CTCF, a novel fusion partner of ETO2 in a post-transplant relapsed acute myeloid leukemia patient

Background ETO2 is a nuclear co-repressor, which plays a critical role in the regulation of the cell cycle, self-renewal capacity, and differentiation of hematopoietic progenitor cells. Methods We identied novel fusion transcripts involving ETO2 and CTCF by RNA-seq in a post-transplant relapsed case. The novel may have prognostic and pathogenic in


Plasmid constructs
The DNA sequence encoding CTCF-ETO2 and reciprocal ETO2-CTCF were synthesized and inserted into both pcDNA3.1 expression vector (Genewiz) with Flag tag fused to their carboxy termini and VENUS-N-FLAG lentiviral vector (Genewiz). Wild-type CTCF and ETO2 were also synthesized and inserted into pcDNA3.1 expression vector with Flag tag fused to their carboxy termini. The empty vector pcDNA3.1 and VENUS-N-FLAG were used as control vectors. All of the plasmids were purchased from Genewiz.

Lentivirus production and transduction
The 293T cells were transfected with lentiviral vector constructs and pPACK packaging plasmids mix by calcium phosphate precipitation method. Cells were then infected with prepared lentivirus and sorted by BD FACS Aria II System for green uorescent protein (GFP)-positive cells.
Antibody staining was monitored with a Novocyte ow cytometer. Data analysis was carried out using FlowJo software.

CCK8 assay
Transduced 32D cells were rinsed with PBS for 3 times and treated with either 10ng/ml or 0 mIL-3 in a 96well cell culture plate. Approximately 15000 cells were seeded in each well. Three replicates were made for each measurement. Then, cells were incubated at 37 °C in a humidi ed atmosphere with 5% CO2 for 24 h, 48 h and 72h. Finally, 10 μL of the CCK-8 reagent (Donjindo, KR675) was added into each well, and OD at 450 nm was measured using a multifunction microplate reader (In nite M200 Pro, Tecan) after incubation for 3 h at 37 °C. The fold each concentration accounted for of the control was presented as relative cell proliferation.

Western blot
Cell lysates were separated by SDS-PAGE gel and transferred to PVDF membrane (Millipore). The membrane was probed with primary antibody and then with secondary antibody. Antibody binding was revealed by using an enhanced chemiluminescence reagent (GE Healthcare Biosciences). ImageJ was used to quantify the density and size of the blots. Statistical analysis Data were shown as mean ± standard deviation (SD). The signi cance of differences between different groups was determined by ANOVA. Data analyses were performed using Graphpad Prism v6.0. Statistical signi cance threshold was set at 0.05; asterisks indicate signi cant differences (*P < .05;**P< .01; and ***P < .001).

Background
In acute myeloid leukemia (AML) patients, cytogenetics represents the single most important prognostic factor for predicting remission rates, relapse risks and overall survival outcomes. In order to better understand the cytogenetic abnormality in a post-transplant relapsed AML patient who had no aberrant results by routine karyotype analysis and multiplex RT-PCR, we performed RNA sequencing and identi ed novel CCCTC-binding factor (CTCF)-eight twenty-one 2 (ETO2) and its reciprocal fusion transcripts.

Case Report
A 19-year-old female was admitted to our hospital on 2 September 2014 due to skin ecchymosis for 4 days and fever for 1 day. Blood tests showed a white blood cell count of 132 × 10 9 /L, a platelet count of 17 × 10 9 /L, and a hemoglobin level of 88 g/L. Bone marrow aspiration showed AML-M2a with 27.5% BM blast (Fig. 1A). Leukemic blasts were positive for CD13, CD34, CD64, CD38 and CD34 with partial weak positivity of HLA-DR and CD45 by ow cytometry. No abnormal results were identi ed by karyotype (Fig. 1B) analysis and multiplex RT-PCR. Therefore, the patient was diagnosed with AML and treated with induction regimens (idarubicin and cytarabine), which yield a complete remission. She went on to receive post-remission intensi cation and subsequent maintenance chemotherapy for a total of 2 years of chemotherapy. During maintenance therapy, the patient experienced rst and second relapse, and underwent haploidentical transplants while in a non-complete remission status. CSF3R T618I , RUNX1 G64R and CEBPA R297P mutations (detected by next-generation sequencing) occurred at that time. In addition, a leukemia-in ltrating breast mass was removed by surgery on 6 September 2016. Unfortunately, on October 2017, she relapsed again and died within 1 month. Subsequent RNAseq on bone marrow samples collected before transplant identi ed novel CTCF-ETO2 and its reciprocal fusion genes. RNAseq results revealed 1 breakpoint in intron 5 of CTCF (ENST00000401394.6) and 1 breakpoint in intron 1 of ETO2 (ENST00000268679.9). Sanger sequencing of the RT-PCR products further con rmed in-frame fusions between CTCF (codon 125) and ETO2 (codon 51) in both chimeric transcripts (Fig. 1C, D). CTCF-ETO2 fusion transcript was formed by the fusion of CTCF exon 5 to ETO2 exon 2, simultaneously, the reciprocal ETO2-CTCF chimeric transcript was composed of the fusion of ETO2 exon 1 to CTCF exon 6. The main domains of ETO2 and CTCF were preserved in fusion proteins (Fig. 1E).
Immuno uorescence analysis of subcellular distributions demonstrated that both CTCF-ETO2 and ETO2-CTCF fusion proteins were localized at nuclear ( Fig. 2A), to execute their functions in promoting cell proliferation. Signi cantly cell growth was detected via cell counting kit-8 assay in CTCF-ETO2 transduced 32D cell lines compared with vector group (Fig. 2C). Strikingly, the improved proliferative strength still had statistical signi cance when retreating murine IL-3 in mIL-3 dependent 32D cells.
Similarly, Thirant et al found that ETO2-GLIS2 confers enhanced self-renewal of progenitor cells, and suggested that the nerve homology region 2 (NHR2) domain of ETO2 proteins in coordination with transcription factor ERG is essential for the self-renewal of ETO2-GLIS2 leukemia cells [1][2][3]. Besides, the activated signal pathway (eg, p-STAT3) and key molecules involving cell proliferation (c-myc) or cell cycle (CDK9, Cyclin T1) may partially account for this result (Fig. 2D). With regard to erythropoiesis, low erythroid-related gene expression in CTCF-ETO2 transduced cells was detected (Fig. 2C), which probably results from dysregulation of GATA1. Besides, neither CTCF-ETO2 nor ETO2-CTCF fusion gene had effect on cell apoptosis (Supplementary Fig), which is correspondent with bcl2 expression in immunoblotting (Fig. 2D).
CTCF is a transcription factor that contains a DNA-binding domain composed of 11 highly conserved zinc ngers (ZF). The nuclear protein is encoded by the CTCF gene located on chromosome 16q22.1. As a transcription repressor or insulator, CTCF negatively regulates MYC, thus providing a mechanism for CTCF to promote erythroid differentiation [4,5]. In addition, CTCF boundary remodels chromatin domain and drives aberrant HOX gene transcription in AML [6]. The present ETO2-CTCF fusion transcripts preserved ZF domain, consistent with previous study, down-regulated MYC protein and up-regulated HOXA9 protein was found in the western blot (Fig. 2D), and higher expression of erythroid gene was identi ed by Q-PCR (Fig. 2C).
Multiple relapses and extramedullary invasion, especially post-transplant relapse gave the patient fatal lethality in the present CTCF-ETO2 and ETO2-CTCF positive AML case. Alternatively, Schuback et al found that positive ETO2-GLIS2 fusion had signi cantly higher relapse rate, worse 5-year overall survival and event-free survival than negative fusion in 193 cytogenetically normal AML patients [14]. Micci et al described that three AML patients with NFIA-ETO2 fusion had poor clinical outcome [10]. In contrast, fouryear event free survival of RUNX1-ETO2 tended to be higher compared with other AML patients (77% vs 51%, P = 0.06) [15]. It seems that the synergetic role of ETO2 and its partner would determine patient-speci c outcome. Better understanding of molecular and clinical characteristics of novel CTCF-ETO2 fusion would be helpful for future treatment.

Conclusions
In conclusion, CTCF has been reported as a translocation partner of ETO2 in AML for the rst time. Both chimeric transcripts, CTCF-ETO2 and ETO2-CTCF, were located in the nuclei and could promote cell growth, but neither had effects on apoptosis. Further research is warranted to investigate functional characterization, prognostic value and potential therapy of this novel fusion in AML. This study was approved by the ethics committee in accordance with the Declaration of Helsinki protocol.

Consent for publication:
All authors give consent for the publication of the manuscript.
Availability of data and material: All data obtained and analyzed in this study were available from the corresponding authors in a reasonable request.  The subcellular localization and proliferative effect of fusion proteins. A. Immuno uorescence analysis of CTCF-ETO2 and ETO2-CTCF fusion protein in Hela cells transfected with expression plasmids. B. CTCF-ETO2 fusion protein played a role in down-regulating erythroid gene expression in 32D cells. Expression levels of ζ-globin, β1-globin, βh1-globin and -globin were assessed by qPCR and mCD71 (immature erythroid surface marker) was showed by ow cytometry. C. Both CTCF-ETO2 and ETO2-CTCF fusion genes promoted 32D cell proliferation. Relative proliferation of 32D cells treated by 0 or 10 ng/ml mIL-3 was detected via CCK-8 assay. D. Western blot analysis of 293T cells expressing pcDNA3.1-tagged vector and fusions. Protein levels of STAT3, phosphorylated (p-) STAT3, STAT5, p-STAT5, AKT, p-AKT, ERK, p-ERK, JNK, p-JNK, BCL2, c-MYC, HOXA9, P53, CDK9 and cyclinT1 were calculated by ImageJ.