Comparative Genomic Identification of Unique Signal Transduction Pathways and Targets In Pediatric Burkitt Lymphoma (PBL)

The prognosis of PBL has dramatically improved over the past 25 years (> 80% 5 yrs EFS) (Cairo et al BJH 2009). This success, however, has come at the cost of the high toxicity associated with intensive multi-agent chemotherapy. A need to identify less toxic targeted therapy exists. Our recent studies show that while almost 100% of childhood BL (238 children) is characterized by an 8q24 (c-myc) gene rearrangement, 70% of the patients also had a secondary chromosomal aberration (Cairo et al Blood 2007), and those children with BL who had a 13q abnormality had a significantly poorer outcome (20% reduction in EFS; Poirel/Cairo et al Leukemia, 2009). Loss of 13q14.3 was associated with a significant decrease in OS (Nelson/Cairo/Perkins/Sanger et al BJH 2010). DLEU1, proven as a Burkitt specific classifier and a cMYC target gene (Dave/Staudt et al NEJM 2006), is located within the region of 13q14.3. DLEU1 is recognized to interact with c-Myc, TUBB2C, UBR1, and other proteins. When DLEU1 is down-regulated by a DLEU1 siRNA, the spontaneous apoptotic rate was decreased with concomitant significantly reduced levels of UBR1 and TUBB2C gene expression (Day/Cairo et al SIOP 2008). Our hypothesis is that deletion of 13q14.3, which contains DLEU1, in pediatric BL may confer a phenotype of chemoimmunotherapy resistance.

To identify potential drug targets using comparative genomic microarray analysis of specimens in patients with PBL.

PBL data from three different research groups were used; i.e. COG ANHL01P1 by our group (n=11), NCBI's GEO GSE10172 and GSE4475 (Klapper et al) (n=16), and GSE4732 (Dave/Staudt et al) (n=15). For comparison at equal weighted level, ANHL01P1 samples were validated by building a prediction model with Support Vector Machines using Klapper PBL as training database. RNAs were subjected to microarray studies (Affymetrix U133A_2) and analyzed by Agilent GeneSpring or Partek. Functional interpretation of the identified PBL genes was analyzed by Ingenuity Pathways Analysis. One-way ANOVA followed by Tukey test was used. To test our hypothesis, we knocked down DLEU1 and investigated cytoxan (CY) and rituximab effect on apoptotic rate of Ramos. Ramos BL cell lines were transiently transfected (24 hrs) with DLEU1 siRNA (5′-AUACUUGGCAUGAAUGAACUUAUGU-3′ and 3′-UAUGAACCGUACUUACUUGAAUACA-5′) (Day/Cairo SIOP 2008). The siRNA transfected cells were then treated with CY (0, 89.5, 895, 8950 nM) or rituximab (0, 4, 40, 400 mg/mL) for additional 4 hrs. Cells were evaluated for percent apoptosis using Annexin V-FITC and Propidium Iodide followed by FACS using BD LSRII.

1565 genes were identified (p<0.05), among which 376 genes showed no significant difference among three groups. High expression of cMYC (27F) and DLEU1 (9.9F) were detected. Pathway analysis indicates that these genes areinvolved in Toll-like receptor signaling (16 genes; p<0.01), including IRAK1 (22.9F), IFNAR2 (5.4F), NFKBIA (15.3F), and STAT1 (9.2F); JAK-STAT signaling (16 genes; p<0.01), including PTPN11 (25.6F), PTPN6 (8.8F), PIM1 (6.4F), and IL21R (2.9F); and MAPK signaling (19 genes; p<0.01) including MAP2K1 (11.8F), MAP3K7 (6.3F), MAPK9 (6.9F), and RAF1 (9.2F). There was a significant reduction in apoptosis in the CY-treated BL transfected DLEU1 siRNA vs mock control cells (89.5 nM CY: 10.26+0.23% reduction, p<0.05 to negative control; 895 nM CY, 10.86+0.67% reduction, p<0.01; 8950 nM, 9.85+0.32% reduction, p<0.05. There was a similar significant reduction in rituximab-induced apoptosis in the BL transfected DLEU1 siRNA vs mock control cells (4 mg/mL Rituximab: 25.45+2.55% reduction, p<0.01 to negative control; 40 mg/mL Rituximab, 18.31+5.13% reduction, p<0.04; 400 mg/mL rituximab, 32.33+1.77% reduction, p<0.02).

Our pathway analysis of the 376 BL signature indicates that Toll-like receptor, JAK-STAT, and MAPK signaling pathways play important roles in gene regulation, anti-apoptosis, and proliferation, respectively, in PBL. The siRNA knock down of DLEU1 resulted in significantly less apoptosis on Ramos with CY or rituximab treatment. These data suggest that DLEU1 may in part play an important role in regulation of programmed cell death in BL and suggest a mechanism whereby BL with a 13q- abnormality may have a more aggressive clinical course. As CY or rituximab does not induce apoptosis in DLEU1 knockdown BL, the absence of DLEU1 may, in part, lead to drug resistance.

No relevant conflicts of interest to declare.