Gene expression profiling identifies emerging oncogenic pathways operating in extranodal NK/T-cell lymphoma, nasal type

Abstract

Biopsies and cell lines of natural killer/T-cell lymphoma, nasal type (NKTCL) were subject to combined gene expression profiling and array-based comparative genomic hybridization analyses. Compared with peripheral T-cell lymphoma, not otherwise specified, NKTCL had greater transcript levels for NK-cell and cytotoxic molecules, especially granzyme H. Compared with normal NKcells, tumors were closer to activated than resting cells and overexpressed several genes related to vascular biology, Epstein-Barr Virus-induced genes, and PDGFRA. Notably, platelet-derived growth factor receptor alpha and its phosphorylated form were confirmed at the protein level, and in vitro the MEC04 NKTCL cell line was sensitive to imatinib. Deregulation of the AKT, Janus kinase-signal transducers and activators of transcription, and nuclear factor-kappaB pathways was corroborated by nuclear expression of phosphorylated AKT, signal transducers and activators of transcription 3, and RelA in NKTCL, and several deregulated genes in these pathways mapped to regions of recurrent copy number aberrations (AKT3 [1q44], IL6R [1q21.3], CCL2 [17q12], TNFRSF21 [6p12.3]). Several features of NKTCL uncovered by this analysis suggest perturbation of angiogenic pathways. Integrative analysis also evidenced deregulation of the tumor suppressor HACE1 in the frequently deleted 6q21 region. This study highlights emerging oncogenic pathways in NKTCL and identifies novel diagnostic and therapeutic targets.

Figure 1

Validation of gene expression profiling by immunohistochemistry. Representative NKTCLs disclosed cytoplasmic staining of neoplastic cells for (A) gzm H, (B) EBI3, (C) VEGFA, nuclear localization of (D) pSTAT3, (E) pAKT, cytoplasmic and nuclear staining of (F) RelA, and positivity for (G) PDGFRα and (H) pPDGFRα. Images were captured with a Zeiss Axioskop2 microscope (Zeiss). Photographs were taken with an Olympus DP70 camera. Image acquisition was performed with Olympus DP Controller 2002, and images were processed with Adobe Photoshop Version 7.0 (Adobe Systems). Original magnification, ×400 (A-H).

Figure 2

Unsupervised clustering of 7 NKTCLs, 2 NKTCL-derived cell lines, and 16 PTCLs, NOS. Dendrogram of 23 lymphoma tissues and 2 cell lines based on principal component analysis demonstrated 2 major clusters of NKTCL including tumor-derived cell lines and PTCL, NOS, respectively. T1-T7: 7 NKTCL samples, SNK6, SNT8; C1-16: 16 PTCL, NOS samples.

Figure 3

Genomic profiles of NKTCL. Pangenomic view of 8 NKTCL tissue samples. The horizontal axis represents the genomic order, and the vertical axis represents the number of samples with gains or losses. Gains are represented as red bars and losses are represented as green bars.

Figure 4

Quantification of selected genes by qRT-PCR analysis. qRT-PCR analysis for (A) MET, (B) TNFRSF21, (C) CCND3, (D) CCL2, and (E) PRDM1, ATG5, AIM1, and HACE1 in NKTCL primary tumors and cell lines. The results are expressed as relative fold change compared with resting CD3⁻/CD56⁺ NK cells sorted from peripheral blood. Quantifications were performed in duplicate, and mean values and SD were calculated for each transcript. (A-D) In agreement with the microarray results, mRNA levels of MET, CCL2, and TNFRSF21 are increased in NKTCL primary tumors compared with normal NK cells, whereas CCND3 mRNA is reduced (E). The analysis of 4 putative tumor suppressors in 6q21 region show that the transcripts levels of HACE1, AIM1, and ATG5 are reduced in primary tumors and cell lines whereas the mRNA level of PRDM1 is increased in primary tumors, due to a wide variation from case to case (not shown).

Figure 5

Cellular programs deregulated in NKTCL. Representative molecular pathways differentially expressed in NKTCLs by comparison with either normal NK cells (JAK-STAT in panel A, angiogenesis in panel B, apoptosis in panel E), normal B cells and ABC-DLBCL (NF-κB in panel C), normal NK cells and normal B cells (AKT in panel D), or PTCL, NOS (cytotoxicity in panel F) were illustrated. For each line, green corresponds to the minimal intensity value (min), red corresponds to the maximal intensity value (max), and black corresponds to (min + max)/2. Several genes related to JAK-STAT, angiogenesis, and apoptosis-related pathways are overexpressed in NKTCLs compared with normal NK cells, and genes related to the cytotoxicity-related pathway also are overexpressed in NKTCLs compared with PTCL, NOS. Many genes in the NF-κB and AKT signaling pathways are overexpressed in NKTCLs compared with normal B cells, with a molecular signature similar to that of ABC-DLBCL in NK-κB pathway. The expression data of PDGFRA in NKTCLs, 2 NKTCL-derived cell lines, and normal NK cells is highlighted. T1-7: 7 NKTCL samples; C1-16: 16 PTCL, NOS samples; AcNK, AcNK02, AcNK08, AcNK24: activated NK cells; ReNK, ReNK: resting NK cells; CB.1-8 and CB.11: 9 centroblasts samples; CC.1-8 and CC.11: 9 centrocytes samples; and ABC.2024-2195: 15 ABC-DLBCL samples.

Figure 6

Proliferation of NKTCL-derived cell lines in the presence of imatinib mesylate. SNK6, MEC04, and U937 cells were incubated for 72 hours in media with or without imatinib mesylate at concentrations of 1, 3, and 6 μM. Imatinib induced concentration-dependent growth inhibition of MEC04 cells. The effect on SNK6 was not significantly different from that observed on the myelomonocytic U937 cell line. The horizontal axis is the concentration of imatinib mesylate. The vertical axis is the percentage of control proliferation. Bars indicate the SEM of triplicate experiments.

Figure 7

Hypothetical representation of signaling pathways involved in NKTCL. The interactions of PDGF, AKT, and JAK-STAT signaling pathways may contribute to the angiogenesis, immunosuppression, proliferation, and survival of NKTCL.