The human family of ETS transcription factors numbers 28 genes which control multiple aspects of development, notably the differentiation of blood and immune cells. Otherwise, aberrant expression of ETS genes is reportedly involved in forming leukemia and lymphoma. Here, we comprehensively mapped ETS gene activities in early hematopoiesis, lymphopoiesis and all mature types of lymphocytes using public datasets. We have termed the generated gene expression pattern lymphoid ETS-code. This code enabled identification of deregulated ETS genes in patients with lymphoid malignancies, revealing 12 aberrantly expressed members in Hodgkin lymphoma (HL). For one of these, ETS gene ETV3, expression in stem and progenitor cells in addition to that in developing and mature T-cells was mapped together with downregulation in B-cell differentiation. In contrast, subsets of HL patients aberrantly overexpressed ETV3, indicating oncogenic activity in this B-cell malignancy. Analysis of ETV3-overexpressing HL cell line SUP-HD1 demonstrated genomic duplication of the ETV3 locus at 1q23, GATA3 as mutual activator, and suppressed BMP-signalling as mutual downstream effect. Additional examination of the neighboring ETS genes ETS1 and FLI1 revealed physiological activities in B-cell development and aberrant downregulation in HL patient subsets. SUP-HD1 showed genomic loss on chromosome 11, del(11)(q22q25), targeting both ETS1 and FLI1, underlying their downregulation. Furthermore, in the same cell line we identified PBX1-mediated overexpression of RIOK2 which inhibited ETS1 and activated JAK2 expression. Collectively, we codified normal ETS gene activities in lymphopoiesis and identified oncogenic ETS members in HL.

NKL homeobox genes encode developmental transcription factors regulating basic processes in cell differentiation. According to their physiological expression pattern in early hematopoiesis and lymphopoiesis, particular members of this homeobox gene subclass constitute an NKL-code. B-cell specific NKL-code genes generate a regulatory network and their deregulation is implicated in B-cell lymphomagenesis. Epstein-Barr virus (EBV) infects B-cells and influences the activity of signalling pathways including JAK/STAT and several genes encoding developmental regulators. Therefore, EBV-infection impacts the pathogenesis and the outcome of B-cell malignancies including Hodgkin lymphoma and diffuse large B-cell lymphoma (DLBCL). Here, we isolated EBV-positive and EBV-negative subclones from the DLBCL derived cell line DOHH-2. These subclones served as models to investigate the role of EBV in deregulation of the B-cell specific NKL-code members HHEX, HLX, MSX1 and NKX6-3. We showed that the EBV-encoded factors LMP1 and LMP2A activated the expression of HLX via STAT3. HLX in turn repressed NKX6-3, SPIB and IL4R which normally mediate plasma cell differentiation. In addition, HLX repressed the pro-apoptotic factor BCL2L11/BIM and hence supported cell survival. Thus, EBV aberrantly activated HLX in DLBCL, thereby disturbing both B-cell differentiation and apoptosis. The results of our study appreciate the pathogenic role of EBV in NKL homeobox gene deregulation and B-cell malignancies.

NKL homeobox genes encode basic transcriptional regulators of cell and tissue differentiation. Recently, we described a hematopoietic NKL-code comprising nine specific NKL homeobox genes expressed in normal hematopoietic stem cells, lymphoid progenitors and during lymphopoiesis, highlighting their physiological role in the development of T-, B- and NK-cells. Here, we identified aberrant expression of the non-hematopoietic neural NKL homeobox gene NKX2-2 in about 12% of both, classical Hodgkin lymphoma (HL) and nodular lymphocyte predominant (NLP) HL patients. The NKX2-2 expressing NLPHL-derived cell line DEV served as a model by analysing chromosomal configurations and expression profiling data to reveal activating mechanisms and downstream targets of this developmental regulator. While excluding chromosomal rearrangements at the locus of NKX2-2 we identified t(3;14)(p21;q32) resulting in overexpression of the IL17 receptor gene IL17RB via juxtaposition with the IGH-locus. SiRNA-mediated knockdown experiments demonstrated that IL17RB activated NKX2-2 transcription. Overexpression of IL17RB-cofactor DAZAP2 via chromosomal gain of 12q13 and deletion of its proteasomal inhibitor SMURF2 at 17q24 supported expression of NKX2-2. IL17RB activated transcription factors FLI1 and FOXG1 which in turn mediated NKX2-2 expression. In addition, overexpressed chromatin-modulator AUTS2 contributed to NKX2-2 activation as well. Downstream analyses indicated that NKX2-2 inhibits transcription of lymphoid NKL homeobox gene MSX1 and activates expression of basic helix-loop-helix factor NEUROD1 which may disturb B-cell differentiation processes via reported interaction with TCF3/E2A. Taken together, our data reveal ectopic activation of a neural gene network in HL placing NKX2-2 at its hub, highlighting a novel oncogenic impact of NKL homeobox genes in B-cell malignancies.

Recently, we have presented a scheme, termed "NKL-code", which describes physiological expression patterns of NKL homeobox genes in early hematopoiesis and in lymphopoiesis including main stages of T-, B- and NK-cell development. Aberrant activity of these genes underlies the generation of hematological malignancies notably T-cell leukemia. Here, we searched for deregulated NKL homeobox genes in main entities of T-cell lymphomas comprising angioimmunoblastic T-cell lymphoma (AITL), anaplastic large cell lymphoma (ALCL), adult T-cell leukemia/lymphoma (ATLL), hepatosplenic T-cell lymphoma (HSTL), NK/T-cell lymphoma (NKTL) and peripheral T-cell lymphoma (PTCL). Our data revealed altogether 19 aberrantly overexpressed genes in these types, demonstrating deregulated NKL homeobox genes involvement in T-cell lymphomas as well. For detailed analysis we focused on NKL homeobox gene MSX1 which is normally expressed in NK-cells. MSX1 was overexpressed in subsets of HSTL patients and HSTL-derived sister cell lines DERL-2 and DERL-7 which served as models to characterize mechanisms of deregulation. We performed karyotyping, genomic and expression profiling, and whole genome sequencing to reveal mutated and deregulated gene candidates, including the fusion gene CD53-PDGFRB. Subsequent knockdown experiments allowed the reconstruction of an aberrant network involved in MSX1 deregulation, including chromatin factors AUTS2 and mutated histone HIST1H3B(K27M). The gene encoding AUTS2 is located at chromosome 7q11 and may represent a basic target of the HSTL hallmark aberration i(7q). Taken together, our findings highlight an oncogenic role for deregulated NKL homeobox genes in T-cell lymphoma and identify MSX1 as a novel player in HSTL, implicated in aberrant NK- and T-cell differentiation.

Recently, we identified deregulated expression of the B-cell specific transcription factor MEF2C in T-cell acute lymphoid leukemia (T-ALL). Here, we performed sequence analysis of a regulatory upstream section of MEF2C in T-ALL cell lines which, however, proved devoid of mutations. Unexpectedly, we found strong conservation between the regulatory upstream region of MEF2C (located at chromosomal band 5q14) and an intergenic stretch at 7q11 located between STAG3L4 and AUTS2, covering nearly 20 kb. While the non-coding gene STAG3L4 was inconspicuously expressed, AUTS2 was aberrantly upregulated in 6% of T-ALL patients (public dataset GSE42038) and in 3/24 T-ALL cell lines, two of which represented very immature differentiation stages. AUTS2 expression was higher in normal B-cells than in T-cells, indicating lineage-specific activity in lymphopoiesis. While excluding chromosomal aberrations, examinations of AUTS2 transcriptional regulation in T-ALL cells revealed activation by IL7-IL7R-STAT5-signalling and MEF2C. AUTS2 protein has been shown to interact with polycomb repressor complex 1 subtype 5 (PRC1.5), transforming this particular complex into an activator. Accordingly, expression profiling and functional analyses demonstrated that AUTS2 activated while PCGF5 repressed transcription of NKL homeobox gene MSX1 in T-ALL cells. Forced expression and pharmacological inhibition of EZH2 in addition to H3K27me3 analysis indicated that PRC2 repressed MSX1 as well. Taken together, we found that AUTS2 and MEF2C, despite lying on different chromosomes, share strikingly similar regulatory upstream regions and aberrant expression in T-ALL subsets. Our data implicate chromatin complexes PRC1/AUTS2 and PRC2 in a gene network in T-ALL regulating early lymphoid differentiation.

Recently, we have documented a hematopoietic NKL-code mapping physiological expression patterns of NKL homeobox genes in early hematopoiesis and in lymphopoiesis, which spotlights genes deregulated in lymphoid malignancies. Here, we enlarge this map to include normal NKL homeobox gene expressions in myelopoiesis by analyzing public expression profiling data and primary samples from developing and mature myeloid cells. We thus uncovered differential activities of six NKL homeobox genes, namely DLX2, HHEX, HLX, HMX1, NKX3-1 and VENTX. We further examined public expression profiling data of 251 acute myeloid leukemia (AML) and 183 myelodysplastic syndrome (MDS) patients, thereby identifying 24 deregulated genes. These results revealed frequent deregulation of NKL homeobox genes in myeloid malignancies. For detailed analysis we focused on NKL homeobox gene NANOG, which acts as a stem cell factor and is correspondingly expressed alone in hematopoietic progenitor cells. We detected aberrant expression of NANOG in a small subset of AML patients and in AML cell line NOMO-1, which served as a model. Karyotyping and genomic profiling discounted rearrangements of the NANOG locus at 12p13. But gene expression analyses of AML patients and AML cell lines after knockdown and overexpression of NANOG revealed regulators and target genes. Accordingly, NKL homeobox genes HHEX, DLX5 and DLX6, stem cell factors STAT3 and TET2, and the NOTCH-pathway were located upstream of NANOG while NKL homeobox genes HLX and VENTX, transcription factors KLF4 and MYB, and anti-apoptosis-factor MIR17HG represented target genes. In conclusion, we have extended the NKL-code to the myeloid lineage and thus identified several NKL homeobox genes deregulated in AML and MDS. These data indicate a common oncogenic role of NKL homeobox genes in both lymphoid and myeloid malignancies. For misexpressed NANOG we identified an aberrant regulatory network, which contributes to the understanding of the oncogenic activity of NKL homeobox genes.

T-cell acute lymphoblastic leukemia (T-ALL) cells represent developmentally arrested T-cell progenitors, subsets of which aberrantly express homeobox genes of the NKL subclass, including TLX1, TLX3, NKX2-1, NKX2-5, NKX3-1 and MSX1. Here, we analyzed the transcriptional landscape of all 48 members of the NKL homeobox gene subclass in CD34+ hematopoietic stem and progenitor cells (HSPCs) and during lymphopoiesis, identifying activities of nine particular genes. Four of these were expressed in HSPCs (HHEX, HLX1, NKX2-3 and NKX3-1) and three in common lymphoid progenitors (HHEX, HLX1 and MSX1). Interestingly, our data indicated downregulation of NKL homeobox gene transcripts in late progenitors and mature T-cells, a phenomenon which might explain the oncogenic impact of this group of genes in T-ALL. Using MSX1-expressing T-ALL cell lines as models, we showed that HHEX activates while HLX1, NKX2-3 and NKX3-1 repress MSX1 transcription, demonstrating the mutual regulation and differential activities of these homeobox genes. Analysis of a public T-ALL expression profiling data set comprising 117 patient samples identified 20 aberrantly activated members of the NKL subclass, extending the number of known NKL homeobox oncogene candidates. While 7/20 genes were also active during hematopoiesis, the remaining 13 showed ectopic expression. Finally, comparative analyses of T-ALL patient and cell line profiling data of NKL-positive and NKL-negative samples indicated absence of shared target genes but instead highlighted deregulation of apoptosis as common oncogenic effect. Taken together, we present a comprehensive survey of NKL homeobox genes in early hematopoiesis, T-cell development and T-ALL, showing that these genes generate an NKL-code for the diverse stages of lymphoid development which might be fundamental for regular differentiation.

Homeobox genes encode transcription factors which regulate basic processes in development and cell differentiation and are grouped into classes and subclasses according to sequence similarities. Here, we analyzed the activities of the 20 members strong TALE homeobox gene class in early hematopoiesis and in lymphopoiesis including developing and mature B-cells, T-cells, natural killer (NK)-cells and innate lymphoid cells (ILC). The resultant expression pattern comprised eleven genes and which we termed TALE-code enables discrimination of normal and aberrant activities of TALE homeobox genes in lymphoid malignancies. Subsequent expression analysis of TALE homeobox genes in public datasets of Hodgkin lymphoma (HL) patients revealed overexpression of IRX3, IRX4, MEIS1, MEIS3, PBX1, PBX4 and TGIF1. As paradigm we focused on PBX1 which was deregulated in about 17% HL patients. Normal PBX1 expression was restricted to hematopoietic stem cells and progenitors of T-cells and ILCs but absent in B-cells, reflecting its roles in stemness and early differentiation. HL cell line SUP-HD1 expressed enhanced PBX1 levels and served as an in vitro model to identify upstream regulators and downstream targets in this malignancy. Genomic studies of this cell line therein showed a gain of the PBX1 locus at 1q23 which may underlie its aberrant expression. Comparative expression profiling analyses of HL patients and cell lines followed by knockdown experiments revealed NFIB and TLX2 as target genes activated by PBX1. HOX proteins operate as cofactors of PBX1. Accordingly, our data showed that HOXB9 overexpressed in HL coactivated TLX2 but not NFIB while activating TNFRSF9 without PBX1. Further downstream analyses showed that TLX2 activated TBX15 which operated anti-apoptotically. Taken together, we discovered a lymphoid TALE-code and identified an aberrant network around deregulated TALE homeobox gene PBX1 which may disturb B-cell differentiation in HL by reactivation of progenitor-specific genes. These findings may provide the framework for future studies to exploit possible vulnerabilities of malignant cells in therapeutic scenarios.