NUP160–SLC43A3 Is a Novel Recurrent Fusion Oncogene In

Cancer Priority Report Research

NUP160–SLC43A3 Is a Novel Recurrent Fusion Oncogene in Angiosarcoma Naoki Shimozono1, Masatoshi Jinnin1, Mamiko Masuzawa2, Mikio Masuzawa3, Zhongzhi Wang1, Ayaka Hirano1, Yukiko Tomizawa1, Tomomi Etoh-Kira1, Ikko Kajihara1, Miho Harada1, Satoshi Fukushima1, and Hironobu Ihn1

Abstract

Angiosarcoma is a malignant vascular tumor originating from dermal microvascular endothelial cells elicited a gene-expres- endothelial cells of blood vessels or lymphatic vessels. The sion pattern mimicking ISO-HAS cells and increased cell pro- specific driver mutations in angiosarcoma remain unknown. In liferation, an effect traced in part to NUP160 truncation. Con- this study, we investigated this issue by transcriptome sequenc- versely, RNAi-mediated attenuation of NUP160 in ISO-HAS ing of patient-derived angiosarcoma cells (ISO-HAS), identify- cells decreased cell number. Confirming the oncogenic effects ing a novel fusion gene NUP160–SLC43A3 found to be of the fusion protein, subcutaneous implantation of NUP160– expressed in 9 of 25 human angiosarcoma specimens that were SLC43A3-expressing fibroblasts induced tumors resembling examined. In tumors harboring the fusion gene, the duration human angiosarcoma. Collectively, our findings advance knowl- between the onset of symptoms and the first hospital visit was edge concerning the genetic causes of angiosarcoma, with poten- significantly shorter, suggesting more rapid tumor progression. tial implications for new diagnostic and therapeutic approaches. Stable expression of the fusion gene in nontransformed human Cancer Res; 75(21); 4458–65. 2015 AACR.

Introduction Materials and Methods Angiosarcoma is a malignant vascular tumor originating from Cell cultures endothelial cells of blood vessels or lymphatics vessels. The ISO-HAS was isolated from a tumor tissue as described in 1992 tumors frequently occur in the liver, breast, and skin. Cutaneous (1). Adult human dermal microvascular endothelial cells angiosarcoma usually arises on the scalp of elderly individuals, (HDMEC) or NIH3T3 cells were obtained from Lonza or ATCC and rapidly metastasizes. We previously established cultured in 2009, respectively. Cells were tested and authenticated by DNA human cutaneous angiosarcoma cell line, ISO-HAS, to clarify the fingerprinting in May 2015. malignant characteristics of the tumor, and found that the cells harbor a TP53 point mutation (1). However, there may be more RNA isolation specific driver mutations in angiosarcoma. Trisomy 5 and a loss of High-quality RNA for transcriptome sequencing was obtain- the Y chromosome were reported in a single case of angiosarcoma ed from cultured cells using the RNeasy Mini Kit. The RNeasy arising within a venous malformation (2). Antonescu and col- FFPE Kit was used for RNA extraction from paraffin-embedded leagues (3) reported some patients with breast angiosarcoma who sections. Institutional review board approval and written harbored KDR mutations. More recently, mutations in PTPRB and informed consent were obtained. PLCG1 were identified in the tumor tissues of angiosarcoma patients (4). Although these changes seem to be functional, they Library preparation and transcriptome analysis are not found in all patients, and the mechanism by which they Transcriptome analysis was performed in accordance with the contribute to tumorigenesis is unclear. protocol provided by Riken Genesis using the Illumina TruSeq In the present study, we describe a novel frequent fusion gene, RNA Sample Preparation Kit. Final cDNA libraries were NUP160–SLC43A3, in cutaneous angiosarcoma patients. sequenced on Illumina HiSeq2000 platform. The clean and trimmed reads were aligned to GRCh37/hg19 using TopHat. The differential expression between the samples was analyzed with 1Department of Dermatology and Plastic Surgery, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan. 2Department of Cuffdiff by calculating the fragments per kilobase per million map Dermatology, Kitasato University School of Medicine, Kitasato, Kana- reads (FPKM). Bioinformatic analysis for detecting mutations was 3 gawa, Japan. Department of Molecular Diagnostics, School of Allied performed using Samtools. Potential gene fusion transcripts were Health Sciences, Kitasato University, Kitasato, Kanagawa, Japan. identified and filtered by TopHat-Fusion. Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/). PCR Corresponding Author: Masatoshi Jinnin, Faculty of Life Sciences, Kumamoto For RT-PCR, cDNA, and primers were mixed with SYBR Premix University, 1-1-1 Honjo, Kumamoto 860-8556, Japan. Phone; 81-96-373-5233; Ex Taq II. PCR was performed for 50 cycles of denaturation for 5 Fax: 81-96-373-5235; E-mail: [email protected] seconds at 95C, annealing for 30 seconds at 60C, and extension doi: 10.1158/0008-5472.CAN-15-0418 for 1 minute at 68C. Mutations of TP53 or KDR were detected as 2015 American Association for Cancer Research. described previously (3, 5, 6).

4458 Cancer Res; 75(21) November 1, 2015

A Figure 1. Chr11 The identification of a novel NUP160– SLC43A3 fusion in angiosarcoma cells. A, top, a schematic diagram showing the NUP160 (red) and SLC43A3 (blue) gene structures in relation to the NUP160 exon26 SLC43A3 exon7 formation of the NUP160–SLC43A3 fusion transcript. Bottom, the partial sequence of the NUP160–SLC43A3 fusion transcript, along with the N AKCLAE LQ SARLSIGKEYLL predicted amino acid sequence. The transcript is an in-frame fusion of the NUP160 exon 26 to SLC43A3 exon 7. B, left, the RT-PCR detection of the NUP160–SLC43A3 fusion. The NUP160-SLC43A3 fusion presence of the NUP160–SLC43A3 fusion transcript was verified by a RT-PCR using the NUP160 forward (red) and SLC43A3 reverse primers B (blue). The fusion gene–specific primer was designed to not amplify wild-type full-length NUP160 or NUP160 exon26 SLC43A3 exon7 SLC43A3. Right, the PCR products and ladder marker were run out on an HDMEC agarose gel containing ethidium bromide. Lane M, 100-bp marker; lane ISO-HAS 400bp 1, HDMEC; lane 2, ISO-HAS; lane 3, 300bp 200bp cultured human dermal fibroblasts; lane 4, cultured human normal keratinocytes. Arrow, PCR product of C the fusion gene–specific primer HDMEC ISO-HAS M HDMEC ISO-HAS Fibroblasts Keratinocytes (240-bp). C, a FISH analysis with a human NUP160- (green) and SLC43A3 (red)-specific probe. HDMEC (left) had two green or red signals, whereas three or more green/red signals were detected in ISO-HAS. White arrow, fusion signal (yellow). D, the results of a chromosome analysis. Left, the karyotypes of HDMEC, which were diploid: 46, XX. Right, ISO-HAS had complex abnormal karyotypes and was aneuploid. The karyotype was D 93104,XY,-X,-Y,add(1)(p13),add(1) (q31),-2,-2,add(2)(q31),add(3)(q21),- 1 23 4 5 4,-4,add(4)(p12),-6,-6,add(6)(q11),-7, 1 23 45 add(7)(p11.2),add(8)(p11.2),add(8) 6 78 9 10 11 12 (p11.2),add(9)(p11),add(9)(p11),þ11, add(11)(p11.2)2,-12,-12,add(12) 6 7 8 9 10 11 12 (q13)2,add(14)(p11.2),add(15)(q22), 13 14 15 16 17 18 add(16)(q24),add(17)(q25),add(18) 13 14 15 16 17 18 (q21),add(20)(q13.1),add(21)(p11.2), 19 20 21 22 XY add(21)(p11.2),add(21)(p11.2), -22,þ 19 20 21 22 XY mar1,þmar2,þmar3,þmar4,þmar5,þ mar1mar2mar3mar4 mar5 mar6 mar7 mar8 mar mar6,þmar7,þmar8[cp20]. Arrows, putative break points. Inset, 11 chromosome 11. 11

For PCR array, cDNA was mixed with RT2 SYBR Green/Rox Empire Genomics. The slides were treated with probes, and were qPCR Master Mix, and the mixture was added to Human Angio- denatured by heating at 72C for 4 minutes. Hybridization was genesis PCR Array. performed by incubation at 37C overnight. The nuclei were stained with DAPI. Fluorescence in situ hybridization and G-banding For G-banding, slides were treated with 0.2% trypsin in FISH analysis was performed according to the protocol pro- Hanks' Balanced Salt Solution. The cells were then washed vided by Nihon Gene Research Laboratories. The cells were ﬁxed immediately with PBS, and were stained for 5 minutes with with freshly made Carnoy's solution. The probes were designed by Giemsa solution.

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 4459

A B Marker Tissue Blood Serum Marker NC Pt no.16 Pt no.17 Pt no.18 Pt no.19 Pt no. 20

Pt no.14 C D

Figure 2. The clinical and histopathologic features of angiosarcoma patients with the fusion gene. A, the RT-PCR detection of specific 240-bp fragments in tumor RNA from angiosarcoma patient nos. 16, 17, and 19, but not in numbers 18 or 20. Marker, 100-bp ladder. NC, PCR-negative control. Arrow, PCR product of the fusion gene–specific primers (240-bp). B, the RT-PCR detection of specific 240-bp fragment in the RNA of angiosarcoma tumor tissue, but not in the RNA of blood or serum from the same individual (patient no. 14). Marker, 100-bp ladder. Arrow, PCR product of the fusion gene–specific primers (240-bp). C, the representative clinical picture and histopathologic findings of a patient with the fusion gene (patient no. 14). Top left, the clinical presentation of the tumor, which included purpuras, tumors, and ulcers on the scalp. Top right, the results of a hematoxylin and eosin staining at low magnification (scale bars, 500 mm). Bottom left, the hematoxylin and eosin staining at high magnification (scale bars, 20 mm) showing poorly differentiated lesion; a mass of large pleomorphic and hyperchromatic tumor cells with luminal differentiation containing erythrocytes was observed. Nuclear atypia and mitotic figures were also seen. Bottom right, the tumor cells were positive for CD31 (scale bars, 500 mm). D, the representative clinical picture and histopathologic findings of a patient with the fusion gene (patient no. 22). Top upper, the clinical presentation of the tumor, which included purpuras, tumors, and ulcers on the scalp. Top right, the results of an hematoxylin and eosin staining at low magnification (scale bars, 500 mm). Bottom left, the hematoxylin and eosin staining at high magnification (scale bars, 50 mm) showing the well-differentiated lesion characterized by irregular anastomosing vascular channels lined by a single layer of endothelial cells with atypical nuclei. Bottom left, the tumor cells were positive for CD31 (scale bars, 50 mm).

Migration assays cells migrated to the undersurface of filter membrane were Migration assays were performed using 24-well plate Trans- counted in five different 400-fold magnification fields under well inserts with a polycarbonate filter membrane. Cells were an inverted microscope. detached by trypsin-EDTA, and were seeded into the upper compartment of Transwell chamber in 100 mLofserum-free Cell count and BrdUrd ELISA medium. The lower chamber was filled with 10% FBS-contain- Cells were detached from the wells by trypsin treatment and ing culture medium as a chemoattractant. After 24 hours, the counted using Coulter Particle Counter. Cell proliferation

4460 Cancer Res; 75(21) November 1, 2015 Cancer Research

A Control SLC43A3

NUP160-SLC43A3 fusion ISO-HAS

Figure 3. The phosphorylation state of the NUP160–SLC43A3 fusion. A, the microscopic characteristics of HDMECs transfected with the vector control, fusion gene or full-length SLC43A3, as well as ISO-HAS. B, a schematic representation of the wild- type full-length NUP160 protein, full- length SLC43A3 protein, and the B predicted NUP160–SLC43A3 fusion protein product identiﬁed in this Wild-type proteins study. P, phosphorylation site. The P numbers indicate the exon numbers. NUP160 1-10 11-20 21-30 31-36 C, the cell lysates were immunoprecipitated with an anti- SLC43A3 antibody to examine the SLC43A3 1345682 7 9 10 11 12 tyrosine phosphorylation levels of P NUP160–SLC43A3. All cell types, including HDMEC, ISO-HAS, and HDMEC, transfected with the control vector and HDMEC transfected with Predicted fusion proteins the fusion gene, expressed the full- length SLC43A3 protein. ISO-HAS 1-10 11-20 21-26 7 8 9 10 11 12 and HDMEC transfected with the P fusion gene expressed the fusion protein, but tyrosine phosphorylation of the fusion was not detected in these cells. C IP: IgG Anti-SLC43A3

WB: anti-phospho-Tyr

NUP160−SLC43A3 WB: anti-SLC43A3 Full-length SLC

HDMEC ISO-HAS Control vector Fusion gene HDMEC activity was confirmed using the bromodeoxyuridine (BrdUrd) to electrophoresis on sodium dodecyl sulfate-polyacrylamide ELISA kit. gels and transferred onto polyvinylidene difluoride (PVDF) filters. The PVDF filters were incubated with primary antibo- Immunoprecipitation dies, followed by incubation with secondary antibodies. Cell lysates were precleared, and then incubated with an anti- SLC43A3 antibody and Protein A/G Plus Agarose overnight at Lentiviral gene transfer 4C. Agarose-bound proteins were extracted by incubation in Lentiviral vector–mediated gene transfer was performed using sample buffer at 95C. The immunoprecipitates were subjected CSII-EF-RfA, pCMV-VSV-G-RSV-Rev, and pHIVgp (7).

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 4461

A B C 80 Cell number BrdU 70 30 10 60 25

) * 50 4 60

) 8 4 40 20 ** * 30 6 Cells/field 15 40 20 4 10 10 HDMEC Cell number (x10 Cell number 20 Relative BrdUrd Cell number (x10 Cell number 0 HDMEC ISO-HAS 5 ISO-HAS 2 n.d. 0 0 0 Passage 6141820 Control Fusion SLC NUP D Marker * E F ) 6 4 6 ) 4 4 4 * NUP160

Cell number (x10 2 2 Control NUP160 Truncated GAPDH Cell number (x10 0 NUP160 siRNA − + NUP160 siRNA − +

G Control NUP160-SLC43A3 fusion

H I Marker Tissue 1 Tissue 2 Cell 1 Cell 2 NC

Figure 4. The functional analysis of the NUP160–SLC43A3 fusion gene. A, for the cell migration analysis, 1 105 HDMECs or ISO-HAS were seeded in serum-free medium into the top chamber of the Transwell system. After 48 hours, the cells remaining on the upper surface were wiped off, and those that had migrated to the lower surface were counted in five different 400-fold magnification fields after staining with hematoxylin. The graph depicts the number of cells that migrated to the bottom of the membrane. B, HDMEC or ISO-HAS (1 105 cells) at the indicated passages were plated and grown in 6-well plates. After 48 hours, the cells were detached from the wells by trypsin treatment and were counted. , P ¼ 0.049 in comparison with HDMEC by the Mann–Whitney U test (n ¼ 3). C, for the proliferation analysis, HDMECs that were stably transfected with a lentiviral control, fusion gene, full-length SLC43A3 (SLC), or full-length NUP160 (NUP) were counted as described in Materials and Methods. Cells were also labeled with BrdUrd and analyzed by ELISA. The white bars indicate the cell numbers, and the black lines represent the relative absorbance as determined by BrdUrd ELISA. , P ¼ 0.049 in comparison with cells transfected with the control vector by the Mann–Whitney U test (n ¼ 3). n.d, not determined because the NUP160-transfected cells did not grow. D, HDMECs were stably transfected with the lentiviral vector control or truncated NUP160. The number of cells was counted. , P ¼ 0.049 in comparison with cells transfected with the control vector by the Mann–Whitney U test (n ¼ 3) as described in Fig. 4B. E and F, ISO-HAS were transfected with control siRNA or NUP160 siRNA. To show the transfection efficiency of NUP160 siRNA, the PCR products obtained using RNA from ISO-HAS and the fusion gene–specific primer pair were run out on agarose gels containing ethidium bromide. (Continued on the following page.)

4462 Cancer Res; 75(21) November 1, 2015 Cancer Research

Transient transfection with HDMEC; the fold-changes in expression were 0.90 (q value; For reverse transfection, siRNAs mixed with Lipofectamine 0.99) and 2.41 (q value; 0.82), respectively. RNAiMAX was added when the cells were plated. To characterize the rearrangements at the chromosomal level, multicolor FISH experiments were performed. In HDMEC, two Tumorigenicity assay separate green (NUP160) and two separate red signals (SLC43A3) NIH3T3 cells were implanted into 6-week-old female athymic were visible (Fig. 1C). On the other hand, the fusion of one pair of nu/nu mice by s.c. injection using 25-gauge needles. The implants signals, appearing as a single yellow signal, was seen in ISO-HAS. were removed 5 weeks after xenografting, fixed in 10% buffered These data confirmed the presence of the NUP160–SLC43A3 formalin, embedded in paraffin, and sliced into sections of 5 mm fusion, which was observed in 244/300 ISO-HAS (81.3%). in thickness. Increased single red/green signals were also observed in ISO-HAS, indicating their aneuploidy. G-banding showed the presence of Statistical analysis aneuploid nuclei in many chromosomes, including Ch11, of ISO- Data presented as bar graphs are the means þ SD from at least HAS (Fig. 1D). three experiments. Values of P < 0.05 were considered to be To determine whether the NUP160–SLC43A3 fusion was recur- statistically significant. rent, we performed fusion-specific RT-PCR using RNA obtained Other methods are described in Supplementary Materials and from paraffin sections of human tumors. We screened a series of Methods. 25 primary angiosarcomas on the scalp (Supplementary Table S5), and detected the fusion gene in nine cases: For example, Results amplification by fusion gene–specific primer pair was observed in the RNA samples from patient nos. 16, 17, and 19 (Fig. 2A), which Identification of angiosarcoma-specific fusion gene were all confirmed by Sanger sequencing. On the other hand, In gene-expression study by transcriptome sequencing, 94 patient no. 14 had the fusion gene in the tumor tissue, but not in genes were estimated to be significantly upregulated and 237 the peripheral blood cells or serum (Fig. 2B), indicating that the genes were estimated to be significantly downregulated in ISO- rearrangement was somatically acquired. NUP160–SLC43A3 was HAS compared with HDMEC (Supplementary Table S1). These not found in normal skin samples or in other tumor tissue included many angiogenesis-related genes (Supplementary Table specimens (Supplementary Table S5). S2). Several changes in the expression of angiogenesis-related The clinical and histopathologic characteristics were basically genes seen in ISO-HAS by the present study (e.g., downregulation similar between the patients with and without the fusion gene; of ITGB3, CTGF, and CLDN1) were compatible with the findings both poorly (Fig. 2C) and well-differentiated lesions (Fig. 2D) were of previous reports (8–10). seen in patients with the fusion. On the other hand, the duration of Next, we determined the presence of ISO-HAS–specific point the disease (between symptom onset and the first hospital visit) mutations in five angiogenesis-related genes and 21 oncogenes was significantly shorter in patients with the fusion gene than in selected on the basis of previous studies (3, 4). Among six putative those without (2.9 vs. 9.1 months, P ¼ 0.028 by Mann–Whitney angiosarcoma-specific mutations identified by bioinformatic U test), suggesting that the patients with the fusion gene had a analysis (Supplementary Table S3), the point mutation of TP53, more rapid progression and/or more severe subjective symp- a T-to-A transition in exon7 at codon 240, was already described in toms (Supplementary Table S5). In additional, at the first visit, ISO-HAS (1). On the other hand, the T-to-C mutation found in metastasis was less frequently detected in patients with the fusion PLCG1 of ISO-HAS was different from previously reported PLCG1 than in those without (0% vs. 37.5%, P ¼ 0.045 by Fisher's mutation (4). exact probability test), which also indicated that patients with the Putative fusion transcripts were predicted by TopHat-Fusion fusion gene visited the hospital earlier (before metastasis). (Supplementary Table S4). Among the 14 candidates, 35 spanning reads and five spanning mate pairs indicated the in-frame fusion between exon26 of NUP160 and exon7 of SLC43A3, which Functional analysis of the fusion gene are located at 11p11.2 and 11q12.1, respectively (Fig. 1A). The To clarify the function of the fusion gene, it was amplified by fusion gene was not found in HDMEC. Specific primers designed PCR and cloned into a lentiviral vector. HDMECs were stably for NUP160 exon26 and SLC43A3 exon7 were used for RT-PCR transfected with the control vector or fusion gene, full-length experiments to validate the finding (Fig. 1B). The expected frag- NUP160 or full-length SLC43A3. To note, the cells transfected ment was amplified by PCR using RNA from ISO-HAS, but not with full-length NUP160 did not grow, and exhibited inhibited RNA from HDMEC, cultured human keratinocytes or dermal cell proliferation. HDMECs transfected with the control vector, fibroblasts. Sanger sequencing of the amplified fragments con- fusion or SLC43A3, and ISO-HAS showed no apparent differences firmed the in-frame fusion between NUP160 and SLC43A3 (Fig. in cell shape (Fig. 3A). 1A). The gene-expression analysis indicated that levels of NUP160 We determined the effects of the fusion gene on the gene ex- and SLC43A3 were not significantly altered in ISO-HAS compared pression using a PCR array (Supplementary Table S6: the complete

(Continued.)TheGAPDH levels were shown as a control. Marker, 100-bp ladder. Arrow, PCR product of the fusion gene primers (240-bp; E). The cells were counted as described in Materials and Methods. , P ¼ 0.049 in comparison to the value in the cells transfected with control siRNA by the Mann–Whitney Utest(n ¼ 3; F). G, NIH3T3 cells transfected with the control vector or fusion gene (1 105 cells) were s.c. injected into mice (n ¼ 3). After 5 weeks, the cells overexpressing the fusion gene (right) showed tumor formation. H, a hematoxylin and eosin staining of the 5-week-old implants of NIH3T3 cells transfected with the fusion gene, showing a mass of pleomorphic and hyperchromatic tumor cells. Left, low magnification; scale bars, 500 mm; right, high magnification; scale bars, 20 mm. I, the RT-PCR detection of specific 240-bp fragments in RNAs from the 5-week-old implants of NIH3T3 cells transfected with the fusion gene (n ¼ 2) and cultured cells from the implants (n ¼ 2). Arrow, PCR product of the fusion gene–specific primers (240-bp). Marker, 100-bp ladder. NC, PCR-negative control.

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 4463

dataset is available at www.ncbi.nlm.nih.gov/geo/, GSE69725). lioid hemangioendotheliomas (WWTR1-CAMTA1 and YAP1- The transfection of the fusion increased or decreased the expression TFE3), and primary cutaneous hidradenomas (TORC1-MAML2, of several angiogenesis-related genes. Twelve genes were included CRTC1-MAML2, MECT1-MAML2, EWSR1-POU5F1, and EWS- in both the transcriptome analysis (Supplementary Table S2) and Oct-4B; refs. 13–15). Because COL1A1-PDGFB is highly specific PCR array (Supplementary Table S6). Among these, the up- or to dermatofibrosarcoma protuberans, its detection is clinically downregulation of nine genes was common to both experiments. useful for the diagnosis. Furthermore, EML4--ALK-driven lung Thus, there is a possibility that the gene-expression pattern of carcinoma shows dramatic response to ALK inhibitors. Thus, the HDMEC stably transfected with the fusion gene mimicked that of identification of fusion genes in cancers is also important for ISO-HAS. We expected that cells transfected with the fusion gene the treatments. Although the detection of fusion genes, includ- would obtain an "angiosarcoma-like" phenotype. ing EWSR1–ATF1 or CEP85L–ROS1, has also been reported in We then examined the phosphorylation levels of the angiosarcoma (16), these were seen in single cases, and were not NUP160–SLC43A3 protein, because NUP214–ABL1 fusion specific to angiosarcoma. Furthermore, their functions were not reportedly causes its autophosphorylation. Full-length NUP160 determined. and SLC43A3 have a serine phosphorylation site in exon29 and On the basis of the orientation of the genes and the appear- a tyrosine phosphorylation site in exon9, respectively (Fig. 3B). ance of Ch11, we suppose that interchromosomal or inter- NUP160–SLC43A3 preserves the SLC43A3 phosphorylation chromatid rearrangement may cause the NUP160–SLC43A3 site, but loses the NUP160 site. Immunoprecipitation revealed fusion. Fusion genes involving the NUP or SLC family have that all cell types, including HDMEC, ISO-HAS, HDMEC trans- been reported in several tumors. NUP160 is one of the con- fected with the control vector and HDMEC transfected with the stituent nucleoporins of NUP107-160, the largest subunit of fusion gene expressed the full-length SLC43A3 protein (Fig. 3C). nuclear pore complex (NPC). The NPC is the regulator of ISO-HAS and HDMEC transfected with fusion gene expressed molecular bidirectional traffic between the cell nucleus and the fusion protein, but tyrosine phosphorylation of the fusion cytoplasm, and NUP107-160 is a subcomplex essential for was not detected in these cells. Taken together, these findings various processes, including interphase NPC assembly and indicate that the fusion protein is not likely to exhibit increased mitotic chromosome segregation. Although NUP160 itself is autophosphorylation. originally implicated in nucleoplasmic transport, the molecule The migration of ISO-HAS was almost identical to that of is also associated with hybrid inviability, female sterility, mor- HDMEC (Fig. 4A). On the other hand, although the cell growth phologic anomalies, and slow development in vivo (17). On the of ISO-HAS and HDMEC was similar until passage 14, the growth other hand, SLC43A3 was originally predicted as a transmem- of HDMEC was significantly reduced after passage 18 (Fig. 4B). brane protein that transports nutrients or metabolites in rapidly Considering the potential role of mitogenesis in the development growing and/or developing tissues (18). However, its functions of angiosarcomas described previously (11), we hypothesized have not been confirmed in animal models. Recent articles that the sustained cell growth of ISO-HAS plays a role in its indicate SLC43A3 as a candidate gene associated with micro- malignant phenotype, and therefore focused on cell proliferation. vascularization, systemic inflammation links to atherosclerosis Stable overexpression of the fusion gene in HDMEC at higher and thyroid cancers (19, 20). Taken together, although the passages led to the significant recovery of cell numbers and the normal functions and pathways of NUP160 and SLC43A3 are BrdUrd incorporation (Fig. 4C), whereas full-length SLC43A3 still unclear, they can be involved in the mitosis, vascular overexpression did not. Thus, the fusion gene may contribute to formation, and carcinogenesis, which may contribute to the sustained cell proliferation. Of note, truncated NUP160 (exons1- pathogenesis of angiosarcoma. Our results suggest that full- 26) could also increase the number of HDMEC (Fig. 4D). Taken length NUP160 inhibits cell proliferation, and the truncation of together, our results suggest that NUP160 truncation, and not the NUP160 stimulates it. Thus, lack of phosphorylation residue phosphorylation of NUP160–SLC43A3, is the cause of the sus- due to the NUP160 truncation in the fusion gene may be the tained cell proliferation of ISO-HAS. On the other hand, NUP160 cause of sustained cell proliferation of ISO-HAS, at least partly. siRNA inhibited the expression of the fusion gene (Fig. 4E), as well Patients with NUP160–SLC43A3 did not have the previously as the expression of full-length NUP160 (not shown) in ISO-HAS, reported mutations in KDR or TP53 (Supplementary Table S5; and significantly decreased the cell number (Fig. 4F), indicating refs. 1, 3), suggesting the possibility that the fusion may occur therapeutic value of the siRNA. independently of KDR/TP53 mutations, that KDR/TP53 is an Finally, we performed a tumorigenicity assay using NIH3T3 alternate oncogenic mechanism for NUP160--SLC3A3 fusion, according to a previous study (12). At 5 weeks after subcutaneous and that these molecules may be associated with each other. implantation, NIH3T3 transfected with the fusion gene showed With regard to the clinical significance of the fusion gene, the tumor formation, whereas the cells with the control vector did not detection of NUP160–SLC43A3 by RT-PCR or variant assays (Fig. 4G). The histopathologic findings of the tumor tissue may be useful for the diagnosis of challenging cases or the showed a mass of pleomorphic and hyperchromatic cells (Fig. evaluation of surgical margins, especially in well-differentiated 4H), the features of which were similar to poorly differentiated angiosarcoma. Because the patients with tumors expressing the angiosarcoma (Fig. 2C). We confirmed that the tumor tissues, as fusion gene visited the hospital earlier than those without, and well as the cells cultured from them, were positive for the fusion thus did not have metastasis, immediate and intensive treat- gene by RT-PCR with fusion-specific primers (Fig. 4I). ments may improve their prognosis. Furthermore, NUP160 siRNA significantly decreased the cell number of ISO-HAS, NUP160 NUP160– Discussion indicating that full-length and/or the SLC43A3 fusion may represent potential therapeutic targets. Among skin tumors or vascular tumors, fusion genes are found Such specific treatments would be beneficial to avoid side in dermatofibrosarcoma protuberans (COL1A1-PDGFB), epithe- effects in elderly patients.

4464 Cancer Res; 75(21) November 1, 2015 Cancer Research

Disclosure of Potential Conﬂicts of Interest Writing, review, and/or revision of the manuscript: N. Shimozono, M. Jinnin, No potential conﬂicts of interest were disclosed. M. Masuzawa, M. Masuzawa, H. Ihn Administrative, technical, or material support (i.e., reporting or organizing Authors' Contributions data, constructing databases): M. Masuzawa, Z. Wang, A. Hirano, Y. Tomizawa, T. Etoh-Kira, I. Kajihara, M. Harada Conception and design: M. Jinnin, H. Ihn Study supervision: M. Masuzawa, H. Ihn Development of methodology: N. Shimozono, M. Jinnin, Z. Wang, A. Hirano, Y. Tomizawa, T. Etoh-Kira, M. Harada Acknowledgments Acquisition of data (provided animals, acquired and managed patients, The authors thank Dr. H. Miyoshi for kindly providing the lentiviral vector provided facilities, etc.): N. Shimozono, M. Jinnin, M. Masuzawa, Z. Wang, and Dr. A. Irie and C. Shiotsu for their valuable technical assistance. A. Hirano, Y. Tomizawa, T. Etoh-Kira, I. Kajihara, M. Harada Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): N. Shimozono, M. Jinnin, Z. Wang, I. Kajihara, Received February 13, 2015; revised June 27, 2015; accepted August 13, 2015; S. Fukushima, H. Ihn published online November 2, 2015.

References 1. Masuzawa M, Fujimura T, Hamada Y, Fujita Y, Hara H, Nishiyama S, et al. vivo and in vitro: differences between human and mouse. Toxicol Appl Establishment of a human hemangiosarcoma cell line (ISO-HAS). Int J Pharmacol 2009;237:83–90. Cancer 1999;81:305–8. 12. Soda M, Choi YL, Enomoto M, Takada S, Yamashita Y, Ishikawa S, et al. 2. Mandahl N, Jin YS, Heim S, Willen H, Wennerberg J, Biorklund€ A, et al. Identification of the transforming EML4–ALK fusion gene in non–small Trisomy 5 and loss of the Y chromosome as the sole cytogenetic anomalies cell lung cancer. Nature 2007;448:561–6. in a cavernous hemangioma/angiosarcoma. Genes Chromosomes Cancer 13. Tanas MR, Sboner A, Oliveira AM, Erickson-Johnson MR, Hespelt J, Han- 1990;1:315–6. wright PJ, et al. Identification of a disease-defining gene fusion in epithe- 3. Antonescu CR, Yoshida A, Guo T, Chang NE, Zhang L, Agaram NP, et al. lioid hemangioendothelioma. Sci Transl Med 2011;3:98ra82. KDR activating mutations in human angiosarcomas are sensitive to specific 14. Antonescu CR, LeLoarer F, Mosquera JM, Sboner A, Zhang L, Chen CL, et al. kinase inhibitors. Cancer Res 2009;69:7175–9. Novel YAP1-TFE3 fusion defines a distinct subset of epithelioid heman- 4. Behjati S, Tarpey PS, Sheldon H, Martincorena I, Van Loo P, Gundem G, gioendothelioma. Genes Chromosomes Cancer 2013;52:775–84. et al. Recurrent PTPRB and PLCG1 mutations in angiosarcoma. Nat Genet 15. Kim S, Lim B, Kim J. EWS-Oct-4B, an alternative EWS-Oct-4 fusion gene, is a 2014;46:376–9. potent oncogene linked to human epithelial tumours. Br J Cancer 2010; 5. Fujita M, Inoue M, Tanizawa O, Iwamoto S, Enomoto T. Alterations of the 102:436–46. p53 gene in human primary cervical carcinoma with and without human 16. Gru AA, Becker N, Pfeifer JD. Angiosarcoma of the parotid gland with a t papillomavirus infection. Cancer Res 1992;52:5323–8. (12;22) translocation creating a EWSR1-ATF1 fusion: a diagnostic dilem- 6. Jinnin M, Medici D, Park L, Limaye N, Liu Y, Boscolo E, et al. Suppressed ma. J Clin Pathol 2013;66:452–4. NFAT-dependent VEGFR1 expression and constitutive VEGFR2 signaling in 17. Maehara K, Murata T, Aoyama N, Matsuno K, Sawamura K. Genetic infantile hemangioma. Nat Med 2008;14:1236–46. dissection of Nucleoporin 160 (Nup160), a gene involved in multiple 7. Tahara-Hanaoka S, Sudo K, Ema H, Miyoshi H, Nakauchi H. Lentiviral phenotypes of reproductive isolation in Drosophila. Genes Genet Syst vector-mediated transduction of murine CD34(-) hematopoietic stem 2012;87:99–106. cells. Exp Hematol 2002;30:11–7. 18. Stuart RO, Pavlova A, Beier D, Li Z, Krijanovski Y, Nigam SK. EEG1, a 8. Mechtersheimer G, Barth T, Hartschuh W, Lehnert T, Moller€ P. In situ putative transporter expressed during epithelial organogenesis: compari- expression of beta 1, beta 3 and beta 4 integrin subunits in non- son with embryonic transporter expression during nephrogenesis. Am J neoplastic endothelium and vascular tumours. Virchows Arch 1994; Physiol Renal Physiol 2001;281:F1148–56. 425:375–84. 19. Sivapalaratnam S, Farrugia R, Nieuwdorp M, Langford CF, van Beem RT, 9. Igarashi A, Hayashi N, Nashiro K, Takehara K. Differential expression of Maiwald S, et al. Identification of candidate genes linking systemic inflam- connective tissue growth factor gene in cutaneous fibrohistiocytic and mation to atherosclerosis; results of a human in vivo LPS infusion study. vascular tumors. J Cutan Pathol 1998;25:143–8. BMC Med Genomics 2011;4:64. 10. Gyorffy H. Study of claudins and prognostic factors in some gastrointes- 20. Abend M, Pfeiffer RM, Ruf C, Hatch M, Bogdanova TI, Tronko MD, et al. tinal diseases. Magy Onkol 2009;53:377–83. Iodine-131 dose dependent gene expression in thyroid cancers and corre- 11. Kakiuchi-Kiyota S, Vetro JA, Suzuki S, Varney ML, Han HY, Nascimento M, sponding normal tissues following the Chernobyl accident. PLoS ONE et al. Effects of the PPARgamma agonist troglitazone on endothelial cells in 2012;7:e39103.

www.aacrjournals.org Cancer Res; 75(21) November 1, 2015 4465

Naoki Shimozono, Masatoshi Jinnin, Mamiko Masuzawa, et al.

Cancer Res 2015;75:4458-4465.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/75/21/4458

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2015/11/10/75.21.4458.DC1

Cited articles This article cites 20 articles, 4 of which you can access for free at: http://cancerres.aacrjournals.org/content/75/21/4458.full#ref-list-1

Citing articles This article has been cited by 8 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/75/21/4458.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/75/21/4458. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.