Page 38 - Haematologica August 2018

P. Shyamsunder et al.
Figure 4. Card10 locus exhibits transcriptional activation signature in granulocytes. H3K27ac, H3K4me1 and H3K4me3 ChIP-seq signals in LT-HSC, CMP, GMP and granulocytes (Gr) and IGV track of ATAC-seq profile of sorted granulocytes (bottom track). Tracks are represented in frame with the CEBPE ChIP-seq peak of Card10 gene.22
1274
We identified that CEBPE binds upstream of Card10 gene (-1.5 kb and -7kb) (Figure 3A). Interestingly, expres- sion profile of Card10 and Cebpe followed a similar pat- tern, with the highest level of expression observed in gran- ulocytes (https://www.ebi.ac.uk/gxa/experiments/E-MTAB- 3079) (Figure 3B). Expression of Card10 was significantly lower in immature granulocytes from Cebpe KO mice compared with the same stage of differentiation of the wild type mice (Figure 3C, Online Supplementary Figure S2B).
On scanning the ChIP-seq data, peaks occurred at the - 7 kb and -1.5 kb regions upstream of Card10 gene. We also verified the binding of Cebpe to know targets such as Lactoferrin (Ltf) and Neutrophilic granule precursor pro- tein (Ngp) (Online Supplementary Figure S3A). Examination of the nucleotide sequence using ConSite (an online soft- ware tool to predict transcription factor motifs) revealed two CEBPE motifs within the -7kb peak and one CEBPE motif was detected in the -1.5kb peak (Online Supplementary Figure S3B). The -7kb region had stronger binding of CEBPE compared with the -1.5 Kb region; and therefore, the region was further studied. CEBPE binding to the -7kb peak was validated using ChIP-PCR in bone marrow cells from Cebpe WT mice; and this binding was significantly reduced in the KO cells (Figure 3D and Online Supplementary Figure S4).
To assess whether the interactions of CEBPE is function- ally relevant, luciferase reporter assay and electrophoretic mobility shift assay (EMSA) were performed. Luciferase reporter assay with a 500 bp fragment containing the CEBPE motif sequence transfected into NIH/3T3 cells revealed that CEBPE was able to transactivate the luciferase reporter in a dose-dependent manner (Figure 3E). Next, biotinylated oligos encompassing the putative CEBP motif were designed and EMSA assays were per- formed. Incubation of biotinylated oligos with nuclear extract from cells ectopically expressing CEBPE caused a shift in migration of the biotinylated oligos. This shift dis-
appeared when the nuclear extract complex was incubat- ed with either 100-fold molar excess of unlabelled com- petitor oligos (cold competition) or with biotinylated oli- gos harboring a mutation in CEBP motif (Figure 3F). Taken together, these results indicate that Card10 is a direct tar- get of CEBPE.
Epigenetic landscape and expression of Card10 reveals an exclusive signature in granulocytes
Comparison of H3K27ac, H3K4me1 and H3K4me3 occupancy at Card10 locus (at the core promoter and -7kb CEBPE binding region) in LT-HSC, CMP, GMP and granu- locytes, revealed transcriptional activation signatures exclusively in granulocytes (Figure 4). ATAC-Seq of sorted granulocytes22 also indicated that the region bound by CEBPE is located in open chromatin (Figure 4). These find- ings along with exclusive expression of Card10 in granulo- cytes suggests a role in myeloid differentiation.
CARD10 regulates neutrophil differentiation in vitro Since Card10 expression is controlled by CEBPE, we hypothesized that CARD10 may have a role in granu- lopoesis. To test this hypothesis, CARD10 expression was stably silenced using short hairpin RNA (shRNA) in NB4 cells, an acute promyelocytic leukemia cell line that can be differentiated into granulocytes in the presence of all-trans retinoic acid (ATRA) (Figure 5A). We observed that knock- down of CARD10 resulted in a reduced proportion of CD11b+ cells upon induction with ATRA (control shRNA: 93% and CARD10 KD: 83% (sh4) and 63% (sh5) 4 days incubation). Although the data are not statistically signifi- cant, we did observe a trend towards impaired differenti- ation following Card10 knock-down (Fig. 5B). In parallel as a control, we also verified granulocytic differentiation after ATRA induction of CEBPE knock-down cells (Online Supplementary Figure S5A and S5B). Viability assay with control and Card10 KD cells revevaled no major effect on
cell proliferation (Online Supplementary Figure S5C).
haematologica | 2018; 103(8)