate that PSF is a major tumor-suppressor protein and VL30-1 RNA is a major tumor-promoter RNA in mice. Although VL30-1 RNA can integrate into the cell genome, tumor promotion by VL30-1 RNA involves a trans effect rather than a cis effect on gene transcription. Expression of VL30-1 RNA is 5- to 8-fold higher in mouse tumor lines than in mouse fibroblast or myoblast lines, whereas expression of PSF mRNA does not decrease in the tumor lines, suggesting that tumorigenesis is driven by an increase of VL30-1 RNA rather than a decrease of PSF. A similar regulatory mechanism functions in human cells, except that human PSF-binding RNAs replace VL30-1 RNA, which is not encoded in the human genome. We propose that PSF protein and PSF-binding RNAs have a central role in the reversible regulation of mammalian cell proliferation and tumorigenesis and that increasing PSF expression or decreasing PSF-binding RNA expression in tumor cells is a potential therapeutic strategy for cancer.

Role of human noncoding RNAs in the control of tumorigenesis.

Related studies showed that the protein PSF represses proto-oncogene transcription, and VL30-1 RNA, a mouse noncoding retroelement RNA, binds and releases PSF from a proto-oncogene, activating transcription. Here we show that this mechanism regulates tumorigenesis in human cells, with human RNAs replacing VL30-1 RNA. A library of human RNA fragments was used to isolate, by affinity chromatography, 5 noncoding RNA fragments that bind to human PSF (hPSF), releasing hPSF from a proto-oncogene and activating transcription. Each of the 5 RNA fragments maps to a different human gene. The tumorigenic function of the hPSF-binding RNAs was tested in a human melanoma line and mouse fibroblast line, by determining the effect of the RNAs on formation of colonies in agar and tumors in mice. (i) Expressing in human melanoma cells the RNA fragments individually promoted tumorigenicity. (ii) Expressing in human melanoma cells a shRNA, which causes degradation of the endogenous RNA from which an RNA fragment was derived, suppressed tumorigenicity. (iii) Expressing in mouse NIH/3T3 cells the RNA fragments individually resulted in transformation to tumorigenic cells. (iv) A screen of 9 human tumor lines showed that each line expresses high levels of several hPSF-binding RNAs, relative to the levels in human fibroblast cells. We conclude that human hPSF-binding RNAs drive transformation and tumorigenesis by reversing PSF-mediated repression of proto-oncogene transcription and that dysfunctional regulation of human hPSF-binding RNA expression has a central role in the etiology of human cancer.

Factor VII-verteporfin for targeted photodynamic therapy in a rat model of choroidal neovascularization.

PURPOSE: To study the efficacy and safety of factor VII (fVII)-verteporfin for targeted photodynamic therapy (TPT) compared with nontargeted photodynamic therapy (PDT) in a rat model of choroidal neovascularization (CNV). fVII-verteporfin binds tightly and specifically to tissue factor, which is expressed on endothelial cells of CNV but not normal vasculature. METHODS: Multiple CNV lesions were induced by laser photocoagulation of the retina in Brown-Norway rats. After 3 weeks, the rats were injected intravenously with fVII-verteporfin (0.5 and 1.0 mg/m(2)) or Visudyne (6.0 mg/ m(2); QLT Inc., Vancouver, BC, Canada). Randomly selected lesions were treated with a 689-nm laser 30 or 60 minutes later. The lesions were evaluated by fluorescein angiography and histopathology. RESULTS: The rats treated with Visudyne PDT showed leakage in 75% of the CNV lesions on day 7 and 100% of lesions on day 14. The rats treated with fVII-verteporfin TPT at a dose of 0.5 mg/m(2) showed leakage in 33% and 36% of the CNV lesions on days 7 and 14, respectively. When the dose was increased to 1.0 mg/m(2) for TPT, leakage was detected in 25% and 23% of the CNV lesions on days 7 and 14, respectively. No ocular side effect was detected by histopathologic evaluation. CONCLUSIONS: The frequency of leakage in CNV lesions was significantly reduced using fVII-verteporfin TPT compared with PDT. The efficacious dose with fVII-verteporfin was approximately 10% of the dose usually used in nontargeted Visudyne PDT. Using fVII-verteporfin for TPT may improve the efficacy and safety of PDT for treating choroidal neovascularization.

Regulatory roles of tumor-suppressor proteins and noncoding RNA in cancer and normal cell functions.

We describe a mechanism for reversible regulation of gene transcription, mediated by a family of tumor-suppressor proteins (TSP) containing a DNA-binding domain (DBD) that binds to a gene and represses transcription, and RNA-binding domains (RBDs) that bind RNA, usually a noncoding RNA (ncRNA), forming a TSP/RNA complex that releases the TSP from a gene and reverses repression. This mechanism appears to be involved in the regulation of embryogenesis, oncogenesis, and steroidogenesis. Embryonic cells express high levels of RNA that bind to a TSP and prevent repression of proto-oncogenes that drive cell proliferation. The level of the RNA subsequently decreases in most differentiating cells, enabling a TSP to repress proto-oncogenes and stop cell proliferation. Oncogenesis can result when the level of the RNA fails to decrease in a proliferating cell or increases in a differentiated cell. This mechanism also regulates transcription of P450scc, the first gene in the steroidogenic pathway.

Targeting tissue factor for immunotherapy of choroidal neovascularization by intravitreal delivery of factor VII-Fc chimeric antibody.

PURPOSE: ICON is a fusion protein composed of factor VII, the natural ligand for tissue factor, conjugated to the Fc domain of a human IgG1 immunoglobulin. It binds to the tissue factor expressed on neovascular endothelia and initiates a cytolytic immune attack that destroys the neovascular tissue. We previously showed that mouse factor VII-Fc chimeric antibody (mICON) dramatically decreases the frequency of choroidal neovascularization in a laser-induced choroidal neovascularization model in mice. Herein, we determined the safety and efficacy of mICON in destroying subretinal choroidal neovascularization in pig eyes. METHODS: mICON (150-1200 microg) was administered into the midvitreous cavity of the pig eye either before (on Day 0) or after (on Day 10) induction of choroidal neovascularization with laser photocoagulation. On Day 14, the incidence of choroidal neovascularization was determined using confocal microscopy. We also determined the binding specificity (% binding to choroidal neovascularization/% binding to non-choroidal neovascularization areas) of mICON to tissue factor expressed on endothelial cells of laser-induced choroidal neovascularization. RESULTS: We observed that mICON selectively destroyed choroidal neovascularization in a dose-dependent manner (r = -0.93; EDB50B = 571.3 microg). Obliteration of the choroidal neovascular complex was more prominent at doses > 300 microg (p < 0.05). No systemic or local complications (including retinal tear/detachment, inflammation, infection, cataract, or glaucoma) were observed. Binding specificities of hICON (2.2 +/- 0.2) and mICON (3.4 +/- 0.4) were significantly higher than that of anti-von Willebrand antibody (0.1 +/- 0.01, p < 0.001). CONCLUSIONS: Both hICON and mICON bound to the neovascular endothelia of choroidal neovascularization with greater specificity than anti-von Willebrand antibody. Furthermore, mICON can selectively obliterate already established choroidal neovascularization, which suggests that it may be useful for immunotherapy in patients with exudative (wet) macular degeneration.

Roles of PSF protein and VL30 RNA in reversible gene regulation.

The mammalian protein PSF contains a DNA-binding domain (DBD) that coordinately represses multiple oncogenic genes in human cell lines, indicating a role for PSF as a human tumor-suppressor protein. PSF also contains two RNA-binding domains (RBD) that form a complex with a noncoding VL30 retroelement RNA, releasing PSF from a gene and reversing repression. Thus, the DBD and RBD in PSF are linked by a mechanism of reversible gene regulation involving a noncoding RNA. This mechanism also could apply to other regulatory proteins that contain both DBD and RBD. The mouse genome has multiple copies of VL30 retroelements that are developmentally regulated, and mouse cells contain VL30 RNAs that have normal and pathological roles in gene regulation. Human chromosome 11 has a VL30 retroelement, and a VL30 EST was identified in human blastocyst cells, indicating that the PSF-VL30 RNA regulatory mechanism also could function in human cells.

Anti-tumor effects of mouse factor VII-Staphylococcus enterotoxin A (mfVII) chimeric protein in mice / 中华肿瘤杂志

<p><b>OBJECTIVE</b>Mouse factor VII (mfVII), ligand of tissue factor (TF) which is frequently over-expressed during neovascularization activated by tumor growth, was fused to staphylococcus enterotoxin A (SEA) that mediates greater intensity of T-cell activation against tumor cells. The anti-tumor effects of the mfVII-SEA chimeric protein were evaluated.</p><p><b>METHODS</b>Fusion of SEA and mfVII cDNA was constructed using adenovirus vector and produced in 293 packaging cell lines. The 293 cells containing the adenovirus were administered subcutaneously to mice. Fluorescence studies at the injection site and the liver were performed 3 days later. Mouse prostatic tumor RM-1 cells and mouse sarcoma MCA 205 H12 cell lines were then used in mice to create lung metastasis and subcutaneous tumor to carry out efficacy evaluation, respectively.</p><p><b>RESULTS</b>Adenovirus released from the injected 293 cells only infected the subcutaneous tissue at the injection site. The in vivo experiments in mice revealed that formation of lung metastasis was strongly inhibited by the mfVII-SEA (23 +/- 8) compared to the vacant vector control group (193 +/- 38) and PBS control group (211 +/- 42) (P < 0.01). The mfVII-SEA also strongly suppressed tumor growth at the subcutaneous injection site (342.6 +/- 107.1) mm(3) compared to that of vacant vector control (2244.3 +/- 350) mm(3) and SEA (1208.3 +/- 210) mm(3) by the 23rd day.</p><p><b>CONCLUSION</b>The chimeric protein mfVII-SEA significantly inhibits lung metastasis formation and local tumor growth.</p>

Binding of mouse VL30 retrotransposon RNA to PSF protein induces genes repressed by PSF: effects on steroidogenesis and oncogenesis.

We describe a mechanism of gene regulation involving formation of a complex between PSF protein and mouse VL30 (mVL30) retrotransposon RNA. PSF represses transcription of the insulin-like growth factor 1 (IGF1)-inducible gene P450scc by binding to an insulin-like growth factor response element (IGFRE) motif in the gene. The complex with mVL30 RNA releases PSF, allowing transcription to proceed. Retrovirally mediated transmission of mVL30 RNA to human tumor cells induced several genes, including oncogenes, which also are induced by IGF1, and promoted metastasis. In mice, steroid synthesis is activated in steroidogenic cells by pituitary hormones, which concomitantly induce transcription of mVL30 RNA in the cells. We showed that steroid synthesis could also be activated in mouse steroidogenic adrenal cells by transfection with cDNA encoding either mVL30 RNA tracts that form a complex with PSF or a small interfering RNA (siRNA) that degrades PSF transcripts. These results suggest that mVL30 RNA regulates steroidogenesis, and possibly other physiological processes of mice, by complex formation with PSF. Retrotransposons such as mVL30 apparently evolved not only as "junk" DNA but also as transcriptionally active noncoding DNA that acquired physiological and pathological functions.

Immunotherapy for choroidal neovascularization in a laser-induced mouse model simulating exudative (wet) macular degeneration.

Age-related macular degeneration (AMD) is the leading cause of blindness after age 55 in the industrialized world. Severe loss of central vision frequently occurs with the exudative (wet) form of AMD, as a result of the formation of a pathological choroidal neovasculature (CNV) that damages the macular region of the retina. We tested the effect of an immunotherapy procedure, which had been shown to destroy the pathological neovasculature in solid tumors, on the formation of laser-induced CNV in a mouse model simulating exudative AMD in humans. The procedure involves administering an Icon molecule that binds with high affinity and specificity to tissue factor (TF), resulting in the activation of a potent cytolytic immune response against cells expressing TF. The Icon binds selectively to TF on the vascular endothelium of a CNV in the mouse and pig models and also on the CNV of patients with exudative AMD. Here we show that the Icon dramatically reduces the frequency of CNV formation in the mouse model. After laser treatment to induce CNV formation, the mice were injected either with an adenoviral vector encoding the Icon, resulting in synthesis of the Icon by vector-infected mouse cells, or with the Icon protein. The route of injection was i.v. or intraocular. The efficacy of the Icon in preventing formation of laser-induced CNV depends on binding selectively to the CNV. Because the Icon binds selectively to the CNV in exudative AMD as well as to laser-induced CNV, the Icon might also be efficacious for treating patients with exudative AMD.

Immunization and immunotherapy for cancers involving infection by a human papillomavirus in a mouse model.

We describe a protocol for generating a potent cellular immune response against viral-infected cells, and demonstrate its efficacy and safety in a mouse model of human cancer associated with infection by a human papillomavirus (HPV). In the mouse model, the mouse tumor TC-1, which expresses the E7 oncoprotein from HPV-16, is used as a surrogate for human tumors infected with HPV-16. The antigen for the protocol is composed of the E7 oncoprotein conjugated to the Fc region of a mouse IgG1 Ig (E7-mFc). The mFc domain should bind to Fc receptors on dendritic cells, enhancing the processing and presentation of E7 peptides by dendritic cells to T cells, which mediate a cellular immune attack against tumors expressing E7. The E7-mFc antigen was encoded in a replication-incompetent adenoviral vector, called Ad(E7-mFc), for infection of the human kidney cell line 293. The infected 293 cells synthesize the E7-mFc antigen and also infectious Ad(E7-mFc) vector particles for approximately equal 2 days, until the cells lyse and the vector particles are released. To test the protocol for immunization against formation of a TC-1 tumor, the mice first were injected s.c. with 293 cells infected with Ad(E7-mFc), followed by two challenges with TC-1 cells. The immunized mice remained healthy and tumor-free for the 6-month duration of the experiment, and the autopsies showed no toxicity. In the control mice immunized with 293 cells infected with an adenoviral vector that does not encode the E7-mFc antigen, the TC-1 tumor grew continuously and the mice had to be killed within 1 month. To test the protocol for immunotherapy, the mice first were injected with TC-1 cells, followed by s.c. injections of 293 cells infected with Ad(E7-mFc). Tumor growth was prevented or strongly retarded in these mice, in contrast to the continuous tumor growth in the controls. These results suggest that the protocol could be adapted for immunization against human cancers associated with an HPV infection, notably cervical cancer, and for immunotherapy to prevent recurrence of a tumor after surgery.

Retroviral-mediated transmission of a mouse VL30 RNA to human melanoma cells promotes metastasis in an immunodeficient mouse model.

Infection of a human melanoma cell line by a retroviral vector resulted in transmission of a mouse VL30 (mVL30-1) retroelement RNA to some of the cells infected by the retrovirus, followed by synthesis, integration, and expression of the mVL30-1 cDNA. One vector carried a tissue factor (TF) transgene that generated high TF melanoma clones, and another vector was a control without the TF transgene that generated low TF clones. Some high TF melanoma clones contained the mVL30-1 retroelement and others did not, and some low TF melanoma clones contained the mVL30-1 retroelement and others did not. Each type of melanoma clone was tested for its metastatic potential in severe combined immunodeficient (SCID) mice, by i.v. injection of the cells to generate lung tumors. None of the low TF clones that either contained or lacked the mVL30-1 retroelement generated lung tumors, consistent with earlier results showing that high TF expression promoted metastasis. The high TF clones containing the mVL30-1 retroelement were strongly metastatic, in contrast to the high TF clones lacking the mVL30-1 retroelement, which were weakly metastatic. Southern hybridization analyses showed that the mVL30-1 cDNA integrated into different genomic sites in different melanoma clones, suggesting that the effect of the mVL30-1 retroelement on metastasis depends not on integration per se but instead on expression of the mVL30-1 RNA. A role for the mVL30-1 RNA in metastasis and possibly other cell functions is an unexpected finding, because the RNA appears to lack significant coding potential for a functional protein. The metastatic effect might be mediated directly by a noncoding mVL30-1 RNA or by a peptide or small protein encoded by one of the short ORFs in the mVL30-1 RNA.