DUX4-r selectively interacts with GTF2I. (A) Heatmap showing DUX4-r-selective interactors which are more expressed in REH cells. Protein levels were expressed as MS spectral counts. The heatmap reports unscaled MS/MS counts. (B) Tandem affinity purification followed by immunoblotting validation of selective GTF2I interaction with DUX4-IGH and DUX4-del50. Five percent of input and first elution and 10% of second elution were loaded and incubated with antibodies specific for GTF2I or wt DUX4 and DUX4-r. (C) Western blot validation of nuclear protein levels of GTF2I in REH, Jurkat, or REH cells using GAPDH as a loading control. (D) Densitometric analysis of signals in Fig. 3D using the ImageLab software (Bio-Rad, ver. 6.1). Bar plots represent the mean of three independent biological replicates (n = 3). Error bars represent ± SD. One-way ANOVA with multiple comparisons against REH. *P ≤ 0.05; ***P ≤ 0.001. (E) CUT&Tag profile plot of GTF2I enrichment on DUX4-r direct targets in HEK EV (red), REH EV (blue), or REH DUX4-IGH (teal) cells. Wilcoxon ranks sum test was performed to compare the signal of GTF2I between HEK EV, REH EV, and REH DUX4-IGH cells. (F) Heatmap showing the expression levels (FPKM, fragments per kilobase of transcript per million fragments mapped) of DUX4-r core genes in different DUX4-r B-ALL patients ordered based on decreasing GTF2I expression levels. The heatmap reports Z scores scaled by row. (G) Pearson's correlation plot showing the positive correlation between GTF2I expression levels and the mean activation of DUX4-r core genes in different DUX4-r B-ALL patients. The correlation coefficient (R) and P value are indicated. See also fig. S3.

Contexts in source publication

Context 1

... identified as wt DUX4-selective interactor (fig. S3C), thus providing a possible molecular explanation for DUX4-r lack of pioneer activity ( Fig. 1, G and H). Among the selective DUX4-r interactors preferentially expressed in REH cells, the most enriched in both DUX4-IGH and DUX4-del50 TAP-MS was general transcription factor IIi (GTF2I/TFII-I) (Fig. 4A). GTF2I is a transcription factor regulated by several signaling pathways, including the pre-B cell receptor (35,36). GTF2I translocations are associated with ALL and acute promyelocytic leukemia (37,38). Moreover, a recurrent GTF2I mutation is the most frequent oncogenic driver in thymic epithelial tumors ...

Context 2

... affinity purifications followed by immunoblotting confirmed GTF2I-selective interaction with both DUX4-r variants, but not wt DUX4 (Fig. 4B). Moreover, immunoblotting of total protein extracts confirmed significantly higher levels of GTF2I in REH compared to HEK and Jurkat cells (Fig. 4, C and D). Coimmunoprecipitation (co-IP) confirmed the interaction between the endogenous DUX4-r and GTF2I in NALM6 cells ( fig. ...

Context 3

... affinity purifications followed by immunoblotting confirmed GTF2I-selective interaction with both DUX4-r variants, but not wt DUX4 (Fig. 4B). Moreover, immunoblotting of total protein extracts confirmed significantly higher levels of GTF2I in REH compared to HEK and Jurkat cells (Fig. 4, C and D). Coimmunoprecipitation (co-IP) confirmed the interaction between the endogenous DUX4-r and GTF2I in NALM6 cells ( fig. ...

Context 4

... performed CUT&Tag for GTF2I in HEK and REH cells lacking expression of DUX4-r and in REH cells expressing DUX4-IGH. For direct DUX4-r targets, we observed a significantly higher GTF2I signal in REH compared to HEK cells, which was further increased upon DUX4-IGH expression (Fig. 4E). No significant difference in GTF2I chromatin association was observed in REH cells expressing DUX4-IGH as compared to EV-expressing REH cells, indicating that DUX4-IGH expression enhances GTF2I binding selectively to DUX4-r target genes ( fig. ...

Context 5

... ability in cell lines, we re-analyzed DUX4-r B-ALL patient datasets (6) to determine the expression levels of GTF2I and the level of activation of the DUX4-r core gene set in each patient. Despite all patients analyzed overexpressed the DUX4-r core gene set compared to healthy B cells or other ALL subtypes (6), they do so at variable levels (Fig. 4F). Notably, we found a significant and positive correlation between the expression levels of GTF2I and the relative activation of DUX4-r core genes (Fig. 4, F and G) strongly supporting the relevance of our ...

Context 6

... DUX4-r core gene set in each patient. Despite all patients analyzed overexpressed the DUX4-r core gene set compared to healthy B cells or other ALL subtypes (6), they do so at variable levels (Fig. 4F). Notably, we found a significant and positive correlation between the expression levels of GTF2I and the relative activation of DUX4-r core genes (Fig. 4, F and G) strongly supporting the relevance of our ...

Context 7

... we performed GTF2I knockdown in REH cells expressing ectopic DUX4-IGH or in NALM6 expressing endogenous DUX4-r (Fig. 5A). In line with a GTF2I requirement for DUX4-r transcriptional activity, we observed significant down-regulation of DUX4-r target genes (Fig. 5B). GTF2I silencing had no significant effect on the activation of wt DUX4 targets ( fig. S4A). ...

Context 8

... DUX4-IGH between EV and GTF2I conditions. (I) Leukemia expansion in peripheral blood of NSG mice 7 and 14 days after NALM6-Lucia cell transplantation. Bar plots show the mean Luciferase signal ± SD of shCTRL or shGTF2I NALM6-Lucia mice. (n = 5). Two-way ANOVA. **P ≤ 0.01. (J) Kaplan-Meier survival analysis of the above mice. *P ≤ 0.05. See also fig. ...

Context 9

... we investigated whether GTF2I is required for the cellular phenotypes induced by DUX4-r. GTF2I knockdown significantly reduced both the adhesion and transwell migration selectively of DUX4-r-expressing B-ALL cells (Fig. 5C and fig. S4B), leading to a significantly decreased tumor spheroid formation selectively in DUX4-r-expressing cells (Fig. 5D and fig. S4, C to E). GTF2I genetic down-regulation significantly reduced the proliferation of REH expressing ectopic DUX4-IGH (Fig. 5E) or NALM6 expressing endogenous DUX4-r ( fig. S4F), significantly increasing their cell ...

Context 10

... we investigated whether GTF2I is required for the cellular phenotypes induced by DUX4-r. GTF2I knockdown significantly reduced both the adhesion and transwell migration selectively of DUX4-r-expressing B-ALL cells (Fig. 5C and fig. S4B), leading to a significantly decreased tumor spheroid formation selectively in DUX4-r-expressing cells (Fig. 5D and fig. S4, C to E). GTF2I genetic down-regulation significantly reduced the proliferation of REH expressing ectopic DUX4-IGH (Fig. 5E) or NALM6 expressing endogenous DUX4-r ( fig. S4F), significantly increasing their cell death (Fig. ...

Context 11

... of DUX4-r-expressing B-ALL cells (Fig. 5C and fig. S4B), leading to a significantly decreased tumor spheroid formation selectively in DUX4-r-expressing cells (Fig. 5D and fig. S4, C to E). GTF2I genetic down-regulation significantly reduced the proliferation of REH expressing ectopic DUX4-IGH (Fig. 5E) or NALM6 expressing endogenous DUX4-r ( fig. S4F), significantly increasing their cell death (Fig. ...

Context 12

... encoding for GTF2I or a control empty vector (EV) in HEK cells together with EV, wt DUX4, or DUX4-IGH. GTF2I was expressed at comparable levels in all cell lines and did not affect wt DUX4 or DUX4-IGH expression levels (Fig. 5G). Notably, in HEK cells, ectopically expressed GTF2I reached levels comparable to endogenous GTF2I in REH cells ( fig. S4G). GTF2I expression in HEK cells significantly increased target gene activation by DUX4-IGH (Fig. 5G), while wt DUX4 activity was unaffected ( fig. ...

Context 13

... in all cell lines and did not affect wt DUX4 or DUX4-IGH expression levels (Fig. 5G). Notably, in HEK cells, ectopically expressed GTF2I reached levels comparable to endogenous GTF2I in REH cells ( fig. S4G). GTF2I expression in HEK cells significantly increased target gene activation by DUX4-IGH (Fig. 5G), while wt DUX4 activity was unaffected ( fig. ...

Context 14

... investigate the molecular determinants of the observed DUX4-r transcriptional "rescue" by GTF2I, we performed CUT&Tag in HEK cells expressing DUX4-IGH in the presence/ absence of transfected GTF2I. In the absence of GTF2I, DUX4-IGH displayed negligent chromatin association genome-wide ( fig. S4I). Instead, upon GTF2I supplementation, DUX4-IGH displayed a significantly enhanced ability to associate with its target gene loci in HEK cells (Fig. 5H). In line with the fact that we identified GTF2I as a selective DUX4-r interactor, wt DUX4 association to its target loci was unaffected by GTF2I overexpression ( fig. ...

Context 15

... genome-wide ( fig. S4I). Instead, upon GTF2I supplementation, DUX4-IGH displayed a significantly enhanced ability to associate with its target gene loci in HEK cells (Fig. 5H). In line with the fact that we identified GTF2I as a selective DUX4-r interactor, wt DUX4 association to its target loci was unaffected by GTF2I overexpression ( fig. ...

Context 16

... blood of GTF2I knockdown animals compared to controls (Fig. 5I). This translated into significantly higher survival rates in mice transplanted with GTF2I shRNA compared to control shRNA leukemia cells (Fig. 5J). At sacrifice, GTF2I knockdown transplants showed also a significantly lower leukemia burden in both BM and spleen compartments (fig . S4K). We found that GTF2I shRNA cells harvested from the mice at sacrifice displayed a milder reduction in the expression level of GTF2I and DUX4-r targets compared to the one that we observed in vitro (compare Fig. 5, A and B, with fig. S4L). These results are in line with the possibility of a negative selection against cells with a ...

Context 17

... At sacrifice, GTF2I knockdown transplants showed also a significantly lower leukemia burden in both BM and spleen compartments (fig . S4K). We found that GTF2I shRNA cells harvested from the mice at sacrifice displayed a milder reduction in the expression level of GTF2I and DUX4-r targets compared to the one that we observed in vitro (compare Fig. 5, A and B, with fig. S4L). These results are in line with the possibility of a negative selection against cells with a strong decrease with GTF2I expression in vivo. All in all, our findings indicate that GTF2I is required for DUX4-r transcriptional and leukemogenic ...