Recurrent rearrangement of the PHF1 gene in ossifying fibromyxoid tumors.

Ossifying fibromyxoid tumor (OFMT) is a soft tissue tumor of unknown lineage. Although most cases are histologically and clinically benign, some show malignant morphological features and local recurrences are not uncommon; a few may even metastasize. In the present study, cytogenetic analysis identified different structural rearrangements of chromosome band 6p21 in tumor cells from three cases of OFMT, including one with typical, one with atypical, and one with malignant morphological features. Mapping of the 6p21 breakpoint by fluorescence in situ hybridization (FISH) indicated that the PHF1 gene was rearranged in all three cases. Further FISH, 5'-rapid amplification of cDNA ends, and RT-PCR analyses disclosed an EP400/PHF1 fusion transcript in one of the cases. Interphase FISH on tumor sections from 13 additional cases of OFMT showed rearrangement of the PHF1 locus in four of four typical, two of three atypical, and one of six malignant lesions. Thus, the PHF1 gene, previously shown to be the 3'-partner of fusion genes in endometrial stromal tumors, is also recurrently involved in the pathogenesis of OFMTs, irrespective of whether they are diagnosed as typical, atypical, or malignant lesions. The PHF1 protein interacts with the polycomb-repressive complex 2 (PRC2), which, in turn, regulates the expression of a variety of developmental genes. Thus, the results indicate that deregulation of PRC2 target genes is crucial for OFMT development.

Fusion of the AHRR and NCOA2 genes through a recurrent translocation t(5;8)(p15;q13) in soft tissue angiofibroma results in upregulation of aryl hydrocarbon receptor target genes.

Soft tissue angiofibroma is a recently delineated tumor type of unknown cellular origin. Cytogenetic analysis of four cases showed that they shared a t(5;8)(p15;q13). In three of them it was the sole change, underlining its pathogenetic significance. FISH mapping suggested the involvement of the aryl hydrocarbon receptor repressor (AHRR) and nuclear receptor coactivator 2 (NCOA2) genes in 5p15 and 8q13, respectively. RT-PCR revealed in-frame AHRR/NCOA2 and NCOA2/AHHR transcripts in all four cases. Interphase FISH on paraffin-embedded tissue from 10 further cases without cytogenetic data showed that three were positive for fusion of AHRR and NCOA2. While AHRR has never been implicated in gene fusions before, NCOA2 is the 3'-partner in fusions with MYST3 and ETV6 in leukemias and with PAX3 and HEY1 in sarcomas. As in the previously described fusion proteins, NCOA2 contributes with its two activation domains to the AHRR/NCOA2 chimera, substituting for the repressor domain of AHRR. Because the amino terminal part of the transcription factor AHRR, responsible for the recognition of xenobiotic response elements in target genes and for heterodimerization, shows extensive homology with the aryl hydrocarbon receptor (AHR), the fusion is predicted to upregulate the AHR/ARNT signaling pathway. Indeed, global gene expression analysis showed upregulation of CYP1A1 as well as other typical target genes of this pathway, such as those encoding toll-like receptors. Apart from providing a diagnostic marker for soft tissue angiofibroma, the results also suggest that this tumor constitutes an interesting model for evaluating the cellular effects of AHR signaling.

FOSL1 as a candidate target gene for 11q12 rearrangements in desmoplastic fibroblastoma.

Desmoplastic fibroblastoma (DF) is a benign fibroblastic/myofibroblastic tumor. Cytogenetic analyses have revealed consistent rearrangement of chromosome band 11q12, strongly suggesting that this region harbors a gene of pathogenetic importance. To identify the target gene of the 11q12 rearrangements, we analyzed six cases diagnosed as DF using chromosome banding, fluorescence in situ hybridization (FISH), single-nucleotide polymorphism array and gene expression approaches. Different structural rearrangements involving 11q12 were found in five of the six cases. Metaphase FISH analyses in two of them mapped the 11q12 breakpoints to an ~20-kb region, harboring FOSL1. Global gene expression profiling followed by quantitative real-time PCR showed that FOSL1 was expressed at higher levels in DF with 11q12 rearrangements than in desmoid-type fibromatoses. Furthermore, FOSL1 was not upregulated in the single case of DF that did not show cytogenetic involvement of 11q12; instead this tumor was found to display a hemizygous loss on 5q, including the APC (adenomatous polyposis coli) locus, raising the possibility that it actually was a misdiagnosed Gardner fibroma. 5'RACE-PCR in two 11q12-positive DF did not identify any fusion transcripts. Thus, in agreement with the finding at chromosome banding analysis that varying translocation partners are involved in the 11q12 rearrangement, the molecular data suggest that the functional outcome of the 11q12 rearrangements is deregulated expression of FOSL1.

EWSR1-ATF1 dependent 3D connectivity regulates oncogenic and differentiation programs in Clear Cell Sarcoma.

Oncogenic fusion proteins generated by chromosomal translocations play major roles in cancer. Among them, fusions between EWSR1 and transcription factors generate oncogenes with powerful chromatin regulatory activities, capable of establishing complex gene expression programs in permissive precursor cells. Here we define the epigenetic and 3D connectivity landscape of Clear Cell Sarcoma, an aggressive cancer driven by the EWSR1-ATF1 fusion gene. We find that EWSR1-ATF1 displays a distinct DNA binding pattern that requires the EWSR1 domain and promotes ATF1 retargeting to new distal sites, leading to chromatin activation and the establishment of a 3D network that controls oncogenic and differentiation signatures observed in primary CCS tumors. Conversely, EWSR1-ATF1 depletion results in a marked reconfiguration of 3D connectivity, including the emergence of regulatory circuits that promote neural crest-related developmental programs. Taken together, our study elucidates the epigenetic mechanisms utilized by EWSR1-ATF1 to establish regulatory networks in CCS, and points to precursor cells in the neural crest lineage as candidate cells of origin for these tumors.

The t(X;6) in subungual exostosis results in transcriptional deregulation of the gene for insulin receptor substrate 4.

Subungual exostosis is a benign bone- and cartilage-forming tumor known to harbor a pathognomonic t(X;6)(q22;q13-14). Using global gene expression analysis and quantitative real-time PCR, we could show that this translocation results in increased expression of the IRS4 gene, presumably due to disruption and/or exchange of regulatory sequences with the translocation partner, the COL12A1 gene. A corresponding deregulation at the protein level could be demonstrated in primary cell cultures using a combination of fluorescence in situ hybridization and immunostaining. As the t(X;6) usually is the sole cytogenetic aberration in subungual exostosis, the deregulated expression of IRS4 is likely to be pathogenetically essential. The exact role of IRS4 is still poorly investigated, but IRS proteins are known to act as mediators of signaling from receptors, such as the insulin and insulin-like growth factor 1 receptors, and thus have an important effect on cell growth and survival.

Bidirectionality and transcriptional activity of the EWSR1 promoter region.

EWSR1 is involved in chimeric proteins which play crucial roles in the development of a variety of bone and soft tissue tumors. Many of the chimeric genes involving EWSR1 have been extensively studied, whereas less is known about the wild-type (wt) gene and its regulation. As the expression of the chimeric gene is driven by the EWSR1 promoter, it is of importance to study the mechanisms regulating wt EWSR1 expression. We estimated the transcriptional activity of the EWSR1 promoter through deletion fragments driving reporter gene expression. This assay identified the 100-bp region immediately downstream of the EWSR1 transcriptional start site (+1) and the downstream region from +100 to +300 as important regions for transcriptional regulation. We also found that EWSR1 and RHBDD3, a gene located directly upstream of EWSR1 that is likely to share regulatory elements with EWSR1, were co-expressed in the tissue panels, Ewing tumor biopsies and cell lines. Thus, our results show that the EWSR1 promoter functions in a bidirectional manner, thereby regulating also RHBDD3, and identifies specific regions that strongly influence promoter activity.

A PCR/restriction digestion assay for the detection of the transcript variants 1 and 2 of POU5F1.

POU5F1 has two alternatively spliced transcripts: a long (variant 1, NM_002701) and a short (variant 2, NM_203289) transcript. Only variant 1 is a key regulator of pluripotency. Hence, it is important to be able to distinguish this transcript from variant 2 and from the many pseudogenes present in the genome. Previous studies on the expression of POU5F1 were, however, usually carried out without considering the existence of the two transcripts and the pseudogenes which could be the source of false positive RT-PCR amplification. Here, we establish an RT-PCR/restriction digestion analysis to distinguish variant 1 of POU5F1 from variant 2 and all its currently known pseudogenes. Variant 1 has ApaI and Tsp45I restriction sites, which are not present in the pseudogenes or in variant 2. Thus, ApaI- and Tsp45I- digestions of POU5F1 PCR fragment, amplified with primers flanking these sites, are sufficient to identify the true variant 1 of POU5F1. To study the expression of variant 2 of POU5F1, two forward primers in the 5'-region that are not present in variant 1 were combined with reverse primers located in exon 3 of POU5F1 common to both transcripts. The assay was applied on 10 samples from peripheral blood leukocytes and commercially available ready-cDNAs from leukocytes and testis. We found that only variant 2 was expressed in leukocytes and testis and that the extracted RNA was not completely DNA free, despite DNAse treatment. This trace amount of DNA is a source of false positive RT-PCR amplifications. This article contains Supplementary Material available at http://www.interscience.wiley.com/jpages/1045-2257/suppmat.

High-resolution molecular cytogenetic analysis of Wilms tumors highlights diagnostic difficulties among small round cell kidney tumors.

Many solid tumors exhibit characteristic gene fusions, which are reflected by balanced translocations at the cytogenetic level. These changes might be useful diagnostic and prognostic tools. In Wilms tumor (WT, nephroblastoma) no fusions genes or recurrent balanced translocations have been described thus far. To screen for cryptic balanced translocations, we have analyzed 17 renal neoplasms, histopathologically classified as WT, by a combination of G-banding, multicolor FISH, and subtelomeric FISH. This approach revealed several submicroscopic chromosomal aberrations and three different seemingly balanced translocations, resulting in a heterozygous deletion of HACE1, an EWSR1/ERG fusion, and an EWSR1/FLI1 fusion, respectively. As EWSR1 rearrangements are known to be a characteristic of Ewing tumors (ET), our findings illustrate the diagnostic problems regarding small cell kidney tumors and strongly argue for the need of adjuvant diagnostic techniques in this group of neoplasms. In summary, our genomic screening approach proved efficient in finding structural chromosomal aberrations. The fact that no recurrent translocations were found in the WTs of this study argues against the presence of a frequent pathognomonic translocation in this disease entity.

The POU5F1P1 pseudogene encodes a putative protein similar to POU5F1 isoform 1.

POU5F1, which encodes a transcriptional factor, has two alternatively spliced transcripts, 1 and 2, as well as six pseudogenes. Transcript 1 is considered to be a key regulator of cellular pluripotency and self-renewal. The POU5F1 pseudogene, POU5F1P1 on 8q24, encodes a protein with 95% homology with the isoform 1 of POU5F1. It is located 15 kbp downstream of the SNP rs6983267, which is strongly associated with an increased risk of prostate and colon cancer, and within the amplified region in a variety of human malignancies. The previous finding of expressed sequence tags suggests that POU5F1P1 can be expressed. We showed that a putative POU5F1P1 protein is localized in the nucleus, acts as a transcriptional activator and regulates the expression in a similar way to the POU5F1 isoform 1. However, POU5F1P1 was a weaker activator than isoform 1 of POU5F1, possibly due to the amino acid substitutions.

Reciprocal modulation of mesenchymal stem cells and tumor cells promotes lung cancer metastasis.

Metastasis is a multi-step process in which direct crosstalk between cancer cells and their microenvironment plays a key role. Here, we assessed the effect of paired tumor-associated and normal lung tissue mesenchymal stem cells (MSCs) on the growth and dissemination of primary human lung carcinoma cells isolated from the same patients. We show that the tumor microenvironment modulates MSC gene expression and identify a four-gene MSC signature that is functionally implicated in promoting metastasis. We also demonstrate that tumor-associated MSCs induce the expression of genes associated with an aggressive phenotype in primary lung cancer cells and selectively promote their dissemination rather than local growth. Our observations provide insight into mechanisms by which the stroma promotes lung cancer metastasis.

Comparison of the proximal promoter regions of the PAX3 and PAX7 genes.

Translocations t(2;13)(q35;q14) and t(1;13)(p36;q14), which fuse PAX3 and PAX7, respectively, to FOXO1A, characterize alveolar rhabdomyosarcoma. Previous studies have suggested that the expression of PAX7-FOXO1A is copy-number dependent, but that of PAX3-FOXO1A is not, which may be due to a weaker PAX7 than PAX3 promoter. The aim of the present study was to compare the transcriptional activities of the PAX3 and PAX7 proximal promoter regions, using the dual-luciferase reporter assay with three vector systems in eight cell lines. The PAX3 promoter was found to have higher transcriptional activity than that of PAX7 irrespective of the vector system or cell line used. These findings are consistent with the idea that an amplification event is required for the PAX7-FOXO1A chimeric transcript to reach a critical expression level.

MUC4 is a highly sensitive and specific marker for low-grade fibromyxoid sarcoma.

Low-grade fibromyxoid sarcoma (LGFMS) is a distinctive fibroblastic neoplasm that is characterized by alternating collagenous and myxoid areas, deceptively bland spindle cell morphology, a whorling architecture, and a t(7;16) translocation involving FUS and CREB3L2. Owing to variable morphology and a lack of discriminatory markers, LGFMS can be difficult to distinguish from benign mesenchymal tumors and other low-grade sarcomas. Gene expression profiling has identified differential upregulation of the mucin 4 (MUC4) gene in LGFMS compared with histologically similar tumors. MUC4 is a transmembrane glycoprotein that functions in cell growth signaling pathways; aberrant MUC4 expression has been reported in various carcinomas. We investigated MUC4 protein expression by immunohistochemistry in LGFMS and in other soft tissue tumors to determine the potential diagnostic use of this novel marker. Whole-tissue sections of 309 tumors were evaluated: 49 LGFMSs (all with FUS gene rearrangement confirmed by fluorescence in situ hybridization), 40 soft tissue perineuriomas, 40 myxofibrosarcomas, 20 cellular myxomas, 20 solitary fibrous tumors, 20 low-grade malignant peripheral nerve sheath tumors, 20 cases of desmoid fibromatosis, 20 neurofibromas, 20 schwannomas, 20 monophasic synovial sarcomas, 20 cases of dermatofibrosarcoma protuberans, 10 myxoid liposarcomas, and 10 extraskeletal myxoid chondrosarcomas. The LGFMS cases included 7 with marked hypercellularity, 4 with prominent hemangiopericytoma-like vessels, 3 with giant collagen rosettes, 3 with epithelioid morphology, 2 with focal nuclear pleomorphism, and 2 with areas of sclerosing epithelioid fibrosarcoma. All 49 LGFMS cases (100%) showed cytoplasmic staining for MUC4, which was usually diffuse and intense. All the other tumor types were negative for MUC4, apart from 6 (30%) monophasic synovial sarcomas. In conclusion, MUC4 is a highly sensitive and quite specific immunohistochemical marker for LGFMS, and can be helpful to distinguish this tumor type from histologic mimics.

FUS-CREB3L2/L1-positive sarcomas show a specific gene expression profile with upregulation of CD24 and FOXL1.

PURPOSE: Low-grade fibromyxoid sarcoma (LGFMS) is typically characterized by the specific translocation t(7;16)(q33;p11) and the corresponding fusion gene FUS-CREB3L2. The present study aimed to extract LGFMS-specific, and putatively FUS-CREB3L2-dependent, gene expression patterns to learn more about the pathogenesis of this tumor. EXPERIMENTAL DESIGN: We carried out single nucleotide polymorphism (SNP) and global gene expression array analyses, and/or immunohistochemical (IHC) analyses on 24 LGFMS tumor biopsies. Tumor types that are important differential diagnoses to LGFMS were included as comparison in the gene and protein expression analyses. In addition, cells that stably expressed FUS-CREB3L2 were analyzed with gene expression array and the influence of FUS-CREB3L2 on gene expression was investigated in vitro. RESULTS: The SNP array analysis detected recurrent microdeletions in association with the t(7;16) chromosomal breakpoints and gain of 7q in cases with ring chromosomes. Gene expression analysis clearly distinguished LGFMS from morphologically similar tumors and MUC4 was identified as a potential diagnostic marker for LGFMS by gene expression and IHC analysis. FOXL1 was identified as the top upregulated gene in LGFMS and CD24 was upregulated in both LGFMS tumors and FUS-CREB3L2 expressing cells. FUS-CREB3L2 was capable of activating transcription from CD24 regulatory sequences in luciferase assays, suggesting an important role for the upregulation of this gene in LGFMS. CONCLUSIONS: The gene expression profile of LGFMS is distinct from that of soft tissue tumors with similar morphology. The data could be used to identify a potential diagnostic marker for LGFMS and to identify possible FUS-CREB3L2 regulated genes.

Fusion of the FUS and CREB3L2 genes in a supernumerary ring chromosome in low-grade fibromyxoid sarcoma.

Low-grade fibromyxoid sarcoma (LGFMS) is a rare, low-grade malignant soft tissue tumor that is often mistaken for either benign or more malignant tumor types. Commonly, this tumor affects young adults and typically arises in the deep proximal extremities or trunk with frequent recurrences and can metastasize to the lungs many years later. Most cases have a recurrent balanced translocation involving chromosomes 7 and 16, t(7;16)(q32-34;p11), which leads to the fusion of the FUS and CREB3L2 genes. However, supernumerary ring chromosomes have been identified in a subset of FUS/CREB3L2-positive LGFMS, but it has not yet been formally demonstrated that such ring chromosomes harbor the FUS/CREB3L2 fusion gene. Here, we report the genetic findings of a supernumerary ring chromosome from an LGFMS from a 77-year-old man. Chromosome banding analysis revealed a supernumerary ring chromosome, and further studies with fluorescence in situ hybridization and reverse transcriptase-polymerase chain reaction (RT-PCR) showed that the ring contained material from chromosomes 7 and 16, that the FUS gene was present in two rearranged copies, and that it expressed the FUS/CREB3L2 fusion gene. Moreover, an assessment of previously reported cases showed that tumors with ring chromosomes relapsed more often than tumors with a balanced t(7;16), suggesting that ring formation in LGFMS is correlated with tumor progression.

Molecular identification of COL6A3-CSF1 fusion transcripts in tenosynovial giant cell tumors.

Tenosynovial giant cell tumors (TGCTs) are benign lesions of the tendon sheaths that primarily affect the fingers, ankles, or feet. Cytogenetic data have shown that 1p13 is most frequently involved in structural aberrations and that 2q37 is its most common translocation partner. The genes involved in the translocation t(1;2)(p13;q37) were recently identified: the colony-stimulating factor-1 (CSF1 or M-CSF1) at 1p13 and the collagen type VI alpha-3 (COL6A3) at 2q37. Based upon the suggestion that a fusion of these genes through the translocation would result in overexpression of CSF1 due to a strong COL6A3 promoter, we performed RT-PCR on six TGCT cases with t(1;2) to search for a putative COL6A3-CSF1 fusion gene. Such fusion transcripts were detected in three cases of which one was an in-frame fusion. In all cases, however, the breakpoints in CSF1 appeared downstream of exon 5, indicating that the amino-terminal part of CSF1, which interacts with its receptor CSF1R, is not encoded by the the chimeric transcripts we identified. The pathogenetic mechanism of these chimeric transcripts is therefore unclear.

Characterization of the native CREB3L2 transcription factor and the FUS/CREB3L2 chimera.

CREB3L2 was first identified as the 3'-partner of FUS in a fusion gene that seems to be specific for low grade fibromyxoid sarcoma. In silico analyses suggest that the predicted CREB3L2 protein is a member of the CREB3 family of transcription factors, with its bZIP domain being highly similar to that in CREB3L1, CREB3L3, CREB3L4, CREB3, and Drosophila Bbf-2. In the present study, the authors assessed various cellular outcomes after transfection of NIH3T3 and HEK-293 cells with constructs containing full-length and truncated versions of CREB3L2 and FUS/CREB3L2. Northern blot of CREB3L2 mRNA revealed a 7.4 kbp band that contains 0.4 kbp and 5.5 kbp untranslated 5' and 3' regions, respectively. CREB3L2 constructs containing the first 120 amino acids (aa) showed the highest transcriptional activation. Much stronger transcriptional activation was consistently seen for the FUS/CREB3L2 constructs than for the corresponding CREB3L2 constructs. Transcriptional activity was achieved through the box-B element, ATF6 and CRE binding sites, as well as the GRP78 promoter. Proteins encoded by full-length CREB3L2 and FUS/CREB3L2 were localized to reticular structures of the cytoplasm, whereas the corresponding, truncated proteins lacking the transmembrane domain and the carboxy-terminal part of CREB3L2 resided within the nucleus. The results of the present study show that CREB3L2 is not only structurally, but also functionally very similar to CREB3L1. Thus, studies regarding the pathways influenced by wild-type CREB3L2 should provide valuable clues to the pathogenetic significance of the FUS/CREB3L2 chimera in low grade fibromyxoid sarcoma.