|Home | About | Journals | Submit | Contact Us | Français|
The diverged homeobox gene TLX1 (for T-cell leukemia homeobox 1, previously known as HOX11 or TCL3) is activated in ~5-10% of childhood and up to 30% of adult T-cell acute lymphoblastic leukemia (T-ALL) cases, most frequently by t(10;14)(q24;q11) and t(7;10)(q35;q24) chromosomal translocations which juxtapose the intact TLX1 coding region downstream of T cell receptor (TCR) δ (TRD@) or TCRβ (TCRB) gene regulatory sequences. TLX1+ T-ALL samples are virtually all arrested at the early cortical (CD1+) CD4+CD8+ “double-positive” (DP) stage of thymocyte development (Ferrando et al, 2002; Asnafi et al, 2004).
We previously showed that retroviral expression of TLX1 in primary murine hematopoietic-repopulating cells induced T-ALL-like malignancies in engrafted mice (Hawley et al, 1997). However, the murine model system did not perfectly reproduce the phenotype of human TLX1+ T-ALL (Owens et al, 2006). Furthermore, a low penetrance and long latency of tumor induction indicated the requirement for additional neoplastic mutations, consistent with a multistep mechanism of leukemogenesis (De Keersmaecker et al, 2006; Van Vlierberghe et al, 2008). In this latter regard, it has been found that the CDKN2A tumor suppressor locus is inactivated by homozygous or hemizygous deletion in most cases of TLX1+ disease (Sulong et al, 2008). Also, >50% of all T-ALL cases — including TLX1+ samples — harbor activating mutations in the NOTCH1 gene (Weng et al, 2004). Other recent work has identified two variant ABL1 fusion genes encoding constitutively activated tyrosine kinases in many TLX1+ T-ALL cases (De Keersmaecker et al, 2006); and tandem duplication of the MYB transcription factor oncogene has also been described (Lahortiga et al, 2007). The importance of these cooperative genetic aberrations in the multistep transformation of TLX1+ T-ALL is underscored by the ALL-SIL cell line (DSMZ no. ACC 511) (Riz & Hawley, 2005), which carries all of these alterations (summarized in Owens et al, 2006).
To gain a better understanding of the contribution of TLX1 to the malignant T-ALL phenotype, we down-regulated TLX1 expression in ALL-SIL cells by lentiviral-mediated expression of short-hairpin RNAs (shRNAs) against TLX1 transcripts. Using a transient cotransfection luciferase reporter assay, we identified two TLX1 shRNA lentiviral vectors targeting the TLX1 coding region (designated 94 and 95), which resulted in ~60-70% knockdown of reporter expression (Supplementary Fig S1). The 94 and 95 TLX1 shRNA lentiviral vectors, together with the pLKO.1-puro control lentiviral vector backbone, were produced as recombinant vector particles and used to stably transduce ALL-SIL cells. Reduced levels of TLX1 (Fig 1A) were associated with decreased growth of TLX1 shRNA-expressing ALL-SIL cell populations (Fig 1B). Of note, annexin V staining revealed increased numbers of apoptotic cells in the TLX1 knockdown cultures (Fig 1C) which correlated with increased cleavage of poly(ADP-ribose) polymerase 1 (Fig 1D). Importantly, the decreased fitness of the TLX1 knockdown ALL-SIL cells could not be rescued by ectopic expression of a fully active form (ICN1) of NOTCH1 (data not shown). The combined data thus demonstrated that despite acquisition of multiple cooperating mutations and adaptation to continuous growth in culture, ALL-SIL cells remained partially dependent on (or were addicted to) TLX1 expression.
We next examined the effects of TLX1 knockdown on global transcript levels by conducting gene expression profiling of ALL-SIL cells stably expressing the 95 TLX1 shRNA and ALL-SIL cells expressing the empty pLKO.1-puro lentiviral vector. Complementary RNA from three independent cultures of the experimental and control ALL-SIL cell lines was hybridized to Affymetrix HG-U133 Plus 2.0 GeneChip oligonucleotide arrays and analysis of expression data was performed as described previously (Riz & Hawley, 2005). Of 46 genes that displayed ≥1.9-fold changes in expression and had P values ≤0.05 for the three biological replicates (Supplementary Tables S1 and S2), 16 exhibited expression patterns following TLX1 knockdown consistent with a partial release of the early DP differentiation block (Supplementary Table S1) by comparison to expression profiles obtained for sorted subpopulations of primary human thymocytes and mature T cells (GEO accession no. GSE1460). This is best exemplified by the changes in the expression patterns of the CD1B and CD1E genes, characteristically expressed by TLX1+ T-ALL clinical samples (Ferrando et al, 2002; Asnafi et al, 2004). The CD1b and CD1e cell surface molecules, members of the lipid-presenting CD1 family of MHC-class-I-like glycoproteins, become down-regulated after TCR-mediated positive selection at the late DP stage of normal thymocyte development. In contrast, the CD55 gene, which encodes a glycosylphosphatidylinositol-anchored complement regulatory protein with signal transduction activity affecting a broad range of cellular properties, is an example of a gene that is more highly expressed in single positive thymocytes and mature T cells. For CD1B and CD55, the changes in mRNA expression detected by microarray profiling were validated at the level of the respective cell surface molecules by flow cytometric analysis (Fig 2A, top). Besides ALL-SIL, there are only a few other TLX1+ T-ALL cell lines that have been described in the literature (K3P, PER-255 and SUP-T4). Among these, the K3P cell line was available for our studies (Riz & Hawley, 2005). We were interested, therefore, in determining whether knockdown of TLX1 in K3P cells would also result in a partial release of the early cortical T-cell differentiation block. As can be seen in Fig 2A (bottom), K3P cells stably expressing the 95 TLX1 shRNA (>90% knockdown of TLX1 transcripts) exhibited irreversible down-modulation of CD1b cell surface levels concomitant with increased CD55 cell surface expression, paralleling the findings with ALL-SIL cells. To obtain further support of a direct relationship between TLX1 protein levels and the changes observed in CD1b and CD55 expression, ALL-SIL cells stably expressing the 95 TLX1 shRNA were viably sorted into CD1bhiCD55lo and CD1bloCD55hi subpopulations. Our presumption was that the residual CD1bhiCD55lo subpopulation was due to less efficient TLX1 knockdown (because of cell-to-cell variability in lentiviral vector-directed TLX1 shRNA expression) and consequently should express higher levels of TLX1 than the phenotypically more mature CD1bloCD55hi subpopulation. This prediction was confirmed by Western blotting for TLX1 (Fig 2B). We interpreted these results as direct evidence for a role of aberrant TLX1 expression in the early cortical DP differentiation block exhibited by TLX1+ T-ALL cells.
We regret that all literature could not be appropriately cited because of space constraints and we apologize to those authors whose work is not cited. We thank Reza Behnam for technical assistance and Eric Hoffman for access to the Molecular Genetics/Proteomics Core Facility of the Center for Genetic Medicine at Children’s National Medical Center. This work was supported in part by National Institutes of Health Grants R01HL66305 and R01HL65519, and by an Elaine H. Snyder Cancer Research Award and a King Fahd Endowed Professorship (to R.G.H.) from The George Washington University Medical Center.